Rwanda - Demographic and Health Survey - 2006

Publication date: 2006

REPUBLIC OF RWANDA Rwanda Demographic and Health Survey 2005 Institut National de la Statistique Ministère des Finances et de la Planification Économique Kigali, Rwanda ORC Macro Calverton, Maryland, USA July 2006 The following people participated in data analysis and the preparation of this report: Jean Philippe Gatarayiha, Apolline Mukanyonga, Dr Eugénie Kayirangwa, Adeline Kabeja, Alphonse Rukundo, Athanasie Kabagwira, Gafishi N. Philippe, Rwabikumba Dévote, Rwakayiro Ignace, Muhoza Ananie, Kalinda Charles, Kayibanda Françoise, Fern Greenwell, Noah Bartlett, Mohamed Ayad, and Monique Barrère. This report presents the findings of the 2005 Rwanda Demographic and Health Survey (RDHS-III), carried out from February to July 2005 by the Direction de la Statistique (renamed Institut National de la Statistique du Rwanda in September 2005). Funding for the RDHS-III was provided by USAID, the Commission Nationale de Lutte contre le SIDA (CNLS) through the World Bank’s Multi-County AIDS Program (MAP), Unicef, UNFPA, DFID and GTZ. Assistance was also provided by other national organizations, such as the Treatment and Research AIDS Center (TRAC), the Laboratoire National de Référence and the Service National de Recensement (SNR). Technical assistance was provided by ORC Macro as part of the Demographic and Health Surveys project (MEASURE DHS). The objective of the MEASURE DHS project is to collect, analyze and disseminate demographic data, especially those related to fertility, family planning, maternal and child health, and HIV/AIDS. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID or other cooperating organizations. Additional information about the survey can be obtained from the Institut National de la Statistique du Rwanda (INSR), BP 6139, Kigali, Rwanda (Telephone: (250) 55104164; e-mail: snr@rwanda1.com). Additional information about the MEASURE DHS project can be obtained from ORC Macro, 11785 Beltsville Drive, Suite 300, Calverton, MD 20705, USA. Telephone: 301-572-0200; Fax: 301-572-0999; e-mail: reports@orcmacro.com; Internet: http://www.measuredhs.com). Recommended citation: Institut National de la Statistique du Rwanda (INSR) and ORC Macro. 2006. Rwanda Demographic and Health Survey 2005. Calverton, Maryland, U.S.A.: INSR and ORC Macro. Contents | iii CONTENTS Page TABLES AND FIGURES . ix FOREWORD . xix ACKNOWLEDGMENTS . xxi ABBREVIATIONS . xxiii SUMMARY OF FINDINGS . xxv MILLENNIUM DEVELOPMENT GOAL INDICATORS. xxxi MAP OF RWANDA.xxxii CHAPTER 1 INTRODUCTION 1.1 Country Profile .1 1.1.1 Geography .1 1.1.2 Economy.2 1.1.3 Population .3 1.1.4 Population Policy .4 1.1.5 Public Health Policy.4 1.2 Objectives and Methodology of the Survey.4 1.2.1 Objectives of the Survey .5 1.2.2 Questionnaires.5 1.2.3 Sample Design .7 1.2.4 Sample Coverage .7 1.2.5 Hemoglobin and HIV Testing .8 1.2.6 Training and Data Collection.9 1.2.7 Data Processing.10 CHAPTER 2 HOUSEHOLD CHARACTERISTICS 2.1 Household Population By Age and Sex .11 2.2 Household Size and Composition.12 2.3 School Attendance and Educational Attainment.13 2.4 Living Conditions.18 2.5 Birth Registration with Civil Authorities .21 iv │ Contents CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS 3.1 Background Characteristics of Respondents .23 3.2 Educational Attainment.25 3.3 Literacy.27 3.4 Exposure to Mass Media .28 3.5 Employment .31 CHAPTER 4 FERTILITY 4.1 Fertility Levels and Differentials .38 4.2 Fertility Trends.41 4.3 Parity and Primary Infertility.44 4.4 Birth Intervals .46 4.5 Age at First Birth .48 4.6 Teenage Fertility .49 CHAPTER 5 FAMILY PLANNING 5.1 Knowledge of Contraception .51 5.2 Use of Contraception.52 5.2.1 Ever Use of Contraception .52 5.2.2 Current Use of Contraception .54 5.3 Number of Children at First Use of Contraception .59 5.4 Knowledge of Fertile Period.59 5.5 Source of Contraception.60 5.6 Future Use of Contraception.61 5.7 Exposure to Family Planning Messages.64 5.8 Contact of Nonusers with Family Planning Providers.51 5.9 Opinions and Attitudes of Couples toward Family Planning .51 5.9.1 Discussion of Family Planning with Husband.51 5.9.2 Attitudes of Couples toward Family Planning.52 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY 6.1 Marital Status.71 6.2 Polygyny.72 6.3 Age at First Union.73 6.4 Age at First Sexual Intercourse .76 6.4 Recent Sexual Activity.78 6.6 Exposure to the Risk of Pregnancy .81 6.7 Menopause .82 Contents | v CHAPTER 7 FERTILITY PREFERENCES 7.1 Desire for (More) Children.85 7.2 Need for Family Planning Services .88 7.3 Ideal Number of Children.90 7.4 Fertility Planning Status .92 CHAPTER 8 MOTHER AND CHILD HEALTH 8.1 Antenatal Care.95 8.2 Delivery Care . 100 8.3 Postnatal Care . 105 8.4 Vaccination of Children . 106 8.5 Childhood Illnesses . 109 8.5.1 Acute Respiratory Infection (ARI) and Fever. 109 8.5.2 Diarrhea . 111 8.6 Problems in Accessing Health Care . 116 8.7 Tobacco Consumption. 117 CHAPTER 9 MALARIA 9.1 Introduction . 119 9.2 Malaria Prevention . 121 9.2.1 Household Possession of Mosquito Nets . 122 9.2.2 Use of Mosquito Nets by Children . 123 9.2.3 Use of Mosquito Nets by Women . 125 9.2.4 Intermittent Preventive Treatment during Pregnancy . 127 9.3 Treatment of Malaria in Children Under the Age of Five . 128 CHAPTER 10 BREASTFEEDING AND NUTRITION OF MOTHERS AND CHILDREN 10.1 Breastfeeding and Supplementation. 131 10.2 Micronutrient Intake And Anemia Prevalence . 137 10.3 Prevalence of Anemia Due to Iron Deficiency. 143 10.4 Nutritional Status of Children. 147 10.5 Nutritional Status of Women . 154 CHAPTER 11 INFANT AND CHILD MORTALITY 11.1 Definition, Methodology, and Data Quality . 157 11.2 Levels and Trends . 158 11.3 Differentials in Infant and Child Mortality . 160 11.4 Perinatal Mortality . 163 vi │ Contents 11.5 High-Risk Fertility Behavior. 165 CHAPTER 12 MATERNAL MORTALITY 12.1 Introduction . 167 12.2 Data Collection. 167 12.3 Data Quality. 168 12.4 Direct Estimates of Adult Mortality. 170 12.5 Direct Estimates of Maternal Mortality . 172 CHAPTER 13 DOMESTIC VIOLENCE 13.1 Methodology . 175 13.2 Domestic Violence . 177 13.2.1 Physical Violence Since Age 15 . 177 13.2.2 Violence during Pregnancy. 178 13.2.3 Marital Control Exercised by the Husband/Partner . 179 13.3 Spousal Violence . 181 13.3.1 Prevalence of Spousal Violence. 181 13.3.2 Frequency of Recent Spousal Violence. 183 13.3.3 Onset of Spousal Violence . 184 13.4 Consequences Of Violence And Help Seeking . 185 13.5 Violence by Spousal Characteristics and Women’s Status Indicators. 187 CHAPTER 14 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR 14.1 Knowledge, Opinions, And Attitudes . 192 14.1.1 Knowledge of HIV Transmission and Prevention Methods . 192 14.1.2 Stigmatization . 198 14.1.3 Opinions. 200 14.2 Higher-Risk Sexual Intercourse and Condom Use . 203 14.3 Testing and Counseling for HIV/AIDS. 206 14.4 Sexually Transmitted Infections (STIs). 209 14.5 Injections from a Health Worker. 211 14.6 Knowledge of HIV/AIDS and Sexual Behavior among Youth . 214 CHAPTER 15 HIV PREVALENCE AND ASSOCIATED FACTORS 15.1 HIV Testing Protocol. 225 15.2 Coverage of HIV Testing . 228 15.3 HIV Prevalence. 231 Contents | vii 15.3.1 HIV Prevalence Distribution According to Sociodemographic Variables. 231 15.3.2 HIV Prevalence by Demographic Variables . 233 15.3.3 HIV Prevalence by Sexual Behavior Characteristics. 234 15.3.4 HIV Prevalence among Youth. 236 15.3.5 HIV Prevalence and Other Risk Factors . 238 15.3.6 HIV Prevalence and Male Circumcision . 239 15.3.7 HIV Prevalence among Couples . 239 15.4 Sentinel Surveillance System and RDHS-III . 241 CHAPTER 16 ORPHANED AND VULNERABLE CHILDREN 16.1 Orphanhood and Children’s Living Arrangements. 243 16.2 Access to Essential Services . 246 16.3 Strengthening Family Capacities to Support And Protect Children. 247 16.3.1 Malnutrition . 247 16.3.2 Early Sexual Intercourse . 248 16.3.3 Succession Planning . 248 16.4 Protection of Vulnerable Children. 249 16.5 Care and Support . 250 16.5.1 Care and Support of the Chronically Ill. 250 16.5.2 Care and Support of OVC. 251 REFERENCES .253 APPENDIX A SAMPLE IMPLEMENTATION A.1 Introduction . 257 A.2 Sample Frame . 257 A.3 Sample Selection . 258 A.4 Sampling Probability . 259 A.5 Survey Results. 259 APPENDIX B ESTIMATES OF SAMPLING ERRORS. 267 APPENDIX C DATA QUALITY TABLES . 279 APPENDIX D RESULTS ACCORDING TO OLD PROVINCES . 285 APPENDIX E PERSONS INVOLVED IN THE 2005 RWANDA DEMOGRAPHIC AND HEALTH SURVEY. 323 APPENDIX F QUESTIONNAIRES . 327 Tables and Figures | ix TABLES AND FIGURES Page CHAPTER 1 INTRODUCTION Table 1.1 Results of the household and individual interviews. 8 CHAPTER 2 HOUSEHOLD CHARACTERISTICS Table 2.1 Household population by age, sex, and residence. 11 Table 2.2 Household composition. 13 Table 2.3.1 Educational attainment of household population: female. 14 Table 2.3.2 Educational attainment of household population: male. 15 Table 2.4 School attendance ratios . 16 Table 2.5 Housing characteristics. 19 Table 2.6 Household durable goods . 20 Table 2.7 Wealth quintiles. 21 Table 2.8 Birth registration of children under age five . 22 Figure 2.1 Population pyramid . 12 Figure 2.2 Age-specific attendance rates . 18 CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS Table 3.1 Age of respondents . 23 Table 3.2 Background characteristics of respondents . 24 Table 3.3.1 Educational attainment by background characteristics: women . 25 Table 3.3.2 Educational attainment by background characteristics: men. 26 Table 3.4.1 Literacy: women. 27 Table 3.4.2 Literacy: men. 28 Table 3.5.1 Exposure to mass media: women. 29 Table 3.5.2 Exposure to mass media: men. 30 Table 3.6 Employment status. 32 Table 3.7.1 Occupation: women. 33 Table 3.7.2 Occupation: men . 34 Table 3.8 Type of employment. 35 Table 3.9 Type of employer. 36 CHAPTER 4 FERTILITY Table 4.1 Current fertility . 38 Table 4.2 Fertility by background characteristics . 40 Table 4.3 Trends in fertility. 41 x | Tables and Figures Table 4.4 Trends in age-specific fertility rates. 42 Table 4.5.1 Children ever born and living : women. 44 Table 4.5.2 Children ever born and living : men. 45 Table 4.6 Birth intervals. 47 Table 4.7 Age at first birth . 48 Table 4.8 Median age at first birth by background characteristics. 49 Table 4.9 Teenage pregnancy and motherhood. 50 Figure 4.1 Age-Specific Fertility Rates, by Residence. 39 Figure 4.2 Total Fertility Rate and Mean Number of Children Ever Born to Women Age 40-49. 40 Figure 4.3 Trends in Age-Specific Fertility Rates, Rwanda 1992, 2000, and 2005 . 42 Figure 4.4 Age-Specific Fertility Rates for Five-Year Periods Preceding the Survey . 43 Figure 4.5 Trends in the Total Fertility Rate among Women Age 15-34, Rwanda 1992, 2000, and 2005 . 43 CHAPTER 5 FAMILY PLANNING Table 5.1.1 Knowledge of contraceptive methods: women. 51 Table 5.1.2 Knowledge of contraceptive methods: men . 52 Table 5.2 Ever use of contraception . 53 Table 5.3 Current use of contraception . 55 Table 5.4 Current use of contraception by background characteristics . 58 Table 5.5 Number of children at first use of contraception . 59 Table 5.6 Knowledge of the fertile period. 60 Table 5.7 Source of contraception. 61 Table 5.8 Future use of contraception . 62 Table 5.9 Reason for not intending to use contraception . 63 Table 5.10 Preferred method of contraception for future use. 64 Table 5.11.1 Exposure to family planning messages: women . 65 Table 5.11.2 Exposure to family planning messages: men. 66 Table 5.12 Contact of nonusers with family planning providers . 67 Table 5.13 Discussion of family planning with husband . 68 Table 5.14 Attitudes towards family planning . 69 Figure 5.1 Contraceptive Use among Currently Married Women Age 15-49. 56 Figure 5.2 Trends in Use of Modern Methods among Currently Married Women . 57 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY Table 6.1 Current marital status . 71 Table 6.2 Number of co-wives and wives . 73 Table 6.3 Age at first marriage . 74 Table 6.4 Median age at first marriage. 75 Table 6.5 Age at first sexual intercourse:. 76 Table 6.6 Median age at first sexual intercourse. 78 Table 6.7.1 Recent sexual activity: women. 79 Tables and Figures | xi Table 6.7.2 Recent sexual activity: men. 80 Table 6.8 Postpartum amenorrhea, abstinence, and insusceptibility. 81 Table 6.9 Median duration of postpartum insusceptibility by background characteristics . 82 Table 6.10 Menopause. 83 Figure 6.1 Percentage of Never-Married Women and Men, by Age . 72 Figure 6.2 Median Age at First Marriage among Women and Men, by Background Characteristics. 75 Figure 6.3 Median Age at First Intercourse and at First Union among Women 25-49, by Background Characteristics . 77 CHAPTER 7 FERTILITY PREFERENCES Table 7.1 Fertility preferences by number of living children . 86 Table 7.2 Desire to limit childbearing . 88 Table 7.3 Need for family planning among currently married women. 89 Table 7.4 Ideal number of children . 91 Table 7.5 Mean ideal number of children. 92 Table 7.6 Fertility planning status. 93 Table 7.7 Wanted fertility rates. 94 Figure 7.1 Proportion of Currently Married Women and Men Who Want No More Children, by Number of Living Children. 87 CHAPTER 8 MATERNAL AND CHILD HEALTH Table 8.1 Antenatal care. 96 Table 8.2 Number of antenatal care visits and timing of first visit . 97 Table 8.3 Components of antenatal care . 98 Table 8.4 Tetanus toxoid injections . 100 Table 8.5 Place of delivery . 101 Table 8.6 Assistance during delivery . 102 Table 8.7 Delivery characteristics . 104 Table 8.8 Postnatal care . 106 Table 8.9 Vaccinations by source of information. 107 Table 8.10 Vaccinations by background characteristics . 109 Table 8.11 Prevalence and treatment of symptoms of ARI and fever. 110 Table 8.12 Prevalence of diarrhea . 112 Table 8.13 Knowledge of ORS packets . 114 Table 8.14 Diarrhea treatment . 115 Table 8.15 Feeding practices during diarrhea . 116 Table 8.16 Problems in accessing health care . 117 Table 8.17 Use of smoking tobacco. 118 Figure 8.1 Trends in Antenatal Care and Delivery, Rwanda 1992, 2000, and 2005 . 97 Figure 8.2 Children Whose Delivery Was Assisted by Trained Personnel . 103 xii | Tables and Figures Figure 8.3 Trends in Vaccination Coverage among Children Age 12-23 Months, Rwanda 1992, 2000, and 2005 . 108 Figure 8.4 Prevalence of ARI, Fever, and Diarrhea, by Age . 113 CHAPTER 9 MALARIA Table 9.1 Household possession of mosquito nets . 122 Table 9.2 Use of mosquito nets by children. 124 Table 9.3 Use of mosquito nets by women. 126 Table 9.4 Use of Intermittent Preventive Treatment by women during pregnancy. 127 Table 9.5 Use of SP/Fansidar by women during pregnancy . 128 Table 9.6 Prevalence and prompt treatment of children with fever . 129 Table 9.7 Type and timing of antimalarial drugs taken by children with fever . 130 Figure 9.1 Household Ownership of Mosquito Nets . 123 Figure 9.2 Use of Mosquito Nets by Children Under Age 5, According to Province. 125 Figure 9.3 Pregnant Women Who Slept Under a Mosquito Net the Night Preceding the Survey. 126 CHAPTER 10 BREASTFEEDING AND NUTRITION OF MOTHERS AND CHILDREN Table 10.1 Initial breastfeeding. 132 Table 10.2 Breastfeeding status by age . 134 Table 10.3 Median duration and frequency of breastfeeding . 136 Table 10.4 Foods consumed by children in the day or night preceding the interview. 137 Table 10.5 Iodization of household salt . 138 Table 10.6 Micronutrient intake among children . 140 Table 10.7 Micronutrient intake among mothers . 142 Table 10.8 Prevalence of anemia in children . 144 Table 10.9 Prevalence of anemia in women . 145 Table 10.10 Prevalence of anemia in children by anemia status of mother . 146 Table 10.11 Prevalence of anemia in men. 147 Table 10.12 Nutritional status of children . 149 Table 10.13 Nutritional status of women . 156 Figure 10.1 Breastfeeding Practices Among Children Under Age 3 . 135 Figure 10.2 Percentage of Children Under Age 5 Who Are Stunted . 150 Figure 10.3 Percentage of Children Under Age 5 Who Are Wasted . 152 Figure 10.4 Trends in malnutrition among Children under 5 Years), Rwanda 1992, 2000, and 2005. 154 CHAPTER 11 INFANT AND CHILD MORTALITY Table 11.1 Early childhood mortality rates . 158 Table 11.2 Early childhood mortality rates by background characteristics. 161 Table 11.3 Early childhood mortality rates by demographic characteristics. 162 Table 11.4 Perinatal mortality. 164 Tables and Figures | xiii Table 11.5 High-risk fertility behavior . 166 Figure 11.1 Trends in Infant and Under-five Mortality, Rwanda 1992, 2000, and 2005. 159 Figure 11.2 Trends in Infant and Under-five Mortality from the RDHS-I, RDHS-II, and RDHS-III . 160 Figure 11.3 Under-five Mortality by Mother’s Background Characteristics . 161 Figure 11.4 Infant Mortality by Reproductive Behavior . 163 CHAPTER 12 MATERNAL MORTALITY Table 12.1 Data on siblings . 168 Table 12.2 Indicators on data quality. 169 Table 12.3 Estimates of age-specific female and male adult mortality. 170 Table 12.4 Maternal mortality . 173 Figure 12.1 Female Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group . 171 Figure 12.2 Male Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group . 172 CHAPTER 13 DOMESTIC VIOLENCE Table 13.1 Experience of beatings or physical mistreatment . 177 Table 13.2 Perpetrators of violence . 178 Table 13.3 Violence during pregnancy . 179 Table 13.4 Marital control exercised by husband. 180 Table 13.5 Marital violence . 182 Table 13.6 Frequency of spousal violence . 184 Table 13.7 Onset of spousal violence . 185 Table 13.8 Physical consequences of spousal violence. 186 Table 13.9 Help seeking. 187 Table 13.10 Spousal violence, women's status, and spousal characteristics. 188 Figure 13.1 Percentage of Ever-Married Women who Have Ever Experienced Specific Forms of Violence from Their Husbands . 183 Figure 13.2 Prevalence of Spousal Violence, by Level of Education of Woman and Her Spouse and Alcohol Consumption of Spouse. 189 CHAPTER 14 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR Table 14.1 Knowledge of AIDS. 192 Table 14.2 Knowledge of HIV prevention methods. 193 Table 14.3.1 Comprehensive knowledge about AIDS : women . 195 Table 14.3.2 Comprehensive knowledge about AIDS : men . 196 Table 14.4 Knowledge of prevention of mother-to-child transmission of HIV. 197 Table 14.5.1 Accepting attitudes toward those living with HIV/AIDS: women. 199 Table 14.5.2 Accepting attitudes toward those living with HIV/AIDS: men. 200 xiv | Tables and Figures Table 14.6 Attitudes toward negotiating safer sexual relations with husband . 201 Table 14.7 Adult support of education about condom use to prevent AIDS . 202 Table 14.8.1 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: women . 204 Table 14.8.2 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: men . 205 Table 14.9.1 Prior HIV testing and knowledge of results: women . 206 Table 14.9.2 Prior HIV testing and knowledge of results: men . 207 Table 14.10 Pregnant women counseled and tested for HIV. 209 Table 14.11 Self-reported prevalence of sexually-transmitted infections (STIs) and STI symptoms . 210 Table 14.12 Prevalence of injections . 212 Table 14.13 Comprehensive knowledge about AIDS and of a source of condoms among youth . 214 Table 14.14 Age at first sexual intercourse among youth. 216 Table 14.15 Condom use at first sexual intercourse among youth. 217 Table 14.16 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth . 218 Table 14.17 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months . 219 Table 14.18 Age-mixing in sexual relationships among women age 15-19 . 221 Table 14.19 Drunkenness during sexual intercourse among youth. 222 Table 14.20 Recent HIV tests among youth . 223 Figure 14.1 Perception and Beliefs about Abstinence and Faithfulness. 203 Figure 14.2 Women and Men Seeking Treatment for STIs . 211 Figure 14.3 Type of Facility where Received Last Medical Injection . 213 Figure 14.4 Percentage whose Last Injection was Given with a Syringe and Needle Taken from a New, Unopened Package. 213 Figure 14.5 Trends in Age at First Sex, Rwanda 2000 and 2005. 217 Figure 14.6 Abstinence, Being Faithful, and Condom Use (ABC) Among Young Women and Men . 220 CHAPTER 15 HIV PREVALENCE AND ASSOCIATED FACTORS Table 15.1 Coverage of HIV testing by residence and province. 228 Table 15.2 Coverage of HIV testing by background characteristics . 230 Table 15.3 HIV prevalence by age. 231 Table 15.4 HIV prevalence by background characteristics . 232 Table 15.5 HIV prevalence and confidence intervals . 233 Table 15.6 HIV prevalence by sociodemographic characteristics . 234 Table 15.7 HIV prevalence by sexual behavior characteristics. 235 Table 15.8 HIV prevalence among young people . 237 Table 15.9 HIV prevalence by other characteristics. 238 Table 15.10 Prior HIV testing by HIV status . 239 Table 15.11 HIV prevalence by male circumcision . 239 Table 15.12 HIV prevalence among couples. 240 Tables and Figures | xv Figure 15.1 HIV Prevalence by Sex and Age . 231 CHAPTER 16 ORPHANED AND VULNERABLE CHILDREN Table 16.1 Children’s living arrangements and orphanhood . 244 Table 16.2 Orphans and vulnerable children (OVC) . 245 Table 16.3 School attendance by survivorship of parents and by OVC status . 246 Table 16.4 Underweight orphans and vulnerable children. 247 Table 16.5 Sexual intercourse before age 15 among orphans and vulnerable children . 248 Table 16.6 Succession planning. 249 Table 16.7 Widows dispossessed of property. 250 Table 16.8 External support for chronically ill persons . 251 Table 16.9 External support for orphans and vulnerable children. 252 APPENDIX A SAMPLE IMPLEMENTATION Table A.1 Distribution of households and enumeration areas (EAs) by old province and according to residence (RGPH, 2002) . 257 Table A.2 Sample allocation by old province and according to residence. 258 Table A.3 Sample implementation: women . 260 Table A.4 Sample implementation: men. 261 Table A.5 Coverage of HIV testing among interviewed women by background characteristics . 262 Table A.6 Coverage of HIV testing among interviewed men by background characteristics . 263 Table A.7 Coverage of HIV testing among women who ever had sex by risk status variables . 264 Table A.8 Coverage of HIV testing among men who ever had sex by risk status variables . 265 APPENDIX B ESTIMATES OF SAMPLING ERRORS Table B.1 List of selected variables for sampling errors . 270 Table B.2 Sampling errors - National sample. 271 Table B.3 Sampling errors - Urban sample . 272 Table B.4 Sampling errors - Rural sample. 273 Table B.5 Sampling errors – City of Kigali. 274 Table B.6 Sampling errors – South Province. 275 Table B.7 Sampling errors – West Province. 276 Table B.8 Sampling errors – North Province. 277 Table B.9 Sampling errors – East Province. 278 APPENDIX C DATA QUALITY TABLES Table C.1 Household age distribution . 279 Table C.2.1 Age distribution of eligible and interviewed women . 280 Table C.2.2 Age distribution of eligible and interviewed men. 280 xvi | Tables and Figures Table C.3 Completeness of reporting . 281 Table C.4 Births by calendar years . 281 Table C.5 Reporting of age at death in days . 282 Table C.6 Reporting of age at death in months. 283 APPENDIX D RESULTS ACCORDING TO OLD PROVINCES Table D.2.3 Educational attainment of household population . 285 Table D.2.4 School attendance ratios . 286 Table D.2.7 Wealth quintiles. 286 Table D.2.8 Birth registration of children under age five . 287 Table D.3.3 Educational attainment . 287 Table D.3.4 Literacy. 288 Table D.3.5 Exposure to mass media. 288 Table D.3.6 Employment status. 289 Table D.3.6 Occupation. 289 Table D.4.2 Fertility by old province . 290 Table D.4.6 Birth Intervals. 290 Table D.4.8 Median age at first birth . 290 Table D.4.9 Teenage pregnancy and motherhood. 291 Table D.5.4 Current use of contraception by background characteristics . 291 Table D.5.11 Exposure to family planning messages . 292 Table D.6.2 Number of co-wives and wives . 292 Table D.6.4 Median age at first marriage. 293 Table D.6.6 Median age at first sexual intercourse. 293 Table D.6.7 Recent sexual activity. 294 Table D.6.9 Median duration of postpartum insusceptibility by background characteristics . 294 Table D.7.2 Desire to limit childbearing . 295 Table D.7.3 Need for family planning among currently married women. 295 Table D.7.5 Mean ideal number of children. 295 Table D.7.7 Wanted fertility rates. 296 Table D.8.1 Antenatal care. 296 Table D.8.3 Components of antenatal care . 297 Table D.8.4 Tetanus toxoid injections . 297 Table D.8.5 Place of delivery . 297 Table D.8.6 Assistance during delivery . 298 Table D.8.7 Delivery characteristics . 298 Table D.8.8 Postnatal care . 299 Table D.8.10 Vaccinations . 299 Table D.8.11 Prevalence and treatment of symptoms of ARI and fever. 300 Table D.8.12 Prevalence of diarrhea . 300 Table D.8.13 Knowledge of ORS packets . 300 Table D.8.14 Diarrhea treatment . 301 Table D.8.16 Problems in accessing health care . 301 Table D.9.1 Household possession of mosquito nets . 302 Table D.9.2 Use of mosquito nets by children. 302 Tables and Figures | xvii Table D.9.3 Use of mosquito nets by women. 303 Table D.9.4 Use of Intermittent Preventive Treatment by women during pregnancy. 303 Table D.9.6 Prevalence and prompt treatment of children with fever . 304 Table D.9.7 Type and timing of antimalarial drugs taken by children with fever . 304 Table D.10.1 Initial breastfeeding. 304 Table D.10.3 Median duration and frequency of breastfeeding . 305 Table D.10.5 Iodization of household salt . 305 Table D.10.6 Micronutrient intake among children . 305 Table D.10.7 Micronutrient intake among mothers . 306 Table D.10.8 Prevalence of anemia in children . 306 Table D.10.9 Prevalence of anemia in women . 307 Table D.10.11 Prevalence of anemia in men. 307 Table D.10.12 Nutritional status of children . 307 Table D.10.13 Nutritional status of women . 308 Table D.11.2 Early childhood mortality rates . 308 Table D.11.4 Perinatal mortality. 308 Table D.13.1 Experience of beatings or physical mistreatment . 309 Table D.13.3 Violence during pregnancy . 309 Table D.13.5 Marital violence . 309 Table D.13.6 Frequency of spousal violence . 310 Table D.14.1 Knowledge of AIDS. 310 Table D.14.2 Knowledge of HIV prevention methods. 310 Table D.14.3 Comprehensive knowledge about AIDS . 311 Table D.14.4 Knowledge of prevention of mother to child transmission of HIV . 311 Table D.14.5 Accepting attitudes toward those living with HIV/AIDS. 312 Table D.14.6 Attitudes toward negotiating safer sexual relations with husband . 312 Table D.14.7 Adult support of education about condom use to prevent AIDS . 313 Table D.14.8 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months . 313 Table D.14.9 Coverage of prior HIV testing . 314 Table D.14.10 Pregnant women counseled and tested for HIV. 314 Table D.14.11 Self-reported prevalence of sexually-transmitted infections (STIs) and STI symptoms . 315 Table D.14.12 Prevalence of injections . 315 Table D.14.13 Comprehensive knowledge about AIDS and of a source of condoms among youth . 316 Table D.14.14 Age at first sexual intercourse among youth. 316 Table D.14.16 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth . 317 Table D.14.17 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months . 317 Table D.14.19 Drunkenness during sexual intercourse among youth. 318 Table D.15.4 HIV prevalence. 318 Table D.15.8 HIV prevalence among young people . 318 Table D.15.12 HIV prevalence among couples. 319 Table D.16.1 Children’s living arrangements and orphanhood . 319 Table D.16.2 Orphans and vulnerable children (OVC) . 320 xviii | Tables and Figures Table D.16.3 School attendance by survivorship of parents and by OVC status . 320 Table D.16.4 Underweight orphans and vulnerable children. 321 Table D.16.6 Succession planning. 321 Table D.16.7 Widows dispossessed of property. 321 Table D.16.8 External support for chronically ill persons . 322 Table D.16.9 External support for orphans and vulnerable children. 322 Foreword | xix FOREWORD In the context of its desire to obtain a database designed to provide reliable indicators to monitor and assess the implementation of the country’s sector programs and policies, the Poverty Reduction Strategy, Vision 2020 and the commitments it has undertaken at the international level, in particular the Millennium Development Goals, the Government of Rwanda has just completed the Third Demographic and Health Survey (EDSR-III 2005). EDSR-III follows the surveys that were successfully conducted in 1992 and 2000, and is part of a broad, worldwide program of socio-demographic and health Surveys conducted in developing countries since the mid-1980’s. In addition to the indicators on fertility, family planning, and maternal and child health which the Survey normally provides, the main innovation of EDSR-III was the integration of a survey module on the seroprevalence of HIV and anemia as well as a module on domestic violence. As such, for the first time, the survey allowed us to determine the prevalence of HIV at the national level. Using this report, the reader will be better able to delineate the socio-demographic challenges the country faces and that it will have to meet, in particular: a maternal and infant mortality rate which remains high despite being in decline, poor utilization of childbirth and post-natal services, a continually high fertility rate, which places pressure on social costs and slows the pace of development, poor utilization of modern contraceptive methods, as well as an alarming nutritional status, above all among children under five years of age and their mothers. The reader could also be alerted to the fact that certain population groups are particularly impacted by a high prevalence of anemia or HIV. Most of these indicators can be improved by increased awareness and heightened responsibility within a couple or among individuals. Without this, the State’s investments would have limited impact. This Survey also draws attention to indicators of an appreciable level that will require strengthening of sustained efforts to maintain, if not to improve, trends. This is particularly the case with regard to the high level of breastfeeding, prenatal visits, vaccination rates of children under five years of age (except for the city of Kigali), and the use of iodized salt. The results of EDSR-III 2005 are thus extremely important because they allow us to assess the progress made in meeting the challenges mentioned above. The results also make it possible to readjust intermediate objectives, identify areas requiring priority attention, and even make projections of future socio-demographic development. The same results represent a daunting challenge to entities providing development financing and call for integrated financing approaches involving multiple sectors of socio- economic life. Accordingly, the Government of Rwanda and in particular the Ministry of Finance and Economic Planning is pleased to provide reliable results to policymakers, planners, and other users in both the public and private sector, based on the current context of the country. May this document be a source of valuable and useful information to all those individuals and organizations active in development who will use it to contribute to an improved quality of life for Rwanda’s population. Signed in Kigali on May 12, 2006 Monique Nsanzabaganwa Minister of State in Charge of Economic Planning at the Ministry of Finance and Economic Planning Acknowledgements | xxi ACKNOWLEDGMENTS This report would not have materialized without the participation of a large number of individuals and organizations. We would like to express our profound thanks to them. First, we extend our thanks to the men and women who generously agreed to respond to all of the questions submitted to them. There was a high response rate both from men (99.2%) and women (98.1%). We would like to express our sincere appreciation to the various Ministries for facilitating the implementation of the Survey. We offer our profound gratitude to the Ministry of Health for its cooperation during the preparation and completion of the survey. We also offer our sincere thanks to the Ministry of Local Government, Good Governance, Community Development and Social Affairs as well as to all of the provincial and district authorities for their assistance and their contribution to the smooth implementation of the Survey. Certainly, without the ongoing support of these various authorities, EDSR- III 2005 could not have been achieved. We also express our gratitude to the International Organizations for their vital financial assistance. Financial contributions from the United States Agency for International Development (USAID/Rwanda), the World Bank through the Support for the Multisectoral AIDS Project (MAP) and through the National AIDS Control Commission (CNLS), the Department For International Development (DFID), the United Nations Children’s Fund (UNICEF), the United Nations Population Fund (UNFPA), and the German Technical Cooperation enterprises (GTZ) to the EDSR-III budget were of immense significance to the effective accomplishment of the survey. We hereby express our profound gratitude to the team from ORC Macro, in particular Mr. Mohamed Ayad, responsible for drafting the project and technical coordination, Mrs. Fern Greenwell, ORC Macro Technical Advisor to EDSR-III 2005, Mr. Noah Bartlett, technical advisor for drafting the reports, and the other ORC Macro officers who contributed to the success of EDSR-III 2005 for their much appreciated technical assistance. The high quality of the analyses presented in this report is evidence of their support. We deeply appreciate the specific technical support of the CNLS, the Treatment and Research Aids Center (TRAC), and the National Reference Laboratory (LNR). Their active participation throughout the conduct of the survey demonstrated the effectiveness of the excellent collaboration between the country’s various institutions. The Third Demographic and Health Survey would not have been accomplished without the unfailing participation of the officers from the National Institute of Statistics who were relentlessly involved, in particular Mr. Philippe Gafishi Ngango, National Director of EDSR-III 2005, Mrs. Apolline Mukanyonga, Technical Director, and Mrs. Athanasie Kabagwira, Associate Technical Director, who, in cooperation with supervisors and administrative support personnel, supplied pertinent technical supervision and contributed to the analysis of the results. We warmly congratulate the cartographers, team leaders, monitors, and the men and women who conducted the surveys, as well as the drivers who were able to overcome the challenges and fatigue inherent in this type of operation. xxii | Acknowledgements We wish to reiterate our sincere thanks to all those, far and wide, who contributed to the completion of this Survey. Lastly, we offer our profound appreciation to the men and women who will use this document, as they have understood the ultimate aim of the production of this valuable report. Dr. Ir. Louis Munyakazi Managing Director of the National Institute of Statistics of Rwanda Abbreviations | xxiii ABBREVIATIONS AD Age at death AIDS Acquired Immunodeficiency Syndrome ANC Antenatal Care AQ Amodiaquine ARI Acute Respiratory Infection ASFR Age-specific Fertility Rate BCG Bacillus of Calmette and Guérin (vaccine against tuberculosis) BMI Body Mass Index CBR Crude Birth Rate CDC Centers for Disease Control and Prevention CNLS Commission Nationale de Lutte contre le Sida CSPro Census and Survey Processing CTS Conflict Tactics Scale DFID Department For International Development DHS Demographic and Health Surveys DPT Diphtheria-Pertussis-Tetanus vaccine EA Enumeration area EDSC Cameroon Demographic and Health Survey EDSBF Burkina Faso Demographic and Health Survey ENF Enquête Nationale sur la Fécondité (National Fertility Survey) EPI Expanded Program of Immunization ESD Enquête sociodémographique (Sociodemographic Survey) FP Family Planning FRw Rwandan Franc GAR Gross Attendance Ratio GDP Gross Domestic Product GFR General Fertility Rate GPI Gender Parity Index GTZ German Technical Cooperation HIV Human Immunodeficiency Virus IEC Information/Education/Communication INSR Institut National de la Statistique du Rwanda IPT Intermittent Preventive Treatment ITN Insecticide-Treated Mosquito Net IUD Intra Uterine Device xxiv | Abbreviations LAM Lactational Amenorrhea Method LNR National Reference Laboratory MAP Multi-country AIDS Program MDG Millennium Development Goals MMR Maternal Mortality Ratio NAR Net Attendance Ratio NCHS National Center for Health Statistics ORS Oral Rehydration Salts ORT Oral Rehydration Therapy OVC Orphaned and Vulnerable Children PNILP Programme National Intégré de Lutte contre le Paludisme (National Malaria Control Program) PRSP Poverty Reduction Strategy Papers PSU Primary Sampling Units RBM Roll Back Malaria RDHS-I First Rwanda Demographic and Health Survey, 1992 RDHS-II Second Rwanda Demographic and Health Survey, 2000 RDHS-III Third Rwanda Demographic and Health Survey, 2005 RGPH Recensement Général de la Population et de l’Habitat (General Population and Housing Census) RHF Recommended Home Fluids SDM Standard Days Method SP Sulfadoxine-Pyrimethamine STI Sexually Transmitted Infection TFR Total Fertility Rate TRAC Treatment and Research AIDS Center TWFR Total Wanted Fertility Rate UNFPA United Nations Population Fund UNDP United Nations Development Programme UNICEF United Nations Children’s Fund USAID United States Agency for International Development USD United States Dollars VCT Voluntary Counseling and Testing Center VIP Ventilation-Improved Pit Latrine WHO World Health Organization YSD Years since death Summary of Findings | xxv SUMMARY OF FINDINGS A total of 10,644 households were selected in the sample for the third Demographic and Health Survey in Rwanda (RDHS-III), and 10,307 of these were contacted at the time of the survey. The survey teams were able to interview individuals in 10,272 households, for a response rate of nearly 100 per- cent. In the 10,272 households surveyed, 11,539 women age 15 to 49 were considered eligible for the individual interview and 11,321 were successfully interviewed. Thus the response rate for women was 98 percent. The male survey was conducted in one out of every two households. A total of 4,959 men age 15 to 59 were identified in the subsample of households. Of the 4,959 men eligible for the indi- vidual interview, 4,820 were successfully inter- viewed, for a response rate of 97 percent among men. The survey results show that 44 percent of the women interviewed were age 15 to 24 at the time of the survey and 43 percent of men were in that age group. Two out of every five women and about one out of two men had never been married. These data indicate that Rwanda’s population is generally young, a fact that needs to be taken into considera- tion by policymakers when designing national de- velopment programs. The proportion of women with no formal education (23 percent) is higher than that of men (17 percent). Only 10 percent of women and 12 percent of men have at least a secondary level of education. The proportion of men and women who do not know how to read is 22 percent and 29 per- cent, respectively. At the national level, more than two out of five women (44 percent) and about one out of five men (19 percent) do not have access to any mass media. Only 8 percent of women and 10 percent of men read a newspaper at least once a week. Very few Rwandan households have electric- ity (5 percent). In rural areas less than 2 percent of households have electricity, compared with 25 per- cent in urban areas. Thirty-nine percent of urban households and 71 percent of rural households do not have reliably clean, potable water (tap water, boreholes, or protected wells). Two out of three households (67 percent) use uncovered latrines. A total of 5 percent have no toilet facilities. FERTILITY Analysis of the 2005 RDHS-III data indicates that the fertility rate of Rwandan women remains high. The total fertility rate (TFR) is 6.1 children for all women, 4.9 for urban women, and 6.3 for rural women. The main background variables for which dif- ferentials in fertility rates can be seen are urban-rural residence, marital status, education, and household wealth. Among the provinces, North and West prov- inces have the highest fertility rates while South province has the lowest. Fertility among adolescent women is negligi- ble, accounting for only 3 percent of national fertil- ity. Women older than age 40 account for 12 percent of fertility. The mean number of children ever born (CEB) among all women age 40 to 49 is 6.6 children per woman. Among urban residents in this age- group, the mean number of children ever born is 5.8, compared with 6.7 for rural residents. Regarding trends in fertility, the youngest and oldest age groups (15-19 and 40-49 years) show a decline from one survey to the next. Women age 20 to 39 account for the largest increase in fertility. A comparison of TFRs across past and current surveys indicates that the fertility rate in Rwanda stabilized at about 6 children per woman in 1992. FAMILY PLANNING Knowledge of Contraception. While almost all married women are aware of contraception, mod- ern methods in particular (98 percent in 2005, com- pared with 97 percent in 2000), relatively few women use them. xxvi | Summary of Findings Knowledge of contraception among men is also almost universal: 98 percent of male respon- dents reported that they knew of at least one modern contraceptive method and 77 percent said they knew of traditional methods. Contraceptive Prevalence. Contraceptive prevalence among currently married women is 17 percent, with 10 percent using modern methods. However, the proportion of married women using contraception has increased in the five years since the last DHS survey in Rwanda (RDHS-II), rising from 13 percent in 2000 to 17 percent in 2005 for all meth- ods and from 4 percent in 2000 to 10 percent in 2005 for modern methods. The modern methods most often used are injectables (5 percent) and pills (2 percent). The survey results show that contraceptive use is lowest among the youngest and oldest age groups: 7 percent for women age 15-24 and 10 percent for women age 45-49. MARRIAGE Forty-nine percent of women age 15-49 were in a union at the time of the survey. The proportion of never-married women decreases as age increases and it is rare to find a woman over 45 who has never been married (2 percent). Therefore, marriage, which remains practically the sole context of pro- creation in Rwanda, is very common. Twelve per- cent of Rwandan women live in polygamous house- holds. Rwandan women tend to marry late: only 19 percent of those age 25 to 49 married before age 18. The median age of first union for women is 20.7 years; the median age of first sex is 20.3 years. Men tend to marry at an older age than women. The median age for the first marriage for men is 25.0 years; the median age of first sex is 20.8 years. FERTILITY PREFERENCES Regarding fertility preferences, 42 percent of women reported that they did not wish to have any more children, while over half (52 percent) wanted a(nother) child. Among the latter group, 12 percent wanted to have the next child within two years, 39 percent wanted a child after two years, and 2 percent wished for another child without specifying the timeframe. The percentage of men (44 percent) who do not want any more children is similar to that of women. Forty percent said they wanted to wait two or more years for another child. The average ideal family size for all women, as well as for married women, is about 4 children. This ideal family size is less than the TFR of 6.1, a finding that partially explains the high percentage of women who do not want to have more children. MATERNAL AND CHILD HEALTH Antenatal Care. The vast majority of expec- tant mothers receive some antenatal care (94 per- cent). However, only 13 percent go for at least four antenatal care (ANC) visits, as recommended by the World Health Organization (WHO) and the Rwan- dan government. The first ANC visit tends to be late in the pregnancy: a median of 6.4 months into the pregnancy. During the ANC consultations, women were rarely informed of the signs of complications that may occur during pregnancy (6 percent). Most often, women were weighed (94 percent) and had their blood pressure taken (71 percent). About half the women (56 percent) said their height was measured. However, routine blood and urine tests were rare. A small percentage of women took iron supplements (28 percent) or antimalaria medication (6 percent). Delivery Care. A large proportion of Rwan- dan women give birth at home (70 percent). Six out of ten were not assisted by trained health providers; 43 percent were assisted by untrained traditional birth attendant, while 17 percent reported giving birth without any assistance. Vaccination Coverage. The objective of Rwanda’s Expanded Program on Immunization—to vaccinate all children within their first 12 months of life—has not yet been met. Only 75 percent of chil- dren age 12-23 months have received all the recom- mended vaccinations. Among these children, only 69 percent received all vaccinations before the age of one year. The dropout rate between the first and third doses of DPT was 10 percent, while for polio it was 13 percent. Childhood Illness. During the two weeks pre- ceding the survey, 17 percent of children under five had acute respiratory infection (ARI); 26 percent had fever, and 14 percent had diarrhea. Summary of Findings | xxvii Medical treatment or advice was sought for 27 percent of the children with ARI or a fever. For those who had diarrhea, only 14 percent received medical treatment. The great majority of mothers (87 percent) know about oral rehydration salts (ORS) for treat- ment for diarrhea. However, during the last episode of diarrhea, only 32 percent of children received either ORS, recommended home fluids, or increased fluids. A similar proportion of children were treated with traditional remedies. Thus, 33 percent of chil- dren with diarrhea received no treatment at all. NUTRITION Breastfeeding Practices. In Rwanda, breast- feeding is nearly universal and is of relatively long duration. Virtually all children under six months are breastfed and at 10-11 months, 97 percent are still being breastfed. The recommendation of exclusive breastfeeding for children up to six months of age is followed by nine out of ten mothers (88 percent). The median duration of breastfeeding is 24.9 months. It is very unusual to see other liquids or com- plementary food being introduced before the age of two months (5 percent). However, the recommended introduction of solid foods at six months is not gen- erally followed: only 69 percent of children age 6-9 months were receiving complementary foods Nutritional Status. More than four out of ten children under age five (45 percent) have chronic malnutrition (stunting) and nearly one out of five (19 percent) has the most severe form. Levels of stunting increase rapidly with age; the highest proportion is found among children age 12-23 months (55 per- cent), but remains fairly high (51 to 53 percent) among older children. The rate of stunting is highest in the North province (52 percent). Stunting tends to be lower among children of mothers with more edu- cation: 50 percent among those with no education, 44 percent among those with primary education, and 43 percent among those of at least secondary level. An estimated 4 percent of children have acute malnutrition (wasting), and 1 percent have the most severe form. The highest prevalence of wasting (9 percent) is found among children age 12-23 months. This corresponds to the period during which the child is most likely to be weaned and is, there- fore, vulnerable to illnesses (such as those linked to the introduction of foods that may have become con- taminated, or picked up as the child crawls around exploring the environment). Interestingly, rates of wasting in the City of Kigali (8 percent) are higher than in the other areas surveyed. Findings show that 22 percent of children in Rwanda are underweight and 4 percent are severely underweight. These figures indicate either chronic or acute malnutrition. Overall, the survey indicates that 56 percent of children age 6-59 months are anemic: 20 percent are mildly anemic, 27 percent are moderately anemic, and 9 percent are severely anemic. In general, chil- dren in urban and rural areas have similar levels of anemia, although the prevalence of severe anemia is higher in urban areas than in rural areas (13 percent versus 8 percent). Children in the City of Kigali experience higher levels of anemiaparticularly the severest form than other children. Women in Rwanda are less afflicted with anemia than the children. Nationally, 33 percent of women suffer from anemia: 19 percent are mildly anemic, 11 percent are moderately anemic, and 3 percent are severely anemic. As with children, ane- mia rates are similar in urban or rural areas; how- ever, women of the City of Kigali have a higher prevalence of moderate and severe anemia than women elsewhere. Vitamin Supplements. The survey results showed that 84 percent of last-born children age 0-3 years had received vitamin A supplements. How- ever, only 33 percent of mothers received vitamin A within the two months following delivery. Also, 71 percent of women did not receive any iron sup- plements during their pregnancy and 24 percent re- ceived supplements for no more than 3 months. Nearly nine out of ten women and children live in households that use adequately iodized salt (15+ ppm). MALARIA Possession of Mosquito Nets. In Rwanda, 18 percent of households own at least one mosquito net. Urban residents, especially in the City of Kigali, xxviii | Summary of Findings show a higher rate (40 percent) of households with at least one net than do rural residents. The percent- age is highest among the wealthiest households (45 percent versus 6 percent among the poorest). However, only 6 percent of the total of households own more than one mosquito net. Overall, almost all households with at least one mosquito net have a net that has been treated at some time. However, there is a difference between those with at least one net and those with an insecti- cide-treated mosquito net (ITN) (18 percent versus 15 percent). The same differential is observed among the households with more than one net (6 percent) and those with more than one ITN (4 percent). Mosquito Net Usage: Only 16 percent of children under the age of five slept under a mosquito net the night before the survey. Among pregnant women, 20 percent slept under a net the night before the survey. INFANT AND CHILD MORTALITY Childhood mortality remains high in Rwanda. Data for the five-year period preceding the survey indicate that for every 1,000 live births, 86 children die before their first birthday (37 between birth and 1 month and 49 between 1 and 12 months). For every 1,000 children who survive to age one, 72 do not reach their fifth birthday. Overall, the mortality risk between birth and exact age five is 152 per 1,000 live births. The RDHS-III results indicate a significant decline in infant and child mortality since the 2000 RDHS-II. However, comparison with the RDHS-I shows that the 2005 infant and under-five mortality rates have actually just returned to the 1992 levels. MATERNAL MORTALITY Maternal mortality continues to be high in Rwanda. According to the RDHS-III, the rate of ma- ternal mortality is about 750 deaths for every 100,000 live births. However, this level of mortality shows a considerable decline since the 2000 RDHS-II, which indicated a maternal mortality rate of 1,071 between 1995 and 1999. DOMESTIC VIOLENCE About one-third of women interviewed (31 percent) reported that they had been victims of physical violence at least once since they were age 15, and 19 percent reported experiencing violence during the past 12 months. Most often, it was the husband or partner who was responsible for the vio- lence. Whether physical or sexual, the violence resulted in serious consequences for the woman: in 22 percent of cases in the past 12 months, the women suffered bruises or wounds, and in 14 per- cent of cases, they experienced bone fractures. In 7 percent of cases, the women had to be treated by a doctor or were treated at a health care facility. STI AND HIV/AIDS-RELATED KNOWL- EDGE, ATTITUDES AND BEHAVIORS Almost all respondents reported that had heard of HIV/AIDS, but only 54 percent of women and 58 percent of men had a comprehensive knowledge of the disease. The level of knowledge about ways to avoid contracting HIV is insufficient: only 73 percent of women and 80 percent of men know it is possible to reduce the risk of getting the AIDS virus by using condoms and by limiting sex to one faithful and un- infected partner. Only 51 percent of men and 46 percent of women expressed positive attitudes towards people living with HIV/AIDS, indicating that the level of stigmatization and discrimination remains high in Rwanda. The survey shows that 8 percent of women and 14 percent of men reported having had higher- risk sex (intercourse with a nonmarital, noncohabit- ing partner). However, only 20 percent of these women and 41 percent of these men had used con- doms during the last higher-risk sex. Among pregnant women, only 22 percent reported receiving counseling on HIV/AIDS during their antenatal care visits or having been tested for HIV and received the results. Summary of Findings | xxix Among youth age 15-24, 51 percent of women and 54 percent of men had comprehensive knowl- edge of HIV/AIDS, and 12 percent of men and 7 percent of women used a condom during their first sexual intercourse. HIV PREVALENCE HIV Testing Rates. Overall, 97 percent of eligible respondents provided blood for HIV testing. The coverage rate was 94 percent in urban areas and 97 percent in rural areas. HIV Prevalence Rates. Survey results indi- cate that 3 percent of adults age 15-49 are infected with HIV. The prevalence rate is higher among women than among men; the ratio of women to men is 1.6. HIV prevalence is significantly higher in urban areas than in rural areas. Among all those age 15-49, the City of Kigali shows the highest level of HIV prevalence (6.7 percent). Among those age 15-24, HIV prevalence in Kigali is 3.4 percent. The lowest HIV prevalence is in the North province (2 percent). According to age and sex, the prevalence of HIV is highest among men age 40 to 44 (7.1 per- cent) and among women age 35 to 39 (6.9 percent). HIV and Associated Factors. HIV preva- lence is very high among respondents who declared having contracted a sexually transmitted infection in the 12 months prior to the survey (15.7 percent). Prevalence is also high among widowed women (15.9 percent) and divorced or separated women (10.9 percent). Fifty-six percent of men and 64 percent of women who tested positive for HIV at the time of the survey had never had an HIV test previously. CARE AND SUPPORT FOR VULNERABLE PERSONS Approximately one out of five children under the age of 18 is an orphan: 4 percent have lost both parents, 13 percent have lost their father, and 3 per- cent have lost their mother. Around 11 percent of children in Rwanda are considered to be vulnerable. Overall, 29 percent of children under age 18 can be classified as orphans or vulnerable children (OVC). The highest proportion of OVC is in the City of Kigali (35 percent) and the lowest is in the North province (25 percent). The RDHS-III results have shown that paren- tal survival status influences the school attendance of children age 10-14. When both parents are alive and the child lives with at least one parent, 91 per- cent attend school. School attendance drops to 75 percent when both parents are dead. In Rwanda, OVC do not seem to suffer more from malnutrition than other children, regardless of age or sex. A ratio of less than 1.0 (0.92) indicates that non-OVC are slightly more likely to be under- nourished than OVC. Early sexual relations seem to be slightly more frequent among OVC (6 percent among girls and 15 percent among boys) than among other children (5 percent among girls and 14 percent among boys). Very few Rwandan households have received assistance to care for sick family members. In only 12 percent of cases did households receive any assis- tance, whether medical, social, material or emo- tional. Less than 1 percent of households received all of these forms of assistance. In 87 percent of cases, households in Rwanda received no external support in caring for OVC. The external assistance that is provided tends to be for paying school fees (9 percent of households). Other types of support are virtually nonexistent. Millennium Development Goal Indicators | xxxi Millennium Development Goal Indicators, Rwanda 2005 Goal Indicator Value 1. Eradicate extreme poverty and hunger Prevalence of underweight children under five years of age Male: 22.9 % Female: 22.1 % Total: 22.5 % Net enrolment ratio in primary education1 Male: 73.8 % Female: 76.6 % Total: 75.2 % Percent of pupils starting grade 1 who reach grade 51 Male: 9.6 % Female: 10.3 % Total: 10.0 % 2. Achieve universal primary education Literacy rate of 15-24 year-olds2 Male: 67.8 % Female: 65.2 % Total: 66.0 % Ratio of girls to boys in primary and secondary education Primary: 1.03 Secondary: 0.81 Ratio of literate women to men, 15-24 years old2 0.96 3. Promote gender equality and empower women Share of women in wage employment in the non-agricultural sector3 8.8 % Under-five mortality rate (per 1,000 live births) 152 per 1,000 Infant mortality rate (per 1,000 live births) 86 per 1,000 4. Reduce child mortality Percent of 1 year-old children immunized against measles Male: 84.9 % Female: 86.4 % Total: 85.6 % Maternal mortality ratio (per 100,000 live births) 750 per 100,000 5. Improve maternal health Percent of births attended by skilled health personnel 38.6 % Condom use to overall modern contraceptive use among currently married women age 15-49 9.2 % Condom use at last higher-risk sex (population age 15-24)4 Male: 39.5 % Female: 26.0 % Percentage of population age 15-24 with comprehensive correct knowledge of HIV/AIDS5 Male: 53.6 % Female: 50.9 % Contraceptive prevalence rate (any modern method, currently married women age 15-49) 10.3 % 6. Combat HIV/AIDS, malaria and other diseases Ratio of school attendance of orphans to school attendance of non- orphans aged 10-14 years 0.82 Percent of population using solid fuels6 Urban: 98.3 % Rural: 99.8 % Total: 99.6 % Percent of population with sustainable access to an improved water source7, urban and rural Urban: 55.0 % Rural: 22.4 % Total: 27.4 % 7. Ensure environmental sustainability Percent of population with access to improved sanitation8, urban and rural Urban: 97.2 % Rural: 96.5 % Total: 96.6 % 1 Excludes children with parental status missing. 2 Refers to respondents who attended secondary school or higher and women who can read a whole sentence. 3 Wage employment includes respondents who receive wages in cash or in cash and kind. 4 Higher risk refers to sexual intercourse with a partner who neither was a spouse nor who lived with the respondent; time frame is 12 months preceding the survey. 5 A person is considered to have a comprehensive knowledge about AIDS when they say that use of condoms for every sexual intercourse and having just one uninfected and faithful partner can reduce the chance of getting the AIDS virus, that a healthy-looking person can have the AIDS virus, and when they reject the two most common local misconceptions. The most common misconceptions in Rwanda are that AIDS can be transmitted through mosquito bites and that a person can become infected with the AIDS virus by sharing food with someone who is infected. 6 Charcoal, firewood, or sawdust. 7 Improved water sources are: household connection (piped), public standpipe, borehole, or protected dug well. 8 Improved sanitation technologies are: flush toilet, traditional pit latrine, or ventilated improved pit latrine. xxxii | Map of Rwanda Country Profile and Survey Introduction | 1 COUNTRY PROFILE AND SURVEY INTRODUCTION 1 1.1 COUNTRY PROFILE 1.1.1 Geography The country of Rwanda is situated in central Africa immediately south of the equator between 1°4' and 2°51' south latitude and 28°63' and 30°54' east longitude. Its total area of 26,338 square kilometers is bordered by Uganda to the north, Tanzania to the east, the Democratic Republic of the Congo to the west, and Burundi to the south. Landlocked, Rwanda lies 1,200 kilometers from the Indian Ocean and 2,000 kilometers from the Atlantic Ocean. Rwanda forms part of the highlands of eastern and central Africa, with mountainous relief and an average elevation of 1,700 meters. However, there are three distinct geographical regions. Western and north-central Rwanda is made up of the mountains and foothills of the Congo-Nile Divide, the Virunga volcano range, and the northern highlands. This region is characterized by rugged mountains intercut by steep valleys, with elevations generally exceeding 2,000 meters. The Divide itself rises to 3,000 meters at its highest point but is dwarfed by the volcano range, whose highest peak, Kalisimbi, reaches 4,507 meters. The Congo-Nile Divide slopes westward to Lake Kivu, which lies 1,460 meters above sea level in the Rift Valley trough. In Rwanda’s center, mountainous terrain gives way to the rolling hills that give the country its nickname, “Land of a Thousand Hills.” Here the average elevation varies between 1,500 and 2,000 meters. This area is also referred to as the central plateau. Further east lies a vast region known as the “eastern plateaus,” where the hills level gradually into flat lowlands interspersed with a few hills and lake-filled valleys. The elevation of this region generally falls below 1,500 meters. Due to its elevation, Rwanda enjoys a temperate, sub-equatorial climate with average yearly temperatures of around 18.5°C. The average annual rainfall is 1,250 millimeters and occurs in two rainy seasons of differing lengths, alternating with one long and one short dry season. The climate varies somewhat from region to region, depending on the altitude, the volcano range and northern highlands being generally cooler and wetter, with average temperatures of 16°C, and average rainfall of above 1,300 millimeters. The maximum rainfall is 1,600 millimeters, above the Divide and the volcanic range. The hilly central region receives an average of between 1,000 and 1,300 millimeters of rain per year, while rainfall on the eastern plateau, whose climate is relatively warmer and drier, generally falls below 1,000 millimeters and can be as low as 800 millimeters. Although Rwanda enjoys more or less constant temperatures, the climate is known to vary from year to year, with extreme variations in rainfall sometimes resulting in flooding or, more often, drought. These extremes have a profound impact on agricultural production, which sometimes falls into recession. Rwanda has a dense network of rivers and streams, draining into the Congo River on the western slope of the Congo-Nile Divide, and into the Nile in the rest of the country via the Akagera River, which 2 | Country Profile and Survey Introduction receives all the streams of this watershed. Water resources also include several lakes surrounded by wetlands. Deforestation due primarily to land clearing for agricultural expansion has resulted in mostly anthropic vegetation with only a few small areas of natural forestland (representing 7 percent of the country) remaining on the Congo-Nile Divide and the slopes of the volcanic range. It should be noted that at the time the survey was conducted in 2005, the country was divided into 11 provinces and the City of Kigali, with the provinces being further subdivided into districts, sectors and cells. Since then, the country’s administrative structure and associated terminology have changed: there are now four geographically-based provinces (North, South, East, and West) and the City of Kigali, these being further subdivided into 30 districts, 415 sectors, cells and, finally, villages (Imidugudu). This report is based on the new administrative divisions (four provinces and the City of Kigali) but also includes the former names (11 provinces and the City of Kigali) for purposes of clarity in referring to the sample and to assist readers as yet unfamiliar with the new administrative entities. 1.1.2 Economy Although regular efforts have been made to develop the service sector and stimulate investment in the industrial sector, the Rwandan economy remains dominated by agriculture. According to the 2002 General Population and Housing Census (RGPH), more than 8 out of 10 people are employed in agriculture, including 81 percent of men and 93 percent of women. However, the agricultural sector is facing major problems: a production system dominated by small farming operations of less than one hectare, rudimentary techniques, and a low rate of investment. Agrarian reforms are being gradually introduced to address these problems, in particular through population resettlement and labor quality improvements focusing on specialized training mainly for women. Efforts are also underway to regionalize crops and fully expand the use of farm inputs. Agriculture accounts for the largest share of Rwanda’s Gross Domestic Product (GDP),1 roughly 45 percent in 2003, followed by services at 36 percent and industry at 19 percent at constant 1995 prices. Nevertheless, agricultural production declined by 4 percent in 2003 in relation to 2002, essentially due to poor rainfall during the two growing seasons. As a direct result, production dropped for grains (-3.4 percent), legumes (-1 percent), tubers (-10.7 percent), and bananas (-13.4 percent) over that of 2002. Among the export crops, coffee production alone dropped by 29 percent in relation to 2002. In 2003, industry value added grew by 7 percent, while mining value added declined significantly (-8.6 percent).2 At the same time, services value added increased by 4 percent in 2003 over the previous year. Financial institutions, transport and communications services, and hotels and restaurants were the main contributors to the increase in value added. The per capita GDP at constant 1995 prices was FRw 76,089 in 2003 compared with FRw 77,631 in 2002. The value added of final consumption expenditure dropped by 0.98 percent due to a decrease in private consumption expenditure, which in 2003 fell from FRw 558,293 million to FRw 537,746 million at constant 1995 francs, a decline of 3.78 percent over 2002. Government consumption expenditure increased by 10.6 percent in 2003 in relation to 2002 (Department of Statistics, 2004). 1 Republic of Rwanda, Ministery of Finances and Economic Planning, Department of Statistics: Rwanda Develop- ment Indicators 2004 2 Rwanda Development Indicators 2004 Country Profile and Survey Introduction | 3 Finally, because of the failure of most development strategies based on structural adjustment programs focusing on growth measured in terms of per capita GDP, the overwhelming majority of development partners are recognizing the need to incorporate social factors into development strategies. Therefore, new initiatives are geared toward pro-poor economic growth and poverty reduction to revive the economies of developing nations. Rwanda has also adopted this new orientation. 1.1.3 Population According to the 2002 Rwanda General Population and Housing Census (RGPH), the country’s population numbers 8,128,553 people. Although Rwanda suffered a major loss of human life (more than one million people) in the 1994 genocide, the population remains essentially the same because more than one million former refugees who had been living for years in exile returned at the end of the war and genocide. The population of Rwanda has increased steadily and rapidly from more than 2,000,000 in 1952, to 7,666,000 in 1996, to 8,128,553 in 2002. The increase is essentially due to rapid demographic growth. The 2002 RGPH estimated the natural growth rate at 2.6 percent and the fertility rate at 5.9. The rate of increase declined significantly to 1.2 percent between 1991 and 2002, compared with 3.1 percent between 1978 and 1991. Population density is high across the country and is increasing steadily: 321 inhabitants per square kilometer in 2002, compared with 283 in 1991 and 191 in 1978. The population is essentially young, with 67 percent of all Rwandans under the age of 20. In terms of gender, the 2002 RGPH shows females to be in the majority (52 percent) while males make up 48 percent of the population. The illiteracy rate remains fairly high: 36 percent of Rwandans age 15 years and older do not know how to read or write and only 4 percent of women are able to read. Sixty percent of the total population is considered literate. The education level of Rwandans age 6 years and above is also low. According to the 2002 RGPH, one in three people is completely uneducated (34 percent) and nearly 60 percent of all Rwandans have received no education beyond primary school. Only 5.8 percent have reached the secondary school level, while those receiving education beyond the secondary level make up less than 1 percent of the population. Under Article 33 of Rwanda’s current Constitution (adopted in 2003), “Freedom of thought, opinion, conscience, religion, worship, and the public manifestation thereof is guaranteed by the State in accordance with conditions determined by law.” Although numerous religions are practiced in Rwanda, Christianity is by far the dominant faith, practiced in some form by 93 percent of the resident population, the majority of whom are Catholic. In the 1991 Census, 90 percent of the resident population identified themselves as Christian. Their number has increased at the expense of those professing no religion, who have declined from 6.8 percent in 1991 to 3.6 percent in 2002. The number of Muslim adherents has risen slightly, from 1.2 percent of the population in the 1991 Census to 1.8 percent in 2002. Nearly all Rwandans speak the same language, Kinyarwanda (spoken by over 99 percent of the population), which is the country’s first official language, followed by French and English. Kiswahili, the third relatively common foreign language, is generally spoken in urban areas and in the provinces bordering on countries where this language is widely spoken (Democratic Republic of the Congo, Tanzania). 4 | Country Profile and Survey Introduction 1.1.4 Population Policy Out of concern for improving the country’s quality of life, the Rwandan government has developed various strategies over the years to ensure an acceptable balance between demographic growth and available resources, particularly since the 1980s. A family planning initiative developed in 1982 provided for training, improved access to family planning services and, in particular, the promotion of family planning through trained communicators known as Abakangurambaga (“Awakeners of the People”). A subsequent policy was adopted in 1990 aimed at curbing demographic growth and reducing fertility through family planning. To create an environment favorable to behavioral changes that result in lower fertility rates, other elements were included in the plan such as increased production, public health improvements, land use planning, training of communicators, the promotion of education and school attendance, and the employment and advancement of women. Following the 1994 genocide, population problems were seen in a new light with respect to both quality of life and population growth. A new national population policy was developed and issued to all development agents in 2003. This policy emphasizes quality of life by providing objectives and strategies used to affect both demographic (fertility, mortality) and socioeconomic factors. Concretely, it emphasizes: slowing demographic growth, managing natural resources sustainability, food safety, access to primary and secondary education for all children—with a focus on technical and vocational instruction and information technologies—good governance, equal opportunity, and participation in development by both men and women. 1.1.5 Public Health Policy The Ministry of Health, in collaboration with its partners, has just developed a policy aimed at the entire health sector. Special emphasis is placed on priority reproductive health issues such as making pregnancy safer, children’s health, family planning, sexually transmitted infections (STIs), HIV/AIDS, teenage reproductive health, prevention and response to sexual violence, and social changes aimed at increasing women’s decisionmaking power. Health indicators have shown clear improvement: the proportion of the population covered by health mutual schemes increased from 4 percent to 7 percent in one year, and the number of doctors and nurses rose by 10 percent and 7 percent, respectively. In addition, the Ministry of Health is developing incentives to encourage highly qualified medical personnel to serve in rural areas. HIV/AIDS is a major problem in Rwanda; for this reason, HIV/AIDS testing was included in the survey. HIV/AIDS affects all population strata, especially young women, sex workers, orphans, prisoners and truck drivers. The price of antiretroviral drugs continues to decline, and the prevention of mother-to- child transmission ((PMTCT) program launched in 2001 has been implemented in all provinces. Government budget allocations for health have increased substantially—by 185 percent between 2002 and 2004 (Finance Law of 2002 and 2004, Government of Rwanda). In 2004, the government allocated 6.1 percent of its budget to health (Department of Statistics, 2004). 1.2 OBJECTIVES AND METHODOLOGY OF THE SURVEY The Rwanda Demographic and Health Survey (RDHS-III, 2005) is the third of its kind, following surveys conducted in 1992 and 2000. Ordered by the Ministry of Finances and Economic Planning, it was carried out by the Department of Statistics (now known as the National Institute of Statistics of Rwanda) Country Profile and Survey Introduction | 5 with the technical assistance of ORC Macro, an American company that supervises the international Demographic and Health Surveys program through the MEASURE DHS project. Financial support for the survey was provided by the United States Agency for International Development (USAID/Rwanda), the United Nations Population Fund (UNFPA), the United Nations Children’s Fund (UNICEF), the Commission Nationale de Lutte contre le SIDA (CNLS) through the World Bank’s Multi-country AIDS Program (MAP), the British Department for International Development (DFID), and the German Technical Cooperation (GTZ). It was conducted on a representative sample of women between the ages of 15 and 49 and men between the ages of 15 and 59. 1.2.1 Objectives of the Survey The main objectives of the RDHS-III were: • At the national level, gather data to determine demographic rates, particularly fertility and infant and child mortality rates, and analyze the direct and indirect factors that determine fertility and child mortality rates and trends. • Evaluate the level of knowledge and use of contraceptives among women and men. • Gather data concerning family health: vaccinations; prevalence and treatment of diarrhea, acute respiratory infections (ARI), and fever in children under the age of five; antenatal care visits; and assistance during childbirth. • Gather data concerning the prevention and treatment of malaria, particularly the possession and use of mosquito nets, and the prevention of malaria in pregnant women. • Gather data concerning child feeding practices, including breastfeeding and, in half the households surveyed, collect anthropometric measurements to evaluate the nutritional status of women and children, and test for anemia in children under the age of five, women between the ages of 15 and 49, and men between the ages of 15 and 59. • Gather data concerning knowledge and attitudes of women and men about STIs and AIDS, and evaluate recent changes in behavior with respect to the use of condoms. • Gather data to determine adult mortality levels at the national level. • Gather quality data concerning domestic violence. • Gather data concerning the types of care and support received by those under the age of 60 who died in the 12 months preceding the survey. • Collect blood samples in half of the households surveyed to estimate the prevalence of HIV in the adult population of reproductive age—anonymous HIV testing of women age 15 to 49 and men age 15 to 59. 1.2.2 Questionnaires Three questionnaires were used in the RDHS-III: the Household Questionnaire, the Women’s Questionnaire, and the Men’s Questionnaire. The content of these questionnaires was based on model questionnaires developed by the MEASURE DHS project. Technical meetings between experts and representatives of the Rwandan government and national and international organizations were held beginning in June 2004 to discuss the content of the questionnaires. The inputs generated by these meetings were used to modify the model questionnaires to reflect the needs of users and the relevant population, family planning, HIV/AIDS, and other health issues in Rwanda. The final questionnaires were 6 | Country Profile and Survey Introduction then translated from French into English and Kinyarwanda. These questionnaires were further refined and then finalized in December 2004 after pretesting and training of field staff. The Household Questionnaire was used to list all of the usual members and visitors in the selected households. Some basic information was collected on the characteristics of each person listed, including age, sex, education, and relationship to the head of the household. The main purpose of the Household Questionnaire was to identify women and men who were eligible for the individual interview. The Household Questionnaire also collected information on characteristics of the household’s dwelling unit such as the main source of drinking water, type of toilet facilities, materials used for the floor of the house, the main energy source used for cooking, ownership of various durable goods, ownership and use of mosquito nets, and the type of salt used for cooking. In addition, questions were asked about the type of assistance or support received by vulnerable members of the population such as the very ill, and orphaned or otherwise vulnerable children. The questionnaire was also used to register people eligible for anthropometric (height and weight) measurements and the collection of samples for hemoglobin and HIV testing. The Women’s Questionnaire was used to collect information on all women of reproductive age (15-49 years) and covered a wide variety of topics, including: • Background characteristics • Reproductive history • Knowledge and use of contraceptive methods • Fertility preferences • Antenatal, childbirth, and postpartum care • Breastfeeding and child feeding practices • Vaccinations and childhood illnesses • Marriage and sexual activity • Women’s work and husband’s background characteristics • AIDS and other sexually transmitted infections • Adult mortality • Domestic violence The Men’s Questionnaire was administered to all men age 15-59 years living in every second household of the RDHS-III sample. The Men’s Questionnaire collected information similar to that of the Women’s Questionnaire, but was shorter because it did not contain questions on reproductive history, maternal and child health, or nutrition. All aspects of RDHS-III data collection, including anemia and HIV testing procedures, were pre- tested between 19 November and 15 December 2004. The 30 team members recruited received four weeks of training on the questionnaires and procedures for collecting blood samples for the anemia and HIV tests. Training in blood sample collection was provided by the Department of Statistics technical team and a representative of the National Reference Laboratory, with the assistance of ORC Macro. The training included a theory section and a practicum section both in the classroom and at health facilities in the city of Kigali. During the pilot survey, approximately 150 households were visited in urban and semi- urban clusters in the city of Kigali and Kigali Ngari. The blood sample collection acceptance rate was sufficiently high during the pretest (over 85 percent for women and men) to indicate the feasibility of conducting such samples during the survey itself. The lessons learned during this pretest were used to finalize the survey instruments and logistical arrangements. Country Profile and Survey Introduction | 7 1.2.3 Sample Design The sample for the RDHS-III survey covered the population residing in ordinary households across the country. A national sample of 10,644 households was selected. The sample was first stratified to provide adequate representation of urban and rural areas as well as all 12 provinces including the “City of Kigali,” the nation’s capital. Decentralization reforms were introduced after this sample was drawn, resulting in new geographically-based divisions that regroup the former districts into five new provinces. However, the sample used posed no obstacle to adequate representation of the new provinces and the data in this report present key indicators corresponding to the five recently created provinces (South, West, North, East, and the City of Kigali). The survey used a two-stage sample design. The first stage involved selecting primary sampling units (PSUs) based on the list of enumeration areas covered in 2002 General Population and Housing Census (RGPH) prepared by the National Census Bureau. These enumeration areas provided the master frame for the drawing of 462 clusters (351 rural and 111 urban), selected with a representative probability proportional to their size. A strictly proportional sample allocation would have resulted in a very low number of urban households in certain provinces such as Umutara. It was therefore necessary to slightly over-sample urban areas in order to survey a sufficient number of households to produce reliable estimates for urban areas. The second stage involved selecting a sample of households in these enumeration areas. In order to adequately guarantee the accuracy of the indicators, it was necessary to control the total size of the households drawn by setting the number of households to be surveyed at 20 in urban clusters and 24 in rural clusters. Because of the nonproportional distribution of the sample among the different strata and the fact that the number of households was set for each cluster, weighting was used to ensure the validity of the sample at both national and regional levels. All women age 15-49 years who were either usual residents of the selected household or visitors present in the household on the night before the survey were eligible to be interviewed (approximately 11,500 women). In addition, in a subsample of every second household selected for the survey, a sample of 5,000 men age 15-59 years was selected to be interviewed. In this subsample, all men and women eligible for the individual survey were also eligible for the HIV test. In addition, in this subsample of households, all women eligible for the survey and all children under the age of five were eligible for the anemia test. Finally, in this same subsample of households, all women eligible for the survey and all children under the age of five were eligible for height and weight measurements to determine their nutritional status. 1.2.4 Sample Coverage All of the 462 clusters selected for the sample were able to be surveyed for the RDHS-III. A total of 10,644 households were selected, of which 10,307 households were identified and occupied at the time of the survey. Among these households, 10,272 completed the Household Questionnaire, yielding a response rate of nearly 100 percent (Table 1.1). In the 10,272 households surveyed, 11,539 women age 15-49 years were identified as being eligible for the individual interview; interviews were completed with 11,321 of these women, yielding a response rate of 98 percent. Male interviews were conducted in every second household. A total of 4,959 men age 15-59 years were identified in the subsample of households. Of these 4,959 men, 4,820 completed the individual interviews, yielding a response rate of 97 percent. The response rates were slightly higher in rural areas for both men and women. 8 | Country Profile and Survey Introduction Table 1.1 Results of the household and individual interviews Number of households, number of interviews, and response rates, according to residence, Rwanda 2005 Residence Result Urban Rural Total WOMEN Household interviews Households selected 2,220 8,424 10,644 Households occupied 2,122 8,185 10,307 Households interviewed 2,107 8,165 10,272 Household 99.3 99.8, 99.7 Interviews with women Number of eligible women 2,689 8,850 11,539 Number of eligible women interviewed 2,616 8,705 11,321 Eligible woman response rate 97.3 98.4 98.1 MEN Household interviews Households selected 1,110 4,212 5,322 Households occupied 1,061 4,095 5,156 Households interviewed 1,053 4,083 5,136 Household response rate 99.2 99.7 99.6 Interviews with men Number of eligible men 1,183 3,776 4,959 Number of eligible men interviewed 1,130 3,690 4,820 Eligible man response rate 95.5 97.7 97.2 1.2.5 Hemoglobin and HIV Testing In every second household, women age 15-49 years, men age 15-59 years, and children under the age of five were eligible for the anemia test. These men and women were also eligible for the HIV test. The anemia and HIV test protocols were approved by the ORC Macro Internal Review Board in Calverton and the National Ethics Committee of Rwanda. Hemoglobin test Checking hemoglobin levels is the primary way of diagnosing anemia. This test is performed with the HemoCue system. An informed consent form is read to the eligible person or parent/responsible adult of the child or teenager between the ages of 15 and 17 years. This consent form explains the objectives of the test, informs the eligible individual (or parent/responsible adult) that the results will be communicated immediately after the test, and asks permission to conduct the test. Country Profile and Survey Introduction | 9 Before collecting the blood, the finger is cleaned with a swab dipped in alcohol and allowed to air dry. Then the tip of the finger (or heel, for children under 6 months, or under one year if very thin) is pricked with a sterile, single-use retractable blood lancet. One drop of blood was collected in a microcuvette and then introduced into the HemoCue photometer, which indicated the level of hemoglobin. These results were recorded on the Household Questionnaire and communicated to the person tested, or to the parent/responsible adult, with an explanation of their meaning. If the person presented severe anemia (hemoglobin below 7 g/dl, or 9 g/dl for pregnant women), the survey conductor provided a reference explaining how and where to seek treatment at a medical facility. HIV test The HIV test was given in the subsample of households selected for the men’s survey. Blood samples were collected from all eligible men and women who volunteered to be tested in these households. The HIV test protocol is based on the anonymous linked protocol developed by the DHS (Demographic and Health Surveys) program and approved by ORC Macro’s Internal Review Board. According to this protocol, names and other personal or geographic information that might identify an individual may not be linked to the blood sample. The anonymous linked protocol was also approved by the National Ethics Committee of Rwanda specifically for the RDHS-III. Because HIV tests are strictly anonymous, it was not possible and will not be possible to inform those surveyed of their test results. All persons eligible for the survey, whether or not they agreed to be HIV tested, received a card allowing them to obtain, if desired, counseling and free testing at a voluntary counseling and testing center (VCT). The card contained a list of 77 VCTs located throughout the country that offer free services to those who present the card. For the purposes of blood sample collection, two “survey technicians” were included on each field team to be specifically responsible for collecting blood samples. In addition to training in conducting the survey, these technicians received special training covering all aspects of the anemia and HIV test protocols. After explaining blood collection procedures, data confidentiality, and test anonymity, the technician sought to obtain the informed consent of each person eligible for the test. At that time, the eligible person was given a voucher for counseling and free testing at a VCT center. For men and women who consented to be tested, the technician collected drops of blood on a filter paper, observing all safety and hygienic precautions. In most cases, the drops of blood were collected from the same finger prick as for the anemia test. A barcode label was attached to each filter paper containing the blood sample. A duplicate label was attached to the Household Questionnaire on the line showing consent for that respondent and a third copy of the same barcode label was affixed to the Blood Sample Transmittal Form. The drops of blood on the filter paper were dried for a minimum of 24 hours in a drying box containing dessicants to absorb moisture. The next day, each dried sample was placed in a waterproof plastic Ziploc bag with a dessicant and moisture indicator for preservation. This kept the individual bags dry during transmittal from the field to the central office of the National Institute of Statistics in Kigali, where they were immediately verified and placed in a dry place prior to being logged and sent on to the National Reference Laboratory. Testing for the HIV antibody and compilation of results were performed by the National Reference Laboratory (LNR) in Kigali. The LNR undergoes rigorous internal quality audits on a regular basis as well as external quality audits. 1.2.6 Training and Data Collection Staff responsible for the survey at the National Institute of Statistics, in collaboration with the technical team, recruited approximately 95 people to participate in data collection during the main survey, 33 of whom were medically qualified to take blood samples. Four weeks of training were provided, from 10 | Country Profile and Survey Introduction 21 January to 21 February, followed by three days of practicum in urban and rural areas not selected for the main survey. After the training, the field agents were divided into 15 teams, each of which contained a team leader, a supervisor, three female interviewers and one male interviewer. One of the three female interviewers and the male interviewer also served as medical technicians. Data collection began on 28 February 2005 in the City of Kigali. This location made it possible to closely monitor the teams before they were dispatched to more distant areas. After two weeks, all teams, except for those remaining to complete the work in the City of Kigali, were deployed to their respective work zones. Data collection was completed on 13 July 2005. 1.2.7 Data Processing Data entry on personal computers began on 23 March 2005, three weeks after the survey was launched in the field. Data were entered by a team of eight data processing personnel recruited and trained for this task, assisted during these operations by 4 others. Completed questionnaires were periodically brought in from the field to the National Institute of Statistics in Kigali, where assigned agents checked them and coded the open-ended questions. Next, the questionnaires were sent to the data entry facility and the blood samples were sent to the National Reference Laboratory to be screened for HIV. Data were entered using CSPro, a program developed jointly by the United States Census Bureau, the ORC Macro MEASURE DHS+ program, and Serpro S.A. All questionnaires were entered twice to eliminate as many data entry errors as possible. In addition, a quality control program was used to detect some of the main data collection errors for each team. This information was shared with field teams during supervisory visits to improve data quality. The data entry and internal consistency verification phase of the survey was completed in October 2005. Household Characteristics | 11 HOUSEHOLD CHARACTERISTICS 2 This chapter presents information on the social, economic, and demographic characteristics of the households sampled, focusing on such background characteristics as age, sex, school attendance, and the educational attainment of the respondents, as well as the physical features of their dwellings and ownership of durable goods. The purpose of this chapter is to present a profile of the households and socioeconomic conditions in which the men, women, and children targeted by this survey live. Such descriptions are essential because socioeconomic and environmental factors are major determinants of the health status and overall living conditions of a country’s population. 2.1 HOUSEHOLD POPULATION BY AGE AND SEX Table 2.1 shows the distribution by age and sex of the household population surveyed, according to urban-rural residence. The household survey involved 46,490 respondents, of which 39,352, or 85 percent, live in rural areas and 7,139, or 15 percent, live in urban areas. Table 2.1 Household population by age, sex, and residence Percent distribution of the de facto household population by five-year age groups, according to sex and residence, Rwanda 2005 Urban Rural Total Age Male Female Total Male Female Total Male Female Total <5 17.3 15.7 16.4 19.1 16.3 17.6 18.8 16.2 17.4 5-9 14.6 13.7 14.1 17.0 14.4 15.6 16.6 14.3 15.4 10-14 12.9 11.9 12.4 14.2 13.3 13.7 14.0 13.1 13.5 15-19 10.4 12.4 11.5 11.6 10.4 11.0 11.4 10.7 11.0 20-24 11.8 11.5 11.6 8.5 9.3 8.9 9.0 9.6 9.4 25-29 8.6 8.4 8.5 5.9 6.9 6.4 6.3 7.1 6.7 30-34 6.5 6.3 6.4 4.4 5.8 5.2 4.8 5.9 5.4 35-39 4.8 5.2 5.0 3.8 4.5 4.2 4.0 4.6 4.3 40-44 4.1 4.3 4.2 3.9 4.6 4.3 3.9 4.6 4.3 45-49 3.0 2.8 2.9 3.5 3.9 3.7 3.4 3.7 3.6 50-54 1.9 2.7 2.3 2.5 3.2 2.9 2.4 3.1 2.8 55-59 1.3 1.4 1.3 1.6 1.7 1.7 1.5 1.7 1.6 60-64 0.9 1.4 1.2 1.1 1.7 1.4 1.0 1.6 1.3 65-69 0.7 1.0 0.9 0.9 1.5 1.2 0.8 1.4 1.1 70-74 0.4 0.6 0.5 0.9 0.9 0.9 0.8 0.9 0.9 75-79 0.3 0.4 0.4 0.5 0.7 0.6 0.5 0.7 0.6 80 + 0.4 0.4 0.4 0.6 0.8 0.7 0.6 0.7 0.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 3,316 3,822 7,139 18,446 20,906 39,352 21,762 24,727 46,490 Table 2.1 shows the distribution by age and sex of the household population as depicted by the age pyramid in Figure 2.1. The age pyramid is wide at the base, narrowing rapidly as it reaches the upper age limits, an indication of a population with high fertility and even higher mortality. In addition, there is a notable gender imbalance: 88 males for every 100 females in the total population. Further analysis reveals structural elements peculiar to the Rwandan population. First, the number of men drops off significantly in the 30-34 and 45-49 age groups, resulting in a shortage of males. The same trend occurs among females, but to a significantly lesser degree. The shortage of men and women may be attributed to the excess mortality caused by the 1994 genocide. Moreover, there is significant falloff in the 10-14 year 12 | Household Characteristics age group, a direct effect of the low birth rate during the years surrounding 1994. The higher proportion of children age 0-4 years reflects the return of fertility rates to their 1992 levels (6.2 compared with 6.1 in 2005). The overrepresentation of women overall is noted in both urban and rural areas. In rural areas, males predominate among those age 0 to 19 years. From age 20-24 on, however, the situation begins to reverse and the gap narrows. In urban areas, males age 0-4 and 10-14 outnumber females, but beginning at age 35-39, the proportion of females is slightly larger. 2.2 HOUSEHOLD SIZE AND COMPOSITION Table 2.2 shows that the mean size of a Rwandan household is 4.6 persons. This mean size varies somewhat: 4.5 in rural areas and 4.8 in urban areas. It is identical to the mean household size of 4.6 found in the previous survey, with variations of 4.5 in rural areas and 5.0 in urban areas. In addition, the results presented in Table 2.2 show that 66 percent of Rwandan households are headed by men. Female-headed households represent 34 percent of households in rural areas and nearly the same percentage in urban areas (33 percent). The percentage of female-headed households increased significantly from 21 percent to 36 percent between 1992 and 2000, but dropped again in 2005 (34 percent). Approximately half of all households contain 3 to 5 people. One-person households make up only 7 percent of the population. Only one in ten households (10 percent) contains 8 or more people. RDHS 2005 Figure 2.1 Population Pyramid 80 + 75-79 70-74 65-69 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5-9 <5 Age 0246810 0 2 4 6 8 10 Male Female Percent Household Characteristics | 13 Table 2.2 Household composition Percent distribution of households by sex of head of household and household size, according to residence, Rwanda 2005 Residence Characteristic Urban Rural Total Sex of head of household Male 67.2 66.0 66.1 Female 32.8 34.0 33.9 Total 100.0 100.0 100.0 Number of usual members 1 8.8 6.2 6.5 2 10.1 11.3 11.1 3 13.1 17.4 16.7 4 17.7 18.4 18.3 5 14.6 16.1 15.9 6 13.5 12.8 12.9 7 9.4 8.1 8.3 8 5.3 5.3 5.3 9+ 7.6 4.4 4.9 Total 100.0 100.0 100.0 Number of households 1,510 8,762 10,272 Mean size 4.8 4.5 4.6 Note: Table is based on de jure members, i.e., usual residents. 2.3 SCHOOL ATTENDANCE AND EDUCATIONAL ATTAINMENT Tables 2.3.1 and 2.3.2 show the percent distribution of the male and female household population according to highest level of educationa attained, by age, residence, province, and household wealth quintile. Educational attainment is important; it contributes to improved living conditions not only for the individual household but for society as a whole. Reproductive behavior, the use of contraception, health habits, school attendance of household members, and habits relating to hygiene and nutrition are all influenced by educational attainment. The data in these two tables show that 29 percent of women and 22 percent of men have never attended school. The percentage of men and women who have completed primary school is nearly identical (8 percent for men, 7 percent for women). As educational attainment increases, the percentage of both men and women in these categories decreases: only 2 percent of men and 1 percent of women have completed secondary level education; less than 1 percent of men and women have attended any education beyond the secondary level. 14 | Household Characteristics Table 2.3.1 Educational attainment of household population: female Percent distribution of the de facto female household population age six and over by highest level of education attended or completed, according to background characteristics, Rwanda 2005 Background characteristic No education Primary Primary complete 1 Secondary incomplete Secondary complete2 Superior Total Number Age 6-9 35.7 63.3 0.0 0.0 0.0 0.0 100.0 2,746 10-14 6.3 92.2 0.6 0.5 0.0 0.0 100.0 3,232 15-19 9.3 75.5 8.5 6.1 0.3 0.0 100.0 2,647 20-24 17.7 57.4 14.1 6.6 3.5 0.7 100.0 2,382 25-29 18.0 53.1 16.5 6.9 3.5 1.6 100.0 1,759 30-34 24.9 53.8 8.9 8.9 2.4 0.7 100.0 1,464 35-39 36.2 44.6 7.0 9.1 2.2 0.7 100.0 1,141 40-44 42.7 38.9 11.0 5.6 1.5 0.2 100.0 1,136 45-49 48.9 37.4 8.7 3.8 0.6 0.3 100.0 921 50-54 65.9 26.6 3.5 2.8 0.8 0.4 100.0 762 55-59 70.7 22.3 3.1 2.3 0.3 0.0 100.0 417 60-64 77.6 18.6 1.8 0.8 0.7 0.0 100.0 403 65+ 87.4 9.8 0.7 0.1 0.0 0.0 100.0 914 Residence Urban 19.4 52.3 9.4 11.5 4.5 2.2 100.0 3,103 Rural 30.9 59.1 6.2 2.8 0.6 0.0 100.0 16,823 Province Kigali city 17.1 49.4 10.9 13.3 5.2 3.1 100.0 1,683 South 27.3 59.4 7.9 3.6 1.2 0.1 100.0 5,261 West 31.6 58.6 5.7 2.7 0.7 0.2 100.0 5,132 North 30.5 58.9 5.1 4.2 1.1 0.1 100.0 3,782 East 32.0 58.2 6.3 2.7 0.6 0.0 100.0 4,069 Wealth quintile Lowest 36.3 57.6 4.8 1.0 0.1 0.0 100.0 4,243 Second 30.0 61.4 6.6 1.6 0.1 0.0 100.0 3,878 Middle 31.8 59.9 5.5 2.0 0.3 0.0 100.0 3,932 Fourth 29.6 59.3 7.1 2.9 0.4 0.0 100.0 3,958 Highest 17.2 51.9 9.8 13.3 5.4 1.8 100.0 3,916 Total 29.1 58.0 6.7 4.1 1.2 0.4 100.0 19,927 1 Completed 6 grades at the primary level 2 Completed 6 grades at the secondary level A comparison of these proportions to those of the previous survey shows no significant improvement, although at the time of the previous survey, 35 percent of women and 28 percent of men had no education at all, compared with 29 percent and 22 percent, respectively, in the current survey. The percentage of men and women who have completed primary school has declined, from 12 percent to 8 percent for men, and from 10 percent to 7 percent for women. However, when compared with previous generations, the figures show significant gains: the proportion of females with no education at all has dropped from 87 percent for women 65 and over to 6 percent for girls between the ages of 10 and 14. The percentage for males in these age groups has dropped from 57 percent to 7 percent. In addition, the gap in educational attainment between the sexes seems to be narrowing in the younger age groups. The percentage of women having completed primary school is the same or close to that of men for all ages up to age 34: 9 percent of women between the ages of 15 and 19 said they had completed primary school, compared with 7 percent of men. This narrowing of the gap in educational attainment between the sexes is also seen at the secondary level: between the ages of 20 and 24, 3 percent of men and 4 percent of women have completed secondary school. This contrasts with the common situation of previous generations, when the proportion of women between the ages of 45 and 49 who had completed primary school was 9 percent, while that of men was 19 percent. Household Characteristics | 15 Table 2.3.2 Educational attainment of household population: male Percent distribution of the de facto male household population age six and over by highest level of education attended or completed, according to background characteristics, Rwanda 2005 Background characteristic No education Primary Primary complete 1 Secondary incomplete Secondary complete2 Superior Total Number Age 6-9 37.1 62.3 0.0 0.1 0.0 0.0 100.0 2,835 10-14 7.4 91.0 0.7 0.4 0.0 0.0 100.0 3,053 15-19 8.7 76.6 6.7 7.3 0.2 0.1 100.0 2,489 20-24 15.4 56.2 14.8 9.2 3.4 0.4 100.0 1,967 25-29 15.8 48.8 19.4 9.4 4.0 2.3 100.0 1,376 30-34 19.4 50.9 9.2 13.5 3.8 2.9 100.0 1,036 35-39 24.1 48.7 8.0 13.3 3.0 2.2 100.0 861 40-44 31.8 39.8 16.1 7.6 2.7 1.4 100.0 847 45-49 29.4 43.8 18.9 5.0 1.9 1.1 100.0 742 50-54 33.4 44.3 14.4 4.9 1.5 0.7 100.0 525 55-59 32.1 46.9 13.6 3.1 2.1 1.4 100.0 336 60-64 41.2 39.8 11.0 4.9 1.4 0.5 100.0 224 65+ 56.7 36.3 4.0 1.0 0.4 0.2 100.0 595 Residence Urban 15.4 52.0 10.0 13.2 4.6 3.6 100.0 2,660 Rural 22.6 64.3 7.7 4.0 0.9 0.2 100.0 14,231 Province Kigali city 13.6 47.1 11.7 16.0 5.3 4.7 100.0 1,536 South 20.8 65.0 8.1 4.4 1.0 0.4 100.0 4,436 West 21.6 63.7 7.8 4.3 1.5 0.4 100.0 4,185 North 21.8 64.3 6.5 5.2 1.2 0.4 100.0 3,137 East 25.2 62.2 8.0 3.6 0.6 0.1 100.0 3,596 Wealth quintile Lowest 27.6 64.9 4.9 2.2 0.1 0.0 100.0 3,226 Second 24.2 65.5 7.2 2.8 0.2 0.0 100.0 3,054 Middle 21.7 66.6 7.4 3.2 0.5 0.0 100.0 3,519 Fourth 22.4 62.5 8.9 4.9 0.7 0.1 100.0 3,477 Highest 12.6 53.1 11.2 13.2 5.5 3.3 100.0 3,614 Total 21.5 62.3 8.0 5.4 1.5 0.7 100.0 16,890 1 Completed 6 grades at the primary level 2 Completed 6 grades at the secondary level By residence, the data show significant gaps in educational attainment. In rural areas, 23 percent of men and 31 percent of women have no education at all, compared with 15 percent of men and 19 percent of women in urban areas. There are also variations between provinces. The City of Kigali has the lowest percentage of residents with no education (17 percent of women and 14 percent of men). Conversely, the East region has the highest percentage of men and women with no education (25 percent and 32 percent, respectively). As the level of educational attainment increases, the gaps between the provinces widen: in the City of Kigali, 5 percent have completed secondary school, compared with 2 percent, at most, for men and 1 percent for women in the other provinces. Results by wealth quintile show that the proportion of both men and women with no education decreases as the household standard of living increases. Conversely, the proportion of people having attained education at any given level increases with household wealth. The data also show that in households in the highest wealth quintile there is practically no gap in educational attainment between males and females, up to the secondary level. 16 | Household Characteristics The level of school attendance of school-age children is the primary indicator of a population’s access to education and, indirectly, the socioeconomic development of the area in which the population lives. The 2005 RDHS-III asked questions concerning school attendance of all respondents between the ages of 5 and 24. Table 2.4 shows net attendance ratios (NAR) and gross attendance ratios (GAR) by sex, residence, and province, by level of educational attainment. Table 2.4 School attendance ratios Net attendance ratios (NAR) and gross attendance ratios (GAR) for the de jure household population by level of schooling and sex, according to background characteristics, Rwanda 2005 Net attendance ratio1 Gross attendance ratio2 Background characteristic Male Female Total Male Female Total Gender parity index3 PRIMARY SCHOOL Residence Urban 79.4 81.6 80.5 134.3 133.6 133.9 0.99 Rural 72.9 75.7 74.3 132.7 138.1 135.4 1.04 Province Kigali city 80.5 82.0 81.2 134.7 133.3 134.0 0.99 South 73.1 75.6 74.4 129.3 132.9 131.0 1.03 West 74.2 76.2 75.2 138.8 142.3 140.6 1.03 North 75.9 78.9 77.4 125.6 127.6 126.6 1.02 East 69.8 74.1 71.9 136.8 148.6 142.5 1.09 Total 73.8 76.6 75.2 132.9 137.5 135.2 1.03 SECONDARY SCHOOL Residence Urban 11.1 12.3 11.8 20.6 20.8 20.7 1.01 Rural 3.3 2.2 2.7 5.6 3.5 4.5 0.62 Province Kigali city 12.8 14.4 13.7 23.7 26.1 25.0 1.10 South 2.9 3.5 3.2 4.8 5.2 5.0 1.08 West 3.4 2.4 2.9 7.3 4.1 5.6 0.56 North 5.3 3.6 4.4 8.4 4.2 6.2 0.50 East 3.7 2.0 2.8 5.5 4.0 4.8 0.72 Total 4.4 3.8 4.1 7.7 6.2 6.9 0.81 1 The NAR for primary school is the percentage of the primary-school-age (6-11 years) population that is attending primary school. The NAR for secondary school is the percentage of the secondary-school-age (12-18 years) population that is attending secondary school. By definition the NAR cannot exceed 100 percent. 2 The GAR for primary school is the total number of primary school students, expressed as a percentage of the official primary-school-age population. The GAR for secondary school is the total number of secondary school students, expressed as a percentage of the official secondary-school-age population. If there are significant numbers of over-age and underage students at a given level of schooling, the GAR can exceed 100 percent. 3 The Gender Parity Index for primary school is the ratio of the primary school GAR for females to the GAR for males. The Gender Parity Index for secondary school is the ratio of the secondary school GAR for females to the GAR for males. Net school attendance ratios (NAR) measure school attendance in children who have reached the official school age. For primary school, the NAR is the percentage of the primary-school-age population (age 7-12 in Rwanda) that is actually attending primary school. This table shows that the primary level NAR is 75 percent for Rwanda, which means that three-quarters of the population between the ages of 7 and 12 are attending primary school. The ratio is higher for urban areas than for rural areas (81 percent compared with 74 percent). In the provinces, the ratio ranges from a high of 81 percent in the City of Household Characteristics | 17 Kigali to a low of 72 percent in the East province. The NAR is also higher for women (77 percent) than for men (74 percent), regardless of residence and province. At the secondary level, which concerns the population between the ages of 13 and 19, the NAR is much lower (4 percent), which means that only 4 percent of the official secondary-school-age population are actually attending school. There is practically no gap between the sexes. However, it is much higher in urban areas than in rural areas (12 percent compared with 3 percent), which may explain the major gap between the City of Kigali, with an NAR of 14 percent, and the other provinces, whose NARs are between 3 percent and 4 percent. Table 2.4 also shows gross school attendance ratios (GAR). Unlike the NAR, the GAR measures school attendance in young people regardless of age. The GAR for primary school is the total number of students of any age attending primary school, expressed as a percentage of the official primary-school-age population, which is 7 to 12 years in Rwanda. Unless there are significant numbers of over-age and under-age students at a given level of schooling, the GAR is always higher than the NAR and can, in some cases, exceed 100 percent. In Rwanda, the GAR is 135 percent, which means that a significant proportion of children who do not fall into the official primary-school-age category are attending school at the primary level. These are likely to be children over the age of 12 who are still attending primary school; in fact, a program exists to reintegrate children who dropped out of primary school for any reason. In addition, the GAR is higher for girls than for boys (138 percent for girls compared with 133 percent for boys). Moreover, there is practically no difference by residence. At the secondary level, the NAR is very low. Only 4 percent of all children of official secondary school age are actually attending school. The ratio is nearly the same for girls and boys. However, it is higher in urban areas than in rural areas (12 percent compared with 3 percent). The GAR is also very low (7 percent), either because official secondary-school-age children are still in primary school, or because they have dropped out of secondary school or have never attended at all. Students who do not pass the national exam at the end of primary school are not allowed to attend state or state-subsidized secondary schools, which are less expensive than private schools. In addition, the GAR for boys is very similar to that of girls, but there is a pronounced difference by residence (21 percent for urban areas compared with 5 percent for rural areas). At 25 percent, the GAR for the City of Kigali stands out from the other provinces, where the GAR is very low (a maximum of 6 percent in the North province). The table includes a third school attendance indicator: the gender parity index (GPI), which is the ratio of the GAR for females to the GAR for males. The narrower the gap between the sexes, the closer the index is to 1. Table 2.4 shows a GPI for primary school of just above 1, which indicates an absence of disparity between the sexes. Curiously, only urban areas, in particular the City of Kigali, show a GPI of slightly below 1. The GPI for secondary school is below one (0.81); this indicates that girls are educationally disadvantaged at this level. The inequality is more pronounced in rural areas, which have a GPI of only 0.62, and in the West (0.56) and North (0.50) provinces. The City of Kigali has the highest GPI (1.10). Figure 2.2 shows that the rate of school attendance, which is low at ages 5 and 6, begins to increase at age 7, the official age for entering primary school. It reaches high levels between the ages of 8 and 13. This period corresponds to the primary school years for children in the normal primary cycle. After age 13, the curve declines steadily, reaching its lowest point at the age of 20. After the age of 13, school attendance rates approaching or exceeding 50 percent do not indicate high school attendance at the secondary level but, rather, that a majority of children are beginning primary school late. 18 | Household Characteristics It should also be noted that the proportion of women is higher between the ages of 6 and 12 everywhere, while the situation reverses itself after this up to age 23, although, paradoxically, the balance is restored at age 24. 2.4 LIVING CONDITIONS The household survey gathered information on certain housing characteristics (access to electricity, drinking water source, type of toilet facilities, roofing and flooring materials). Information was also sought concerning ownership of various modern durable goods (radio, television, refrigerator, bicycle, motorcycle/scooter, car/truck). These characteristics are used to evaluate the socioeconomic conditions of the household. Table 2.5 shows that, at the national level, very few households have access to electricity (5 percent). The situation has not changed much compared with 2000, when the proportion was 6 percent. The results show large disparities between urban and rural areas. In rural areas, only around 1 percent of households have electricity, compared with 25 percent in urban areas. With respect to drinking water, at the national level, almost 33 percent of households use spring water and one-quarter of households use a public tap; 14 percent of households use uncovered public wells as a source of drinking water and 22 percent consume water from a public tap. Overall, 19 percent of households use water that is considered unhealthy, leaving the population open to increased risk of contracting diseases related to unclean drinking water. The proportion of households with access to running water in their dwelling or courtyard remains low, approximately 3 percent. In rural areas, more than half of the households use unsafe drinking water because 55 percent draw their water from springs (35 percent), rivers/streams (12 percent), or ponds/lakes (8 percent). Figure 2.2 Age-specific Attendance Rates (Percentage of the population age 5-24 years attending school, by age and sex) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Age 0 10 20 30 40 50 60 70 80 90 100 Percent Male Female RDHS 2005 Household Characteristics | 19 Table 2.5 Housing characteristics Percent distribution of households by housing characteristics, according to residence, Rwanda 2005 Residence Housing characteristic Urban Rural Total Electricity Yes 25.1 1.3 4.8 No 74.7 98.6 95.1 Total 100.0 100.0 100.0 Source of drinking water Piped into dwelling/compound/plot 14.1 0.5 2.5 Public tap 41.3 22.1 24.9 Open well in compound/plot 0.5 0.1 0.1 Open public well 12.0 13.8 13.5 Covered well in compound/plot 0.1 0.0 0.0 Covered public well 5.2 6.3 6.2 Spring 18.5 35.2 32.7 River, stream 4.9 12.2 11.2 Pond, lake 1.5 8.0 7.1 Dam 0.4 0.9 0.9 Rainwater 0.1 0.3 0.3 Tanker truck 0.1 0.0 0.0 Bottled water 0.1 0.0 0.0 Other 1.2 0.5 0.6 Total 100.0 100.0 100.0 Time to water source Percentage < 15 minutes 47.9 27.0 30.1 Median time to source (in minutes) 14.3 28.0 24.4 Sanitation facility Flush toilet 5.4 0.2 1.0 Traditional pit toilet 44.0 70.8 66.9 Ventilated improved pit (VIP) latrine 47.1 24.1 27.5 No facility, bush, field 3.4 4.8 4.6 Other 0.1 0.1 0.1 Total 100.0 100.0 100.0 Flooring material Earth, mud, sand 51.8 92.1 86.2 Dung 0.5 0.9 0.8 Parquet, polished wood 0.0 0.0 0.0 Vinyl, asphalt strips 0.0 0.0 0.0 Ceramic tiles 1.0 0.0 0.2 Cement 46.5 6.9 12.7 Carpet 0.2 0.0 0.0 Other 0.1 0.1 0.1 Total 100.0 100.0 100.0 Number of households 1,510 8,762 10,272 In urban areas, public taps constitute the main water source and are used by 41 percent of the households surveyed. 19 percent of urban households use spring water and 14 percent have running water in their dwellings or courtyards. Finally, 12 percent draw water from uncovered public wells. The situation has not improved since 2000. The proportion of households that have running water in their dwelling units has decreased by 3 percent. The number of households using water from a public tap has dropped by 4 percent. Table 2.5 shows that 30 percent of households are within 15 minutes of their water source. This proportion is lower in rural areas (27 percent) than in urban areas (48 percent). The median time to drinking water source is 24 minutes for the country as a whole, 28 minutes for rural areas and 14 minutes for urban areas. 20 | Household Characteristics Compared with 2000, the proportion of households less than 15 minutes from their water source has increased by 5 percent (from 25 percent to 30 percent). However, the change is insignificant in terms of the median time to drinking water source, which was 26 minutes in 2000 and is 24 minutes now. With respect to type of toilet facilities, Table 2.5 shows a high proportion of households with access only to open pits or uncovered latrines (67 percent); 28 percent of households use covered latrines. In the country as a whole, rural areas have more rudimentary latrines (71 percent) than ventilated improved pit (VIP) latrines (24 percent), while in urban areas the proportion of VIP latrines (47 percent) and rudimentary latrines (44 percent) are similar to one another. Very few households have flush toilets: 1 percent in the country as a whole, 5 percent in urban areas, and an insignificant percentage in rural areas. It should also be noted that 5 percent of households have no toilet facilities at all. Compared with the previous survey, the proportion of households using VIP latrines has increased significantly, from 7 percent to approximately 28 percent. The proportion of households with no facilities at all has remained the same. The type of material used for flooring is extremely important because some materials are a propagation factor for certain disease-causing germs and parasites. The great majority of Rwandan households use earth/sand/dung flooring (86 percent). The proportion is higher in rural areas (92 percent) than in urban areas (52 percent). It should also be noted that 13 percent of households have cement floors. However, this type of flooring occurs much more frequently in urban than in rural areas (47 percent compared with 7 percent). To evaluate households’ socioeconomic level, the survey gathered information about ownership of certain durable goods considered indicative of higher socioeconomic living standards. Table 2.6, shows that half of Rwandan households own none of the goods listed. The proportion is higher in rural areas than in urban areas (56 percent for rural, 32 percent for urban). However, it has declined in relation to 2000, when 63 percent of households owned none of the goods listed. Overall, the most frequently owned durable good is a radio (46 percent), which is found more often in urban households than in rural areas (65 percent compared with 43 percent). The proportion of house- holds owning radios has increased overall in relation to 2000, when only 35 percent of households owned a radio. Bicycles are used as a means of transportation in 11 percent of households in both rural and urban areas. In addition, in urban areas, 24 percent of households own a mobile telephone, and 14 percent own a television; in rural areas, these goods are more or less nonexistent. Table 2.7 shows the percent distribution of households by wealth quintile. The wealth index was developed on the basis of household goods data, using principal components analysis. The information on household goods comes from responses to questions about ownership of certain durable goods (television, radio, car, etc.) and questions about certain housing characteristics such as access to electricity, source of drinking water, type of toilet facilities, type of flooring material, number of rooms used for sleeping, type of cooking fuel, etc. The index was developed as follows: • Each durable good or housing characteristic is assigned a weight (score or coefficient) generated by principal components analysis. Table 2.6 Household durable goods Percentage of households possessing various durable consumer goods, by residence, Rwanda 2005 Residence Durable consumer good Urban Rural Total Radio 65.2 42.5 45.8 Television 14.0 0.3 2.3 Mobile telephone 24.1 1.3 4.6 Non-mobile telephone 4.8 0.1 0.8 Refrigerator 7.7 0.1 1.2 Bicycle 10.5 11.1 11.0 Motorcycle/scooter 1.8 0.3 0.5 Car, truck 4.2 0.1 0.7 None of the above 31.7 55.5 52.0 Number of households 1,510 8,762 10,272 Household Characteristics | 21 • The resulting scores for durable goods are standardized according to a normal distribution assuming a mean of 0 and a standard deviation of 1 (Gwatkin et al., 2000). • Each household is assigned a score for each durable good and these scores are added together to obtain a total for each household. • The households are classified in increasing order of total score and divided into 5 equal categories, quintiles. This yields a scale from 1 (poorest quintile) to 5 (richest quintile). • The score for each household is assigned to the individuals in that household. The individuals are thus distributed among the categories. The results show that in urban areas, 60 percent of households fall into the richest quintile, while in rural areas only 12 percent fall into this quintile. The proportion of rich households is highest in the City of Kigali (69 percent). Conversely, in urban areas, only 6 percent of households fall into the poorest quintile. In fact, the preceding tables showing ownership of durable goods, housing characteristics, and source of drinking water have already established that the population of Rwanda is generally poor. Table 2.7 only confirms the previous results and explains the relative lack of variation between provinces. Table 2.7 Wealth quintiles Percent distribution of households by wealth quintile, according to residence and province, Rwanda 2005 Wealth quintile Residence/ province Lowest Second Middle Fourth Highest Total Number Residence Urban 6.1 7.7 11.6 14.8 59.7 100.0 1,510 Rural 24.2 20.4 22.2 21.5 11.7 100.0 8,762 Province Kigali city 6.4 4.4 10.8 9.8 68.5 100.0 864 South 21.4 20.8 19.0 22.4 16.5 100.0 2,722 West 23.5 15.3 21.6 24.5 15.1 100.0 2,522 North 27.3 19.5 22.6 18.6 12.0 100.0 1,946 East 20.5 24.2 23.6 19.5 12.2 100.0 2,218 Total 21.6 18.6 20.6 20.5 18.7 100.0 10,272 2.5 BIRTH REGISTRATION WITH CIVIL AUTHORITIES Registering a child with civil authorities establishes the child’s legal family ties and his or her right to a name and nationality prior to the age of majority. It confers on the child the right to be recognized by his or her parents and the right to state protection if his or her rights are abused by the parents. It gives the child access to social assistance through the parents, including health insurance, and establishes family lineage. It is therefore an essential formality. 22 | Household Characteristics Registration of a child with civil authorities, if performed correctly, also provides a reliable source of sociodemographic statistics. For this reason, the survey asked whether children had been registered with the civil authorities. Table 2.8 shows that a majority of children have been registered with the civil authorities (82 percent); only 18 percent of children (less than one in five) have not been registered. Of those children declared with the civil authorities, 78 percent possess birth certificates. Children’s age and sex have little to do with whether or not they are registered with the civil authorities. Also, the level of wealth does not seem to influence the prevalence of birth registration. Children in the second and middle wealth quintiles showed the highest levels of registration (84 percent in both of these quintiles). There is some discrepancy by residence with, curiously, rural areas showing a higher percentage of birth registrations (83 percent compared with 79 percent in urban areas). Results by province show that households in the North and South provinces are the most likely to have declared their children with the civil authorities (85 percent and 89 percent, respectively). Table 2.8 Birth registration of children under age five Percentage of de jure children under five years of age whose births are registered with the civil authorities, according to background characteristics, Rwanda 2005 Percentage of children whose births are registered: Background characteristic Had a birth certificate Did not have a birth certificate Total registered Number of children Age <2 79.6 2.8 82.4 3,411 2-4 76.6 5.7 82.3 4,711 Sex Male 77.3 4.5 81.8 4,103 Female 78.5 4.5 82.9 4,019 Residence Urban 74.0 4.6 78.6 1,170 Rural 78.5 4.5 83.0 6,952 Province Kigali city 74.1 5.2 79.2 596 South 81.7 3.2 84.9 2,013 West 73.8 5.0 78.8 2,166 North 83.8 5.2 89.0 1,622 East 74.4 4.5 78.8 1,725 Wealth quintile Lowest 76.9 4.9 81.8 1,687 Second 80.1 4.0 84.1 1,640 Middle 79.9 4.4 84.2 1,697 Fourth 76.0 4.7 80.7 1,623 Highest 76.3 4.5 80.9 1,475 Total 77.9 4.5 82.4 8,123 Characteristics of Household Respondents | 23 CHARACTERISTICS OF SURVEY RESPONDENTS 3 The purpose of this chapter is to provide a sociodemographic profile of the women age 15-49 and men age 15-59 who responded to this survey. This information is important for understanding the behavior of the population with respect to contraception, STIs, HIV/AIDS, and fertility preferences. Like the household questionnaire, the individual questionnaires gathered information concerning respondents’ age, place of residence, marital status, and educational attainment. This chapter will also analyze results with respect to literacy, exposure to mass media, and employment of the men and women surveyed. These characteristics will be used to interpret findings in the rest of the report. 3.1 BACKGROUND CHARACTERISTICS OF RESPONDENTS Given the importance of age in analyzing demographic phenomena, special attention was paid to making sure this statistic was accurately recorded in the survey. Prior to taking down any information, the interviewer asked respondents to gather all official documents providing information about themselves and other members of the household. If no official documents were available, the interviewer confirmed the age information provided by the respondent through reference to major life events (age at the time of marriage, age of first child, etc.) or well-known national or regional events. Table 3.1 shows no major disparities in the distribution of women age 15-49 and men age 15-59 grouped by five-year age increments. Proportions decline with increasing age. For women, the percentages range from 23 percent for the 15-19 age group, to 8 percent for the 45-49 age group. For men, the percentages range from 23 percent for ages 15-19, to 3 percent for ages 55-59. Table 3.1 Age of respondents Percent distribution of women and men by age, Rwanda 2005 Women Men Age Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number 15-19 22.8 2,585 2,595 22.9 1,102 1,079 20-24 20.8 2,354 2,356 19.6 946 951 25-29 15.4 1,738 1,745 13.1 632 647 30-34 12.9 1,466 1,460 10.6 509 515 35-39 10.0 1,134 1,133 9.2 442 435 40-44 10.0 1,135 1,127 8.4 404 408 45-49 8.0 910 905 7.8 378 378 50-54 na na na 5.4 260 261 55-59 na na na 3.1 147 146 Total 15-49 100.0 11,321 11,321 91.5 4,413 4,413 Total 15-59 na na na 100.0 4,820 4,820 na = Not applicable 24 | Characteristics of Household Respondents All men and women in the sample were asked their marital status. For the RDHS-III, all men and women were considered “married” if they were in union with a partner, whether the union was formal (legally married) or informal (“living together”). According to this definition, Table 3.2 shows that more than 4 in 10 women (44 percent) had never been married at the time of the survey, and more than half of the women (54 percent) were married. Nearly five in ten men were single (46 percent) and more than half of the men (52 percent) were married. In addition, 2 percent of the women were divorced, separated or widowed at the time of the survey, compared with 3 percent of the men. Table 3.2 Background characteristics of respondents Percent distribution of women and men by selected background characteristics, Rwanda 2005 Women Men Background characteristic Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number Marital status Never married 44.0 4,983 4,961 45.6 2,196 2,214 Married 54.1 6,126 6,138 51.9 2,500 2,478 Divorced/separated 1.4 158 167 1.9 89 92 Widowed 0.5 54 55 0.7 35 36 Residence Urban 17.0 1,921 2,616 17.4 840 1,130 Rural 83.0 9,400 8,705 82.6 3,980 3,690 Old province Kigali 8.0 900 1,085 8.8 426 511 Kigali Ngali 9.9 1,118 945 9.3 449 387 Gitarama 10.8 1,219 930 10.8 522 400 Butare 9.6 1,090 945 9.4 452 381 Gikongoro 5.7 650 885 5.7 275 371 Cyangugu 7.5 852 1,010 8.0 386 461 Kibuye 5.7 649 921 5.1 244 344 Gisenyi 10.4 1,179 938 10.1 488 385 Ruhengeri 10.4 1,180 940 9.9 478 376 Byumba 7.7 873 893 8.2 395 398 Umutara 4.9 554 897 5.6 271 425 Kibungo 9.3 1,057 932 9.0 433 381 Province Kigali city 10.0 1,127 1,329 10.8 523 619 South 26.1 2,958 2,760 25.9 1,250 1,152 West 24.9 2,824 2,971 24.6 1,185 1,237 North 18.2 2,063 1,821 17.5 845 746 East 20.7 2,348 2,440 21.1 1,017 1,066 Education No education 23.4 2,646 2,603 17.4 839 819 Primary 67.1 7,591 7,497 70.3 3,389 3,357 Secondary 9.0 1,018 1,134 10.9 526 566 More than secondary 0.6 66 87 1.4 66 78 Wealth quintile Lowest 21.4 2,421 2,327 18.0 867 826 Second 20.5 2,325 2,195 18.3 884 819 Middle 18.5 2,099 1,988 20.3 978 927 Fourth 18.8 2,133 2,151 20.8 1,004 1,012 Highest 20.7 2,342 2,660 22.6 1,087 1,236 Religion Catholic 45.3 5,126 4,975 51.5 2,482 2,416 Protestant 37.5 4,247 4,382 31.9 1,539 1,586 Adventist 13.2 1,498 1,490 12.0 578 585 Muslim 1.8 207 234 2.0 96 112 Other 0.9 97 97 - 0 0 No religion 1.3 146 143 2.6 126 121 Total 100.0 11,321 11,321 100.0 4,820 4,820 Characteristics of Household Respondents | 25 The distribution of respondents by residence shows that the majority of the Rwandan population is living in rural areas (83 percent of women and men). Similarly, the data by province shows a relatively uniform distribution, with no significant disparities between men and women. The tabulation of respondents by religion indicates a majority of Catholic adherents (45 percent of women and 52 percent of men), with Protestant religions coming in second (38 percent of women and 32 percent of men). The Adventist faith is the next most common religion (13 percent of women and 12 percent of men), followed by the Muslim faith (2 percent of women and 2 percent of men). Table 3.2 also shows the distribution of men and women according to household wealth quintile. The development of this index is explained in Chapter 2. Table 3.2 provides educational attainment data for the respondents. The proportion of women with no education is significantly higher than that of men (23 percent for women, 17 percent for men). However, the gap between males and females is not very wide at the primary and secondary levels. 3.2 EDUCATIONAL ATTAINMENT Tables 3.3.1 and 3.3.2 show the distribution of respondents by highest level of education attained. The proportions of educated men are only slightly higher than those of women: 70 percent have completed primary school, compared with 67 percent of women. At the secondary level, the proportions are 11 percent for men and 9 percent for women. It should be noted that proportions for both men and women drop significantly from the primary to secondary and secondary to post-secondary levels. Table 3.3.1 Educational attainment by background characteristics: women Percent distribution of women by highest level of schooling attended or completed, according to background characteristics, Rwanda 2005 Highest level of schooling attended or completed Background characteristic No education Primary Secondary More than secondary Total Number of women Age 15-19 8.6 84.4 7.0 0.0 100.0 2,585 20-24 17.5 71.6 10.2 0.8 100.0 2,354 25-29 17.8 69.9 10.9 1.5 100.0 1,738 30-34 24.6 63.5 11.3 0.6 100.0 1,466 35-39 36.4 51.7 11.1 0.8 100.0 1,134 40-44 41.8 50.7 7.2 0.2 100.0 1,135 45-49 50.0 45.8 4.0 0.2 100.0 910 Residence Urban 13.5 58.9 24.3 3.4 100.0 1,921 Rural 25.4 68.7 5.9 0.0 100.0 9,400 Province Kigali city 11.3 58.6 26.0 4.1 100.0 1,127 South 20.3 71.4 8.0 0.3 100.0 2,958 West 28.1 65.7 5.9 0.3 100.0 2,824 North 25.4 65.2 9.3 0.1 100.0 2,063 East 25.5 68.9 5.5 0.1 100.0 2,348 Wealth quintile Lowest 32.3 65.7 2.0 0.0 100.0 2,421 Second 25.9 71.3 2.7 0.0 100.0 2,325 Middle 25.6 70.2 4.2 0.0 100.0 2,099 Fourth 22.5 71.2 6.3 0.0 100.0 2,133 Highest 10.4 57.6 29.2 2.8 100.0 2,342 Total 23.4 67.1 9.0 0.6 100.0 11,321 26 | Characteristics of Household Respondents Table 3.3.2 Educational attainment by background characteristics: men Percent distribution of men by highest level of schooling attended or completed, according to background characteristics, Rwanda 2005 Highest level of schooling attended or completed Background characteristic No education Primary Secondary More than secondary Total Number of men Age 15-19 6.8 86.2 7.0 0.0 100.0 1,102 20-24 12.8 74.5 12.2 0.6 100.0 946 25-29 15.9 68.2 13.4 2.5 100.0 632 30-34 20.3 60.6 16.3 2.9 100.0 509 35-39 22.9 57.9 15.7 3.5 100.0 442 40-44 29.7 56.7 11.6 1.9 100.0 404 45-49 25.9 67.1 5.9 1.1 100.0 378 50-54 29.6 62.9 6.6 0.9 100.0 260 55-59 29.5 62.9 6.6 1.0 100.0 147 Residence Urban 9.5 59.5 24.9 6.2 100.0 840 Rural 19.1 72.6 8.0 0.4 100.0 3,980 Province Kigali city 9.9 56.3 26.4 7.4 100.0 523 South 16.4 73.1 9.8 0.6 100.0 1,250 West 17.8 71.6 9.8 0.7 100.0 1,185 North 20.1 70.2 8.7 1.0 100.0 845 East 19.8 72.6 7.3 0.3 100.0 1,017 Wealth quintile Lowest 25.5 71.6 2.9 0.0 100.0 867 Second 22.3 72.3 5.4 0.0 100.0 884 Middle 18.1 76.0 5.9 0.0 100.0 978 Fourth 16.0 73.8 10.1 0.1 100.0 1,004 Highest 7.7 59.3 27.0 6.0 100.0 1,087 Total 17.4 70.3 10.9 1.4 100.0 4,820 The data by age show that the proportion of men and women with no education has decreased significantly from previous generations. For men, this proportion has dropped from 30 percent in the 55-59 age group to 7 percent in the 15-19 age group. For women, the proportions for these age groups are 50 percent and 9 percent, respectively. The gap between men and women in the previous generations has narrowed significantly: among men age 45 to 49 years, 26 percent have no education, compared with 50 percent for women in the same age group. For those age 15-19 years, the proportions are 7 percent for men and 9 percent for women. Similarly, in the 15-19 age group, the proportion of girls who have completed primary school is not significantly different from that of boys (84 percent for girls, 86 percent for boys), although the percentage of boys is still slightly higher. In addition, 11 percent of young men have completed secondary school, compared with 9 percent of young women. The gaps are due to early marriage and pregnancy, which often prevent girls from pursuing a regular course of education. The educational attainment of respondents varies by residence. The proportion of men and women with no education is higher in rural areas (19 percent for men, 25 percent for women) than in urban areas (10 percent for men, 14 percent for women). Urban areas also have the highest proportion of men and women at every level of education except primary. Results by province show a wide gap between the City of Kigali and the rest of the country. In the City of Kigali, 11 percent of women and 10 percent of men have no education; in the other provinces the Characteristics of Household Respondents | 27 proportions are nearly twice as high. The West province has the highest percentage of women with no education (28 percent); the North and East provinces have the highest proportion of uneducated men (20 percent each). The data in this table show a positive relationship between educational attainment and household wealth: the proportion of men and women with no education decrease as household wealth increases. 3.3 LITERACY For this survey, literacy was established by asking respondents who reported not having attended school and or having attended only primary school to read a sentence that was presented to them. Respondents were then classified into one of the following three levels: cannot read at all; can read part of a sentence; can read a whole sentence. The test was given only to men and women who had less than a secondary education; those with secondary or postsecondary educations (10 percent of women and 12 percent of men) were considered literate. Tables 3.4.1 and 3.4.2 show that a higher proportion of women than men cannot read (29 percent of women; 22 percent of men). Conversely, 78 percent of men and 70 percent of women are considered literate; that is, they have attended secondary school or, if they have attended only primary school, they are able to read all or part of a sentence. Table 3.4.1 Literacy: women Percent distribution of women by level of schooling attended and by level of literacy, and percent literate, according to background characteristics, Rwanda 2005 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all Total1 Number of women Percent literate2 Age 15-19 7.0 60.3 12.2 20.2 100.0 2,585 79.6 20-24 11.0 51.6 12.1 25.1 100.0 2,354 74.6 25-29 12.4 55.0 11.1 21.4 100.0 1,738 78.4 30-34 11.9 50.7 9.8 27.1 100.0 1,466 72.4 35-39 11.9 41.7 10.0 36.0 100.0 1,134 63.6 40-44 7.4 33.8 11.7 46.5 100.0 1,135 52.9 45-49 4.2 30.9 9.2 55.0 100.0 910 44.3 Residence Urban 27.7 48.3 7.8 15.6 100.0 1,921 83.8 Rural 5.9 49.8 11.9 32.1 100.0 9,400 67.6 Province Kigali city 30.1 47.5 7.4 14.7 100.0 1,127 85.0 South 8.3 55.9 8.6 26.8 100.0 2,958 72.8 West 6.2 47.0 12.6 33.8 100.0 2,824 65.8 North 9.4 45.7 14.4 30.3 100.0 2,063 69.5 East 5.5 49.0 11.7 33.3 100.0 2,348 66.2 Wealth quintile Lowest 2.0 44.6 13.3 39.9 100.0 2,421 59.8 Second 2.7 51.2 12.8 32.9 100.0 2,325 66.8 Middle 4.2 51.2 12.2 32.0 100.0 2,099 67.7 Fourth 6.3 54.1 10.6 28.7 100.0 2,133 71.0 Highest 32.0 47.5 7.0 13.1 100.0 2,342 86.5 Total 9.6 49.6 11.2 29.3 100.0 11,321 70.3 1 Includes those with missing information 2 Refers to women who attended secondary school or higher and women who can read a whole sentence or part of a sentence. 28 | Characteristics of Household Respondents Table 3.4.2 Literacy: men Percent distribution of men by level of schooling attended and by level of literacy, and percent literate, according to background characteristics, Rwanda 2005 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all Total1 Number of men Percent literate2 Age 15-19 7.0 60.6 13.0 19.0 100.0 1,102 80.7 20-24 12.7 54.7 11.3 20.9 100.0 946 78.8 25-29 15.9 57.4 9.5 17.2 100.0 632 82.8 30-34 19.2 50.1 9.4 21.0 100.0 509 78.7 35-39 19.2 51.1 9.3 20.4 100.0 442 79.6 40-44 13.5 49.9 7.8 28.4 100.0 404 71.2 45-49 7.0 51.3 9.9 30.7 100.0 378 68.1 50-54 7.5 51.8 9.8 29.7 100.0 260 69.1 55-59 7.6 49.2 13.8 28.0 100.0 147 70.6 Residence Urban 31.0 47.3 7.6 13.3 100.0 840 86.0 Rural 8.3 56.1 11.3 23.9 100.0 3,980 75.8 Province Kigali city 33.8 44.4 7.8 13.6 100.0 523 86.0 South 10.5 53.7 10.4 24.9 100.0 1,250 74.6 West 10.6 56.3 10.4 22.2 100.0 1,185 77.3 North 9.7 54.1 12.2 23.7 100.0 845 76.0 East 7.5 59.3 11.5 21.4 100.0 1,017 78.4 Wealth quintile Lowest 2.9 52.9 12.7 31.1 100.0 867 68.5 Second 5.4 53.6 12.5 27.9 100.0 884 71.5 Middle 5.9 57.7 12.4 23.6 100.0 978 76.1 Fourth 10.2 59.1 10.5 19.9 100.0 1,004 79.8 Highest 33.0 49.7 6.2 10.6 100.0 1,087 88.9 Total 12.3 54.6 10.7 22.0 100.0 4,820 77.5 1 Includes those with missing information 2 Refers to men who attended secondary school or higher and women who can read a whole sentence or part of a sentence. The level of illiteracy varies appreciably by residence. Illiteracy is higher in rural areas than in urban areas. The results by province show a gap between the City of Kigali and the rest of the country: in Kigali, 86 percent of men and 85 percent of women are literate, compared with a maximum of 78 percent of men in the East province and 73 percent of women in the South province. In addition, results by wealth quintile show that the level of illiteracy decreases considerably from the poorest to the richest quintile, dropping for women from 40 percent in the lowest quintile to 13 percent in the highest quintile and for men from 31 percent in the lowest quintile to 11 percent in the highest quintile. 3.4 EXPOSURE TO MASS MEDIA Data on the exposure of men and women to mass media are especially important to the development of education programs and the dissemination of all types information, particularly information about health and family planning. Tables 3.5.1 and 3.5.2 present data on the exposure of men and women to mass media (print or broadcast). It should be stated at the outset that it is not necessary for a household to own a radio or television or to buy a newspaper in order to have access to these media because many people listen to the radio or watch television at the homes of friends and neighbors. Characteristics of Household Respondents | 29 Table 3.5.1 Exposure to mass media: women Percentage of women who usually read a newspaper at least once a week, watch television at least once a week, and listen to the radio at least once a week, by background characteristics, Rwanda 2005 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media No media Number of women Age 15-19 11.2 6.9 56.7 1.5 39.7 2,585 20-24 8.6 6.7 57.0 1.9 40.4 2,354 25-29 8.8 5.2 56.0 1.8 42.3 1,738 30-34 7.5 4.2 53.7 1.2 44.4 1,466 35-39 6.8 4.2 50.0 1.3 47.5 1,134 40-44 6.7 2.6 48.6 0.7 49.9 1,135 45-49 3.6 2.6 48.2 0.8 50.7 910 Residence Urban 15.4 22.9 73.5 6.7 23.5 1,921 Rural 6.9 1.6 50.2 0.4 47.7 9,400 Province Kigali city 15.4 30.2 76.6 7.5 19.9 1,127 South 8.7 2.8 57.5 1.2 40.9 2,958 West 10.5 2.9 42.4 1.0 53.8 2,824 North 7.6 2.7 50.3 0.6 47.6 2,063 East 2.4 1.2 56.6 0.1 42.4 2,348 Education No education 0.8 1.1 37.7 0.1 61.5 2,646 Primary 8.1 3.3 55.6 0.5 41.7 7,591 Secondary or higher 28.2 28.2 83.5 11.3 12.7 1,084 Wealth quintile Lowest 4.1 0.4 22.6 0.0 75.1 2,421 Second 5.2 1.0 54.3 0.0 43.5 2,325 Middle 6.3 1.0 52.4 0.0 45.7 2,099 Fourth 7.1 1.6 63.4 0.2 34.8 2,133 Highest 18.7 21.3 79.6 6.7 17.0 2,342 Total 8.3 5.2 54.1 1.4 43.6 11,321 Table 3.5.1 shows that, at the national level, more than two in five women (44 percent) and approximately one in five men (19 percent) are not exposed to any media. However, there has been a significant improvement since the 2000 RDHS-II, which reported that 59 percent of women and 35 percent of men were not exposed to any media. Radio is the most common form of media exposure: more than half of the women (54 percent) and four out of five of the men (80 percent) reported listening to the radio at least once a week. One in twenty women (5 percent) and one in ten men (11 percent) watch television at least once a week. Men also reported reading a newspaper a little more frequently than women: only 8 percent of women, compared with 10 percent of men, reported reading a newspaper at least once a week. The proportions of men and women who are exposed to all three media are very low: only 1 percent of women and 4 percent of men. The data by age show that the younger generations are relatively more exposed to mass media than older people. In fact, the proportions of women who are not exposed to any media vary from 40 percent for women age 15-19 to 51 percent for women age 45-49. For men, the age differences are narrow and uneven. 30 | Characteristics of Household Respondents Table 3.5.2 Exposure to mass media: men Percentage of men who usually read a newspaper at least once a week, watch television at least once a week, and listen to the radio at least once a week, by background characteristics, Rwanda 2005 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media No media Number of men Age 15-19 6.5 11.1 77.5 2.3 21.3 1,102 20-24 10.7 14.3 84.0 5.4 15.0 946 25-29 14.4 13.2 81.4 6.5 16.7 632 30-34 12.5 14.1 79.2 7.8 19.8 509 35-39 12.6 8.7 80.1 5.1 18.8 442 40-44 9.1 6.2 77.3 2.7 21.8 404 45-49 9.3 6.3 76.4 2.5 22.3 378 50-54 7.9 6.4 77.4 3.0 21.0 260 55-59 6.9 3.7 78.4 2.2 21.6 147 Residence Urban 27.3 37.5 89.8 19.2 9.3 840 Rural 6.5 5.2 77.5 1.3 21.3 3,980 Province Kigali city 33.4 47.6 89.2 25.9 9.5 523 South 6.1 7.4 76.3 2.0 23.1 1,250 West 8.6 6.1 73.9 1.6 24.7 1,185 North 6.7 5.6 82.9 1.9 16.6 845 East 7.5 5.9 82.7 1.6 15.1 1,017 Education No education 0.5 3.0 67.1 0.1 32.4 839 Primary 7.4 7.9 80.2 2.1 18.5 3,389 Secondary or higher 38.7 38.7 94.4 23.8 4.6 592 Wealth quintile Lowest 3.1 2.4 62.3 0.3 36.7 867 Second 5.0 2.6 75.3 0.3 23.0 884 Middle 4.3 3.7 80.1 0.6 18.6 978 Fourth 7.4 6.0 85.9 1.1 13.5 1,004 Highest 27.4 35.1 90.8 17.4 7.8 1,087 Total 10.1 10.8 79.6 4.4 19.2 4,820 Results by residence reveal significant differentials. In rural areas, the percentage of women who are not exposed to any media is twice as high as in urban areas (48 percent compared with 24 percent). In rural areas, women often have no access to media at all even if media exists in the household. This is because housework takes up the majority of their time and the radio is often considered the property of the man, who may take it with him when he leaves the house. The differential is also wide for men: the proportion of men not exposed to any media varies from 21 percent in rural areas to 9 percent in urban areas. Results by province show significant differences between the City of Kigali and other provinces: indeed, in Kigali more than one-quarter of men are exposed to all three media, compared with approxi- mately 2 percent of men elsewhere. For women, the proportion varies from a maximum of 8 percent in Kigali to less than 1 percent in the North and East provinces. Educational attainment has a significant impact on the level of media exposure. For both men and women, those who have completed secondary or postsecondary levels are the most likely to be exposed to all three media: 11 percent of women and 24 percent of men, compared with only 1 percent of women and 2 percent of men who have completed only Characteristics of Household Respondents | 31 primary school. In addition, the results show that 62 percent of women with no education are not exposed to any media, compared with 13 percent of women with secondary or postsecondary educations. For men, 32 percent of those with no education are not exposed to any media, while only 5 percent of those with secondary or postsecondary educations are not exposed to any media. As in the case of educational attainment, there is a positive relationship between household wealth and media exposure. Men and women in the richest households have the highest levels of exposure to all three media: 7 percent of women and 17 percent of men, compared with less than 1 percent of men and 0 percent of women in the poorest households. 3.5 EMPLOYMENT The RDHS-III asked both men and women whether they were employed at the time of the survey. Respondents who reported having held a job, paid or unpaid, in any sector during the 12 months preceding the survey were considered employed. Table 3.6 shows that, at the national level, 9 percent of women were not working at the time of the survey even if they reported working in the preceding 12 months. More than three in five women (64 percent) were employed at the time of the survey. The percentage of women working at the time of the survey increases steadily with age, rising from 44 percent at age 15-19 to 76 percent at age 40-44. Women who were separated, divorced or widowed (74 percent) and married women (72 percent) were the most likely to be working at the time of the survey. The number of children also affects a woman’s level of employment. As the number of children increases, the proportion of women who work also increases, from 70 percent among women with only one or two children, to 74 percent among women with three children or more. Data by residence show that rural areas have the highest proportion of women working at the time of the survey (66 percent, compared with 54 percent in urban areas). The City of Kigali has the lowest percentage of women working (44 percent). In the provinces, the proportion of employed women ranges from 59 percent in the West province, to 61 percent in the North province, to a maximum of 72 percent in the South province and 74 percent in the East province. Results by educational attainment show that women with no education (70 percent) are proportionally more likely to be employed than women who have completed primary school (64 percent) and women who have completed secondary or postsecondary educations (53 percent). Finally, women in households in the two poorest wealth quintiles are more likely to be employed (73 percent and 77 percent) than women in the richest households (52 percent). The results for men show that 52 percent of men had some form of employment at the time of the survey. This is lower than for women (64 percent). As with women, the percentage of men working at the time of the survey increases with age, from 29 percent for those age 15 to 19, to 61 percent for those age 50 to 54. With respect to marital status, the results show married men and separated or divorced men being proportionally more likely to be working (59 percent for married men, 57 percent for separated, divorced, or widowed men) than other men. With respect to residence, urban areas had the highest proportion of men working at the time of the survey: 61 percent, compared with 50 percent in rural areas. With respect to educational attainment, the results show men with no education (59 percent) being proportionally more likely to be employed than men with primary educations (49 percent) and men with secondary or postsecondary education (54 percent). Finally, unlike women, the proportion of men working is lower in the poorest households than in the richest households (51 percent compared with 60 percent). 32 | Characteristics of Household Respondents Table 3.6 Employment status Percent distribution of women and men by employment status, according to background characteristics, Rwanda 2005 Women Men Employed in the 12 months preceding the survey Employed in the 12 months preceding the survey Background characteristic Currently employed Not currently employed Not employed in the 12 months preceding the survey Total Number Currently employed Not currently employed Not employed in the 12 months preceding the survey Total Number Age 15-19 43.5 6.2 50.0 100.0 2,585 29.4 3.5 66.3 100.0 1,102 20-24 62.8 10.0 27.2 100.0 2,354 57.0 5.2 37.4 100.0 946 25-29 70.2 8.4 21.4 100.0 1,738 56.9 7.5 35.3 100.0 632 30-34 72.0 10.4 17.6 100.0 1,466 58.8 6.7 34.6 100.0 509 35-39 73.8 8.7 17.5 100.0 1,134 59.9 6.7 33.3 100.0 442 40-44 75.9 10.6 13.5 100.0 1,135 58.2 3.8 37.8 100.0 404 45-49 75.1 10.4 14.5 100.0 910 58.3 5.6 35.9 100.0 378 50-54 na na na na na 61.0 8.0 30.6 100.0 260 55-59 na na na na na 60.0 3.6 36.3 100.0 147 Marital status Never married 50.4 7.8 41.7 100.0 4,263 43.4 5.0 51.1 100.0 2,196 Married 71.8 9.3 18.9 100.0 5,510 58.7 5.8 35.4 100.0 2,500 Divorced, separated, widowed 74.3 10.7 15.0 100.0 1,548 56.5 6.0 37.4 100.0 125 Number of living children 0 51.4 7.7 40.8 100.0 4,363 52.1 5.4 42.2 100.0 1,928 1-2 69.7 9.2 21.1 100.0 2,722 51.1 5.5 43.0 100.0 1,306 3-4 74.1 9.1 16.8 100.0 2,266 52.0 4.6 43.2 100.0 1,014 5 + 73.1 11.0 15.9 100.0 1,970 50.9 6.9 41.9 100.0 571 Residence Urban 53.5 8.9 37.4 100.0 1,921 60.9 8.7 29.6 100.0 840 Rural 66.3 8.9 24.8 100.0 9,400 49.7 4.7 45.3 100.0 3,980 Province Kigali city 43.5 10.6 45.7 100.0 1,127 55.7 10.8 33.1 100.0 523 South 71.7 11.4 16.7 100.0 2,958 43.0 6.2 50.6 100.0 1,250 West 58.8 7.8 33.5 100.0 2,824 51.5 7.5 40.7 100.0 1,185 North 60.6 9.5 29.9 100.0 2,063 52.8 2.9 43.7 100.0 845 East 73.9 5.8 20.3 100.0 2,348 59.4 1.4 38.9 100.0 1,017 Education No education 70.2 9.8 20.0 100.0 2,646 59.3 5.1 35.2 100.0 839 Primary 63.6 8.9 27.5 100.0 7,591 49.3 5.2 45.3 100.0 3,389 Secondary or higher 53.0 6.8 39.9 100.0 1,084 54.1 7.1 37.6 100.0 592 Wealth quintile Lowest 73.3 7.7 18.9 100.0 2,421 50.8 3.5 45.5 100.0 867 Second 76.9 8.4 14.7 100.0 2,325 52.4 3.6 44.1 100.0 884 Middle 62.9 9.4 27.7 100.0 2,099 50.6 6.7 42.4 100.0 978 Fourth 54.0 10.5 35.4 100.0 2,133 44.1 6.1 49.2 100.0 1,004 Highest 52.2 8.8 38.8 100.0 2,342 59.6 6.7 33.2 100.0 1,087 Total 64.1 8.9 26.9 100.0 11,321 51.7 5.4 42.6 100.0 4,820 na = Not applicable Characteristics of Household Respondents | 33 Table 3.7.1 shows women’s occupations. The majority of women who were employed at the time of the survey, or who had worked during the 12 months preceding it, were employed in agriculture (86 percent). Among those working in other occupations (13 percent), 5 percent worked in the sales and services; 4 percent performed unskilled manual labor, and 1 percent performed skilled manual labor. Only 3 percent reported working in a technical or administrative occupation. As expected, the data by residence show that the proportion of women working in agriculture is higher in rural areas (92 percent, compared with 44 percent in urban areas). It is much lower in the City of Kigali (27 percent). Outside the City of Kigali, the lowest proportion of women working in agriculture is 88 percent. With respect to educational attainment, 94 percent of women with no education work in agriculture compared with 39 percent of women with secondary or postsecondary education. Table 3.7.1 Occupation: women Percent distribution of women employed in the 12 months preceding the survey by occupation, according to background characteristics, Rwanda 2005 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agri- culture Missing Total Number of women Age 15-19 0.8 0.1 4.5 0.9 12.0 79.4 2.4 100.0 1,285 20-24 1.7 0.8 5.3 1.4 6.5 83.7 0.6 100.0 1,712 25-29 4.7 1.1 6.3 1.3 2.4 84.0 0.2 100.0 1,367 30-34 4.3 0.8 6.1 1.1 1.8 85.6 0.2 100.0 1,207 35-39 3.8 1.1 3.9 1.4 2.4 87.1 0.4 100.0 935 40-44 2.7 0.7 3.7 0.6 1.5 90.5 0.4 100.0 981 45-49 1.4 0.4 2.7 0.5 1.4 93.5 0.1 100.0 778 Marital status Never married 2.7 1.0 5.4 1.7 10.6 77.0 1.6 100.0 2,480 Married 2.7 0.7 4.6 0.7 1.4 89.8 0.2 100.0 4,470 Divorced, separated, Widowed 3.1 0.4 4.8 1.3 3.1 86.8 0.6 100.0 1,316 Residence Urban 9.9 3.7 17.6 3.7 19.2 44.4 1.5 100.0 1,199 Rural 1.6 0.2 2.7 0.6 1.9 92.4 0.5 100.0 7,067 Province Kigali city 13.9 5.4 18.9 5.2 27.0 26.9 2.8 100.0 609 South 1.9 0.5 2.2 0.8 2.7 91.2 0.7 100.0 2,460 West 1.7 0.2 6.6 0.6 2.9 87.3 0.5 100.0 1,879 North 2.7 0.4 4.3 1.2 3.0 88.1 0.3 100.0 1,446 East 1.4 0.2 2.4 0.5 2.0 93.1 0.3 100.0 1,872 Education No education 0.5 0.0 2.4 0.5 2.7 93.7 0.2 100.0 2,116 Primary 0.7 0.1 4.8 1.0 5.0 87.8 0.6 100.0 5,503 Secondary or higher 27.7 8.0 13.2 4.0 5.5 39.1 2.5 100.0 648 Wealth quintile Lowest 0.2 0.1 0.9 0.3 1.4 96.7 0.4 100.0 1,962 Second 0.2 0.1 1.4 0.5 0.6 96.9 0.3 100.0 1,984 Middle 1.0 0.1 4.5 0.9 2.4 90.5 0.6 100.0 1,517 Fourth 1.0 0.1 4.9 1.4 2.9 89.2 0.6 100.0 1,376 Highest 13.6 3.7 15.4 2.9 17.6 45.2 1.6 100.0 1,427 Total 2.8 0.7 4.8 1.1 4.4 85.5 0.7 100.0 8,266 34 | Characteristics of Household Respondents Table 3.7.2 shows men’s occupations. Like women, the majority of men work in agriculture (62 percent). One in seven men performs unskilled manual labor (14 percent), and 11 percent perform skilled manual labor. As expected, the proportion of men working in agriculture is higher in the rural areas (73 percent compared with 18 percent in urban areas). However, the proportion of men performing skilled or unskilled manual labor is significantly higher in urban areas than in rural areas (23 percent compared with 8 percent in rural areas for skilled, and 28 percent compared with 10 percent in rural areas for unskilled). With respect to educational attainment, the results show that, like women, the majority of men with no education work in agriculture (78 percent, compared with 22 percent of those with at least a secondary education). However, of those with the highest educational attainment levels, 37 percent work in managerial or technical occupations. Results by wealth quintile show that a majority of men in the poorest households work in agriculture (86 percent). Conversely, in the richest quintile, only 21 percent of men work in agriculture, and 18 percent work in managerial or technical occupations. Table 3.7.2 Occupation: men Percent distribution of men employed in the 12 months preceding the survey by occupation, according to background characteristics, Rwanda 2005 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agri- culture Missing Total Number of men Age 15-19 0.9 0.0 4.5 3.7 32.2 57.9 0.8 100.0 363 20-24 3.4 0.1 6.9 8.7 21.1 57.6 2.1 100.0 588 25-29 8.9 0.4 7.7 13.8 12.5 55.4 1.3 100.0 407 30-34 9.8 1.9 8.0 14.3 6.9 58.0 1.0 100.0 333 35-39 7.5 1.2 4.1 15.6 7.9 63.3 0.5 100.0 295 40-44 9.1 1.5 7.5 10.0 6.0 65.8 0.0 100.0 251 45-49 4.1 1.1 3.8 13.2 7.0 70.8 0.0 100.0 242 50-54 5.1 2.3 0.0 11.2 6.6 74.8 0.0 100.0 179 55-59 9.5 0.0 2.1 13.4 4.1 70.9 0.0 100.0 94 Marital status Never married 5.8 0.3 7.9 8.8 23.9 51.5 1.8 100.0 1,063 Married 6.2 1.1 4.5 12.5 7.8 67.4 0.4 100.0 1,611 Divorced, separated, widowed 4.3 2.9 0.0 11.9 7.5 73.3 0.0 100.0 78 Residence Urban 14.3 2.7 11.6 23.2 27.9 18.1 2.1 100.0 584 Rural 3.7 0.3 4.1 7.8 10.3 73.2 0.6 100.0 2,168 Province Kigali city 16.2 3.2 12.5 25.7 32.4 7.0 2.9 100.0 348 South 4.9 0.3 5.6 11.5 13.0 63.6 1.2 100.0 615 West 5.8 0.6 6.1 8.6 10.0 68.5 0.4 100.0 700 North 5.3 0.5 3.9 13.6 15.5 60.8 0.3 100.0 471 East 2.1 0.6 2.9 3.2 8.2 82.5 0.5 100.0 619 Education No education 0.3 0.0 2.1 5.3 13.7 78.2 0.4 100.0 540 Primary 1.5 0.2 6.4 11.3 15.6 64.2 0.8 100.0 1,849 Secondary or higher 37.2 5.3 7.6 18.4 6.6 22.4 2.5 100.0 362 Wealth quintile Lowest 0.5 0.0 1.9 7.1 4.7 85.7 0.2 100.0 471 Second 0.7 0.0 2.8 7.4 10.7 78.1 0.3 100.0 495 Middle 1.6 0.4 3.3 8.3 10.0 75.9 0.6 100.0 560 Fourth 4.1 0.0 5.1 10.6 15.2 64.4 0.4 100.0 505 Highest 17.9 2.9 12.5 18.5 24.8 20.9 2.5 100.0 721 Total 6.0 0.8 5.7 11.0 14.0 61.5 0.9 100.0 2,752 Characteristics of Household Respondents | 35 Table 3.8 shows the distribution of women employed during the 12 months preceding the survey by type of earnings, type of employer, and continuity of employment. Overall, 57 percent of women were not paid for their work, 16 percent were paid in cash and in kind, 15 percent were paid in cash only, and 12 percent were paid in kind only. Women in nonagricultural occupations were more likely to be paid in cash (82 percent) than those working in agriculture (4 percent). In the majority of cases (73 percent), women are self-employed, regardless of their occupations. Women who work in agriculture are more likely to work for a family member than women in nonagricultural occupations (17 percent compared with 3 percent). Finally, 75 percent of all women work all year, with the largest proportion working in agriculture (77 percent). Table 3.8 Type of employment Percent distribution of women employed in the 12 months preceding the survey by type of earnings, type of employer, and continuity of employment, according to type of employment (agricultural or nonagricultural), Rwanda 2005 Employment characteristic Agricultural work Nonagricultural work Total Type de earnings Cash only 4.0 81.8 14.8 Cash and in-kind 17.4 5.7 15.7 In-kind only 14.4 0.7 12.4 Not paid 64.2 11.7 57.0 Total 100.0 100.0 100.0 Type of employer Employed by family member 16.8 3.4 14.8 Employed by non-family member 7.0 40.4 11.6 Self-employed 76.2 55.5 73.4 Total 100.0 100.0 100.0 Continuity of employment All year 76.5 67.8 75.2 Seasonal 15.3 14.3 15.2 Occasional 8.2 17.7 9.5 Total 100.0 100.0 100.0 Number of women 7,066 1,146 8,266 Note: Total includes women with missing information on type of employment. Table 3.9 shows the distribution of women employed in the 12 months preceding the survey by type of employer, according to background characteristics. Approximately three-quarters of women are self-employed (73 percent). This proportion increases with age, from 63 percent in the age group 20-24 years to 90 percent among those age 45-49 years. The proportion of self-employed women is highest in rural areas (76 percent) and among those having no education (82 percent). Moreover, nearly one in eight women (12 percent) is employed by a non-family member. This is especially true of women in urban areas (29 percent) and women with a secondary or postsecondary education (33 percent). In addition, more than one in eight women (15 percent) works for a family member, a situation affecting 43 percent of the youngest women, 16 percent of women in rural areas, and 19 percent of women with a primary education. 36 | Characteristics of Household Respondents Table 3.9 Type of employer Percent distribution of women employed in the 12 months preceding the survey by type of employer, according to background characteristics, Rwanda 2005 Type of employer Background characteristic Employed by family member Employed by non-family member Self- employed Total1 Number of women Age 15-19 43.0 17.0 39.5 100.0 1,285 20-24 23.6 13.2 63.0 100.0 1,712 25-29 9.2 10.7 79.8 100.0 1,367 30-34 4.6 10.2 85.1 100.0 1,207 35-39 4.1 10.3 85.7 100.0 935 40-44 2.9 8.5 88.5 100.0 981 45-49 2.2 8.1 89.5 100.0 778 Residence Urban 9.1 29.2 61.3 100.0 1,199 Rural 15.8 8.6 75.5 100.0 7,067 Education No education 7.0 10.9 81.9 100.0 2,116 Primary 18.7 9.2 71.8 100.0 5,503 Secondary or higher 6.9 33.3 59.5 100.0 648 Total 14.8 11.6 73.4 100.0 8,266 1 Includes those with missing information Fertility | 37 FERTILITY 4 For more than 20 years, Rwanda has been collecting sociodemographic data to evaluate the fertility levels and characteristics of its population. These efforts include the 1978 RGPH (General Population and Housing Census), the 1983 ENF (National Fertility Survey), the 1991 RGPH, the 1992 RDHS-I (Rwanda Demographic and Health Survey), the 1996 ESD (Socio-demographic Survey), the 2000 RDHS-II, the 2002 RGPH, and the current survey, the 2005 RDHS-III. Information on fertility obtained by the RDHS-III is used to estimate fertility levels, to determine the timing of births, and to describe the fertility characteristics of such variables as residence and educational attainment. It provides recent indicators of fertility rates and birth spacing not only at the national level, but also by province and residence. Fertility is one of the three principle components of population dynamics, the others being mortality and migration (United Nations, 1973). For this reason, the collection of data on fertility levels, trends, and differentials has been a prime objective of the Demographic and Health Surveys program since its inception. The continued collection of fertility data has been essential to recognizing the important role that fertility plays in Rwanda’s overall population growth equation. Rwanda has been conducting national fertility surveys since 1983, using them as the primary basis for developing its population policies. This chapter analyzes the fertility data gathered by the RDHS-III, which have been used to estimate fertility levels, trends, and differentials according to selected background characteristics. The chapter also presents results for age at first birth and birth intervals, and concludes with an analysis of teenage fertility, which has become critical to the issue of the fertility transition, and is a special emphasis of the National Reproductive Health Policy Declaration. Fertility data were obtained by posing a series of questions to all eligible women respondents. During the interview, interviewers recorded the total number of children to whom the woman had given birth, the gender of each child, the number of children currently living with the mother, the number of children living elsewhere, the number of children who had died, and the number still living. A complete birth history was compiled, from the earliest to the most recent birth. In addition, the following information was gathered for each birth: type of birth (single or multiple), sex of child, date of birth, and survival status. For living children, respondents were asked the current age of the child and whether the child was living with its mother or elsewhere. For children who had died, respondents were asked the age at the time of death. At the end of the interview, the interviewer verified that the number of children reported by the mother initially (for each category: living and dead) was consistent with the number of children obtained from the birth history. Because this is a retrospective survey, the data can be used to estimate not only current fertility levels, but also fertility trends over the past 20 years. Despite the organization and controls established to ensure the achievement of survey objectives (including training, instructions to field and data processing personnel, and quality controls at all levels), the data obtained may be subject to various types of errors, primarily errors inherent in all retrospective surveys, including: 38 | Fertility • Underreporting of births, in particular, the omission of children living elsewhere and children who died very young (a few days or hours after birth), which can result in under- estimation of fertility levels. • Misreporting of date of birth and/or age, in particular, the tendency to round off ages or year of birth, which can result in under- or overestimation of fertility at certain ages and/or for certain periods. • Selective survival bias or selectivity effect: the women surveyed are those who have survived. Assuming that the fertility of women who died prior to the survey differs from the fertility of the survivors, the fertility levels obtained by the survey may be slightly biased. Finally, for the men’s survey, as for the women’s survey, information was gathered concerning total fertility by asking men a series of questions, including: the number of children they had, the gender of each child, the number of children living with them, the number living elsewhere, the number of children who had died, and the number still living. However, the men were not asked to provide a complete birth history. 4.1 FERTILITY LEVELS AND DIFFERENTIALS Current fertility levels are measured in terms of age-specific fertility rates (ASFRs) and the total fertility rate (TFR). ASFRs are calculated by dividing the number of births in each age group into the total number of women for that age group. The TFR, a common measurement of current fertility, is the average of all of the ASFRs. It corresponds to the number of children the average woman would bear in her lifetime if fertility rates were to remain constant at the level prevailing during the period under consideration, in this case, the three years preceding the survey. Table 4.1, illustrated by Figure 4.1, indicates that, at the national level, general fertility rates (GFRs) by age group follow the classic pattern of high fertility countries. In Rwanda, this pattern is characterized by relatively high early fertility (42 births per 1,000 for women age 15-19), followed by a rapid increase to very high levels for women age 20-24 (235 per 1,000), 25-29 (305 per 1,000), and 30-34 (273 per 1,000). This high fertility is sustained over a very long period (211 per 1,000 at age 35-39), before declining precipitously at the very end of the childbearing years (32 per 1,000 at age 45-49). These data show that the fertility of Rwandan women remains very high: at the end of her childbearing years, a Rwandan woman has an average of 6.1 children, nearly identical to the TFR of the 1992 RDHS-I (6.2), and even slightly higher than the TFR of the 2000 RDHS-II (5.8). However, the 1994 genocide seems to have had the effect of slowing the significant decline in fertility observed since the National Fertility Survey of 1983 (TFR of 8.5). The data in Table 4.1 show clear differentials in fertility by residence: women in urban areas have lower fertility than those in rural areas. The TFR, estimated at 6.1 children per woman for the country as a whole, ranges from 4.9 in urban areas to 6.3 in rural areas. This means that, if Table 4.1 Current fertility Age-specific and cumulative fertility rates, the general fertility rate, and the crude birth rate for the three years preceding the survey, by urban-rural residence, Rwanda 2005 Residence Age Urban Rural Total 15-19 35 43 42 20-24 172 249 235 25-29 269 313 305 30-34 228 283 273 35-39 170 218 211 40-44 90 121 117 45-49 17 34 32 TFR 4.9 6.3 6.1 GFR 152 198 190 CBR 39.8 43.8 43.2 Note: Rates for age group 45-49 may be slightly biased because of truncation. TFR: Total fertility rate for ages 15-49, expressed per woman GFR: General fertility rate (births divided by the number of women age 15-44), expressed per 1,000 women CBR: Crude birth rate, expressed per 1,000 popu- lation Fertility | 39 current fertility levels were to remain constant, by the end of her childbearing years a woman living in a rural area would have an average of 1.4 children more than a woman living in an urban area. This differential in fertility levels is seen at all ages (Figure 4.1). Childbearing begins very early for women in rural areas: 43 per 1,000 for women age 15 to 19, compared with 35 per 1,000 for this age group in urban areas. At ages 20 to 24, 1,000 women in rural areas give birth to an average of 249 children, compared with 172 for women in urban areas. However, women reach their peak fertility between the ages of 25 and 29 in both rural (313 births per 1,000) and urban (269 births per 1,000) areas. Table 4.1 shows the crude birth rate (CBR), or average number of live births annually in the total population, estimated at 43 per 1,000 for the country as a whole, and the general fertility rate (GFR), that is, the average number of live births per 1,000 women of reproductive age (15-44), estimated here at 190 per 1,000. Like the TFR, these two indicators vary significantly by residence. Rural areas have a GFR of 198 per 1,000, which means that 1,000 women in rural areas are giving birth to an average of 46 more children annually than their urban counterparts (GFR of 152 per 1,000). Similarly, the CBR for rural areas (44 per 1,000) is 4 points higher than the CBR for urban areas (40 per 1,000). Table 4.2 presents fertility rates by background characteristic. The TFR varies considerably by province, ranging from a high of 6.6 children per woman in the West province to a low of 4.3 children per woman in the City of Kigali. In other words, women in the West province have an average of 2.3 more children than women in the City of Kigali. The TFR is strongly correlated with level of educational attainment, varying from a low of 4.3 children for women with secondary educations or higher, to 6.9 for women with no education. This means that a woman with no education (6.9) has an average of 0.8 more children than a woman who has attended primary school (6.1), and an average of 2.6 more children than a woman who has attended secondary school or higher (4.3). Figure 4.1 Age-Specific Fertility Rates, by Residence RDHS 2005 , , , , , , , + + + + + + + * * * * * * * 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 Births per 1,000 women Urban Rural Total* + , 40 | Fertility Table 4.2 and Figure 4.2 show the mean number of live births for women age 40 to 49. This figure is an indicator of completed, or cumulative fertility. Unlike the TFR, which measures the current or recent fertility of women age 15 to 49, cumulative fertility shows the past fertility of women surveyed at the end of their childbearing years. In a population whose fer- tility does not change, the cumulative fertility rate more or less coincides with the TFR. But TFRs that are lower than the mean number of children ever born to women at the end of their childbearing years indicate a downward trend in fertility. In Rwanda, the total cumulative fertility rate is estimated at 6.6 children. This is slightly higher than the TFR (6.1). The difference, though small, suggests a slight decline in fer- tility. In the 1992 RDHS-I, the difference be- tween the two was 1.5 children; in the 2000 RDHS-II, it was 1 child. As stated previously, the significant downward trend observed be- tween 1983 (ENF – National Fertility Survey) and 1992 (RDHS-I) did not continue. Table 4.2 Fertility by background characteristics Total fertility rate for the three years preceding the survey, percentage of women 15-49 currently pregnant, and mean number of children ever born to women age 40-49 years, by background characteristics, Rwanda 2005 Background characteristic Total fertility rate1 Percentage currently pregnant 1 Mean number of children ever born to women age 40-49 Residence Urban 4.9 6.3 5.8 Rural 6.3 8.3 6.7 Province Kigali city 4.3 6.9 5.9 South 5.6 7.6 6.1 West 6.6 7.9 7.1 North 6.4 7.9 6.7 East 6.5 9.0 6.6 Education No education 6.9 8.4 6.8 Primary 6.1 8.1 6.5 Secondary or higher 4.3 6.2 4.9 Wealth quintile Lowest 6.1 8.4 6.8 Second 6.3 7.7 6.6 Middle 6.7 8.2 6.5 Fourth 6.4 9.8 6.8 Highest 5.0 5.9 6.1 Total 6.1 8.0 6.6 1 Women age 15-49 years Figure 4.2 Total Fertility Rate and Mean Number of Children Ever Born to Women Age 40-49 6.1 4.9 6.3 6.9 6.1 4.3 6.1 6.3 6.7 6.4 5.0 6.6 5.8 6.7 6.8 6.5 4.9 6.8 6.6 6.5 6.8 6.1 RWANDA RESIDENCE Urban Rural EDUCATION No education Primary Secondary or higher WEALTH QUINTILE Lowest Second Middle Fourth Highest 0.0 2.0 4.0 6.0 8.0 Number of children per woman TFR Mean number of children ever born RDHS 2005 Fertility | 41 The fertility results by background characteristic show cumulative fertility rates above the TFR for all categories except women with no education, indicating that fertility is declining for all women, regardless of residence or province. However, the difference between cumulative fertility (number of children ever born) and the TFR is greater in the City of Kigali (1.6 children) and in the wealthiest households (1.1 children) than anywhere else. Table 4.2 shows the percentage of women who reported being pregnant at the time of the survey. At the national level, 8 percent of women reported being pregnant. This is likely to be an underestimate because women in the early stages of pregnancy may be unaware or unsure of their pregnancy status. Age, residence, culture, and/or beliefs may also affect a woman’s willingness to report her condition. In Rwanda, women generally declare their pregnancies only when their condition they becomes visible. For these reasons, the differentials in pregnancy rates shown here must be interpreted with a great deal of caution. It should be noted, however, that the findings are generally consistent with current fertility levels. In fact, the lowest pregnancy rates are observed for women living in the wealthiest households (6 percent), women with a secondary education or higher (6 percent), and women living in the City of Kigali (7 percent), which are the groups that also have the lowest fertility levels. 4.2 FERTILITY TRENDS Two national demographic data collection efforts are conducted regularly in Rwanda: the General Population and Housing Census and the Demographic and Health Survey (DHS). The Censuses of 1978, 1991, and 2002 gathered information on natural population dynamics and were used to estimate fertility levels for those years by asking questions about births that occurred in the 12 months preceding the survey. However, this method generally results in underestimates of fertility levels. The DHS surveys employ a more accurate method (women’s birth histories), which yield more reliable results. Yet the various RDHS surveys (1992, 2000, and 2005) and the Censuses of 1991 and 2002 have produced more or less similar results with respect to the TFR, which fluctuates around 6. This means that fertility has remained relatively stable in Rwanda since the RDHS-I. Table 4.3 shows age-specific fertility rates (ASFRs) for the three DHS surveys. Figure 4.3 examines past fertility trends based on the results of the 1992 RDHS-I, the 2000 RDHS-II, and the 2005 RDHS-III. The three ASFR curves follow a similar pattern: they increase rapidly with age and reach their peak between the ages of 25 and 29, then taper off steadily as they move toward the age group 45 to 49. It should be emphasized that the decline slows with this age group, demonstrating high levels of late fertility. However, the curve for the current survey (2005 RDHS-III) drops lower after age 40 than the other two curves, indicating a trend toward declining fertility in women of these generations. Table 4.3 Trends in fertility Age-specific fertility rates (per 1,000 women) and total fertility rates, 1992 RDHS, 2000 RDHS, and 2005 RDHS Age group 1992 RDHS 2000 RDHS 2005 RDHS 15-19 60 52 42 20-24 227 240 235 25-29 294 272 305 30-34 270 257 273 35-39 214 190 211 40-44 135 123 117 45-49 46 33 32 TFR 6.2 5.8 6.1 Note: Age-specific fertility rates are per 1,000 women. 42 | Fertility The data collected in the RDHS-III were used to track fertility trends for five-year periods preceding the survey based on women’s ASFRs (Table 4.4 and Figure 4.4). Fertility rates declined fairly steadily between the earliest period (15-19 years prior to the survey) and the most recent period, except for the youngest age groups (15-19 and 20-24) in the period 5-9 years preceding the survey (1996- 2000). The slight increase in the TFR in 2005 seems to be the result of an increase in fertility among women age 20 to 35. In other words, fertility among teenagers (age 15-19) has been declining steadily from one survey to the next, as has the fertility of women age 40 and over, especially in the recent periods. The ASFRs for the RDHS-III were used to calculate the TFR for women age 15 to 34—when the greatest number of births occur—for each five-year period. These data, presented in Figure 4.5, were positioned in relation to the central year of each period for which fertility was calculated. This figure also includes similar data from the RDHS-I (by four-year periods) and the RDHS-II (by five-year periods). The data reveal no general trends and no significant changes in fertility levels in Rwanda. In addition, there have been no changes in the factors that generally affect fertility. On the contrary, use of contraception, although it has risen since 2000, has not reached 1992 levels. There are no changes in median age at first birth, or in the other determinants influencing exposure to the risk of pregnancy: age at first marriage has remained relatively stable since 1992, and age at first sexual intercourse has risen only slightly. Table 4.4 Trends in age-specific fertility rates Age-specific fertility rates for five-year periods preceding the survey, by mother's age at the time of the birth, Rwanda 2005 Number of years preceding survey Age group 0-4 5-9 10-14 15-19 15-19 44 71 53 62 20-24 232 264 236 257 25-29 292 310 321 338 30-34 261 283 289 [334] 35-39 207 232 [259] - 40-44 118 [166] - - 45-49 [34] - - - Note: Age-specific fertility rates are per 1,000 women. Estimates in brackets are truncated. Figure 4.3 Trends in Age-Specific Fertility Rates, Rwanda 1992, 2000, and 2005 , , , , , , , ( ( ( ( ( ( ( # # # # # # # 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 Births per 1,000 women 1992 RDHS 2000 RDHS 2005 RDHS# ( , Fertility | 43 Figure 4.4 Age-Specific Fertility Rates for Five-Year Periods Preceding the Survey + + + + $ $ $ $ $ ! ! ! ! ! ! # # # # # # # 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 400 Births per 1,000 women Years preceding survey 0-4 years 5-9 years 10-14 years 15-19 years# ! $ + RDHS 2005 Figure 4.5 Trends in the Total Fertility Rate among Women Age 15-34, Rwanda 1992, 2000, and 2005 ( ( ( ( # # # # & & & & 79 83 87 88 91 93 94 98 03 Central year of estimates 0 1 2 3 4 5 6 Children per woman 1992 RDHS 2000 RDHS 2005 RDHS& # ( 2005 RDHS 1992 RDHS 2000 RDHS 44 | Fertility 4.3 PARITY AND PRIMARY INFERTILITY Women’s average parity by age group is calculated on the basis of the total number of children ever born in their lifetime. Table 4.5.1 presents these parities for all women and for currently married women. For all women, parity increases steadily and rapidly with age: from an average of 0.04 children at age 15 to 19, parity increases to 0.8 children at age 20 to 24, and to 7.0 children at age 45 to 49, the end of the childbearing years. In addition, the distribution of women by number of children ever born shows relatively late childbearing. Only 3.3 percent of women under the age of 20 have given birth to at least one child. Even at ages 20 to 24, less than half the women (46 percent) have given birth to at least one child, and only 16 percent of the women in this age group have given birth twice. The fertility level accelerates between age 25 and 29: more than one-fifth (23 percent) of women in this age group have given birth to at least 3 children. However, nearly one-quarter of the women in the 30-34 age group have had at least 4 births. Finally, at age 45 to 49, the end of the reproductive period, 16 percent of women have given birth to at least 10 children. Table 4.5.1 Children ever born and living : women Percent distribution of all women and currently married women by number of children ever born, and mean number of children ever born and mean number of living children, according to age group, Rwanda 2005 Number of children ever born Age group 0 1 2 3 4 5 6 7 8 9 10+ Total Number of women Mean number of children ever born Mean number of living children ALL WOMEN 15-19 96.7 3.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 2,585 0.04 0.03 20-24 54.2 22.8 16.1 5.5 1.2 0.1 0.0 0.0 0.0 0.0 0.0 100.0 2,354 0.77 0.67 25-29 16.6 15.5 21.1 22.9 15.8 5.7 1.6 0.8 0.0 0.0 0.0 100.0 1,738 2.33 1.96 30-34 5.6 8.5 12.1 18.2 23.4 17.1 8.5 4.6 1.3 0.4 0.2 100.0 1,466 3.66 3.01 35-39 3.1 3.9 9.3 10.1 13.9 16.0 17.8 13.4 7.8 2.7 2.0 100.0 1,134 4.96 3.92 40-44 2.9 2.0 3.9 6.6 9.9 11.2 14.4 17.6 12.4 9.3 9.9 100.0 1,135 6.20 4.89 45-49 2.2 1.6 2.0 4.5 5.1 11.0 12.8 16.3 14.0 14.6 15.9 100.0 910 7.02 5.24 Total 37.4 9.6 9.7 9.1 8.5 6.7 5.6 5.1 3.3 2.4 2.5 100.0 11,321 2.68 2.14 CURRENTLY MARRIED WOMEN 15-19 42.9 50.5 4.2 2.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 65 0.66 0.57 20-24 14.7 37.4 32.7 12.2 2.6 0.3 0.1 0.0 0.0 0.0 0.0 100.0 980 1.52 1.32 25-29 4.8 13.7 23.7 27.1 19.8 7.7 2.2 1.1 0.0 0.0 0.0 100.0 1,254 2.80 2.36 30-34 1.2 5.2 10.3 19.0 26.1 19.8 10.3 5.8 1.7 0.4 0.3 100.0 1,112 4.08 3.38 35-39 1.0 1.8 6.0 7.1 14.2 16.3 20.8 16.6 10.1 3.4 2.8 100.0 807 5.55 4.41 40-44 1.4 1.7 1.9 5.0 6.5 8.4 14.5 18.7 15.5 12.5 13.9 100.0 739 6.87 5.46 45-49 1.3 1.4 0.6 3.5 3.3 7.8 12.4 13.8 16.5 18.6 20.8 100.0 554 7.63 5.74 Total 4.9 12.1 14.5 14.3 13.5 10.1 8.8 7.7 5.6 4.1 4.4 100.0 5,510 4.24 3.42 The results for married women do not vary significantly from the results for all women, except for the younger age groups. More than half of married women between the ages of 15 and 19 (57 percent) have had at least one child, compared with 3.3 percent of all women. However, as for all women, the fertility of young married women age 15 to 19 remains relatively low: only 7 percent of the women in this age group have had two or more children. Fertility | 45 Even at age 20-24, when more than half of women are still never-married (53 percent: see Marital Status in Chapter 6), there is still a wide gap between the proportion of married women who have had at least one child (85 percent) and the proportion of all women in this age group who have had at least one child (46 percent). Women who remain childless voluntarily are relatively rare in Rwanda, where the population is still very pro-natal (see Chapter 7, Fertility Preferences). For this reason, zero parity among married women age 35 to 49 would be an indicator of total or primary infertility. In Rwanda, only 1.2 percent of married women age 35 to 49 years (when the arrival of a first child is unlikely) have never had a child and can be considered infertile. This shows that the level of primary infertility has remained stable at low levels since 1983, when it was calculated at 1.5 percent. The percentage was as low as 0.7 percent in the RDHS-I, and was 1.2 percent in the RDHS-II. It should be noted that the level of primary infertility observed in Rwanda in 2005 is lower than the level found in some sub-Saharan countries such as Cameroon (3.6 percent in the 2004 EDSC), but is similar to that of other countries in this region such as Burkina Faso (1 percent in the 2003 EDSBF). The average number of children by age group for men is calculated on the basis of the total number of children ever born to men in their lifetime. Table 4.5.2 shows the number of children ever born for all men and for married men. Table 4.5.2 Children ever born and living : men Percent distribution of all men and currently married men by number of children ever born, and mean number of children ever born and mean number of living children, according to age group, Rwanda 2005 Number of children ever born Age group 0 1 2 3 4 5 6 7 8 9 10+ Total Number of men Mean number of children ever born Mean number of living children ALL MEN 15-19 99.8 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 1,102 0.00 0.00 20-24 83.5 9.9 4.8 1.4 0.3 0.0 0.0 0.1 0.0 0.0 0.0 100.0 946 0.26 0.23 25-29 39.9 17.8 17.9 16.0 5.2 2.0 0.8 0.4 0.1 0.0 0.0 100.0 632 1.40 1.22 30-34 15.7 11.2 15.5 19.5 18.7 12.1 3.2 2.7 0.7 0.7 0.0 100.0 509 2.86 2.38 35-39 8.2 6.1 9.1 14.6 16.8 14.6 13.3 9.4 3.5 3.0 1.5 100.0 442 4.26 3.50 40-44 2.9 2.3 4.7 8.5 11.8 13.1 13.6 16.8 11.1 7.6 7.5 100.0 404 5.90 4.72 45-49 1.7 1.1 3.7 5.7 3.8 8.0 15.4 13.1 12.4 10.8 24.4 100.0 378 7.43 5.79 50-54 1.3 0.8 1.6 2.1 1.6 5.9 7.6 12.0 15.8 13.0 38.4 100.0 260 8.63 6.50 55-59 0.8 1.2 0.7 0.0 3.8 3.8 3.3 14.0 13.3 15.8 43.3 100.0 147 9.28 6.59 Total 47.3 6.4 6.6 7.0 5.8 5.0 4.5 4.7 3.6 3.0 6.1 100.0 4,820 2.76 2.18 CURRENTLY MARRIED MEN 15-19 * * * * * * * * * * * * 2 * * 20-24 20.2 44.3 25.6 7.4 1.9 0.0 0.0 0.5 0.0 0.0 0.0 100.0 173 1.29 1.15 25-29 7.6 25.5 27.9 25.5 8.3 3.1 1.2 0.7 0.1 0.0 0.0 100.0 394 2.20 1.92 30-34 3.7 11.2 17.7 22.8 22.0 14.1 3.5 3.3 0.8 0.8 0.0 100.0 429 3.32 2.75 35-39 3.4 5.5 9.3 14.4 17.5 16.1 14.7 10.4 3.6 3.4 1.7 100.0 400 4.56 3.77 40-44 1.0 1.7 4.8 8.1 11.2 13.9 13.6 17.8 11.8 8.1 8.0 100.0 381 6.11 4.90 45-49 0.2 0.6 3.1 4.9 3.3 8.2 14.7 13.7 12.7 11.8 26.6 100.0 346 7.77 6.07 50-54 0.3 0.5 1.0 2.3 1.4 4.8 7.0 12.2 15.6 13.8 41.1 100.0 235 8.90 6.79 55-59 0.5 0.8 0.7 0.0 4.0 3.4 3.5 13.5 13.2 15.9 44.5 100.0 139 9.40 6.69 Total 4.0 10.4 12.0 12.9 10.5 9.4 8.1 8.9 6.5 5.7 11.5 100.0 2,500 5.10 4.06 Note : An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed 46 | Fertility For all men, the average number of children ever born and living increases steadily and rapidly with age, from an average of 0.3 children at age 20 to 24, to 1.4 children at age 25 to 29, to 2.9 children at age 30 to 34, and to 9.3 children at age 55 to 59. For married men, the average number of children is higher in the younger age groups. A comparison of these results with those for married women shows that the average number of children increases more rapidly for married women than for married men. However, in the older age groups (45 to 49), the average number of children ever born is much higher for married men than married women. 4.4 BIRTH INTERVALS Examination of birth intervals, defined as the length of time between two successive live births, is important not only for their impact on the health status of both mother and child, but for their role in fertility analysis and the design of reproductive health programs. Currently, short birth intervals (less than 24 months) are considered harmful to the health and nutritional status of children, increasing their risk of death. In addition, short birth intervals diminish a woman’s physiological capacity, exposing her to a greater risk of complications during and after pregnancy (miscarriage, eclampsia, etc.), and are also associated with high cumulative fertility. Table 4.6 shows the distribution of non-first births in the five years preceding the survey by number of months since the preceding birth, according to background characteristics. Table 4.6 shows that 8 percent of births occur less than 18 months after the preceding birth and that 15 percent of children are born between 18 and 24 months after the birth of their immediately older sibling. In 23 percent of all cases, the birth interval is less than two years. However, a large proportion of births (41 percent) occur between 2 and 3 years after the preceding birth, and more than one-third of all children (36 percent) are born three years or more after the birth of their immediately older sibling. The mean duration of the birth interval is slightly more than two and a half years (31.3 months), which means that half of all births take place 31.3 months after the preceding birth. With respect to age, birth intervals are shorter for younger women; that is, the younger the woman, the shorter the birth interval: the mean duration increases from 27.7 months at age 20 to 29, to 37.1 months at age 40 to 49. The results also show a significant increase in the length of birth intervals associated with birth order, from 29.9 months for birth orders 2-3, to 32.7 months for birth orders 7 and higher. However, differentials by gender are not significant (31.6 months for boys; 30.9 months for girls). Survival of the preceding child is an important factor. When the preceding child has died, the birth interval between that birth and the next birth is a median of 26.4 months. When the preceding child is living, the birth interval is a median of 32.1 months, or approximately six months later than the birth following the death of the preceding sibling. The median interval between births is lowest in rural areas (31.4 months, compared with 29.9 in urban areas). In 2005, the differential between rural and urban areas was 1.5 months; in 2000, it was 3.2 months. This is not a very significant difference; it is to be expected that women in urban areas, who have greater access to family planning services, will have much longer birth intervals than women in rural areas. With respect to provinces, the birth interval varies from a low of 30.1 months in the City of Kigali to a high of 32.6 months in the South province. Fertility | 47 Table 4.6 Birth intervals Percent distribution of non-first births in the five years preceding the survey by number of months since preceding birth, according to background characteristics, Rwanda 2005 Months since preceding birth Background characteristic 7-17 18-23 24-35 36-47 48+ Total Number of non-first births Median number of months since preceding birth Mother’s age 15-19 * * * * * * 7 * 20-29 12.2 20.1 43.4 14.6 9.7 100.0 2,578 27.7 30-39 6.2 13.7 41.0 21.3 17.7 100.0 3,243 32.5 40-49 4.2 8.6 34.1 24.2 29.0 100.0 1,245 37.1 Birth order 2-3 10.6 17.2 38.9 16.3 17.0 100.0 2,880 29.9 4-6 6.3 14.0 42.3 20.7 16.7 100.0 2,753 31.9 7+ 6.3 13.3 41.4 22.7 16.4 100.0 1,441 32.7 Sex of preceding birth Male 7.9 15.0 40.4 19.5 17.2 100.0 3,611 31.6 Female 8.2 15.4 41.0 19.2 16.2 100.0 3,463 30.9 Survival of preceding birth Living 5.4 14.2 42.7 20.5 17.1 100.0 5,834 32.1 Dead 20.4 19.5 31.1 14.1 15.0 100.0 1,240 26.4 Residence Urban 11.5 16.4 36.3 15.7 20.1 100.0 968 29.9 Rural 7.5 15.0 41.4 19.9 16.2 100.0 6,106 31.4 Province Kigali city 12.7 16.3 32.6 16.6 21.9 100.0 486 30.1 South 6.9 14.9 39.1 21.1 18.0 100.0 1,708 32.6 West 7.9 16.6 41.7 19.2 14.6 100.0 1,874 30.2 North 6.8 13.4 46.3 18.7 14.8 100.0 1,445 31.8 East 9.1 15.1 38.6 19.1 18.1 100.0 1,562 31.3 Education No education 7.3 15.0 37.1 21.1 19.5 100.0 2,128 32.7 Primary 8.3 14.9 43.0 18.9 15.0 100.0 4,368 30.8 Secondary or higher 9.4 17.6 37.1 15.7 20.2 100.0 578 30.1 Wealth quintile Lowest 6.2 13.0 40.4 20.9 19.6 100.0 1,513 32.8 Second 6.8 13.0 42.8 20.4 17.0 100.0 1,474 32.1 Middle 8.4 16.1 40.5 20.0 15.0 100.0 1,465 30.9 Fourth 8.4 14.7 41.9 19.4 15.7 100.0 1,395 31.0 Highest 11.1 19.9 37.3 15.4 16.3 100.0 1,226 28.8 Total 8.1 15.2 40.7 19.3 16.7 100.0 7,074 31.3 Note: First-order births are excluded. The interval for multiple births is the number of months since the preceding pregnancy that ended in a live birth. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Regarding mother’s level of educational attainment, birth intervals for women with no education are longer (32.7 months) than birth intervals for women with a secondary education or higher (30.1). This is explained, among other things, by the fact that the median duration of breastfeeding is longer for women with no education than for women with some education (see Table 10.3). Household wealth data follow a similar pattern: 50 percent of children in the poorest quintile are born 32.8 months after the birth of their immediately older sibling, or 4 months later than children in the richest quintile (28.8). 48 | Fertility 4.5 AGE AT FIRST BIRTH The age at which childbearing begins is an important demographic indicator because it has a direct bearing on a women’s cumulative fertility, particularly when there is little or no contraceptive use. The younger a woman is when she begins childbearing, the greater her likelihood of having many children. At the same time, having children at too young an age can have negative repercussions on the mother’s health and can put her children at risk of dying. Table 4.7 shows the distribution of women by age at first birth and median age at first birth by age at the time of the survey. The results show that median age at first birth has remained practically unchanged from one generation to the next (from a low of 21.7 to a high of 22.2) and no trends indicate a rise or fall in this median age. Table 4.7 Age at first birth Among all women, the percentage who first gave birth by exact age, and median age at first birth, by current age, Rwanda 2005 Percentage who gave birth by exact age: Age group 15 18 20 22 25 Percentage who have never given birth Number of women Median age at first birth 15-19 0.2 na na na na 96.7 2,585 a 20-24 0.6 7.8 22.3 na na 54.2 2,354 a 25-29 0.6 10.2 31.1 53.4 75.4 16.6 1,738 21.7 30-34 1.3 8.7 24.1 48.3 77.8 5.6 1,466 22.1 35-39 1.3 10.0 25.8 48.2 73.9 3.1 1,134 22.2 40-44 0.9 8.8 28.8 51.1 77.5 2.9 1,135 21.9 45-49 1.1 9.0 25.8 50.6 80.3 2.2 910 22.0 na = Not applicable a = Omitted because less than 50 percent of women had a birth before reaching the beginning of the age group. Table 4.8 shows a median age at first birth of 22.0 years for women age 25 to 49; this is identical to the median age observed for women the same age in the 2000 RDHS-II. Table 4.8 shows median age at first birth according to various socioeconomic characteristics. The first child arrives at a younger age for women in rural areas (21.9 years) than for those in urban areas (22.4 years). The South province has the highest median age at first birth (22.9 years), followed by the City of Kigali (22.4 years). In the other provinces, median age at first birth varies from a low of 21.2 years in the East province to a high of 21.7 years in the West province. Women’s level of educational attainment affects the median age at first birth: women with no education (21.4 years) and women with primary education (22.0 years) have a lower median age at first birth than women with secondary or higher education (23.9 years). Results by household wealth show that the first birth occurs later among women in the richest quintile (22.7 years) than among those in the poorest quintile (21.8 years). Fertility | 49 Table 4.8 Median age at first birth by background characteristics Median age at first birth among women age 25-49 years, by current age and background characteristics, Rwanda 2005 Current age Background characteristic 25-29 30-34 35-39 40-44 45-49 Women age 25-49 Residence Urban 22.2 22.5 23.0 22.1 21.6 22.4 Rural 21.6 22.1 22.0 21.9 22.0 21.9 Province Kigali city 22.4 23.0 23.7 21.2 20.7 22.4 South 22.9 23.1 23.4 22.8 22.5 22.9 West 21.2 22.1 21.4 21.6 22.1 21.7 North 21.2 21.4 22.2 21.7 22.1 21.6 East 21.0 21.3 21.3 21.6 21.1 21.2 Education No education 20.8 21.3 21.2 21.3 21.8 21.4 Primary 21.6 22.2 22.3 22.1 22.1 22.0 Secondary or higher 23.9 23.6 24.5 24.0 22.9 23.9 Wealth quintile Lowest 21.4 21.9 22.2 21.5 21.8 21.8 Second 21.0 22.0 21.2 21.9 22.2 21.7 Middle 21.9 21.9 22.2 21.8 21.9 21.9 Fourth 21.7 22.1 21.9 22.1 21.9 21.9 Highest 22.3 22.9 23.2 22.6 21.9 22.7 Total 21.7 22.1 22.2 21.9 22.0 22.0 4.6 TEENAGE FERTILITY Teenage fertility is an important demographic factor for many reasons. First, children born to very young mothers run a greater risk of illness and death. Second, teenage mothers are more likely to suffer complications during pregnancy and less likely to treat them, exposing them to greater risk of complications during delivery and greater risk of dying for reasons related to childbearing. Third, early childbearing seriously affects a woman’s ability to pursue an education, thereby limiting her job opportunities. In Rwanda, teenagers make up 23 percent of all women of childbearing age, but only 3 percent have had at least one birth. Table 4.9 shows the proportion of teenagers who have already had one or more children, as well as those currently in their first pregnancy. Together these two subgroups make up the proportion of teenagers who have already begun childbearing: 4 percent of young women between the ages of 15 and 19 (3 percent are already mothers; 0.8 percent are pregnant for the first time). At age 15, 0.3 percent of women have begun childbearing, but the percentage increases steadily and rapidly with age: at age 17, 2 percent of women have already had at least one child or are pregnant for the first time. At age 19, this proportion reaches 13 percent, of which 10 percent are those who have already had at least one child. 50 | Fertility Table 4.9 Teenage pregnancy and motherhood Percentage of women age 15-19 who are mothers or pregnant with their first child, by background characteristics, Rwanda 2005 Percentage who are: Background characteristic Mothers Pregnant with first child Percentage who have begun childbearing Number of women Age 15 0.3 0.0 0.3 536 16 1.0 0.0 1.0 591 17 2.1 0.3 2.4 506 18 5.0 1.2 6.2 520 19 9.5 3.3 12.8 432 Residence Urban 4.1 0.9 5.0 472 Rural 3.1 0.8 3.9 2,113 Province Kigali city 6.4 0.6 7.0 277 South 2.6 1.2 3.8 648 West 3.2 0.8 4.0 686 North 1.3 0.4 1.8 453 East 4.4 0.9 5.3 521 Education No education 9.7 0.3 10.0 222 Primary 2.7 0.8 3.5 2,182 Secondary or higher 2.3 1.8 4.1 181 Wealth quintile Lowest 2.5 0.4 2.9 557 Second 3.8 0.9 4.7 509 Middle 3.0 0.7 3.7 444 Fourth 3.3 1.6 4.8 477 Highest 3.8 0.8 4.6 599 Total 3.3 0.8 4.1 2,585 Table 4.9 shows that teenagers residing in urban areas begin childbearing much earlier than their rural counterparts. In fact, 5 percent of teenagers in urban areas have begun childbearing, compared with 4 percent in rural areas. Similar differences are seen between provinces: the proportion of teenagers who have begun childbearing varies from a low of 1.8 percent in the North province to a high of 7 percent in the City of Kigali. Early childbearing occurs more frequently among teenagers with no education (10 percent) than among those who are educated (4 percent). However, differentials by wealth quintile are not significant: the proportion of teenagers who have begun childbearing varies from 3 percent in the poorest quintile to 5 percent in the two highest quintiles. These narrow differentials mean that standard of living has no bearing on the behavior of Rwandan teenagers with regard to procreation. Finally, it should be noted that the proportion of teenagers who have begun childbearing has decreased from 11 percent in 1992, to 7 percent in 2000, to 4 percent in 2005. Family Planning | 51 FAMILY PLANNING 5 During the RDHS-III, men and women were asked to name any means of contraception they knew about. They were then asked questions concerning their own past and/or current use of contraception, and their intended use in the future. Finally, men and women were asked if they knew where to procure the various methods of contraception. 5.1 KNOWLEDGE OF CONTRACEPTION The use of contraception presupposes prior knowledge of at least one contraceptive method, as well as a source of contraceptive supply. The different methods covered by the questionnaire fall into two categories: • Modern methods. These include female sterilization, male sterilization, the pill, the IUD (intrauterine device), injectables, implants (Norplant), the male condom, the female condom, the diaphragm, vaginal methods (spermicides, foams and jellies), emergency contraception, the lactational amenorrhea method (LAM), and the standard days method (SDM)/beads. • Traditional methods. These include the rhythm or periodic abstinence method, withdrawal, and so-called “folk” methods such as herbs, etc. As previously indicated, information concerning knowledge of contraceptive meth- ods was gathered in two ways: first, each respondent was asked to spontaneously name the contraceptive methods he or she knew about. If a respondent failed to mention a par- ticular method covered by the questionnaire, the interviewer briefly described the method and recorded whether or not the respondent had heard of it. A method was considered to be known by a respondent if he or she mentioned it spontaneously or recognized it after it was described. The results show that knowledge of family planning is nearly universal in Rwanda: 95 percent of women age 15-49 reported having knowledge of at least one method of contraception (Table 5.1.1). In general, women are more familiar with modern methods than with traditional or folk methods: 95 percent of women have heard of at least one modern method; 67 percent have heard of at least one traditional method; and 0.2 percent know of a folk method. Table 5.1.1 Knowledge of contraceptive methods: women Percentage of all women, of currently married women, and of sexually active unmarried women who know any contraceptive method, by specific method, Rwanda 2005 Method All women Currently married women Sexually active unmarried women Any method 94.9 97.9 94.6 Any modern method 94.5 97.5 94.6 Female sterilization 62.7 71.1 68.8 Male sterilization 23.4 30.1 20.0 Pill 77.9 89.4 84.8 IUD 31.3 39.7 27.1 Injectables 80.2 92.0 82.7 Implants 38.2 49.7 34.8 Male condom 88.7 91.0 92.0 Female condom 37.7 40.6 36.4 Diaphragm 3.8 5.0 2.2 Foam/jelly 5.6 6.3 5.6 Lactational amenorrhea method (LAM) 35.4 47.2 34.2 Emergency contraception 7.7 9.3 7.5 Standard days method/beads 33.7 42.9 35.4 Any traditional method 67.0 79.7 75.0 Rhythm or periodic abstinence 58.8 68.9 62.9 Withdrawal 47.1 63.3 58.7 Local traditional method 0.2 0.3 0.0 Mean number of methods known 6.3 7.5 6.5 Number of women 11,321 5,510 136 52 | Family Planning Knowledge of contraceptive methods among sexually active unmarried women is very high (95 percent for any method), although it has declined since 2000, when it was 100 percent. Knowledge of any contraceptive method among married women has increased slightly, from 97 percent in 2000 to 98 percent in 2005. With respect to specific methods, Table 5.1.1 indicates that male condoms constitute the method best known by all women (89 percent), followed by injectables (80 percent), and the pill (78 percent). Diaphragms constitute the least known method for all categories of women (4 percent). With respect to traditional methods, 69 percent of married women reported knowing about the rhythm or periodic abstinence method, and 63 percent had heard of withdrawal. The mean number of methods known is highest among married women (7.5). Table 5.1.2 shows knowledge of con- traception among men. As with women, knowledge of contraceptive methods is high: 98 percent of men reported having knowledge of at least one modern method, compared with 77 percent for traditional methods. With respect to specific methods, Table 5.1.2 shows that, like women, men are most familiar with the male condom (97 percent), followed by injectables among all men and currently married men (at least 75 percent). The pill is the second most commonly known method among sexually active unmarried men (85 percent). Like women, few men have heard of the diaphragm (6 percent). With respect to traditional methods, rhythm or periodic abstinence and withdrawal methods are known in roughly the same percentages (65 percent and 63 percent). 5.2 USE OF CONTRACEPTION Increasing the use of contraception is the ultimate aim of family planning programs, and contraceptive prevalence serves as a key measure for assessing the success of such programs. RDHS-III data have been used to estimate “ever use” of contraception and the current level of use, that is, at the time of the survey. 5.2.1 Ever Use of Contraception Women who said that they had heard of a contraceptive method were asked if they had ever used that method. This information was used to measure the level of contraceptive use at any time in the woman’s reproductive life (ever use), according to specific method. Table 5.2 presents the results for all women, married women, and sexually active unmarried women. Table 5.1.2 Knowledge of contraceptive methods: men Percentage of all men, of currently married men, and of sexually active unmarried men who know any contraceptive method, by specific method, Rwanda 2005 Method All men Currently married men Sexually active unmarried men Any method 98.1 99.5 100.0 Any modern method 98.0 99.5 100.0 Female sterilization 71.3 82.4 79.5 Male sterilization 34.3 43.2 38.3 Pill 70.7 83.4 84.5 IUD 36.6 47.1 43.4 Injectables 75.1 87.8 80.9 Implants 34.1 46.6 39.8 Male condom 96.6 98.0 100.0 Female condom 51.9 56.5 67.1 Diaphragm 5.6 7.2 5.3 Foam/jelly 10.7 12.6 16.3 Lactational amenorrhea method (LAM) 27.1 37.3 22.8 Emergency contraception 13.2 16.4 21.3 Standard days method/beads 39.6 50.3 41.6 Any traditional method 76.6 92.1 92.0 Rhythm or periodic abstinence 65.4 79.9 81.3 Withdrawal 63.3 81.9 70.5 Mean number of methods known 7.0 8.3 7.9 Number of men 4,820 2,500 57 T ab le 5 .2 E ve r u se o f c on tra ce pt io n P er ce nt ag e of a ll w om en , cu rr en tly m ar rie d w om en , an d se xu al ly a ct iv e un m ar rie d w om en w ho h av e ev er u se d an y co nt ra ce pt iv e m et ho d, b y sp ec ifi c m et ho d an d ag e, Rw an da 2 00 5 M od er n m et ho d Tr ad iti on al m et ho d A ge gr ou p An y m et ho d An y m od er n m et ho d Fe m al e st er ili z- at io n M al e st er ili z- at io n Pi ll In je ct - ab le s M al e co nd om LA M Em er - ge nc y co nt ra - ce pt io n St an da rd da ys m et ho d/ be ad s O th er m od er n m et ho ds An y tra di - tio na l m et ho d Pe rio di c ab st i- ne nc e W ith - dr aw al O th er N um be r of w om en AL L W O M EN 1 5- 19 1. 3 1. 2 0. 0 0. 0 0. 1 0. 0 1. 2 0. 1 0. 0 0. 0 0. 0 0. 3 0. 3 0. 1 0. 0 2, 58 5 2 0- 24 12 .3 7. 8 0. 0 0. 0 2. 3 2. 6 3. 3 0. 9 0. 1 0. 3 0. 0 6. 7 3. 6 4. 1 0. 0 2, 35 4 2 5- 29 27 .3 19 .3 0. 2 0. 0 6. 4 8. 8 4. 5 2. 9 0. 1 1. 6 0. 7 14 .3 8. 3 9. 5 0. 0 1, 73 8 3 0- 34 33 .1 22 .3 0. 4 0. 1 7. 9 12 .6 3. 4 3. 6 0. 0 2. 0 1. 1 17 .9 11 .6 9. 6 0. 1 1, 46 6 3 5- 39 36 .6 25 .7 0. 6 0. 0 9. 8 14 .3 3. 8 4. 1 0. 2 1. 8 1. 5 17 .9 11 .6 10 .0 0. 3 1, 13 4 4 0- 44 36 .3 27 .5 0. 6 0. 0 12 .3 17 .0 1. 5 2. 3 0. 0 1. 4 1. 7 14 .8 9. 9 8. 1 0. 0 1, 13 5 4 5- 49 34 .2 25 .8 1. 2 0. 2 10 .7 17 .1 0. 3 2. 2 0. 0 1. 4 1. 6 15 .6 9. 9 9. 4 0. 3 91 0 T ot al 21 .4 15 .2 0. 3 0. 0 5. 6 8. 0 2. 6 1. 9 0. 1 1. 0 0. 7 10 .5 6. 5 6. 1 0. 1 11 ,3 21 C U RR EN TL Y M A RR IE D W O M EN 1 5- 19 7. 9 7. 9 0. 0 0. 0 2. 9 0. 8 6. 0 1. 9 1. 9 1. 9 0. 0 1. 9 1. 9 1. 9 0. 0 65 2 0- 24 22 .7 12 .9 0. 0 0. 0 4. 7 6. 1 2. 9 2. 1 0. 1 0. 7 0. 1 13 .5 7. 1 8. 2 0. 0 98 0 2 5- 29 32 .4 22 .3 0. 3 0. 0 7. 8 11 .0 4. 0 3. 4 0. 1 2. 1 0. 9 17 .8 10 .0 12 .1 0. 0 1, 25 4 3 0- 34 37 .0 24 .3 0. 5 0. 1 8. 6 14 .1 2. 8 3. 8 0. 0 2. 4 1. 3 20 .1 12 .8 11 .2 0. 2 1, 11 2 3 5- 39 40 .8 28 .1 0. 8 0. 0 10 .9 15 .5 3. 0 5. 2 0. 0 2. 1 1. 7 21 .4 13 .7 11 .9 0. 4 80 7 4 0- 44 41 .6 31 .1 0. 8 0. 0 12 .8 19 .8 1. 8 2. 6 0. 0 2. 2 1. 8 18 .7 12 .4 10 .7 0. 0 73 9 4 5- 49 40 .2 29 .8 1. 5 0. 3 12 .3 19 .8 0. 2 2. 5 0. 0 2. 3 1. 5 19 .4 12 .8 11 .6 0. 4 55 4 T ot al 34 .6 23 .7 0. 5 0. 0 8. 9 13 .4 2. 8 3. 3 0. 1 1. 9 1. 1 18 .1 11 .1 10 .8 0. 1 5, 51 0 SE XU A LL Y AC TI V E U N M A RR IE D W O M EN 1 T ot al 28 .0 22 .3 0. 4 0. 0 6. 3 9. 0 14 .9 1. 6 0. 9 0. 0 0. 6 17 .4 10 .6 11 .2 0. 0 13 6 L A M = L ac ta tio na l a m en or rh ea m et ho d 1 W om en w ho h ad s ex ua l i nt er co ur se in th e m on th p re ce di ng th e su rv ey 53Family Planning | 54 | Family Planning The results show that 21 percent of women have used a method of contraception at some time. Modern methods were used more frequently than traditional methods (15 percent for modern; 11 percent for traditional) and, among the modern methods, injectables and the pill were used more frequently than other methods (8 percent for injectables, 6 percent for the pill; 3 percent for male condoms). Among the traditional methods, withdrawal and periodic abstinence were used in the same proportions (6 percent). Ever use of contraception is considerably higher among married women than all women: 35 percent of married women have used a method at some time, 24 percent a modern method, and 18 percent a traditional or folk method. For all women, the percentages are 21 percent, 15 percent, and 11 percent, respectively. Sexually active unmarried women have higher levels of ever use of contraception than all women, but lower levels of ever use than married women. The male condom was the method most frequently used by sexually active unmarried women (15 percent), followed by injectables (9 percent), and the pill (6 percent). Among traditional methods, these women opted for periodic abstinence and withdrawal in the same proportions as married women (11 percent for both methods). 5.2.2 Current Use of Contraception Table 5.3 shows current contraceptive prevalence. Women who were not pregnant and had heard of at least one contraceptive method were asked whether they were currently using any method of contraception to avoid pregnancy. The responses to this question were used to assess current contraceptive prevalence, that is, the proportion of women who were using a method of contraception at the time of the survey. Table 5.3 shows that, among all women age 15-49 who were not pregnant at the time of the survey, 10 percent were using at least one method of contraception, 6 percent were using a modern method, and 4 percent were using a traditional method. Results according to age show that prevalence is lowest among the youngest women, age 15 to 24 (7 percent at most), and the oldest women, age 45 to 49 (10 percent). The most frequently used modern method is injectables (2 percent). The rate of use for other methods remains very low. Periodic abstinence and withdrawal are the most frequently used traditional methods (2 percent each). Contraceptive prevalence among married women at the time of the survey was 17 percent for any method and 10 percent for any modern method. Seven percent of married women reported using a traditional method at the time of the survey. The most frequently used modern methods were injectables (5 percent) and the pill (2 percent) (Figure 5.1). Periodic abstinence (4 percent) and withdrawal (3 percent) were the most frequently used traditional methods. The variation in contraceptive prevalence by age found among all women is also found among married women: prevalence is lowest for younger women (3 percent for age 15 to 19) and older women (14 percent for age 45 to 49). T ab le 5 .3 C ur re nt u se o f c on tra ce pt io n P er ce nt d ist rib ut io n of a ll w om en , cu rr en tly m ar rie d w om en , an d se xu al ly a ct iv e un m ar rie d w om en b y co nt ra ce pt iv e m et ho d cu rr en tly u se d, a cc or di ng t o ag e, Rw an da 2 00 5 M od er n m et ho d Tr ad iti on al m et ho d A ge gr ou p An y m et ho d An y m od er n m et ho d Fe m al e st er ili z- at io n Pi ll In je ct - ab le s M al e co nd om LA M St an da rd da ys m et ho d/ be ad s O th er m od er n m et ho ds An y tra di - tio na l m et ho d Pe rio di c ab st i- ne nc e W ith - dr aw al N ot cu rr en tly us in g To ta l N um be r of w om en AL L W O M EN 1 5- 19 0. 4 0. 3 0. 0 0. 1 0. 0 0. 2 0. 0 0. 0 0. 0 0. 1 0. 1 0. 0 99 .6 10 0. 0 2, 58 5 2 0- 24 6. 5 4. 1 0. 0 1. 1 1. 4 1. 3 0. 2 0. 1 0. 0 2. 4 1. 4 1. 0 93 .5 10 0. 0 2, 35 4 2 5- 29 13 .9 8. 6 0. 2 2. 0 3. 9 1. 1 0. 7 0. 4 0. 3 5. 3 2. 7 2. 6 86 .1 10 0. 0 1, 73 8 3 0- 34 17 .0 10 .5 0. 4 2. 3 4. 8 1. 2 0. 9 0. 4 0. 4 6. 5 4. 2 2. 4 83 .0 10 0. 0 1, 46 6 3 5- 39 16 .2 10 .3 0. 6 2. 2 3. 9 1. 0 1. 3 0. 2 1. 0 5. 9 3. 5 2. 4 83 .8 10 0. 0 1, 13 4 4 0- 44 14 .2 7. 1 0. 6 1. 7 3. 3 0. 7 0. 1 0. 6 0. 1 7. 1 5. 2 1. 9 85 .8 10 0. 0 1, 13 5 4 5- 49 9. 7 4. 0 1. 2 0. 4 1. 7 0. 0 0. 1 0. 2 0. 2 5. 7 3. 7 2. 0 90 .3 10 0. 0 91 0 T ot al 9. 6 5. 6 0. 3 1. 3 2. 4 0. 8 0. 4 0. 2 0. 2 3. 9 2. 4 1. 5 90 .4 10 0. 0 11 ,3 21 C U RR EN TL Y M A RR IE D W O M EN 1 5- 19 3. 2 3. 2 0. 0 2. 9 0. 0 0. 3 0. 0 0. 0 0. 0 0. 0 0. 0 0. 0 96 .8 10 0. 0 65 2 0- 24 12 .7 7. 6 0. 0 2. 4 3. 4 1. 3 0. 3 0. 1 0. 0 5. 1 2. 7 2. 4 87 .3 10 0. 0 98 0 2 5- 29 17 .3 10 .9 0. 3 2. 6 5. 2 0. 8 1. 0 0. 6 0. 5 6. 5 3. 0 3. 4 82 .7 10 0. 0 1, 25 4 3 0- 34 20 .3 12 .7 0. 5 2. 8 6. 2 1. 1 1. 1 0. 5 0. 6 7. 7 4. 7 3. 0 79 .7 10 0. 0 1, 11 2 3 5- 39 20 .4 13 .1 0. 8 2. 9 5. 1 0. 8 1. 9 0. 3 1. 2 7. 3 4. 4 3. 0 79 .6 10 0. 0 80 7 4 0- 44 19 .7 10 .0 0. 8 2. 3 4. 7 0. 9 0. 1 1. 0 0. 1 9. 7 6. 7 3. 0 80 .3 10 0. 0 73 9 4 5- 49 14 .1 5. 8 1. 5 0. 7 2. 7 0. 1 0. 2 0. 4 0. 3 8. 2 4. 9 3. 3 85 .9 10 0. 0 55 4 T ot al 17 .4 10 .3 0. 5 2. 4 4. 7 0. 9 0. 8 0. 5 0. 4 7. 1 4. 2 3. 0 82 .6 10 0. 0 5, 51 0 SE XU A LL Y AC TI V E U N M A RR IE D W O M EN 1 T ot al 11 .6 6. 2 0. 4 0. 5 1. 5 3. 8 0. 0 0. 0 0. 0 5. 4 3. 3 2. 1 88 .4 10 0. 0 13 6 N ot e: If m or e th an o ne m et ho d is us ed , o nl y th e m os t e ffe ct iv e m et ho d is co ns id er ed in th is ta bu la tio n. LA M = L ac ta tio na l a m en or rh ea m et ho d 1 W om en w ho h av e ha d se xu al in te rc ou rs e in th e m on th p re ce di ng th e su rv ey 55Family Planning | 56 | Family Planning Current use of contraception among sexually active unmarried women falls between the two other categories: 12 percent for any method, 6 percent for any modern method, and 5 percent for any traditional method. This represents a decline compared with 2000, when contraceptive prevalence among sexually active unmarried women was 22 percent for any method, 15 percent for any modern method, and 7 percent for any traditional method. Male condoms (4 percent) constitute the modern method used most frequently by these women. Periodic abstinence (3 percent) is the most frequently used traditional method. Comparison of the current survey results with those of the previous two surveys reveals an overall decline in contraceptive prevalence1 (Figure 5.2). Among married women, current use of modern methods dropped significantly from 13 percent to 4 percent between 1992 and 2000. Although it has increased since the 2000 survey (10 percent in 2005), current use of contraception nevertheless remains lower than the level observed in 1992 (13 percent). Among sexually active unmarried women, the rate of condom use was 11 percent in 2000; it has dropped to 4 percent in the current survey. 1 During the two previous surveys, LAM and SDM/beads were not included among modern contraceptive methods. If these were not included among modern methods in the RDHS-III, modern contraceptive prevalence among currently married women drops to 9 percent. Figure 5.1 Contraceptive Use among Currently Married Women Age 15-49 RDHS 2005 17 10 5 2 1 7 4 3 Any method Modern method Injectables Pill Male condom Traditional method Periodic abstinence Withdrawal 0 5 10 15 20 25 Percent Family Planning | 57 Table 5.4 shows the distribution of currently married women by method of contraception currently used, according to background characteristics. Contraceptive prevalence is noticeably higher in urban areas (32 percent) than in rural ones (15 percent). It is considerably higher in the City of Kigali (36 percent) than in the other provinces (19 percent at most in the East province). Percentages in the South and West (15 percent each) and North (16 percent) provinces are lower than the national average (17 percent). The prevalence of modern contraceptive methods also varies by province, being highest in the City of Kigali (23 percent), and ranging between 8 percent and 10 percent in the other provinces. Contraceptive prevalence varies by level of education. Married women with no education have a lower level of contraceptive use—11 percent for any method, 6 percent for any modern method, and 5 percent for any traditional method—than educated women. The use of contraception increases with the number of children, from 3 percent among nulliparous women, to 14 percent among women with 1 or 2 children, to 21 percent among women with 3 children or more. Similarly, women in the poorest wealth quintile use contraception less frequently (11 percent) than women in the other quintiles, particularly the richest quintile (32 percent). Figure 5.2 Trends in Use of Modern Methods among Currently Married Women 13 20 13 4 14 3 10 21 9 Rwanda Urban Rural 0 5 10 15 20 25 30 Percent 1992 RDHS 2000 RDHS 2005 RDHS T ab le 5 .4 C ur re nt u se o f c on tra ce pt io n by b ac kg ro un d ch ar ac te ris tic s P er ce nt d ist rib ut io n of c ur re nt ly m ar rie d w om en b y co nt ra ce pt iv e m et ho d cu rr en tly u se d, a cc or di ng to b ac kg ro un d ch ar ac te ris tic s, R w an da 2 00 5 M od er n m et ho d Tr ad iti on al m et ho d B ac kg ro un d ch ar ac te ris tic An y m et ho d An y m od er n m et ho d Fe m al e st er ili - za tio n Pi ll In je ct - ab le s M al e co nd om LA M St an da rd da ys m et ho d/ be ad s O th er m od er n m et ho ds An y tra di - tio na l m et ho d Pe rio di c ab st i- ne nc e W ith - dr aw al N ot cu rr en tly us in g To ta l N um be r of w om en R es id en ce U rb an 31 .6 21 .2 1. 1 4. 2 7. 3 4. 0 1. 5 1. 4 1. 9 10 .4 6. 9 3. 5 68 .4 10 0. 0 74 4 Ru ra l 15 .2 8. 6 0. 5 2. 2 4. 3 0. 4 0. 7 0. 3 0. 2 6. 6 3. 7 2. 9 84 .8 10 0. 0 4 76 6 P ro vi nc e Ki ga li C ity 35 .5 23 .2 1. 3 4. 2 6. 6 5. 2 2. 2 1. 4 2. 2 12 .3 7. 9 4. 4 64 .5 10 0. 0 40 7 So ut h 14 .8 8. 4 0. 4 1. 7 3. 9 0. 7 0. 6 0. 7 0. 3 6. 5 3. 4 3. 1 85 .2 10 0. 0 1, 41 1 W es t 14 .5 10 .3 1. 0 2. 2 4. 7 0. 4 1. 3 0. 5 0. 2 4. 2 2. 3 1. 9 85 .5 10 0. 0 1, 42 7 N or th 16 .0 9. 8 0. 2 2. 8 5. 5 0. 7 0. 1 0. 3 0. 3 6. 1 3. 5 2. 6 84 .0 10 0. 0 1, 05 8 Ea st 18 .9 8. 5 0. 2 2. 7 4. 1 0. 4 0. 7 0. 1 0. 4 10 .4 6. 5 3. 9 81 .1 10 0. 0 1, 20 8 E du ca tio n N o ed uc at io n 10 .8 5. 9 0. 4 1. 4 3. 1 0. 3 0. 4 0. 2 0. 1 5. 0 3. 0 1. 9 89 .2 10 0. 0 1, 64 0 Pr im ar y 17 .3 9. 7 0. 4 2. 2 4. 9 0. 7 1. 0 0. 3 0. 2 7. 6 4. 1 3. 4 82 .7 10 0. 0 3, 39 2 Se co nd ar y or h ig he r 40 .4 29 .1 1. 7 7. 9 8. 5 4. 0 0. 9 2. 6 3. 5 11 .2 8. 1 3. 2 59 .6 10 0. 0 47 9 N um be r of li vi ng c hi ld re n 0 2. 6 1. 1 0. 0 0. 0 0. 7 0. 3 0. 0 0. 0 0. 0 1. 5 0. 5 0. 9 97 .4 10 0. 0 33 7 1- 2 13 .9 8. 0 0. 3 2. 5 2. 9 1. 1 0. 6 0. 4 0. 2 5. 9 3. 4 2. 4 86 .1 10 0. 0 1, 87 4 3- 4 21 .1 13 .0 0. 7 2. 8 6. 2 1. 0 1. 0 0. 4 0. 7 8. 1 4. 4 3. 7 78 .9 10 0. 0 1, 73 5 5+ 20 .7 12 .0 0. 8 2. 5 5. 9 0. 6 1. 0 0. 7 0. 5 8. 7 5. 5 3. 2 79 .3 10 0. 0 1, 56 5 W ea lth q ui nt ile Lo w es t 11 .0 6. 0 0. 4 1. 1 3. 3 0. 2 0. 8 0. 1 0. 1 5. 0 3. 3 1. 7 89 .0 10 0. 0 1, 13 6 Se co nd 15 .2 7. 4 0. 5 1. 7 3. 8 0. 4 0. 7 0. 2 0. 1 7. 8 4. 4 3. 3 84 .8 10 0. 0 1, 12 3 M id dl e 15 .7 8. 5 0. 2 2. 2 4. 2 0. 2 1. 1 0. 4 0. 1 7. 3 4. 4 2. 8 84 .3 10 0. 0 1, 11 2 Fo ur th 14 .8 8. 5 0. 4 1. 8 4. 9 0. 5 0. 4 0. 3 0. 2 6. 4 2. 7 3. 7 85 .2 10 0. 0 1, 14 4 H ig he st 31 .8 22 .4 1. 2 5. 9 7. 4 3. 4 1. 2 1. 4 1. 8 9. 4 6. 2 3. 2 68 .2 10 0. 0 99 5 T ot al 17 .4 10 .3 0. 5 2. 4 4. 7 0. 9 0. 8 0. 5 0. 4 7. 1 4. 2 3. 0 82 .6 10 0. 0 5, 51 0 N ot e: If m or e th an o ne m et ho d is us ed , o nl y th e m os t e ffe ct iv e m et ho d is co ns id er ed in th is ta bu la tio n. LA M = L ac ta tio na l a m en or rh ea m et ho d 58 | Family Planning Family Planning | 59 5.3 NUMBER OF CHILDREN AT FIRST USE OF CONTRACEPTION The use of contraception for the first time meets different needs depending primarily on the number of living children: • Delaying first birth - contraceptive use begins before the woman has any children. • Birth spacing - contraceptive use begins when the number of living children is low. • Limiting cumulative fertility - contraceptive use begins after the desired number of children has been reached. Table 5.5 shows the distribution of women who have ever used contraception by number of living children at the time of first use of contraception, according to age. Overall, 5 percent of women who have used contraception began use before they gave birth, i.e., to delay the first birth (compared with only 1 percent in 2000); 26 percent began using contraception after having one child (12 percent in 2000); 25 percent began using contraception after having two children (9 percent in 2000); 17 percent began using after having 3 children (5 percent in 2000); and 26 percent began using at higher parities (4 children or more) (8 percent in 2000), almost certainly to limit cumulative fertility. The proportion of women using contraception for the first time before having any children has risen from previous generations: only 2 percent of women between the ages of 30 and 34 began using contraception before having children; for women age 25 to 29, this proportion is 4 percent, and for women age 20 to 24, it is 20 percent. Table 5.5 Number of children at first use of contraception Percent distribution of women who have ever used contraception by number of living children at the time of first use of contraception, according to age, Rwanda 2005 Number of living children at time of first use of contraception Age group 0 1 2 3 4+ Total1 Number 15-19 (78.2) (16.4) (3.7) (0.0) (0.0) (100.0) 34 20-24 19.8 48.6 26.2 4.0 0.2 100.0 289 25-29 4.3 38.1 32.1 17.5 7.4 100.0 475 30-34 1.6 26.8 31.8 21.3 18.5 100.0 485 35-39 0.7 24.0 26.4 21.2 27.6 100.0 415 40-44 0.1 12.7 19.8 18.6 48.5 100.0 411 45-49 0.0 8.3 13.1 16.7 61.8 100.0 311 Total 4.8 26.2 25.4 17.1 26.1 100.0 2,421 Note: Figures in parentheses are based on 25-49 unweighted cases. 1 Includes those with missing information 5.4 KNOWLEDGE OF FERTILE PERIOD Table 5.3 shows that among currently married women, injectables (5 percent) and periodic abstinence (4 percent), are the most frequently used methods of contraception in Rwanda. However, the effective use of periodic abstinence depends largely on an understanding of when during the menstrual cycle a woman is most likely to conceive. To assess this understanding, the survey asked all women if 60 | Family Planning there were certain days during the menstrual cycle when they were more likely to become pregnant if they had sexual intercourse. Those who answered yes were asked when those days occurred during the cycle. The question provided four explicit responses: “just before her period begins,” “right after her period has ended,” “during her period,” and “halfway between two periods.” Respondents could also give a different response or state that they did not know when this occurred. These responses can be grouped into three categories of decreasing knowledge: • Correct knowledge: “halfway between two periods;” the middle of the cycle. • Possibly correct knowledge: “just before her period begins,” “right after her period has ended.” These responses are too vague to be considered accurate but, depending on how a woman views “right after her period has ended” or “just before her period begins,” these answers could indicate the fertile period. • Incorrect knowledge: “during her period,” “no specific time,” “other,” and “don’t know.” Table 5.6 indicates that only 13 percent of women have correct knowledge about when the fertile period occurs; 15 percent have possibly correct knowledge; and 72 percent have incorrect knowledge or don’t know that there is a time during the menstrual cycle when a woman is more likely to conceive. Knowledge of the fertile period among users of periodic abstinence as a family planning method is considerably higher (33 percent) than for nonusers (13 per- cent). However, four in ten users of peri- odic abstinence (41 percent) have only possibly correct knowledge of the fertile period, and 26 percent do not know when a woman should stop having sexual intercourse in order to avoid becoming pregnant, or do not know that such a time exists. These results show that, in 2005, as in 2000, nearly seven in ten users of periodic abstinence (67 percent) do not know how to use the method correctly because they have only possibly correct knowledge of the fertile period. This has major implications for family planning, especially given that contraceptive prevalence is so low in Rwanda (17 percent) and periodic abstinence is the most frequently used method after injectables. 5.5 SOURCE OF CONTRACEPTION To assess the contribution of public and private medical service providers to the sale or distribution of the various modern methods of contraception, the women surveyed were asked where they obtained the method they use. The RDHS-III also asked where they had most recently obtained the contraceptive methods they were using at the time of the survey. Table 5.7 shows that in Rwanda, the majority of women obtain modern methods of contraception from the public sector (73 percent, compared with 69 percent in 2000); 13 percent of women obtain their method from a government hospital; 58 percent from a health facility; and 1 percent from another public entity. In addition, 14 percent of contraceptive users obtain their method from the private medical sector, particularly pharmacies and other private medical sources (6 percent each). The nonmedical private sector (shops, kiosks, friends, relatives) supplies 8 percent of contraceptive needs, and other sources provide 5 percent. Table 5.6 Knowledge of the fertile period Percent distribution of women by knowledge of the fertile period during the ovulatory cycle, according to current use/non use of rhythm or periodic abstinence, Rwanda 2005 Perceived fertile period Users of rhythm or periodic abstinence Nonusers of rhythm or periodic abstinence All women Halfway between two periods 32.7 12.7 13.1 Just before her period begins 3.0 2.0 2.0 Right after her period has ended 37.9 12.6 13.2 During her period 3.0 1.1 1.1 No specific time 21.2 62.9 61.9 Don't know 2.1 8.8 8.6 Total 100.0 100.0 100.0 Number of women 276 11,045 11,321 Family Planning | 61 Table 5.7 Source of contraception Percent distribution of current users of modern contraceptive methods by most recent source of method, according to specific method, Rwanda 2005 Source Female sterilization Pill Injectables Male condom Standard days method/beads Other modern methods Total Public sector (92.1) 80.9 86.2 19.0 (54.5) (69.4) 72.6 Government hospital (67.6) 7.9 9.2 4.1 (3.0) (53.0) 13.0 Government health centre (22.0) 71.3 76.7 14.2 (42.2) (16.4) 58.3 Other public (2.5) 1.7 0.3 0.7 (9.4) (0.0) 1.2 Private medical sector (2.4) 14.5 8.8 29.6 (11.9) (26.4) 14.0 Private hospital or clinic (2.4) 1.9 0.6 6.5 (0.0) (10.1) 2.3 Pharmacy (0.0) 7.8 1.7 20.4 (0.0) (0.0) 5.9 Other private medical (0.0) 4.8 6.5 2.7 (11.9) (16.3) 5.8 Other source (5.5) 4.6 4.9 51.4 (33.5) (4.2) 13.4 Shop/kiosk (0.0) 0.6 0.1 41.1 (4.8) (4.2) 7.0 Friends, relatives (0.0) 0.0 0.0 4.0 (8.5) (0.0) 1.0 Other (5.5) 4.0 4.8 6.3 (20.2) (0.0) 5.4 Total (100.0) 100.0 100.0 100.0 (100.0) (100.0) 100.0 Number of women 34 144 269 93 27 26 592 Note: Table excludes lactational amenorrhea method (LAM). Figures in parentheses are based on 25-49 unweighted cases. For methods designed to be used directly by women and requiring procurement, the great majority of women who use modern methods turn to the public sector (81 percent for the pill, and 86 percent for injectables), while condoms, designed to be used by the partner, are most often obtained from the private sector. Women obtain condoms either from the private medical sector (30 percent) or from other sources such as shops and kiosks (45 percent). Because the number of women using female sterilization is so low, a discussion of the variations would not be meaningful. This is also the case for the standard days, or beads method, which was only recently introduced in Rwanda and is therefore used by relatively few women. 5.6 FUTURE USE OF CONTRACEPTION Married women who were not using a contraceptive method at the time of the survey were asked whether they planned to use one in the future. The reason given by those who do not plan to use contraception in the future is useful in developing family planning marketing strategies. Also, the methods preferred by those who do plan to use contraception in the future is useful in assessing the demand for family planning. Overall, nearly six in ten women (59 percent) reported that they plan to use a contraceptive method in the future, 7 percent were not sure, and 34 percent reported that they did not intend to use contraception (Table 5.8). The number of children a woman has affects her decision on whether to use contraception in the future. Nearly half (46 percent) of women who do not have any children reported intending to use a family planning method in the future. Among women with one or two children, the proportion is 65 percent; among those with three children and with four or more children, the proportions are lower (61 percent for three children; 54 percent for four or more children). 62 | Family Planning Table 5.8 Future use of contraception Percent distribution of currently married women who are not using a contraceptive method by intention to use in the future, according to number of living children, Rwanda 2005 Number of living children1 Intention 0 1 2 3 4+ Total Intends to use 45.8 64.6 65.1 60.7 53.6 58.5 Unsure 12.7 10.4 6.5 7.5 5.3 7.0 Does not intend to use 41.5 25.0 28.2 31.6 40.8 34.3 Total2 100.0 100.0 100.0 100.0 100.0 100.0 Number of women 186 715 880 789 1,982 4,552 1 Includes current pregnancy 2 Includes those with missing information Women who were not using contraception and do not plan to use it in the future were asked to give their reason. Table 5.9 shows the variety of reasons given. Forty-four percent of women gave reasons relating to fertility, in particular, infrequent sex/no sex (8 percent), menopause/hysterectomy (14 percent), the desire to have as many children as possible (10 percent), and low fertility/infertility (12 percent). A little more than one in five women (22 percent) said they were opposed to the use of contraception, this opposition being motivated by religious prohibitions (10 percent), opposition of the husband/partner (4 percent), or opposition of the respondent herself (7 percent). Twenty-two percent of women gave health concerns and fear of side effects as reasons for not intending to use contraception. The proportion of women who gave reasons related to lack of knowledge is negligible (2 percent). The frequency with which reasons were reported by women varies according to age group. Women age 15 to 29 were less likely to give fertility related reasons (19 percent) than women age 30 to 49 (51 percent). Women age 15 to 29 gave reasons more frequently related to opposition to contraception (34 percent) or to the method of contraception (31 percent) than older women age 30 to 49 (18 percent and 19 percent, respectively). Family Planning | 63 Table 5.9 Reason for not intending to use contraception Percent distribution of currently married women who are not using a contraceptive method and who do not intend to use in the future by main reason for not intending to use, according to age, Rwanda 2005 Age Reason 15-29 30-49 Total Fertility-related reasons 19.0 51.1 43.6 Infrequent sex/no sex 3.9 8.7 7.6 Menopausal/had hysterectomy 0.0 18.0 13.8 Subfecund/infecund 3.4 15.0 12.3 Wants as many children as possible 11.6 9.4 9.9 Opposition to use 33.6 18.0 21.7 Respondent opposed 9.9 5.8 6.7 Husband/partner opposed 7.6 3.3 4.3 Others opposed 1.5 0.5 0.7 Religious prohibition 14.7 8.4 9.9 Lack of knowledge 3.3 2.0 2.3 Knows no method 2.8 1.0 1.4 Knows no source 0.5 1.0 0.9 Method-related reasons 30.9 19.3 22.0 Health concerns 3.7 3.9 3.8 Fear of side effects 20.9 12.2 14.2 Lack of access/too far 0.0 0.1 0.1 Costs too much 2.3 0.4 0.9 Inconvenient to use 1.7 0.7 1.0 Interfere with body's normal processes 2.3 2.0 2.1 Other 11.4 7.6 8.5 Don’t know/missing 1.8 2.0 1.9 Total 100.0 100.0 100.0 Number of women 366 1,198 1,563 To assess the potential demand for specific contraceptive methods, married women who reported intending to use contraception in the future were asked to state their preferred method. Table 5.10 indicates that most women prefer modern methods regardless of their age; in particular, injectables (36 percent) and the pill (20 percent). Male condoms and female sterilization were mentioned by only 4 percent of women. Among traditional methods, periodic abstinence was cited most frequently (13 percent). 64 | Family Planning Table 5.10 Preferred method of contraception for future use Percent distribution of currently married women who are not using a contraceptive method but who intend to use in the future by preferred method, according to age, Rwanda 2005 Age Method 15-29 30-49 Total Female sterilization 1.8 7.2 4.3 Male sterilization 0.1 0.0 0.1 Pill 21.4 17.7 19.7 IUD 2.7 2.0 2.4 Injectables 37.0 35.7 36.4 Implants 2.5 4.1 3.2 Male condom 4.6 3.5 4.1 Female condom 0.1 0.1 0.1 Diaphragm 0.0 0.1 0.1 Foam/jelly 0.0 0.1 0.1 LAM 0.2 0.2 0.2 Standard days/beads 2.3 2.8 2.5 Rhythm/periodic abstinence 13.9 12.3 13.2 Withdrawal 1.4 2.7 2.0 Other 3.0 3.1 3.1 Unsure 8.9 8.2 8.6 Missing 0.0 0.1 0.0 Total 100.0 100.0 100.0 Number of women 1,433 1,230 2,663 5.7 EXPOSURE TO FAMILY PLANNING MESSAGES Information on the level of exposure to sources of information about family planning can be very important to those developing family planning programs. This information allows them to design strategies to reach specific target populations and to effectively disseminate information about contraceptive use. For this reason, the survey asked women age 15 to 49 and men age 15 to 59 whether they had heard or seen anything about family planning on the radio or on television, or from newspapers/magazines, or posters/ brochures, during the past few months. Table 5.11.1 shows that 59 percent of women did not see or hear a family planning message in newspapers/magazines, or on radio or television. However, 41 percent of women did hear a family planning message on the radio, and 4 percent did see one on television. Only 5 percent of women saw a family planning message in a newspaper or magazine in the past few months. Family Planning | 65 Table 5.11.1 Exposure to family planning messages: women Percentage of women who heard or saw a family planning message on the radio or television, or in a newspaper/magazine in the past few months, according to background characteristics, Rwanda 2005 Background characteristic Radio Television Newspaper/ magazine None of these three media sources Number of women Age 15-19 32.4 3.1 3.8 67.4 2,585 20-24 42.2 4.2 5.1 57.3 2,354 25-29 46.9 4.0 5.9 52.6 1,738 30-34 44.1 4.1 5.6 55.6 1,466 35-39 42.5 3.3 3.9 57.3 1,134 40-44 41.9 2.9 3.9 58.1 1,135 45-49 40.4 2.5 3.1 59.4 910 Residence Urban 56.5 12.7 12.1 42.7 1,921 Rural 37.6 1.7 3.1 62.3 9,400 Province Kigali city 56.1 16.0 14.2 43.0 1,127 South 37.5 2.6 4.9 62.3 2,958 West 32.3 3.0 4.4 67.4 2,824 North 47.5 1.6 2.7 52.4 2,063 East 41.9 1.1 1.6 58.0 2,348 Education No education 32.6 1.2 1.2 67.3 2,646 Primary 39.3 2.3 3.3 60.5 7,591 Secondary or higher 71.3 17.8 22.0 27.3 1,084 Wealth quintile Lowest 22.9 1.0 1.6 77.1 2,421 Second 38.0 0.7 1.7 61.9 2,325 Middle 39.5 1.5 2.6 60.5 2,099 Fourth 46.0 1.9 4.1 53.6 2,133 Highest 58.4 12.4 12.8 40.8 2,342 Total 40.8 3.5 4.6 58.9 11,321 Exposure to family planning messages in the media varies by background characteristics. Women age 15 to 19 and 45 to 49 had the highest levels of nonexposure to family planning messages in the media during the 12 months preceding the survey (67 percent and 59 percent, respectively). The results also show disparities by residence, with women in rural areas having higher rates of nonexposure than women in urban areas (62 percent for rural, 43 percent for urban). Similarly, women with no education were less exposed (67 percent with no exposure) than those with a secondary education or higher (27 percent with no exposure); and women in the poorest households were less exposed (77 percent with no exposure) than women in the wealthiest households (41 percent with no exposure). In the provinces, the West (67 percent) and South (62 percent) provinces had the highest levels of nonexposure to family planning messages. Radio is by far the most frequent source of family planning messages (41 percent). Only a small percentage of women reported seeing a family planning message on television (4 percent) or in a newspaper (5 percent). With respect to men, Table 5.11.2 shows that nearly four in ten men (39 percent)—a smaller proportion than for women (59 percent)—had no exposure to a family planning message in the past few 66 | Family Planning months through any of the various media (radio, television, newspapers/magazines). However, 61 percent of men reported having heard a family planning message on the radio; at least 6 percent had seen one on television; and at least 12 percent had seen one in a newspaper or magazine. Younger men were the least exposed to family planning messages—age 15 to 19, 52 percent had no exposure—regardless of the media source. Like women, men in rural areas were more likely to report not having been exposed to family planning messages, regardless of the source (41 percent for rural areas compared with 28 percent for urban areas). Similarly, men with no education (45 percent) were more likely to have had no exposure than those with a secondary education or higher (23 percent); and men in the poorest households (53 percent) were more likely to have had no exposure to family planning messages than those in the richest households (28 percent). Results by province show that 76 percent of men in the City of Kigali have heard or seen a family planning message, compared with 51 percent in the West province. Table 5.11.2 Exposure to family planning messages: men Percentage of men who heard or saw a family planning message on the radio or television, or in a newspaper/magazine in the past few months, according to background characteristics, Rwanda 2005 Background characteristic Radio Television Newspaper/ magazine None of these three media sources Number of men Age 15-19 48.1 5.3 8.2 51.7 1,102 20-24 61.0 6.7 12.2 38.3 946 25-29 65.4 7.6 15.2 33.5 632 30-34 70.5 9.2 14.4 28.9 509 35-39 64.6 6.0 16.1 34.6 442 40-44 62.3 6.0 14.1 37.5 404 45-49 65.7 4.9 13.2 33.9 378 50-54 66.4 6.6 11.7 33.6 260 55-59 60.9 3.3 6.7 39.1 147 Residence Urban 70.4 21.1 25.2 28.2 840 Rural 58.7 3.3 9.6 41.0 3,980 Province Kigali city 75.7 26.7 28.7 22.6 523 South 62.0 4.9 11.6 37.7 1,250 West 51.2 2.8 9.7 48.2 1,185 North 62.9 3.9 10.8 37.0 845 East 60.9 4.1 9.2 38.8 1,017 Education No education 55.2 2.4 6.8 44.7 942 Primary 58.4 3.9 9.1 41.2 2,955 Secondary or higher 75.2 19.8 30.5 23.3 850 Wealth quintile Lowest 47.1 1.9 6.2 52.5 928 Second 56.9 2.4 7.4 43.0 970 Middle 59.7 3.5 10.6 40.1 940 Fourth 68.0 4.1 11.6 31.7 958 Highest 70.9 19.0 24.8 27.7 1,024 Total 60.7 6.4 12.3 38.8 4,820 Family Planning | 67 5.8 CONTACT OF NONUSERS WITH FAMILY PLANNING PROVIDERS Information on contact of women who do not use contraception with family planning service providers is important for determining effective family planning outreach activities. For this reason, the RDHS-III asked women whether they had been visited in the past 12 months by a health fieldworker who spoke to them about family planning. Women who had visited a health facility in the past 12 months were also asked whether medical personnel had spoken to them about family planning methods. Table 5.12 shows that in the 12 months preceding the survey, nine out of ten women who were nonusers of contraception (91 percent) had not discussed family planning with a fieldworker or at a health facility. Nearly one in five women (19 percent) had visited a health facility but had not discussed family planning issues. Only 3 percent had been visited by a fieldworker who discussed family planning with them, and only 7 percent had discussed family planning at a health facility. There are no significant differences by residence: 92 percent of women in urban areas and 90 percent in rural areas had not discussed family planning with a fieldworker or at a health facility. The results show no significant differentials by level of education. Table 5.12 Contact of nonusers with family planning providers Percentage of women who are not using contraception who were visited by a fieldworker who discussed family planning, who visited a health facility and discussed family planning, and who visited a health facility but did not discuss family planning, in the 12 months preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Women visited by fieldworker who discussed family planning Women visited health facility and discussed family planning Women visited health facility but did not discuss family planning Did not discuss family planning with field- worker or at a health facility Number of women Age 15-19 1.3 0.6 8.3 98.2 2,576 20-24 3.9 6.5 20.0 90.7 2,201 25-29 4.2 12.9 25.3 84.7 1,497 30-34 4.8 13.2 24.9 84.5 1,216 35-39 3.5 11.4 24.7 87.0 951 40-44 4.5 10.0 20.8 87.5 974 45-49 4.0 4.4 16.1 92.8 822 Residence Urban 2.6 5.6 24.9 92.4 1,637 Rural 3.6 7.7 17.4 90.1 8,599 Education No education 4.4 7.6 18.0 89.7 2,448 Primary 3.0 7.4 17.8 90.8 6,931 Secondary or higher 4.0 6.9 27.0 90.3 857 Total 3.4 7.4 18.6 90.5 10,237 5.9 OPINIONS AND ATTITUDES OF COUPLES TOWARD FAMILY PLANNING 5.9.1 Discussion of Family Planning with Husband The RDHS-III asked married women how many times they had discussed family planning with their husband/partner in the 12 months preceding the survey. 68 | Family Planning Overall, 30 percent of women said they had not discussed contraception with their husband during the 12 months preceding the survey (Table 5.13); 27 percent had discussed it at least once or twice; and 41 percent had discussed it at least three times. The proportion of women who discussed family planning with their husband has grown considerably since the RDHS-II survey, from 57 percent in 2000 to 68 percent in 2005. Results by age show that older women (age 45-49) were the most likely to have never discussed family planning with their husband/partner (47 percent). Women age 25-29 were the most likely to have discussed family planning at least three times (48 percent). Table 5.13 Discussion of family planning with husband Percent distribution of currently married women who know a contraceptive method by the number of times they discussed family planning with their husband in the past year, according to age, Rwanda 2005 Number of times family planning discussed with husband in the past 12 months Age group Never One or two Three or more Missing Total Number of women 15-19 33.3 23.0 43.3 0.3 100.0 57 20-24 26.8 32.6 39.3 1.3 100.0 953 25-29 24.3 26.1 47.6 2.0 100.0 1,232 30-34 25.4 27.1 45.3 2.2 100.0 1,095 35-39 32.2 25.6 39.3 3.0 100.0 799 40-44 33.7 21.3 40.1 4.8 100.0 722 45-49 46.6 23.7 25.8 3.9 100.0 537 Total 29.7 26.5 41.2 2.6 100.0 5,394 5.9.2 Attitudes of Couples toward Family Planning In general, contraceptive use by women is influenced by the attitude of the couple. In Rwandan society, decisionmaking power with respect to family size rests most often with the husband. Depending on the society, other family members may also influence this decision. However, joint decisionmaking by both spouses/partners can result in changes in men’s behavior and a more favorable attitude toward contraception. For this reason, the RDHS-III examined couples’ attitudes toward family planning. Women were asked whether they approved or disapproved of couples who use a family planning method. They were then asked if they thought their husband approved or disapproved of family planning. The combined responses to these two questions were used to reveal differences in attitudes between the spouses. The results are presented in Table 5.14. Table 5.14 shows that, overall, 87 percent of women approve of family planning, 10 percent do not approve, and approximately 4 percent are not sure. In addition, spouses in 59 percent of couples have the same opinion and approve of family planning; 10 percent of women approve but their husband does not; and among those who do approve, nearly 18 percent of women do not know their husband’s opinion. The proportion of couples in which both spouses approve of family planning varies according to the woman’s age. It is lowest among couples in which the woman is age 45 to 49 (49 percent) or 15 to 19 (51 percent). In addition, the proportion of couples in which both spouses approve is lower in rural areas (58 percent) than in urban areas (67 percent), and lower among couples in which the woman has no education (47 percent) than among those with at least a secondary education (77 percent). By wealth Family Planning | 69 quintile, the proportion of couples in which both spouses approve is lowest in the poorest quintile (53 percent) and highest in the richest quintile (69 percent). Couples whose opinions diverge represent 11 percent of all couples. In 10 percent of couples, women approve of contraception but their husbands do not, but the reverse is true in only 1 percent of couples. This shows the role of the man in decisionmaking: when the man is favorable to family planning, the woman is too, while the reverse is not always true. The proportion of couples with diverging opinions varies only slightly by women’s background characteristics. Finally, in 21 percent of couples, the woman has no idea of her husband’s opinion, which shows a lack of dialogue on the subject between some spouses. Table 5.14 Attitudes towards family planning Percent distribution of currently married women who know of a method of family planning,, by approval of family planning and their perception of their husband's attitude towards family planning, according to background characteristics, Rwanda 2005 Respondent approves of family planning Respondent disapproves of family planning Background characteristic Husband approves Husband disapproves Husband's attitude unknown/ missing Husband approves Husband disapproves Husband's attitude unknown/ missing Respondent unsure Total Number of women Age 15-19 51.2 12.0 30.9 0.0 4.7 0.0 1.2 100.0 57 20-24 62.2 9.0 15.8 1.6 5.5 1.7 4.2 100.0 953 25-29 64.3 8.9 14.2 0.9 4.6 3.7 3.3 100.0 1,232 30-34 62.0 10.1 14.5 1.7 5.5 3.7 2.6 100.0 1,095 35-39 55.9 11.6 19.5 1.1 5.8 2.0 4.2 100.0 799 40-44 55.9 8.4 22.1 0.6 6.1 3.2 3.7 100.0 722 45-49 49.0 10.7 24.7 0.7 5.3 4.7 4.9 100.0 537 Residence Urban 66.6 9.3 11.6 2.0 5.9 3.1 1.6 100.0 738 Rural 58.3 9.8 18.5 1.0 5.3 3.1 4.0 100.0 4,656 Education No education 47.4 11.4 23.4 1.5 6.3 4.4 5.6 100.0 1,572 Primary 62.5 9.3 16.4 0.9 5.2 2.6 3.2 100.0 3,343 Secondary or higher 77.4 7.1 6.9 1.7 3.9 2.2 0.8 100.0 479 Wealth quintile Lowest 52.9 11.4 21.4 1.2 5.1 2.6 5.4 100.0 1,104 Second 59.8 8.9 19.2 0.7 5.2 3.0 3.2 100.0 1,097 Middle 58.5 9.7 18.4 0.9 5.9 3.4 3.4 100.0 1,093 Fourth 57.8 10.7 18.2 1.2 5.4 3.3 3.5 100.0 1,116 Highest 69.2 7.6 10.1 1.8 5.4 3.3 2.6 100.0 984 Total 59.4 9.7 17.6 1.1 5.4 3.1 3.6 100.0 5,394 Other Proximate Determinants of Fertility | 71 OTHER PROXIMATE DETERMINANTS OF FERTILITY 6 This chapter addresses the key factors that define the risk of becoming pregnant. These include age at first marriage, age at first sexual intercourse, sexual activity, and postpartum abstinence and amenorrhea. 6.1 MARITAL STATUS In Rwanda, formal unions (“married”) or informal unions (“living together”) between men and women are the sole socially permissible context for sexual activity. Marital status can therefore be considered the primary factor initiating exposure to the risk of pregnancy. In the data discussed in this chapter, the term “married” refers to men and women bound together legally, while “living together” refers to couples cohabiting in informal unions. People are considered “never married” if they are not currently married, widowed, separated, or divorced. Table 6.1 shows the distribution of men and women by marital status according to age at the time of the survey. Of the 11,321 women surveyed, 49 percent were in union, 29 percent of these were formal marriages and 20 percent were informal unions. The proportion of women who were never married is 38 percent. Divorced women make up only 0.9 percent of women, separated women make up 9 percent, and widows are 4 percent. The proportion of never-married women has increased since the RDHS-II, from 34 percent to 38 percent. The largest increase occurred in the 15 to 19 age group, of whom 90 percent were never married in 1992, 93 percent in 2000, and 97 percent in 2005. The number of married women has remained relatively stable since the last survey. The proportion of widowed women has dropped by half, from 8 percent in 2000 to 4 percent in 2005. Table 6.1 Current marital status Percent distribution of women and of men by current marital status, according to age, Rwanda 2005 Marital status Age Never married Married Living together Divorced Separated Widowed Total Number WOMEN 15-19 97.1 0.2 2.3 0.0 0.4 0.0 100.0 2,585 20-24 53.2 15.4 26.2 0.3 4.6 0.3 100.0 2,354 25-29 16.2 40.2 31.9 0.8 9.9 1.0 100.0 1,738 30-34 8.0 46.6 29.4 1.2 11.6 3.3 100.0 1,466 35-39 5.0 46.0 25.1 1.4 14.4 8.1 100.0 1,134 40-44 2.7 46.1 19.0 1.9 15.9 14.5 100.0 1,135 45-49 1.7 48.3 12.5 2.2 17.2 18.1 100.0 910 Total 37.7 28.6 20.1 0.9 8.5 4.3 100.0 11,321 MEN 15-19 99.8 0.0 0.2 0.0 0.0 0.0 100.0 1,102 20-24 80.7 6.1 12.2 0.0 1.0 0.0 100.0 946 25-29 35.9 36.2 26.2 0.1 1.5 0.0 100.0 632 30-34 13.6 52.1 32.2 0.5 1.6 0.0 100.0 509 35-39 4.4 59.0 31.4 1.6 2.6 0.8 100.0 442 40-44 1.8 67.1 27.1 0.9 1.9 1.2 100.0 404 45-49 1.4 70.7 20.9 1.5 2.4 3.2 100.0 378 50-54 1.9 69.1 21.3 2.2 1.3 4.2 100.0 260 55-59 0.0 73.9 20.3 0.9 2.5 2.3 100.0 147 Total 45.6 34.0 17.8 0.6 1.3 0.7 100.0 4,820 72 | Other Proximate Determinants of Fertility Among the 4,820 men surveyed, 46 percent were never married, and 52 percent were in union, 34 percent were in formal marriages and 18 percent were “living together.” In addition, 2 percent were either separated or divorced (1.3 percent separated, 0.6 percent divorced,). Less than 1 percent of the men were widowed. A comparison of these data with the results of the previous survey shows no change in proportions of never-married men and married men. Figure 6.1, shows the percentage of never-married men and women according to age, indicates that the proportions of never-married men and women decrease with age: at age 15-19, nearly all men and women are never married (97 percent of women and 100 percent of men). Beginning at age 30 however, fewer than one man or woman in ten falls into this category. 6.2 POLYGYNY The survey asked currently married women whether their partners had any other wives besides them. Table 6.2 shows the percent distribution of married women by number of co-wives according to background characteristic. Polygyny is not very common in Rwanda. However, although illegal, it affects 12 percent of married women. The proportion of women with at least one co-wife increases steadily with age, from 6 percent at age 15-19, to 19 percent at age 45-49. The extent of polygyny does not differ substantially by residence, the percentage of married women living in polygynous unions ranging from 10 percent in urban areas to 12 percent in rural areas. Similarly, variations between the provinces are only slight and there are no substantial differences by wealth quintile. However, women’s level of education does affect the frequency of this practice: the percentage of married women living in polygynous unions is twice as high among women with no education (16 percent) as among those with a secondary education or higher (8 percent). Table 6.2 also gives results on polygyny for men. The rate of polygyny, that is the ratio of polygynous married men to all married men, is 5 percent. Results by age are inconsistent for polygamously married men and, like women, there are no significant differentials by background characteristics. Figure 6.1 Percentage of Never-Married Women and Men, by Age 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age 0 20 40 60 80 100 Percent Women Men RDHS 2005 Other Proximate Determinants of Fertility | 73 Table 6.2 Number of co-wives and wives Percent distribution of currently married women by number of co-wives and percent distribution of currently married men by number of wives, according to background characteristics, Rwanda 2005 Women Men Background characteristic 0 1 2+ Total Number 1 2 3+ Total Number Age 15-19 93.6 0.0 6.4 100.0 65 * * * * 2 20-24 93.6 0.0 6.3 100.0 980 97.1 2.5 0.0 100.0 173 25-29 91.3 0.3 8.2 100.0 1,254 98.2 1.6 0.2 100.0 394 30-34 88.0 0.0 11.9 100.0 1,112 94.9 4.4 0.3 100.0 429 35-39 84.5 0.2 15.1 100.0 807 96.5 3.4 0.2 100.0 400 40-44 85.1 0.0 14.5 100.0 739 92.8 6.7 0.5 100.0 381 45-49 80.8 0.0 19.0 100.0 554 91.2 8.5 0.3 100.0 346 50-54 na na na na na 92.3 7.1 0.6 100.0 235 55-59 na na na na na 93.0 6.1 0.9 100.0 139 Residence Urban 89.7 0.5 9.6 100.0 744 96.5 3.3 0.0 100.0 352 Rural 87.9 0.0 11.8 100.0 4,766 94.3 5.2 0.4 100.0 2,147 Province Kigali city 89.4 0.7 9.9 100.0 407 95.4 4.2 0.0 100.0 198 South 89.0 0.1 10.7 100.0 1,411 96.5 2.9 0.7 100.0 631 West 87.1 0.0 12.7 100.0 1,427 93.6 6.2 0.0 100.0 664 North 90.0 0.1 9.9 100.0 1,058 96.4 3.4 0.0 100.0 474 East 86.4 0.0 13.2 100.0 1,208 91.8 7.4 0.8 100.0 533 Education No education 83.6 0.0 16.2 100.0 1,640 94.3 5.0 0.6 100.0 593 Primary 89.9 0.1 9.8 100.0 3,392 94.4 5.2 0.3 100.0 1,621 Secondary or higher 92.0 0.2 7.8 100.0 479 96.8 2.9 0.0 100.0 285 Wealth quintile Lowest 88.4 0.1 11.3 100.0 1,136 95.5 3.8 0.7 100.0 481 Second 87.5 0.0 11.9 100.0 1,123 93.3 6.0 0.5 100.0 505 Middle 87.5 0.1 12.4 100.0 1,112 93.7 6.1 0.1 100.0 526 Fourth 88.3 0.0 11.7 100.0 1,144 94.8 4.8 0.4 100.0 551 Highest 89.3 0.4 10.3 100.0 995 96.3 3.6 0.0 100.0 437 Total 88.2 0.1 11.5 100.0 5,510 94.6 4.9 0.3 100.0 2,500 na = Not applicable Note: An asterisk indicates that the figure is based on fewer than 25 unweighted cases and has been suppressed. 6.3 AGE AT FIRST UNION Marriage remains the legally sanctioned context for sexual intercourse in Rwanda. Therefore, despite the existence of prenuptial intercourse, age at first marriage constitutes the beginning of exposure to the risk of pregnancy. For this reason, analysis of this variable is very important. Tables 6.3 and 6.4 show the percentage of currently married men and women by age first marriage according to current age. The proportion of girls who are already in union by age 15-19 is very low (3 percent). At age 18, the proportion is significantly higher (19 percent). At age 20, more than two in five women are married; at age 25, the proportion is 82 percent. The median age at first union is 20.7 years, which is relatively late. This has remained more or less unchanged since 1992, when the median age at first union was 20 years. 74 | Other Proximate Determinants of Fertility Table 6.3 Age at first marriage Percentages of women age 15-49 and of men age 15-59 who were first married by specific exact ages and median age at first marriage, according to current age, Rwanda 2005 Percentage first married by exact age: Age 15 18 20 22 25 Percentage never married Number Median age at first marriage WOMEN 15-19 0.2 na na na na 97.1 2,585 a 20-24 1.1 13.3 29.1 na na 53.2 2,354 a 25-29 2.2 20.1 44.5 61.9 79.1 16.2 1,738 20.6 30-34 2.7 14.9 35.7 59.0 81.7 8.0 1,466 21.1 35-39 2.5 18.3 39.2 58.7 80.0 5.0 1,134 21.0 40-44 2.5 21.5 45.8 65.2 84.7 2.7 1,135 20.4 45-49 3.3 22.8 45.3 70.8 88.1 1.7 910 20.3 25-49 2.6 19.2 41.9 62.5 82.1 7.9 6,383 20.7 MEN Percentage first married by exact age: Age 18 20 22 25 28 Percentage never married Number Median age at first marriage 15-19 0.2 na na na na 99.8 1,102 a 20-24 2.0 7.6 na na na 80.7 946 a 25-29 2.5 12.1 29.1 52.6 na 35.9 632 24.6 30-34 1.3 5.8 19.4 54.9 70.8 13.6 509 24.5 35-39 2.6 9.5 21.0 43.6 71.9 4.4 442 25.5 40-44 5.1 11.1 25.1 45.5 64.6 1.8 404 25.9 45-49 4.9 12.8 31.7 60.1 78.1 1.4 378 23.8 50-54 6.3 19.6 37.2 64.0 78.6 1.9 260 23.5 55-59 5.4 30.0 53.3 73.7 82.5 0.0 147 21.7 25-59 3.5 12.1 27.9 53.8 na 12.0 2,772 24.6 30-59 3.8 12.1 27.5 54.1 72.9 5.0 2,141 24.5 na = Not applicable a = Omitted because less than 50 percent of the women or men began living with their husbands, wives or partners for the first time before reaching the beginning of the age group According to the data, men marry at a later age than women: it is not until age 28 that three- quarters of all men are in union (73 percent). The median age at first union is 24.5 years among men age 30-59, nearly identical to the estimate from the preceding survey (24.3 years). Table 6.4 and Figure 6.2 show median age at first union for men and women according to background characteristics. In rural areas, the median age at first marriage is slightly lower than in urban areas for all age groups: 20.6 years in rural areas , compared with 21.5 years in urban areas, for women age 25-49 (Figure 6.2), and 24.2 years in rural areas, compared with 26.9 years in urban areas, for men age 30-59. The data show variations by province: among women, the East province has the earliest age at first union (19.9 years), and the South province and City of Kigali have the latest ages (21.8 years and 21.6 years, respectively). Level of education is the variable that most affects age at first union: among women with no education, the median age is 19.9 years; it is 20.8 years for those with a primary education and 23.2 years for those with a secondary education or higher, indicating that remaining in the school system allows women to delay marriage. Results according to wealth quintile show virtually no differences between the four lowest quintiles; however, women in the richest quintile enter into first union later than women in the other quintiles (age 21.9 years, compared with 20.3 years for the poorest quintile). Other Proximate Determinants of Fertility | 75 Table 6.4 Median age at first marriage Median age at first marriage among women age 25-49 and men age 30-59, by current age and background characteristics, Rwanda 2005 Current age Background characteristic 25-29 30-34 35-39 40-44 45-49 Women 25-49 Men 30-59 Residence Urban 22.0 21.9 22.1 20.7 20.4 21.5 26.9 Rural 20.3 21.0 20.8 20.4 20.3 20.6 24.2 Province Kigali city 22.6 22.4 23.0 19.9 19.6 21.6 27.5 South 21.9 22.1 22.3 21.4 21.0 21.8 25.3 West 20.0 21.0 20.3 20.4 20.4 20.4 23.5 North 19.9 20.5 21.1 20.0 20.3 20.2 24.5 East 19.8 20.2 19.9 20.1 19.5 19.9 24.0 Education No education 19.4 19.8 19.9 19.8 20.3 19.9 23.6 Primary 20.5 21.2 21.2 20.4 20.3 20.8 24.3 Secondary or higher 23.9 23.0 23.6 23.0 (21.9) 23.2 26.8 Wealth quintile Lowest 20.1 20.8 21.0 20.0 19.9 20.3 23.9 Second 20.0 20.8 20.1 20.3 20.5 20.4 23.9 Middle 20.6 21.0 20.9 20.3 20.3 20.6 24.4 Fourth 20.4 20.9 20.8 20.4 20.4 20.6 24.1 Highest 22.0 22.3 22.5 21.4 20.7 21.9 26.7 Total women 20.6 21.1 21.0 20.4 20.3 20.7 na Total men 24.6 24.5 25.5 25.9 23.8 na 24.5 Note: Figures in parentheses are based on 25-49 unweighted cases. na = Not applicable Figure 6.2 Median Age at First Marriage among Women and Men, by Background Characteristics Note: Women age 25-49; men age 30-59 20.7 21.5 20.6 19.9 20.8 23.2 24.5 26.9 24.2 23.6 24.3 26.8 RWANDA RESIDENCE Urban Rural EDUCATION No education Primary Secondary+ 15 17 19 21 23 25 27 29 Age in years Women Men RDHS 2005 76 | Other Proximate Determinants of Fertility The data for men show the same variations as for women. Men in rural areas enter into union for the first time a little earlier than those in urban areas (median 24.2 years for rural, compared with 26.9 years for urban). According to province, men also enter union later in the City of Kigali. Unlike women, however, their age at first union is earliest in the West province (23.5 years). In addition, like women, men’s age at first union rises with level of education: median age of 23.6 years for men with no education, 24.3 years for those with a primary education, and 26.8 years for those with the highest levels of education. Results according to wealth quintile show the same differential between the richest quintile and the four others as seen for the women, with the richest quintile having the highest age at first union (26.7 years compared with 23.9 years in the poorest quintile). 6.4 AGE AT FIRST SEXUAL INTERCOURSE Although marriage is still considered the only socially sanctioned context for sexual activity, prenuptial sex is nevertheless increasingly common. For this reason, the survey asked respondents their age at the time they first had sexual intercourse. Table 6.5 shows percentages for women and men according to age at first sexual intercourse, and the median age at first intercourse for both sexes. Table 6.5 Age at first sexual intercourse: Percentage of women and men who had first sexual intercourse by specific exact ages and median age at first intercourse, according to current age, Rwanda 2005 Percentage who had first sexual intercourse by exact age: Age 15 18 20 22 25 Percentage who never had intercourse Number Median age at first intercourse WOMEN 15-19 5.2 na na na na 87.9 2,585 a 20-24 2.6 19.1 38.6 na na 41.1 2,354 a 25-29 3.8 24.5 50.2 67.8 83.9 10.1 1,738 20.0 30-34 4.0 20.8 41.6 64.3 84.3 3.4 1,466 20.6 35-39 3.8 22.6 44.3 64.2 83.6 1.6 1,134 20.5 40-44 3.3 23.7 49.4 67.8 85.0 1.3 1,135 20.1 45-49 4.0 24.1 48.8 71.9 87.6 0.7 910 20.1 25-49 3.8 23.1 46.8 66.9 84.6 4.1 6,383 20.3 MEN 15-19 15.3 na na na na 77.4 1,102 a 20-24 10.8 26.3 42.2 na na 41.8 946 a 25-29 5.7 24.5 43.1 60.8 79.0 13.0 632 20.6 30-34 2.1 15.0 32.9 54.2 74.3 3.8 509 21.5 35-39 3.7 18.1 37.2 56.5 73.2 1.4 442 21.0 40-44 3.8 24.6 42.9 60.8 75.6 0.8 404 20.6 45-49 1.6 15.8 38.3 60.5 76.7 0.0 378 20.8 50-54 2.9 24.2 50.5 65.9 82.6 0.4 260 19.9 55-59 2.6 23.3 52.1 71.5 83.0 0.0 147 19.8 25-59 3.5 20.5 40.8 59.9 76.9 4.0 2,772 20.8 na = Not applicable a = Omitted because less than 50 percent of the women or men had intercourse for the first time before reaching the beginning of the age group In Rwanda very few women have sexual intercourse at an early age (4 percent by exact age 15). A little more than one in five women (23 percent) first had sexual intercourse before the age of 18. At age 20, nearly half the women have had sexual intercourse. The median age at first sexual intercourse is estimated at 20.3 years, a slight increase from the first survey in 1992, when it was 19.7 years for women age 25-49. However, there has been virtually no change since the 2000 survey (20.1 years). In addition, Other Proximate Determinants of Fertility | 77 the median age at first intercourse is nearly identical to the median age at first union, which seems to confirm that Rwandan women have their first sexual intercourse at the time of their first union. With respect to men, there are also very few who have sexual intercourse for the first time prior to age 15 (4 percent). However, three-quarters of men have had sexual intercourse by age 25 (77 percent). The median age at first sexual intercourse is 20.8 years for men age 25-59. For women age 15-49, the median age has remained more or less unchanged since the last survey. However, unlike women, men’s age at first sexual intercourse is 3.7 years younger than their age at first union. Table 6.6 shows the median age at first sexual intercourse according to background characteristic for both men and women. The results show the greatest variation in median age at first intercourse is by level of education: for women and men alike, the higher the level of education, the later the median age at first sexual intercourse. Among women, this median age ranges from 19.4 years for those with no education to 22.2 years for those with secondary education or higher. Among men, it ranges from 20.4 to 21.3 years, respectively. There is virtually no variation by residence. In the provinces, the median age at first intercourse for women varies slightly from 19.6 years in the East province to 20.8 years in the City of Kigali; for men it varies from 20.3 years in the East province to 20.8 years in the City of Kigali. Results according to wealth quintile show some variation, particularly among women: women in the richest quintile tend to have intercourse for the first time at a later age (21.1 years) than women in the other quintiles, especially the poorest quintile (19.9). For men the differences are marginal. The median age at first sexual intercourse rises with household wealth, from 20 years in the poorest households to 21 years in the richest households, for both women age 25-49 and men age 25-59. Figure 6.3 Median Age at First Intercourse and at First Union among Women 25-49, by Background Characteristics RDHS 2005 20.3 20.6 20.2 19.4 20.3 22.2 20.7 21.5 20.6 19.9 20.8 23.2 RWANDA RESIDENCE Urban Rural EDUCATION No education Primary Secondary+ 15 16 17 18 19 20 21 22 23 24 Age in years Median age at First intercourse First union 78 | Other Proximate Determinants of Fertility Table 6.6 Median age at first sexual intercourse Median age at first sexual intercourse among women age 25-49 and men age 25-59, by current age and background characteristics, Rwanda 2005 Current age Background characteristic 25-29 30-34 35-39 40-44 45-49 Women 25-49 Men 25-59 Residence Urban 20.3 20.8 21.3 20.4 19.9 20.6 20.5 Rural 19.9 20.5 20.4 20.0 20.1 20.2 20.8 Province Kigali city 20.5 21.7 22.0 19.7 19.4 20.8 20.8 South 21.1 21.4 21.4 20.9 20.6 21.1 21.3 West 19.5 20.6 19.9 20.0 20.1 20.0 20.5 North 19.6 20.1 20.8 19.6 20.3 19.9 21.1 East 19.5 19.7 19.5 19.8 19.2 19.6 20.3 Education No education 18.9 19.5 19.2 19.4 20.0 19.4 20.4 Primary 20.0 20.7 20.7 20.1 20.1 20.3 20.8 Secondary or higher 21.7 21.9 23.1 22.5 21.1 22.2 21.3 Wealth quintile Lowest 19.6 20.1 20.3 19.5 19.8 19.9 20.7 Second 19.5 20.5 19.2 20.1 20.3 20.0 20.7 Middle 20.2 20.5 20.4 20.0 19.8 20.2 20.8 Fourth 20.0 20.5 20.5 20.1 20.2 20.3 20.9 Highest 20.7 21.4 21.8 21.0 20.4 21.1 20.6 Total women 20.0 20.6 20.5 20.1 20.1 20.3 na Total men 20.6 21.5 21.0 20.6 20.8 na 20.8 na = Not applicable 6.4 RECENT SEXUAL ACTIVITY Frequency of sexual intercourse is a direct determinant of fertility. Therefore, the survey asked all men and women, regardless of marital status, how long it had been since they last had sexual intercourse. Table 6.7.1 shows the data on most recent sexual activity for women according to background characteristics. Forty-four percent of all women had sexual intercourse in the four weeks preceding the survey. Recent sexual activity was most common among women age 25 to 39, more than 60 percent of whom reported being sexually active in the past four weeks, although there was some decrease at age 35. The results also show that married women are most likely to have been sexually active in the past four weeks (87 percent). Recent sexual activity decreases with marital duration, from a high of 93 percent for marital durations of 0-4 years, to a low of 82 percent for marital durations of 25 years or more. Other Proximate Determinants of Fertility | 79 Table 6.7.1 Recent sexual activity: women Percent distribution of women by timing of last sexual intercourse, according to background characteristics, Rwanda 2005 Timing of last sexual intercourse Background characteristic Within the past 4 weeks Within 1 year1 One or more years Missing Never had sexual intercourse Total Number of women Age 15-19 3.2 2.8 5.9 0.2 87.9 100.0 2,585 20-24 39.9 8.5 8.9 1.7 41.1 100.0 2,354 25-29 66.0 11.6 10.4 2.0 10.1 100.0 1,738 30-34 68.3 10.3 14.1 3.9 3.4 100.0 1,466 35-39 61.6 12.0 21.6 3.2 1.6 100.0 1,134 40-44 54.6 9.1 29.6 5.5 1.3 100.0 1,135 45-49 51.2 6.1 37.0 4.9 0.7 100.0 910 Marital status Never married 1.2 4.7 10.9 1.0 82.2 100.0 4,263 Married 87.4 7.9 3.5 1.1 0.0 100.0 5,510 Divorced/separated/widowed 5.4 18.4 65.0 11.2 0.0 100.0 1,548 Marital duration among women married only once2 0-4 years 92.9 6.0 0.5 0.6 0.0 100.0 1,143 5-9 years 88.5 8.1 1.7 1.6 0.0 100.0 1,158 10-14 years 87.0 8.4 3.2 1.4 0.0 100.0 938 15-19 years 83.1 10.5 5.3 1.1 0.0 100.0 558 20-24 years 82.4 8.5 7.8 1.2 0.0 100.0 520 25 + years 82.3 7.2 9.3 1.3 0.0 100.0 433 Married more than once 87.7 7.7 3.7 0.8 0.0 100.0 760 Residence Urban 34.0 10.3 18.7 2.8 34.2 100.0 1,921 Rural 45.7 7.7 13.9 2.4 30.3 100.0 9,400 Province Kigali city 31.1 10.5 20.2 3.2 35.0 100.0 1,127 South 41.7 7.8 17.0 2.5 30.9 100.0 2,958 West 46.5 6.7 12.0 2.4 32.5 100.0 2,824 North 46.9 8.2 13.1 2.7 29.0 100.0 2,063 East 46.2 8.9 13.8 2.1 29.0 100.0 2,348 Education No education 54.3 9.7 19.8 3.4 12.7 100.0 2,646 Primary 40.6 7.4 12.7 2.1 37.1 100.0 7,591 Secondary or higher 39.8 9.1 16.0 2.7 32.5 100.0 1,084 Current contraceptive method Sterilization (80.8) (2.4) (16.8) (0.0) (0.0) (100.0) 34 Pill 88.6 8.3 2.5 0.7 0.0 100.0 144 Male condom 52.9 40.3 6.4 0.4 0.0 100.0 93 Rhythm/periodic abstinence 69.1 12.2 14.5 3.8 0.3 100.0 276 Other method 90.9 7.1 2.0 0.0 0.0 100.0 538 Not currently using 39.7 7.8 15.6 2.6 34.2 100.0 10,237 Total 43.7 8.1 14.7 2.5 31.0 100.0 11,321 Note: Figures in parentheses are based on 25-49 unweighted cases. 1 Excludes women who had sexual intercourse within the past 4 weeks 2 Excludes women who are not currently married 80 | Other Proximate Determinants of Fertility Women in rural areas reported a significantly higher level of sexual activity in the past four weeks (46 percent) than women in urban areas (34 percent). The percentage of women who had sexual intercourse during the past four weeks decreases as level of education increases (54 percent for those with no education, 41 percent for those with primary, and 40 percent for those with secondary education or higher). Table 6.7.2 presents information on recent sexual activity among men according to background characteristics. The data indicate that 48 percent of men had sexual intercourse in the four weeks preceding the survey. The proportion of men who are sexually active increases with age and then begins declining at age 45. Sexual activity peaks between the ages of 35 and 44 (84 percent). The results show that, like women, married men are more sexually active (90 percent). Results by marital duration, although less consistent than those for women, show decreasing sexual activity with increasing marital duration, from 93 percent for durations of 0-4 years, to 83 percent for durations of 25 years or more. Table 6.7.2 Recent sexual activity: men Percent distribution of men by timing of last sexual intercourse, according to background characteristics, Rwanda 2005 Timing of last sexual intercourse Background characteristic Within the past 4 weeks Within 1 year1 One or more years Missing Never had sexual intercourse Total Number of men Age 15-19 0.9 4.6 17.1 0.0 77.4 100.0 1,102 20-24 19.2 10.7 28.2 0.1 41.8 100.0 946 25-29 59.5 11.9 15.5 0.1 13.0 100.0 632 30-34 77.7 9.9 8.4 0.2 3.8 100.0 509 35-39 84.4 10.5 3.6 0.0 1.4 100.0 442 40-44 84.4 11.2 3.7 0.0 0.8 100.0 404 45-49 82.1 11.3 6.7 0.0 0.0 100.0 378 50-54 75.7 16.5 7.3 0.0 0.4 100.0 260 55-59 76.1 15.4 8.5 0.0 0.0 100.0 147 Marital status Never married 1.7 9.1 27.3 0.1 61.9 100.0 2,196 Married 89.7 9.3 1.0 0.0 0.0 100.0 2,500 Divorced/separated/widowed 14.6 36.2 48.6 0.0 0.7 100.0 125 Marital duration among men married only once2 0-4 years 93.0 6.8 0.0 0.2 0.0 100.0 458 5-9 years 92.4 7.4 0.2 0.0 0.0 100.0 471 10-14 years 89.4 10.3 0.3 0.0 0.0 100.0 370 15-19 years 90.8 8.5 0.7 0.0 0.0 100.0 227 20-24 years 87.0 12.1 0.9 0.0 0.0 100.0 219 25 + years 82.5 14.3 3.2 0.0 0.0 100.0 300 Married more than once 89.2 8.8 2.0 0.0 0.0 100.0 455 Residence Urban 37.5 16.0 21.0 0.1 25.3 100.0 840 Rural 49.8 8.6 12.7 0.0 28.8 100.0 3,980 Province Kigali city 31.6 20.4 22.6 0.2 25.3 100.0 523 South 46.7 9.0 15.6 0.0 28.6 100.0 1,250 West 53.1 7.1 12.3 0.1 27.4 100.0 1,185 North 51.0 8.9 9.5 0.1 30.4 100.0 845 East 47.8 9.6 14.2 0.0 28.3 100.0 1,017 Education No education 62.4 13.3 9.1 0.1 15.1 100.0 839 Primary 44.4 8.6 14.7 0.0 32.2 100.0 3,389 Secondary or higher 45.3 12.3 18.3 0.1 23.9 100.0 592 Total 47.6 9.9 14.2 0.1 28.2 100.0 4,820 1 Excludes men who had sexual intercourse within the last 4 weeks 2 Excludes men who are not currently married Other Proximate Determinants of Fertility | 81 Results by residence show a sizeable differential in the frequency of sexual activity between rural (50 percent) and urban (38 percent) areas. 6.6 EXPOSURE TO THE RISK OF PREGNANCY Women are less exposed to the risk of pregnancy for a period of time following childbirth. Exposure to the risk of pregnancy depends on several factors including the duration of postpartum amenorrhea—the period between childbirth and the return of ovulation—and the period when a woman abstains from sexual intercourse (postpartum abstinence). These two factors jointly determine which women are insusceptible to becoming pregnant and the length of the period of insusceptibility. Women are considered insusceptible if they are abstaining from intercourse following childbirth and/or are amenorrheic. In the latter case, the risk of pregnancy is negligible even if sexual activity is resumed without contraceptive protection. Table 6.8 shows the percentage of births in the three years preceding the survey for which mothers are postpartum amenorrheic, abstaining, and insusceptible, by the number of months since the birth. It also shows median and mean durations for these indicators. In Rwanda, 42 percent of women who gave birth during the three years preceding the survey were amenorrheic. A little more than seven in ten women (84 percent) remained amenorrheic for 5 months; approximately seven in ten women (73 percent) were still amenorrheic at 9 months; and 11 percent remained so at 26-27 months. Beyond 28 months, the proportion of women for whom ovulation had not yet returned varied between 2 percent and 7 percent. The median duration of postpartum amenorrhea is 14.3 months, and the mean is 15.4 months. The duration, intensity, and frequency of exclusive breastfeeding, which affects the return of ovulation (see Chapter 10 - Nutrition), is partly responsible for these relatively long durations, which have changed little since 2000. Postpartum abstinence is not tradition- ally practiced in Rwanda. Only 10 percent of women had not resumed sexual intercourse 4-5 months following the birth of their last child. The median and mean durations for postpartum abstinence are very short (0.6 months and 4.4 months, respectively). Mothers were insusceptible to the risk of pregnancy for 46 percent of births in the three years preceding the survey. The mean duration of the period of insusceptibility is 16.8 months. The median duration is 15.3 months. Table 6.8 Postpartum amenorrhea, abstinence, and insusceptibility Percentage of births in the three years preceding the survey for which mothers are postpartum amenorrheic, abstaining, and insusceptible, by number of months since birth, and median and mean durations, Rwanda 2005 Percentage of births for which the mother is: Months since birth Amenorrheic Abstaining Insusceptible Number of births < 2 98.6 41.8 98.6 270 2-3 90.4 16.4 91.3 339 4-5 83.5 10.4 86.0 320 6-7 75.8 10.5 76.7 283 8-9 72.9 12.7 76.7 290 10-11 67.2 8.4 69.9 300 12-13 53.0 7.6 56.3 363 14-15 50.7 8.2 54.4 291 16-17 43.5 8.5 47.4 280 18-19 29.3 7.1 33.4 284 20-21 24.1 6.0 28.3 273 22-23 19.9 7.4 23.7 285 24-25 16.3 9.7 21.7 317 26-27 10.7 7.5 14.8 339 28-29 7.3 8.7 14.8 272 30-31 7.6 7.6 13.7 271 32-33 5.1 8.3 12.6 330 34-35 1.8 5.2 7.0 363 Total 41.8 10.5 45.7 5,469 Median 14.3 0.6 15.3 na Mean 15.4 4.4 16.8 na Note: Estimates are based on status at the time of the survey. na = Not applicable 82 | Other Proximate Determinants of Fertility Table 6.9 shows the median duration of postpartum amenorrhea, abstinence, and insusceptibility following births in the three years preceding the survey by background characteristics. Although entirely dependant on the duration of amenorrhea and abstinence, the duration of postpartum insusceptibility varies with age. Women 30 years of age and older have longer periods of insusceptibility (15.7 months for amenorrhea, 0.6 months for abstinence, and 16.3 months for insusceptibility) than women under the age of 30 (13.2 months, 0.7 months, and 14.1 months, respectively). The median duration of amenorrhea is longer in rural areas (14.8 months) than in urban areas (12 months). However, the median duration of abstinence is longer in urban areas (1.8 months) than in rural areas (0.6 months). The period of insusceptibility is longer in rural areas (15.6 months, compared with 13.2 months for urban areas). By province, the City of Kigali has the shortest period of amenorrhea (9.4 months) and the longest period of abstinence (2.5 months). Results differ according to level of education: women with the highest levels of education have the shortest periods of amenorrhea (10.0 months); women with no education have the longest periods of amenorrhea (13.9 months). Table 6.9 Median duration of postpartum insusceptibility by background characteristics Median number of months of postpartum amenorrhea, postpartum abstinence, and postpartum insusceptibility following births in the three years preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Postpartum amenorrhea Postpartum abstinence Postpartum insusceptibility Number of births Age 15-29 13.2 0.7 14.1 2,809 30-49 15.7 0.6 16.3 2,660 Residence Urban 12.0 1.8 13.2 759 Rural 14.8 0.6 15.6 4,711 Province Kigali city 9.4 2.5 10.5 410 South 15.8 0.6 16.9 1,306 West 15.5 0.6 15.9 1,441 North 14.3 0.6 15.1 1,078 East 12.7 0.6 13.4 1,234 Education No education 13.9 0.7 14.9 1,520 Primary 15.0 0.6 15.8 3,516 Secondary or higher 10.0 1.7 10.2 433 Total 14.3 0.6 15.3 5,469 Note: Medians are based on current status. 6.7 MENOPAUSE Women cease being exposed to the risk of pregnancy when they reach menopause. For the survey, women were considered menopausal if they were neither pregnant nor postpartum amenorrheic but had not had a menstrual period in the six months preceding the survey, or if they reported themselves as having entered menopause. Other Proximate Determinants of Fertility | 83 Table 6.10 shows the percentage of women age 30 to 49 who are menopausal. Overall, 6 percent of women age 30 to 49 reported being menopausal. The proportion increases with age from 1 percent for women age 30 to 34, to 7 percent at age 45, to 32 percent for women age 48 to 49. Table 6.10 Menopause Percentage of women age 30-49 who are menopausal, by age, Rwanda 2005 Age Percentage menopausal1 Number of women 30-34 1.3 1,466 35-39 2.0 1,134 40-41 3.7 448 42-43 4.0 496 44-45 6.7 400 46-47 14.5 404 48-49 31.7 296 Total 5.6 4,645 1 Percentage of all women who are not pregnant and not postpartum amenorrheic whose last menstrual period occurred six or more months preceding the survey Fertility Preferences | 85 FERTILITY PREFERENCES 7 Data on fertility preferences is used to evaluate the effectiveness of couples’ efforts to control their own fertility and to assess Rwanda’s future contraceptive needs not only for birth spacing, but to limit the total number of births. To obtain information about fertility preferences, the RDHS-III asked women how many additional children they wanted to have in the future, how long they wanted to wait before having their next child, and the total number of children desired. Data on attitudes and opinions about procreation have always been somewhat controversial. Some researchers believe responses to questions about fertility preferences are subject to three potential flaws: first, they represent viewpoints that are subject to change rather than firm convictions; second, they do not take into account the effects of social pressure and the attitudes of other family members, particularly the husband, who can have enormous influence over reproductive decisions; and third, the data are obtained from a sample of women of differing ages with differing birth histories. Their responses relate to medium- or long-term goals that may change over time or be of limited predictive value for young and/or recently-married women. The responses of older women and/or women at the end of their childbearing years are inevitably influenced by their birth histories. Despite possible problems of interpretation, the data on fertility preferences can assist in understanding the factors affecting fertility in Rwanda, where contraceptive prevalence remains low and fertility levels remain high. This analysis covers only men and women who were married at the time of the survey. 7.1 DESIRE FOR (MORE) CHILDREN The desire to have (more) children in the future generally correlates with a woman’s age and the number of living children she and/or her husband have. The RDHS-III asked currently married women a series of questions designed to discern their desire to delay the next birth or to stop having children. The results are presented in Table 7.1 by number of living children (including the current pregnancy) at the time of the survey. A little more than two in five women (42 percent) reported wanting no more children, while more than half (52 percent) wanted to have another child. Among the women who wanted (more) children in the future, 12 percent wanted another child within two years, 39 percent wanted to delay the next birth by two or more years, and 2 percent wanted to have another child but were uncertain as to when. In general, over three-quarters of married women in Rwanda (83 percent) can be considered potential candidates for family planning: those who do not want any more children (42 percent), and those who want to delay their next birth (41 percent). The percentage of women who want no more children has increased compared with the previous survey (RDHS-II), from 33 percent in 2000 to 42 percent in 2005. 86 | Fertility Preferences Table 7.1 Fertility preferences by number of living children Percent distribution of currently married women and currently married men by desire for children, according to number of living children, Rwanda 2005 Number of living children 1 Desire for children 0 1 2 3 4 5 6+ Total WOMEN Have another soon2 85.4 20.2 13.7 10.0 5.0 4.0 1.1 11.8 Have another later3 3.9 69.1 59.7 49.1 31.6 21.8 7.1 38.8 Have another, undecided when 2.6 2.7 1.6 1.9 1.4 1.2 0.4 1.6 Undecided 0.4 0.7 2.0 1.5 2.0 2.8 3.2 2.0 Want no more 1.1 6.0 21.3 34.8 56.8 64.3 81.1 42.2 Sterilized4 0.0 0.0 0.5 0.7 0.7 0.9 0.6 0.5 Declared infecund 6.5 1.4 1.0 1.8 2.6 5.0 6.3 3.1 Missing 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of women 195 809 1,046 982 830 589 1,060 5,510 MEN Have another soon2 46.1 18.8 13.9 8.7 7.9 6.5 2.8 10.7 Have another later3 43.5 72.8 59.9 52.1 34.4 28.3 12.7 39.7 Have another, undecided when 5.3 2.8 1.9 2.1 2.0 1.2 1.4 2.0 Undecided 0.0 0.0 0.6 1.5 2.1 1.6 0.5 0.9 Want no more 2.3 5.2 20.9 34.3 51.3 59.3 77.2 43.7 Sterilized4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Declared infecund 2.8 0.4 2.0 1.2 1.2 1.7 5.0 2.4 Missing 0.0 0.0 0.8 0.2 1.1 1.5 0.4 0.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of men 127 313 393 393 325 243 706 2,500 1 Includes current pregnancy (for women) 2 Wants next birth within 2 years 3 Wants to delay next birth for 2 or more years 4 Includes both female and male sterilization Unlike most countries in sub-Saharan Africa, the proportion of men in Rwanda who want no more children (44 percent) is similar to that of women (42 percent). The same is true for the proportion of men who want (more) children later (52 percent for men and women). Like women, the proportion of men who want (more) children soon decreases as parity increases, and the proportion of men who want no more children increases as parity increases (Figure 7.1). In fact, the percentage of men who want more children ranges from 76 percent among those with two children to 44 percent among those with four children, to 17 percent among those who have six children or more. It should be noted that at each parity level (Table 7.1) the differences between men and women who want more children are minimal. As expected, the proportion of women who want no more children increases considerably with the number of living children, from 1 percent for women with no children, to 35 percent for women with three children, to 81 percent for those with six children or more. Women who want no more children have presumably reached their desired family size, or cumulative fertility, and should be using a contraceptive method to avoid unwanted pregnancies. Finally, the data show that 92 percent of women with no children would like to have a child, and the majority of these women (85 percent) would like to have one soon. Fertility Preferences | 87 Table 7.2 shows the percentages of women and men who want no more children by background characteristics. Results by residence show that the proportions of women and men who want no more children are somewhat higher in urban areas (49 percent for women; 48 percent for men) than in rural areas (42 percent for women; 43 percent for men). By province, the proportion of women who want no more children ranges from a low of 40 percent in the West province to a high of 52 percent in the City of Kigali. Results by level of education show that women with no education are more likely to want to limit births (48 percent) than women with primary (40 percent) or secondary education (46 percent). In addition, with respect to wealth quintile, only the richest quintile stands out with a significantly higher proportion of women wanting no more children (47 percent) than the other quintiles (38 to 44 percent). Unlike women, higher levels of education for men correlate with higher proportions wanting no more children (47 percent, compared to 43 percent among men with a primary education). The results according to province for men are similar to those for women: the City of Kigali has the highest proportion of men who have reached their desired number of children (50 percent). In addition, for men, the desire to limit births correlates closely with household standard of living: the proportion of men who want no more children increases from the poorest quintile (41 percent) to the richest quintile (52 percent). Married women who do not use contraception and who reported not wanting any more children (desiring, therefore, to limit births) or who reported wanting to wait two or more years before their next birth (desiring, therefore, to space births) are considered to have unmet family planning need. Women who reported having unmet need and women currently using contraception make up the total potential demand for family planning. Figure 7.1 Proportion of Currently Married Women and Men Who Want No More Children, by Number of Living Children RDHS 2005 0 1 2 3 4 5 6 + Number of living children 0 20 40 60 80 100 Percent Men Women 88 | Fertility Preferences Table 7.2 Desire to limit childbearing Percentage of currently married women who want no more children, by number of living children and the percentage of currently married women and currently married men who want no more children by background characteristics, Rwanda 2005 Number of living children 1 Background characteristic 0 1 2 3 4 5 6+ Women Men Residence Urban (3.5) 11.2 26.5 49.9 68.4 73.6 86.3 49.3 47.6 Rural 0.8 5.3 20.9 33.0 55.6 64.0 81.2 41.7 43.1 Province Kigali city * 15.2 33.4 55.0 80.1 (91.4) 92.2 52.0 50.2 South (1.7) 2.5 16.6 33.8 55.1 71.8 88.4 40.7 42.0 West (0.0) 5.7 19.7 28.6 56.1 53.2 70.3 39.5 43.7 North (0.0) 3.9 26.3 30.8 49.3 59.7 83.5 44.1 47.9 East (2.7) 8.5 21.6 42.4 62.7 69.4 88.7 44.6 39.5 Education No education (2.1) 10.2 23.3 36.9 58.9 60.2 78.8 48.3 44.2 Primary 1.0 5.0 20.8 33.9 55.6 65.6 82.8 40.3 43.0 Secondary or higher * 4.2 26.8 43.0 65.6 79.6 90.7 45.6 47.3 Wealth quintile Lowest (0.0) 6.8 21.0 35.4 57.5 67.4 82.2 42.7 40.7 Second (0.0) 5.8 22.6 33.3 55.1 67.3 83.7 43.7 43.8 Middle * 3.3 24.0 39.3 54.7 57.8 77.8 42.3 41.9 Fourth (0.0) 5.2 14.9 26.8 57.2 61.2 82.7 38.2 41.3 Highest (1.9) 9.6 26.8 45.6 63.3 73.4 82.5 47.4 52.2 Total 1.1 6.0 21.8 35.5 57.5 65.2 81.8 42.7 43.7 Note: Women and men who have been sterilized are considered to want no more children. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 1 Includes current pregnancy 7.2 NEED FOR FAMILY PLANNING SERVICES Table 7.3 presents estimates for unmet need, met need, and total demand for family planning for currently married women by background characteristics. Unmet need for family planning remains high: nearly two in five married women (38 percent) have expressed need for family planning. The majority of these women would be using contraception to space births (25 percent), while 13 percent would use contraception to limit births. If married women with unmet need for contraception were able to satisfy this need, that is, if they were to use contraception, contraceptive prevalence would reach 55 percent. This is approximately three times the current rate. By way of comparison, the expressed need for contraception in the RDHS-II survey of 2000 was 49 percent. The total potential demand for family planning—the proportion of women with unmet need plus women who are already using contraception—is broken down into two categories: need for birth spacing (32 percent), and need for limiting births (23 percent). Among currently married women, only 31 percent of the total potential demand for family planning is being met. However, this is an increase from 27 percent in the RDHS-II. Fertility Preferences | 89 Table 7.3 Need for family planning among currently married women Percentage of currently married women with unmet need for family planning, and with met need for family planning, and the total demand for family planning, by background characteristics, Rwanda 2005 Unmet need for family planning1 Met need for family planning (currently using) 2 Total demand for family planning3 Background characteristic For spacing For limiting Total For spacing For limiting Total For spacing For limiting Total Percentage of demand satisfied Number of women Age 15-19 18.7 3.2 21.9 1.1 2.2 3.2 19.8 5.4 25.1 12.9 65 20-24 33.0 3.0 36.0 9.8 2.9 12.7 42.8 5.9 48.7 26.0 980 25-29 34.4 5.9 40.3 10.8 6.6 17.3 45.2 12.5 57.6 30.1 1,254 30-34 30.9 9.9 40.8 9.9 10.4 20.3 40.8 20.3 61.1 33.3 1,112 35-39 20.1 21.6 41.7 6.3 14.2 20.4 26.4 35.7 62.1 32.9 807 40-44 8.5 31.3 39.8 2.1 17.6 19.7 10.6 48.9 59.4 33.1 739 45-49 2.4 21.4 23.8 0.4 13.7 14.1 2.8 35.1 37.9 37.1 554 Residence Urban 20.5 13.9 34.4 13.5 18.1 31.6 34.0 32.0 66.0 47.9 744 Rural 25.1 13.3 38.4 6.5 8.7 15.2 31.6 22.0 53.6 28.3 4,766 Province Kigali city 16.8 14.1 30.9 14.3 21.2 35.5 31.1 35.3 66.4 53.5 407 South 24.5 13.0 37.5 6.9 8.0 14.8 31.4 21.0 52.4 28.3 1,411 West 25.4 13.2 38.7 6.9 7.5 14.5 32.4 20.8 53.2 27.3 1,427 North 26.1 13.9 40.1 6.4 9.6 16.0 32.6 23.5 56.0 28.5 1,058 East 24.4 13.4 37.8 7.3 11.6 18.9 31.8 25.0 56.8 33.3 1,208 Education No education 22.1 17.9 40.0 2.8 7.3 10.1 24.9 25.2 50.1 20.1 1,359 Primary 26.0 12.3 38.3 7.8 9.3 17.1 33.8 21.6 55.4 30.9 3,672 Secondary or higher 19.4 9.2 28.6 18.1 22.3 40.4 37.4 31.5 68.9 58.5 479 Wealth quintile Lowest 26.0 14.0 40.0 4.1 6.9 11.0 30.1 20.9 51.1 21.6 1,136 Second 24.5 13.0 37.5 5.3 9.9 15.2 29.8 22.9 52.7 28.9 1,123 Middle 24.8 14.7 39.5 8.0 7.7 15.7 32.8 22.4 55.2 28.5 1,112 Fourth 25.6 12.5 38.1 7.2 7.6 14.8 32.8 20.1 52.9 28.0 1,144 Highest 21.1 12.8 33.9 13.3 18.6 31.8 34.4 31.4 65.8 48.4 995 Total 24.5 13.4 37.9 7.4 9.9 17.4 31.9 23.3 55.3 31.4 5,510 1 Unmet need for spacing includes pregnant women whose pregnancy was mistimed; amenorrheic women who are not using family planning and whose last birth was mistimed, or whose last births was unwanted but now say they want more children; and fecund women who are neither pregnant nor amenorrheic, who are not using any method of family planning, and say they want to wait 2 or more years for their next birth. Also included in unmet need for spacing are fecund women who are not using any method of family planning and say they are unsure whether they want another child or who want another child. Unmet need for limiting refers to pregnant women whose pregnancy was unwanted; amenorrheic women who are not using family planning, whose last child was unwanted and who do not want any more children; and fecund women who are neither pregnant nor amenorrheic, who are not using any method of family planning, and who want no more children. Excluded from the unmet need category are pregnant and amenorrheic women who became pregnant while using a method (these women are in need of a better method of contraception). 2 Using for spacing is defined as women who are using some method of family planning and say they want to have another child or are undecided whether to have another. Using for limiting is defined as women who are using and who want no more children. Note that the specific methods used are not taken into account here. 3 Nonusers who are pregnant or amenorrheic whose pregnancy was the result of a contraceptive failure are not included in the category of unmet need, but are included in total demand for contraception (since they would have been using had their method not failed). The need for family planning varies according to background characteristic. With respect to age, unmet need is lower among younger women age 20-24 (36 percent) and among older women age 45-49 (24 percent). In the other age groups, the proportions are approximately 40 percent. Up until the age of 34, unmet need for family planning relates essentially to birth spacing while, after age 40, women express greater need for contraception to limit births. Results by residence show that the proportion of women with unmet need is somewhat higher in rural areas (38 percent) than urban areas (34 percent). Because women in rural areas use contraception far less, the total demand for family planning services satisfied is much higher in urban areas (48 percent) 90 | Fertility Preferences than rural areas (28 percent). The need for contraception to space births is always much greater than the need for contraception to limit births, regardless of residence. The total potential demand has risen, regardless of residence, compared with 2000 RDHS-II levels, which were 61 percent for urban areas (66 percent in 2005) and 47 percent for rural areas (53 percent in 2005). By province, the proportion of women with unmet need for family planning ranges from a low of 31 percent in the City of Kigali to a high of 40 percent in the North province. The City of Kigali also has the highest total potential demand (66 percent); the South province has the lowest (52 percent). With respect to level of education, unmet need for family planning is greater among women with no education (40 percent) than among women with a primary education (38 percent) and women with a secondary education or higher (29 percent). Because of the positive correlation between family planning and level of education, the total demand for family planning services satisfied is much higher among women with a secondary education or higher (59 percent) than among women with a primary education (31 percent) or women with no education (20 percent). According to wealth quintile, unmet need seems to be especially higher for women in the lowest four quintiles (around 40 percent). The total potential demand, however, is greater among women in the richest households (66 percent) and is 48 percent satisfied. The lowest demand satisfied is found among women in the poorest households (22 percent). 7.3 IDEAL NUMBER OF CHILDREN Women’s reproductive behavior can be influenced by the ideal number of children they would like to have and the ideal number their husband/partner would like to have. In order to determine this ideal number, the RDHS-III asked all women surveyed one of the following two questions: • To women with no living children: If you could choose the exact number of children you would like to have in your lifetime, how many would you have? • To women with living children: If you could go back to the time when you had no children and choose the exact number of children you would like to have in your lifetime, how many would you have chosen? These seemingly simple questions may be embarrassing, particularly for women with living children who may specify an ideal number that differs from the number of children they already have. It may also be difficult for a woman to specify an ideal number that is lower than her current cumulative fertility. The responses to these questions are presented in Table 7.4. Four percent of women did not give a numeric response, giving instead a general answer such as “However many God gives me,” “I don’t know,” or “any number.” The proportion of women who gave this type of response varies between 3 and 6 percent. The average ideal number of children for all women and for married women at the time of the survey was 4.3 and 4.5, respectively. This ideal number of children lower than the TFR (6.1), which means that women would like to have a lower cumulative fertility. An examination of the distribution of reported ideal family size shows that the ideal number of children for 40 percent of women is 4. However, 16 percent of women have an ideal number of 6 or more. This proportion ranges from 14 percent among women with no living children to 20 percent among women with 4 living children, to 29 percent among those with at least 6 living children. Fertility Preferences | 91 In general, there is a positive correlation between current family size and ideal family size, which ranges from 4 children for all women with no children, to 4.8 for those with 6 children or more. Among women who were married at the time of the survey, ideal family size varies inconsistently from 4.4 children for women with no children to 4.9 for women with 6 or more children. The ideal number of children for men is approximately 4 (all men and married men). As with women, men’s ideal number of children is lower than the TFR. Table 7.4 Ideal number of children Percent distribution of all women and all men by ideal number of children, and mean ideal number of children for all women and currently married women and for all men and currently married men, according to the number of living children, Rwanda 2005 Number of living children 1 Ideal number of children 0 1 2 3 4 5 6+ Total WOMEN 0 1.4 1.1 0.3 0.4 0.4 1.1 1.0 1.0 1 0.6 1.9 0.3 0.8 0.5 0.2 0.4 0.7 2 9.3 11.6 11.4 6.9 6.6 7.4 6.9 8.9 3 13.9 20.0 13.2 13.2 5.3 8.0 7.4 12.5 4 38.4 38.7 49.8 40.7 48.0 33.0 35.3 40.3 5 18.2 13.7 11.9 18.9 17.0 21.2 14.6 16.7 6+ 14.1 9.7 9.7 16.2 19.6 25.6 28.8 16.3 Non-numeric responses 4.0 3.3 3.3 2.9 2.7 3.5 5.6 3.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of women 4,201 1,312 1,395 1,282 1,070 781 1,280 11,321 Mean ideal number of children2 for: All women 4.2 3.9 4.0 4.3 4.5 4.7 4.8 4.3 Currently married women 4.4 4.1 4.1 4.4 4.6 4.8 4.9 4.5 MEN 0 8.1 0.7 0.4 0.5 0.4 0.6 0.7 4.2 1 0.3 0.3 0.3 1.0 1.5 0.6 0.4 0.5 2 10.6 12.1 11.6 8.9 10.0 10.6 11.0 10.7 3 18.7 25.0 18.9 14.7 8.5 14.6 14.6 17.3 4 34.9 41.0 48.7 42.1 41.8 31.0 43.2 38.7 5 15.6 12.3 12.4 19.3 18.4 14.4 10.2 14.7 6+ 11.2 7.8 7.3 13.0 18.2 25.8 18.5 13.1 Non-numeric responses 0.6 0.7 0.4 0.5 1.2 2.4 1.4 0.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of men 2,317 366 415 409 338 255 721 4,820 Mean ideal number of children2 for: All men 3.8 3.8 3.9 4.2 4.3 4.5 4.3 4.0 Currently married men 3.8 3.8 3.9 4.2 4.3 4.5 4.4 4.2 1 Includes current pregnancy 2 Means are calculated excluding respondents who gave non-numeric responses. Table 7.5 shows the mean ideal number of children for all women and all men according to current age and background characteristics. The ideal number of children does not vary much by age: for women age 20 to 29 it is 4.1 children and for women age 40 to 49 it is 4.5 children. 92 | Fertility Preferences However, this ideal number varies significantly by residence, province, level of education, and household standard of living. For women in rural areas, the ideal family size is larger (4.4) than for women in urban areas (3.8). With respect to province, the mean ideal number of children is lower among women in the City of Kigali (3.7) than among those in the other provinces, where it ranges from 4.2 in the East province to 4.6 in the West province. Also, the higher the level of education, the lower the mean ideal number of children: 4.6 for women with no education, compared with 3.6 for women with a secondary education or higher. As with level of education, the desired cumulative fertility decreases as household wealth increases, from 4.5 children in the poorest households to 3.9 in the richest. Table 7.5 Mean ideal number of children Mean ideal number of children for all women and all men, by age and background characteristics, Rwanda 2005 Age Background characteristic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 All women All men Residence Urban 3.9 3.6 3.6 4.0 4.0 4.0 4.3 3.8 3.7 Rural 4.4 4.3 4.3 4.4 4.5 4.6 4.6 4.4 4.0 Province Kigali city 3.6 3.6 3.5 3.7 3.8 4.5 3.9 3.7 3.2 South 4.3 4.2 4.1 4.3 4.3 4.3 4.6 4.3 4.2 West 4.6 4.5 4.4 4.6 4.8 4.7 4.7 4.6 4.4 North 4.2 4.0 4.1 4.3 4.4 4.8 4.7 4.3 3.8 East 4.2 4.1 4.3 4.3 4.3 4.2 4.2 4.2 3.7 Education No education 4.5 4.4 4.4 4.5 4.7 4.8 4.6 4.6 4.3 Primary 4.3 4.2 4.2 4.3 4.4 4.4 4.5 4.3 4.0 Secondary or higher 3.4 3.5 3.4 3.7 3.8 4.0 (4.1) 3.6 3.5 Wealth quintile Lowest 4.5 4.4 4.3 4.5 4.3 4.7 4.4 4.5 4.1 Second 4.3 4.3 4.2 4.3 4.5 4.6 4.7 4.4 4.1 Middle 4.4 4.3 4.2 4.4 4.4 4.6 4.6 4.4 4.0 Fourth 4.3 4.2 4.3 4.3 4.5 4.4 4.6 4.3 4.1 Highest 3.9 3.7 3.7 4.0 4.2 4.1 4.3 3.9 3.6 All women 4.3 4.1 4.1 4.3 4.4 4.5 4.5 4.3 na All men 4.0 3.7 3.7 4.1 4.3 4.1 4.1 na 4.0 Note: Figures in parentheses are based on 25-49 unweighted cases. na = Not applicable 7.4 FERTILITY PLANNING STATUS For each child born in the five years preceding the survey and for the current pregnancy (if the respondent was pregnant), each mother was asked if she had wanted to be pregnant at that time, if she would have preferred to be pregnant later, or had if she not wanted to become pregnant at all. The responses to these questions are used to measure couples’ effectiveness in controlling their fertility. Such questions require a woman to concentrate in order to remember her desires accurately at one or more specific times during the past five years. The data may be subject to rationalization, as an undesired pregnancy often results in the birth of a child to which the mother has become attached. Fertility Preferences | 93 Table 7.6 shows that more than four in five births (84 percent) were wanted. Most of these births (60 percent) occurred at the desired time; 24 percent occurred earlier than the women would have liked. Unwanted pregnancies represented approximately 15 percent of the births. The great majority of births are desired and arrive according to the desired timing, regardless of birth order. However, first births are better planned than births 2, 3, 4, or higher. In the RDHS-III, 82 percent of first births arrived according to the desired timing, compared with 63 percent of second births, and 50 percent of births 4 or higher. With respect to age of the mother, the best planned births occurred among women who had their children before the age of 20 and between the age of 20 and 29. Conversely, births among women who had children when they were older (age 45 to 49) seem to be less well planned: 37 percent arrived according to the desired timing, 8 percent arrived later, and 56 percent were unwanted. Table 7.6 Fertility planning status Percent distribution of births in the five years preceding the survey (including current pregnancy), by planning status of the birth, according to birth order and mother’s age at birth, Rwanda 2005 Planning status of birth Birth order and mother’s age at birth Wanted then Wanted later Wanted no more Total1 Number of births Birth order 1 82.3 4.5 12.8 100.0 1,755 2 63.1 28.4 8.2 100.0 1,665 3 60.0 30.0 9.8 100.0 1,548 4+ 50.1 28.6 20.9 100.0 4,648 Mother’s age at birth <20 67.6 11.4 20.9 100.0 559 20-24 67.4 22.7 9.5 100.0 2,609 25-29 60.4 30.1 9.3 100.0 2,545 30-34 56.9 29.1 13.5 100.0 1,905 35-39 50.4 22.4 27.1 100.0 1,254 40-44 50.3 12.6 36.7 100.0 668 45-49 36.8 7.5 55.7 100.0 76 Total 59.8 24.4 15.4 100.0 9,615 1 Includes those with missing information Table 7.7 compares the total wanted fertility rate (TWFR) with the current total fertility rate (TFR) for the five years preceding the survey. Calculation of the TWFR is the same as for the TFR, except that unwanted births are omitted. If all unwanted births were eliminated, the TFR for Rwandan women would be 4.6 children, rather than 6.1 children. 94 | Fertility Preferences The TWFR is higher in rural areas (4.8) than in urban areas (3.6) and, in particular, the City of Kigali (3.4). It decreases as level of education and wealth quintile increase. The lowest TWFRs are found among women with the highest levels of education (3.3 compared with 5.4 for women with no education) and the greatest household wealth (3.6 for the richest quintile; 4.7 to 5.0 for the other quintiles). Table 7.7 Wanted fertility rates Total wanted fertility rates and total fertility rates for the three years preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Total wanted fertility rate Total fertility rate Residence Urban 3.6 4.9 Rural 4.8 6.3 Province Kigali city 3.4 4.3 South 4.4 5.6 West 4.9 6.6 North 4.8 6.4 East 4.8 6.5 Education No education 5.4 7.0 Primary 4.6 6.1 Secondary or higher 3.3 4.3 Wealth quintile Lowest 4.8 6.1 Second 4.7 6.3 Middle 5.0 6.7 Fourth 4.8 6.4 Highest 3.6 5.0 Total 4.6 6.1 Note: Rates are calculated based on births to women age 15-49 in the period 1-36 months preceding the survey. The total fertility rates are the same as those presented in Table 4.2. Maternal and Child Health | 95 MATERNAL AND CHILD HEALTH 8 The RDHS-III collected information about the health of mothers and their children born in the five years preceding the survey. This chapter covers antenatal, postnatal, and delivery care, characteristics of neonates, childhood vaccination coverage, and the prevalence and treatment of common childhood illnesses, specifically, respiratory infections, fever, and diarrhea. The findings in this chapter help identify the most important problems in maternal and child health and reproductive health. Comparison of the results with those of previous surveys assists in the planning and evaluation of national health policies and programs. 8.1 ANTENATAL CARE Monitoring of pregnant women through antenatal care visits helps reduce risks and complications during pregnancy and delivery. For this reason, the RDHS-III asked women who had had a live birth in the five years preceding the survey if they had received antenatal care (ANC). Table 8.1 shows the distribution of the women’s most recent live births in the past five years according to type of medical personnel consulted by the mother during the pregnancy and the mother’s background characteristics. During the RDHS-III, all categories of ANC providers consulted by the mother were recorded; however, if more than one provider was mentioned, only the provider with the highest qualifications was considered in the tabulations. For the most recent live births in the five years preceding the survey, nearly all of the mothers (94 percent) received antenatal care from trained personnel. This proportion has remained relatively stable since 1992, when 94 percent of births benefited from antenatal care (Figure 8.1). In the RDHS-III, ANC was mainly provided by nurses or midwives, auxiliary nurses/midwives, trained traditional birth attendants (88 percent) or, in very low percentages, doctors (7 percent). In the current Rwandan health system, ANC at public or certified health facilities is almost always provided by nurses (doctors only intervene if complications are noticed in the mother in the course of the ANC visit). The data do not vary much by background characteristics: the proportion of mothers who received antenatal care is greater than 90 percent for all variables. However, the proportion of women who consulted with a doctor during these visits is higher in urban areas (15 percent) than in rural areas (5 percent), higher among women in the City of Kigali (19 percent) than among those in the other provinces (2 to 9 percent), and higher among women with a secondary education or higher (18 percent, compared with 4 percent for mothers with no education). The proportion of those who consulted with a doctor is also higher among women in the richest quintile (14 percent compared with 4 to 6 percent in the other quintiles). These results can be explained by the concentration of doctors in urban areas, particularly the City of Kigali. To be effective, antenatal care must be sought early in the pregnancy and, more importantly, must continue regularly through to delivery. The World Health Organization (WHO) recommends at least four ANC visits at regular intervals throughout the pregnancy, as does the Rwandan health system. 96 | Maternal and Child Health Table 8.1 Antenatal care Percent distribution of women who had a live birth in the five years preceding the survey by antenatal care (ANC) provider during pregnancy for the most recent birth, according to background characteristics, Rwanda 2005 Background characteristic Doctor Nurse/midwife/ auxiliary nurse/midwife/ trained traditional birth attendant Trained personnel Untrained traditional birth attendant/ other No one Total1 Number Mother’s age at birth <20 7.6 84.7 92.3 0.0 7.7 100.0 276 20-34 6.8 88.4 95.2 0.0 4.6 100.0 3,777 35-49 6.8 85.9 92.8 0.0 7.0 100.0 1,372 Birth order 1 6.9 87.7 94.6 0.0 5.1 100.0 875 2-3 7.7 87.7 95.4 0.1 4.4 100.0 1,706 4-5 6.9 87.6 94.5 0.0 5.3 100.0 1,349 6+ 5.8 87.3 93.1 0.0 6.7 100.0 1,495 Residence Urban 15.4 77.4 92.8 0.2 7.0 100.0 774 Rural 5.4 89.3 94.7 0.0 5.1 100.0 4,651 Province City of Kigali 18.8 73.7 92.5 0.2 7.3 100.0 427 South 6.7 88.3 95.0 0.0 5.0 100.0 1,357 West 9.2 83.6 92.9 0.0 6.7 100.0 1,395 North 1.9 94.7 96.6 0.0 2.9 100.0 1,052 East 4.3 90.0 94.3 0.0 5.7 100.0 1,194 Education No education 4.2 87.6 91.8 0.0 7.6 100.0 1,552 Primary 6.5 88.7 95.2 0.0 4.7 100.0 3,404 Secondary or higher 18.1 79.2 97.3 0.2 2.5 100.0 469 Wealth quintile Lowest 6.0 85.7 91.6 0.0 8.1 100.0 1,163 Second 4.3 90.1 94.4 0.0 5.4 100.0 1,124 Middle 5.8 90.1 95.9 0.0 3.8 100.0 1,097 Fourth 4.6 90.6 95.2 0.0 4.5 100.0 1,069 Highest 14.4 80.9 95.3 0.1 4.6 100.0 972 Total 6.8 87.6 94.4 0.0 5.3 100.0 5,425 Note: If more than one source of ANC was mentioned, only the provider with the highest qualifications is considered in this tabulation. 1 Includes those with missing information Maternal and Child Health | 97 Table 8.2 shows the number of ANC visits and the timing of the first visit. Although the great majority of Rwandan mothers sought antenatal care, the number of visits was below the standard set by WHO and Rwandan health officials. Only 13 percent of women who had a live birth in the five years preceding the survey met the standard of at least four ANC visits. More than two-thirds of the women had 2 or 3 ANC visits (68 percent). This percentage has remained virtually unchanged since 2000, when it was 69 percent. It should also be noted that 13 percent of mothers had only one visit and 5 percent of mothers had no ANC visits at all. This situation has also remained unchanged since 2000. Results by residence show that the propor- tion of women who made at least 4 ANC visits is slightly higher in urban areas (18 percent) than in rural areas (13 percent). It should be noted that Rwandan women seek their first prenatal visit late in pregnancy. In fact, half of the women did not have an ANC visit until their sixth or seventh month of pregnancy; 27 percent had their first visit between the fourth and fifth month; and 9 percent did not receive antenatal care until the eighth month or later. Only 5 percent of women made their first visit before the fourth month of pregnancy, and this proportion is twice as high in urban areas (14 percent) as in rural areas (7 percent). The median number of months of pregnancy at the first ANC visit is 6.4 Table 8.2 Number of antenatal care visits and timing of first visit Percent distribution of women who had a live birth in the five years preceding the survey by number of antenatal care (ANC) visits for the most recent birth, and by the timing of the first visit, and among women with ANC, median months pregnant at first visit, according to residence, Rwanda 2005 Residence Number and timing of ANC visits Urban Rural Total Number of ANC visits None 7.0 5.1 5.4 1 9.5 13.5 13.0 2-3 65.5 68.6 68.1 4+ 17.6 12.6 13.3 Total1 100.0 100.0 100.0 Number of months pregnant at time of first ANC visit No antenatal care 7.0 5.1 5.4 <4 13.5 7.0 7.9 4-5 26.5 27.4 27.3 6-7 45.3 50.5 49.8 8+ 6.9 9.7 9.3 Total1 100.0 100.0 100.0 Median months pregnant at first ANC visit 6.2 6.5 6.4 Number of women 774 4,651 5,425 1 Includes those with missing information Figure 8.1 Trends in Antenatal Care and Delivery, Rwanda 1992, 2000, and 2005 94 26 25 92 31 27 94 39 28 Antenatal care Delivery assisted by trained personnel Delivery in a health facility 0 20 40 60 80 100 Percent RDHS-I 1992 RDHS-II 2000 RDHS-III 2005 98 | Maternal and Child Health for the country as a whole, 6.2 in urban areas, and 6.5 in rural areas. The lateness of the first ANC visit can be explained by a Rwandan tradition whereby women do not speak of their pregnancy until it is visible. Also, it may be that women wait until the sixth month of pregnancy to have their first prenatal visit in order to receive a tetanus vaccination. Components of ANC The effectiveness of antenatal care depends not only on the type of examinations performed at the visit, but also on the counseling and preventive measures given to avoid the risk of miscarriage and other pregnancy complications. The RDHS-III collected data on this important aspect of prenatal monitoring by asking women if, during their ANC visits for the most recent birth: they were told about the danger signs of pregnancy complications, they received specific medical examinations (weight, height, and blood pressure measurements), and they were given blood and urine tests. In addition, women were asked if they had received iron supplements and antimalarial drugs. The answers to these questions are presented in Table 8.3 by background characteristics. Table 8.3 Components of antenatal care Percentage of women with a live birth in the five years preceding the survey who received antenatal care for the most recent birth, by content of antenatal care, and percentage of women with a live birth in the five years preceding the survey who received iron tablets or syrup or antimalarial drugs for the most recent birth, according to background characteristics, Rwanda 2005 Among women who received antenatal care Background characteristic Informed of signs of pregnancy complications Weight measured Height measured Blood pressure measured Urine sample taken Blood sample taken Number of women who received antenatal care Received iron tablets or syrup Received anti- malarial drugs Number of women Mother’s age at birth <20 5.7 95.4 52.0 70.1 10.9 32.4 254 25.1 4.4 276 20-34 5.5 93.4 56.6 71.4 7.8 25.7 3 597 27.9 6.2 3,777 35-49 8.1 94.4 53.0 71.1 6.2 19.6 1 273 29.4 4.9 1,372 Birth order 1 6.8 93.0 55.8 71.5 11.6 32.4 828 27.2 7.6 875 2-3 6.5 93.8 55.8 71.2 8.6 25.9 1 629 27.2 6.0 1,706 4-5 4.1 94.5 57.6 73.0 6.1 23.3 1 275 29.6 5.7 1,349 6+ 7.3 93.3 53.0 69.6 5.5 19.4 1 392 28.5 4.6 1,495 Residence Urban 7.8 96.1 58.4 88.1 21.8 63.6 720 33.8 9.8 774 Rural 5.9 93.3 55.0 68.5 5.3 18.2 4 404 27.2 5.1 4,651 Province City of Kigali 8.5 97.2 59.5 84.1 28.4 62.8 396 31.0 8.7 427 South 7.2 96.2 64.1 86.2 7.5 24.1 1 289 36.2 8.3 1,357 West 8.2 90.9 48.6 67.2 6.8 27.1 1 296 33.6 4.0 1,395 North 3.4 95.5 48.4 62.8 3.6 16.6 1 017 20.6 2.6 1,052 East 4.3 91.4 58.5 62.0 4.9 16.0 1 126 18.3 6.8 1,194 Education No education 6.9 92.5 50.2 65.6 5.6 19.4 1 425 27.4 4.5 1,552 Primary 5.6 93.9 57.8 71.9 6.3 24.4 3 241 27.1 5.8 3,404 Secondary or higher 7.8 96.1 55.7 84.0 23.0 41.8 458 38.3 9.8 469 Wealth quintile Lowest 5.0 92.4 53.8 67.5 4.6 17.2 1 065 24.0 3.2 1,163 Second 5.4 93.9 56.2 68.1 4.4 18.3 1 061 22.9 5.0 1,124 Middle 6.2 94.1 52.9 65.5 4.6 21.9 1 052 27.4 5.3 1,097 Fourth 6.2 93.9 56.4 72.6 6.9 23.0 1 018 31.1 7.0 1,069 Highest 8.3 94.4 58.4 84.3 18.9 45.0 927 36.8 9.0 972 Total 6.2 93.7 55.5 71.3 7.6 24.6 5 124 28.2 5.8 5,425 Maternal and Child Health | 99 Very few women (6 percent) were informed of the signs of pregnancy complications, a situation that has remained unchanged since 2000, when the proportion of women who received this information was also 6 percent. There is little variation in this percentage by background characteristic. Weight is by far the most common ANC measurement taken (94 percent), regardless of the mother’s background characteristics. Only 71 percent of women reported having their blood pressure measured; 56 percent said their height was measured. Taking blood and urine samples for testing was least likely to occur during an ANC visit (25 percent and 8 percent, respectively). Overall, women in rural areas, women with no education, and women living in the poorest households are the least likely to receive blood pressure measurements or blood and urine analyses as part of their ANC visits. The proportion of women who receive iron supplements and antimalarial drugs is very low: 28 percent receive iron supplements and 6 percent receive antimalarial medication. However, it should be noted that nutritional iron supplements are not systematically prescribed for pregnant women in Rwanda except in the case of anemia. It should also be noted that the practice of giving antimalarial drugs preventively has been introduced only recently. The results by residence and wealth quintile reveal large disparities. In rural areas, 27 percent of the women reported receiving iron tablets or syrups and 5 percent said they received antimalarial drugs; the levels are higher in urban areas (34 percent for iron tablets and 10 percent for antimalarial drugs). Results by wealth quintile reveal similar differentials: in the poorest households, 24 percent of women received iron supplements, compared with 37 percent in the richest households; 3 percent received antimalarial medication, compared with 9 percent in the richest households. Results by province show that the East and North provinces have the lowest rates for iron supplementation: 21 percent for the East and 18 percent for the North, compared with a high of 36 percent for the South province. Women in the North (3 percent) and West (4 percent) provinces were the least likely to have received antimalarial drugs. Results for some ANC components have changed little since 2000: weight measurement (93 percent); information on inherent pregnancy risks (6 percent); and preventive treatment by antimalarial drugs (8 percent, compared with 6 percent currently), although proportions have increased for the other types of examinations. Tetanus vaccinations Neonatal tetanus is a major cause of death among newborns in most developing countries. Tetanus toxoid injections given to the mother during pregnancy protect both mother and child against this disease. To be fully protected, a pregnant woman should receive two doses of the vaccine during her pregnancy; however, if she has already been vaccinated, for example during a previous pregnancy, one more dose is sufficient. It is important to note that the information presented here does not take into account the woman’s “vaccination history;” some women may have received the vaccine prior to the period under consideration. If the vaccination was received within the past 10 years, the woman will retain some immunity. Table 8.4 shows that antitetanus vaccination coverage for pregnant mothers remains low, and it has dropped since the last survey. Only 63 percent of women who had a live birth in the five years preceding the survey received one or two or more doses of antitetanus vaccine during their most recent pregnancy, compared with 70 percent in 2000. Those who are fully protected (along with their newborns) because they received two or more doses of antitetanus vaccine, represent only 22 percent of pregnant women; those who are partially protected (unless they were vaccinated previously) by receiving one dose 100 | Maternal and Child Health of the vaccine, represent 41 percent of the mothers surveyed. The age of the mother seems to be an important factor in tetanus coverage: the proportion of women who received one or two or more doses is higher among younger mothers (84 percent for the youngest age group; 33 percent for the oldest). Similarly, first births are better protected than higher order births: 85 percent for first births, compared with 26 percent for births order 6 and above. In addition, mothers in rural areas (62 percent, compared with 71 percent in urban areas), mothers in the South province (64 percent), and mothers with no education (54 percent, compared with 73 percent for women with a secondary education or higher) are less likely to receive the tetanus vaccine. The data by wealth quintile show no major variations with respect to vaccination coverage. Table 8.4 Tetanus toxoid injections Percent distribution of women who had a live birth in the five years preceding the survey by number of tetanus toxoid injections received during pregnancy for the most recent birth, according to background characteristics, Rwanda 2005 Background characteristic None One injection Two or more injections Don’t know/ missing Total Number Mother’s age at birth <20 15.1 45.2 39.1 0.6 100.0 276 20-34 25.9 47.7 25.4 1.0 100.0 3,777 35-49 66.4 22.2 10.3 1.1 100.0 1,372 Birth order 1 13.5 37.9 47.2 1.4 100.0 875 2-3 16.0 55.9 27.1 1.0 100.0 1,706 4-5 33.4 50.4 15.3 1.0 100.0 1,349 6+ 72.9 17.9 8.4 0.9 100.0 1,495 Residence Urban 26.4 43.9 27.4 2.3 100.0 774 Rural 37.1 40.7 21.4 0.8 100.0 4,651 Province City of Kigali 21.3 42.8 33.2 2.7 100.0 427 South 35.1 46.4 17.9 0.5 100.0 1,357 West 38.6 39.0 20.9 1.6 100.0 1,395 North 38.0 40.0 21.0 1.0 100.0 1,052 East 35.6 38.0 25.9 0.4 100.0 1,194 Education No education 45.1 34.0 19.6 1.3 100.0 1,552 Primary 32.8 44.2 22.3 0.6 100.0 3,404 Secondary or higher 24.4 42.1 30.5 2.9 100.0 469 Wealth quintile Lowest 37.6 41.3 20.5 0.6 100.0 1,163 Second 40.8 38.5 19.8 0.9 100.0 1,124 Middle 35.9 41.4 22.0 0.7 100.0 1,097 Fourth 33.5 43.1 22.4 0.9 100.0 1,069 Highest 29.0 41.4 27.4 2.2 100.0 972 Total 35.6 41.1 22.3 1.0 100.0 5,425 8.2 DELIVERY CARE Place of delivery Because every pregnancy may be subject to complications, women are advised to deliver their babies in a health facility so they will have access to emergency services if needed during labor and delivery. For this reason, the RDHS-III asked women where they had given birth and who had assisted the delivery. Table 8.5 shows that less than one-third of the women delivered their babies at a health facility. In fact, 70 percent of the births in the five years preceding the survey took place at home. The Maternal and Child Health | 101 incidence of home births increases with the age of the mother: 59 percent among mothers under the age of 20; 78 percent among mothers age 35 to 49. The proportion of home births also increases with the child’s birth order: 49 percent of first births took place at home, compared with 80 percent of births order 6 and above. In addition, home births were more frequent in rural areas (75 percent, compared with 44 percent in urban areas), and among women with no education (81 percent) or only a primary education (71 percent) than among women with a secondary education or higher (32 percent). By province, with the exception of the City of Kigali, where only 42 percent of births take place at home, the proportion of home births ranges from a low of 69 percent in the North to 78 percent in the East province. Moreover, mothers who have not received ANC were more likely to give birth at home (89 percent, compared with 49 percent for women who made four or more ANC visits). Finally, the proportion of women who delivered at home decreases as household wealth increases, from 82 percent for women in the poorest households, to 40 percent for those in the richest households. Table 8.5 Place of delivery Percent distribution of live births in the five years preceding the survey by place of delivery, according to background characteristics, Rwanda 2005 Health facility Background characteristic Public sector Private sector Home Total1 Number of births Mother’s age at birth <20 37.5 2.7 58.8 100.0 533 20-34 28.1 1.4 69.3 100.0 6,366 35-49 19.7 0.6 77.9 100.0 1,815 Birth order 1 47.9 2.3 48.5 100.0 1,616 2-3 26.0 1.5 71.3 100.0 2,905 4-5 21.5 1.2 76.1 100.0 2,05 6+ 17.6 0.4 80.4 100.0 2,138 Residence Urban 49.7 5.2 44.1 100.0 1,228 Rural 23.2 0.6 74.8 100.0 7,487 Province City of Kigali 50.1 7.4 41.7 100.0 655 South 27.5 0.7 70.7 100.0 2,122 West 24.1 0.5 73.4 100.0 2,290 North 28.3 1.5 69.3 100.0 1,716 East 20.6 0.6 77.5 100.0 1,932 Mother’s education No education 17.1 0.7 80.6 100.0 2,470 Primary 26.8 0.9 71.0 100.0 5,513 Secondary or higher 61.2 5.9 31.7 100.0 732 Antenatal care visits2 None 8.8 1.1 88.9 100.0 291 1-3 26.0 1.3 71.4 100.0 4,400 4 or more 46.4 2.6 48.9 100.0 724 Wealth quintile Lowest 16.0 0.6 82.0 100.0 1,845 Second 19.0 0.3 79.6 100.0 1,794 Middle 22.4 0.9 75.3 100.0 1,785 Fourth 27.2 0.7 70.6 100.0 1,742 Highest 54.1 4.4 40.3 100.0 1,548 Total 26.9 1.3 70.4 100.0 8,715 1 Includes those with missing information 2 Includes only the most recent birth in the five years preceding the survey 102 | Maternal and Child Health Conversely, in urban areas, more than 55 percent of births took place at a health facility; in the City of Kigali, this proportion is 58 percent. Similarly, 67 percent of women with a secondary education or higher delivered their babies at a health facility. Finally, it should be noted that these results show no change from the two previous DHS surveys with respect to place of delivery for women in Rwanda (Figure 8.1). Assistance during delivery To avoid the risk of maternal death, women should be assisted during delivery by personnel who have received training in normal childbirth and who are able, if needed, to diagnose, treat, and refer complications. Table 8.6 shows the distribution of births in the five years preceding the survey by person providing assistance during the delivery. These results show that still too few women are assisted by trained personnel during childbirth. This is a crucial problem that threatens the health of both mother and child. Six in ten women (61 percent) were not assisted by trained personnel during delivery (43 percent were assisted by untrained traditional birth attendants, and 17 percent received no assistance at all). Table 8.6 Assistance during delivery Percent distribution of live births in the five years preceding the survey by person providing assistance during delivery, according to background characteristics, Rwanda 2005 Background characteristic Doctor Nurse/midwife/ auxiliary midwife/trained traditional birth attendant Trained personnel Untrained traditional birth attendant Relative/ other No one Total1 Number of births Mother’s age at birth <20 6.0 44.0 50.0 42.1 0.2 7.7 100.0 533 20-34 5.3 34.7 40.0 44.0 0.4 15.2 100.0 6,366 35-49 3.7 26.9 30.6 41.0 0.6 27.6 100.0 1,815 Birth order 1 9.3 51.8 61.1 34.0 0.1 4.6 100.0 1,616 2-3 5.1 33.0 38.1 47.5 0.3 13.9 100.0 2,905 4-5 3.2 29.9 33.1 45.4 0.6 20.5 100.0 2,056 6+ 3.4 24.4 27.8 42.6 0.7 28.7 100.0 2,138 Residence Urban 13.6 49.5 63.1 26.6 0.8 9.2 100.0 1,228 Rural 3.6 31.0 34.6 46.0 0.4 18.7 100.0 7,487 Province City of Kigali 15.0 46.7 61.8 26.9 1.2 9.8 100.0 655 South 6.4 33.5 39.9 43.0 0.2 16.9 100.0 2,122 West 5.2 29.2 34.4 45.1 0.7 19.4 100.0 2,290 North 2.2 31.9 34.1 50.3 0.5 14.9 100.0 1,716 East 2.4 36.1 38.5 40.9 0.1 20.2 100.0 1,932 Mother’s education No education 2.7 24.5 27.2 46.4 0.4 25.9 100.0 2,470 Primary 4.6 34.7 39.2 44.8 0.5 15.1 100.0 5,513 Secondary or higher 16.3 56.6 72.9 21.6 0.0 5.4 100.0 732 Wealth quintile Lowest 2.1 25.1 27.2 51.1 0.6 20.9 100.0 1,845 Second 2.9 27.3 30.2 49.0 0.7 19.9 100.0 1,794 Middle 3.1 30.8 33.9 46.6 0.3 18.5 100.0 1,785 Fourth 3.8 35.9 39.7 44.1 0.2 15.9 100.0 1,742 Highest 14.5 51.9 66.4 22.7 0.4 10.4 100.0 1,548 Total 5.0 33.6 38.6 43.3 0.4 17.3 100.0 8,715 Note: If the respondent mentioned more than one person attending during delivery, only the most qualified person is considered in this tabulation. 1 Includes those with missing information Maternal and Child Health | 103 Although only 39 percent of births were delivered with the assistance of qualified personnel, this proportion has increased since 1992 (Figure 8.1). The proportion of women who received no assistance increases with age of the mother (8 percent for women under age 20, 28 percent for women age 35-49) and with birth order (5 percent for first births, compared with 29 percent for birth order 6 or above). Unassisted deliveries are more frequent in rural areas (19 percent) than in urban areas (9 percent). Similarly, in the provinces, the proportion of unassisted deliveries ranges from a high of 20 percent in the East province to a low of 10 percent in the City of Kigali. A woman’s level of education is related to the delivery conditions: 26 percent of women with no education delivered without assistance, compared with 15 percent of women with a primary education and 5 percent of women with higher educations. In addition, results by household wealth quintile show that deliveries assisted by trained personnel are more than twice as frequent in the richest quintile as in the poorest (66 percent, compared with 27 percent) (see Figure 8.2). Delivery characteristics For live births in the five years preceding the survey, mothers were asked if the delivery took place by caesarean section (C-section); they were also asked the child’s birth weight and size. It should be noted that Rwandan health officials hold that C-sections should not exceed 10 percent of deliveries in a health facility. Table 8.7 shows that only 3 percent of live births were delivered by C-section, a figure well below the Rwandan health stipulation. As expected, the frequency of this intervention, although very low, is higher among younger women, first births, births in urban areas, births among educated women, and births among women in the richest wealth quintile. Figure 8.2 Children Whose Delivery Was Assisted by Trained Personnel RDHS 2005 39 63 35 27 39 73 27 30 34 40 66 RWANDA RESIDENCE Urban Rural MOTHER'S EDUCATION None Primary Secondary or higher WEALTH QUINTILE Lowest Second Middle Fourth Highest 0 20 40 60 80 Percent 104 | Maternal and Child Health Table 8.7 Delivery characteristics Percentage of live births in the five years preceding the survey delivered by caesarean section, and percent distribution by birth weight and by mother’s estimate of baby’s size at birth, according to background characteristics, Rwanda 2005 Birth weight Size of child at birth Background characteristic Delivery by C-section Not weighed Less than 2.5 kg 2.5 kg or more Total1 Very small Smaller than average Average or larger Total1 Number of births Mother’s age at birth <20 3.5 61.1 2.7 32.9 100.0 3.2 11.9 84.5 100.0 533 20-34 3.3 67.2 1.7 29.5 100.0 3.2 9.4 86.9 100.0 6,366 35-49 1.5 73.4 1.2 23.9 100.0 3.7 9.6 86.4 100.0 1,815 Birth order 1 5.6 50.9 3.7 42.6 100.0 5.1 13.1 81.2 100.0 1,616 2-3 3.3 69.2 1.5 27.9 100.0 2.6 8.6 88.4 100.0 2,905 4-5 1.8 71.8 1.2 25.5 100.0 3.1 8.5 87.6 100.0 2,056 6+ 1.6 76.1 0.8 21.5 100.0 3.2 9.1 87.4 100.0 2,138 Residence Urban 7.5 37.3 2.5 58.1 100.0 3.8 8.8 86.7 100.0 1,228 Rural 2.2 73.1 1.6 23.7 100.0 3.3 9.7 86.7 100.0 7,487 Province City of Kigali 9.2 32.9 2.9 62.2 100.0 4.1 8.0 87.3 100.0 655 South 3.4 70.8 2.0 26.0 100.0 4.7 9.3 86.0 100.0 2,122 West 2.7 74.4 1.0 22.4 100.0 2.7 10.5 86.3 100.0 2,290 North 1.7 68.2 1.4 28.7 100.0 2.8 8.0 88.7 100.0 1,716 East 1.7 69.5 2.1 27.0 100.0 2.9 10.6 85.8 100.0 1,932 Mother’s education No education 2.1 78.2 1.0 18.8 100.0 3.4 10.6 85.5 100.0 2,470 Primary 2.6 68.7 1.8 27.9 100.0 3.2 9.4 86.9 100.0 5,513 Secondary or higher 8.7 29.5 3.3 66.0 100.0 3.6 7.2 89.0 100.0 732 Wealth quintile Lowest 1.3 81.0 1.4 16.1 100.0 3.4 10.2 85.9 100.0 1,845 Second 2.2 76.0 1.2 21.6 100.0 3.3 9.6 86.9 100.0 1,794 Middle 1.7 72.4 1.5 23.9 100.0 2.8 10.1 86.5 100.0 1,785 Fourth 2.4 69.4 1.9 27.0 100.0 4.0 9.6 86.0 100.0 1,742 Highest 7.8 37.0 2.5 58.4 100.0 3.3 8.1 88.1 100.0 1,548 Total 2.9 68.1 1.7 28.5 100.0 3.3 9.5 86.7 100.0 8,715 1 Includes those with missing information Table 8.7 shows results for birth weight. According to mothers’ reports, for 68 percent of live births, the infants were not weighed, the reason being that most of them were born at home. The proportion of children not weighed was particularly high for mothers age 35 to 49 (73 percent) and for birth order six and above (76 percent). Similarly, nearly three-quarters of children in rural areas were not weighed at birth (73 percent). The proportion not weighed among infants whose mothers had no education was 78 percent, and the highest proportion not weighed was found in the poorest quintile (81 percent). Because of the high proportion of births for which data are not available, and the wide variations by background characteristics, the figure for low-birth-weight babies is heavily biased (almost certainly underestimated) and therefore should be viewed with caution. Maternal and Child Health | 105 Mothers were also asked if they believed their child was very large, larger than average, average, smaller than average, or very small at birth. Eighty-seven percent of the mothers said they believed their child was average or larger than average. This belief does not vary significantly by respondents’ background characteristics. Ten percent of mothers said their child was smaller than average and 3 percent said it was very small. Births believed to be smaller than average were reported most frequently for mothers under the age of 20 at the time of the birth (12 percent), first births (13 percent), mothers in rural areas (10 percent), mothers in the West (11 percent) and East (11 percent) provinces, mothers with no education (11 percent), and mothers in the poorest quintile (10 percent). 8.3 POSTNATAL CARE A significant proportion of maternal and newborn deaths in the neonatal period take place within 48 hours following delivery. For this reason, Safe Motherhood programs have recently placed special emphasis on the importance of postnatal checkups, recommending that all women have a postnatal visit within two days following the delivery. During the survey, therefore, women whose most recent birth took place outside a health facility were asked if they had received a postnatal checkup, and the timing of this checkup following delivery. Table 8.8 shows that more than one in four women (29 percent) delivered their babies in a health facility; it is presumed that these women received postnatal care prior to leaving the facility. However, practically none of the women who delivered outside a health facility received a postnatal checkup within the 42 days immediately following the delivery (95 percent), and this proportion remains very high for all background characteristics. Only 4 percent of women who did not deliver at a health facility received a postnatal checkup within two days following the delivery. The proportions who received postnatal care, though low, are highest in the City of Kigali (5 percent), among the most educated women (10 percent), and among women in the richest quintile (8 percent). The proportion of mothers who did not receive a postnatal checkup has remained stable since 2000, when it was 96 percent, compared with 95 percent in 2005. 106 | Maternal and Child Health 8.4 VACCINATION OF CHILDREN To assess Rwanda’s Expanded Program on Immunization (EPI), the RDHS-III gathered information on vaccinations for all children who were born in the five years preceding the survey. The EPI largely follows the World Health Organization’s (WHO) guidelines for vaccinating children. These guidelines stipulate that, to be considered fully immunized, children should receive the following vaccines by the age of 12 months: one dose of BCG (against tuberculosis), three doses of DPT (against diphtheria, pertussis, and tetanus), three doses of the oral polio vaccine, and one dose of the measles vaccine. Vaccines against Haemophilus influenza and hepatitis were introduced in Rwanda in January 2001. Each child who is vaccinated receives a card on which all of the vaccines received are recorded. The information on vaccinations was gathered from two sources: where vaccination cards were available, Table 8.8 Postnatal care Percentage of live births in the five years preceding the survey for which the mother delivered in a health facility, and percent distribution of women whose last live birth in the five years preceding the survey occurred outside a health facility by timing of postnatal care, according to background characteristics, Rwanda 2005 Timing of first postnatal checkup for births occurring outside a health facility Background characteristic Delivered in a health facility Number of births 0-2 days after delivery 3-6 days after delivery 7-41 days after delivery Don’t know/ missing Did not receive postnatal checkup1 Total Number of births occurring outside a health facility Mother’s age at birth <20 44.6 276 2.6 0.0 0.5 0.0 96.9 100.0 153 20-34 31.2 3,777 3.6 0.4 0.6 0.3 95.1 100.0 2,600 35-49 21.1 1,372 4.1 0.1 0.4 0.4 95.1 100.0 1,083 Birth order 1 54.2 875 3.9 0.5 1.4 0.3 93.9 100.0 400 2-3 28.9 1,706 3.5 0.3 0.5 0.2 95.5 100.0 1,214 4-5 25.6 1,349 4.0 0.4 0.3 0.5 94.7 100.0 1,004 6+ 18.5 1,495 3.6 0.1 0.4 0.3 95.7 100.0 1,218 Residence Urban 55.0 774 5.4 0.7 1.1 0.6 92.2 100.0 348 Rural 25.0 4,651 3.5 0.2 0.4 0.3 95.5 100.0 3,487 Province City of Kigali 58.1 427 5.2 0.9 0.9 1.3 91.6 100.0 179 South 29.2 1,357 3.9 0.2 0.5 0.2 95.2 100.0 961 West 25.8 1,395 4.3 0.2 0.5 0.4 94.5 100.0 1,036 North 30.3 1,052 2.9 0.5 0.1 0.6 95.9 100.0 734 East 22.4 1,194 3.1 0.1 0.7 0.0 96.2 100.0 926 Education No education 19.1 1,552 2.8 0.2 0.4 0.3 96.2 100.0 1,256 Primary 28.6 3,404 3.8 0.2 0.5 0.4 95.2 100.0 2,429 Secondary or higher 67.8 469 9.9 1.6 1.1 0.0 87.5 100.0 151 Wealth quintile Lowest 17.8 1,163 2.7 0.1 0.2 0.3 96.7 100.0 956 Second 21.2 1,124 2.5 0.3 0.1 0.0 97.1 100.0 886 Middle 22.5 1,097 3.3 0.3 0.6 0.5 95.2 100.0 850 Fourth 28.5 1,069 4.6 0.4 0.6 0.6 93.7 100.0 764 Highest 61.0 972 7.9 0.2 1.5 0.4 90.0 100.0 379 Total 29.3 5,425 3.7 0.3 0.5 0.3 95.2 100.0 3,836 1 Includes women who received the first postnatal checkup after 41 days Maternal and Child Health | 107 the interviewer copied the information directly onto the questionnaire; where cards were not available— because the mother never had one, or it was unavailable at the time of the survey, or she had lost it— mothers were asked to recall whether or not the child had received each of the vaccines covered by the survey. Table 8.9 presents vaccination coverage results by source of information for children age 12 to 23 months, thereby including only children who had reached the age by which they should be fully immunized. According to the vaccination cards, 66 percent of children age 12 to 23 months are fully immunized. When information from both information sources is considered, the percentage of children fully immunized reaches 75 percent. Vaccination coverage based solely on the mother’s report occurred in only 9 percent of cases. Of the fully immunized children, 69 percent received their vaccinations before their first birthday as recommended by WHO and the Rwanda EPI. Only 3 percent of children age 12 to 23 months had not received any vaccinations at the time of the survey. Table 8.9 Vaccinations by source of information Percentage of children age 12-23 months who received specific vaccines at any time before the survey, by source of information (vaccination card or mother’s report), and percentage vaccinated by 12 months of age, Rwanda 2005 DPT Polio1 Source of information BCG 1 2 3 0 1 2 3 Measles All2 No vacci- nations Number of children Vaccinated at any time before the survey Vaccination card 75.1 75.7 74.8 72.7 61.4 75.6 74.6 72.7 66.9 65.9 0.0 1,234 Mother’s report 21.3 21.1 18.6 14.3 12.5 20.9 18.4 11.6 18.7 9.3 2.5 392 Either source 96.5 96.8 93.4 87.0 73.9 96.5 93.0 84.3 85.6 75.2 2.5 1,626 Vaccinated by 12 months of age3 96.4 96.5 93.0 86.4 73.8 96.2 92.6 83.7 79.4 69.3 3.8 1,626 1 Polio 0 is the polio vaccination given at birth. 2 BCG, measles and three doses each of DPT and polio vaccine (excluding polio vaccine given at birth) 3 For children whose information was based on the mother’s report, the proportion of vaccinations given during the first year of life was assumed to be the same as for children with a written record of vaccination. According to the vaccination cards, 75 percent of children age 12 to 23 months have received the BCG vaccine; 21 percent more have received it based on mothers’ reports. Therefore, a total of 97 percent of children had been immunized against tuberculosis at the time of the survey, almost all of them before their first birthday (12 months). According to both sources of information, the proportion of children who received the first dose of DPT is also very high (97 percent); however, DPT vaccination coverage gradually declines for subsequent doses, from 97 percent for the first dose, to 93 percent for the second dose, to 87 percent for the third dose. These figures represent a dropout rate of 10 percent between the first and third doses of DPT. Because polio vaccine is given at the same time as DPT, its levels are expected to be similar, which is the case in Rwanda. For this vaccine as well, coverage gradually declines for subsequent doses, from 96 percent for the first dose, to 93 percent for the second dose, to 84 percent for the third dose. The dropout rate is 13 percent between the first and third doses. According to both sources of information, just under three-quarters of children received polio dose 0 at birth (74 percent). 108 | Maternal and Child Health According to both sources of information, 86 percent of children received the measles vaccine; however, only 79 percent received it before the age of 12 months. Although the proportion of fully immunized children had declined between the two previous surveys, from 87 percent in 1992 to 76 percent in 2000, the results of the current survey show some improvement in vaccination coverage, which has maintained its 2000 level (76 percent) (Figure 8.3). Table 8.10 shows the results for vaccination coverage among children age 12 to 23 months according to background characteristics of mother and child. The data show practically no disparity by sex (75 percent for males and females). However, complete coverage declines with children’s birth order: 79 percent for the first birth; 75 percent for birth orders 2-3 and 4-5; and 73 percent for children of birth order 6 and above. By residence, complete vaccination coverage is higher in rural areas (76 percent) than in urban areas (71 percent), primarily because the City of Kigali has the lowest vaccination coverage in the country (62 percent). This low proportion in the City of Kigali is due in part to the high dropout rate between polio doses (22 percent between the first and third doses). The East province has the second lowest coverage rate (67 percent) after the City of Kigali. Complete vaccination coverage increases steadily with the mother’s level of education, although the differentials are not great: 72 percent for children whose mothers have no education; 76 percent for children whose mothers have a primary education; and 78 percent for children whose mothers have a secondary education or higher. However, the proportion of vaccinated children varies little according to household wealth: it is highest in the fourth quintile (79 percent); in the other quintiles the proportions are all approximately 74 percent. Figure 8.3 Trends in Vaccination Coverage among Children Age 12-23 Months, Rwanda 1992, 2000, and 2005 BCG DPT 3 Polio 3 Measles All vaccines 0 20 40 60 80 100 Percent RDHS-I 1992 RDHS-II 2000 RDHS-III 2005 97 97 97 91 86 87 91 88 84 91 87 86 87 76 75 Maternal and Child Health | 109 Table 8.10 Vaccinations by background characteristics Percentage of children age 12-23 months who received specific vaccines at any time before the survey (according to a vaccination card or the mother’s report), and percentage with a vaccination card, by background characteristics, Rwanda 2005 DPT Polio1 Background characteristic BCG 1 2 3 0 1 2 3 Measles All2 No vacci- nations Per- centage with a vacci- nation card seen Number of children Sex Male 95.6 96.0 92.2 85.3 72.4 95.4 91.7 83.3 84.9 75.0 3.4 75.1 844 Female 97.4 97.6 94.8 88.8 75.5 97.6 94.4 85.3 86.4 75.4 1.7 76.8 782 Birth order 1 95.9 96.1 92.2 87.6 76.8 96.4 93.4 84.8 90.7 79.0 3.1 77.5 324 2-3 97.7 97.7 96.3 88.8 73.8 97.1 93.7 84.1 85.5 74.6 1.7 75.6 519 4-5 96.2 97.2 92.6 85.9 73.4 96.9 93.7 84.2 85.6 74.7 2.4 75.5 380 6+ 95.6 95.7 91.5 85.2 72.2 95.2 91.3 84.2 81.7 73.3 3.4 75.3 402 Residence Urban 97.6 96.4 90.7 84.9 81.9 98.3 93.6 81.0 89.6 71.0 1.5 69.3 214 Rural 96.3 96.8 93.8 87.3 72.7 96.2 92.9 84.8 85.0 75.8 2.7 76.9 1,412 Province City of Kigali 97.4 96.2 89.4 80.6 83.1 98.3 91.5 76.4 85.4 61.7 1.7 69.0 103 South 98.3 98.1 96.9 92.5 73.3 97.7 94.7 88.8 94.1 84.3 1.1 76.4 393 West 96.7 98.4 92.6 84.4 71.4 97.1 92.0 82.6 82.5 72.0 1.6 76.0 440 North 99.0 98.8 95.7 90.3 79.1 98.8 97.5 86.6 92.1 81.2 0.2 76.6 340 East 91.4 91.4 89.5 82.6 69.9 91.4 88.5 81.4 73.9 67.0 7.9 76.6 350 Education No education 94.2 94.7 91.0 83.7 69.7 94.1 90.5 80.4 82.6 71.8 4.7 71.0 423 Primary 97.2 97.5 94.8 88.3 75.0 97.2 93.6 86.0 86.0 76.2 1.8 78.8 1,067 Secondary or higher 98.2 97.2 90.3 86.4 78.2 98.2 96.9 82.8 92.0 77.7 1.8 67.9 135 Wealth quintile Lowest 95.8 96.0 92.9 85.7 68.6 96.0 91.8 82.3 84.9 74.3 3.7 71.4 335 Second 96.3 95.8 91.5 84.8 71.9 94.3 90.2 82.4 83.9 73.8 3.0 76.6 345 Middle 95.1 96.4 94.4 88.1 74.9 96.4 93.6 85.5 84.1 75.0 3.0 78.3 339 Fourth 97.0 97.7 95.3 90.1 71.3 97.7 95.1 87.6 88.1 78.7 1.4 77.7 329 Highest 98.5 98.1 93.0 86.0 84.6 98.3 94.9 83.7 87.6 74.0 1.3 75.3 277 Total 96.5 96.8 93.4 87.0 73.9 96.5 93.0 84.3 85.6 75.2 2.5 75.9 1,626 1 Polio 0 is the polio vaccination given at birth. 2 BCG, measles and three doses each of DPT and polio vaccine (excluding polio vaccine given at birth) 8.5 CHILDHOOD ILLNESSES 8.5.1 Acute Respiratory Infection (ARI) and Fever Acute Respiratory Infections (ARI), particularly pneumonia, constitute one of the main causes of child deaths in developing countries. To assess the prevalence of these infections, mothers were asked if their children under five years had been ill with a cough during the two weeks preceding the survey. If the answer was yes, they were asked if the cough had been accompanied by short, rapid breathing. Fever is the primary symptom of many illnesses including malaria and measles, which cause numerous deaths in developing countries. For this reason, mothers were asked whether their children had suffered from a fever during the two weeks preceding the interview. In addition, for children who had presented 110 | Maternal and Child Health symptoms of ARI and fever, information was gathered concerning whether or not treatment or advice had been sought. The results are presented in Table 8.11. Among children under the age of five, 17 percent had been ill with a cough accompanied by short, rapid breathing in the two weeks preceding the survey. These respiratory infections were the most frequent among children age 6-11 months (28 percent) and 12-23 months (21 percent) (see Figure 8.4). There is no notable difference in ARI prevalence between boys and girls (17 percent for both). The prevalence of AIR is similar in rural and urban areas (17 percent and 18 percent, respectively). Table 8.11 Prevalence and treatment of symptoms of ARI and fever Percentage of children under five years who had a cough accompanied by short, rapid breathing (symptoms of ARI), and percentage of children who had fever in the two weeks preceding the survey, and percentage of children with symptoms of ARI and/or fever for whom treatment was sought from a health facility or provider, by background characteristics, Rwanda 2005 Background characteristic Percentage of children with symptoms of ARI Percentage of children with fever Number of children Among children with symptoms of ARI and/or fever, percentage for whom treatment was sought from a health facility/provider1 Number of children Age in months <6 15.5 19.5 891 24.4 228 6-11 27.5 38.9 830 35.8 374 12-23 21.3 36.9 1,626 30.9 684 24-35 15.8 24.0 1,732 26.8 500 36-47 14.2 20.8 1,373 18.6 362 48-59 11.2 18.4 1,346 18.8 302 Sex Male 17.2 26.5 3,959 27.8 1,258 Female 16.9 26.0 3,839 26.0 1,192 Residence Urban 18.4 25.3 1,144 40.6 362 Rural 16.9 26.4 6,653 24.5 2,088 Province City of Kigali 17.4 25.2 599 43.6 188 South 17.7 29.5 1,909 28.1 652 West 15.5 23.6 2,075 20.0 593 North 14.9 22.9 1,571 32.1 437 East 20.4 29.3 1,644 23.2 580 Education No education 18.6 28.3 2,172 23.7 719 Primary 16.7 26.0 4,938 26.5 1,549 Secondary or higher 14.7 21.0 687 43.0 183 Wealth quintile Lowest 18.1 27.8 1,612 21.7 531 Second 16.3 24.8 1,605 24.5 481 Middle 17.0 25.8 1,620 23.9 505 Fourth 16.7 27.5 1,525 23.8 492 Highest 17.4 25.2 1,436 42.7 441 Total 17.1 26.2 7,797 26.9 2,450 1 Excludes pharmacy, shop, and traditional practitioner Maternal and Child Health | 111 Results according to province show a higher prevalence of ARI in the East (20 percent) and South (18 percent) provinces and in the City of Kigali (17 percent) than elsewhere. Results according to mother’s level of education vary somewhat: from a high of 19 percent for children of mothers with no education, to 17 percent for children of mothers with primary education, to 15 percent for children of mothers with secondary or higher education. ARI prevalence does not vary much by wealth quintile. In the two weeks preceding the survey, just over one-quarter of the children had had a fever (26 percent). As with ARI, age seems to be the most important factor affecting fever prevalence: children age 6-11 months (39 percent) and 12-23 months (37 percent) were the most likely to have had a fever (Figure 8.4). Fever prevalence does not vary much by gender of child (27 percent for boys; 26 percent for girls) or residence (25 percent for urban; 26 percent for rural), and there are only slight variations between provinces, prevalence being highest in the South (30 percent) and East (29 percent) provinces and in the City of Kigali (27 percent). Similarly, children whose mothers have no education (28 percent) were more likely to have suffered from fever (28 percent, compared with 21 percent for those whose mothers have a secondary education or higher). Household wealth does not significantly affect the prevalence of fever in children under the age of five. The table also shows the proportion of children for whom treatment was sought. Treatment or advice was sought from a health facility or provider for only 27 percent of children with the symptoms of acute respiratory infection and/or fever. Treatment was sought most often for children age 6-11 months (35 percent) and 12-23 months (31 percent), who, as seen above, have the highest prevalence of fever and ARI. Whether or not treatment is sought from a health facility for ARI or fever is influenced by residence, mother’s level of education, and wealth quintile. In urban areas, treatment was sought for 41 percent of children, compared with only one in four children in rural areas (25 percent). Similarly, treatment or advice was sought for 43 percent of children whose mothers have a secondary education or higher, compared with only 27 percent of children whose mothers have a primary education, and 24 percent of those whose mothers have no education. Finally, treatment was sought for 43 percent of children in the richest households, while in the poorest households, this proportion was only 22 percent. The data for treatment seeking show no significant variation by gender of child. The results according to province show that seeking treatment is not necessarily linked to prevalence of ARI or fever. Apart from the City of Kigali, which has a high proportion of children for whom treatment or advice was sought (44 percent), treatment was most often sought in the North province (32 percent), which has the lowest prevalence of ARI and/or fever. However, the proportion of children for whom advice or treatment was sought was only 23 percent in the East province, which has relatively high levels of ARI and fever. 8.5.2 Diarrhea Prevalence of diarrhea Diarrheal diseases constitute one of the main causes of death among young children in developing countries because of associated dehydration and malnutrition. To combat the effects of dehydration, WHO promotes the use of oral rehydration therapy (ORT), which includes a prepared solution of oral rehydration salts (ORS), from packets; a solution prepared at home using water, sugar, and salt (recommended home fluids, or RHF); or simply increased intake of fluids. 112 | Maternal and Child Health To assess the prevalence of diarrheal diseases in children under the age of five, mothers were asked whether their children had suffered from diarrhea during the two weeks preceding the survey (Table 8.12). Information was also gathered on the percentage of mothers who had heard of ORS packets (Table 8.13), the percentage of children for whom treatment or advice was sought, and the type of treatment used to treat the diarrhea. Regarding treatment, mothers were asked whether they had used ORS packets and/or RHF, or other treatments during the diarrheal episodes (Table 8.14). Table 8.12 shows that, according to mothers’ reports, 14 percent of children had suffered from diarrhea in the two weeks preceding the survey. The prevalence of diarrhea is especially high among children age 6-23 months (24 percent) (Figure 8.4). These high-prevalence ages are also the ages at which children begin to be weaned and consume foods other than breast milk. They also correspond to the ages at which children begin to explore their environment, resulting in greater exposure to pathogens. Diarrhea prevalence seems to have little relation to a child’s gender or residence: 15 percent of male children suffered from diarrhea, compared with 13 percent of female children, and 13 percent of children in rural areas were affected by diarrhea, compared with 14 percent in urban areas. By province, the City of Kigali has the lowest diarrhea prevalence (11 percent); varia- tions are small among the other provinces, the proportion of children with diarrhea ranging between 14 percent and 15 percent. However, mother’s level of education seems to play an important role, with prevalences being higher among children whose mothers have no education or have a primary education than among those whose mothers have a secondary education or higher (15 per- cent, compared with 9 percent). Moreover, children who drink piped (tap) water have the lowest prevalence of diarrhea (12 percent). Although unclean water is an increased risk factor for contracting diarrheal diseases, surprisingly, the prevalence of diarrhea among children in households that drink water from open wells (14 per- cent) or surface water (from lakes or marsh creeks— 15 percent) does not differ substantially from the preva- lence of diarrhea among children who consume piped water (12 percent). Moreover, children who drink pro- tected well water have the highest prevalence of diarrhea (16 percent). There also does not appear to be a strong link between diarrhea prevalence and household wealth. In households in the poorest quintile, 16 percent of chil- dren had diarrhea in the two weeks preceding the survey, compared with 11 percent among children in the richest quintile, but diarrhea prevalence in the fourth quintile is identical to that of the poorest quintile (16 percent). Table 8.12 Prevalence of diarrhea Percentage of children under five years with diarrhea in the two weeks preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Diarrhea in the two weeks preceding the survey Number of children Age in months <6 9.0 891 6-11 24.2 830 12-23 24.3 1,626 24-35 12.8 1,732 36-47 8.7 1,373 48-59 6.4 1,346 Sex Male 15.2 3,959 Female 13.1 3,839 Residence Urban 12.7 1,144 Rural 14.4 6,653 Province City of Kigali 11.2 599 South 14.5 1,909 West 13.7 2,075 North 14.5 1,571 East 15.1 1,644 Mother’s education No education 15.1 2,172 Primary 14.5 4,938 Secondary or higher 8.5 687 Source of drinking water Piped 12.1 2,216 Protected well 15.8 484 Open well 14.0 1,046 Surface 15.3 3,975 Other 5.2 76 Wealth quintile Lowest 16.0 1,612 Second 14.2 1,605 Middle 13.6 1,620 Fourth 16.0 1,525 Highest 10.8 1,436 Total 14.1 7,797 Maternal and Child Health | 113 Knowledge of ORS packets Table 8.13 shows that 87 percent of women with births in the five years preceding the survey reported knowing about oral rehydration salt (ORS) packets. This proportion is slightly higher than that of the RDHS-II survey (86 percent). The level of knowledge of ORS packets increases with the age of the mother, ranging from a low of 63 percent for mothers age 15 to 19, to a high of 91 percent for mothers age 35 to 49. There is not much variation between urban and rural areas (90 percent and 86 percent, respectively). Neither is there any significant difference between the provinces, knowledge of ORS ranging between 83 percent and 89 percent. However, knowledge of ORS packets increases with mother’s level of education and household wealth. The proportion of women who had heard of ORS packets increases from 85 percent among women with no education to 95 percent among the most educated women; similarly, it rises from 84 percent among the poorest women to 91 percent among women in the richest quintile. Figure 8.4 Prevalence of ARI, Fever, and Diarrhea, by Age 16 28 21 16 14 11 20 39 37 24 21 18 9 24 24 13 9 6 < 6 months 6-11 months 12-23 months 24-35 months 36-47 months 48-59 months 0 10 20 30 40 50 Percent ARI Fever Diarrhea RDHS 2005 114 | Maternal and Child Health Table 8.13 Knowledge of ORS packets Percentage of mothers with births in the five years preceding the survey who know about ORS packets for treatment of diarrhea, by background characteristics, Rwanda 2005 Background characteristic Percentage of mothers who know about ORS packets Number of mothers Age 15-19 63.2 84 20-24 79.0 1,060 25-29 85.9 1,359 30-34 90.3 1,175 35-49 91.1 1,747 Residence Urban 90.0 774 Rural 86.3 4,651 Province City of Kigali 88.3 427 South 87.8 1,357 West 82.9 1,395 North 89.3 1,052 East 87.7 1,194 Education No education 84.5 1,552 Primary 86.9 3,404 Secondary or higher 94.5 469 Wealth quintile Lowest 84.4 1,163 Second 84.3 1,124 Middle 88.9 1,097 Fourth 86.8 1,069 Highest 90.5 972 Total 86.9 5,425 ORS = Oral rehydration salts Treatment of diarrhea Table 8.14 shows that advice or treatment was sought for only 14 percent of children with diarrhea. Treatment was most often sought for children age 12-35 months (17 percent). Only 12 percent of children age 6-11 months—who have the highest prevalence of diarrhea—received treatment. Boys (16 percent) were more likely to be taken to health facilities for treatment than girls (12 percent). There is little difference in treatment seeking for diarrhea between urban (16 percent) and rural (14 percent) areas. However, there are major differences with respect to provinces; the proportion of children taken to a health facility ranges from a high of 23 percent in the North province to a low of 10 percent in the East province. Children whose mothers have a secondary education or higher (24 percent, compared with 16 percent for those whose mothers have no education) and those living in the richest households (18 percent, compared with 13 percent in the poorest quintile) received treatment more frequently than other children. Maternal and Child Health | 115 Table 8.14 Diarrhea treatment Percentage of children under five years who had diarrhea in the two weeks preceding the survey taken for treatment to a health provider, percentage who received oral rehydration therapy (ORT), and percentage given other treatments, according to background characteristics, Rwanda 2005 Oral rehydration therapy (ORT) Other treatments Background characteristic Percentage taken to a health provider1 ORS packets RHF Either ORS or RHF Increased fluids ORS, RHF, or increased fluids Pill/ syrup Injection Home remedy/ other No treat- ment Number of children Age in months <6 7.1 2.5 7.4 9.9 7.4 16.0 9.7 0.0 29.0 52.1 80 6-11 11.7 12.8 9.3 19.3 17.0 30.9 13.5 1.3 36.5 36.1 201 12-23 16.6 12.0 9.1 19.5 17.9 31.9 20.2 1.3 31.3 31.8 395 24-35 15.5 16.0 6.7 21.1 22.5 36.2 22.4 0.5 34.4 25.0 222 36-47 14.7 9.0 8.5 16.6 25.3 36.1 24.8 0.3 28.7 30.0 119 48-59 10.5 7.3 9.6 16.9 16.6 32.1 10.3 0.0 29.2 42.4 86 Sex Male 16.1 11.5 7.7 17.7 18.4 31.0 19.9 1.0 31.9 34.5 600 Female 11.8 11.7 9.5 19.6 18.9 33.1 16.6 0.6 32.6 31.9 503 Residence Urban 16.2 14.6 13.0 26.8 25.4 39.5 25.0 0.9 26.8 27.9 145 Rural 13.8 11.1 7.8 17.3 17.6 30.8 17.4 0.8 33.1 34.1 958 Province City of Kigali 18.6 20.3 11.6 30.4 26.6 40.9 23.3 1.3 27.1 28.5 67 South 10.9 6.6 11.8 17.7 24.7 35.9 14.4 0.8 28.9 35.0 277 West 13.2 17.0 7.9 22.5 14.6 32.8 17.6 0.9 27.9 35.9 284 North 22.5 14.7 4.9 18.8 14.2 27.6 22.1 0.9 26.8 36.8 227 East 9.9 5.6 8.0 11.7 18.3 28.0 19.0 0.7 47.4 26.5 248 Mother’s education No education 16.0 12.8 6.8 18.6 16.1 29.9 19.8 0.9 30.9 33.3 328 Primary 12.5 10.4 9.1 18.0 18.3 31.5 16.5 0.7 32.9 34.7 717 Secondary or higher 24.4 18.4 11.0 26.1 36.9 48.7 33.9 1.5 32.6 16.4 58 Wealth quintile Lowest 13.3 10.6 7.0 15.2 15.8 27.1 16.4 0.3 34.0 36.0 257 Second 11.5 6.6 8.1 13.0 18.7 27.2 16.1 1.2 36.2 31.9 227 Middle 13.5 10.2 6.5 15.2 18.8 29.3 15.7 0.0 29.4 39.7 220 Fourth 15.4 14.3 10.7 24.5 16.0 37.9 16.6 1.2 33.3 31.7 243 Highest 18.3 18.1 10.9 27.8 26.9 41.3 31.6 1.7 26.0 24.4 155 Total 14.1 11.6 8.5 18.6 18.6 31.9 18.4 0.8 32.3 33.3 1,103 Note: ORT includes solution prepared from oral rehydration salt (ORS) packets, recommended home fluids (RHF), and increased fluids. 1 Excludes pharmacy, shop and traditional practitioner During diarrheal episodes, only 12 percent of children received ORS, 9 percent received RHF, and 19 percent received either ORS or RHF. In addition, 19 percent of children received increased fluids. Overall, 32 percent of children were treated with some form of oral rehydration. In addition, 18 percent of children received pills or syrup, and a very small proportion of children (1 percent) received treatment by injection. The proportion of children treated with traditional remedies is high (33 percent), and nearly identical to that of children who received ORT (32 percent). One-third of children (33 percent) received no treatment at all. This proportion is particularly high among children younger than 6 months (52 percent). 116 | Maternal and Child Health Feeding practices during diarrhea During diarrheal episodes, it is recommended that children consume more food and liquids than usual. Table 8.15 shows that 42 percent of children who had diarrhea were offered the same amount of liquids as usual while they were ill; 22 percent were offered less than usual; and 8 percent were offered much less than usual. Only 19 percent of children were offered more liquids than usual. Nine percent of children were offered no liquids at all. Regarding food intake, 36 percent of children with diar- rhea were offered the same amount of food as usual, 29 percent were offered less than usual, and 11 percent were offered much less than usual. Only 6 percent of children were offered more food than usual. Finally, 2 percent were not given any food. 8.6 PROBLEMS IN ACCESSING HEALTH CARE Access to health care is a key priority for improving a country’s overall health status. Therefore, the survey sought to obtain information on the problems women perceive as barriers to accessing health care. The results are presented in Table 8.16. First, 71 percent of women reported lack of money for treatment as the primary barrier. The extent of this problem increases with age, with the oldest women encountering this problem more frequently than the youngest women (68 percent at age 15-19, compared with 76 percent at age 40- 49). Divorced, separated, and widowed women (83 percent) reported having this problem more frequently than married women (70 percent) and never-married women (68 percent). Lack of money was also more of a barrier for women in rural areas (73 percent) than for women in urban areas (60 percent). With respect to provinces, women in the West province were proportionately more likely to mention this problem (82 percent). Similarly, women with no education mentioned this problem more often (82 percent) than women with a secondary education or higher (42 percent), and women in the poorest wealth quintile were more affected by lack of money (83 percent) than women in the richest quintile (52 percent). Forty percent of women mentioned distance to the health facility as a problem, and 39 percent of women mentioned having to take public transport. These problems were much more frequent in rural areas than in urban areas, and even more frequent among women with little or no education and women in poorer households. This confirms the fact that women with no education who live in rural areas are in the parts of the country that are the least equipped to provide adequate health care. Overall, more than eight in ten women (81 percent) reported having at least one of the problems mentioned. Divorced, separated, and widowed women (90 percent), women in rural areas (83 percent), women with no education (88 percent), women in the poorest households (89 percent), and women performing unpaid labor (84 percent) were the most likely to encounter barriers to accessing health care. Table 8.15 Feeding practices during diarrhea Percent distribution of children under five years who had diarrhea in the two weeks preceding the survey by amount of liquids and food offered compared with normal practice, Rwanda 2005 Liquid/food offered Percentage Amount of liquids offered Same as usual 41.9 More 18.6 Somewhat less 21.7 Much less 7.8 None 8.8 Total1 100.0 Amount of food offered Same as usual 36.2 More 5.7 Somewhat less 28.8 Much less 11.2 None 1.6 Never gave food 15.6 Total1 100.0 Number of children 1,103 1 Includes those with missing information Maternal and Child Health | 117 Table 8.16 Problems in accessing health care Percentage of women who reported they have big problems in accessing health care for themselves when they are sick, by type of problem and background characteristics, Rwanda 2005 Problems in accessing health care Background characteristic Knowing where to go for treatment Getting permission to go for treatment Getting money for treatment Distance to health facility Having to take transport Not wanting to go alone Concern there may not be a female provider Any of the specified problems Number of women Age 15-19 7.6 6.2 68.1 40.0 38.4 17.3 18.5 79.9 2,585 20-29 3.8 2.9 69.6 38.8 37.6 15.4 8.4 79.8 4,092 30-39 3.5 1.4 71.2 39.5 39.4 17.4 4.9 80.1 2,600 40-49 4.2 1.4 76.4 42.9 42.1 19.2 4.9 84.7 2,045 Number of living children 0 6.4 5.2 67.5 40.0 38.5 16.9 15.5 79.3 4,363 1-2 4.0 2.1 71.9 39.4 39.8 17.5 6.2 81.9 2,722 3-4 3.4 1.6 72.6 39.2 38.0 17.5 4.8 80.7 2,266 5 or more 3.2 1.2 74.7 41.5 40.4 16.2 4.9 82.7 1,970 Marital status Never married 6.3 5.3 68.2 39.1 37.5 16.8 15.3 79.4 4,263 In union 3.2 1.8 69.5 39.1 38.9 15.6 5.7 79.3 5,510 Divorced, separated, widowed 5.3 1.4 82.9 45.3 43.8 22.8 5.3 89.6 1,548 Residence Urban 5.4 4.5 59.6 28.5 30.0 16.4 10.0 70.5 1,921 Rural 4.5 2.7 73.1 42.3 40.9 17.1 9.1 82.9 9,400 Province City of Kigali 5.8 3.5 62.0 35.1 35.3 17.2 9.2 72.5 1,127 South 3.5 2.5 70.6 44.3 43.9 17.1 7.4 83.6 2,958 West 5.4 4.4 81.8 44.1 43.6 19.1 12.2 89.4 2,824 North 3.4 1.4 59.5 22.6 23.1 11.3 5.7 66.9 2,063 East 5.8 3.3 72.2 47.1 43.0 19.3 11.3 83.1 2,348 Education No education 5.0 2.7 82.1 43.5 42.6 18.7 8.6 88.2 2,646 Primary 4.8 3.2 71.1 40.6 39.5 16.9 10.0 81.9 7,591 Secondary or higher 2.8 2.6 41.7 26.5 26.6 13.2 5.4 54.8 1,084 Employment Not employed 5.4 4.6 69.4 36.9 34.9 14.9 10.8 78.3 3,055 Working for cash 3.9 3.4 64.9 38.6 37.5 16.4 8.0 77.2 2,522 Working, not for cash 4.6 2.0 74.2 42.2 41.9 18.4 9.0 83.7 5,738 Wealth quintile Lowest 5.8 2.8 83.1 46.4 45.3 20.7 10.0 89.0 2,421 Second 4.4 2.5 74.1 44.3 43.3 17.5 10.2 85.0 2,325 Middle 4.2 2.7 74.7 40.7 39.0 15.8 8.4 83.0 2,099 Fourth 4.4 3.2 70.4 40.3 39.3 16.4 8.3 82.6 2,133 Highest 4.5 3.9 51.9 28.0 28.0 14.4 9.2 64.5 2,342 Total 4.7 3.0 70.8 40.0 39.0 17.0 9.3 80.8 11,321 8.7 TOBACCO CONSUMPTION The consumption of tobacco has a negative impact on children’s health, because it affects the health of those who consume it and the health of those around people who consume it. For this reason, the RDHS-III asked questions to determine the level of tobacco consumption among the women surveyed. 118 | Maternal and Child Health Table 8.17 shows that the vast majority of women in Rwanda do not smoke tobacco (95 percent). The proportion of women who smoke cigarettes is insignificant, although 3 percent of women reported smoking a pipe and 2 percent consume tobacco in other forms. The oldest women (7 percent), women in rural areas (3 percent), and women with no education (6 percent) smoke pipes or consume tobacco more frequently than other women. The proportion of pregnant or breastfeeding women who smoke is very low. Table 8.17 Use of smoking tobacco Percentage of women who smoke cigarettes or tobacco, according to background characteristics and maternity status, Rwanda 2005 Background characteristic Cigarettes Pipe Other tobacco Does not use tobacco Number of women Age 15-19 0.0 0.0 0.1 99.7 2,585 20-34 0.2 1.3 1.4 97.1 5,557 35-49 0.7 6.7 3.9 88.7 3,179 Residence Urban 0.4 0.4 0.9 98.2 1,921 Rural 0.3 2.9 2.0 94.7 9,400 Education No education 0.3 6.1 3.9 89.6 2,646 Primary 0.2 1.6 1.3 96.8 7,591 Secondary or higher 0.6 0.1 0.2 99.1 1,084 Maternity status Pregnant 0.0 2.0 2.5 95.1 901 Breastfeeding (not pregnant) 0.1 2.9 2.3 94.5 3,867 Neither 0.4 2.3 1.4 95.8 6,553 Total 0.3 2.5 1.8 95.3 11,321 Malaria | 119 MALARIA 9 9.1 INTRODUCTION Malaria is a potentially fatal parasitic disease found in intertropical regions. It is caused by protozoa of the genus Plasmodium transmitted to humans through the bite of the female Anopheles mosquito. Malaria is one of the world’s major public health concerns, particularly in sub-Saharan Africa. Each year it afflicts 300 to 500 million people worldwide, killing between one and two million. More than 80 percent of these cases, and over 90 percent of the deaths, occur in Africa. Malaria also has an enormous negative socioeconomic impact in countries with endemic wetlands (losses estimated at USD 3.6 billion and 1.3 percentage points in GDP growth annually), and is a major contributor to school absenteeism. It aggravates poverty, contributes to inequality, and impedes development. Malaria affects males and females of all ages. However, its most serious consequences are felt by pregnant women and children under the age of five. In pregnant women, malaria can lead to severe anemia, loss of a pregnancy, and a greater likelihood of low birth weight babies. In young children, it increases the risk of anemia, delays physical and mental growth and, all too often, results in death. Combating malaria in Africa In October 1998, WHO, UNICEF, UNDP, and the World Bank launched the worldwide “Roll Back Malaria” (RBM)1 initiative. One of its aims is to promote social and economic growth in Africa by combating malaria. Its goals are to reduce mortality directly related to malaria by 50 percent by 2010, 30 percent by 2015, and 20 percent by 2025. If the program is successful, by 2030 malaria should cease to be a major cause of sickness, death , and socioeconomic loss in Africa. The currently recommended strategies for combating malaria are: (1) Prompt access to effective treatment; (2) Increased use of insecticide-treated mosquito nets (ITNs); (3) Improved prevention and treatment of malaria in pregnant women, and (4) Early detection and response to malaria epidemics. In April 2000, a summit of African Heads of State and Government held in Abuja, Nigeria, renewed political will in this struggle and established the following goals for 2005: • Provide access to the most effective preventive treatment measures to at least 60 percent of children under the age of 5 and 60 percent of pregnant women. • Provide appropriate treatment within 24 hours of the onset of symptoms to at least 60 percent of children suffering from malaria. The malaria situation in Rwanda The Rwandan plains are considered malaria-endemic, while the high plateaus are subject to malaria epidemics (Ivorra, 1967; Vermylen, 1967). 1 In French, Faire Reculer le Paludisme. 120 | Malaria Malaria is the main cause of morbidity and mortality in Rwanda. In 2005, approximately 30 percent of all cases were among children under the age of five. Hospitals reported more than 80,000 cases of severe malaria, approximately 900 of them resulting in death. Thirty-five percent of those who died were children under the age of five. These data reflect reported cases, which represent only a fraction of the overall number of cases. Malaria stratification mapping and predisposing factors The country is divided into four natural “malarial ecozones” based on elevation, climate, plasmodic index (Plasmodium infestation),2 and disease vectors (Meyus et al., 1962). The first stratum extends from Lake Kivu to the Congo-Nile Divide at elevations ranging between 1,460 and 1,800 meters. The plasmodic indices among children here are generally between 5 percent and 30 percent. The second stratum consists of a north-south band 160 km long and 20 to 50 km wide, located east of the first stratum between the elevations of 1,800 and 3,000 meters. The plasmodic index here is under 2 percent. The third stratum is situated on the central plateau between the elevations of 1,000 and 2,000 meters. The plasmodic indices vary widely here, ranging from 10 to 50 percent. This area is at risk of malaria epidemics, many of which have been recorded at elevations ranging between 1,675 and 1,860 meters. Malaria-endemic pockets in the valleys provide the starting points for these epidemics. The fourth stratum covers the lower eastern shelf of the central plateau at elevations ranging between 1,000 and 1,500 meters, where malaria is endemic and appears to be stable. Within these four large strata, micro stratification is also possible because of topographical variations and agricultural activity in the valleys (Rusanganwa, 1999). Malaria is now present in sectors and at altitudes where the disease was not previously a major public health concern. Residents in these locations are poorly prepared to combat malaria and are therefore highly predisposed to malaria epidemics. Combating malaria in Rwanda In 1999, the Government established the National Malaria Control Program, or PNILP (Programme National Intégré de Lutte contre le Malaria) with strategies and activities focused on: managing cases of malarial illness, malaria prevention, epidemiological surveillance, IEC (health education) and community participation, and operational research. a. Management of malarial illness focuses on early and effective diagnosis, early and appropriate treatment, training, staff supervision, and monitoring of drug efficacy. Because of increasing disease resistance resulting in high rates of therapeutic failure, chloroquine was discontinued as a treatment at the beginning of 2002. Now an AQ+SP (amodiaquine, sulfadoxine-pyrimethamine) combination regimen is used to treat uncomplicated cases of malaria (PNILP, 2001). This is only a transitional measure because SP is used for Intermittent Preventive Treatment (IPT) (WHO, 2003). In addition, in 2005, Rwanda began using artemisinin-based combination therapies, which, although costly, are the most effective and long-lasting solutions (Attaran et al., 2004; Yamey, 2003). Quinine is used for 2 Percentage of subjects examined having malaria parasites in the blood. Malaria | 121 the most serious cases with an initial loading dose. Community-based malaria management is currently operational under the RBM 2004-2010 Strategic Plan. b. Malaria prevention. Malaria is prevented through the use of ITNs, indoor residual spraying of homes with insecticides, and the destruction of mosquito breeding grounds. Mosquito nets are distributed through two channels: health care facilities and PSI/Rwanda, which involves the private sector. ITNs have been distributed to pregnant women during antenatal care visits at a cost of FRw 200, subsidized by UNICEF. The promotion of Long-Lasting Insecticide-treated Mosquito Nets (LLITNs) is also operational, along with IPT (Intermittent Preventive Treatment) for pregnant women. c. Epidemiological surveillance consists of monitoring areas at risk for malaria epidemics by collecting and analyzing data from health care facilities and representative sentinel sites.3 The 20 areas at risk for epidemics are located in the former provinces of Byumba (1 zone), Cyangugu (4 zones), Gikongoro (2 zones), Gisenyi (3 zones), Gitarama (1 zone), Kibuye (4 zones), Kigali Ngali (1 zone), and Ruhengeri (4 zones). This means that the former provinces of Butare, Kibungo, and Umutara, which are malaria-endemic, contain no zones considered at risk of malaria epidemics. d. IEC and community participation. The aim of this activity is to inform the population about combating malaria by targeting political and governmental authorities and the most vulnerable groups. Messages in Kinyarwanda are distributed through various media, including radio, television, and local newspapers, and are reinforced on Africa Malaria Day. However, IEC is not yet fully deployed and needs to be scaled up, particularly in rural communities. e. Operational research. Research is the weak link in the Rwanda health system. In order to increase control of malaria, more study and interventions are needed in biomedical research (entomological, parasitological, clinical, reevaluation of Plasmodia resistance to antimalarial drugs and Anopheles to resistance pyrethroids) and socioanthropology (disease presentation, mosquito net acceptability, etc.). The third Demographic and Health Survey gathered data on the use of malaria prevention methods. The results are presented in this chapter. 9.2 MALARIA PREVENTION The survey asked each household whether it owned a mosquito net, how many mosquito nets it possessed, and how long it had owned each mosquito net. Respondents were then asked if the mosquito net had been treated with an insecticide and how long it had been since it was last treated. Respondents were also asked whether the mosquito net had been washed since it was last treated. For this section, mosquito nets were grouped into three categories: any type of mosquito net; ever-treated mosquito nets, i.e., factory-treated mosquito nets that do not require further treatment or nets that were not pretreated but were soaked in insecticide at some time; and insecticide-treated nets (ITNs), i.e., factory-treated nets that do not require further treatment, pretreated nets obtained within the previous 12 months, or nets that were soaked in insecticide within the past 12 months. 3 An epidemic threshold has been defined corresponding to twice the monthly average over the past three years. 122 | Malaria 9.2.1 Household Possession of Mosquito Nets Table 9.1 shows the responses to questions about net ownership. Table 9.1 and Figure 9.1 show that 18 percent of Rwandan households possess at least one mosquito net. The percentage varies by province, reaching a high of 40 percent in the City of Kigali, and a low of 10 percent in the North province. Net possession varies by residence from 40 percent in urban areas to 14 percent in rural areas. Households in the richest wealth quintile were the most likely to own at least one mosquito net (45 percent, compared with 6 percent in the poorest quintile). Table 9.1 Household possession of mosquito nets Percentage of households with at least one and more than one mosquito net (treated or untreated), ever-treated mosquito net, and insecticide- treated net1 (ITN), and the average number of nets per household, by background characteristics, Rwanda 2005 Any type mosquito net Ever-treated mosquito net1 Insecticide-treated mosquito nets (ITNs)2 Background characteristic Percentage with at least one Percentage with more than one Average number per household Percentage with at least one Percentage with more than one Average number per household Percentage with at least one Percentage with more than one Average number per household Number of households Residence Urban 40.3 18.5 0.7 39.7 18.1 0.7 31.6 13.9 0.5 1,510 Rural 14.4 3.4 0.2 14.3 3.4 0.2 11.8 2.5 0.2 8,762 Province Kigali city 40.2 21.2 0.7 39.7 20.7 0.7 32.2 15.9 0.6 864 South 19.8 4.5 0.3 19.7 4.5 0.3 16.0 3.2 0.2 2,722 West 16.7 4.9 0.2 16.5 4.9 0.2 14.0 3.8 0.2 2,522 North 10.0 2.6 0.1 9.9 2.5 0.1 7.9 1.9 0.1 1,946 East 16.7 4.4 0.2 16.6 4.3 0.2 13.0 3.4 0.2 2,218 Wealth quintile Lowest 6.1 0.7 0.1 6.1 0.7 0.1 4.8 0.5 0.1 2,217 Second 13.7 2.0 0.2 13.6 2.0 0.2 11.1 1.1 0.1 1,907 Middle 11.7 1.8 0.1 11.6 1.8 0.1 8.8 1.1 0.1 2,119 Fourth 17.8 3.4 0.2 17.6 3.3 0.2 14.5 2.6 0.2 2,105 Highest 44.5 21.6 0.8 44.0 21.0 0.8 36.5 16.8 0.6 1,925 Total 18.2 5.6 0.3 18.1 5.5 0.3 14.7 4.2 0.2 10,272 1 An ever-treated net is (1) a pretreated net or (2) a nonpretreated net that which has subsequently been soaked with insecticide at some time. 2 An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatment or (2) a pretreated net obtained within the past 12 months or (3) a net that has been soaked with insecticide within the past 12 months. Table 9.1 shows the proportion of households that reported owning at least one ever-treated mosquito net, i.e. a mosquito net that had been soaked in insecticide at some time, and the proportion of households that possessed at least one insecticide-treated net (ITN), i.e., a factory-treated net that does not require further treatment, a pretreated net obtained within the past 12 months, or a mosquito net that was soaked in insecticide within the past 12 months. Only 18 percent of all households reported owning an ever-treated mosquito net. However, 40 percent of households in Kigali reported owning one. Forty-four percent of the richest households reported owning a pretreated mosquito net, more than twice as high as the national average (18 percent). The percentage of households owning at least one ever-treated mosquito net is higher in urban areas than in rural areas (40 percent compared with 14 percent). The percentage of all households owning an ITN at the time of the survey is lower (15 percent), although it reaches as high as 32 percent in Kigali and 37 percent in the richest quintile, compared with 12 percent in rural areas and 5 percent in the poorest quintile. Malaria | 123 9.2.2 Use of Mosquito Nets by Children Households that reported owning at least one mosquito net were asked who had slept under the net the night before the survey. Results are shown for all women age 15 to 49, and for two particularly vulnerable groups: pregnant women and children under the age of five (Tables 9.2 and 9.3). Table 9.2 and Figure 9.2 show the proportion of children under age five who slept under a mosquito net the night before the survey (16 percent) The results do not show major differences by age group: at most, older children can be said to be somewhat less likely to have slept under a mosquito net than younger children (over 15 percent of children age 24 to 35 months, compared with 14 percent of children age 36 to 47 months, and 10 percent of children age 48 to 59 months). There are no differences with respect to gender (16 percent for both sexes). However, the percentage of children who slept under a mosquito net varies widely by residence: in the City of Kigali, 31 percent of children under the age of five had slept under a mosquito net, compared with only 20 percent in the South province, 14 percent in the West and East provinces, and 9 percent in the North province. By residence, the percentages are from 33 percent in urban areas and 13 percent in rural areas. The richest households show the highest proportion of children under the age of five who had slept under a mosquito net the night before the survey (37 percent). Figure 9.1 Household Ownership of Mosquito Nets RDHS 2005 40 20 17 10 17 18 32 16 14 8 13 15 City of Kigali South West North East Rwanda 0 10 20 30 40 50 Percent At least one mosquito net At least one insecticide-treated mosquito net (ITN) 124 | Malaria Table 9.2 Use of mosquito nets by children Percentage of children under five years of age who slept under a mosquito net (treated or untreated), an ever-treated mosquito net1, and an insecticide-treated net2 (ITN) the night before the survey, by background characteristics, Rwanda 2005 Background characteristic Percentage who slept under any net the preceding night Percentage who slept under an ever-treated net1 the preceding night Percentage who slept under an ITN2 the preceding night Number of children Age < 12 19.4 19.3 16.2 1,709 12-23 19.5 19.4 15.9 1,601 24-35 14.7 14.7 11.9 1,665 36-47 13.5 13.4 11.2 1,292 48-59 9.9 9.9 8.5 1,267 Sex Male 15.8 15.8 12.6 3,833 Female 15.8 15.6 13.5 3,701 Residence Urban 32.6 32.0 25.7 1,075 Rural 13.0 13.0 10.9 6,459 Province Kigali city 30.9 29.8 24.0 544 South 20.1 20.1 16.1 1,864 West 14.3 14.3 12.5 2,012 North 8.7 8.7 7.5 1,527 East 14.2 14.2 11.6 1,588 Wealth quintile Lowest 5.9 5.9 4.5 1,575 Second 12.5 12.5 10.5 1,547 Middle 9.9 9.9 8.1 1,577 Fourth 16.4 16.4 13.5 1,478 Highest 37.1 36.6 31.0 1,357 Total 15.8 15.7 13.0 7,534 1 An ever-treated net is (1) a pretreated net or (2) a non-pretreated net that has subsequently been soaked with insecticide at some time. 2 An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatment or (2) a pretreated net obtained within the past 12 months or (3) a net that has been soaked with insecticide within the past 12 months. The proportion of children under the age of five who slept under an ever-treated mosquito net the night before the survey is slightly higher than the proportion who slept under an ITN: 16 percent for ever- treated nets and 13 percent for ITNs. In the City of Kigali, the percentages are 30 percent for ever-treated nets and 24 percent for ITNs; for households in the richest quintile they are 37 percent and 31 percent, respectively. Malaria | 125 9.2.3 Use of Mosquito Nets by Women Table 9.3 shows the percentage of all women and pregnant women age 15 to 49 who slept under a mosquito net the night before the survey (Figure 9.3). A total of 13 percent of women slept under a mosquito net. The proportion of pregnant women who did so is higher (20 percent). In rural areas, 10 percent of all women slept under a mosquito net while in urban areas, 27 percent did so. The percentages for pregnant women are 18 percent for rural areas and 35 percent for urban areas. The results show that women were more likely to have slept under a mosquito net in the City of Kigali, although the use of mosquito nets by pregnant women does not seem widespread there (24 percent). Women with higher levels of education and women in the richest quintile were proportionally more likely to have protected themselves against malaria by sleeping under a mosquito net (22 percent and 30 percent, respectively) than other women. Eleven percent of all women slept under an ITN; the percentage of pregnant women using ITNs is slightly higher, 17 percent, but still low. These results indicate that in Rwanda, mosquito nets are not being used by pregnant women—who are more vulnerable to infection—in significantly greater numbers than by women in general. This is one of the major challenges to be addressed by the PNILP. Figure 9.2 Use of Mosquito Nets by Children Under Age 5, According to Province 31 20 14 9 14 16 24 16 13 8 12 13 City of Kigali South West North East Rwanda 0 10 20 30 40 50 Percent Children who slept under a mosquito net the night preceding the survey Children who slept under an ITN the night preceding the survey RDHS 2005 126 | Malaria Table 9.3 Use of mosquito nets by women Percentage of all women age 15-49 and pregnant women age 15-49 who slept under a mosquito net (treated or untreated), an ever-treated mosquito net1, and an insecticide-treated net2 (ITN) the night before the survey, by background characteristics, Rwanda 2005 Percentage of all women age 15-49 who Percentage of pregnant women age 15-49 who Background characteristic Slept under any net the preceding night Slept under an ever- treated net1 the preceding night Slept under an ITN2 the preceding night Number of women Slept under any net the preceding night Slept under an ever- treated net1 the preceding night Slept under an ITN2 the preceding night Number of women Residence Urban 26.5 26.4 21.6 1,890 34.6 34.6 28.6 118 Rural 10.1 10.0 8.3 9,388 17.8 17.7 15.5 776 Province Kigali city 25.5 25.3 21.1 1,106 24.1 24.1 22.5 76 South 14.5 14.5 11.6 2,959 21.8 21.8 19.1 224 West 11.3 11.3 9.6 2,804 18.8 18.4 16.4 221 North 7.0 7.0 5.9 2,053 14.1 14.1 11.7 161 East 11.7 11.7 9.3 2,356 22.6 22.6 18.4 212 Education No education 9.9 9.9 8.3 2,534 13.7 13.7 9.6 183 Primary 12.8 12.7 10.5 7,861 19.8 19.7 17.5 643 Secondary or higher 21.8 21.7 17.5 884 39.9 39.9 35.3 67 Wealth quintile Lowest 4.0 4.0 3.1 2,414 9.4 9.4 7.6 203 Second 9.3 9.3 7.5 2,329 18.7 18.7 16.8 178 Middle 8.3 8.2 6.4 2,092 16.6 16.2 12.1 170 Fourth 12.8 12.7 10.8 2,128 20.5 20.5 18.7 207 Highest 29.7 29.6 24.9 2,315 41.1 41.1 36.3 136 Total 12.8 12.8 10.5 11,278 20.0 20.0 17.2 894 1 An ever-treated net is (1) a pretreated net or (2) a non-pretreated net that has subsequently been soaked with insecticide at some time. 2 An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatment or (2) a pretreated net obtained within the past 12 months or (3) a net that has been soaked with insecticide within the past 12 months. Figure 9.3 Pregnant Women Who Slept Under a Mosquito Net the Night Preceding the Survey 13 26 15 11 7 12 10 13 22 Rwanda PROVINCE City of Kigali South West North East EDUCATION None Primary Secondary or higher 0 10 20 30 40 Percent RDHS 2005 Malaria | 127 9.2.4 Intermittent Preventive Treatment during Pregnancy Rwanda has adopted a new malaria prevention policy for pregnant women involving a change in therapy from weekly preventive doses of chloroquine to Intermittent Preventive Treatment (IPT) with SP Fansidar, with one restriction: the new treatment is not given to pregnant women in the first trimester. The RDHS-III asked women who had had a live birth in the past five years several questions about whether or not they had taken antimalarial drugs preventively during their last pregnancy, and what type of antimalarial drugs they had taken. According to Table 9.4, 6 percent of pregnant women took antimalarial drugs preventively during their last pregnancy. The percentages are higher in urban areas (10 percent), in the City of Kigali (9 percent), among women with at least a secondary education (10 percent), and among women in the richest quintile (9 percent). Table 9.4 Use of Intermittent Preventive Treatment by women during pregnancy Percentages of women who took any antimalarial drugs for prevention, who took SP/Fansidar, and who received Intermittent Preventive Treatment (IPT), during the pregnancy for their last live birth in the five years preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Percentage of women who took any antimalarial drug to prevent or treat malaria during an ANC visit during the last pregnancy Percentage of women who received Intermittent Preventive Treatment during an ANC visit1 Number of last- born children in the five years preceding the survey Residence Urban 9.8 0.6 774 Rural 5.1 0.2 4,651 Province Kigali city 8.7 0.9 427 South 8.3 0.4 1,357 West 4.0 0.0 1,395 North 2.6 0.2 1,052 East 6.8 0.2 1,194 Education No education 4.5 0.4 1,552 Primary 5.8 0.2 3,404 Secondary or higher 9.8 0.5 469 Wealth quintile Lowest 3.2 0.1 1,163 Second 5.0 0.3 1,124 Middle 5.3 0.2 1,097 Fourth 7.0 0.1 1,069 Highest 9.0 0.5 972 Total 5.8 0.3 5,425 1 Intermittent Preventive Treatment is treatment with at least two doses of SP/Fansidar during antenatal care (ANC) visits. Table 9.5 shows that, of the women surveyed, 31 percent had taken Fansidar preventively during their last pregnancy in the five years preceding the survey. Nearly half took it once, 26 percent twice, and 24 percent took it at least three times. Women in urban areas (36 percent), uneducated women (41 percent), and women in the middle wealth quintile (37 percent) took the drug most frequently. 128 | Malaria Table 9.5 Use of SP/Fansidar by women during pregnancy Among women who took antimalarial drugs for prevention of malaria during the pregnancy for their last live birth in the 5 years preceding the survey, percentage who took SP/Fansidar, the number of times taken, by background characteristics, Rwanda 2005 Among those who took any SP/Fansidar, number of times taken Background characteristic Percentage who took any SP/Fansidar Number of women who took any antimalarial medication Once Twice Three times or more Number of women who took any SP/Fansidar Residence Urban 36.1 76 (50.0) (23.1) (26.9) 27 Rural 29.1 238 46.9 27.0 22.7 69 Education No education 41.2 70 (48.8) (32.1) (19.0) 29 Primary 27.9 198 48.2 25.6 26.2 55 Secondary or higher (27.2) 46 * * * 12 Wealth quintile Lowest (33.5) 37 * * * 13 Second 25.5 57 * * * 14 Middle 37.3 58 * * * 22 Fourth 32.4 74 * * * 24 Highest 27.3 88 * * * 24 Total 30.8 314 47.8 25.9 23.9 97 Note: An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 9.3 TREATMENT OF MALARIA IN CHILDREN UNDER THE AGE OF FIVE In addition to questions on the availability of mosquito nets and preventive antimalarial treatment in pregnant women, the RDHS-III asked whether children under the age of five had had a fever in the two weeks prior to the survey. If the answer was affirmative, respondents were asked questions about how the fever was treated, including whether or not antimalarial drugs were given and when they were given for the first time. The results are shown in Tables 9.6 and 9.7. Table 9.6 shows the percentage of children under age five who had a fever, the percentage of those with fever who received any type of antimalarial drug, and the percentage of those who took an antimalarial drug who took the drug promptly after the fever appeared. In Rwanda, more than one-quarter of children under the age of five (26 percent) had a fever with or without convulsions in the two weeks preceding the survey. Results according to age show a higher prevalence of fever in children age 6 to 11 months (39 percent) than among those age 48 to 59 months (18 percent). However, analysis by residence shows no significant differential between rural (26 percent) and urban (25 percent) areas. In the provinces, however, there are significant differentials: of the 30 percent of children who had a fever, the highest prevalence was in the South province (30 percent); the lowest was in the North province (23 percent). by level of education, the highest prevalence of fever was among children whose mothers had no education (28 percent). Prevalence by wealth quintile showed only minor, inconsistent variations. Malaria | 129 Table 9.6 Prevalence and prompt treatment of children with fever Percentage of children under age five with fever in the two weeks preceding the survey, and among children with fever, the percentage who took antimalarial drugs and the percentage who took the drugs the same or next day following the onset of fever, by background characteristics, Rwanda 2005 Among children under age five: Among children under age five with fever: Background characteristic Percentage with fever in the two weeks preceding the survey Number of children Percentage who took antimalarial drugs Percentage who took antimalarial drugs same or next day Number of children Age in months < 6 19.5 891 5.2 1.1 174 6-11 38.9 830 13.5 3.8 323 12-23 36.9 1,626 13.9 2.4 600 24-35 24.0 1,732 13.5 2.5 416 36-47 20.8 1,373 11.5 3.2 286 48-59 18.4 1,346 10.6 1.0 247 Residence Urban 25.3 1,144 10.5 1.3 289 Rural 26.4 6,653 12.6 2.7 1,757 Province Kigali city 25.2 599 14.8 0.5 151 South 29.5 1,909 16.2 3.4 563 West 23.6 2,075 6.5 1.5 490 North 22.9 1,571 4.6 1.0 360 East 29.3 1,644 18.5 4.1 482 Education No education 28.3 2,172 10.0 1.3 616 Primary 26.0 4,938 13.7 3.0 1,286 Secondary or higher 21.0 687 9.0 3.3 145 Wealth quintile Lowest 27.8 1,612 11.7 2.0 448 Second 24.8 1,605 12.5 2.1 398 Middle 25.8 1,620 11.3 2.9 418 Fourth 27.5 1,525 12.7 2.3 420 Highest 25.2 1,436 13.3 3.2 361 Total 26.2 7,797 12.3 2.5 2,046 With respect to treatment, the results show that of all the children who had a fever, only 12 percent received antimalarial drugs and only 3 percent took them early, that is, either the day the fever appeared or the following day. This means that, in Rwanda, a very small proportion of children with fever receive effective treatment. Results according to age show little variation regarding antimalarial treatment, except for children under six months (5 percent) who were treated less frequently than older children (11 percent of those age 48-59 months). Although the results do not show significant differences with respect to fever prevalence, the proportion of children treated is higher in rural (13 percent) than in urban (11 percent) areas. The same trend is seen with respect to early administration of treatment (3 percent for rural areas, compared with 1 percent for urban areas). By province, the East (29 percent) and South (30 percent) provinces have the highest fever prevalences. These provinces also have the highest proportions of children who received antimalarial treatment (19 percent in the East, 16 percent in the South), and the highest proportions of children who received treatment promptly (4 percent in the East, 3 percent in the South). Finally, children whose mothers attended primary school not only benefited most frequently from antimalarial treatment, but also benefited from it earliest (3 percent). Results do not vary significantly by wealth quintile. 130 | Malaria Table 9.7 shows the type and timing of antimalarial treatment received by children with fever. Six percent of those who had a fever in the two weeks preceding the survey were treated with amodiaquine. Only 1 percent took this medication the same or next day after the fever appeared. This drug was given far less frequently to younger children under the age of 6 months (2 percent) than to children in the other age groups (7 percent on average). Use of this drug was more widespread in rural areas (7 percent) than in urban ones (3 percent). By province, the proportion of children treated with amodiaquine ranges from 10 percent in the East province to 3 percent in the North province. Results by level of education and wealth quintile reveal no significant differentials. Five percent of all children with fever received quinine but less then 1 percent received it promptly. Finally, 4 percent of children were treated with SP/Fansidar, but the proportion of those who were treated promptly is negligible (less than 1 percent); this is true for all variables. Overall, it appears that Rwandan households are only infrequently observing any of the procedures for treatment of malaria in children under the age of five. Table 9.7 Type and timing of antimalarial drugs taken by children with fever Among children under age five with fever in the two weeks preceding the survey, the percentage who took specific antimalarial drugs and the percentage who took each type of drug the same or next day after developing fever, by background characteristics, Rwanda 2005 Percentage of children who took drug: Percentage or children who took drug the same or next day: Background characteristic SP/Fansidar Amodiaquine Quinine SP/Fansidar Amodiaquine Quinine Number of children with fever Age in months < 6 1.2 1.9 3.4 0.0 0.0 1.1 174 6-11 3.1 7.2 5.1 1.2 2.5 0.4 323 12-23 4.3 5.4 7.7 0.5 0.8 1.5 600 24-35 5.3 7.2 4.9 1.6 1.3 0.8 416 36-47 2.6 6.8 4.6 1.0 2.8 0.4 286 48-59 4.8 7.8 2.2 0.0 0.8 0.2 247 Residence Urban 1.6 2.8 7.1 0.6 0.2 0.7 289 Rural 4.2 6.8 4.9 0.8 1.6 0.9 1,757 Province Kigali city 4.0 6.3 7.7 0.0 0.0 0.5 151 South 3.9 7.2 7.4 1.3 1.7 0.9 563 West 2.4 4.2 2.3 0.2 1.0 0.5 490 North 1.1 2.6 2.0 0.0 0.2 0.8 360 East 7.3 9.9 7.4 1.8 2.9 1.2 482 Education No education 3.5 5.0 3.3 0.0 0.8 0.5 616 Primary 4.3 7.2 6.2 1.3 1.7 0.9 1,286 Secondary or higher 1.4 3.5 5.2 0.0 1.2 2.1 145 Wealth quintile Lowest 3.1 5.3 5.4 0.3 1.6 0.4 448 Second 5.1 6.5 4.4 0.7 1.5 0.6 398 Middle 4.1 6.6 4.8 1.2 0.8 1.7 418 Fourth 4.3 6.8 5.2 1.4 1.1 0.3 420 Highest 2.8 6.0 6.6 0.4 2.0 1.2 361 Total 3.9 6.2 5.3 0.8 1.4 0.8 2,046 Breastfeeding and Nutrition of Mothers and Children | 131 BREASTFEEDING AND NUTRITION OF MOTHERS AND CHILDREN 10 As stated in the Health Sector Strategic Plan 2005-2009, malnutrition is not only a leading and direct cause of death, particularly among women and children, it is also the underlying cause of numerous other health problems affecting Rwandans. Malnutrition is the result of inadequate food consumption due to inappropriate feeding practices1 and infectious and parasitic diseases that develop under conditions of poor hygiene at the environmental, collective, and individual levels. This chapter analyzes feeding practices for children born in the five years preceding the survey and women and children’s nutritional status. It is divided into three parts: the first part discusses feeding practices including breastfeeding and supplementary feeding; the second part analyzes micronutrient deficiencies (iodine, vitamin A) and anemia; and the third part discusses women and children’s nutritional status based on anthropometric indices (height and weight measurements). 10.1 BREASTFEEDING AND SUPPLEMENTATION Knowledge of feeding practices is crucial to determining children’s nutritional status, which in turn determines their morbidity and mortality. Among these practices, breastfeeding plays a pivotal role. Breast milk has special properties—it is sterile, transmits antibodies from mother to child, and contains all of the nutrients children need during the first six months of life—that prevent nutritional deficiencies and limit the prevalence of diarrhea and other diseases. In addition, prolonged breastfeeding on demand extends postpartum amenorrhea, thereby limiting the mother’s risk of becoming pregnant again too soon and, by lengthening the birth interval, further safeguarding both the health of the mother and the development of the child. Because of the importance of breastfeeding to infant nutrition, mothers were asked whether they had breastfed those of their children who were born in the five years preceding the survey and how old their children were when they initiated breastfeeding. In addition, mothers were asked how long they had breastfed, how frequently, the children’s age when they were introduced to supplementary foods, the type of supplementary foods they were given and, finally, how frequently the different types of foods were given to the child. Mothers were also asked if they had fed their children using a bottle. Initiation of breastfeeding Table 10.1 shows the percentage of children born in the five years preceding the survey who were breastfed and, among breastfed children, the percentage who were breastfed within one hour or within one day following birth, according to background characteristics. Nearly all children born in the five years preceding the survey were breastfed (97 percent); this is true regardless of background characteristic. The proportion is lower only for children whose mothers delivered outside a health facility or at home (92 percent). The high proportion of breastfed children has remained stable since the RDHS-I and RDHS-II surveys (97 percent in 1992 and 2000). 1 Inappropriate feeding practices refer not only to the quality and quantity of food given to children, but also to the timing of introduction of these foods into children’s diets. 132 | Breastfeeding and Nutrition of Mothers and Children Although breastfeeding is widespread, only 41 percent of Rwandan children began breastfeeding within one hour of birth and only 56 percent began within on day of birth. Table 10.1 Initial breastfeeding Percentage of children born in the five years preceding the survey who were ever breastfed, and for last-born children, the percentage who started breastfeeding within one hour and within one day of birth, and the percentage who received a prelacteal feed, by background characteristics, Rwanda 2005 Background characteristic Percentage ever breastfed Number of children Percentage who started breastfeeding within 1 hour of birth Percentage who started breastfeeding within 1 day of birth1 Percentage who received a prelacteal feed2 Number of breastfed children Sex Male 96.9 4,428 41.5 56.6 23.9 4,289 Female 97.4 4,287 40.4 55.7 23.9 4,175 Residence Urban 96.6 1,228 43.8 58.0 21.4 1,186 Rural 97.2 7,487 40.5 55.8 24.3 7,277 Province Kigali city 96.3 655 45.9 59.2 21.3 631 South 97.1 2,122 45.9 60.5 24.2 2,061 West 97.2 2,290 35.0 51.6 23.6 2,226 North 97.5 1,716 41.2 52.7 23.9 1,673 East 97.0 1,932 40.7 58.9 24.8 1,873 Mother’s education No education 97.1 2,470 39.1 54.3 26.1 2,398 Primary 97.2 5,513 41.3 56.5 23.3 5,361 Secondary or higher 96.3 732 44.4 59.7 20.9 704 Assistance at delivery Health personnel 3 96.0 2,479 44.8 58.6 19.3 2,379 Traditional birth attendant 97.6 4,662 39.2 55.1 25.7 4,549 Other (99.1) 38 (45.0) (59.0) (13.5) 38 No one 97.6 1,511 40.6 56.0 26.0 1,475 Place of delivery Health facility 96.0 2,460 45.0 58.7 19.0 2,360 At home 97.7 6,139 39.6 55.0 25.6 5,995 Other 91.7 94 34.7 68.7 40.6 86 Wealth quintile Lowest 96.8 1,845 40.2 56.5 26.5 1,785 Second 97.8 1,794 41.4 55.8 26.4 1,755 Middle 97.2 1,785 41.3 55.9 21.6 1,734 Fourth 96.5 1,742 38.7 55.0 23.9 1,682 Highest 97.3 1,548 43.5 57.7 20.6 1,507 Total4 97.1 8,715 41.0 56.1 23.9 8,464 Note: Table is based on all births whether the children were living or dead at the time of the survey. 1 Includes children who started breastfeeding within one hour of birth. 2 Children given something other than breast milk during the first three days of life before the mother started breastfeeding regularly 3 Doctor, nurse/midwife, or auxiliary midwife 4 Total includes 23 cases where assistance at delivery and place of delivery is unknown. Unfortunately, these percentages represent a decline compared with the RDHS-II 2000 survey in which 48 percent of children were breastfed within one hour of birth and 73 percent were breastfed within one day of birth. In the RDHS-III, three-fifths of children (59 percent) did not begin breastfeeding within one hour of birth and more than four in ten children (44 percent) did not receive breast milk within one Breastfeeding and Nutrition of Mothers and Children | 133 day of birth. This trend can have negative consequences for children, even affecting their chances of survival. This is because the breast milk that is produced in the first twenty-four hours following birth contains colostrum, which transmits the mother’s antibodies to the child, providing crucial resistance to numerous diseases. In addition, newborns who are not breastfed within 24 hours of birth are usually given other liquids in place of breast milk, and these may carry pathogens. Overall, these results indicate that a major effort is needed to inform mothers of the benefits of breastfeeding in the first hours of a child’s life. Although breastfeeding is widely practiced across all subgroups of women, the timing of initial breastfeeding varies by background characteristics. The results show that in urban areas, 44 percent of children are breastfed within one hour of delivery, compared to 41 percent in rural areas. With respect to provinces, the lowest proportion of children breastfed within one hour of birth occurs in the West province (35 percent), followed by the East and North provinces (41 percent for both). The City of Kigali and the South province have the highest proportions of children breastfed within one hour of birth (46 percent for both). The place of delivery seems to be associated with the timing of initial breastfeeding: children born at a health facility (45 percent) are more likely to begin breastfeeding within one hour of birth than children who are born at home (40 percent). Children born outside of a health facility or at home are the most disadvantaged in this regard. The proportion of children breastfed also varies according to the type of assistance received by the mother during childbirth. Children whose birth was assisted by a health professional are more likely to begin breastfeeding in the first 24 hours of life (59 percent) and in the first hour of life (45 percent). Among those whose birth was assisted by a traditional birth attendant, the proportions are lower, 55 percent and 39 percent, respectively. The mother’s level of education affects breastfeeding practices. Children whose mothers have no education are less likely to be breastfed within one hour of birth (39 percent) or within one day of birth (54 percent). As a result, these children are more likely to receive some form of prelacteal food (26 percent). However, children whose mothers have a secondary education or higher—who are also more likely to be born in a health facility with the assistance of trained personnel (see Chapter 8, Maternal and Child Health)—are more likely to begin breastfeeding within one hour of birth (44 percent) and one day of birth (60 percent); these children are also less likely to receive prelacteal food (21 percent). Similar results are seen according to wealth quintile, where differentials in the timing of initial breastfeeding between the poorest and the richest quintiles can be explained by differences in place of delivery and type of assistance received during delivery. Overall, one-quarter of Rwandan children (24 percent) received some form of prelacteal food. The proportion varies from 21 percent in urban areas to 24 percent in rural areas, and from a low of 21 percent in the City of Kigali to a high of 25 percent in the East province. The proportion of children receiving prelacteal food is also higher among children born at home (26 percent), among children in the poorest wealth quintile (27 percent), and among those whose mothers have no education (26 percent). Introduction of supplementary foods According to the recommendations of WHO and UNICEF (which have been adopted by Rwanda), all children should be breastfed exclusively for the first six months of life. Introducing supplementary foods earlier is not recommended because it exposes children to pathogens, thereby increasing their risk of contracting infectious diseases, particularly diarrhea. In addition, it reduces the amount of milk taken from the breast, thereby reducing suckling, which in turn causes a reduction in milk production. Finally, in poorer populations, supplementary foods are often of poor nutritional value. 134 | Breastfeeding and Nutrition of Mothers and Children After six months, breast milk alone does not cover all of the child’s nutritional needs. It must be supplemented with other appropriate foods to satisfy the child’s needs and to support optimum growth. Information concerning supplemental feeding was obtained by asking the mother whether her child was breastfeeding and what type of food (solid or liquid) it had consumed in the past 24 hours. Although questions about breastfeeding were asked for all children born in the five years preceding the survey, questions on nutritional supplementation were asked only for the most recently born child, and results are presented only for children under the age of three; this is because about half of all children are weaned by the age of three. Table 10.2 and Figure 10.1 show that nearly all children are breastfed at birth and that this practice continues for a long time: at 32-35 months, more than half of all children are still breastfeeding (55 percent). It should also be noted that a high proportion of children under the age of six months are breastfed exclusively (88 percent). While few children under six months receive anything other than breast milk, it should be emphasized that approximately 12 percent of children were not breastfed in accordance with the international recommendations to which Rwanda subscribes. Table 10.2 Breastfeeding status by age Percent distribution of youngest children under three years living with the mother by breastfeeding status and percentage of all children under three years using a bottle with a nipple, according to age in months, Rwanda 2005 Breastfeeding and consuming Percentage using a bottle with a nipple1 Age in months Not breastfed Exclusively breastfed Plain water only Water- based liquids/ juice Other milk Comple- mentary foods Total Number of children Number of children <2 0.0 94.5 0.5 1.6 1.5 1.8 100.0 260 0.3 261 2-3 0.6 91.7 0.9 1.3 4.4 1.1 100.0 322 2.2 324 4-5 0.3 79.7 2.4 2.0 8.2 7.5 100.0 303 6.4 305 6-7 1.3 30.6 1.9 1.6 9.5 55.0 100.0 273 8.4 274 8-9 1.1 6.7 2.1 0.5 6.6 83.0 100.0 275 7.4 279 10-11 3.4 4.5 0.3 0.0 0.3 91.5 100.0 276 4.9 277 12-15 3.6 0.9 1.1 0.5 0.2 93.7 100.0 589 3.2 595 16-19 10.9 1.3 1.0 0.3 0.8 85.7 100.0 497 2.5 532 20-23 22.9 0.5 0.8 0.2 0.0 75.5 100.0 446 2.2 499 24-27 30.5 0.2 0.5 0.0 0.2 68.6 100.0 470 0.9 600 28-31 41.0 0.3 0.3 0.0 0.2 58.2 100.0 315 2.0 490 32-35 45.1 0.0 0.4 0.0 0.3 54.3 100.0 308 0.5 643 <6 0.3 88.4 1.3 1.6 4.9 3.5 100.0 885 3.1 891 6-9 1.2 18.6 2.0 1.0 8.0 69.1 100.0 548 7.9 553 Note: Breastfeeding status refers to a “24-hour” period (yesterday and the past night). Children classified as breastfeeding and consuming plain water only consume no supplements. The categories of not breastfeeding, exclusively breastfed, breastfeeding and consuming plain water, water-based liquids/juice, other milk, and complementary foods (solids and semi-solids) are hierarchical and mutually exclusive, and their percentages add to 100 percent. Thus children who receive breast milk and water-based liquids and who do not receive complementary foods are classified in the water-based liquid category even though they may also get plain water. Any children who get complementary food are classified in that category as long as they are breastfeeding as well. 1 Based on all children under three years Breastfeeding should continue until the child turns two. However, because breast milk alone does not meet all of the infant’s nutritional needs after six months, it must be supplemented with appropriate foods to support normal growth and development. The results of the survey show that 31 percent of children age 6 to 9 months do not receive supplementary foods and, for this reason, are not being adequately nourished. Breastfeeding and Nutrition of Mothers and Children | 135 Feeding with a bottle is not recommended for young children because it is often associated with increased risk of diseases, particularly diarrheal diseases. Inadequately cleansed bottles with poorly sterilized nipples cause gastric disorders, diarrhea, and vomiting in babies. Table 10.2 shows that Rwandan mothers rarely use bottles: only 0.3 percent of children under the age of two months were fed with a bottle in the 24 hours preceding the survey. This proportion reaches a high of 8 percent among children age 6 to 7 months. Overall, 3 percent of children under the age of six months and 8 percent of children 6-9 months were fed with a bottle. Duration and frequency of breastfeeding The median duration of breastfeeding is calculated for most recently born children under the age of three. Table 10.3 indicates that Rwandan children are breastfed for a long period of time. Half of all children are breastfed for 25.2 months. The median durations of exclusive breastfeeding (5.6 months) and predominant breastfeeding (5.9 months) are fairly high. There is no significant difference with respect to gender (26.1 months for boys, 24.4 months for girls). Children are breastfed longer in rural areas (25.6 months) than in urban areas (21.9 months). Results by province show that the median duration of any breastfeeding varies from a high of 27.5 months in the South province to a low of 21.5 months in the City of Kigali. The median duration drops slightly as the mother’s level of education rises: from 25.9 months for children whose mothers have no education, to 25.1 months for children whose mothers have a primary education, to 23.9 months for children whose mothers have a secondary education or higher. Finally, the median duration of any breastfeeding decreases with household wealth. It is highest in the two poorest quintiles (26.7 and 27.7 months) and lowest in the richest quintile (23 months). Overall, the median duration of any breastfeeding has dropped significantly from 32.6 months in 2000 to 25.2 in 2005, a decline of 7.4 months. The mean duration of breastfeeding in Rwanda is 24.9 months, making it one of the longest durations among the sub-Saharan countries surveyed that calculate mean duration in the same way. Figure 10.1 Breastfeeding Practices Among Children Under Age 3 RDHS 2005 >2 2-3 4-5 6-7 8-9 10-11 12-15 16-19 20-23 24-27 28-31 32-35 Age (months) 0% 20% 40% 60% 80% 100% Percent Exclusively breastfed Breastfeeding and complementary foods Breastfeeding and other liquids Not breastfed 136 | Breastfeeding and Nutrition of Mothers and Children Table 10.3 Median duration and frequency of breastfeeding Median duration of any breastfeeding, exclusive breastfeeding, and predominant breastfeeding among children born in the three years preceding the survey, percentage of breastfeeding children under six months of age living with the mother who were breastfed six or more times in the 24 hours preceding the survey, and mean number of feeds (day/night), by background characteristics, Rwanda 2005 Median duration (months) of breastfeeding Frequency of breastfeeding among children under six months of age Background characteristic Any breastfeeding Exclusive breastfeeding Predominant breastfeeding Number of children Percentage breastfed 6+ times in past 24 hours Mean number of day feeds Mean number of night feeds Number of children Sex Male 26.1 5.7 5.9 2,828 98.1 8.1 6.0 431 Female 24.4 5.6 5.8 2,708 98.2 7.8 5.7 443 Residence Urban 21.9 4.8 5.1 771 97.7 7.2 5.9 114 Rural 25.6 5.8 6.0 4,765 98.2 8.0 5.9 760 Province Kigali city 21.5 4.2 4.6 416 98.2 6.8 5.6 58 South 27.5 5.7 5.8 1,324 98.0 7.9 4.7 200 West 25.9 5.6 6.0 1,454 97.3 7.2 5.9 239 North 24.9 6.2 6.3 1,088 99.4 8.8 6.9 171 East 22.0 5.6 5.9 1,253 98.0 8.3 6.2 207 Mother’s education No education 25.9 5.7 5.9 1,538 97.5 8.3 6.4 233 Primary 25.1 5.8 6.0 3,558 98.3 7.9 5.7 581 Secondary or higher 23.9 4.2 4.7 439 98.6 7.0 5.6 60 Wealth quintile Lowest 26.7 6.2 6.4 1,174 98.1 7.8 5.9 194 Second 27.7 5.7 5.9 1,140 97.4 8.3 6.0 174 Middle 24.6 5.8 6.1 1,156 99.2 7.9 6.2 190 Fourth 26.0 5.5 5.7 1,123 98.7 8.1 5.5 174 Highest 23.0 5.0 5.3 943 97.0 7.3 5.7 142 Total 25.2 5.6 5.9 5,535 98.1 7.9 5.9 874 Mean for all children 24.9 6.3 6.7 na na na na na Note: Median and mean durations are based on current status. na = Not applicable Table 10.3 shows that 98 percent of breastfeeding children under six months were breastfed six or more times in the 24 hours preceding the survey. The mean number of feedings is higher during the day than at night (an average of 7.9 times during the day compared with 5.9 times at night). The proportion of children breastfed six or more times in the past 24 hours varies little by background characteristics. This is also true for the mean number of feedings, day or night. Type of supplementary food Table 10.4 shows the types of food consumed by most recently born children under the age of three, according to breastfeeding status. In Rwanda, prior to the age of six months, the introduction of liquids other than breast milk and solid or semi-solid foods is relatively rare. Only 3 percent of children under two months received other liquids and 2 percent received infant formula. Among children age 2-3 months, 3 percent received infant formula and 2 percent received solid or semi-solid foods. Among children age 4-5 months, 9 percent consumed solid or semi-solid foods and 7 percent consumed infant formula. At 6-9 months, only 47 percent of children were receiving fruits and vegetables rich in vitamin Breastfeeding and Nutrition of Mothers and Children | 137 A, but by age 12-15 months, the great majority of children (over three-quarters) were receiving foods rich in vitamin A (77 percent). The introduction of solid or semi-solid foods is recommended starting at the age of six months. Since several types of complementary foods can be given at once, the total of the various percentages can exceed 100 percent. In Rwanda, only a small proportion (4 percent) of children are already consuming solid or semi-solid foods prior to the age of six months. And at 6-7 months, 42 percent of children are not consuming any solid or semi-solid foods as a supplement to breast milk. However, at 6-9 months, 52 percent are consuming fruits and/or vegetables, 40 percent are eating grain-based foods, 32 percent are consuming food made from roots/tubers, and 6 percent are eating meat, poultry, fish, and/or eggs. When the data are limited to children in the 6-7 month age group—the age at which it is generally recommended that supplementary foods be introduced—the proportions are only 38 percent for fruits and/or vegetables, 32 percent for grain-based foods, 19 percent for root/tuber-based foods, and 5 percent for meat, poultry, fish and/or eggs. In the 16-19, 20-23, and 24-35 month age groups, the proportions of nonbreastfeeding children consuming these different types of foods are, in general, slightly higher than those for breastfeeding children, except for grains in the 20-23 month age group. Table 10.4 Foods consumed by children in the day or night preceding the interview Percentage of youngest children under three years of age living with the mother who consumed specific types of food groups in the day or night preceding the interview, by breastfeeding status and age, Rwanda 2005 Age in months Infant formula Other milk/ cheese/ yogurt Other liquids1 Food made from grains Fruits/ vegetables2 Food made from roots/ tubers Food made from legumes Meat/fish/ shellfish/ poultry/ eggs Food made with oil/fat/ butter Fruits and vegetables rich in vitamin A3 Any solid or semisolid food Number of children BREASTFEEDING CHILDREN <2 1.9 1.2 2.6 0.9 1.5 1.5 1.5 0.6 1.5 1.5 1.8 260 2-3 2.8 2.0 1.4 0.4 0.6 0.4 0.6 0.0 0.6 0.6 2.4 320 4-5 7.0 6.9 2.9 3.7 4.4 2.8 1.7 0.3 1.0 3.9 8.7 302 6-7 44.5 13.3 17.8 32.3 37.6 18.8 15.0 4.9 11.9 33.4 58.4 269 8-9 57.9 15.6 28.4 46.6 66.1 44.5 44.3 7.8 33.5 60.1 87.0 272 10-11 61.1 18.1 33.7 49.1 78.1 57.0 63.4 13.5 44.8 72.6 96.7 266 12-15 63.6 14.8 38.4 55.1 82.1 63.1 75.2 14.2 49.6 77.4 98.8 568 16-19 59.4 12.1 37.3 47.4 80.3 65.8 74.3 11.2 47.0 77.2 98.4 443 20-23 56.6 11.1 37.5 50.6 82.9 60.6 77.3 13.5 46.7 78.8 99.4 344 24-35 50.2 10.5 35.9 45.5 84.7 64.3 79.8 10.4 46.9 79.1 99.4 682 <6 4.0 3.4 2.3 1.7 2.1 1.5 1.2 0.3 1.0 2.0 4.4 882 6-9 51.2 14.5 23.2 39.5 51.9 31.7 29.7 6.3 22.8 46.8 72.8 541 NONBREASTFEEDING CHILDREN 0-11 * * * * * * * * * * * 19 12-15 (60.4) (41.1) (59.2) (52.2) (80.2) (51.8) (66.5) (33.4) (52.9) (63.6) (100.0) 21 16-19 70.7 32.9 38.8 65.5 77.5 55.7 78.8 11.1 60.0 67.3 100.0 54 20-23 74.7 16.4 42.8 43.5 81.8 68.1 80.5 13.8 49.0 79.0 96.9 102 24-35 60.7 18.3 46.0 55.2 83.0 64.2 79.8 19.8 57.2 77.7 98.6 411 Note: Breastfeeding status and food consumed refer to a “24-hour” period (yesterday and the past night). An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based o n 25-49 unweighted cases. 1 Does not include plain water 2 Includes fruits and vegetables rich in vitamin A 3 Includes pumpkin, red or yellow yams or squash, carrots, red sweet potatoes, green leafy vegetables, mangoes, papayas, and other locally grown fruits and vegetables that are rich in vitamin A 10.2 MICRONUTRIENT INTAKE AND ANEMIA PREVALENCE Deficiencies in micronutrients such as vitamin A, iodine, iron, calcium, and zinc, are the root cause of various health disorders the symptoms for which can often appear simultaneously. Vitamin A deficiency can cause night blindness; iodine deficiency can cause goiter and impaired mental function; 138 | Breastfeeding and Nutrition of Mothers and Children and insufficient iron causes anemia. These deficiencies also have less visible effects, in particular the weakening of the immune system. Household intake of iodized salt Low iodine consumption is often the source of serious and sometimes irreversible health problems that can increase the risk of miscarriage, perinatal and infant mortality, premature childbirth, congenital anomalies, stunted growth, learning disabilities, impaired mental function, and goiter (the latter being the most visible manifestation of iodine deficiency). Insufficient iodine in food is generally due to poor iodine content in the soil that produced it. In Rwanda, salt falls under ministerial regulation, which maintains tight control over the mandatory import of iodized kitchen salt. During the survey, interviewers tested the kitchen salt of each household they visited. This rapid test, performed with a kit, provides an immediate measurement of iodine content, determining whether or not the salt is iodized and, if it is, whether it contains 15 parts per million (ppm) of iodine. Salt containing at least 15 ppm is considered adequately iodized; salt containing less than 15 ppm is considered inadequately iodized. It should be noted that salt was tested in 87 percent of all households; 10 percent did not have any salt at the time the survey team visited. Table 10.5 shows that 99 percent of households that had salt were using iodized salt: 88 percent of the samples were adequately iodized; 11 percent were inadequately iodized (less than 15 ppm). The percentage of households with adequately iodized salt is slightly higher in urban areas (90 percent) than in rural areas (87 percent), and varies considerably by region, from a low of 71 percent in the West province to a high of 96 percent in the North province. The results do not vary significantly by household wealth. Overall, the proportion of households using iodized salt has risen compared with the 2000 level of 92 percent. Table 10.5 Iodization of household salt Percent distribution of households with salt tested for iodine content, by level of iodine in salt (parts per million), percentage of households tested, and percentage of households with no salt, according to background characteristics, Rwanda 2005 Iodine content among households tested Percentage of households tested Percentage of households with no salt Number of households Background characteristic None (0 ppm) Inadequate (<15 ppm) Adequate (15+ ppm) Total Number of households Residence Urban 0.4 9.3 90.3 100.0 1,263 83.7 10.4 1,510 Rural 1.3 11.3 87.4 100.0 7,652 87.3 9.8 8,762 Province Kigali city 0.3 10.3 89.4 100.0 722 83.5 8.6 864 South 0.4 5.3 94.3 100.0 2,411 88.6 9.3 2,722 West 3.3 26.1 70.6 100.0 2,116 83.9 12.1 2,522 North 0.6 3.8 95.6 100.0 1,706 87.7 10.1 1,946 East 0.6 8.3 91.2 100.0 1,960 88.4 8.6 2,218 Wealth quintile Lowest 1.1 10.6 88.2 100.0 1,899 85.7 11.4 2,217 Second 1.4 8.5 90.1 100.0 1,680 88.1 9.5 1,907 Middle 1.5 11.4 87.1 100.0 1,825 86.1 10.8 2,119 Fourth 1.2 13.0 85.8 100.0 1,865 88.6 9.1 2,105 Highest 0.5 11.3 88.2 100.0 1,646 85.5 8.5 1,925 Total 1.2 11.0 87.8 100.0 8,915 86.8 9.9 10,272 Breastfeeding and Nutrition of Mothers and Children | 139 Micronutrient intake by children Vitamin A deficiency is the main cause of preventable blindness in Africa and a contributor to morbidity and mortality. Even moderate deficiencies in vitamin A affect the immune system, reducing resistance to infection. Vitamin A is indispensable to growth, vision, and the maintenance of epithelial cells. Groups that are vulnerable to vitamin A deficiency include children under the age of five, pregnant women, and nursing mothers. UNICEF and WHO recommend systematic vitamin A supplementation according to a defined protocol for countries whose child mortality rates exceed 70 per thousand. Rwanda follows these main strategies for combating vitamin A deficiency: • Supplementation using vitamin A capsules. • Promotion of the consumption of foods rich in vitamin A. • Promotion of the cultivation of foods rich in vitamin A. Table 10.6 shows the percentage of most recently born children under age three who consumed foods rich in vitamin A in the seven days preceding the survey.2 It also shows the percentage of children age 6-59 months who received at least one dose of vitamin A (capsule or ampoule) in the 6 months preceding the survey. In Rwanda, 84 percent of children age 6-59 months have received vitamin A supplements. There are no significant variations by background characteristics. The youngest children, age 6-9 months, were less likely to receive supplements (75 percent) than children age 10-11 months (88 percent) and 12-23 months (87 percent). In the provinces, the proportions range from a low of 76 percent in the West, to a high of 90 percent in the North province. The data by level of education show that children whose mothers have a secondary education or higher were most likely to benefit from this nutritional supplement (87 percent). However, the proportion of breastfeeding children (85 percent) who received a vitamin A supplement is virtually the same as for nonbreastfeeding children (84 percent). There are no differentials by gender of child or residence. The data vary slightly by wealth quintile: the second wealth quintile has the highest proportion of children who received a vitamin A supplement (87 percent); the fourth quintile and the poorest quintile have the lowest proportions (81 percent and 82 percent, respectively). Finally, there are differentials by age of the mother at the birth of the child, proportions ranging from a low of 80 percent for women under age 20, to a high of 86 percent for women age 25 to 29. To avoid vitamin A deficiency, it is also recommended that children consume foods rich in vitamin A. Nearly six in ten (58 percent) of the most recently born children under age three consumed foods rich in vitamin A in the seven days preceding the survey. The consumption of foods rich in vitamin A increases with age, from 2 percent at under 6 months to 79 percent at age 24-35 months. There is no differential by gender of child. It should be emphasized that breastfeeding children (55 percent) are less likely to consume foods rich in vitamin A than nonbreastfeeding children (77 percent). For this reason, breastfeeding children have an increased risk of vitamin A deficiency, especially if the foods given in place of breast milk during weaning are not rich in this micronutrient. 2 Foods rich in vitamin A are listed in a footnote to tables 10.4 and 10.6. 140 | Breastfeeding and Nutrition of Mothers and Children Table 10.6 Micronutrient intake among children Percentage of youngest children under age three living with the mother who consumed fruits and vegetables rich in vitamin A in the seven days preceding the survey, percentage of children age 6-59 months who received vitamin A supplements in the six months preceding the survey, and percentage of children under five living in households using adequately iodized salt, by background characteristics, Rwanda 2005 Background characteristic Consumed fruits and vegetables rich in vitamin A1 Number of children under age 3 Received vitamin A supplement Number of children age 6-59 months Lives in household using adequately iodized salt2 Number of children under age 5 Age in months <6 2.0 885 na na 88.6 812 6-9 47.1 548 75.0 553 83.3 484 10-11 73.2 276 88.1 277 91.7 253 12-23 77.2 1,532 86.9 1,626 87.0 1,470 24-35 78.5 1,093 84.1 1,732 88.2 1,577 36-47 na na 83.7 1,373 87.5 1,243 48-59 na na 84.1 1,346 86.0 1,215 Sex Male 57.8 2,238 83.9 3,519 87.0 3,589 Female 58.4 2,095 84.3 3,387 87.6 3,465 Breastfeeding status Breastfeeding 55.2 3,725 84.5 3,117 87.0 3,631 Not breastfeeding 76.7 601 83.7 3,721 87.6 3,356 Residence Urban 61.8 594 85.1 1,028 89.1 1,025 Rural 57.6 3,740 83.9 5,879 87.0 6,029 Province Kigali city 64.1 318 79.9 540 88.0 542 South 60.6 1,056 89.2 1,708 94.4 1,758 West 56.3 1,140 75.7 1,831 70.5 1,829 North 61.8 864 90.4 1,397 96.4 1,398 East 52.4 956 84.2 1,431 90.7 1,527 Mother’s education No education 56.8 1,195 82.2 1,935 87.2 1,902 Primary 57.6 2,781 84.5 4,348 87.1 4,509 Secondary or higher 67.0 358 87.0 624 89.1 643 Mother’s age at birth <20 53.3 204 80.3 408 87.9 402 20-24 55.6 1,136 84.1 1,868 87.5 1,913 25-29 58.4 1,156 86.3 1,821 87.6 1,871 30-34 58.5 882 83.7 1,357 86.0 1,392 35-49 61.5 956 82.9 1,453 87.7 1,476 Wealth quintile Lowest 58.0 919 82.0 1,415 87.0 1,413 Second 57.9 914 86.7 1,428 89.7 1,446 Middle 56.7 906 85.3 1,426 86.2 1,451 Fourth 58.5 861 81.2 1,346 86.5 1,415 Highest 59.9 733 85.2 1,292 87.0 1,329 Total3 58.1 4,333 84.1 6,907 87.3 7,054 Note: Information on vitamin A supplements is based on mother’s recall. na = Not applicable 1 Includes pumpkin, red or yellow yams or squash, carrots, red sweet potatoes, green leafy vegetables, mango, papaya, and other locally grown fruits and vegetables that are rich in vitamin A 2 Salt containing 15 ppm of iodine or more. Excludes children in households in which salt was not tested. 3 Includes children for whom breastfeeding status is unknown Breastfeeding and Nutrition of Mothers and Children | 141 The data by residence show a higher proportion of children who consume foods rich in vitamin A in urban areas (62 percent) than in rural areas (58 percent). By province, the proportion of children who consume foods rich in vitamin A varies from a low of 52 percent in the East province to a high of 64 percent in the City of Kigali. Children whose mothers have a secondary education or higher (67 percent) consume more vitamin A-rich foods that those whose mothers have no education or only a primary education (57 percent for both). There appears to be a positive correlation between the age of the mother and child’s consumption of foods rich in vitamin A. The proportions of children who consume this type of food increases with the age of the mother, from a low of 53 percent for children whose mothers were under age 20 when they were born, to a high of 62 percent for those whose mothers were age 35-49. Results according to household wealth show no significant differentials, the proportion of children consuming foods rich in vitamin A ranging from 58 percent in the poorest quintile to 60 percent in the richest quintile. Table 10.6 shows that 87 percent of children under age five live in households with adequately iodized salt. The proportion is highest in the North province (96 percent). Micronutrient intake and night blindness in women Mothers who gave birth in the five years preceding the survey were asked whether they had received a dose of vitamin A in the two months following childbirth. Thirty-four percent of mothers had received the supplement (Table 10.7). The proportion of mothers who received vitamin A varies considerably by province and level of education. In the South province, 43 percent of women received vitamin A within two months of childbirth. In the West province, the proportion was 25 percent. There are significant differentials by level of education: 40 percent of women with a secondary education received this nutritional supplement, compared with only 31 percent of women with no education. Results by other background characteristics show no significant differentials. The first clinical manifestation of vitamin A deficiency is night blindness, which is essentially caused by insufficient vitamin A in the diet. This disorder primarily affects children, pregnant women, and nursing mothers. During the survey, women were asked whether they had suffered from night blindness during pregnancy; that is, whether they had problems seeing at dawn or dusk. Table 10.7 indicates that 8 percent of women who gave birth in the five years preceding the survey reported having vision difficulties at dawn or dusk while pregnant. Some women reported also having vision difficulties during the day. These women appear to have eye problems that are not necessarily related to night blindness. To eliminate these cases, an adjusted night blindness prevalence was calculated. The adjusted night blindness prevalence is 3 percent. Table 10.7 shows the proportion of women who took iron tablets during pregnancy. Overall, nearly three-quarters of the women (71 percent) took no iron during pregnancy. Among those who did take it, 24 percent took it for less than 60 days, 0.6 percent took it for two to three months, and 0.5 percent took it for three months or more. There are differentials in iron consumption by residence: the proportion of women who reported taking iron for less than 60 days is 28 percent in urban areas, compared with 23 percent in rural areas. The data according to province also show significant differentials, from a low of 16 percent in the North province, to a high of 32 percent in the South province. 142 | Breastfeeding and Nutrition of Mothers and Children Table 10.7 Micronutrient intake among mothers Percentage of women with a birth in the five years preceding the survey who received a vitamin A dose in the first two months after delivery, percentage who experienced night blindness during pregnancy, percentage who took iron tablets or syrup for specific numbers of days, and percentage who live in households using adequately iodized salt, by background characteristics, Rwanda 2005 Background characteristic Received vitamin A dose postpartum1 Experienced night blindness during pregnancy Number of days took iron tablets or syrup during pregnancy Number of women Lives in household using adequately iodized salt3 Number of women Reported Adjusted2 None <60 60-89 90+ Missing Mother’s age at birth <20 32.2 4.8 2.4 74.9 23.0 0.0 0.0 2.1 276 87.5 245 20-24 33.4 7.5 2.1 72.0 23.1 0.3 0.7 3.9 1,331 87.6 1,203 25-29 35.1 8.2 3.1 70.3 25.0 1.0 0.5 3.2 1,344 88.0 1,226 30-34 32.0 6.8 1.7 72.0 23.3 0.2 0.5 3.9 1,102 86.7 988 35-49 33.4 9.3 3.2 70.0 24.5 1.0 0.4 4.1 1,372 86.9 1,242 Number of living children 1 32.8 7.3 2.2 71.9 24.1 0.6 0.5 3.0 875 85.8 792 2-3 32.8 6.3 2.1 72.3 23.7 0.3 0.6 3.2 1,706 88.9 1,553 4-5 36.6 8.2 3.1 69.8 24.3 0.8 0.5 4.6 1,349 86.9 1,220 6 or more 31.8 9.6 2.8 70.8 23.9 0.9 0.5 3.9 1,495 86.8 1,339 Residence Urban 34.3 7.2 2.0 65.8 27.9 0.6 0.5 5.2 774 89.3 686 Rural 33.3 8.0 2.7 72.1 23.3 0.6 0.5 3.4 4,651 87.0 4,217 Province Kigali city 28.0 8.2 2.6 68.5 24.4 0.2 0.6 6.3 427 88.0 383 South 42.8 6.3 2.2 63.6 31.7 1.5 0.5 2.7 1,357 94.2 1,249 West 25.3 11.5 3.4 65.2 28.6 0.5 0.6 5.1 1,395 70.3 1,221 North 32.8 5.8 2.3 78.6 16.2 0.3 0.6 4.2 1,05 96.3 942 East 35.1 7.1 2.3 81.2 16.5 0.3 0.2 1.8 1,194 90.3 1,109 Education No education 30.7 8.7 3.1 71.7 23.4 0.5 0.6 3.8 1,552 87.2 1,365 Primary 33.8 7.8 2.4 72.4 23.3 0.7 0.4 3.3 3,404 87.3 3,098 Secondary or higher 40.0 5.2 1.6 60.6 31.0 1.1 0.8 6.5 469 88.0 441 Wealth quintile Lowest 30.8 7.2 3.0 75.3 20.1 0.5 0.5 3.6 1,163 87.5 1,022 Second 34.2 6.7 2.3 76.3 20.1 0.7 0.0 2.9 1,124 90.0 1,012 Middle 33.9 8.4 2.7 71.9 23.9 0.5 0.4 3.3 1,097 86.3 986 Fourth 34.0 8.6 2.7 68.5 27.1 0.8 0.8 2.8 1,069 85.8 987 Highest 34.9 8.6 2.0 62.5 29.7 0.7 1.0 6.1 972 86.9 897 Total 33.5 7.9 2.6 71.2 24.0 0.6 0.5 3.7 5,425 87.3 4,904 Note: For women with two or more live births in the five-year period, data refer to the most recent birth. 1 In the first two months after delivery 2 Women who reported night blindness but did not report difficulty with vision during the day 3 Salt containing 15 ppm of iodine or more. Excludes women in households in which salt was not tested. With respect to education, women with higher education have the highest level of iron supplementation (31 percent, compared with 23 percent among women with no education). There is also a positive correlation between iron consumption during pregnancy and household wealth: the proportion of women who received iron supplements during pregnancy increases with wealth, from 20 percent in the poorest households to 30 percent in the richest. Eighty-seven percent of women live in households with adequately iodized salt. By province, the West province has the lowest percentage (70 percent) and the North province has the highest (96 percent). Breastfeeding and Nutrition of Mothers and Children | 143 10.3 PREVALENCE OF ANEMIA DUE TO IRON DEFICIENCY Insufficient iron is the most widespread micronutrient deficiency in the world, affecting more than 3.5 billion people in developing countries (ACC/SCN, 2000). Anemia is characterized by a reduced number of red blood cells and lower concentrations of hemoglobin in the blood. It is generally the result of diets deficient in iron, vitamin B12, and other nutrients. Although anemia can be caused by parasites, hemorrhaging, and congenital or chronic diseases, it is most often due to nutritional deficiencies based on insufficient iron (DeMaeyer, 1989; Yip, 1994). However, in parasite endemic zones such as Rwanda (see Chapter 9, Malaria), malaria and other parasitic diseases contribute to a high prevalence of anemia. Iron deficiency in children increases the risk of impaired coordination and motor development, learning disabilities, and reduced physical activity. Anemia in women can cause lowered resistance, fatigue and, particularly for pregnant women, increased risk of maternal and fetal morbidity and mortality, and low-birth-weight babies. During the survey, men, women, and children in half of the households surveyed were measured for height and weight and asked to give blood samples to assess hemoglobin content. Samples were collected in the following manner: a) capillary blood was taken by pricking the finger with a retractable blade (Tenderlette); b) a drop of blood was squeezed into a microcuvette, which was then introduced into a portable hemoglobin reader (HemoCue), and the reader produced a hemoglobin value in grams per deciliter of blood (g/dl) in less than one minute; c) the value given was recorded on the questionnaire. There is a three-level classification system for anemia based on blood hemoglobin content that was developed by researchers at WHO (DeMaeyer, 1989). For children over the age of five, nonpregnant women, and men, anemia is considered severe if the hemoglobin content per deciliter of blood is less than 7.0 g/dl; it is considered moderate if the value is between 7.0 and 9.9 g/dl; and it is considered mild if the value is between 10.0 and 10.9 g/dl. The amount of hemoglobin in the blood increases with altitude. This is because the partial pressure of oxygen decreases at high altitudes, as does blood oxygen saturation. There is also a compensation factor that causes increased production of red blood cells to ensure adequate oxygen carrying capacity in the blood (CDC, 1998). In other words, the higher the altitude, the more hemoglobin needed by the blood. Because three-quarters of Rwanda’s population live at high altitudes, the hemoglobin values were adjusted for altitude according to CDC formulas. Prevalence of anemia in children Table 10.8 indicates that more than half of Rwandan children age 6 to 59 months (52 percent) have anemia: 22 percent are mildly anemic, 27 percent are moderately anemic, and 2 percent are severely anemic. About three-quarters of children age 6-9 months are anemic (74 percent). At age 12-23 months, 5 percent are severely anemic, which may be explained by improper weaning. The proportion of children who are anemic is higher in rural areas (52 percent) than in urban areas (47 percent). There are variations by province: the West and East provinces have the highest proportion of anemic children (56 and 58 percent); the North province has the lowest proportion (44 percent). 144 | Breastfeeding and Nutrition of Mothers and Children Table 10.8 Prevalence of anemia in children Percentage of children age 6 to 59 months classified as having anemia, by background characteristics, Rwanda 2005 Anemia status by hemoglobin level Background characteristic Any anemia Mild (10.0- 10.9 g/dl) Moderate (7.0- 9.9 g/dl) Severe (<7.0 g/dl) Number of children Age in months 6-9 74.2 23.0 48.2 3.0 254 10-11 67.7 25.4 41.1 1.2 149 12-23 59.5 22.5 32.3 4.6 796 24-35 50.1 23.5 24.4 2.1 898 36-47 46.0 23.2 21.9 0.8 708 48-59 38.9 17.1 20.9 1.0 732 Sex Male 53.0 23.5 27.0 2.5 1,741 Female 50.1 20.4 27.7 2.0 1,797 Residence Urban 46.6 17.8 26.8 2.0 495 Rural 52.3 22.6 27.5 2.3 3,042 Province Kigali city 54.6 16.6 35.2 2.7 226 South 47.0 20.8 24.0 2.2 908 West 58.2 27.4 30.2 0.5 933 North 43.5 19.6 22.1 1.9 729 East 55.7 20.4 30.8 4.6 741 Mother’s education1 No education 54.4 22.4 29.2 2.8 923 Primary 53.0 22.4 28.5 2.1 1,656 Secondary or higher 47.7 21.0 24.6 2.1 588 Wealth quintile Lowest 54.2 21.9 29.3 2.9 721 Second 56.1 24.9 28.2 2.9 755 Middle 51.1 20.9 28.1 2.1 733 Fourth 50.7 21.0 27.9 1.7 740 Highest 44.1 20.5 22.3 1.3 588 Total 51.5 21.9 27.4 2.2 3,537 Note: Table is based on children who stayed in the household the night before the interview. Prevalence is adjusted for altitude using CDC formulas (1998). g/dl = grams per deciliter 1 For women who were not interviewed, information is taken from the Household Questionnaire. The prevalence of anemia varies somewhat by mother’s level of education: it is lower among children whose mothers have a secondary education or higher (48 percent) than among children whose mothers have no education (54 percent) or only a primary education (53 percent). The data according to household wealth show that anemia prevalence decreases as wealth increases, from 54 percent in the poorest quintile, to 44 percent in the richest. The majority of children who are anemic are moderately so, and they share practically the same characteristics as all anemic children. Breastfeeding and Nutrition of Mothers and Children | 145 Prevalence of anemia in women Table 10.9 shows the results of anemia tests among women. One quarter of the women (26 percent) have anemia: 19 percent are mildly anemic, 6 percent are moderately anemic, and 1 percent are severely anemic. The results according to age show the highest prevalence of anemia among women age 35 and older (29-30 percent). There are differentials between women with no children (23 percent) and those with children, particularly those who have 6 children or more (29 percent). Breastfeeding is not significantly associated with increased risk of anemia. Table 10.9 Prevalence of anemia in women Percentage of women with anemia, by background characteristics, Rwanda 2005 Anemia status Background characteristic Any anemia Mild anemia Moderate anemia Severe anemia Number of women Age1 15-19 21.8 17.0 3.8 1.0 1,317 20-24 25.2 19.0 5.3 0.9 1,145 25-29 25.3 19.2 5.3 0.8 826 30-34 24.7 18.2 5.5 1.0 811 35-39 30.1 20.2 9.5 0.5 536 40-44 29.2 21.9 6.5 0.8 555 45-49 30.3 21.4 8.0 1.0 466 Number of children ever born2 None 22.6 17.4 4.2 1.0 2,142 1 26.8 18.6 7.2 1.0 539 2-3 25.8 20.8 4.5 0.6 1,028 4-5 27.8 18.6 8.1 1.2 876 6 or more 29.0 21.3 7.1 0.6 1,072 Maternity status2 Pregnant 28.8 14.2 13.6 1.0 432 Breastfeeding 25.8 19.9 5.1 0.8 1,923 Neither 25.1 19.2 5.0 0.9 3,302 Residence Urban 22.6 16.7 5.2 0.8 938 Rural 26.2 19.5 5.8 0.9 4,719 Province Kigali city 24.8 18.4 5.6 0.8 547 South 28.3 20.9 6.3 1.2 1,518 West 22.8 17.8 4.5 0.5 1,397 North 17.7 13.1 3.9 0.7 1,020 East 32.7 23.6 8.0 1.1 1,175 Education1 No education 29.2 20.5 7.9 0.8 1,273 Primary 24.9 18.8 5.2 0.9 3,824 Secondary or higher 22.7 17.7 3.9 1.1 560 Wealth quintile Lowest 28.3 19.4 8.1 0.8 1,197 Second 27.2 20.8 5.2 1.2 1,197 Middle 25.9 19.9 4.9 1.1 1,044 Fourth 25.4 18.5 6.2 0.7 1,115 Highest 21.0 16.5 3.9 0.6 1,103 Total 25.6 19.0 5.7 0.9 5,657 Note: Table is based on women who stayed in the household the night before the interview. Prevalence is adjusted for altitude and for smoking status if known using CDC formulas (1998). Women with <7.0 g/dl of hemoglobin have severe anemia, women with 7.0-9.9 g/dl have moderate anemia, and pregnant women with 10.0-10.9 g/dl and nonpregnant women with 10.0-11.9 g/dl have mild anemia. 1 For women who were interviewed, information is taken from the Household Questionnaire. 2 Excludes women who were not interviewed 146 | Breastfeeding and Nutrition of Mothers and Children The prevalence of anemia in women varies according to province. The highest prevalence is observed in the East province (33 percent). The lowest prevalence is found in the North province (18 percent). Anemia prevalence varies according to level of education, from a high of 29 percent among women with no education, to a low of 23 percent among women with higher educational levels. The data show no major differentials by wealth quintile, the proportion of anemic women varying from a low of 21 percent in the richest quintile, to highs of 28 percent in the poorest quintiles. Table 10.10 shows anemia prevalence among children according to the mother’s level of anemia. Anemia measurements exist for both children and their mothers in a total of 3,285 cases. Overall, the prevalence of anemia is higher among children whose mothers are anemic than among all children (62 percent, compared with 52 percent, respectively). Twenty-one percent of children whose mothers are anemic are mildly anemic, 37 percent are moderately anemic, and 1 percent are severely anemic. Results according to the mother’s severity of anemia show correspondence between the anemia status of mother and that of the child. Approximately one-fifth of children whose mothers have mild anemia are also mildly anemic (22 percent). When the mother is moderately anemic, 71 percent of the children show some form of anemia, 20 percent of children are mildly anemic, and 47 percent are moderately anemic. Table 10.10 Prevalence of anemia in children by anemia status of mother Percentage of children age 6-59 months classified as having anemia, by anemia status of mother, Rwanda 2005 Anemia status of child Anemia status of mother Any anemia Mild anemia Moderate anemia Severe anemia No anemia Number of children Any anemia 62.2 21.2 36.6 4.4 37.8 815 Mild anemia 59.6 21.5 34.0 4.1 40.4 625 Moderate anemia 71.0 20.0 45.7 5.3 29.0 172 Severe anemia * * * * 30.5 18 No anemia 48.5 22.2 24.8 1.5 51.5 2 470 Total 51.9 22.0 27.7 2.2 48.1 3,285 Note: An asterisk indicates that the figure is based on fewer than 25 unweighted cases and has been suppressed. Table is based on children who stayed in the household the night before the interview. Prevalence is adjusted for altitude (and for smoking in the case of mothers with information on smoking status) using CDC formulas (1998). Tables includes only cases with anemia measurements for both mothers and children. Prevalence of anemia in men Table 10.11 shows the prevalence of anemia in men. Approximately one in five men (22 percent) are anemic: 10 percent are mildly anemic, 11 percent are moderately anemic, and 1 percent are severely anemic. The proportion of men who are anemic varies widely by age, but anemia prevalence is highest in the youngest and oldest age groups: 29 percent of teenagers and about three in ten men age 45 and above are anemic (27 percent at age 45-49; 30 percent at age 50-59 ). The results show variation by residence: the proportion of men with anemia is 24 percent in rural areas, 15 percent for urban areas. Results by province show the highest prevalences in the East province (29 percent). Breastfeeding and Nutrition of Mothers and Children | 147 Table 10.11 Prevalence of anemia in men Percentage of men age 15-59 with anemia, by background characteristics, Rwanda 2005 Anemia status by hemoglobin level Background characteristic Any anemia Mild (12.0- 12.9 g/dl) Moderate (9.0- 11.9 g/dl) Severe (< 9.0 g/dl) Number of men Age1 15-19 29.3 12.8 15.7 0.8 1,082 20-24 17.9 8.6 8.5 0.8 918 25-29 16.2 6.6 8.3 1.3 615 30-34 14.4 5.7 7.4 1.3 486 35-39 16.9 7.0 9.1 0.7 432 40-44 22.3 10.4 10.4 1.5 398 45-49 27.1 12.4 12.3 2.5 373 50-54 30.3 12.3 15.6 2.4 256 55-59 28.6 11.5 12.3 4.8 145 Residence Urban 14.5 6.8 7.0 0.7 782 Rural 23.5 10.2 11.8 1.4 3,922 Province Kigali city 13.0 5.8 6.5 0.7 476 South 26.6 11.4 13.8 1.4 1,230 West 18.6 8.3 9.5 0.8 1,161 North 16.5 8.1 7.3 1.1 838 East 29.0 12.1 14.7 2.3 1,000 Education1 No education 25.7 9.7 14.3 1.7 827 Primary 23.1 10.2 11.5 1.4 3,317 Secondary or higher 9.5 6.0 3.2 0.3 560 Wealth quintile Lowest 26.6 11.6 13.3 1.7 846 Second 26.0 9.8 14.2 2.1 877 Middle 23.9 10.2 12.4 1.3 963 Fourth 21.2 9.6 10.8 0.8 988 Highest 13.6 7.3 5.5 0.8 1,031 Total 22.0 9.6 11.0 1.3 4,705 Note: Table is based on men who stayed in the household the night before the interview. Prevalence is adjusted for altitude using CDC formulas (1998). 1 For men who were not interviewed, information is taken from the Household Questionnaire. Anemia prevalence varies according to men’s level of education. Men with no education are more likely to have anemia (26 percent) than men with the highest levels of education (10 percent). According to household wealth, the prevalence of anemia decreases as wealth increases, from 27 percent and 26 percent in the two lowest quintiles, to 14 percent in the richest quintile. 10.4 NUTRITIONAL STATUS OF CHILDREN Indicators of child nutritional status were developed to assist in evaluating progress toward meeting the objectives of 20/20 Vision, the Millennium Development Goals (MDG), and the Poverty Reduction Strategy Papers. 148 | Breastfeeding and Nutrition of Mothers and Children Methodology Nutritional status depends both on feeding practices that affect the child’s nutrient consumption and the child’s exposure to infectious diseases. Malnourished children are also more vulnerable to infectious diseases and, for this reason, have an increased risk of morbidity. Nutritional status is evaluated by means of anthropometric indices calculated on the basis of the child’s age and height and weight measurements taken during the survey. Weight and height measurements are used to develop three indices: height in relation to age (height-for-age), weight in relation to height (weight-for-height), and weight in relation to age (weight-for-age). During the survey, all children under age five who were present in the households surveyed were weighed and measured. Data were collected for 3,859 children meeting the defined criteria. Evaluation of child nutritional status follows the recommendations of WHO, based on the rationale that in a well nourished population there is a statistically predictable distribution of height and weight among children of a given age. The survey data are compared with an international reference population known as the NCHS/CDC/WHO3 standard population. This international reference is based on a population of American children under age five in good health, and is applicable to all children of a given age group. It was standardized to follow a normal distribution in which the median and mean are identical. Each of the three indices analyzed is expressed in standard deviations from the median of the international reference population. Children whose nutritional status is below minus two (- 2) standard deviations from the median of the reference population are considered moderately malnourished; children below minus three (- 3) standard deviations are considered severely malnourished. Table 10.12 shows the percentage of children with malnutrition (based on the three anthropometric indices) by background characteristics. Among the 3,859 children for whom data on nutritional status were analyzed, 3,679 lived in the same household as their mother. The mothers of 3,623 of these children were surveyed. For these mothers, nutritional status was also analyzed according to birth interval and the mother’s level of education, using the women’s individual questionnaire. For the 236 other children (6 percent of the 3,859 children analyzed), the mother was not surveyed. In addition, in 54 cases, the mother lived in the same household as her child but was not surveyed because she was either absent or ill at the time of the survey. In 180 cases, the mother was not surveyed because she lived elsewhere or was dead. This latter category is of particular interest because of the assumption that children whose mothers do not live in the household will have different living conditions from children who live with their mother. Stunting Children who suffer from chronic undernourishment (in terms of protein-energy consumption), or chronic malnutrition, are short for their age, or stunted. Stunting reflects failure to receive adequate nourishment over a long period of time and may also be caused by chronic or recurrent illness. Beyond the age of two, children have “little chance of improving growth no matter what interventions are taken” (Delpeuch, 1991). Thus, stunting at the earliest ages is almost never reversed. Height-for-age, therefore, reflects the quality of a child’s environment and, more generally, the society’s level of socioeconomic development. However, children who are short for their age may have weights that correspond to their height. For this reason, chronic malnutrition is not always immediately discernible in a population because a stunted three-year-old may look like a well-fed two-year-old child. Therefore, the height-for-age index, which measures a child’s height in relation to his or her age, is a measure of the long-term effects of malnutrition in a population and does not vary appreciably with the season in which the data were collected. 3 NCHS: the U.S. National Center for Health Statistics; CDC: the U.S. Centers for Disease Control and Prevention; WHO: the World Health Organization. Breastfeeding and Nutrition of Mothers and Children | 149 Table 10.12 Nutritional status of children Percentage of children under five years classified as malnourished according to three anthropometric indices of nutritional status: height-for-age, weight-for-height, and weight-for-age, by background characteristics, Rwanda 2005 Height-for-age Weight-for-height Weight-for-age Background characteristic Percentage below -3 SD Percentage below -2 SD1 Percentage below -3 SD Percentage below -2 SD1 Percentage below -3 SD Percentage below -2 SD1 Number of children Age in months <6 1.4 8.4 0.7 2.3 0.0 2.3 387 6-9 5.1 20.6 1.1 5.4 1.8 17.0 253 10-11 11.3 34.0 0.7 6.6 6.3 26.9 146 12-23 25.3 54.9 1.6 8.6 7.4 35.4 781 24-35 23.0 50.7 0.9 3.2 6.0 27.0 888 36-47 21.8 52.7 0.0 1.2 2.3 17.5 693 48-59 22.2 52.2 0.7 2.1 3.9 19.4 712 Sex Male 19.7 46.3 1.1 4.2 4.8 22.9 1,898 Female 18.9 44.4 0.6 3.6 4.0 22.1 1,961 Birth interval in months2 First birth 16.7 42.3 2.2 6.2 5.1 22.5 597 <24 20.9 48.8 0.3 2.7 4.4 21.7 660 24-47 19.4 44.8 0.7 4.0 3.8 23.3 1,839 48+ 17.0 44.4 0.7 3.2 4.5 21.7 527 Size at birth Very small 36.5 65.0 2.9 9.7 12.9 54.3 87 Small 21.5 48.1 0.5 5.9 6.0 32.4 323 Average or larger 18.0 44.2 0.9 3.7 3.8 20.8 3,197 Residence Urban 13.6 33.1 0.7 3.8 3.2 16.2 543 Rural 20.3 47.3 0.9 3.9 4.6 23.5 3,316 Province Kigali city 12.9 29.2 2.5 7.5 5.0 14.4 250 South 21.5 44.8 1.4 5.0 5.4 27.6 987 West 19.0 46.9 0.5 2.8 3.2 20.3 999 North 22.8 52.2 0.2 2.9 5.7 23.6 793 East 15.8 42.4 0.8 3.8 3.1 20.2 831 Mother’s education No education 21.9 50.3 0.8 3.7 6.5 25.4 1,017 Primary 19.0 44.3 0.9 4.0 3.7 22.4 1,829 Secondary or higher 15.8 43.3 1.3 5.5 3.5 22.1 633 Mother’s age 15-19 (10.7) (27.1) (5.4) (5.4) (3.7) (18.6) 34 20-24 16.7 43.2 1.2 5.5 3.4 23.0 621 25-29 18.5 43.9 0.7 3.1 4.9 20.7 943 30-34 20.7 47.2 1.0 5.0 3.7 22.0 953 35-49 19.5 45.9 0.7 3.2 5.0 24.6 1,127 Wealth quintile Lowest 27.4 55.1 1.0 4.0 7.4 30.5 792 Second 19.7 48.3 1.5 5.8 4.6 25.8 822 Middle 17.7 45.1 0.7 3.6 4.1 22.0 805 Fourth 20.0 45.4 0.5 2.9 4.0 21.8 798 Highest 10.1 29.7 0.4 3.1 1.0 9.7 642 Mother’s status Interviewed 18.9 45.0 0.9 4.0 4.3 22.6 3,623 Not interviewed, but in household 28.1 48.5 2.2 6.4 10.2 20.3 54 Not interviewed, and not in household 26.1 50.5 0.0 1.7 4.7 20.0 180 Total 19.3 45.3 0.9 3.9 4.4 22.5 3,859 Note: Table is based on children who stayed in the household the night before the interview. Each of the indices is expressed in standard deviation units (SD) from the median of the NCHS/CDC/WHO International Reference Population. The percentage of children who are more than three or more than two standard deviations below the median of the International Reference Population (-3 SD and -2 SD) are shown according to background characteristics. Table is based on children with valid dates of birth (month and year) and valid measurement of both height and weight. Figures in parentheses are based on 25-49 unweighted cases. 1 Includes children who are below –3 SD 2 First born twins (triplets, etc.) are counted as first births because they do not have a previous birth interval. 150 | Breastfeeding and Nutrition of Mothers and Children Table 10.12 shows height-for-age results, which indicate that 45 percent of Rwandan children under age five have moderate chronic malnutrition (height-for-age below -2 standard deviations from the median of the reference population) and 19 percent have severe chronic malnutrition (height-for-age below -3 standard deviations from the median of the reference population). These proportions are much higher than would be expected in a well-nourished population (2.3 percent below -2 standard deviations and 0.1 percent below -3 standard-deviations). The nutritional status of children whose mothers do not live in the same household is more worrisome than that of children who live with their mother: 51 percent (compared with 45 percent) have moderate chronic malnutrition; and 26 percent (compared with 19 percent) have severe chronic malnutrition. According to age, the data show large variations in the prevalence of chronic malnutrition, both moderate and severe. The proportion of children with moderate stunting increases steadily with age. It is 8 percent at under 6 months, the period during which children benefit from exclusive breastfeeding. But at 6-9 months, the prevalence is more than twice as high: 21 percent, indicating inadequate introduction of supplementary foods. Between 10 and 11 months, the prevalence of stunting reaches 34 percent, revealing once more the negative impact of inappropriate introduction of supplementary foods. Beginning at the age of 12 months, 51 to 55 percent of children are stunted. The period following, age 12 to 59 months, is critical: not only are children’s increased nutritional needs not being met, but children are also more susceptible to infection at this time (Figure 10.2). The severe form of chronic malnutrition affects less than 1 percent of children under the age of 6 months, 5 percent of those age 6 to 9 months, and 11 percent of those age 10 to 11 months. However, starting at 12 months, more than one in five children (22 to 25 percent) have severe chronic malnutrition. The nutritional status of these children is particularly worrisome insofar as stunting is considered irreversible after two years. Moderate chronic malnutrition affects boys slightly more (46 percent) than girls (44 percent). Figure 10.2 Percentage of Children Under Age 5 Who Are Stunted 45 8 21 34 55 51 53 52 33 47 50 44 43 RWANDA AGE (MONTHS) < 6 6-9 10-11 12-23 24-35 36-47 48-59 RESIDENCE Urban Rural MOTHER'S EDUCATION None Primary Secondary or higher 0 10 20 30 40 50 60 Percent RDHS 2005 Breastfeeding and Nutrition of Mothers and Children | 151 Birth intervals affect the prevalence of moderate chronic malnutrition. Children born less than two years after an older sibling are slightly more affected by malnutrition than other children: 49 percent, compared with 45 percent when the birth interval is 24-47 months. Moderate chronic malnutrition is associated with child’s size at birth. The smaller the size at birth, the higher the prevalence of moderate chronic malnutrition: 65 percent for very small children, 48 percent for small children, and 44 percent for average or large children. The same trend is observed for the severe form of chronic malnutrition. Level of chronic malnutrition varies significantly by residence. Moderate chronic malnutrition affects nearly half the children in rural areas (47 percent), compared with 33 percent in urban areas. For severe malnutrition, the proportion of children who are stunted varies from 20 percent in rural areas to 14 percent in urban areas. By province, the highest prevalence of chronic malnutrition is in the North province (52 percent), followed by the West province (47 percent). Severe chronic malnutrition is highest in the North (23 percent) and the South (22 percent) provinces. Mother’s level of education influences the nutritional status of children, although less than expected. Children whose mothers have no education (50 percent) suffer most from moderate stunting; for those whose mothers have a primary or a secondary education, prevalence is more or less identical (44 percent and 43 percent, respectively). For severe malnutrition, prevalence is highest among children whose mothers have no education (22 percent). By age of the mother, chronic malnutrition is highest among children whose mothers are age 30 to 34 (47 percent for moderate; 21 percent for severe). With respect to household wealth, the results show a strong decrease in chronic moderate malnutrition as wealth increases (from 55 percent in the poorest households to 30 percent in the richest). The results for severe malnutrition are less consistent, although prevalence is more than twice as high in the poorest quintile (27 percent) as in the richest (10 percent). Wasting Table 10.12 also shows results for acute malnutrition, represented by the weight-for-height index. This index, which measures body mass in relation to height, reflects current nutritional status (at the time of the survey). It can therefore vary considerably with the season in which the data are collected. Infectious diseases (measles, diarrhea, etc.), drought, and hunger periods (during food shortages) can affect children’s weight and height. These factors are all very sensitive to seasonal variations. Acute malnutrition reflects insufficient nourishment during the period immediately preceding the survey, or weight loss resulting from illness (severe diarrhea, measles, or anorexia, for example). A child with this form of malnutrition is too thin for his height, or wasted. Children whose weight-for-height is below -2 standard deviations from the median of the reference population have moderate acute malnutrition; those whose weight-for-height is below -3 standard deviations have severe acute malnutrition. In Rwanda, the proportion of children under age five with acute malnutrition, though relatively low, is nearly twice as high for the moderate form (3.9 percent), and more than ten times as high for the severe form (0.9 percent), as would be expected in a well-nourished population (2.3 percent for moderate acute malnutrition; 0.1 percent for severe acute malnutrition). Overall, 4 percent of children under age five are affected by moderate acute malnutrition and nearly 1 percent are affected by the severe form. With respect to age, children age 10 to 23 months suffer most from acute malnutrition, particularly those in the 12-23 month age group (9 percent for moderate; 2 percent for severe) (Figure 10.3). After 24 months, the proportions decrease with age, dropping to 2 152 | Breastfeeding and Nutrition of Mothers and Children percent at age 48 to 59 months for moderate acute malnutrition, and to less than 1 percent for the severe form. This form of malnutrition particularly affects children who do not receive supplementary foods of sufficient quantity and/or quality to meet their increased nutritional needs, resulting in nutritional deficiencies that weaken their resistance to infection. This age group also corresponds to the developmental stage when children begin to explore their immediate environment and place objects in their mouths, making them particularly vulnerable to pathogens. The fact that the proportion of wasted children decreases after the second birthday is not necessarily a sign of improved nutritional status. It can also reflect high mortality among the most vulnerable children, the less vulnerable having survived past their second birthday. There is no difference in the level of wasting by gender (4 percent for both sexes). Results by birth interval show only slight, inconsistent variations. As with chronic malnutrition, the smaller the size of the child at birth, the higher the prevalence of acute malnutrition. Thus, in its moderate form, acute malnutrition prevalence varies from 10 percent among very small children, to 6 percent among small children, to 4 percent among average or large children. There is no variation in wasting by residence for either moderate or severe acute malnutrition. With respect to province, the results show a higher prevalence of moderate acute malnutrition in the City of Kigali (8 percent) and the South province (5 percent) than in other provinces. Results by level of education show no major differentials. However, contrary to expectation, children whose mothers have a secondary education or higher have the highest prevalence of wasting (6 percent). This proportion is four times higher than that of the reference population (Figure 10.3). Finally, it should be noted that children who live with their mothers suffer more from moderate acute malnutrition (at least 4 percent) than those whose mothers do not live in the household (2 percent). It should also be emphasized that the proportion of wasted children is twice as high in the second quintile (6 percent) as in the two richest quintiles (3 percent). 4 2 5 7 9 3 1 2 4 4 4 4 6 RWANDA AGE (MONTHS) < 6 6-9 10-11 12-23 24-35 36-47 48-59 RESIDENCE Urban Rural MOTHER'S EDUCATION None Primary Secondary or higher 0 2 4 6 8 10 Percent Figure 10.3 Percentage of Children Under Age 5 Who Are Wasted RDHS 2005 Breastfeeding and Nutrition of Mothers and Children | 153 Underweight Table 10.12 shows the nutritional status of children by weight-for-age (underweight). This is a composite index of height-for-age and weight-for-height and thus does not distinguish between acute malnutrition (wasting) and chronic malnutrition (stunting) A child can be underweight for his age because he is stunted, wasted, or both. Weight-for-age is a useful tool in clinical settings for continual assessment of nutritional progress and growth. However, its use is limited because it does not distinguish long-term nutritional deficiencies (stunting) from recent ones (wasting). Like weight-for-height, this index is sensitive to seasonal variations and its value is limited when there is only one measurement over time. It is presented here for comparison with the results of studies on growth monitoring that use this measurement. Children whose weight-for-age is below -2 standard deviations from the median of the reference population are classified as moderately underweight; those whose weight-for-age is below -3 standard deviations from the median of the reference population are classified as severely underweight. Nearly one in four children (23 percent) under age five in Rwanda is moderately underweight; 4 percent are severely underweight. This situation is worrisome, because these proportions are significantly higher than those expected in a well-nourished population (2.3 percent for moderately underweight; 0.1 percent for severely underweight). There are significant differentials in underweight by background characteristic. Variations by age are similar to those for stunting. Like the two other indices, this form of malnutrition, which is seen already in infants (approximately 2 percent at under 6 months), increases rapidly, affecting more than one-quarter of children by the time they reach age 10 to 11 months (27 percent) and more than one-third of children age 12 to 23 months (35 percent). The data show no significant variations by gender of child or birth interval. However, children living in rural areas are more likely to be moderately underweight (24 percent) than those in urban areas (16 percent). By province, the results show that more than one-quarter of children in the South province (28 percent) and one-quarter in the North province (24 percent) are moderately underweight. Mother’s level of education has a slight influence on the prevalence of underweight: 22 percent of children whose mothers have a secondary education or higher and 22 percent of children whose mothers have a primary education are moderately underweight, compared with 25 percent of those whose mothers have no education. Differences according to the age of the mother are slight: prevalence varies from 25 percent for children whose mother is between ages 35 and 49, to 21 percent for children whose mothers are age 25 to 29. Finally, in the poorest households, 31 percent of children are moderately underweight, compared with 10 percent in the richest households. Trends by background characteristics for severely underweight children follow the same pattern as for moderately underweight children. 154 | Breastfeeding and Nutrition of Mothers and Children Trends in nutritional status of children Figure 10.4 shows the level of malnutrition among children under the age of three in the RDHS-I (1992), RDHS-II (2000), and RDHS-III (2005) surveys. The findings indicate that the nutritional status of children has not improved significantly since 2000, particularly with respect to stunting. The prevalence of stunting in children age 0-5 years was 48 percent in 1992, 43 percent in 2000, and 45 percent in 2005 (19 percent of which represents the severe form). The prevalence of wasting, which had increased from 4 percent in 1992 to 7 percent in 2000, seems to have declined slightly between 2000 and 2005 to 4 percent, which is the same as the 1992 level. Similarly, the prevalence of underweight decreased slightly from 29 percent in 1992, to 24 percent in 2000, to 23 percent in 2005, nearly the same as the 2000 level. 10.5 NUTRITIONAL STATUS OF WOMEN The nutritional status of women age 15 TO 49 years is a determining factor FOR maternal mortality because it has a major impact on the development and outcome of a pregnancy. It also plays a major role in morbidity and mortality among young children. The nutritional status of mothers is conditioned by dietary intake, health status, and birth spacing. Fertility rates and mortality rates are therefore closely related to the nutritional status of women. For these reasons, it is especially important to assess the nutritional status of women of reproductive age, in order to identify high-risk groups. Although genetic factors contribute to height variations in all populations, short stature can result from chronic malnutrition during childhood, and it is an indirect indicator of a woman’s socioeconomic status. Moreover, given the relationship between height and pelvis size, a woman’s height can be used to predict the risk of complications during pregnancy and delivery. Short women are also more likely to give birth to underweight children. While the cutoff point below which women can be considered at risk varies, it is generally taken to be between 140 and 150 centimeters. To determine the nutritional status of women, the RDHS-III measured the height and weight of all women age 15 to 49 in half of the households surveyed. Table 10.13 shows that the average height of Figure 10.4 Trends in Malnutrition among Children Under 5 Years), Rwanda 1992, 2000, and 2005 48 4 29 43 7 24 45 4 23 Stunting Wasting Underweight 0 10 20 30 40 50 60 Percent RDHS-I 1992 RDHS-II 2000 RDHS-III 2005 Breastfeeding and Nutrition of Mothers and Children | 155 Rwandan women (156.6 centimeters). Only 4 percent of women have a height under 145 centimeters, which is considered the cutoff point. The proportion of short women is higher among very young women, age 15 to 19 years (9 percent), women in rural areas (4 percent, compared with 3 percent to for urban areas), women in the West and South provinces (5 percent), and women in the fourth wealth quintile (5 percent). Being underweight at the start of a pregnancy is a major risk factor affecting pregnancy development and outcome. However, because weight varies considerably according to height, the height- weight relationship must be factored in using an indicator known as the Quetelet or Body Mass Index (BMI).4 This index controls for height in order to distinguish underweight and overweight and has the added advantage of doing away with the reference tables needed to assess weight-for-height. A cutoff point of 18.5 is used to define underweight or undernutrition. A BMI of 25 or above usually indicates overweight or obesity. In Rwanda, the average BMI is 21.8, with a relatively high proportion of women (10 percent) being below the cutoff point of 18.5, indicating chronic undernourishment, 7 percent show the mild form; 2 percent show the moderate form; and 1 percent show the severe form. Low BMI levels correlate with low birth weight and malnourishment in children under age five. In general, the average BMI for women does not vary significantly by background characteristics. However, there are differentials in the proportion below the cutoff point of 18.5. The highest levels of chronic undernourishment are found among the youngest women (age 15 to 19: 17 percent) and the oldest women (age 45 to 49: 13 percent). There is no variation by residence (10 percent for urban and rural), but the prevalence of undernourished women by province varies from a low of 7 percent in the North province to a high of 13 percent in the South province. According to level of education, women with a primary education (10 percent) and women with no education (9 percent) are relatively more likely to be undernourished than women with a secondary education or higher (7 percent). Household wealth also impacts this indicator: women in the poorest households (11 percent) are more likely to be undernourished than women in the richest households (7 percent). Just as chronic undernourishment can be dangerous to overall health, obesity is a risk factor for numerous diseases, including hypertension, cardiovascular disease, and diabetes. Overweight affects only a minority of Rwandan women. Table 10.13 shows that just over one in ten women (12 percent) have a high BMI of 25 or more, and are therefore considered overweight or obese. Overweight mainly affects women age 20 to 34 (12 percent to 14 percent). The problem is more widespread in urban areas (19 percent) than in rural areas (10 percent). 4 The BMI is calculated by dividing weight in kilograms by height in meters squared (kg/m2). 156 | Breastfeeding and Nutrition of Mothers and Children By province, women in the City of Kigali (22 percent) and women in the North (13 percent) and East (12 percent) provinces are more likely to be overweight. The problem seems to be more widespread among women with a secondary education or higher (23 percent) and among those in the richest quintile (23 percent). Table 10.13 Nutritional status of women Among women, mean height, the percentage under 145 cm, mean body mass index (BMI), and the percentage with specific BMI levels, by background characteristics, Rwanda 2005 Height BMI (kg/m2) 1 Background characteristic Mean (in cm) Per- centage < 145 cm Number of women Mean Body Mass Index (BMI) 18.5-24.9 (normal) <18.5 (thin) 17.0-18.4 (mildly thin) 16.0-16.9 (mod- erately thin) <16.0 (severely thin) ≥25.0 (over- weight/ob ese) Number of women Age 15-19 154.1 8.9 1,316 21.3 73.6 16.8 11.5 3.5 1.8 9.6 1,300 20-24 156.4 3.3 1,140 22.4 82.1 4.1 3.5 0.4 0.3 13.8 1,001 25-29 156.9 2.6 839 22.3 82.0 5.8 3.9 1.2 0.7 12.3 672 30-34 157.6 1.4 809 22.1 79.2 7.1 5.6 1.0 0.4 13.7 684 35-39 157.8 2.8 540 21.8 80.4 8.4 6.8 0.9 0.8 11.2 462 40-44 158.5 1.1 553 21.6 79.2 10.3 8.8 1.5 0.0 10.5 523 45-49 158.1 1.5 466 21.3 77.7 13.4 10.2 2.3 0.9 9.0 458 Residence Urban 158.3 2.6 934 22.6 70.9 9.9 6.8 2.3 0.7 19.3 862 Rural 156.3 4.1 4,729 21.7 80.3 9.8 7.4 1.6 0.8 9.9 4,238 Province Kigali city 158.1 2.5 539 22.7 68.1 9.7 6.8 2.3 0.6 22.2 493 South 156.5 4.5 1,514 21.3 79.8 13.1 8.7 3.1 1.3 7.1 1,367 West 155.9 4.9 1,405 21.9 81.5 8.1 6.1 1.2 0.7 10.4 1,280 North 156.9 3.1 1,021 22.2 80.1 6.6 5.5 0.6 0.5 13.3 905 East 156.4 3.0 1,184 21.7 77.4 10.5 8.7 1.2 0.6 12.0 1,055 Education No education 156.3 3.9 1,269 21.8 79.9 9.3 7.9 1.0 0.4 10.8 1,122 Primary 156.2 4.3 3,838 21.7 79.5 10.4 7.4 2.0 1.0 10.1 3,462 Secondary or higher 159.8 0.6 556 22.9 70.3 7.2 5.3 1.2 0.7 22.5 516 Wealth quintile Lowest 155.5 4.4 1,200 21.6 80.6 10.8 8.1 2.0 0.8 8.6 1,080 Second 156.4 3.5 1,194 21.5 81.4 11.1 8.0 2.4 0.6 7.6 1,081 Middle 156.5 3.8 1,046 21.6 80.3 9.8 7.2 1.4 1.2 10.0 928 Fourth 156.3 4.6 1,117 21.6 80.9 10.0 7.6 1.4 1.0 9.0 992 Highest 158.3 2.8 1,106 22.8 70.1 7.3 5.5 1.4 0.5 22.6 1,019 Total 156.6 3.8 5,663 21.8 78.7 9.8 7.3 1.7 0.8 11.5 5,100 1 Excludes pregnant women and women with a birth in the past 2 months Infant and Child Mortality | 157 INFANT AND CHILD MORTALITY 11 This chapter presents information on levels, trends, and differentials in neonatal, postneonatal, infant, child and under-five mortality The information provides mortality statistics to policymakers, program managers and researchers for use in assessing the impact of health policies and programs, and to identify sectors of the population that are at high risk. Estimates of infant and child mortality also serve as a necessary parameters for population projections, particularly if the level of adult mortality can be inferred with reasonable confidence. Finally, indices of childhood mortality are widely accepted as indicators of the overall living conditions of a population. 11.1 DEFINITION, METHODOLOGY, AND DATA QUALITY The primary causes of childhood mortality change as children age. A large component of early infant mortality is due to congenital diseases and other biological factors related to conditions in early infancy. Child mortality (1-4 years), on the other hand, is primarily due to environmental causes which are more susceptible to control, such as infectious disease, malnutrition and accidents. As under-five mortality declines over time, it is often observed that child mortality declines to a greater degree than infant mortality; this phenomenon is mainly due to improvements in children’s environments brought about by public health interventions or general improvements in living standards (Sullivan et al., 1994). In this chapter, age-specific mortality measures are defined as follows: Neonatal mortality: the probability of dying in the first month of life. Postneonatal mortality: the probability of dying between the neonatal period and the first birthday; calculated as the difference between infant and neonatal mortality. Infant mortality: the probability of dying before the first birthday. Child mortality: the probability of dying between the first and fifth birthdays. Under-five mortality: the probability of dying before the fifth birthday. All measures are expressed per 1,000 live births, except for child mortality, which is expressed per 1,000 children surviving to 12 months of age. There are several methods that can be used for the direct calculation of infant and child mortality rates, e.g., period approach, true cohort approach, and synthetic cohort approach. It is beyond the scope of this report to describe the differences between the main approaches, but a technical explanation can be found in the Guide to DHS Statistics (Rutstein and Rojas, 2003). DHS uses the synthetic cohort approach, which calculates mortality probabilities for small age segments, and then combines these component probabilities for the full age segment of interest. The advantage to this method is that mortality rates can be calculated for time periods close to the survey date while still respecting the principle of correspondence. The data needed for the calculations are in the birth history section of the Women’s Questionnaire and include the month and year of birth for all of a woman’s children, their sex and survival status, and the current age at the time of the interview if the child was alive, or age at death if the child has died. The quality of mortality estimates calculated from retrospective birth histories depends on the completeness with which births and deaths are reported and recorded. Potentially the most serious data quality problem is the selective omission from the birth history of children who did not survive, which 158 | Infant and Child Mortality can lead to underestimation of mortality rates. Other potential problems include displacement of birth dates, which may cause a distortion of mortality trends, and misreporting of age at death, which may distort the age pattern of mortality. When selective omission of childhood deaths occurs, the impact is usually most severe for deaths in early infancy. If early neonatal deaths are selectively underreported, the result is an unusually low ratio of deaths occurring in the first seven days to all neonatal deaths, and an unusually low ratio of neonatal to infant deaths. Underreporting of early infant deaths is most commonly observed for births that occurred long before the survey; hence it is useful to examine the ratios over time. An examination of the ratios (see Appendix Tables C.5 and C.6) shows that no significant number of early infant deaths was omitted in the 2005 RDHS. The proportion of neonatal deaths occurring in the first week of life (71 percent) is close to the proportions reported in the 2000 RDHS (72 percent) and the 1992 RDHS (64 percent). Moreover, the proportions are roughly constant over the 20 years preceding the survey (between 67 and 71 percent). The proportion of infant deaths that occur during the first month of life is entirely plausible (47 percent); it is almost the same as the proportion reported in the 2000 RDHS (43 percent) and the 1992 RDHS (48 percent). The proportions are also stable over the 20 years preceding the survey (varying between 53 and 47 percent). This inspection of the mortality data reveals no evidence of selective underreporting or misreporting of age at death that would significantly compromise the quality of the RDHS rates for childhood mortality. 11.2 LEVELS AND TRENDS Table 11.1 shows the variation in neonatal, postneonatal, infant, child, and under-five mortality rates for three successive five-year periods preceding the survey. For the most recent five-year period, infant mortality is 86 deaths per 1,000 live births, and under-five mortality is 152 deaths per 1,000 live births. This means that about one in twelve children born in Rwanda dies before the first birthday, and one in seven children dies before attaining the fifth birthday. Neonatal mortality is 37 deaths per 1,000 live births in the most recent five-year period, while postneonatal mortality is 49 deaths per 1,000 live births. This pattern shows that about 43 percent of deaths under one year of age occur in the neonatal period, and about one-quarter of child deaths under five years occur in the neonatal period. Table 11.1 Early childhood mortality rates Neonatal, postneonatal, infant, child, and under-five mortality rates for five-year periods preceding the survey, Rwanda 2005 Years preceding the survey Neonatal mortality (NN) Postneonatal mortality (PNN)1 Infant mortality (1q0) Child mortality (4q1) Under-five mortality (5q0) 0-4 37 49 86 72 152 5-9 52 69 121 109 217 10-14 56 62 118 91 198 1 Computed as the difference between the infant and neonatal mortality rates Infant and Child Mortality | 159 Figure 11.1 compares infant mortality and under-five child mortality for the five-year period preceding the 1992 RDHS-I, the 2000 RDHS-II, and the 2005 RDHS-III. Results of the RDHS-III show a significant drop in both infant and under-five mortality rates since the 2000 RDHS-II. Comparing the RDHS-III results with those of the 1992 RDHS-I, it can be seen that the rates for these two surveys are almost the same: infant mortality was 85 per 1,000 in 1992 and 86 per 1,000 in 2005; similarly, under- five child mortality was 151 per 1,000 in 1992 and 152 per 1,000 in 2005. These trends suggest that, after the tragic events of 1994, which had negative repercussions on childhood mortality in the mid- and late 1990s, the situation has begun to improve in the past five years. Figure 11.2 shows in more detail the evolution of infant and under-five mortality trends for several five-year periods preceding the RDHS-I, the RDHS-II, and the RDHS-III. Under-five mortality rates, and to a lesser extent infant mortality rates, decreased from the mid- to late 1970s into the mid- to late 1980s. In the 1990s, there was a pronounced deterioration in mortality when it again hovered at or above levels in the 1970s. This deterioration corresponds to periods of civil unrest in the early 1990s, and especially the culmination of this unrest in 1994, which resulted in widespread disintegration of the social and health infrastructure. The first half of the present decade shows a distinct improvement in infant and under-five mortality rates. Results from the RDHS-III indicate that levels of mortality have returned to the relatively lower levels of the late 1980s, providing reason for optimism that socioeconomic conditions are regaining ground lost during the period of conflict: under-five mortality rates decreased from 217 deaths per 1,000 live births in the period 5-9 years before the survey (i.e., 1995-1999) to 152 deaths per 1,000 live births for the period 0-4 years before the survey (i.e., 2000-2005); similarly, infant mortality rates decreased from 121 deaths per 1,000 live births in the period 5-9 years before the survey (i.e., 1995-1999) to 86 deaths per 1,000 live births for the period 0-4 years before the survey (i.e., 2000-2005). This represents about a 43 percent decrease in under-five mortality and a 41 percent decrease in infant mortality in the past five years. Figure 11.1 Trends in Infant and Under-five Mortality, Rwanda 1992, 2000, and 2005 85 151 107 196 86 152 Infant mortality Under-five mortality 0 50 100 150 200 250 Deaths per 1,000 RDHS-I 1992 RDHS-II 2000 RDHS-III 2005 160 | Infant and Child Mortality The infant mortality estimates from the RDHS-III are generally comparable to estimates from other sources. For example, the infant mortality rate published in the U.S. Census Bureau International Data Base is 91 deaths per 1,000 live births for 2005 (U.S. Census Bureau, 2005); the infant mortality rate based on the 2002 Rwanda population census is 107 deaths per 1,000 live births (PRB, 2005); and the official Rwanda government estimate for 2000 is 100 deaths per 1,000 live births (MINALOC, 2001, p. 32). In making such comparisons of mortality data, it is important to keep in mind that estimation techniques vary between sources, and that sampling errors can be fairly large. For example, the 95 percent confidence intervals for the RDHS-III infant mortality estimate of 86 deaths per 1,000 live births are 78 and 94 per 1,000 (Appendix B), indicating that, given the sample size, the true value may be 8 points higher or lower than the estimated rate of 86 per 1,000. 11.3 DIFFERENTIALS IN INFANT AND CHILD MORTALITY Mortality differentials by residence, province, educational level of the mother, and wealth quintile are presented in Table 11.2 and Figure 11.3. In order to have a sufficient number of births to study mortality differentials across population subgroups, period-specific rates are presented for the ten-year period preceding the survey (mid-1995 to mid-2005). Childhood mortality is higher in rural areas than in urban areas: the under-five mortality rate in rural areas (192 per 1,000) is 57 percent higher than that of urban areas (122 per 1,000). There are large differentials by province. The highest levels of mortality are found in the East province, which has an infant mortality rate of 125 per 1,000 and an under-five mortality rate of 233 per 1,000. The lowest levels are found in the City of Kigali (68 per 1,000 for infant mortality; 124 per 1,000 for under-five mortality). Variations in mortality by province should be interpreted with caution because of the relatively large sampling errors when the sample is stratified by province or other background characteristics (see Appendix B). " " " * * * ' ' ' $ $ $ ! ! ! # ## 1979 1984 1987 1989 1992 1997 2002 2005 Year on which estimates are centered 0 50 100 150 200 250 Deaths per 1,000 RDHS-I infant mortality RDHS-I under-five mortality RDHS-II infant mortality RDHS-II under-five mortality RDHS-III infant mortality RDHS-III under-five mortality # ! $ ' * " Figure 11.2 Trends in Infant and Under-five Mortality from the RDHS-1, RDHS-II, AND RDHS-III 225 176 151 141 219 217 196 152 129 121 107 86 118 85 86 95 110 198 Infant and Child Mortality | 161 Table 11.2 Early childhood mortality rates by background characteristics Neonatal, postneonatal, infant, child, and under-five mortality rates for the 10-year period preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Neonatal mortality (NN) Postneonatal mortality (PNN)1 Infant mortality (1q0) Child mortality (4q1) Under-five mortality (5q0) Residence Urban 32 37 69 57 122 Rural 46 62 108 94 192 Province Kigali city 27 40 68 60 124 South 48 59 107 80 178 West 43 57 100 87 179 North 42 47 89 77 160 East 48 76 125 123 233 Education No education 46 71 117 106 210 Primary 45 55 101 87 179 Secondary or higher 28 36 64 34 95 Wealth quintile Lowest 51 63 114 110 211 Second 49 62 111 94 195 Middle 37 53 90 88 170 Fourth 48 72 121 95 204 Highest 33 41 73 52 122 1 Computed as the difference between the infant and neonatal mortality rates 122 192 210 179 95 211 195 170 204 122 RESIDENCE Urban Rural EDUCATION None Primary Secondary or higher WEALTH QUINTILE Lowest Second Middle Fourth Highest 0 40 80 120 160 200 240 Deaths per 1,000 live births Figure 11.3 Under-five Mortality by Mother’s Background Characteristics RDHS 2005 162 | Infant and Child Mortality Mother’s level of education is inversely related to a child’s risk of dying. Higher levels of educational attainment are usually associated with lower mortality rates, in part because education exposes mothers to information about better nutrition and adequate spacing between births, as well as better knowledge about childhood illness and treatment. Specifically, significant differences exist between the mortality rates of children of women who have attained secondary education and above and those with only primary education or no formal education. In Figure 11.3, the under-five mortality rate of children born to mothers with no education are the highest (210 deaths per 1,000 live births) followed by that of mothers with primary education (179 per 1,000 live births) and mothers with no formal education (95 deaths per 1,000 live births). The same monotonic decrease is evident for infant mortality rates. Under-five mortality rates by wealth quintile generally show the expected direction, with children in poorer households having a higher probability of dying than children in the richest households. Children in fourth-quintile households, however, have about the same survival chances as children in the poorest households. This result merits deeper analysis. Childhood mortality rates by sex of child, age of mother at birth, birth order, previous birth interval, and size at birth are presented in Table 11.3. Differences in mortality at birth between male and female children are found in nearly all populations. The results show that female mortality is lower than male mortality at all ages up to five years. Table 11.3 Early childhood mortality rates by demographic characteristics Neonatal, postneonatal, infant, child, and under-five mortality rates for the 10-year period preceding the survey, by demographic characteristics, Rwanda 2005 Demographic characteristic Neonatal mortality (NN) Postneonatal mortality (PNN)1 Infant mortality (1q0) Child mortality (4q1) Under-five mortality (5q0) Child’s sex Male 46 60 106 90 187 Female 42 57 99 87 177 Mother’s age at birth <20 64 75 139 102 227 20-29 40 60 99 92 182 30-39 44 54 98 82 173 40-49 56 51 107 78 176 Birth order 1 53 60 113 87 190 2-3 38 62 99 94 184 4-6 39 54 93 86 172 7+ 57 58 115 84 189 Previous birth interval2 <2 years 70 79 149 113 245 2 years 35 53 88 90 170 3 years 24 48 71 72 138 4+ years 30 47 78 71 143 Birth size3 Small/very small 56 51 108 na na Average or larger 31 47 78 na na na = Not applicable 1 Computed as the difference between the infant and neonatal mortality rates 2 Excludes first-order births 3 Rates for the five-year period before the survey Infant and Child Mortality | 163 The relationship between mother’s age at birth and infant mortality shows the expected U-shaped pattern, with infants of the youngest and oldest women having the greatest risk of dying. Neonatal mortality shows a similar U-shaped pattern. Under-five mortality rates show a slightly weaker pattern: children under five born to the youngest women (under 20 years) still have the greatest risk of dying but children born to older women have a relatively better chance of survival. The length of the birth interval has a significant impact on a child’s chances of survival, with short birth intervals increasing the risk of dying. As the birth interval gets longer, the mortality risk is reduced considerably. Children born less than two years after a prior sibling have substantially greater risk of dying than children born after an interval of two or more years. For example, the infant mortality rate is 149 deaths per 1,000 live births for children born after an interval of less than two years, compared with 71 deaths per 1,000 for children born after an interval of three years. Size of child at birth has a bearing on childhood mortality rates. Children whose birth size is small or very small have a 38 percent greater risk of dying before their first birthday than those whose birth size is average or larger. The same trend can be seen for neonatal and postnatal births, but not as strong. 11.4 PERINATAL MORTALITY Pregnancy losses occurring after seven completed months of gestation (stillbirths) plus deaths to live births in the first seven days of life (early neonatal deaths) constitute perinatal deaths. The perinatal mortality rate is derived when the total number of perinatal deaths is divided by the total number of pregnancies reaching seven months gestation. The distinction between a stillbirth and an early neonatal death may be a fine one, depending often on the observed presence or absence of some faint signs of life after delivery. The causes of stillbirths and early neonatal deaths are overlapping, and examining just one or the other can understate the true level of mortality around delivery. 139 99 98 107 113 99 93 115 149 88 71 78 MOTHER'S AGE <20 years 20-29 years 30-39 years 40-49 years BIRTH ORDER 1 2-3 4-6 7+ PREVIOUS BIRTH INTERVAL <2 years 2 years 3 years 4+ years 0 20 40 60 80 100 120 140 160 Deaths per 1,000 births Figure 11.4 Infant Mortality by Reproductive Behavior RDHS 2005 164 | Infant and Child Mortality Table 11.4 shows the number of stillbirths and early neonatal deaths, and the perinatal mortality rate for the five-year period preceding the survey by background characteristics. The results indicate that the perinatal mortality rate is 44 deaths per 1,000 pregnancies. Pregnancies with an inter-pregnancy interval of less than 15 months have a higher perinatal risk (79 deaths per 1,000 pregnancies) than other pregnancies. Table 11.4 Perinatal mortality Number of stillbirths and early neonatal deaths, and the perinatal mortality rate for the five-year period preceding the survey, by background characteristics, Rwanda 2005 Background characteristic Number of stillbirths 1 Number of early neonatal deaths 2 Perinatal mortality rate 3 Number of pregnancies of 7+ months duration Mother’s age at birth <20 6 22 52 539 20-29 90 111 42 4,740 30-39 46 67 39 2,896 40-49 18 27 64 696 Previous pregnancy interval (in months) First pregnancy 32 59 59 1,551 <15 17 34 79 645 15-26 33 55 36 2,440 27-38 32 49 34 2,398 39+ 45 29 40 1,838 Residence Urban 19 19 30 1,247 Rural 140 208 46 7,625 Province Kigali city 8 12 30 663 South 48 43 42 2,170 West 36 65 44 2,325 North 25 43 39 1,740 East 43 64 54 1,973 Education No education 43 78 48 2,511 Primary 108 135 43 5,620 Secondary or higher 9 15 32 741 Wealth quintile Lowest 29 63 49 1,873 Second 41 41 44 1,835 Middle 43 35 43 1,827 Fourth 24 60 48 1,767 Highest 23 28 32 1,570 Total 160 227 44 8,872 1 Stillbirths are fetal deaths in pregnancies lasting seven or more months. 2 Early neonatal deaths are deaths at age 0-6 days among live-born children. 3 The sum of the number of stillbirths and early neonatal deaths divided by the number of pregnancies of seven or more months’ duration. Infant and Child Mortality | 165 As with neonatal mortality, perinatal mortality is significantly higher in rural areas (46 per 1,000) than in urban areas (30 per 1,000). Results by province show the lowest rate in the City of Kigali (30 per 1,000) and the highest rate in the East province (54 per 1,000). As expected, results by mother’s educational attainment show a higher risk of perinatal death for mothers with no education than for other mothers (48 per 1,000, compared with 43 per 1,000 for women with a primary education and 32 per 1,000 for women with a secondary education or higher). Results by wealth quintile show the highest rate of perinatal mortality among women living in the poorest households (49 per 1,000, compared with 32 per 1,000 for the richest households). However, here too, the mortality rate for the fourth quintile is nearer that of the poorest quintiles than that of the richest quintile. A closer examination of the data is needed to establish the significance of this result. 11.5 HIGH-RISK FERTILITY BEHAVIOR Numerous studies have found a strong relationship between children’s chances of dying and certain fertility behaviors. Typically, the probability of dying in early childhood is much greater if children are born to mothers who are too young or too old, if they are born after a short birth interval, or if they are born to mothers with high parity. Very young mothers may experience difficult pregnancies and deliveries because of their physical immaturity. Older women may also experience age-related problems during pregnancy and delivery. For purposes of this analysis, a mother is classified as “too young” if she is less than 18 years of age and “too old” if she is over 34 years of age at the time of delivery; a “short birth interval” is defined as a birth occurring within 24 months of a previous birth; and a “high-order” birth is one occurring after three or more previous births (i.e., birth order four or higher). First-order births may be at increased risk of dying, relative to births of other orders; however, this distinction is not included in the risk categories in Table 11.5 because it is not considered avoidable fertility behavior. Also, for the short birth interval category, only children with a preceding interval of less than 24 months are included. Short succeeding birth intervals are not included—even though they can influence the survivorship of a child—because of the problem of reverse causal effect (i.e., a short succeeding birth interval can be the result of the death of a child rather than being the cause of the death of a child). Table 11.5 presents the distribution of children born in the five years preceding the survey by categories of increased risk of mortality. Column 2 shows the percentage of children falling into specific categories. Column 3 shows the risk ratio of dying for children by comparing the proportion dead among children in each high-risk category with the proportion dead among children not in any high-risk category (i.e., those whose mothers were age 18-34 at delivery, who were born 24 or more months after the previous birth, or who are of birth order two or three). Sixty percent of children in Rwanda fall into a high-risk category, with 33 percent in a single high-risk category and 27 percent in a multiple high-risk category. High risks are associated with birth intervals of less than 24 months, births to mothers older than 34 years, births of parity higher than three, and births to mothers younger than 18 years under the single high-risk category. Contrary to what might be expected, risk ratios are higher for children in an unavoidable risk category than for children born into a single or multiple risk categories. This may be explained by the fact that mothers with a high-risk pregnancy may seek better prenatal or delivery care than other mothers, thus ensuring greater chances of survival for their child despite the risks. The highest risk (1.4) is associated with fourth and higher births that occur less than 24 months after a previous birth; 7 percent of births fall into this multiple high-risk category. Another 9 percent of births in Rwanda have a short birth interval as the sole risk factor; these children run a 30 percent greater chance of dying than children who are not in any high-risk category. 166 | Infant and Child Mortality The last column of Table 11.5 addresses the question of what percentage of currently married women have the potential for a high-risk birth. This was obtained by simulating the distribution of currently married women according to the risk category in which a birth would fall if a woman were to conceive at the time of the survey. Although many women are protected from conception because of postpartum insusceptibility, prolonged abstinence, and the use of family planning, for simplicity only those who have been sterilized are included in the “not in any high-risk category.” Overall, 82 percent of currently married women have the potential for having a high-risk birth, with 29 percent falling into a single high-risk category and 54 percent falling into a multiple high-risk category. Table 11.5 High-risk fertility behavior Percent distribution of children born in the five years preceding the survey by category of elevated risk of mortality and the risk ratio, and percent distribution of currently married women by category of risk if they were to conceive a child at the time of the survey, Rwanda 2005 Births in the 5 years preceding the survey Risk category Percentage of births Risk ratio Percentage of currently married women1 Not in any high-risk category 23.3 1.00 13.6a Unavoidable risk category First-order births between ages 18 and 34 years 17.2 1.29 4.5 Single high-risk category Mother’s age <18 1.3 1.45 0.0 Mother’s age >34 0.8 0.66 2.2 Birth interval <24 months 9.1 1.29 11.5 Birth order >3 21.2 0.79 14.8 Subtotal 32.5 0.96 28.5 Multiple high-risk category Age <18 and birth interval <24 months 2 0.1 * 0.0 Age >34 and birth interval <24 months 0.1 * 0.2 Age >34 and birth order >3 17.2 0.91 28.2 Age >34 and birth interval <24 months and birth order >3 2.7 1.86 9.4 Birth interval <24 months and birth order >3 7.0 1.44 15.7 Subtotal 27.0 1.14 53.5 In any avoidable high-risk category 59.5 1.04 82.0 Total 100.0 na 100.0 Number 8,715 na 5,510 Note: Risk ratio is the ratio of the proportion dead among births in a specific high-risk category to the proportion dead among births not in any high-risk category. An asterisk indicates that a figure is based on fewer than 250 births and has been suppressed. na = Not applicable 1 Women are assigned to risk categories according to the status they would have at the birth of a child if they were to conceive at the time of the survey: current age less than 17 years and 3 months or older than 34 years and 2 months, latest birth less than 15 months ago, or latest birth being of order 3 or higher 2 Includes the category age <18 and birth order >3 a Includes sterilized women Maternal Mortality | 167 MATERNAL MORTALITY 12 12.1 INTRODUCTION Maternal mortality has become an important measure of human and social development. It is particularly revealing of women’s overall status, access to health care, and the responsiveness of the health care system to their needs. Therefore, knowledge of maternal mortality levels is very important not only for identifying the risks associated with pregnancy and childbearing, but also for what it says about women’s health and, indirectly, their economic and social status. Determining the level maternal mortality and the associated risk factors is necessary for both diagnosing issues and assessing the progress and effectiveness of existing programs. The 2000 RDHS-II was the first DHS survey to collect data for use in estimating maternal mortality using the direct sisterhood method. The same methodology was used to collect data on maternal mortality in the 2005 RDHS-III. Maternal mortality is calculated using data on the sisters of respondents. The information gathered on each of the respondent’s sisters included current age and, if the sister was dead, age at death (AD), and the number of years since the death (YSD). For dead sisters, additional questions were asked to determine whether the death was related to childbearing, i.e., whether the death occurred during pregnancy, childbirth, or within two months of the end of a pregnancy or childbirth. The direct method of calculating maternal mortality presented here relies on detailed information about respondents’ sisters, including the current age of all surviving sisters, the age at death of dead sisters, and the number of years since the death occurred. To obtain well defined reference periods, the data are aggregated to determine the number of person-years of exposure to mortality risk and the number of maternal deaths occurring in the defined reference periods. Maternal mortality rates are then estimated directly by dividing the number of maternal deaths by the number of person-years of exposure. The result of this calculation is the proportion of sisters, among all of the respondents’ sisters, who died from maternity- related causes. This estimate of the probability of dying from maternity-related causes is unbiased, provided that the risk of dying is identical for all sisters (Trussel and Rodriguez, 1990). 12.2 DATA COLLECTION The questionnaire used to gather data on maternal mortality is presented in Appendix F (Section 10 of the individual questionnaire). First, the woman is asked to list all of her siblings, i.e., all of the children born to her biological mother beginning with the first born. Next, the respondent is asked the survival status of each of her siblings, and the current age of those still living. For dead siblings, the respondent is asked the age of the sibling at death and the number of years since the death occurred. If the exact age or number of years could not be obtained, interviewers were authorized to accept approximate answers. For sisters who died at the age of 12 or older, the respondent is asked further questions to determine whether the death was maternity-related: 168 | Maternal Mortality - Was (NAME) pregnant when she died? If the answer is no or don’t know, the respondent is asked: - Did (NAME) die during childbirth? If the answer is no, the respondent is asked: - Did (NAME) die within two months of the birth of a child or termination of a pregnancy? These questions are structured to encourage the respondent to report all deaths following a pregnancy regardless of the outcome, including a pregnancy ending in induced abortion, while avoiding posing direct questions regarding such events. All such deaths are considered maternal deaths. 12.3 DATA QUALITY The estimation of maternal mortality rates requires accurate reporting of the number of sisters the respondent ever had, the number who died, and the number who died of maternity-related causes. There is no definitive procedure for establishing the completeness or accuracy of retrospective data on sister survivorship. The direct approach requires not only accurate data on sister survivorship, but on age at death and number of years since the death of a sister—information that may embarrass respondents or require them to provide details they do not have. The number of brothers and sisters reported by the respondent and the completeness of the reported data on current age, age at death, and years since death are presented in Table 12.1. Table 12.1 Data on siblings Number of siblings reported by survey respondents and completeness of the reported data on age, age at death (AD), and years since death (YSD), Rwanda 2005 Sisters Brothers Total Sibling status and completeness of reporting Number Percentage Number Percentage Number Percentage All siblings 35,963 100.0 36,405 100.0 72,368 100.0 Living 25,688 71.4 23,374 64.2 49,062 67.8 Dead 10,074 28.0 12,504 34.3 22,577 31.2 Status unknown 201 0.6 527 1.4 729 1.0 Living siblings 25,688 100.0 23,374 100.0 49,062 100.0 Age reported 25,470 99.2 23,145 99.0 48,614 99.1 Age missing 218 0.8 230 1.0 448 0.9 Dead siblings 10,074 100.0 12,504 100.0 22,577 100.0 AD and YSD reported 9,546 94.8 11,869 94.9 21,415 94.9 Missing only AD 82 0.8 117 0.9 199 0.9 Missing only YSD 342 3.4 375 3.0 716 3.2 Missing both AD and YSD 104 1.0 143 1.1 247 1.1 Maternal Mortality | 169 Complete data were obtained for nearly all sisters, regard- less of survival status. Current age was reported for nearly all surviving sisters (99 percent), and age at death as well as number of years since death were reported for 95 percent of dead sisters. These percentages are indicative of good data quality. Rather than exclude siblings with missing data from further analysis, information on the birth order of siblings was used in conjunction with other information to impute the missing data.1 Sibling survivorship data, including cases with imputed values, were used to directly estimate adult and maternal mortality. Missing date information is only one indicator of overall data quality. Completeness of basic information, such as number of siblings, is much more important. Table 12.2 shows other indicators of data completeness. First, it is expected that the distribution of respondents’ birth years will be roughly equivalent to that of their sibship.2 A median sibship year of birth that is much later than that the respondents median birth year indicates that older siblings have been systematically omitted, perhaps because some of them died before the respondent was born. Such omissions would affect adult mortality estimates. For Rwanda, respondents and siblings have the same median year of birth,3 1970, indicating that there is no serious underreporting of siblings. However, for maternal mortality assessments, the completeness of sibling reporting is not what’s most important; rather, it is the completeness of data relating to those who are exposed to the risk of dying from maternity-related causes: sisters of childbearing age. It is crucial that these data be as complete as possible. Two other tests, sex ratio of births (defined as number of males per 100 females) and mean sibship size, can be used to assess the completeness of sibling reporting. The results appear in Table 12.2. 1 The imputation procedure is based on the assumption that the reported birth order of brothers and sisters is correct. The first step is to calculate birth years for each living sibling whose age is known, and for each dead sibling for whom the age at death and years since death are known. For siblings missing these data, a date of birth is imputed within a range defined by the birth dates of the “bracketing” siblings. In the case of living siblings, an age was then calculated from the imputed birth date. In the case of dead siblings, if either the age at death or years since death was reported, this information was combined with the imputed birth date to produce the missing information. If both pieces of information were missing, the distribution of age at death of siblings for whom years since death were unreported but age at death was known, was used as the basis for imputing age at death. 2 The term sibship used here refers to all of a respondent’s siblings born of the same biological mother. 3 It should be noted that the distribution of birth years is not the same for siblings as for respondents: respondent birth years are distributed over 35 years (1955 to 1990); sibling birth years are distributed over 76 years (1927 to 2005). Table 12.2 Indicators on data quality Percent distribution of respondents and siblings by year of birth, median birth year, mean sibship size and sex ratio of births, Rwanda 2005 Percent distribution Birth year Respondents Siblings Before 1955 0.0 4.7 1955-59 6.7 5.5 1960-64 9.8 8.1 1965-69 9.7 10.9 1970-74 12.8 12.7 1975-79 15.0 14.6 1980-84 20.6 14.9 1985 or later 25.3 28.6 Total 100.0 100.0 Interval 1955-1990 1927-2005 Median 1970 1970 Number 11,321 70,411 Respondent’s year of birth Mean sibship size Sex ratio at birth of siblings 1955-59 7.3 99.7 1960-64 7.5 102.7 1965-69 7.8 102.2 1970-74 7.6 101.4 1975-79 7.5 102.7 1980-84 7.4 100.4 1985-90 7.0 100.4 Total 7.4 101.2 170 | Maternal Mortality For all siblings, the sex ratio of births is 101 males to 100 females. This is slightly lower than generally observed, because the sex ratio of births is around 105 males per 100 females, with only slight variations, for all populations. In Rwanda, the sex ratio of births varies little by respondent’s year of birth, from 100 to 103. Given the well known variability of sex ratios in small sample sizes, this indicates there has been no serious underreporting of sisters. The data indicate a mean sibship size (including the respondent) of 7.4, which is very close to the past final parity of Rwandan women. Variations in sibship size by respondent’s year of birth range from 7.0 to 7.8 children. Fertility begins to decline slightly in the 1965-69 period, confirming actual trends in Rwandan fertility. Thus, the relative stability of mean sibship size suggests, as with the previous results, there has been no serious underreporting of siblings. General and maternal mortality estimates cover the past five years (i.e., 0-4 years preceding the survey). This five-year reference period was chosen to obtain the most recent estimate of maternal mortality while still retaining a sufficient number of maternal deaths (which, nevertheless, remains relatively low) to reduce sampling errors to a minimum and ensure a reliable estimate. 12.4 DIRECT ESTIMATES OF ADULT MORTALITY The total number of deaths (613 brothers and 659 sisters) occurring between the ages of 15 and 49 in the five-year reference period (i.e., 0-4 years preceding the survey) is sufficiently large to ensure a reliable estimate of adult mortality. The data for this period are presented in Table 12.3. The results show a relatively high rate of adult mortality: 6.86 per 1,000 for all women and 7.39 per 1,000 for all men. As a comparison, adult mortality in the 2000 RDHS-II was 10.21 per 1,000 for women and 15.18 per 1,000 for men, indicating a significant decline in adult mortality (33 percent for women, 51 percent for men) between the two surveys. It is important to evaluate the reliability of direct estimates of adult mortality because the data on sister mortality serve as the basis of maternal mortality data. If the adult mortality estimate is incorrect, the maternal mortality estimate will also be erroneous. In the absence of precise mortality data for Rwanda, the reliability of the adult mortality estimate is assessed by comparing it to a series of direct rates extrapolated from United Nations model life tables (United Nations, 1982). Table 12.3 Estimates of age-specific female and male adult mortality Direct estimates of age-specific female and male adult mortality based on the survivorship of siblings of survey respondents, for the period 2000-2004, and model life table rates, Rwanda 2005 2000-2004 Age Deaths Years of exposure Mortality rates (‰) Model life table rates WOMEN 15-19 63 19,172 3.28 3.25 20-24 117 20,920 5.58 4.48 25-29 113 17,192 6.58 5.23 30-34 125 14,632 8.51 6.15 35-39 106 11,522 9.16 7.13 40-44 86 7,932 10.83 8.34 45-49 51 4,850 10.44 10.51 15-49 659 96,220 6.86a MEN 15-19 71 18,730 3.79 3.52 20-24 90 19,099 4.69 5.01 25-29 94 15,531 6.02 5.67 30-34 108 12,520 8.63 6.76 35-39 120 9,644 12.49 8.42 40-44 80 6,642 12.07 10.90 45-49 50 3,819 13.06 14.50 15-49 613 85,986 7.39a Note: The model life table rates come from the United Nations Model Life Tables for Developing Countries, “General” mortality pattern, using a level of mortality approximately corresponding to a probability of dying between birth and exact age 5 estimated for the ten years preceding the survey (i.e., 5q0 of 176 per 1,000 female births and 188 per 1,000 male births). a Age adjusted Maternal Mortality | 171 Age-specific mortality rates obtained from model life tables are presented in Table 12.3. The model life table rates are taken from the United Nations “General” pattern because these most closely approximate the infant and child mortality models of Rwanda. They correspond to the probability of dying between birth and exact age five (5q0) estimated for the ten years preceding the survey.4 Underreporting of events and erroneous dating of reported events can affect the validity of retrospective data. The estimates in this survey are subject to underreporting, especially for less recent events. Although the quality assessments indicate no problem of this type, a closer evaluation is required. Evaluation by comparison with United Nations mortality models confirms the quality of the data concerning sibling survivorship, and the general mortality estimates based on these data are sufficiently plausible to be used in estimates of maternal mortality (Figures 12.1 and 12.2). 4 The probability of dying between birth and exact age 5 (5q0) estimated for the ten years preceding the survey is 176 per 1,000 female births and 188 per 1,000 male births (see Chapter 11). # # # # # # # 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age (years) 0 2 4 6 8 10 12 Deaths per 1,000 women Calculated rate Model life table rate# Figure 12.1 Female Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group 172 | Maternal Mortality 12.5 DIRECT ESTIMATES OF MATERNAL MORTALITY Direct estimates of maternal mortality obtained from reports of sister survivorship are presented in Table 12.4. The number of maternal deaths among women age 15-49 is estimated at 130 for the period 0-4 years preceding the survey. Age-specific proportions dying of maternal causes display a consistent pattern, increasing with age up to age 30-34, then decreasing in the older age groups, except for age 40-44. Given the relatively low number of events, the method used was to estimate a single rate corresponding to the reproductive years. The estimate for all mortality due to maternal causes, expressed per 1,000 women-years of exposure to maternal risk, is 1.29 for the 2000-2004 period. This estimate is significantly lower than that of the RDHS-II 2000, survey, which was 1.88 for the 1995-1999 period. The maternal mortality rate can be converted to a maternal mortality ratio (MMR), expressed per 100,000 live births, by dividing the rate by the general fertility rate associated with the same time period (Table 12.4). This brings out the obstetrical risks of pregnancy and childbearing. Using this method, the MMR is estimated to be 750 maternal deaths per 100,000 live births for the period 0-4 years preceding the survey. This ratio has dropped substantially compared with the 2000 RDHS-II, which showed a ratio of 1,071 for the 1995-1999 period. # # # # # # # 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age (years) 0 2 4 6 8 10 12 14 16 Deaths per 1,000 men Calculated rate Model life table rate# Figure 12.2 Male Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group Maternal Mortality | 173 The estimated age-specific proportions of deaths due to maternal causes (Table 12.4) for the 1995- 2004 period display a plausible pattern, being higher at age 30 to 34, when nearly three in ten deaths (29 percent) are related to maternal causes. Unlike the other measures of mortality presented earlier, these proportions are not affected by underreporting because it can be assumed that underreporting does not affect maternal deaths any more than deaths due to other causes. Therefore, it can be estimated that one in five deaths (20 percent) among women of childbearing age (15 to 49) is due to maternal causes. This represents a slight increase compared with the 2000 RDHS-II, which showed an estimate of 16 percent. In conclusion, there has been a significant decline in adult mortality since the 2000 survey (33 percent for women, 51 percent for men), which, in turn, has directly affected maternal mortality. Table 12.4 Maternal mortality Maternal mortality rates for the period 2000-2004, based on the survivorship of sisters of survey respondents, Rwanda 2005 Age Maternal deaths Years of exposure Mortality rates (‰) Proportion dying of maternal causes 15-19 2 19,172 0.11 3.3 20-24 25 20,920 1.18 21.1 25-29 25 17,192 1.43 21.7 30-34 37 14,632 2.50 29.3 35-39 18 11,522 1.59 17.3 40-44 19 7,932 2.40 22.1 45-49 5 4,850 1.01 9.7 15-49 130 96,220 1.29 a 19.7 General Fertility Rate (GFR)a 172 Maternal mortality ratio (MMR)b 750 Lifetime risk of maternal death (LTR)c 0.044 a Age adjusted b Per 100,000 births; calculated as maternal mortality rate divided by the general fertility rate c Per woman; calculated as: (1-LTR) = (1-MMR/100 000)TFR, where TFR represents the total fertility rate. For the period 2000-2004, the TFR is estimated to be 5.9 children per woman. Domestic Violence | 175 DOMESTIC VIOLENCE 13 Domestic violence is, essentially, a form of violence against women. It cuts across all national boundaries and social backgrounds. Long considered a private family matter, domestic violence is increasingly recognized as a serious violation of human rights that should be punished. In its Declaration on the Elimination of Violence against Women adopted in 1993, the United Nations General Assembly testified to the international recognition of domestic violence as a form of discrimination against women (United Nations General Assembly, 1993). In addition, it recommended that member states take certain steps to prevent domestic violence and better understand its various aspects. Improvement of domestic violence statistics is included in this panel of recommendations. For this reason, a domestic violence module was included in the 2005 RDHS-III survey. It contains questions designed to assist in estimating the prevalence of domestic violence and describing its characteristics in Rwanda. The results are presented in this chapter. 13.1 METHODOLOGY The domestic violence module was administered in half of the households. In the selected households, only one woman was interviewed, chosen at random (using the Kish grid). Because domestic violence is a sensitive subject, female interviewers were instructed to proceed with a great deal of tact. It was important for them to establish a good rapport with the respondent, draw her into their confidence, and ensure her that her responses would be completely confidential. This climate of trust was crucial to ensuring the validity of the data collected. It was also essential to respect the privacy of the interview in order to ensure the respondent’s safety. Asking a woman questions about domestic violence, especially in households where the perpetrator of the violence may be present during the interview, could lead to additional acts of violence. The 2005 RDHS-III covered three types of domestic violence: physical, sexual, and emotional. Physical violence Two levels of severity are assessed for this type of violence: moderate and severe. • Moderate physical violence was assessed using the following questions: Does/Did your (last) husband/partner ever: - Push you, shake you, or throw something at you? - Slap you or twist your arm? - Strike you with his fist or with something that could hurt you? - Kick you or drag you? • Severe physical violence was assessed using the following questions: Does/Did your (last) husband/partner ever: - Try to strangle or burn you? - Threaten you with a knife, gun, or other type of weapon? - Attack you with a knife, gun, or other type of weapon? 176 | Domestic Violence Sexual violence This type of violence was assessed using the following questions: Does/Did your (last) husband/partner ever: - Physically force you to have sexual intercourse even when you do/did not want to? - Force you to perform other types of sexual acts that you do/did not want to do? Emotional violence This type of violence was assessed using the following question: Does/Did your (last) husband/partner ever: - Say or do something to humiliate you in front of others? - Threaten you or someone close to you with harm? Violence was measured using an abbreviated version of the Conflict Tactics Scale (CTS) developed by Strauss (1990). The CTS scale has been found to be not only effective in measuring domestic violence but also easily adaptable to different situations and cultures. This approach, which consists of asking separately about specific acts, has the advantage of not being affected by varying understandings of what constitutes violence. A woman is asked if she has ever been slapped, not whether she has ever experienced violence, and all women would probably agree on what constitutes a slap. This approach also has the advantage of giving the respondent multiple opportunities to disclose any experience of violence. The RDHS-III survey also gathered data on spousal violence, i.e., violence perpetrated by one spouse against the other, in particular by a husband/partner against his wife/partner. Research on violence suggests that spousal violence is the most common form of domestic violence for adults. The population for which the questions on spousal violence are applicable consists of married or cohabiting women (violence on the part of their husbands/partners) and divorced, separated, or widowed women (violence on the part of their last husband/partner). Women who answered “yes” to any question were also asked about the frequency of this type of violence in the 12 months preceding the survey. In addition to spousal violence, women were asked whether they had experienced any type of physical violence at the hands of anyone other than their current or last husband/partner since the age of 15. The question was formulated as follows: From the time you were 15 years old, has anyone other than your (current/last) husband/partner hit, slapped, kicked, or done anything else to hurt you physically? Women who responded “yes” to this question were asked who had done this and how many times it had happened in the 12 months preceding the survey. In this way, the RDHS-III employed different approaches to measure domestic violence, focusing particularly on spousal violence. Using different approaches, giving a woman several opportunities to disclose acts of violence, and taking precautions to ensure privacy during the interview keep under- reporting of domestic violence to a minimum. However, the possibility of differential underreporting by women in the different subgroups cannot be ruled out. For this reason, caution should be exercised in interpreting the differences observed by background characteristics, although a large proportion undoubt- edly reflect actual differences in the prevalence of violence. Domestic Violence | 177 13.2 DOMESTIC VIOLENCE 13.2.1 Physical Violence Since Age 15 Table 13.1 shows the percentage of women who reported having experienced physical violence since age 15, committed either by their husband/ partner or by someone else, and the percentage of women who experienced physical violence in the 12 months preceding the survey. The results are pre- sented according to background characteristics. The results show that in Rwanda, nearly one third of women (31 percent) have experienced physical violence since age 15, and 19 percent experienced it in the 12 months preceding the survey. This means that 61 percent of Rwandan women who have ever suffered violence have experienced it recently. The prevalence of this violence varies by background characteristic. The proportion of women who reported experiencing acts of violence, whether in the past 12 months or not, are higher among women age 30 to 49 than among the youngest age group. With respect to recent violence, this propor- tion varies from a low of 16 percent at age 15 to 19, to a high of 22 percent among women age 40 to 49. According to marital status, the results show signifi- cantly higher proportions experiencing violence, both past (46 percent) and recent (32 percent), among divorced or separated women. Thirty-seven percent of married or cohabiting women have experienced physical violence since age 15, and 26 percent reported recent violence. The data by residence show a slightly higher prevalence of recent violence in rural areas (20 percent) than in urban areas (17 percent). The proportion of women confronted with recent acts of violence varies by province, from a low of 17 percent in the City of Kigali, to a high of 23 percent in the East province. The prevalence of recent violence decreases as women’s level of education increases: the prevalence among women with no education is twice as high (21 percent) as the prevalence among women with a secondary education or higher (10 percent). There are no major differentials by employment status. Also, the data show no strong relationship between household wealth and physical violence; at most, women living in households in the second wealth quintile can be said to have a relatively higher level of recent physical violence (23 percent), while women in the richest households have a relatively lower level (16 percent). Table 13.1 Experience of beatings or physical mistreatment Percentage of ever-married women who have experienced violence since age 15 and percentage who have experienced violence during the 12 months prior to the survey, by background characteristics, Rwanda 2005 Percentage who have experienced violence: Background characteristic Since age 15 In past 12 months Number Age 15-19 22.9 15.8 957 20-29 30.6 18.4 1,392 30-39 33.3 21.9 946 40-49 37.2 22.4 771 Marital status Never married 20.2 10.5 1,560 In union 36.9 25.5 1,963 Divorced/separated 46.0 32.4 375 Widowed 20.5 1.1 168 Residence Urban 30.1 17.3 682 Rural 30.8 19.8 3,384 Province City of Kigali 27.6 16.8 400 South 31.1 18.9 1,081 West 27.2 18.4 1,015 North 30.9 19.0 727 East 35.5 22.8 842 Education No education 30.8 20.9 760 Primary 31.5 20.2 2,901 Secondary or higher 24.4 10.4 405 Employment status Employed for cash 30.5 19.2 777 Employed, not for cash 33.3 20.2 1,639 Not employed 28.1 18.6 1,645 Wealth quintile Lowest 30.2 20.9 856 Second 33.8 22.8 849 Middle 29.5 16.5 754 Fourth 30.7 20.1 798 Highest 29.0 16.2 809 Total 30.7 19.4 4,066 178 | Domestic Violence Perpetrators of physical violence Table 13.2 shows the distribution of women who reported having experienced acts of physical violence since age 15 according to the perpetrator of the violence. The data are presented according to the marital status. Overall, 47 percent of the time, the perpetrator of the acts of violence is the husband/partner only. This proportion is 80 percent for married women; for 76 percent of divorced or separated women, it is the previous husband/partner. Over one-third of women (34 percent) reported that the acts of violence were committed by someone other than the husband/partner. Finally, 8 percent of women reported that the acts were perpetrated by the husband/partner and others. Altogether, the husband/partner is the perpetrator of the violence 66 percent of the time. Table 13.2 Perpetrators of violence Percent distribution of women reporting any physical violence by perpetrator of the violence, according to current marital status, Rwanda 2005 Perpetrator Marital status Husband only Previous husband only Husband and others Person(s) other than husband Perpetrator unknown Total Number of women Never married na na na 98.7 1.3 100.0 315 In union 80.1 1.7 8.8 9.5 0.0 100.0 724 Divorced/separated 0.0 76.2 12.6 11.2 0.0 100.0 173 Widowed (0.0) (15.9) (22.5) (61.6) (0.0) (100.0) 35 Total 46.5 12.0 7.5 33.8 0.3 100.0 1,247 Note: Figures in parentheses are based on 25-49 unweighted cases. na = Not applicable 13.2.2 Violence during Pregnancy Domestic violence takes a serious toll on women’s physical and mental well-being, no matter what their age or period of life. However, violence during pregnancy exposes women to greater risks, not only those affecting their own health and survival but also the health and survival of their unborn children. To assess the magnitude of this violence, currently pregnant or previously pregnant women were asked if they had experienced physical violence during this pregnancy or these period(s) of their life. If the answer was “yes,” they were asked who had perpetrated these acts of violence. Table 13.3 shows the percentage of women who are or have ever been pregnant who reported having experienced physical violence during their pregnancy; results are broken down by perpetrator of the violence. Overall, 10 percent of women reported having experienced violence while they were pregnant. This proportion does not vary significantly according to the age of the woman. However, the data according to marital status show that divorced or separated women reported having experienced acts of violence during pregnancy more frequently than other women (17 percent, compared with 9 percent for both married and never-married women). There is practically no difference in violence by residence (9 percent for urban areas, 10 percent for rural). Among the provinces, the South has the highest proportion of women who experienced acts of violence during pregnancy (15 percent); the East province has the lowest (8 percent). In addition, women with the highest level of education were less likely to experience violence during a pregnancy than other women (7 percent, compared with 10 percent for women with no education, and 11 percent for women with a primary education). Finally, the results show no differences by employment status (approximately 10 percent, for all three categories). Domestic Violence | 179 When asked about the perpetrator of these acts of violence, 70 percent of women who had experienced violence reported the husband/partner only. Approximately one in five women (19 percent) reported that these acts of violence had been perpetrated by person(s) other than their husband, and 12 percent reported that the perpetrator was the previous husband/partner; among divorced or separated women this proportion is 24 percent. Table 13.3 Violence during pregnancy Percentage of women who have experienced physical violence during pregnancy and the percent distribution of these women by perpetrator of violence, according to background characteristics, Rwanda 2005 Perpetrator Background characteristic Percentage experiencing violence during pregnancy Number of women ever pregnant Husband only Previous husband only Husband and others Person(s) other than husband Total Number of women who experienced violence during pregnancy Age 15-19 (5.9) 27 * * * * * 2 20-29 9.7 913 66.5 7.3 0.0 26.2 100.0 88 30-39 9.8 901 70.2 11.4 1.2 17.2 100.0 88 40-49 11.5 752 71.4 16.2 0.0 12.4 100.0 87 Marital status Never married 8.8 143 * * * * * 13 In union 9.0 1 919 81.6 6.8 0.6 11.0 100.0 173 Divorced/separated 16.5 364 63.8 24.1 0.0 12.2 100.0 60 Widowed 11.6 168 * * * * * 19 Residence Urban 9.3 382 (61.4) (3.1) (0.0) (35.5) (100.0) 36 Rural 10.4 2 211 70.8 12.9 0.5 15.9 100.0 229 Province City of Kigali 8.8 212 * * * * * 19 South 15.2 695 80.1 6.7 0.0 13.2 100.0 105 West 8.6 634 71.6 10.4 0.0 18.0 100.0 54 North 8.5 495 (57.1) (23.8) (1.3) (17.8) (100.0) 42 East 7.9 557 60.3 15.4 1.2 23.1 100.0 44 Education No education 10.1 657 70.2 13.9 1.7 14.2 100.0 66 Primary 10.7 1 700 70.5 11.3 0.0 18.2 100.0 182 Secondary or higher 6.8 236 * * * * * 16 Employment status Employed for cash 10.7 499 65.4 11.7 1.0 21.9 100.0 53 Employed, not for cash 10.1 1 248 67.8 12.9 0.0 19.3 100.0 125 Not employed 10.1 846 74.7 9.5 0.7 15.2 100.0 86 Total 10.2 2 593 69.5 11.6 0.4 18.5 100.0 265 Note: An asterisk indicates that the figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 13.2.3 Marital Control Exercised by the Husband/Partner Spousal violence is frequently associated with certain dominating behaviors used by the husband/partner to control various aspects of a woman’s life. Such behaviors can be precursors to acts of violence. To measure the level of control exercised by husbands/partners over their wives, currently married or ever-married women were asked if their husband/partner had displayed certain of these behaviors. The results are presented in Table 13.4 according to background characteristics. 180 | Domestic Violence Table 13.4 Marital control exercised by husband Percentage of currently married women and divorced or separated women whose current or previous husband displayed specific controlling behaviors, by background characteristics, Rwanda 2005 Percentage of women whose husband: Background characteristic Is jealous if she talks to other men Accuses her of being unfaithful Does not permit meetings with girlfriends Tries to limit contact with family Insists on knowing where she is at all times Doesn’t trust her with money Displays at least 3 of these behaviors Displays none of these behaviors Number of women Age 15-19 (32.8) (13.2) (11.0) (13.2) (39.3) (1.3) (17.0) (47.3) 21 20-29 29.0 7.6 16.8 15.1 41.9 16.2 19.6 41.4 862 30-39 25.6 9.1 13.5 13.9 39.1 18.5 20.2 42.4 827 40-49 22.7 9.2 11.6 10.3 34.2 18.5 15.2 40.3 628 Marital status In union 24.9 7.2 13.5 12.2 39.6 16.5 16.8 47.0 1,963 Only once 23.9 6.6 12.8 12.4 39.9 15.5 16.3 47.4 1,659 More than once 30.2 10.4 17.2 11.1 37.9 21.9 19.3 44.5 303 Divorced/separated 32.7 15.8 17.9 19.5 34.8 22.9 28.1 13.0 375 Number of living children 0 32.6 7.5 16.6 18.8 44.4 15.0 21.7 39.3 144 1-2 29.4 9.4 15.9 14.6 41.9 17.5 21.9 40.9 814 3-4 23.4 8.6 12.8 12.5 34.3 16.9 16.4 42.4 740 5 or more 23.8 7.8 13.1 11.7 38.7 18.7 16.2 41.8 641 Education No education 25.0 9.3 13.5 12.5 34.4 16.5 17.7 41.9 588 Primary 26.1 8.6 14.1 13.9 40.0 18.6 18.8 41.2 1,565 Secondary or higher 30.4 6.2 17.2 12.2 42.7 11.6 19.2 43.1 185 Employment status Employed for cash 28.2 9.2 17.7 18.3 38.6 18.4 21.9 36.5 430 Employed, not for cash 24.4 7.6 11.2 11.4 37.0 15.0 15.3 43.7 1,160 Not employed 27.7 9.8 16.8 13.7 41.6 20.8 21.8 41.0 748 Husband’s education No education 23.4 11.0 13.3 12.6 38.6 16.5 18.8 39.8 666 Primary 25.5 7.7 14.3 14.2 37.1 18.3 18.0 43.4 1,355 Secondary or higher 34.1 5.0 16.9 11.0 47.5 14.1 20.0 38.3 265 Unknown/missing 38.3 17.4 9.0 15.9 40.4 28.0 24.9 30.9 52 Interspousal age difference Wife older than husband 24.4 9.3 18.4 14.2 46.0 21.7 21.3 41.4 102 About the same age (1-2 years difference) 24.3 5.9 11.4 11.2 37.6 13.8 14.6 49.2 622 3-4 years 21.9 4.7 10.9 10.9 37.0 15.5 14.5 50.6 364 5-9 years 23.8 7.6 12.2 11.9 39.1 16.4 16.3 46.2 487 10+ years 30.5 10.7 19.2 14.4 44.3 20.2 21.6 42.4 383 Not currently married 32.7 15.8 17.9 19.5 34.8 22.9 28.1 13.0 375 Total 26.2 8.6 14.2 13.4 38.8 17.5 18.6 41.5 2,338 Note: Figures in parentheses are based on 25-49 unweighted cases. Altogether, 19 percent of women reported that their husband/partner had displayed at least three of the behaviors cited. The proportion is highest among divorced or separated women (28 percent). The data show no significant variations by other background characteristics. At most it can be said that this proportion is somewhat higher among women with no children and women with one or two children (22 percent for both) than among women who have more children (16 percent for women with 3 to 4 children or more). Thirty-nine percent of women reported that their husband/partner insisted on knowing where they were at all times. One-quarter of women (26 percent) reported that their husband/partner was Domestic Violence | 181 jealous when they spoke to other men; 18 percent said he didn’t trust her with money. The other types of controlling behaviors were reported less frequently. 13.3 SPOUSAL VIOLENCE Research on violence suggests that spousal violence is the most common form of domestic violence among adults. It can assume several forms: emotional, physical, sexual, or a combination of these three. This section discusses different aspects of this form of violence. 13.3.1 Prevalence of Spousal Violence As explained earlier, the prevalence of physical, sexual, and emotional violence was measured using a method that describes different acts of violence on a scale from less to more severe. Table 13.5 shows the percentage of currently married and ever-married women who have experienced acts of physical, sexual, and/or emotional violence by their current husband/partner (or the most recent husband/partner, for divorced or separated women). The results show that in Rwanda, 31 percent of women have been confronted with acts of physical violence on the part of their husband/partner: 26 percent of these were moderate acts of violence; 3 percent were severe. Thirteen percent experienced acts of sexual violence, and a total of 34 percent experienced physical or sexual violence. In addition, 12 percent of women reported having experienced emotional violence. Altogether, more than one-third of Rwandan women (35 percent) reported having experienced acts of spousal violence—physical, sexual, or emotional. Four percent of women have experienced all three types of violence. The results by background characteristics show that divorced or separated women have experienced spousal violence most frequently, and in all forms: 36 percent physical violence, 17 percent sexual violence, and 22 percent emotional violence. Overall, 40 percent of divorced or separated women have suffered some form of spousal violence. The prevalence of spousal violence is also higher among women age 40 to 49 (39 percent) and among women in the East province (39 percent). The proportion of women who have experienced spousal violence increases with the number of children, for all forms of violence, ranging from 22 percent among women with no children, to 38 percent among women with at least five children. By level of education, the lowest proportion of spousal violence is found among women with the highest level of education (27 percent, compared with at least 36 percent for the other educational levels). Figure 13.1 shows that more than one-quarter of women who experienced violence reported having had their arm twisted or having been slapped (26 percent). 182 | Domestic Violence Table 13.5 Marital violence Percentage of currently married women and divorced or separated women who have ever experienced physical, sexual, or emotional violence from their husband, by background characteristics, Rwanda 2005 Type of violence Background characteristic Less severe physical violence More severe physical violence Physical violence (severity unknown) Physical violence (total) Sexual violence Physical or sexual violence Emotional violence Physical, sexual, or emotional violence Physical, sexual, and emotional violence Number of women Age 15-19 (26.1) (0.0) (0.0) (26.1) (14.2) (26.1) (11.8) (26.1) (5.8) 21 20-29 25.3 2.9 0.3 28.4 14.3 32.8 12.3 34.4 4.6 862 30-39 24.2 2.8 1.4 28.4 11.7 31.9 12.2 33.5 3.6 827 40-49 30.4 3.7 2.8 36.8 12.5 38.0 11.7 39.2 5.0 628 Marital status In union 26.8 2.2 0.7 29.7 12.1 32.9 10.3 34.4 3.3 1,963 Only once 26.9 2.0 0.5 29.4 12.5 32.6 9.6 34.0 3.3 1,659 More than once 26.1 3.2 2.1 31.5 9.9 34.6 13.7 36.7 3.1 303 Divorced/separated 23.7 7.3 4.7 35.8 17.0 38.8 21.7 40.1 10.0 375 Residence Urban 22.2 4.3 1.4 27.8 19.8 33.0 16.2 34.2 7.5 312 Rural 26.9 2.8 1.4 31.1 11.9 33.9 11.5 35.5 3.9 2,026 Province City of Kigali 21.1 5.6 1.4 28.2 20.1 31.1 18.5 32.7 9.6 169 South 28.2 3.2 1.3 32.8 14.7 35.6 12.3 36.9 5.4 614 West 23.4 2.8 0.8 27.0 10.5 30.3 11.3 31.8 2.9 586 North 28.3 2.6 1.9 32.7 7.9 34.0 7.4 34.9 2.3 447 East 27.2 2.6 1.6 31.4 15.4 36.4 14.6 38.6 4.9 523 Number of living children 0 14.9 1.1 0.6 16.6 11.3 19.5 8.1 22.4 3.2 144 1-2 24.3 3.7 0.7 28.7 14.2 33.1 12.8 34.9 4.9 814 3-4 27.0 2.7 2.4 32.1 12.2 34.9 12.7 35.7 4.1 740 5 or more 30.5 3.0 1.2 34.7 12.6 36.7 11.3 38.3 4.3 641 Education No education 27.1 2.8 1.5 31.4 11.1 33.5 11.8 35.7 4.2 588 Primary 26.9 3.1 1.4 31.5 13.7 34.9 12.0 36.2 4.4 1,565 Secondary or higher 18.0 2.7 0.4 21.1 12.0 25.5 13.5 26.8 5.1 185 Employment status Employed for cash 22.9 4.2 2.3 29.5 15.4 34.7 14.7 36.2 5.6 430 Employed, not for cash 28.1 2.6 1.1 31.8 13.5 34.4 11.8 36.0 4.7 1,160 Not employed 25.4 2.9 1.3 29.5 10.6 32.4 11.1 33.7 3.1 748 Total 26.3 3.0 1.4 30.7 12.9 33.8 12.1 35.3 4.4 2,338 Note: Figures in parentheses are based on 25-49 unweighted cases. Domestic Violence | 183 13.3.2 Frequency of Recent Spousal Violence To determine the frequency of recent spousal violence (physical or sexual), women who reported having experienced physical or sexual violence from their husband/partner were asked the number of times they had experienced such acts in the past 12 months. Eighty percent of the women had experienced acts of spousal violence recently: 38 percent at least three times in the past year, and more than one-third (36 percent) once or twice in the past year (Table 13.6). The frequency of recent spousal violence is highest among divorced or separated women, 67 percent of whom had experienced acts of spousal violence at least three times in the past year. By age, the frequency of spousal violence is highest among young women age 20 to 29: 41 percent at least three times, compared with 34 percent for women age 40 to 49. Fifty percent of women in rural areas experi- enced spousal violence at least three times in the past year, compared with 36 percent in urban areas. By province, the data show that 53 percent of women in the City of Kigali experienced violence at least three times, compared with a low of 31 percent in the North province. It should also be noted that the frequency of spousal violence is higher among women with the highest level of education— 40 percent experienced violence at least three times in the past year compared with 36 percent of women with no education—and among women who work for cash—48 percent, compared with 33 percent of women who are employed but not for cash, and 40 percent of women who are not employed. Figure 13.1 Percentage of Ever-Married Women who Have Ever Experienced Specific Forms of Violence from Their Husbands RDHS 2005 26 16 14 8 2 2 1 12 6 MODERATE PHYSICAL VIOLENCE Slapped/arm twisted Pushed/shaken/thrown Punched Kicked/dragged SEVERE PHYSICAL VIOLENCE Strangled/burned Threatened with weapon Attcked with weapon SEXUAL VIOLENCE Forced to have intercourse Forced to perform other sexual acts 0 5 10 15 20 25 30 35 Percent 184 | Domestic Violence Table 13.6 Frequency of spousal violence Percent distribution of currently married women and divorced or separated women who reported physical or sexual violence by their husband by frequency of any form of such violence in the 12 preceding the survey, according to background characteristics, Rwanda 2005 Frequency of any type of physical or sexual violence in the 12 months preceding the survey Background characteristic 0 times 1-2 times 3-5 times More than 5 times Don’t know/ missing Total Number of women Age 15-19 * * * * * * 5 20-29 17.3 38.0 15.7 24.9 4.2 100.0 283 30-39 18.7 36.6 17.5 20.8 6.5 100.0 264 40-49 25.0 32.7 12.8 21.0 8.5 100.0 239 Marital status In union 22.0 41.9 18.1 13.3 4.7 100.0 645 Divorced/separated 10.9 9.2 3.8 63.0 13.1 100.0 146 Residence Urban 21.1 21.5 17.0 33.1 7.3 100.0 103 Rural 19.8 38.1 15.3 20.8 6.1 100.0 688 Province City of Kigali 17.5 21.5 15.3 37.4 8.3 100.0 52 South 16.5 37.5 17.8 22.5 5.7 100.0 219 West 17.7 37.8 22.3 14.7 7.5 100.0 178 North 25.6 37.8 7.8 23.1 5.8 100.0 152 East 22.1 34.7 12.6 24.9 5.6 100.0 190 Number of living children 0 (18.5) (33.6) (24.1) (20.7) (3.0) (100.0) 28 1-2 17.0 34.3 13.3 29.3 6.1 100.0 269 3-4 18.4 37.8 17.3 18.3 8.3 100.0 258 5 or more 25.1 36.0 14.9 19.3 4.6 100.0 235 Education No education 22.1 36.0 17.7 18.7 5.5 100.0 197 Primary 18.9 36.1 14.9 23.4 6.7 100.0 547 Secondary or higher 23.0 33.1 12.7 27.0 4.1 100.0 47 Employment status Employed for cash 12.4 32.5 22.5 25.3 7.3 100.0 149 Employed, not for cash 23.5 38.2 11.3 21.5 5.5 100.0 399 Not employed 18.7 34.2 18.1 22.3 6.8 100.0 243 Total 19.9 35.9 15.5 22.4 6.2 100.0 791 Note: An asterisk indicates that the figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 13.3.3 Onset of Spousal Violence To determine when spousal violence was first initiated, women who reported having experienced physical or sexual violence on the part of their husband/partner were asked how many years they had been married when the first episode of violence occurred. Table 13.7 shows that for the majority of women, spousal violence began very early in the marriage: 77 percent reported the first episode occurring in the first five years of marriage, including 17 percent who said it had occurred in the first year of marriage. For 15 percent of women, the acts of violence began after 10 years of marriage. The median number of years of marriage before the first episode of violence was 2.9 years. Domestic Violence | 185 Table 13.7 Onset of spousal violence Percent distribution of currently married women and divorced or separated women who have experienced physical or sexual violence by their husband by number of years between marriage and first episode of violence, according to current marital status and duration since first marriage, Rwanda 2005 Years between marriage and first experience of violence Before marriage Less than 1 year 1-2 years 3-5 years 6-9 years 10+ years After divorce Don’t know/ missing Total Median number of years Number of women Marital status Currently in union 0.4 14.4 33.4 28.6 7.0 14.4 0.0 1.9 100.0 3.1 645 Only once 0.4 14.1 32.9 28.9 7.3 14.5 0.0 1.9 100.0 3.1 540 More than once 0.0 15.6 35.9 27.0 5.3 13.8 0.0 2.3 100.0 2.8 105 Divorced/separated 0.0 27.5 25.2 23.9 3.7 17.1 1.1 1.5 100.0 2.0 146 Duration since first marriage1 1-5 years 2.0 24.0 56.1 13.6 0.0 0.7 0.0 3.6 100.0 1.7 117 6-9 years 0.0 15.1 34.1 41.7 7.6 0.0 0.0 1.5 100.0 3.0 103 10+ years 0.0 10.2 24.0 30.4 9.9 24.2 0.0 1.3 100.0 4.0 320 Total 0.3 16.8 31.9 27.7 6.4 14.9 0.2 1.8 100.0 2.9 791 1 For women married only once. 13.4 CONSEQUENCES OF VIOLENCE AND HELP SEEKING All women were asked the following questions, independent of specific acts of violence: As a result of something done to you deliberately by your (last) husband/partner, did you ever: ƒ have bruises and aches? ƒ have an injury or broken bone? ƒ have to visit a doctor or health facility? This sequence of questions has two objectives: first, to assess the physical consequences of violence; second, to provide women who may still be reluctant another opportunity to disclose acts of violence. In some cases, women are more willing to disclose something that happened to them than something their husband/partner did. The results are presented in Table 13.8. Six percent of all currently married or ever-married women reported having had bruises and aches in the past 12 months. In addition, 4 percent reported having had an injury or broken bone during the same period. In contrast, 22 percent of women who reported having experienced physical or sexual violence in the past 12 months said they had had bruises and aches, and 14 percent said they had had an injury or broken bone. Two percent of all ever-married women reported visiting a doctor or health facility to receive care as a result of something done to them by a husband/partner. In contrast, 7 percent of women who reported having experienced acts of physical or sexual violence in the past 12 months reported visiting a doctor or health facility. 186 | Domestic Violence Table 13.8 Physical consequences of spousal violence Percentage of currently married women and divorced or separated women who reported specific physical consequences that resulted from something their husband did to them, by type of violence reported, Rwanda 2005 Had bruises and aches Had injury or broken bone Had to visit health facility Type of violence experienced Ever In the past 12 months Ever In the past 12 months Ever In the past 12 months Number of women Physical violence Ever 26.7 19.6 17.6 12.6 10.5 6.2 685 At least once in past 12 months 28.7 24.6 20.3 16.5 10.6 7.6 507 Sexual violence Ever 24.4 16.0 19.5 14.5 12.4 8.1 302 At least once in past 12 months 24.2 19.3 20.9 18.0 13.3 9.8 244 Physical or sexual violence Ever 24.2 17.8 16.2 11.6 9.7 5.8 760 At least once in past 12 months 26.2 21.7 17.9 14.4 10.6 7.1 599 Total 7.9 5.8 5.3 3.8 3.2 1.9 2,338 Help seeking Women who reported ever having experienced acts of physical or sexual violence since age 15, were asked if they had sought help and from whom they sought it. The results are presented in Table 13.9 according to the perpetrator of the violence and the frequency of the violence in the past 12 months. Among women who had ever experienced physical or sexual violence, 44 percent reported seeking help. Only a small proportion of these women sought help from their immediate family (14 percent). More than two-thirds (69 percent) sought help from other family and friends. In addition, 13 percent of women asked for help from the police, a lawyer, or religious leaders; very few sought assistance from medical personnel (5 percent). The results by perpetrator of violence show that when the husband/partner is not involved in the acts of violence, 24 percent of women seek help from their immediate family, compared with only 11 percent when the husband alone is responsible for the physical or sexual assault. When the husband alone is involved, 71 percent of women seek help from other family and friends. Variations according to frequency of violence are relatively minor and inconsistent; therefore it does not appear that help seeking was influenced by frequency of violence in the past 12 months. Domestic Violence | 187 Table 13.9 Help seeking Among women who reported any physical or sexual violence, percentage who sought help, and among those who sought help, the percentage who sought help from specific sources, by the person who perpetrated the violence and the frequency of violence in the 12 months preceding the survey, Rwanda 2005 Person from whom help was sought Perpetrator of violence/ Frequency of violence Percentage who sought help Number of women who reported any physical or sexual violence Immediate family In-laws/ other family by marriage Other family/ friends Medical personnel Police/ lawyer/ religious figure Other Number of women who sought help Perpetrator of violence Husband only 47.1 580 11.0 32.9 70.8 5.6 12.7 0.2 273 Previous husband only 64.6 149 12.1 33.3 82.7 2.0 14.7 0.0 96 Husband and other(s) 58.0 93 10.8 37.2 70.4 8.2 21.2 3.3 54 Other(s) only 28.4 421 24.3 25.5 55.5 4.0 11.0 4.1 120 Frequency of violence in the past 12 months 0 41.2 472 11.7 37.3 59.8 7.1 11.5 2.8 195 1 time 46.4 268 15.2 21.4 73.5 3.6 10.3 1.5 124 2-3 times 38.2 187 17.5 42.4 73.5 7.6 21.3 0.0 71 4 or more times 48.8 266 14.1 29.5 78.7 1.8 16.9 0.4 130 Don’t know/missing 48.4 54 (16.2) (17.9) (64.3) (1.3) (3.9) (3.4) 26 Total 43.8 1,247 14.0 31.6 69.4 4.8 13.4 1.6 546 Note: Figures in parentheses are based on 25-49 unweighted cases. 13.5 VIOLENCE BY SPOUSAL CHARACTERISTICS AND WOMEN’S STATUS INDICATORS The data presented in Table 13.10 and Figure 13.2 examine the variations in spousal violence according to characteristics of spouses, women’s status indicators, and type of family structure. Since the perpetrators of spousal violence are usually the husbands, it is important to examine the variations in the proportion of women exposed to spousal violence according to the characteristics of husbands. The findings indicate that the husband’s level of education strongly affects the level of spousal violence: the proportion of women who reported having experienced no violence increases with the husband’s level of education, from 62 percent for women whose husbands have no education, to 64 percent for husbands with a primary education, to 77 percent for women whose husbands have a secondary education or higher. This pattern is observed for all types of violence. Results according to interspousal age difference show no major variations. However, the preva- lence of spousal violence is higher among couples in which the woman has more education than her husband/partner (28 percent). Excessive alcohol consumption by the husband/partner appears to be a determining factor in the frequency of recent marital violence. The proportion of women who have experienced no acts of spousal violence drops from a high of 82 percent when the husband/partner drinks but never gets drunk, to 25 percent when the husband is often drunk. The negative effects of alcohol abuse are observed for all types of violence: 5 percent of women whose husband/partner never drinks reported acts of emotional violence, compared with 31 percent of women whose husband gets drunk often. The proportions of women reporting physical or sexual violence is 17 percent for those whose husbands never drink, and 60 percent for those whose husbands get drunk often. 188 | Domestic Violence Table 13.10 Spousal violence, women's status, and spousal characteristics Percentage of currently married women and divorced or separated women who experienced specific types of violence from their husband (ever and in the 12 months preceding the survey), and percentage who have been violent toward their husband, by selected women’s status, spousal, and household characteristics, Rwanda 2005 Physical or sexual violence Emotional violence Violence against husband by respondent Characteristic Ever In past 12 months Ever In past 12 months Experienced no physical, sexual, or emotional violence Ever In past 12 months Number of women Husband’s education No education 36.5 27.3 13.4 10.9 61.9 0.8 0.6 666 Primary 35.1 26.9 11.3 8.3 63.6 1.1 0.6 1,355 Secondary or higher 21.7 14.9 10.8 8.0 76.8 0.0 0.0 265 Interspousal age difference Woman older than husband 31.4 28.4 13.9 10.3 64.9 0.0 0.0 102 About the same age (1-2 years difference) 32.4 24.2 8.9 6.2 66.7 1.1 0.3 622 3-4 years 34.7 25.8 8.6 5.8 64.5 0.3 0.3 364 5-9 years 32.9 24.3 10.4 7.2 65.1 1.1 1.0 487 10+ years 32.0 23.7 12.5 9.4 66.0 1.1 0.5 383 Divorced/separated 38.8 30.3 21.7 19.4 59.9 0.6 0.6 375 Interspousal education difference Husband has more education 33.4 25.0 11.5 8.7 65.2 0.8 0.6 1,003 Wife has more education 36.6 27.8 12.5 9.0 62.2 1.4 0.6 734 Both have equal education 29.6 23.6 11.7 9.7 69.1 0.7 0.7 202 Neither educated 33.3 24.9 12.0 10.4 64.6 0.2 0.2 311 Alcohol consumption by husband Never drinks 24.0 17.4 7.2 5.0 74.7 0.8 0.6 589 Drinks but never gets drunk 16.8 11.8 3.3 2.1 81.9 0.6 0.2 301 Gets drunk sometimes 35.0 26.3 10.2 7.5 63.7 0.4 0.2 923 Gets drunk often 71.6 60.4 38.0 31.0 25.4 2.7 1.8 355 Woman can refuse sex with husband Yes for all reasons 33.6 25.6 11.2 8.6 65.5 0.9 0.5 1,383 No for one or more reasons 34.2 25.6 13.3 10.1 63.5 0.7 0.5 955 Number of households decisions respondent participates in No decisions 40.0 32.6 17.5 13.9 56.4 0.6 0.6 165 1-2 decisions 35.0 26.3 12.2 10.2 63.4 1.1 0.9 585 3-4 decisions 32.7 24.6 11.5 8.4 66.0 0.8 0.4 1,588 Index of marital harmony Least harmonious 44.1 34.8 23.0 19.4 53.7 1.2 1.0 720 Middle 46.5 36.8 14.2 10.6 51.9 1.0 0.3 404 Most harmonious 23.5 16.5 4.9 2.8 75.5 0.6 0.3 1,214 Index of marital control exercised by husband 0 point (least control) 22.7 16.3 5.0 3.4 76.4 0.4 0.3 971 1-2 points 37.2 27.7 10.2 7.1 61.3 0.8 0.4 770 3-4 points 53.8 45.7 28.8 24.0 42.7 1.2 1.2 319 5-6 points (most control) 40.4 29.4 22.6 18.8 58.3 2.0 0.8 279 Family structure Nuclear 34.9 26.4 11.2 8.3 63.8 0.9 0.6 1,899 Non-nuclear 29.2 22.4 15.9 13.3 68.7 0.6 0.3 439 Total1 33.8 25.6 12.1 9.2 64.7 0.9 0.5 2,338 1 The total includes 52 women for whom the husband’s education is not known, 5 women for whom the interspousal age difference is not known, 88 women for whom the interspousal education difference is not known, and 170 women for whom husband’s alcohol consumption is not known Domestic Violence | 189 The results are also presented according to two indicators of women’s status. The data show no variation in the prevalence of spousal violence according to whether or not the wife believes that a woman can refuse sex with her husband for certain reasons. However, there seems to be a correlation between the number of household decisions made by the woman and the prevalence of spousal violence: the prevalence of physical or sexual violence drops from 33 percent when the woman participates in no decisionmaking, to 26 percent when she is involved in 1 or 2 household decisions, to 25 percent when she is involved in 3 or 4 household decisions. Table 13.10 also presents results according to marital harmony. The marital harmony index was developed on the basis of responses to the following questions: In your relationship with your (last) husband/partner do/did the following occur frequently, sometimes, or never? • Does/Did he spend his free time with you? • Does/Did he consult with you on issues affecting the household? • Is/was he affectionate with you? • Does/did he respect you and take your desires into account? The marital harmony index is based on how many of the above behaviors occurred frequently. If a woman reported that none of the behaviors occurred frequently, the marriage is considered inharmonious. If a woman reported that three or four of the behaviors occurred frequently, the marriage is considered very harmonious. The results show that the more harmonious the marriage, the lower the frequency of spousal violence. The prevalence of recent physical or sexual violence drops from 35 percent for the least harmonious marriages to 17 percent for the most harmonious marriages. However, the overall prevalence of spousal violence is still high even in the most harmonious households, where one in four women reported experiencing spousal violence at some time. Figure 13.2 Prevalence of Spousal Violence, by Level of Education of Woman and Her Spouse and Alcohol Consumption of Spouse 34 35 26 27 27 15 17 12 26 60 WOMAN'S EDUCATION None Primary Secondary or higher SPOUSE'S EDUCATION None Primary Secondary or higher SPOUSE'S ALCOHOL CONSUMPTION Never drinks Drinks, but never gets drunk Gets drunk sometimes Gets drunk often 0 10 20 30 40 50 60 Percent Note: Physical or sexual violence occurring in the 12 months preceding the survey. RDHS 2005 190 | Domestic Violence The results also show the relationship between controlling behaviors and the frequency of spousal violence, which ranges from 16 percent for the lowest levels of marital control to 46 percent for a control index of 3 to 4 points. Finally, the last characteristic presented in Table 13.10 concerns family structure: nuclear or non-nuclear. It appears that the frequency of spousal violence (physical or sexual) is a little lower in non-nuclear families (22 percent) than in nuclear families (26 percent). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 191 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR 14 HIV infection is a major public health concern in Rwanda, where it is a primary cause of mortality with negative social and economic consequences impacting everyone in the country. In 2001, the Rwandan government created the TRAC (Treatment and Research AIDS Center) and the CNLS (Commission National de Lutte contre le SIDA, or National AIDS Commission) to focus efforts to combat the disease. Current strategies in the fight against AIDS in Rwanda are found in the Plan stratégique national de lutte contre le SIDA au Rwanda (Rwandan National AIDS Plan). They include expansion of the epidemiological surveillance system for HIV/AIDS and STIs established in 2001 to focus on making information available to everyone involved in evaluating trends in the disease, predicting the magnitude of the epidemic, and assessing the impact of various AIDS interventions. During its first ten years, the HIV epidemiological surveillance system relied on, as its primary information source, data on HIV prevalence among pregnant women seeking care through a network of sentinel ANC and AIDS notification sites. However, the system is ill equipped to reflect the epidemic’s diversity. It is limited, in particular, with respect to qualitative data. In “generalized epidemic” countries such as Rwanda, the surveillance system must monitor HIV infection and high-risk behaviors both in the general population and specific subgroups. The effectiveness of prevention measures depends not only on knowing the pace and magnitude of the spread of the disease, but also on identifying problem behaviors, attitudes, and sociocultural factors impacting the disease. For this reason, the 2005 RDHS-III devoted a large part of its efforts to gathering data on the HIV/AIDS pandemic and other STIs. The aim of this chapter is to determine STI and HIV/AIDS-related knowledge, perceptions, attitudes, and behaviors at the national and provincial levels and for certain subgroups of the population. In Rwanda, as in most African countries, the principle mode of transmission of AIDS is through sexual contact. Most of the men and women interviewed for the RDHS-III survey (men age 15 to 59 and women age 15 to 49) are sexually active, making them primary targets of the national Information, Education, and Communication (IEC) plans launched by the CNLS. To assess the impact of Rwanda’s anti-AIDS program, data were collected on the level of knowledge of the means of transmission and prevention of HIV infection, stigmatization of those suffering from the disease, and risk factors, particularly sexual behavior. The information gathered is essential for adjusting current programs and setting up new AIDS information, education and communication campaigns. Survey results cover these main areas: • Knowledge of the existence of HIV/AIDS, its modes of transmission, and ways to avoid it; and knowledge and rejection of misconceptions concerning prevention of the infection. • Knowledge of mother-to-child transmission. • Acceptance of people living with HIV/AIDS. • Attitudes of men and women toward negotiating safer sex with a spouse. • Higher-risk sexual intercourse and condom use during the most recent higher-risk sexual intercourse. • Age at first sexual intercourse for young people age 15-24. • Higher-risk sexual intercourse and condom use during the most recent higher-risk sexual intercourse among young people age 15-24. • Premarital sex and condom use among young people age 15-24. 192 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior • Knowledge of STIs and their symptoms. • Treatment sought for STIs. • Knowledge of injections and syringes. In addition, the RDHS-III conducted HIV testing across the entire population covered by the survey (see Chapter 15). 14.1 KNOWLEDGE, OPINIONS, AND ATTITUDES How much a population knows about a disease influences attitudes and behaviors with respect to that disease. For this reason, the 2005 RDHS-III collected data to determine the level of knowledge of HIV/AIDS in the population. Table 14.1 shows that knowledge of HIV/AIDS is almost universal in Rwanda. The proportion of men and women who have heard of HIV/AIDS has remained relatively stable since the 2000 RDHS-II survey. In addition, the level of knowledge is uniform; nearly every re- spondent reported having heard of HIV/AIDS, regardless of background characteristics. 14.1.1 Knowledge of HIV Transmission and Prevention Methods To effectively fight the AIDS virus, the population must be aware of ways to prevent its spread. Table 14.2 shows that 80 percent of women and 90 percent of men know that the risk of contracting HIV/AIDS can be limited by using condoms. In addition, when asked if they could avoid contracting HIV/AIDS by limiting sexual intercourse to one uninfected partner, 87 percent of women and 87 percent of men answered affirmatively. In all, 73 percent of women and 80 percent of men recognized both of these methods of prevention. It should also be noted that 82 percent of women and 88 percent of men also recognized abstaining from sexual inter- course as a means of preventing HIV/AIDS. The data by age show that knowledge of both methods of prevention is lowest in the 15-19 age group, for both men (75 percent) and women (68 percent). Knowledge of both methods of prevention increases with the level of education, from 78 percent for men with no education and 67 percent for women, to 80 percent for men with a primary education, and 73 percent for women, to 81 percent for men with a secondary education or higher, 79 percent for women. Table 14.1 Knowledge of AIDS Percentage of women and men age 15-49 who have heard of AIDS by background characteristics, Rwanda 2005 Women Men Background characteristic Has heard of AIDS Number of women Has heard of AIDS Number of men Age 15-24 99.8 4,938 99.9 2,048 15-19 99.7 2,585 99.8 1,102 20-24 100.0 2,354 100.0 946 25-29 100.0 1,738 100.0 632 30-39 99.9 2,600 100.0 951 40-49 99.9 2,045 100.0 783 Marital status Never married 99.8 4,263 99.9 2,191 Ever had sex 99.8 758 99.9 833 Never had sex 99.8 3,505 99.9 1,358 In union 99.9 5,510 100.0 2,126 Divorced/separated/ widowed 99.9 1,548 100.0 96 Residence Urban 99.9 1,921 100.0 784 Rural 99.9 9,400 99.9 3,629 Province Kigali city 99.8 1,127 100.0 495 South 100.0 2,958 99.9 1,139 West 99.9 2,824 99.9 1,065 North 99.8 2,063 99.9 777 East 99.9 2,348 100.0 937 Education No education 99.8 2,193 100.0 558 Primary 99.9 8,044 99.9 3,293 Secondary or higher 100.0 1,084 100.0 561 Wealth quintile Lowest 99.8 2,421 99.9 799 Second 99.9 2,325 100.0 794 Middle 99.9 2,099 99.8 892 Fourth 99.9 2,133 100.0 900 Highest 99.9 2,342 100.0 1,028 Total 99.9 11,321 99.9 4,413 HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 193 Table 14.2 Knowledge of HIV prevention methods Percentage of women and men age 15-49 who, in response to a prompted question, say that people can reduce the risk of getting the AIDS virus by using condoms every time they have sexual intercourse, by having one sex partner who is not infected and has no other partners, and by abstaining from sexual intercourse, by background characteristics, Rwanda 2005 Women Men Background characteristic Using condoms Limiting sexual intercourse to one uninfected partner Using condoms, and limiting sexual intercourse to one uninfected partner Abstaining from sexual intercourse Number of women Using condoms Limiting sexual intercourse to one uninfected partner Using condoms, and limiting sexual intercourse to one uninfected partner Abstaining from sexual intercourse Number of men Age 15-24 79.5 85.2 71.4 80.6 4,938 88.4 84.8 77.6 87.6 2,048 15-19 76.9 83.1 68.1 79.8 2,585 86.9 81.5 74.8 86.2 1,102 20-24 82.4 87.5 75.1 81.5 2,354 90.2 88.7 80.8 89.2 946 25-29 81.6 88.6 74.8 81.4 1,738 91.0 87.4 80.6 88.0 632 30-39 82.8 88.5 75.8 83.1 2,600 91.9 89.3 83.1 87.6 951 40-49 76.3 86.8 69.8 83.1 2,045 89.1 89.6 81.5 89.8 783 Marital status Never married 78.6 84.1 69.9 81.5 4,263 88.8 83.9 77.2 87.5 2,191 Ever had sex 86.5 88.5 79.5 84.4 758 93.0 87.9 82.9 89.4 833 Never had sex 76.8 83.2 67.8 80.8 3,505 86.3 81.5 73.7 86.2 1,358 In union 81.8 89.0 75.3 81.5 5,510 90.3 90.2 82.5 88.9 2,126 Divorced/separated/ widowed 77.6 85.9 70.7 83.6 1,548 93.8 85.8 83.4 83.9 96 Residence Urban 84.3 88.4 76.4 82.0 1,921 88.6 83.1 75.4 84.8 784 Rural 79.1 86.4 71.9 81.7 9,400 89.9 87.8 80.9 88.7 3,629 Province Kigali city 84.2 87.0 75.8 84.8 1,127 87.7 80.7 72.1 82.3 495 South 83.4 90.6 78.2 87.1 2,958 88.9 90.6 83.3 91.4 1,139 West 71.2 80.9 60.9 76.7 2,824 84.7 80.9 71.0 84.3 1,065 North 76.8 86.4 70.5 73.4 2,063 95.5 93.6 89.9 93.4 777 East 87.0 89.1 80.1 87.1 2,348 92.4 87.3 81.7 86.8 937 Education No education 74.2 84.7 66.9 81.7 2,193 85.7 87.4 77.9 88.5 558 Primary 80.6 87.1 73.3 82.0 8,044 90.0 86.9 80.0 88.4 3,293 Secondary or higher 87.2 88.0 79.4 80.1 1,084 91.5 87.1 81.0 85.3 561 Wealth quintile Lowest 75.7 87.0 69.7 81.0 2,421 88.1 89.8 80.4 88.7 799 Second 80.1 87.2 74.1 81.6 2,325 91.3 87.9 82.6 90.4 794 Middle 80.7 85.4 72.4 82.1 2,099 89.8 87.4 80.4 88.0 892 Fourth 80.0 86.4 71.6 81.8 2,133 90.2 86.6 80.5 88.7 900 Highest 83.6 87.6 75.4 82.4 2,342 89.1 84.2 76.4 85.2 1,028 Total 80.0 86.8 72.7 81.8 11,321 89.7 87.0 79.9 88.1 4,413 Knowledge of both methods of prevention is lower among women in rural areas (72 percent) than among women in urban areas (76 percent). However, the situation is the reverse for men: 81 percent of men in rural areas know about both methods, compared with only 75 percent of men in urban areas. By province, the results show the West province has the lowest proportion of men (71 percent) and women (61 percent) who had heard of both methods of prevention. By marital status, never-married men, and women who have never had sexual intercourse, were the least likely to have heard of these two ways of avoiding HIV/AIDS infection (68 percent for women; 74 percent for men). Misconceptions about HIV infection and AIDS influences attitudes and behaviors toward the disease. During the survey, a series of questions was asked of respondents to assess their level of correct 194 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior knowledge concerning the transmission and prevention of the AIDS virus. The results are presented in Table 14.3.1 for women and in Table 14.3.2 for men. More than four in five women (84 percent) know that a person who looks healthy can have the AIDS virus. In addition, 81 percent know that AIDS cannot be transmitted by mosquito bites. More than nine in ten women (92 percent) know that AIDS cannot be transmitted by supernatural means, and more than 89 percent of women recognized that a person cannot become infected by sharing food with a person who has AIDS. Overall, a little more than two in three women (68 percent) reject the two most common misconceptions concerning AIDS transmission (i.e., that a person cannot contract AIDS through mosquito bites or by sharing a meal with someone who is infected), and know that a person who looks healthy can have the AIDS virus. The second-to-last column of Table 14.3.1 shows the percentage of women who have what is considered “comprehensive” knowledge of HIV/AIDS: they know that using condoms and limiting sexual intercourse to one faithful uninfected partner can reduce the chance of contracting AIDS, they reject the two most common misconceptions about AIDS transmission, and they know that a healthy- looking person can have the AIDS virus. A little over half of the women (54 percent) and men (58 percent) surveyed have a comprehensive knowledge of AIDS. The proportion of women with comprehensive AIDS knowledge varies according to background characteristics. By age, the percentage is lowest among women age 15 to 19 (45 percent). The proportion increases with the level of education, from a low of 42 percent among women with no education, to 73 percent among those with a secondary education or higher. By residence, the proportion of women with comprehensive knowledge is higher in urban areas (64 percent) than in rural areas (51 percent). There are differences according to marital status: never- married women who have never had sex (49 percent) and divorced, separated, or widowed women (52 percent) have the lowest levels of comprehensive knowledge. Never-married women who have had sex (58 percent) and married women (57 percent) are the best informed. With respect to household wealth, the results show that less than half of the women in the poorest quintile (46 percent) have comprehensive knowledge of AIDS; the proportion fluctuates around 53 percent in the three middle quintiles, and reaches a high of 63 percent in the richest quintile. By province, the City of Kigali has the highest proportion of women with comprehensive knowledge (66 percent), with the West province having the lowest (40 percent). Table 14.3.2 shows the same results for men. Overall, men are more likely than women to have correct knowledge of HIV/AIDS transmission: more than nine in ten men (92 percent) know that a healthy-looking person can have the AIDS virus. In addition, 78 percent know that AIDS cannot be transmitted by mosquito bites, 92 percent reject the misconception that HIV/AIDS can be transmitted by supernatural means, and 92 percent reject the notion that AIDS can be transmitted by sharing food with an infected person. Overall, 70 percent of men, compared with 68 percent of women, reject the two most common misconceptions about AIDS transmission and know that a healthy-looking person can have the AIDS virus. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 195 Table 14.3.1 Comprehensive knowledge about AIDS: women Percentage of women age 15-49 who say that a healthy-looking person can have the AIDS virus and who, in response to prompted questions, correctly reject local misconceptions about AIDS transmission or prevention, and the percentage with a comprehensive knowledge about AIDS by background characteristics, Rwanda 2005 Percentage of women who say that: Background characteristic A healthy- looking person can have the AIDS virus AIDS cannot be transmitted by mosquito bites AIDS cannot be transmitted by supernatural means A person cannot become infected by sharing food with a person who has AIDS Percentage who say that a healthy-looking person can have the AIDS virus and who reject the two most common local misconceptions1 Percentage with a compre- hensive knowledge about AIDS2 Number of women Age 15-24 81.1 81.0 90.9 88.2 65.2 50.9 4,938 15-19 75.8 80.4 89.2 86.9 60.2 45.3 2,585 20-24 86.9 81.6 92.8 89.7 70.6 57.1 2,354 25-29 86.1 81.5 93.6 91.5 69.5 55.7 1,738 30-39 87.4 83.3 93.5 91.1 72.6 59.3 2,600 40-49 83.8 78.9 91.1 87.9 66.0 50.9 2,045 Marital status Never married 80.9 82.2 90.9 89.0 65.9 50.2 4,263 Ever had sex 86.9 82.7 93.1 92.3 70.6 57.8 758 Never had sex 79.6 82.1 90.4 88.3 64.9 48.6 3,505 In union 85.7 80.9 92.8 90.0 69.4 56.7 5,510 Divorced/separated/ widowed 85.2 79.5 92.0 87.9 66.6 51.7 1,548 Residence Urban 94.0 88.3 94.7 95.3 81.7 64.2 1,921 Rural 81.7 79.8 91.4 88.1 64.8 51.4 9,400 Province Kigali city 94.1 90.4 95.4 95.7 84.0 65.9 1,127 South 86.1 86.8 94.4 92.2 73.8 60.0 2,958 West 76.4 75.5 90.2 83.9 57.3 39.6 2,824 North 83.4 77.6 88.6 88.3 65.5 53.4 2,063 East 85.3 79.7 92.4 90.2 66.5 56.5 2,348 Education No education 73.4 72.4 87.5 81.1 54.0 41.6 2,193 Primary 84.8 82.0 92.7 90.5 68.4 54.2 8,044 Secondary or higher 97.9 93.0 95.8 97.4 89.8 73.0 1,084 Wealth quintile Lowest 76.3 76.1 89.6 85.0 57.6 45.7 2,421 Second 81.8 80.3 91.6 89.5 65.5 53.6 2,325 Middle 83.4 80.3 92.2 87.5 66.7 52.3 2,099 Fourth 85.2 81.2 91.6 90.1 68.5 53.5 2,133 Highest 92.7 88.2 94.9 94.6 80.4 62.8 2,342 Total 83.8 81.2 92.0 89.3 67.7 53.6 11,321 1 Two most common local misconceptions: transmission by mosquito bites and sharing food with an infected person. 2 Comprehensive knowledge means knowing that use of condoms and having just one uninfected faithful partner can reduce the chances of getting the AIDS virus, knowing that a healthy-looking person can have the AIDS virus, and rejecting the two most common local misconceptions about AIDS transmission and prevention. 196 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.3.2 Comprehensive knowledge about AIDS: men Percentage of men age 15-49 who say that a healthy-looking person can have the AIDS virus and who, in response to prompted questions, correctly reject local misconceptions about AIDS transmission or prevention, and the percentage with a comprehensive knowledge about AIDS by background characteristics, Rwanda 2005 Percentage of men who say that: Background characteristic A healthy- looking person can have the AIDS virus AIDS cannot be transmitted by mosquito bites AIDS cannot be transmitted by supernatural means A person cannot become infected by sharing food with a person who has AIDS Percentage who say that a healthy-looking person can have the AIDS virus and who reject the two most common local misconceptions1 Percentage with a compre- hensive knowledge about AIDS2 Number of men Age 15-24 87.2 76.4 91.1 91.1 65.5 53.6 2,048 15-19 83.2 74.7 88.2 90.1 60.7 49.0 1,102 20-24 92.0 78.3 94.4 92.3 71.1 59.0 946 25-29 95.2 79.3 94.0 93.4 73.4 60.5 632 30-39 96.8 79.2 92.7 93.0 75.2 63.3 951 40-49 94.0 77.3 92.3 90.4 69.7 58.0 783 Marital status Never married 88.3 78.2 91.1 91.9 67.9 54.8 2,191 Ever had sex 94.7 80.6 95.2 95.6 75.7 63.9 833 Never had sex 84.4 76.7 88.6 89.7 63.1 49.3 1,358 In union 95.0 77.0 93.0 91.7 71.3 60.1 2,126 Divorced/separated/ widowed 94.5 74.3 91.7 86.2 64.4 57.6 96 Residence Urban 96.2 87.0 95.4 96.0 81.8 63.0 784 Rural 90.7 75.5 91.3 90.8 66.8 56.3 3,629 Province Kigali city 95.7 89.3 95.9 94.4 82.4 60.1 495 South 95.5 81.8 93.8 93.3 76.9 66.6 1,139 West 88.0 74.1 88.7 88.0 62.7 47.2 1,065 North 88.7 71.8 89.0 91.7 64.0 59.6 777 East 91.5 74.9 94.2 92.6 65.8 54.8 937 Education No education 87.7 64.3 84.9 82.3 52.4 41.6 558 Primary 91.1 77.2 92.3 92.2 68.8 57.4 3,293 Secondary or higher 98.6 92.9 97.5 98.3 90.3 73.4 561 Wealth quintile Lowest 88.1 68.8 88.9 86.8 59.4 50.8 799 Second 90.8 76.4 90.5 91.7 66.9 56.8 794 Middle 90.0 76.4 91.5 90.5 67.2 57.4 892 Fourth 92.1 77.8 92.3 92.8 70.2 58.2 900 Highest 96.1 86.0 95.9 95.6 80.6 62.6 1,028 Total 91.7 77.5 92.0 91.7 69.5 57.5 4,413 1 Two most common local misconceptions: transmission by mosquito bites and sharing food with an infected person. 2 Comprehensive knowledge means knowing that use of condoms and having just one uninfected faithful partner can reduce the chances of getting the AIDS virus, knowing that a healthy-looking person can have the AIDS virus, and rejecting the two most common local misconceptions about AIDS transmission and prevention. With respect to comprehensive knowledge, men are better informed than women: 58 percent of men, compared with 54 percent of women, have a comprehensive knowledge of AIDS. Never-married men who have had sex, along with married men, are the best informed (64 percent and 60 percent, respectively), but the data vary considerably by level of education, residence, and household wealth. Only 42 percent of men with no education and 57 percent of men with a primary education have a comprehensive knowledge of HIV/AIDS, compared with 73 percent of men with a secondary education or higher. In rural areas, 56 percent of men have a comprehensive knowledge of AIDS, compared with 63 percent in urban areas. By household wealth, comprehensive knowledge ranges from 51 percent for men in the poorest households, to 63 percent for those in the richest households. As with women, the West province has the lowest proportion of men with comprehensive knowledge about AIDS (47 percent). The South province has the highest proportion (67 percent). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 197 During the survey, all respondents were asked whether they knew that the virus that causes AIDS can be transmitted from mother to child by breastfeeding and that the risks of maternal transmission can be reduced if the mother takes special drugs during pregnancy. The results are presented in Table 14.4. Table 14.4 Knowledge of prevention of mother-to-child transmission of HIV Percentage of women and men age 15-49 who know that HIV can be transmitted from mother to child by breastfeeding and that the risk of mother- to-child transmission (MTCT) of HIV can be reduced by the mother taking special drugs during pregnancy, by background characteristics, Rwanda 2005 Women who know that: Men who know that: Background characteristic HIV can be transmitted by breast- feeding Risk of MTCT can be reduced by mother taking special drugs during pregnancy HIV can be transmitted by breastfeeding and risk of MTCT can be reduced by mother taking special drugs during pregnancy Number of women HIV can be transmitted by breast- feeding Risk of MTCT can be reduced by mother taking special drugs during pregnancy HIV can be transmitted by breastfeeding and risk of MTCT can be reduced by mother taking special drugs during pregnancy Number of men Age 15-24 76.3 68.1 58.3 4,938 80.6 77.0 65.8 2,048 15-19 71.5 61.6 52.1 2,585 77.5 72.4 61.0 1,102 20-24 81.5 75.2 65.2 2,354 84.2 82.3 71.4 946 25-29 82.3 79.0 69.6 1,738 84.3 87.0 74.8 632 30-39 82.2 78.3 68.7 2,600 82.9 83.1 71.3 951 40-49 82.7 73.0 65.4 2,045 84.0 80.3 69.7 783 Marital status Never married 74.4 66.9 56.8 4,263 81.1 78.0 67.0 2,191 Ever had sex 77.0 75.2 65.1 758 84.1 87.8 75.6 833 Never had sex 73.8 65.1 55.0 3,505 79.2 72.0 61.7 1,358 In union 83.1 77.9 68.8 5,510 83.6 82.8 71.3 2,126 Divorced/separated/ widowed 82.3 72.2 64.4 1,548 78.2 77.6 62.8 96 Currently pregnant Yes 85.4 79.8 71.6 901 na na na na No/not sure 79.2 72.4 63.0 10,420 na na na na Residence Urban 86.5 87.3 78.9 1,921 85.8 88.3 78.3 784 Rural 78.3 70.1 60.6 9,400 81.4 78.6 67.0 3,629 Province Kigali city 86.6 84.1 77.7 1,127 87.5 87.2 78.8 495 South 80.5 77.0 65.0 2,958 86.6 81.8 73.0 1,139 West 79.0 65.3 57.5 2,824 79.8 70.2 59.8 1,065 North 77.4 70.2 61.4 2,063 73.0 84.0 65.0 777 East 78.4 74.4 64.7 2,348 84.5 83.2 72.6 937 Education No education 76.7 61.4 54.3 2,193 80.0 66.9 58.1 558 Primary 79.6 73.5 63.7 8,044 82.2 80.4 68.9 3,293 Secondary or higher 87.2 92.7 82.4 1,084 84.8 93.3 80.3 561 Wealth quintile Lowest 77.1 61.1 53.3 2,421 80.7 71.9 61.3 799 Second 78.4 72.7 62.2 2,325 82.1 79.4 68.3 794 Middle 79.0 71.7 62.0 2,099 81.3 77.9 67.0 892 Fourth 80.0 75.0 65.4 2,133 79.8 82.4 68.5 900 Highest 84.1 85.0 75.9 2,342 86.4 87.7 77.6 1,028 Total 79.7 73.0 63.7 11,321 82.2 80.3 69.0 4,413 na = Not applicable 198 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior The data show no major differences in the proportion of men and women who reported knowing that HIV can be transmitted from mother to child by breastfeeding (82 percent of men; 80 percent of women). However, men are more likely than women to know that the risk of mother-to-child transmission can be reduced if the mother takes special drugs during pregnancy (80 percent of men; 73 percent of women). Overall, 69 percent of men and 64 percent of women reported knowing both of these aspects of mother-to-child transmission of HIV. The data vary by background characteristic. The women most likely to be aware of this information are those age 25 to 39 (at least 69 percent); married women (69 percent); divorced, separated, or widowed women (64 percent); and women who were pregnant at the time of the survey (72 percent). The proportion of women who are aware of this information is higher in urban areas (79 percent) than in rural areas (61 percent). It is also highest among women with higher levels of education (82 percent), women in the richest wealth quintile (76 percent), and women living in the City of Kigali (78 percent). The data for men follow the same patterns with respect to background characteristics. 14.1.2 Stigmatization The behavior or attitudes a person would adopt toward someone living with HIV/AIDS in certain situations reveal his or her beliefs about the risk of HIV transmission, beliefs which, in daily life, can translate into stigmatization of infected people. During the 2005 RDHS-III, respondents were asked whether they would be willing to take on the care of a relative with HIV/AIDS in their own household, whether they would buy fresh vegetables from a shopkeeper who had HIV/AIDS, whether they believed that a female teacher living with HIV/AIDS should be allowed to continue teaching and, finally, whether they would want to keep secret that a family member had been infected with HIV/AIDS. The results are presented in Table 14.5.1 for women, and in Table 14.5.2, for men. Forty-six percent of women expressed accepting attitudes in all four of the situations presented. Those who were most accepting toward people living with HIV/AIDS in the specific situations presented are women age 25 to 29 (51 percent), never-married women who have had sex (50 percent), women in urban areas (63 percent), women in Kigali City (62 percent), women with a secondary education or higher (69 percent), and women in the richest wealth quintile (61 percent). The proportion of men who expressed accepting attitudes in all four situations is higher than that of women (51 percent, compared with 46 percent). Like women, men age 25 to 29 (56 percent) and men with higher educations (61 percent) were more likely to express accepting attitudes. However, unlike women, the most accepting attitudes are found not among never-married men who had had sex, but among married men. Also unlike women, tolerance in all four situations was most frequently expressed by men in rural (53 percent) rather than urban areas (42 percent). By province, the highest percentage of men expressing acceptance in all four situations is the South province (64 percent), not the City of Kigali, as was the case for women. Finally, results by wealth quintile show the highest level of acceptance is in the fourth quintile for men (55 percent), not the richest quintile, as was the case for women. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 199 Table 14.5.1 Accepting attitudes toward those living with HIV/AIDS: women Among women age 15-49 who have heard of AIDS, percentage expressing specific accepting attitudes toward people with AIDS, by background characteristics, Rwanda 2005 Percentage of women who: Background characteristic Are willing to care for a family member with the AIDS virus in the respondent’s home Would buy fresh vegetables from shopkeeper who has the AIDS virus Say that a female teacher with the AIDS virus and is not sick should be allowed to continue teaching Would not want to keep secret that a family member got infected with the AIDS virus Percentage expressing accepting attitudes on all four indicators Number of women who have heard of AIDS Age 15-24 91.9 66.5 72.6 76.7 43.6 4,929 15-19 90.3 62.3 69.1 74.5 39.1 2,577 20-24 93.7 71.0 76.4 79.2 48.5 2,353 25-29 94.8 73.7 77.0 77.4 51.0 1,738 30-39 94.8 71.9 76.9 76.8 48.4 2,597 40-49 94.5 66.7 73.2 78.3 44.9 2,044 Marital status Never married 92.6 67.8 74.1 76.3 45.1 4,255 Ever had sex 94.7 73.1 79.5 76.3 50.0 757 Never had sex 92.1 66.6 72.9 76.3 44.0 3,498 In union 94.2 69.3 74.6 78.0 46.6 5,506 Divorced/separated/ widowed 93.4 70.2 74.6 76.4 46.9 1,546 Residence Urban 97.9 86.5 88.9 77.8 63.1 1,919 Rural 92.6 65.3 71.4 77.0 42.6 9,389 Province Kigali city 97.1 86.9 88.5 75.8 61.8 1,125 South 96.7 74.9 81.8 81.0 55.8 2,958 West 89.3 61.2 68.1 74.0 35.3 2,821 North 93.3 64.4 71.8 75.7 40.9 2,060 East 93.0 65.8 68.1 77.9 43.8 2,344 Education No education 88.9 55.0 62.2 77.2 33.5 2,189 Primary 94.1 69.4 75.3 76.8 46.4 8,035 Secondary or higher 98.3 92.9 92.1 79.1 69.1 1,084 Wealth quintile Lowest 91.4 57.8 65.9 77.8 37.6 2,417 Second 93.4 67.8 74.0 76.4 44.4 2,324 Middle 92.0 65.8 71.3 76.4 42.1 2,097 Fourth 93.5 68.7 73.8 76.9 44.9 2,131 Highest 97.1 84.3 86.8 78.1 61.2 2,340 Total 93.5 68.9 74.4 77.1 46.1 11,308 200 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.5.2 Accepting attitudes toward those living with HIV/AIDS: men Among men age 15-49 who have heard of AIDS, percentage expressing specific accepting attitudes toward people with AIDS, by background characteristics, Rwanda 2005 Percentage of men who: Background characteristic Are willing to care for a family member with the AIDS virus in the respondent’s home Would buy fresh vegetables from shopkeeper who has the AIDS virus Say that a female teacher with the AIDS virus and is not sick should be allowed to continue teaching Would not want to keep secret that a family member got infected with the AIDS virus Percentage expressing accepting attitudes on all four indicators Number of men who have heard of AIDS Age 15-24 93.5 75.9 75.0 72.2 46.7 2,045 15-19 91.5 70.1 69.5 71.0 42.2 1,099 20-24 95.8 82.6 81.3 73.6 52.0 946 25-29 97.4 85.8 85.2 73.9 56.3 632 30-39 97.5 84.1 84.1 73.5 54.8 951 40-49 98.4 80.1 81.3 75.6 53.4 783 Marital status Never married 94.0 77.6 76.7 71.2 47.0 2,189 Ever had sex 96.4 84.9 81.5 69.2 49.2 832 Never had sex 92.6 73.1 73.8 72.5 45.6 1,356 In union 97.6 82.2 82.5 75.4 55.1 2,126 Divorced/separated/ widowed 95.5 77.8 77.3 76.5 51.7 96 Residence Urban 96.4 90.3 89.0 53.6 42.3 784 Rural 95.7 77.6 77.5 77.6 52.9 3,626 Province Kigali city 97.4 88.4 89.2 40.2 29.0 495 South 95.5 82.7 85.0 83.8 63.8 1,138 West 92.4 72.1 73.1 67.7 41.6 1,065 North 97.2 75.6 74.1 79.9 51.2 776 East 97.9 84.1 79.6 79.2 57.7 937 Education No education 92.6 66.8 69.2 75.5 40.7 558 Primary 96.0 79.5 78.7 73.8 51.0 3,290 Secondary or higher 97.9 94.8 94.7 68.9 61.4 561 Wealth quintile Lowest 95.7 71.1 71.6 74.1 44.3 799 Second 95.5 77.8 78.5 77.0 52.6 794 Middle 95.1 77.1 74.5 80.2 52.8 890 Fourth 95.5 81.1 82.0 75.1 54.7 900 Highest 96.8 89.5 88.7 62.6 50.2 1,028 Total 95.8 79.8 79.5 73.4 51.0 4,410 14.1.3 Opinions The promotion of safe sexual behaviors is a primary means of controlling the AIDS epidemic. Because women are more vulnerable than men to HIV infection, it is important to know whether women are able to refuse higher-risk sexual contact with their husbands/partners. For this reason, the RDHS-III asked women whether they believed that a wife is justified in refusing to have sex with her husband if she knows he has an STI, and whether she is justified in asking him to use a condom under the same circumstances. The results of the survey show that a majority of women (96 percent) believe that a wife is justified in refusing sexual contact or in asking her husband to use a condom if he has an STI (Table 14.6). The proportion of women professing this view is high for all background characteristics. However, it is somewhat lower among young women age 15 to 19 (90 percent), never-married women who have never had sex (93 percent), and women living in the West and North provinces (94 percent for both). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 201 Table 14.6 Attitudes toward negotiating safer sexual relations with husband Percentage of women age 15-49 who believe that, if a husband has a sexually transmitted disease, his wife is justified in refusing to have sexual relations with him or asking that they use a condom, by background characteristics, Rwanda 2005 Background characteristic Refusing to have sexual relations Asking that they use a condom Refusing sexual relations or asking that they use a condom Number of women Age 15-24 89.1 83.5 93.5 4,938 15-19 85.2 79.2 90.1 2,585 20-24 93.3 88.3 97.2 2,354 25-29 92.9 88.4 96.9 1,738 30-39 93.0 89.8 97.6 2,600 40-49 93.3 85.3 96.6 2,045 Marital status Never married 88.5 83.1 93.3 4,263 Ever had sex 91.6 90.1 96.9 758 Never had sex 87.8 81.6 92.5 3,505 In union 93.1 88.3 97.0 5,510 Divorced/separated/widowed 92.8 85.9 96.3 1,548 Residence Urban 92.1 91.6 97.8 1,921 Rural 91.2 84.9 95.1 9,400 Province Kigali city 91.9 93.1 98.6 1,127 South 94.1 85.9 97.0 2,958 West 88.3 82.5 93.5 2,824 North 90.2 85.3 94.0 2,063 East 92.2 87.8 96.0 2,348 Education No education 90.4 81.1 94.1 2,193 Primary 91.2 86.3 95.5 8,044 Secondary or higher 93.7 94.1 98.5 1,084 Total 91.3 86.0 95.5 11,321 During the survey, women and men were asked if they believed that children age 12 to 14 should be taught about using condoms to avoid AIDS. The results for this question are presented in Table 14.7. Overall, the proportion of men who believe that condom use should be taught to young people (82 percent) is a little higher than the proportion of women who share this view (80 percent). The widest differentials are between women and men with a secondary education or higher (88 percent of women and 84 percent of men favorable to condom education) and those with no education (74 percent of women and 77 percent of men favorable). A favorable opinion is more widespread among women and men in urban areas than in rural areas (86 percent of women and 85 percent of men in urban areas, compared with 79 percent of women and 82 percent of men in rural areas). Similarly, women in the richest quintile (86 percent) are more likely to be favorable to condom education than women in the poorest quintile (78 percent). The difference for men is much smaller (85 percent in the richest quintile, 82 percent in the poorest). 202 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.7 Adult support of education about condom use to prevent AIDS Percentage of women and men age 18-49 who agree that children age 12-14 years should be taught about using a condom to avoid AIDS, by background characteristics, Rwanda 2005 Women Men Background characteristic Percentage who agree Number of women Percentage who agree Number of men Age 18-19 81.3 951 81.8 400 20-24 82.6 2,353 85.1 946 25-29 83.8 1,738 83.4 632 30-39 79.9 2,597 83.1 951 40-49 74.3 2,044 77.8 783 Marital status Never married 82.0 2,638 84.2 1,490 In union 79.9 5,500 80.8 2,126 Divorced/separated/ widowed 78.3 1,544 90.0 96 Residence Urban 85.9 1,635 84.9 695 Rural 79.0 8,047 81.8 3,016 Province Kigali city 82.2 968 88.0 444 South 81.5 2,567 80.4 939 West 78.1 2,378 77.2 880 North 77.8 1,750 80.5 651 East 82.2 2,018 88.8 798 Education No education 74.3 2,094 77.3 520 Primary 80.9 6,587 83.0 2,666 Secondary or higher 88.2 1,001 84.2 525 Wealth quintile Lowest 77.6 2,052 82.0 672 Second 78.7 2,005 80.7 660 Middle 78.8 1,820 83.5 730 Fourth 79.8 1,826 80.5 758 Highest 86.2 1,979 84.5 892 Total 18-49 80.2 9,682 82.4 3,711 Perceptions and opinions about abstinence and fidelity were gathered by asking women and men a series of questions (see Figure 14.1). According to the results, women and men generally share the same perceptions and opinions with regard to abstinence and fidelity, except with respect to the fidelity of men known by the respondents. The proportion of women who said they believed that most men they knew were faithful (19 percent) is lower that that of men (27 percent). Nearly all women and men believe that young people should delay sexual intercourse until marriage: 98 percent of women and 97 percent of men believe that young men should wait; 98 percent of women and 96 percent of men believe that young women should wait. Nearly all women and men agreed that married men and women should have sexual intercourse only with their spouse (at least 96 percent for both). However, the percentage who said that most men they knew had sex only with their spouse is much lower (19 percent of women, 27 percent of men). The proportion who said that most married women they knew were faithful is somewhat higher (36 percent of women, 35 percent of men). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 203 14.2 HIGHER-RISK SEXUAL INTERCOURSE AND CONDOM USE Changing behavior associated with the spread of HIV is essential to curtailing the spread of the disease. For this reason, the RDHS-III asked respondents a series of questions about their behavior with respect to sexual intercourse. Higher-risk sexual intercourse was determined by the type of partner reported by the respondent. Sexual intercourse with a partner who was neither a spouse nor living with the respondent was considered higher risk. Table 14.8.1 shows the proportion of women who engaged in higher-risk intercourse in the 12 months preceding the survey, and the proportion of women who reported using a condom during their last higher-risk sexual intercourse. The results show that 8 percent of women who were sexually active in the 12 months preceding the survey had engaged in higher-risk sexual intercourse. All sexually active never- married women had higher-risk intercourse by definition, because their partners were neither spouses nor cohabiting with them. The proportion of young women age 15 to 19 who engaged in higher-risk intercourse is high (53 percent) because at this age most women have never been married. The proportion is high among never-married women for the same reason. More than half (56 percent) of divorced, separated, or widowed women had higher-risk intercourse in the past 12 months. It should also be noted that the proportion of women who had higher-risk intercourse is significantly higher in urban areas (15 percent) than in rural areas (7 percent). By level of education, the proportion is highest among women with a secondary education (11 percent, compared with 8 percent for a primary education and 6 percent for women with no education). By wealth quintile, the proportion is highest among women in the richest households (12 percent, compared with 8 percent in the first two quintiles). Of all those who engaged in higher-risk intercourse in the past 12 months, only 20 percent used a condom. Condom use was higher among women who had higher proportions of higher-risk intercourse, i.e., women in urban areas (35 percent), women with a secondary education or higher (47 percent), and women in the richest wealth quintile (38 percent). Figure 14.1 Perception and Beliefs about Abstinence and Faithfulness RDHS 2005 98 98 96 19 97 36 97 96 96 27 96 35 0 20 40 60 80 100 Women Men Young men should wait until they are married to have sexual intercourse Young women should wait until they are married to have sexual intercourse Married men should only have sex with their wives Most married men they know only have sex with their wives Married women should only have sex with their husbands Most married women they know only have sex with their husbands Percent 204 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.8.1 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: women Among women age 15-49 who had sexual intercourse in the past 12 months, the percentage who had intercourse with more than one partner and the percentage who had higher-risk sexual intercourse1 in the past 12 months, and among those having higher-risk intercourse in the past 12 months, the percentage reporting that a condom was used at last higher-risk intercourse, and the mean number of sexual partners during her lifetime for women who ever had sexual intercourse, by background characteristics, Rwanda 2005 Among women who had sexual intercourse in the past 12 months: Among women who had higher-risk intercourse1 in the past 12 months: Among women who ever had sexual intercourse: Background characteristic Percentage who had 2+ partners in the past 12 months Percentage who had higher-risk intercourse1 in the past 12 months Number of women Percentage who reported using a condom at last higher-risk intercourse1 Number of women Mean number of sexual partners in lifetime Number of women Age 15-24 1.0 15.3 1,287 26.4 197 1.2 1,697 15-19 2.3 53.0 151 27.6 80 1.3 311 20-24 0.8 10.3 1,136 25.5 117 1.2 1,385 25-29 0.5 6.1 1,354 21.6 82 1.3 1,563 30-39 0.6 6.4 1,997 15.6 128 1.6 2,532 40-49 0.3 5.5 1,249 6.2 68 1.6 2,024 Marital status Never married 4.0 100.0 246 24.2 246 1.7 758 In union 0.2 0.5 5,279 (9.8) 25 1.3 5,510 Divorced/separated/ widowed 3.3 56.1 362 15.6 203 1.9 1,548 Residence Urban 1.5 15.4 854 34.5 131 1.6 1,265 Rural 0.5 6.8 5,033 14.1 343 1.4 6,551 Education No education 0.4 6.4 1,405 14.0 89 1.5 1,916 Primary 0.6 8.2 3,952 16.3 325 1.4 5,168 Secondary or higher 1.2 11.3 531 46.8 60 1.5 732 Wealth quintile Lowest 0.5 7.9 1,202 6.0 96 1.5 1,709 Second 0.2 7.6 1,191 15.5 90 1.4 1,605 Middle 0.3 6.5 1,168 16.5 76 1.5 1,497 Fourth 0.9 7.0 1,218 15.1 85 1.4 1,493 Highest 1.0 11.5 1,108 38.1 128 1.5 1,512 Total 0.6 8.1 5,887 19.7 475 1.5 7,816 Note: Figures in parentheses are based on 25-49 unweighted cases. 1 Sexual intercourse with a nonmarital, noncohabiting partner Table 14.8.2, which presents the same data for men, shows that 14 percent of men had higher-risk sexual intercourse in the 12 months preceding the survey. The proportion who used a condom during their last higher-risk sexual intercourse was 41 percent. Nearly all young men age 15 to 19 had engaged in higher-risk sexual intercourse in the 12 months preceding the survey (96 percent). However, the highest percentage of condom use at last higher-risk intercourse was not in this age group (37 percent); instead it was for men age 25 to 29 (62 percent). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 205 Table 14.8.2 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: men Among men age 15-49 who had sexual intercourse in the past 12 months, the percentage who had intercourse with more than one partner and the percentage who had higher-risk sexual intercourse1 in the past 12 months, and among those having higher-risk intercourse in the past 12 months, the percentage reporting that a condom was used at last higher-risk intercourse, and the mean number of sexual partners during his lifetime for men who ever had sexual intercourse, by background characteristics, Rwanda 2005 Among men who had sexual intercourse in the past 12 months: Among men who had higher-risk intercourse1 in the past 12 months: Among men who ever had sexual intercourse: Background characteristic Percentage who had 2+ partners in the past 12 months Percentage who had higher-risk intercourse1 in the past 12 months Number of men Percentage who reported using a condom at last higher-risk intercourse1 Number of men Mean number of sexual partners in lifetime Number of men Age 15-24 4.4 48.0 343 39.5 165 2.1 800 15-19 4.9 96.4 61 37.0 59 1.6 249 20-24 4.3 37.6 282 40.8 106 2.3 550 25-29 4.7 15.0 450 61.8 67 2.6 549 30-39 5.0 6.3 866 37.2 54 3.1 925 40-49 5.6 5.3 740 (16.0) 39 4.0 780 Marital status Never married 6.9 99.3 234 43.4 232 2.6 833 In union 4.7 3.0 2,114 38.6 63 3.0 2,126 Divorced/separated/ widowed 11.7 60.2 51 (26.5) 31 5.3 95 Residence Urban 5.2 27.5 401 62.9 111 4.1 572 Rural 5.0 10.8 1,997 29.6 215 2.7 2,482 Education No education 3.7 10.2 391 (30.5) 40 2.8 444 Primary 5.6 13.2 1,694 32.9 224 2.8 2,190 Secondary or higher 3.8 19.8 313 76.4 62 4.1 420 Wealth quintile Lowest 3.1 10.3 466 (21.8) 48 2.6 568 Second 7.3 10.1 443 (28.3) 45 2.6 540 Middle 5.7 11.4 493 27.6 56 2.6 594 Fourth 4.6 9.9 496 (35.5) 49 2.8 629 Highest 4.7 25.5 500 60.4 128 4.0 723 Total 5.1 13.6 2,399 40.9 326 3.0 3,053 Note: Figures in parentheses are based on 25-49 unweighted cases. 1 Sexual intercourse with a nonmarital, noncohabiting partner As with women, the proportion of men who had higher-risk sexual intercourse in the last 12 months increases with level of education, from a low of 10 percent among those with no education, to a high of 20 percent among those with a secondary education or higher. This last category also shows a high rate of condom use (76 percent). By marital status, nearly all never-married men (99 percent) had engaged in higher-risk sexual intercourse in the 12 months preceding the survey. However, condom use among men in this category is low (43 percent). Higher-risk sexual contact is more frequent among men in urban areas (28 percent) than men in rural areas (11 percent). Condom use follows the same pattern: it is significantly higher in urban areas (63 percent) than in rural areas (30 percent). Five percent of men reported having had at least 2 sexual partners in the past 12 months. Overall, Rwandan men have an average of 3 sexual partners in their lifetime, compared with 1.5 for women. 206 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior 14.3 TESTING AND COUNSELING FOR HIV/AIDS Knowledge of HIV status can help limit the spread of the AIDS epidemic because it helps individuals make decisions that will protect themselves and their partners. The 2005 RDHS-III asked respondents whether they had ever been tested to see if they had the AIDS virus, and whether they had received results from the last HIV test taken in the 12 months preceding the survey. Table 14.9.1 shows that 76 percent of the women surveyed had never been tested. Only 21 percent of those who were tested had received the results. The proportion of women who received the results of the last HIV/AIDS test taken in the past 12 months is only 12 percent. Table 14.9.1 Prior HIV testing and knowledge of results: women Percent distribution of women by whether they were ever tested for HIV and by whether they received the results of the last test, and the percentage of women who received their test results the last time they were tested for HIV in the past 12 months, according to background characteristics, Rwanda 2005 Ever tested Background characteristic Received results Did not receive results Never tested Total1 Percentage who received results from last HIV test taken in the past 12 months Number of women Age 15-24 17.1 1.9 80.8 100.0 10.4 4,938 15-19 6.2 0.9 92.5 100.0 4.8 2,585 20-24 29.2 2.9 67.8 100.0 16.6 2,354 25-29 32.0 5.1 62.7 100.0 16.4 1,738 30-39 26.9 3.4 69.3 100.0 14.1 2,600 40-49 14.6 2.1 83.1 100.0 6.9 2,045 Marital status Never married 11.9 1.3 86.6 100.0 8.3 4,263 Ever had sex 29.5 3.0 67.2 100.0 18.9 758 Never had sex 8.0 0.9 90.8 100.0 6.0 3,505 In union 27.8 3.9 68.0 100.0 14.1 5,510 Divorced/separated/widowed 23.5 2.8 73.5 100.0 11.6 1,548 Residence Urban 43.1 3.7 52.8 100.0 23.0 1,921 Rural 16.7 2.6 80.5 100.0 9.2 9,400 Province Kigali city 45.2 3.4 51.1 100.0 24.4 1,127 South 18.3 3.2 78.4 100.0 9.1 2,958 West 17.6 2.9 79.1 100.0 10.3 2,824 North 19.8 2.2 77.7 100.0 11.1 2,063 East 18.9 2.1 78.7 100.0 10.3 2,348 Education No education 15.7 2.3 81.5 100.0 9.4 2,193 Primary 19.9 2.8 77.0 100.0 11.0 8,044 Secondary or higher 41.7 3.5 54.7 100.0 20.0 1,084 Wealth quintile Lowest 14.2 1.9 83.5 100.0 8.2 2,421 Second 15.0 2.7 82.1 100.0 8.5 2,325 Middle 19.2 2.4 78.1 100.0 10.8 2,099 Fourth 22.3 3.0 74.3 100.0 12.1 2,133 Highest 35.3 3.7 60.7 100.0 18.3 2,342 Total 21.2 2.8 75.8 100.0 11.6 11,321 1 Includes women with missing information HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 207 Results by age show that nearly all young women age 15 to 19 have never been tested for HIV/AIDS (93 percent). The proportion of women who were never tested is also high among women with no education (82 percent) and women in rural areas (81 percent). Nearly three-quarters of divorced, separated, or widowed women (74 percent) have never been tested, compared with only 68 percent of married women. By province, the data show a large difference between the City of Kigali (51 percent never tested) and the other provinces (at least 78 percent never tested). The proportion of women who received the results of the last HIV/AIDS test taken in the past 12 months is highest for women 20 to 29 (16 percent), never-married women who have ever had sex (19 percent), urban women (23 percent), women in the City of Kigali (24 percent), women with secondary or higher education (20 percent), and women in households in the highest wealth quintile (18 percent). Table 14.9.2 Prior HIV testing and knowledge of results: men Percent distribution of men by whether they were ever tested for HIV and by whether they received the results of the last test, and the percentage of men who received their test results the last time they were tested for HIV in the past 12 months, according to background characteristics, Rwanda 2005 Ever tested Background characteristic Received results Did not receive results Never tested Total1 Percentage who received results from last HIV test taken in the past 12 months Number of men Age 15-24 12.1 1.2 86.6 100.0 8.2 2,048 15-19 4.4 0.4 94.9 100.0 3.6 1,102 20-24 21.1 2.1 76.9 100.0 13.6 946 25-29 39.3 2.1 58.5 100.0 18.3 632 30-39 27.2 1.8 71.0 100.0 13.6 951 40-49 16.7 2.7 80.5 100.0 9.2 783 Marital status Never married 13.9 1.1 84.9 100.0 9.7 2,191 Ever had sex 23.7 1.8 74.5 100.0 15.3 833 Never had sex 7.8 0.7 91.3 100.0 6.2 1,358 In union 25.9 2.2 71.8 100.0 12.1 2,126 Divorced/separated/widowed 32.7 4.3 63.0 100.0 16.5 96 Residence Urban 34.8 2.5 62.6 100.0 19.9 784 Rural 16.9 1.6 81.5 100.0 9.0 3,629 Province Kigali city 39.5 1.9 58.7 100.0 22.3 495 South 17.8 1.5 80.6 100.0 7.7 1,139 West 16.4 1.8 81.7 100.0 9.6 1,065 North 19.7 0.7 79.6 100.0 13.2 777 East 17.1 2.7 80.2 100.0 8.8 937 Education No education 11.8 2.0 86.2 100.0 6.8 558 Primary 17.7 1.6 80.6 100.0 9.7 3,293 Secondary or higher 42.2 2.1 55.7 100.0 22.6 561 Wealth quintile Lowest 13.8 0.9 85.2 100.0 8.2 799 Second 13.5 1.0 85.5 100.0 7.2 794 Middle 15.7 2.3 81.8 100.0 8.8 892 Fourth 21.2 1.9 76.9 100.0 11.0 900 Highest 32.8 2.3 64.9 100.0 17.8 1,028 Total 20.1 1.7 78.1 100.0 11.0 4,413 1 Includes men with missing information 208 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior The highest proportions of women who received results from the last HIV test taken in the past 12 months are found among women in urban areas (23 percent), women in the City of Kigali (24 percent), women with a secondary education or higher (20 percent), and women in the richest households (18 percent). Table 14.9.2 shows prior HIV testing and knowledge of results for men. Seventy-eight percent of the men surveyed had never been tested for HIV. Twenty percent had been tested at some time and received the results. The proportion of those who received the results of the last HIV test taken in the past 12 months was only 11 percent. By age, a very high proportion of the youngest men have never been tested (95 percent for age 15 to 19), although previous tables showed that 96 percent of men in this age group had engaged in higher- risk sexual intercourse in the 12 months preceding the survey. A high proportion of men who have never been tested for HIV are found in rural areas (82 percent) and among those with no education (86 percent). The proportions who received the results of the last HIV test taken in the past 12 months follow a similar pattern to that of women, the highest proportions being among men in urban areas (20 percent), men in the City of Kigali (22 percent), men with higher educations (23 percent), and men in the richest households (18 percent). Women who had given birth in the two years preceding the survey were asked whether they had received HIV/AIDS counseling during an antenatal care (ANC) visit, whether they had taken a voluntary AIDS test during an ANC visit, and whether they had received the results of this test. The answers to these questions are presented in Table 14.10. Nearly six in ten women (56 percent) reported having received HIV/AIDS counseling, i.e., they were told about mother-to-child transmission of HIV and the importance of HIV/AIDS testing. Twenty-three percent of women took a voluntary HIV/AIDS test and received the results. Overall, 22 percent received counseling, took an HIV/AIDS test, and received the results. This proportion is much higher among some groups of women: women in urban areas (58 percent), women living in the City of Kigali (56 percent), and women with a secondary education or higher (37 percent). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 209 Table 14.10 Pregnant women counseled and tested for HIV Among all women who gave birth in the two years preceding the survey, percentage who received HIV counseling during antenatal care for their most recent birth, and percentage who accepted an offer of HIV testing and whether they received the test results, according to background characteristics, Rwanda 2005 Percentage who were offered and accepted an HIV test during antenatal care and who: Background characteristic Percentage who received HIV counseling during antenatal care Received results Did not receive results Percentage who were counseled, were offered and who accepted an HIV test, and who received results Number of women who gave birth in the last 2 years Age 15-24 53.9 26.0 3.1 23.3 899 15-19 55.8 33.2 3.5 29.7 73 20-24 53.7 25.4 3.1 22.7 827 25-29 56.6 23.8 3.9 22.0 965 30-39 56.5 22.5 3.0 20.9 1,209 40-49 55.9 17.8 4.0 17.4 363 Residence Urban 76.3 62.8 6.1 58.0 456 Rural 52.6 17.3 3.0 15.9 2,980 Province Kigali city 69.7 62.9 6.4 56.4 245 South 55.2 19.5 3.8 18.1 820 West 56.1 23.7 4.0 22.0 920 North 55.8 19.8 2.2 19.0 671 East 51.6 17.4 2.3 15.6 780 Education No education 50.2 18.0 2.9 16.1 779 Primary 56.5 23.0 3.6 21.5 2,388 Secondary or higher 65.5 41.3 3.0 37.2 269 Total 55.8 23.3 3.4 21.5 3,436 14.4 SEXUALLY TRANSMITTED INFECTIONS (STIS) The 2005 RDHS-III also sought to determine whether women and men who had ever had sexual intercourse had had an STI and/or the symptoms of an STI in the 12 months preceding the survey. The total self-reported STI prevalence (according to spontaneous declarations and symptoms) for women who had ever had intercourse is 5 percent. However, this figure should be taken as an order of magnitude rather than a precise estimate because the presence of the various signs or symptoms is not always proof of an STI (Table 14.11). The proportion of men who reported having an STI and/or the symptoms of an STI in the 12 months preceding the survey was 3 percent. 210 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.11 Self-reported prevalence of sexually-transmitted infections (STIs) and STI symptoms Among women and men age 15-49 who ever had sexual intercourse, the percentage reporting having an STI and/or symptoms of an STI in the past 12 months, by background characteristics, Rwanda 2005 Percentage of women who reported having in the past 12 months: Percentage of men who reported having in the past 12 months: Background characteristic STI Bad smelling/ abnormal genital discharge Genital sore or ulcer STI/genital discharge/ sore or ulcer Number of women who ever had sexual intercourse STI Bad smelling/ abnormal genital discharge Genital sore or ulcer STI/genital discharge/ sore or ulcer Number of men who ever had sexual intercourse Age 15-24 0.7 3.4 1.9 4.5 1,697 0.3 2.2 0.8 3.0 800 15-19 0.6 3.7 1.7 4.5 311 0.7 3.8 0.0 4.1 249 20-24 0.8 3.4 1.9 4.5 1,385 0.2 1.5 1.2 2.5 550 25-29 1.0 3.2 2.4 4.6 1,563 1.3 1.9 1.0 3.1 549 30-39 1.7 3.9 3.1 5.6 2,532 1.4 1.1 2.3 2.9 925 40-49 1.2 3.9 2.9 5.1 2,024 0.9 0.3 1.6 2.1 780 Marital status Never married 1.3 4.0 3.0 5.7 758 0.4 2.2 0.7 2.8 833 In union 1.0 3.4 2.4 4.6 5,510 1.2 1.0 1.9 2.8 2,126 Divorced/separated/ widowed 2.1 4.7 3.5 6.2 1,548 0.8 0.8 0.0 0.8 95 Circumcised Yes na na na na na 1.5 1.2 1.9 2.9 341 No/missing na na na na na 0.9 1.3 1.4 2.7 2,712 Residence Urban 1.9 5.4 3.3 7.3 1,265 1.9 1.4 2.3 3.4 572 Rural 1.1 3.4 2.5 4.6 6,551 0.8 1.3 1.3 2.6 2,482 Province Kigali city 1.2 5.3 2.7 6.9 733 1.1 1.6 2.0 3.2 362 South 1.2 2.9 2.2 4.1 2,044 0.7 0.9 1.1 2.1 781 West 1.5 4.6 3.3 6.1 1,907 1.2 1.6 2.0 3.7 742 North 0.9 1.8 2.0 2.6 1,464 1.2 2.0 0.9 2.6 519 East 1.3 4.5 3.2 6.2 1,667 0.7 0.9 1.5 2.3 649 Education No education 1.1 3.4 2.9 4.8 1,916 1.6 1.2 2.0 3.6 444 Primary 1.2 3.6 2.5 4.9 5,168 0.9 1.3 1.1 2.4 2,190 Secondary or higher 1.8 4.8 3.0 6.5 732 0.9 1.7 3.0 3.6 420 Total 1.2 3.7 2.7 5.0 7,816 1.0 1.3 1.5 2.7 3,053 na = Not applicable Those who reported having had an STI and/or the symptoms of an STI in the past 12 months were asked if they had sought counseling and/or treatment from any source. Half of the women and men responded affirmatively (Figure 14.2). Only a little more than one in ten sought advice or treatment from a health professional (12 percent of women and 14 percent of men). HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 211 14.5 INJECTIONS FROM A HEALTH WORKER Injections given without compliance to aseptic standards can be a source of contamination. It is therefore important to know whether the population is able to receive injections from approved health workers. Table 14.12 shows that a total of 12 percent of women and 9 percent of men received an injection from a health worker in the 12 months preceding the survey. Ninety-five percent of women and 89 percent of men received their last injection from a syringe and needle taken from a newly opened package. Figure 14.2 Women and Men Seeking Treatment for STIs 12 14 5 12 49 5251 48 Women Men 0 20 40 60 80 Percent Clinic/hospital/health professional Advice or medicine from shop/pharmacy Advice or treatment from any source No advice or treatment RDHS 2005 212 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.12 Prevalence of injections Percentage of women and men age 15-49 who received at least one injection from a health worker in the last 12 months, the average number of medical injections per person and, among those who received an injection, the percentage whose health worker took the syringe and needle from a new and unopened package for the last injection, by background characteristics, Rwanda 2005 Women Men Background characteristic Percentage who received an injection from a health worker in the past 12 months Average number of medical injections per year Number of women Last injection, syringe and needle taken from newly opened package Number of women receiving injections from a health worker in the last 12 months Percentage who received an injection from a health worker in the past 12 months Average number of medical injections per year Number of men Last injection, syringe and needle taken from newly opened package Number of men receiving injections from a health worker in the last 12 months Age 15-24 10.7 2.3 4,938 95.6 530 10.0 2.4 2,048 89.6 205 15-19 8.4 2.3 2,585 95.6 217 9.3 2.4 1,102 88.8 103 20-24 13.3 2.3 2,354 95.6 313 10.8 2.3 946 90.5 102 25-29 17.5 2.1 1,738 97.2 305 10.3 2.5 632 88.7 65 30-39 11.4 2.0 2,600 94.0 296 8.4 4.3 951 91.7 80 40-49 8.5 2.9 2,045 88.4 174 7.9 3.4 783 86.5 62 Residence Urban 14.7 2.4 1,921 95.0 283 14.3 3.0 784 94.4 112 Rural 10.9 2.2 9,400 94.6 1,021 8.3 2.9 3,629 87.6 300 Province Kigali city 15.4 2.6 1,127 94.6 173 15.6 3.5 495 92.6 77 South 10.9 2.0 2,958 95.9 324 9.4 2.4 1,139 87.6 107 West 12.0 2.3 2,824 91.7 340 8.9 3.2 1,065 87.9 95 North 10.3 2.1 2,063 96.3 212 9.3 3.1 777 88.3 73 East 10.9 2.3 2,348 95.6 255 6.4 2.3 937 92.3 60 Education No education 9.5 2.1 2,193 91.7 208 6.4 3.3 558 (77.6) 36 Primary 11.3 2.1 8,044 94.9 911 9.1 2.8 3,293 90.2 301 Secondary or higher 17.1 2.9 1,084 96.6 185 13.4 3.1 561 91.9 75 Wealth quintile Lowest 9.2 1.8 2,421 95.1 223 8.7 3.2 799 84.8 69 Second 8.8 2.0 2,325 97.1 204 6.4 3.6 794 89.8 51 Middle 11.3 2.0 2,099 93.3 236 8.6 2.1 892 88.0 77 Fourth 13.6 2.7 2,133 93.3 291 7.3 3.6 900 93.0 66 Highest 14.9 2.5 2,342 95.0 350 14.5 2.6 1,028 90.5 149 Total 11.5 2.2 11,321 94.7 1,304 9.3 2.9 4,413 89.4 412 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figure 14.3 shows the proportions of women and men age 15 to 49 who received an injection from a health worker in the 12 months preceding the survey, according to source of the last injection. The public sector (84 percent for women and 73 percent for men), mainly health centers (68 percent for women and 57 percent for men), was by far the primary source of injections. Approximately 11 percent of women received injections at a private sector health facility; the corresponding proportion for men is 25 percent. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 213 Figure 14.4 shows that nearly all injections received from a health worker were given with a needle and syringe taken from a newly opened package: 95 percent for women and 88 percent for men. There is no difference between public and private sector as far as the women’s data are concerned. Figure 14.3 Type of Facility where Received Last Medical Injection 84 73 14 15 68 57 0 11 25 2 54 16 Women Men 0 20 40 60 80 100 PUBLIC SECTOR Govt. hospital Govt. health center PRIVATE MEDICAL SECTOR Private hospital Private health worker RDHS 2005 Figure 14.4 Percentage whose Last Injection was Given with a Syringe and Needle Taken from a New, Unopened Package 95 88 94 9495 88 0 93 9392 92 Women Men 0 20 40 60 80 100 Percent PUBLIC SECTOR Govt. hospital Govt. health center PRIVATE MEDICAL SECTOR Private health worker RDHS 2005 214 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior 14.6 KNOWLEDGE OF HIV/AIDS AND SEXUAL BEHAVIOR AMONG YOUTH Table 14.13 shows that, overall, the proportion of young people age 15 to 24 who have a comprehensive knowledge of HIV/AIDS is not very high: only 51 percent of young women and 54 percent of young men are shown to have a comprehensive knowledge of the means of prevention and transmission of HIV/AIDS. The proportion increases with age, from 44 percent of women age 15 to 17, to 58 percent at age 23 to 24; and from 45 percent of men age 15 to 17, to 62 percent at age 23 to 24. Table 14.13 Comprehensive knowledge about AIDS and of a source of condoms among youth Percentage of young women and young men age 15-24 with comprehensive knowledge about AIDS and percentage with knowledge of a source of condoms, by background characteristics, Rwanda 2005 Women 15-24 Men 15-24 Background characteristic Percentage with comprehensive knowledge of AIDS1 Percentage who know a condom source Number of women Percentage with comprehensive knowledge of AIDS1 Percentage who know a condom source Number of men Age 15-19 45.3 31.3 2,585 49.0 65.8 1,102 15-17 43.5 27.1 1,633 45.1 60.4 701 18-19 48.4 38.5 952 55.8 75.1 400 20-24 57.1 43.2 2,354 59.0 81.5 946 20-22 56.6 42.4 1,437 57.6 80.4 614 23-24 57.8 44.5 917 61.5 83.5 332 Marital status Never married 49.3 35.7 3,762 53.3 73.1 1,863 Ever had sex 56.9 52.4 520 62.2 86.7 615 Never had sex 48.1 33.0 3,242 48.9 66.3 1,248 Ever married 55.9 41.3 1,176 57.0 72.9 185 Residence Urban 63.3 58.9 910 58.6 84.7 345 Rural 48.1 32.1 4,028 52.6 70.7 1,703 Province Kigali city 67.6 63.5 554 54.8 88.7 221 South 56.8 37.0 1,231 62.2 72.3 548 West 36.0 29.0 1,274 45.5 59.5 499 North 51.8 32.2 859 54.4 73.7 344 East 52.6 36.7 1,020 50.9 81.1 436 Education No education 41.8 24.1 553 44.3 58.7 174 Primary 50.1 34.2 3,947 52.6 71.9 1,676 Secondary or higher 69.8 78.8 439 70.3 95.2 198 Wealth quintile Lowest 42.3 23.6 1,015 50.4 60.8 364 Second 52.8 32.4 1,006 52.9 68.2 359 Middle 49.6 33.8 847 56.3 71.7 435 Fourth 50.1 35.3 952 52.8 77.1 419 Highest 58.7 57.3 1,118 54.7 83.8 471 Total 15-24 50.9 37.0 4,938 53.6 73.0 2,048 1 Comprehensive knowledge means knowing that use of condoms and having just one uninfected faithful partner can reduce the chances of getting the AIDS virus, knowing that a healthy-looking person can have the AIDS virus, and rejecting the two most common local misconceptions about AIDS transmission and prevention. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 215 Comprehensive knowledge of AIDS among young people increases with educational attainment for both women and men. Among those with no education, 42 percent of women and 44 percent of men have a comprehensive knowledge of AIDS, compared with 70 percent with a secondary education or higher for both sexes. By marital status, the highest proportions of comprehensive knowledge are among never- married young people who have had sexual intercourse (57 percent of women and 62 percent of men) and young people who are married (56 percent of women and 57 percent of men). Like adults, young people in urban areas (63 percent of women, 59 percent of men) are more likely to have comprehensive knowledge of AIDS than those in rural areas (48 percent of women, 53 percent of men). Among young people, there is a wide gap in knowledge of a source of condoms between men and women (73 percent for men, 37 percent for women). The next-largest differential is by level of education: 95 percent of men and 79 percent of women with a secondary education or higher know where to obtain condoms; only 59 percent of men and 24 percent of women with no education know a source for condoms. Youth in urban areas (59 percent of women, 85 percent of men) are more likely to know a condom source than youth in rural areas (32 percent of women, 71 percent of men). Knowledge of a source is higher for never-married youth who have had sexual intercourse and youth who are married. The proportion of youth who know where to obtain condoms is highest in the richest quintile (57 percent of women and 84 percent of men). Age at first intercourse as a determinant of sexual activity among young people age 15 to 24 is perhaps more important for HIV/AIDS prevention than any other variable. For this reason, Table 14.14 presents the findings for men and women age 15 to 24 who have ever had sexual intercourse whose age at first intercourse was below age 15 and below age 18. Approximately 4 percent of women age 15 to 24 had intercourse before the age of 15, and almost one in five (18 percent) had intercourse before the age of 18. A much higher proportion of men (13 percent) than women (4 percent) had sexual intercourse before age 15. The proportion of men who had sexual intercourse for the first time before age 18 is 27 percent, compared with 18 percent for women. 216 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.14 Age at first sexual intercourse among youth Percentage of young women and of young men age 15-24 who had sexual intercourse before age 15 and percentage of young women and of young men age 18-24 who had sexual intercourse before age 18, by background characteristics, Rwanda 2005 Women Men Background characteristic Percentage who have had sexual intercourse before age 15 Number of women age 15-24 Percentage who have had sexual intercourse before age 18 Number of women age 18-24 Percentage who have had sexual intercourse before age 15 Number of men age 15-24 Percentage who have had sexual intercourse before age 18 Number of men age 18-24 Age 15-19 5.2 2,585 na na 15.3 1,102 na na 15-17 5.4 1,633 na na 14.2 701 na na 18-19 4.7 952 13.8 952 17.2 400 29.3 400 20-24 2.6 2,354 19.1 2,354 10.8 946 26.3 946 20-22 2.3 1,437 19.1 1,437 12.6 614 29.7 614 23-24 2.9 917 19.1 917 7.4 332 20.2 332 Marital status Never married 4.3 3,762 9.2 2,138 13.8 1,863 27.3 1,161 Ever married 2.8 1,176 32.8 1,167 7.1 185 26.5 185 Knows condom source Yes 4.5 1,828 18.2 1,384 14.7 1,496 29.7 1072 No 3.6 3,111 17.1 1,921 9.2 552 17.6 274 Residence Urban 5.8 910 18.2 626 12.2 345 26.7 255 Rural 3.5 4,028 17.4 2,679 13.4 1,703 27.3 1,091 Province Kigali city 5.2 554 18.9 397 8.7 221 23.3 170 South 4.2 1,231 14.3 840 17.1 548 29.9 348 West 2.9 1,274 15.4 829 15.8 499 29.1 313 North 3.6 859 19.0 548 8.3 344 21.3 218 East 4.5 1,020 22.2 691 11.5 436 28.6 297 Education No education 3.9 553 29.8 455 8.1 174 23.9 135 Primary 4.0 3,947 16.3 2,494 13.3 1,676 27.8 1,049 Secondary or higher 2.9 439 10.6 355 16.7 198 26.0 162 Wealth quintile Lowest 3.7 1,015 18.2 648 15.6 364 28.0 236 Second 3.1 1,006 18.5 685 13.8 359 26.0 225 Middle 3.8 847 19.2 571 10.0 435 22.0 273 Fourth 4.2 952 17.0 645 12.5 419 29.1 277 Highest 4.8 1,118 15.4 756 14.4 471 30.1 334 Total 3.9 4,938 17.6 3,305 13.2 2,048 27.2 1,346 na = Not applicable A comparison of this data with those of the previous survey (2000 RDHS-II) shows an increase in the proportion of young women and men having early intercourse, i.e., before the age of 15 (Figure 14.5). However, the proportion of youth having intercourse before age 18 has declined. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 217 Table 14.15 shows results for condom use at first intercourse among youth age 15 to 24. The proportion of youth who used a condom at first inter- course is somewhat higher for men (12percent) than for women (7 percent). The highest proportions of condom use at first intercourse are seen in the highest level of educational attainment (21 per- cent of men, 22 percent of women), urban areas (26 percent of men, 19 per- cent of women), youth who know of a condom source (14 percent of men, 11 percent of women), and youth in the richest quintile (20 percent of men, 15 percent of women). It should be noted that the data according to age show the highest proportion of condom use at first intercourse to be among women age 15 to 17 (16 percent) and men age 18 to 19 (13 percent). Never-married young people comprise an at-risk population because, during this period in their life, sexual relations are generally unstable and prone to multiple partnership. For this reason, the RDHS-III sought to assess the behavior of young people age 15 to 24 with respect to HIV/AIDS prevention. Table 14.15 Condom use at first sexual intercourse among youth Percentage of young women and young men age 15-24 who used a condom the first time they had sexual intercourse, by background characteristics, Rwanda 2005 Women Men Background characteristic Percentage who used a condom at first sexual intercourse Number of women who have ever had sexual intercourse Percentage who used a condom at first sexual intercourse Number of men who have ever had sexual intercourse Age 15-19 13.3 311 10.7 249 15-17 16.2 131 7.9 124 18-19 11.2 181 13.4 125 20-24 5.2 1,385 12.0 550 20-22 6.1 709 11.8 313 23-24 4.2 676 12.3 237 Marital status Never married 17.4 520 13.3 615 Ever married 1.9 1,176 6.0 185 Knows condom source Yes 10.8 758 13.7 668 Non 3.4 938 0.7 132 Residence Urban 18.5 311 26.4 157 Rural 4.0 1,385 8.0 642 Education No education 2.3 309 5.5 75 Primary 6.1 1,257 11.1 640 Secondary or higher 22.1 131 20.8 84 Wealth quintile Lowest 2.9 347 4.6 147 Second 5.5 337 4.7 125 Middle 5.3 294 11.6 153 Fourth 3.9 360 12.3 170 Highest 15.4 357 20.3 204 Total 15-24 6.7 1,697 11.6 800 Figure 14.5 Trends in Age at First Sex, Rwanda 2000 and 2005 3 9 26 30 5 15 19 26 0 5 10 15 20 25 30 35 Percent RDHS-II 2000 RDHS-III 2005 Women age 15-19 who had sex before age 15 Men age 15-19 who had sex before age 15 Women age 20-24 who had sex before age 18 Men age 20-24 who had sex before age 18 218 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.16 shows the proportion of never-married youth age 15 to 24 who have ever had sexual intercourse, and the proportion who used condoms at last sexual intercourse. Approximately 5 percent of never-married women age 15 to 24 had sexual intercourse in the 12 months preceding the survey. Among these women, 25 percent used a condom at their last sexual intercourse. Among never-married men age 15 to 24, approximately 9 percent reported having had sexual intercourse in the past 12 months and, among these, 39 percent used a condom at their last sexual intercourse. Table 14.16 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth Among never-married women and men age 15-24, the percentage who have never had sexual intercourse, the percentage who have had sexual intercourse in the past 12 months, and, among those who have had premarital sexual intercourse in the past 12 months, the percentage who used a condom at the last sexual intercourse, by background characteristics, Rwanda 2005 Women Men Background characteristic Percentage who have never had sexual intercourse Percentage who have had sexual intercourse in the past 12 months Number of never- married women Percentage who used a condom at last sexual intercourse Number of never- married women who have had sexual intercourse in the past 12 months Percentage who have never had sexual intercourse Percentage who have had sexual intercourse in the past 12 months Number of never- married men Percentage who used a condom at last sexual intercourse Number of never- married men who have had sexual intercourse in the past 12 months Age 15-19 90.6 3.2 2,510 27.4 80 77.5 5.3 1,100 37.0 59 15-17 92.5 2.7 1,624 (27.6) 43 82.3 4.0 701 (27.5) 28 18-19 87.0 4.1 886 (27.2) 37 69.1 7.7 398 (45.7) 31 20-24 77.4 7.7 1,252 23.3 96 51.8 13.5 763 40.5 103 20-22 79.3 7.0 917 20.9 64 55.0 11.7 547 31.4 64 23-24 71.9 9.6 335 (28.1) 32 43.7 18.2 216 (55.1) 39 Knows condom source Yes 79.7 7.5 1,342 33.7 101 60.8 10.9 1,361 42.9 148 Non 89.8 3.1 2,420 13.8 76 83.7 2.8 502 * 14 Residence Urban 77.6 8.5 771 38.5 66 57.3 14.9 327 65.3 49 Rural 88.4 3.7 2,991 17.2 111 69.1 7.4 1,536 27.9 113 Education No education 81.8 8.3 298 (11.5) 25 69.7 9.8 141 * 14 Primary 87.4 4.0 3,079 22.3 122 67.8 8.1 1,529 32.3 124 Secondary or higher 79.8 7.6 385 (48.5) 29 59.0 12.6 193 (82.6) 24 Wealth quintile Lowest 87.6 3.6 763 (0.0) 28 67.1 7.8 322 (29.1) 25 Second 87.1 3.7 767 (22.2) 29 72.7 6.8 321 * 22 Middle 90.1 3.2 614 * 20 74.0 6.0 382 * 23 Fourth 86.7 5.6 683 (25.7) 38 65.5 8.3 380 (33.0) 31 Highest 81.3 6.6 936 41.1 62 58.3 13.2 458 56.6 61 Total 15-24 86.2 4.7 3,762 25.2 176 67.0 8.7 1,863 39.2 162 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 219 Table 14.17 shows the proportion of youth age 15 to 24 who had higher-risk sexual intercourse and the proportions who used condoms at last higher-risk intercourse. Fifteen percent of young women and 48 percent of young men had higher-risk sexual intercourse in the 12 months preceding the survey. Among the women, 26 percent used a condom at last higher-risk sexual intercourse. The proportion for men is 40 percent. Table 14.17 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months Among young women and men age 15-24 who had sexual intercourse in the past 12 months, the percentage who had higher-risk sexual intercourse in the past 12 months, and among those having higher-risk intercourse in the past 12 months, the percentage reporting that a condom was used at last higher-risk intercourse, by background characteristics, Rwanda 2005 Women 15-24 Men 15-24 Background characteristic Percentage who had higher-risk intercourse in the past 12 months Number of women who have had sexual intercourse in the past 12 months Percentage who reported using a condom at last higher-risk intercourse Number of women who have had higher-risk intercourse in the past 12 months Percentage who had higher-risk intercourse in the past 12 months Number of men who have had sexual intercourse in the past 12 months Percentage who reported using a condom at last higher-risk intercourse Number of men who have had higher-risk intercourse in the past 12 months Age 15-19 53.0 151 28.0 80 96.4 61 37.0 59 15-17 83.0 52 (28.0) 43 (100.0) 28 (27.5) 28 18-19 37.0 99 (27.0) 37 (93.3) 33 (45.7) 31 20-24 10.0 1,136 26.0 117 37.6 282 40.8 106 20-22 13.0 561 24.0 72 52.0 129 32.2 67 23-24 8.0 576 (29.0) 44 25.4 153 (55.7) 39 Marital status Never married 100.0 176 25.0 176 99.0 162 39.6 160 In union 0.0 1,038 * 3 0.8 175 * 1 Divorced/separated/ widowed 24.0 73 * 17 * 6 * 3 Knows condom source Yes 19.0 564 35.0 110 53.1 279 43.8 148 Non 12.0 723 15.0 87 25.6 64 * 16 Residence Urban 36.0 193 39.0 70 75.9 64 67.7 48 Rural 12.0 1,094 20.0 126 41.6 279 27.7 116 Education No education 11.0 266 (20.0) 29 (32.5) 46 * 15 Primary 15.0 942 23.0 137 47.1 268 32.8 126 Secondary or higher 38.0 79 (50.0) 30 (80.6) 29 (85.7) 23 Wealth quintile Lowest 12.0 266 (5.0) 31 39.5 67 (27.7) 26 Second 13.0 251 (28.0) 33 41.5 59 * 25 Middle 10.0 242 * 24 30.6 76 * 23 Fourth 15.0 294 (22.0) 44 46.0 68 (33.0) 31 Highest 28.0 234 42.0 64 81.2 73 58.2 59 Total 15-24 15.0 1,287 26.0 197 48.0 343 39.5 165 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figure 14.6 shows the distribution of youth according to their risk of contracting HIV. The following are the three risk categories: • Youth who were not exposed to the risk of sexually transmitted HIV because they had had no sexual contact (66 percent of women age 15 to 24 and 61 percent of men age 15 to 24). 220 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior • Youth who have had sexual intercourse but who were not exposed to the risk of sexually transmitted HIV in the past 12 months, either because they had had no sexual contact or because they had engaged in healthy, responsible sexual behaviors (single partner and use of condom at last sexual intercourse). • Youth at risk of contracting HIV because they engaged in higher-risk sexual behavior in the past 12 months (24 percent of women, 14 percent of men). This category includes young people who had only one partner but did not use a condom at last intercourse (25 percent of women and 13 percent of men), youth who used a condom but had multiple partners (less than 1 percent), and youth who had intercourse with multiple partners without using a condom (less than one percent). Youth in this last category have the greatest risk of contracting HIV. Women who have sexual intercourse with older men who, by virtue of their age, have a greater chance of being infected with the AIDS virus, are at increased risk of contracting HIV. Table 14.18 shows that among women age 15 to 19 who had extramarital intercourse in the 12 months preceding the survey, nearly 5 percent reported having had intercourse with a man at least 10 years older than themselves. The proportion is higher for the younger age group (9 percent for women age 15 to 17; 2 percent for women age 18 to 19). Figure 14.6 Abstinence, Being Faithful, and Condom Use (ABC) Among Young Women and Men 15-19 20-24 15-24 15-19 20-24 15-24 WOMEN MEN 0 20 40 60 80 100 Percent Never had sex No partners in last year One partner, used condom at last sex One partner, no condom use at last sex Multiple partners, used condom at last sex Multiple partners, no condom use at last sex Note: Number of partners refers to the 12 months preceding the survey. RDHS 2005 HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 221 Table 14.18 Age-mixing in sexual relationships among women age 15-19 Percentage of women age 15-19 who had higher-risk sexual intercourse in the past 12 months with a man who was 10 or more years older than themselves, by background characteristics, Rwanda 2005 Background characteristic Percentage of women who had higher-risk intercourse with a man 10+ years older Number of women 15-19 who had higher-risk intercourse in the past 12 months Age 15-17 9.5 52 18-19 2.0 99 Marital status Never married 8.7 80 In union 0.0 65 Divorced/separated/ widowed * 6 Knows condom source Yes 1.3 63 Non 7.0 88 Residence Urban (0.0) 32 Rural 5.9 118 Education No education (7.5) 28 Primary 4.2 116 Secondary or higher * 7 Total 15-19 4.6 151 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. It is generally recognized that excessive alcohol consumption clouds judgment and increases the likelihood of risky behavior. In addition, risky behavior under the influence of alcohol is most common at younger ages. The RDHS-III asked respondents age 15 to 24 whether they or their partners had consumed alcohol the last time they had sexual intercourse. Nearly 1 percent of women and 10 percent of men reported that they had consumed alcohol the last time they had sexual intercourse (Table 14.19). Among men, this behavior was more frequent for the 15 to 19 age group (15 percent), never-married men (16 percent), and men living in households in the first two wealth quintiles (13 percent). 222 | HIV/AIDS-Related Knowledge, Attitudes, and Behavior Table 14.19 Drunkenness during sexual intercourse among youth Among young women and men age 15-24 who had sexual intercourse in the past 12 months, the percentages who had sexual intercourse while being drunk, by background characteristics, Rwanda 2005 Women 15-24 Men 15-24 Background characteristic Percentage who had sexual intercourse in the past 12 months when drunk Percentage who had sexual intercourse in the past 12 months when drunk or with a partner who was drunk Number of women who had sexual intercourse in the past 12 months Percentage who had sexual intercourse in the past 12 months when drunk Percentage who had sexual intercourse in the past 12 months when drunk or with a partner who was drunk Number of men who had sexual intercourse in the past 12 months Age 15-19 1.8 3.3 151 14.6 14.6 61 15-17 0.0 0.0 52 (13.0) (13.0) 28 18-19 2.7 5.1 99 (15.9) (15.9) 33 20-24 0.7 6.2 1,136 8.5 8.5 282 20-22 0.5 7.1 561 9.8 9.8 129 23-24 0.9 5.3 576 7.5 7.5 153 Marital status Never married 0.4 6.7 176 15.6 15.6 162 In union 0.9 4.9 1,038 4.4 4.4 175 Divorced/separated/ widowed 0.0 16.5 73 * * 6 Knows condom source Yes 0.7 6.0 564 10.5 10.5 279 Non 0.8 5.7 723 5.8 5.8 64 Residence Urban 0.4 4.8 193 6.8 6.8 64 Rural 0.9 6.0 1,094 10.2 10.2 279 Education No education 1.8 4.8 266 (11.1) (11.1) 46 Primary 0.6 6.2 942 9.9 9.9 268 Secondary or higher 0.0 4.4 79 (4.3) (4.3) 29 Wealth quintile Lowest 1.0 6.6 266 12.8 12.8 67 Second 0.6 5.4 251 13.4 13.4 59 Middle 1.5 5.8 242 6.5 6.5 76 Fourth 0.8 6.6 294 8.1 8.1 68 Highest 0.0 4.5 234 8.2 8.2 73 Total 15-24 0.8 5.8 1,287 9.6 9.6 343 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. HIV/AIDS-Related Knowledge, Attitudes, and Behavior | 223 The preceding results indicate that many young people are sexually active and that their sexual intercourse is often high-risk. It is therefore important to know what percentage of these young people are consenting to HIV testing and are receiving the results. Table 14.20 shows that 21 percent of young women age 15 to 24 have been tested and received the results in the past 12 months. The proportion is 16 percent for men. By age, the highest proportions who took an HIV test and received the results are among women age 18 to 19 (34 percent) and men age 20 to 22 (18 percent). The proportions are higher among never-married youth (29 percent of women, 18 percent of men), youth who know of a condom source (26 percent of women, 19 percent of men), and youth in urban areas (43 percent of women, 29 percent of men). Young people with a secondary education or higher (39 percent of women) and youth in the richest quintile (34 percent of women, 26 percent of men) are also likely to have taken an HIV test and received the results. Table 14.20 Recent HIV tests among youth Among young women and young men age 15-24 who have had sexual intercourse in the past 12 months, the percentage who have had an HIV test in the past 12 months and received the results of the test, by background characteristics, Rwanda 2005 Women 15-24 Men 15-24 Background characteristic Percentage who have been tested for HIV and received results in the past 12 months Number of women who have had sexual intercourse in the past 12 months Percentage who have been tested for HIV and received results in the past 12 months Number of men who have had sexual intercourse in the past 12 months Age 15-19 26.9 151 12.7 61 15-17 12.9 52 (5.3) 28 18-19 34.4 99 (19.0) 33 20-24 20.1 1,136 16.9 282 20-22 22.3 561 17.6 129 23-24 17.9 576 16.4 153 Marital status Never married 29.0 176 18.0 162 In union 19.3 1,038 13.5 175 Divorced/separated/widowed 23.2 73 * 6 Knows condom source Yes 26.3 564 18.8 279 Non 16.6 723 4.9 64 Residence Urban 43.0 193 29.3 64 Rural 17.0 1,094 13.2 279 Education No education 22.1 266 (13.0) 46 Primary 19.0 942 14.9 268 Secondary or higher 39.3 79 (33.5) 29 Wealth quintile Lowest 18.8 266 13.9 67 Second 16.0 251 7.7 59 Middle 18.0 242 15.7 76 Fourth 19.2 294 15.9 68 Highest 33.5 234 26.0 73 Total 15-24 20.9 1,287 16.2 343 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. HIV Prevalence and Associated Factors | 225 HIV PREVALENCE AND ASSOCIATED FACTORS 15 Rwanda has long been considered one of the African countries most affected by the AIDS virus. In fact, the estimated prevalence rates derived from the first survey conducted on a national level in 1986 were 17.8 percent in urban areas and 1.3 percent in rural areas. In 1988, Rwanda established an HIV sentinel surveillance system among pregnant women attending antenatal clinics and among STI-clinic patients. In 1988 and 1991, the first sets of surveillance data were made available. The 1991 data indicated an HIV prevalence of 27 percent in urban areas, 8.5 percent in semi-urban areas, and 2.2 percent in rural areas. After the April 1994 genocide, a new HIV surveillance system was set up in 1996 with ten sentinel sites. The data gathered that year indicated even higher infection rates: 27 percent among urban residents, 13 percent among semi-urban residents, and 6.9 percent among rural residents. A 1997 study sampled 4,800 people and provided an HIV prevalence rate of 11.1 percent (10.8 percent for men and 11.3 percent for women). In 2002, the national sentinel surveillance system was expanded, increasing the number of sites to 24, thus providing more precise data than before. The 2002 data showed that prevalence varied between 2.6 percent and 3.6 percent in rural areas and between 7.0 percent and 8.5 percent in urban areas. These prevalence rates do not differ significantly from the 2003 rates, which were between 2.7 percent and 3.6 percent in rural areas and between 6.9 percent and 8.3 percent in urban areas. HIV testing was included in the 2005 RDHS-III to estimate HIV prevalence using a nationally- representative sample of men and women. In addition, because the test results are linked to socio- demographic and behavioral data on the individuals interviewed, the RDHS-III allows the identification of factors associated with HIV prevalence. 15.1 HIV TESTING PROTOCOL The third Rwandan Demographic and Health Survey (RDHS-III) was the first to include a blood test to determine HIV prevalence in the general population. Funded by the Ministry of Finance and Economic Planning, the survey was conducted by the Direction de la Statistique (currently, the Institut National de la Statistique du Rwanda or INSR) with the technical assistance of ORC Macro, the U.S. organization in charge of the international Demographic and Health Survey program. The purpose of including the HIV testing in a population-based survey was to estimate HIV prevalence among women age 15 to 49 and among men age 15 to 59. The protocol for HIV testing was based on the “anonymous-linked” protocol developed by DHS and approved by the Institutional Review Board at ORC Macro, as well as the National Committee on Ethics of Rwanda. Since the HIV tests were completely anonymous, it was not possible to inform the respondents of their results. However, a voucher listing the 77 voluntary testing facilities (VCTs) operating throughout the nation was distributed to all eligible respondents, whether or not they had agreed to be tested for HIV. The testing centers would offer free counseling and HIV testing to anyone presenting the card. 226 | HIV Prevalence and Associated Factors The blood drawing for the HIV test was conducted among the 5,322 households selected for the male survey. Blood was drawn from men age 15 to 59 and women 15 to 49 who had voluntarily accepted the testing. Training of the survey interviewers Those responsible for the survey at the INSR, in collaboration with the technical team, recruited 95 people to collect the data during the main survey. Among these, 63 were medically qualified to draw blood. A four-week training workshop covering all aspects of the survey was conducted from January 21 to February 21, 2005. The program included a detailed explanation of the survey questionnaire contents, a presentation of interviewing techniques, instructions on how to fill out the questionnaire form, and training in taking anthropometric measurements. The training included lectures and practice interviews, both in class and in the field. Each trainee conducted at least five interviews during the workshop. A special one-week training session was organized for the people in charge of administering the tests for anemia and HIV. The training dealt with the procedure for obtaining voluntary consent, techniques used for blood drawing, the use of the HemoCue for anemia, referral procedures for those needing treatment for anemia, and referral procedures for VCT facilities. In addition, the session included procedures for handling and storing blood specimens on filter paper prior to their transport to the Laboratoire National de Référence (LNR), as well as the procedure for the disposal of bio-contaminated waste. The training also included a detailed presentation on the transfer of dried blood spots from the field to the laboratory. All the office and laboratory staff involved in testing participated in this phase of the training, as did all the field workers. The LNR agents were trained in how to record the test results and how to return these to the INSR once the survey activities were completed. One-half day was devoted to informing the RDHS-III personnel about the AIDS epidemic, including the means of prevention and the reasons for including the HIV test in the survey. Issues of stigmatization, misconceptions, and confidentiality were touched on during the training. An additional day was devoted to training the team leaders and field editors how to observe field interviews, edit questionnaires that had been filled out, and monitor the blood draw. At the end of the workshop, the field workers were divided into 15 teams, each consisting of a team leader, a field editor, three female interviewers (one of whom was a health technician) and one male interviewer (also a health technician). Data collection The data collection began on February 28, 2005 in the districts of the city of Kigali. Starting in the capital city allowed close monitoring of the teams before they continued the survey in the other regions of the country. After two weeks, all the teams—with the exception of two assigned to work in Kigali—were sent out to their respective districts. The data collection was completed on July 13, 2005. The blood used for HIV testing was obtained using the same finger prick as the anemia test and was collected on filter paper. A label with a bar code was attached to each paper. A second label with the same bar code was attached to the corresponding household questionnaire next to the line indicating the consent of the person tested. A third label with the same bar code was attached to the laboratory transmission slip. The specimens were dried for a minimum of 24 hours in a box containing humidity- absorbing desiccants. The next day each specimen was closed in a Ziploc bag with desiccants and a humidity indicator. The individual bags preserved the specimens until they could be transferred to the INSR in Kigali where they were verified and recorded before being transferred to the LNR. HIV Prevalence and Associated Factors | 227 HIV testing procedure The LNR was responsible for testing the dried blood spot specimens for HIV antibodies and for the delivery of results to the INSR. The algorithm that was used consisted of testing the specimens with ELISA 1 (Vironostika HIV Uniform II Plus 0 Version 3.3 from Biomerieux BV). This ELISA is the third generation of Sandwich type, which allows the detection of HIV-1, HIV-2, and HIV-1 Group 0. As a highly sensitive detection system, it was used in the first round of testing. Any specimen that presented an optic density (OD) less than the threshold value (T) was considered negative; all above the threshold were considered positive. The specimens found to be positive using ELISA 1 (Vironostika), as well as 10 percent of the negative samples, were subsequently analyzed with a second ELISA test: Enzygnost Anti-HIV ½ Plus from Dade Behring AG. This ELISA 2 test was used as confirmation because of its specificity in detecting HIV-1 and HIV-2. The antigens used were recombinant proteins. The results were automatically calculated using the ELISA program developed by Dynex Technologies. All specimens that tested positive using both ELISA 1 and 2 were declared positive. Any discordant results underwent a third test: HIV Blot 2.2. Data processing and delivery of results The LNR was provided with the CSPro program developed by ORC Macro and designed especially for the HIV-testing algorithm. As data were entered, the program automatically calculated all entries (number of blood tests, number of positives and negatives according to the different test kits used). Throughout the survey, the LNR furnished the INSR and ORC Macro with aggregated results to monitor the testing process and to detect any abnormal results. Each specimen transferred to the LNR was identified by a bar code and only this code was entered into the CSPro program with the test results. This confidential file remained the responsibility of the LNR until the end of the survey. Once data entry was complete, and the data files at the INSR had been cleaned and the data had been weighted, a data file was prepared at LNR containing only weighting factors of the respondents (gender, age, residence, marriage status) and was compared with the file at the INSR to verify coherence of the two data banks. to guarantee anonymity, any information allowing identification of the respondents (by cluster or household number) were deleted before merging the two files. The files were then merged to calculate the sociodemographic and behavioral indicators of HIV prevalence. Internal quality control Each blood test was recorded in the lab workers’ notebooks. Each entry included the date, the name of the technician conducting the test, and the test used with its lot number and expiration date. The LNR used its usual internal control mechanism to monitor the testing: each slide was incorporated into an aliquot (HIV+ or HIV-) and frozen to -70 degrees centigrade. Of the 10 percent negative specimens that were tested, 100 percent proved negative. External quality control Since 2001 the LNR has participated in a program of external quality control. This consists of putting HIV antibodies on a coded panel that is sent to an external monitor. The monitoring for the 228 | HIV Prevalence and Associated Factors RDHS-III specimens was done by the Centers for Disease Control and Prevention (CDC) in Atlanta; 100 percent of the negative control samples tested negative. 15.2 COVERAGE OF HIV TESTING Table 15.1 shows coverage rates for the HIV test among women age 15-49 and men age 15-59 grouped by residence (province and urban-rural), along with the reasons for which the blood draw was not conducted. Overall, 96.5 percent of eligible respondents provided blood for the HIV test, 1.5 percent refused to have blood drawn, and 1.7 percent were absent, the great majority of whom (1.5 percent) were also absent during the interview. The results showed higher coverage in rural areas than in urban areas (97.4 percent versus 93.6 percent). The higher coverage level among rural residents holds true for both sexes: in rural areas 97.7 percent of women and 97.1 percent of men accepted being tested while in urban areas 95.8 percent of women and 91.0 percent of men were tested. Table 15.1 Coverage of HIV testing by residence and province Percent distribution of women age 15-49 and men age 15-59 eligible for HIV testing by testing status, according to residence and province (unweighted), Rwanda 2005 Residence Province Testing status Urban Rural Kigali city South West North East Total Women 15-49 Tested 95.8 97.7 94.2 98.4 96.2 96.7 99.4 97.3 Refused 2.3 0.7 3.1 0.5 1.6 1.0 0.1 1.1 Absent for testing 1.5 1.4 2.0 0.8 1.9 2.3 0.6 1.4 Interviewed in survey 0.2 0.1 0.3 0.1 0.0 0.2 0.3 0.2 Not interviewed 1.3 1.3 1.7 0.7 1.9 2.1 0.2 1.3 Other/missing 0.4 0.2 0.7 0.3 0.3 0.0 0.0 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Unweighted number 1,339 4,498 687 1,431 1,533 938 1,248 5,837 Men 15-59 Tested 91.0 97.1 87.3 96.7 95.8 96.6 98.7 95.6 Refused 5.4 0.8 7.9 1.1 1.5 0.8 0.4 1.9 Absent for testing 2.8 1.8 3.8 1.6 2.4 2.5 0.7 2.1 Interviewed in survey 0.4 0.2 0.6 0.2 0.1 0.3 0.2 0.2 Not interviewed 2.4 1.7 3.2 1.4 2.4 2.2 0.6 1.8 Other/missing 0.8 0.3 1.1 0.6 0.3 0.1 0.2 0.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Unweighted number 1,183 3,776 659 1,180 1,274 769 1,077 4,959 Total Tested 93.6 97.4 90.8 97.6 96.0 96.7 99.1 96.5 Refused 3.8 0.7 5.4 0.8 1.6 0.9 0.2 1.5 Absent for testing 2.1 1.6 2.9 1.1 2.1 2.4 0.6 1.7 Interviewed in survey 0.3 0.1 0.4 0.1 0.0 0.2 0.3 0.2 Not interviewed 1.8 1.5 2.5 1.0 2.1 2.2 0.4 1.5 Other/missing 0.6 0.3 0.9 0.5 0.3 0.1 0.1 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Unweighted number 2,522 8,274 1,346 2,611 2,807 1,707 2,325 10,796 HIV Prevalence and Associated Factors | 229 The following four categories explain the cases in which the blood draw for the test did not take place. • Those who refused to have blood drawn (in total, 1.5 percent). Urban male residents account for the highest level of refusal (5.4 percent) while rural female residents account for the lowest level (0.7 percent). • Those who responded to the survey but were not at home when blood was drawn: 0.3 percent of urban residents and 0.1 percent of rural residents. In all, 0.2 percent of respondents were absent during the blood draw. • Those who were not at home for the survey interview or the blood test: 1.8 percent in urban areas and 1.5 percent in rural areas. • Those who were not tested for other reasons (such as inability to give informed consent or technical difficulties in drawing blood): 0.6 percent among urban residents and 0.3 percent among rural residents. Table 15.2 shows coverage rates of the HIV test according to age, education level, and household wealth quintile. Overall, these results show few significant differences in the HIV test coverage by sociodemographic characteristics, for either women or men. The proportion of women who participated in the HIV testing varied from 96.1 percent among those age 15 to 19 to 98.8 percent among those age 40 to 44. There were minimal differences according to household wealth; these varied from 94.5 percent among women in wealthier households to 98.1 percent among women in the second quintile. Education levels showed little difference in participation, varying from 96.3 percent among women having at least secondary education to 97.5 percent among those who attended only primary school. The coverage rates among men ranged from 92.2 percent among those age 30-34 to 98.5 percent among those age 50-54. As with women, men in the wealthiest households have the lowest participation rates (90.9 percent) while men in the poorest households have the highest rates (97.9 percent). Distributed by the level of education, coverage among male respondents shows a clear difference from female respondents, although the difference is minimal (92.4 percent among those with secondary education or higher and 96.6 percent among those with only primary school). Tables A.5 and A.6 in Appendix A show participation rates distributed according to background characteristics of the respondents. Overall, analysis of these rates shows no systematic relation between participation in the test and variables associated with higher risk of HIV infection. These results indicate that the estimated prevalence rates from the 2005 RDHS-III provide an unbiased measure of HIV prevalence in the general population. 230 | HIV Prevalence and Associated Factors Table 15.2 Coverage of HIV testing by background characteristics Percent distribution of women age 15-49 and men age 15-59 eligible for HIV testing by testing status, according to background characteristics (unweighted), Rwanda 2005 Tested Refused Absent for testing Other/missing Background characteristic Interviewed in survey Not interviewed Interviewed in survey Not interviewed Interviewed in survey Not interviewed Interviewed in survey Not interviewed Total Unweighted number WOMEN Age 15-19 96.1 0.1 1.2 0.1 0.1 2.2 0.1 0.1 100.0 1,372 20-24 96.4 0.2 0.7 0.3 0.4 1.6 0.2 0.3 100.0 1,178 25-29 96.3 0.7 1.4 0.5 0.1 0.7 0.2 0.1 100.0 870 30-34 98.5 0.1 0.7 0.2 0.0 0.1 0.0 0.2 100.0 824 35-39 96.7 0.2 1.6 0.2 0.2 1.2 0.0 0.0 100.0 570 40-44 98.8 0.2 0.0 0.2 0.0 0.9 0.0 0.0 100.0 561 45-49 98.1 0.4 0.0 0.0 0.0 1.3 0.0 0.2 100.0 462 Education No education 96.6 0.5 0.5 0.2 0.2 2.1 0.0 0.0 100.0 1,312 Primary 97.5 0.2 0.8 0.1 0.2 1.1 0.1 0.2 100.0 3,298 Secondary or higher 96.3 0.2 1.5 0.5 0.2 0.9 0.2 0.2 100.0 1,227 Wealth quintile Lowest 97.8 0.2 0.3 0.2 0.0 1.3 0.1 0.2 100.0 1,178 Second 98.1 0.4 0.2 0.0 0.1 1.1 0.1 0.1 100.0 1,138 Middle 97.5 0.1 0.7 0.3 0.2 1.3 0.0 0.0 100.0 1,031 Fourth 97.7 0.1 0.5 0.1 0.1 1.3 0.2 0.1 100.0 1,156 Highest 94.5 0.4 2.4 0.4 0.4 1.4 0.1 0.3 100.0 1,334 Total 97.0 0.2 0.9 0.2 0.2 1.3 0.1 0.1 100.0 5,837 MEN Age 15-19 95.6 0.2 1.1 0.4 0.0 2.4 0.1 0.3 100.0 1,109 20-24 95.3 0.3 1.8 0.0 0.3 1.8 0.1 0.3 100.0 982 25-29 94.9 0.6 1.5 0.6 0.4 1.5 0.0 0.4 100.0 668 30-34 92.2 0.4 3.3 0.7 0.7 2.2 0.0 0.4 100.0 540 35-39 95.3 0.2 1.6 0.5 0.0 2.0 0.2 0.2 100.0 443 40-44 95.3 0.0 1.2 0.5 0.2 2.1 0.0 0.7 100.0 422 45-49 97.7 0.3 0.8 0.3 0.0 0.8 0.0 0.3 100.0 384 50-54 98.5 0.0 0.4 0.0 0.0 0.8 0.4 0.0 100.0 265 55-59 95.9 0.7 2.1 0.0 0.0 0.7 0.0 0.7 100.0 146 Education No education 95.1 0.5 1.1 0.1 0.0 2.8 0.0 0.5 100.0 852 Primary 96.6 0.2 1.1 0.2 0.2 1.4 0.0 0.3 100.0 2,963 Secondary or higher 92.4 0.4 3.0 0.9 0.4 2.3 0.3 0.3 100.0 1,144 Wealth quintile Lowest 97.9 0.0 0.5 0.1 0.0 1.3 0.2 0.0 100.0 838 Second 96.3 0.1 0.5 0.1 0.1 2.0 0.0 0.8 100.0 845 Middle 96.4 0.3 0.7 0.4 0.3 1.7 0.0 0.1 100.0 951 Fourth 97.1 0.3 1.1 0.1 0.0 1.1 0.0 0.4 100.0 1,031 Highest 90.9 0.5 3.9 0.8 0.5 2.8 0.2 0.4 100.0 1,294 Total 95.3 0.3 1.6 0.3 0.2 1.8 0.1 0.3 100.0 4,959 HIV Prevalence and Associated Factors | 231 15.3 HIV PREVALENCE 15.3.1 HIV Prevalence Distribution According to Sociodemographic Variables According to the 2005 RDHS-III, HIV prevalence in the Rwandan population age 15-49 is 3 percent (Table 15.3). HIV prevalence among women age 15-49 (3.6 percent) is higher than that of men in the same age group (2.3 percent). The infection ratio between women and men is therefore equal to 1.6, which means that 160 women are infected for every 100 men. Table 15.3 HIV prevalence by age Percentage HIV positive among women age 15-49 and men age 15-59 by age, Rwanda 2005 Women 15-49 Men 15-59 Total Age Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number 15-19 0.6 1,316 0.4 1,087 0.5 2,403 20-24 2.5 1,142 0.5 939 1.6 2,080 25-29 3.4 833 2.1 628 2.9 1,461 30-34 5.9 806 4.2 497 5.2 1,303 35-39 6.9 540 2.3 432 4.8 972 40-44 6.3 554 7.1 401 6.6 955 45-49 4.1 464 5.3 378 4.6 842 50-54 na na 1.7 259 na na 55-59 na na 0.8 143 na na Total 15-49 3.6 5,656 2.3 4,361 3.0 10,016 Total 15-59 na na 2.2 4,763 na na na = Not applicable Figure 15.1 shows that for both women and men, HIV prevalence increases with age. However, the highest prevalence among women is in the 35-39 age group (6.9 percent), whereas among men it is in the 40-44 age group (7.1 percent). Up until age 35-39, the proportion of infected women is higher than the proportion of infected men. Afterward, this pattern is reversed (at age 45-49, 5.3 percent of men are positive, compared with 4.1 percent of women). Figure 15.1 HIV Prevalence by Sex and Age RDHS 2005 ) )) ) ) ) ) ) ) * * * * * * * 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 Age 0 1 2 3 4 5 6 7 8 9 10 Percent Women Men* ) 232 | HIV Prevalence and Associated Factors Table 15.4 shows HIV prevalence according to sociodemographic characteristics. The prevalence rate is higher in urban areas than in rural areas (7.3 percent versus 2.2 percent). The differential is seen for both women and men: 8.6 percent versus 2.6 percent for women and 5.8 percent versus 1.6 percent for men. Table 15.4 HIV prevalence by background characteristics Percentage HIV positive among women and men age 15-49 who were tested, by background characteristics, Rwanda 2005 Women Men Total Background characteristic Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number Residence Urban 8.6 946 5.8 774 7.3 1,720 Rural 2.6 4,710 1.6 3,587 2.2 8,297 Province Kigali city 8.0 556 5.2 487 6.7 1,043 South 3.1 1,501 2.0 1,126 2.7 2,627 West 3.7 1,406 2.4 1,051 3.2 2,458 North 2.6 1,019 1.1 773 2.0 1,792 East 2.9 1,173 2.1 923 2.5 2,096 Education No education 3.3 1,278 3.0 716 3.2 1,994 Primary 2.8 3,251 1.8 2,668 2.3 5,919 Secondary or higher 6.4 1,127 3.2 977 4.9 2,104 Employment Currently working 4.0 3,386 2.7 2,209 3.5 5,594 Not currently working 3.0 2,245 1.8 2,127 2.4 4,371 Wealth quintile Lowest 2.6 1,204 1.3 791 2.1 1,994 Second 2.2 1,193 1.7 788 2.0 1,981 Middle 3.6 1,042 2.0 881 2.9 1,923 Fourth 3.4 1,110 2.1 892 2.8 2,001 Highest 6.5 1,108 4.1 1,010 5.4 2,117 Religion Catholic 3.9 2,574 2.4 2,201 3.2 4,775 Protestant 3.3 2,123 2.3 1,423 2.9 3,546 Adventist 2.5 711 2.1 531 2.3 1,242 Muslim 11.4 102 1.6 87 6.9 188 Other/missing 3.2 146 2.9 119 3.1 265 Total 3.6 5,656 2.3 4,361 3.0 10,016 By province, HIV prevalence is higher in the city of Kigali than in the rest of the country. In Kigali, 8.0 percent of women are seropositive, while prevalence ranges from 2.6 percent in North province to 3.7 percent in West province. Among men, the prevalence in Kigali is estimated at 5.2 percent, while in the interior, it ranges from 1.1 percent in North province to 2.4 percent in West province. Results by level of education show higher prevalence among women with at least secondary education (6.4 percent) compared with those with primary education (2.8 percent). Among men, as with women, the lowest prevalence is found among men who attended primary school (1.8 percent). However, the difference between men with no schooling and those with secondary or higher education is insignificant (3.0 percent versus 3.2 percent). HIV infection rates also vary by employment status. With women as with men, those who were employed at the time of the survey showed a slightly higher HIV Prevalence and Associated Factors | 233 prevalence than those who were not (4.0 percent versus 3.0 percent for women and 2.7 percent versus 1.8 percent for men). Looking at household wealth, the highest HIV prevalence is found in the wealthiest quintile: 6.5 percent for women and 4.1 percent for men. By religion, prevalence ranges from 2.5 percent among Adventist women to 11.4 percent among Muslim women. For men, the differences are smaller, varying from 1.6 percent among Muslims to 2.4 percent among Catholics. Table 15.5 shows HIV prevalence with 95 percent confidence intervals for certain background characteristics. Table 15.5 HIV prevalence and confidence intervals Percentage HIV positive among women and men age 15-49 and 95 % confidence intervals, by age and residence, Rwanda 2005 Women Men Total Background characteristic -2 SD Value +2 SD -2 SD Value +2 SD -2 SD Value +2 SD Age 15-19 0,2 0,6 1,1 0,0 0,4 0,8 0,2 0,5 0,9 20-24 1,6 2,5 3,4 0,0 0,5 0,9 1,0 1,6 2,1 25-29 2,1 3,4 4,7 1,0 2,1 3,3 2,0 2,9 3,7 30-34 4,3 5,9 7,5 2,2 4,2 6,2 3,9 5,2 6,5 35-39 4,8 6,9 9,0 0,9 2,3 3,7 3,4 4,8 6,3 40-44 4,3 6,3 8,4 4,4 7,1 9,7 5,0 6,6 8,2 45-49 2,1 4,1 6,1 3,0 5,3 7,6 3,1 4,6 6,2 Residence Urban 6,9 8,6 10,3 4,2 5,8 7,3 6,0 7,3 8,6 Rural 2,1 2,6 3,1 1,1 1,6 2,1 1,8 2,2 2,6 Total 3,1 3,6 4,1 1,8 2,3 2,8 2,6 3,0 3,5 15.3.2 HIV Prevalence by Demographic Variables There are large variations in HIV prevalence by marriage status (Table 15.6). A total of 1.6 percent of never-married women are HIV positive, versus 2.8 percent of married women. Rates rise to 10.9 percent among divorced or separated women and 15.9 percent among widows. Similarly, divorced men show higher prevalence than married men (5.1 percent of divorced men versus 3.5 percent of married men and 0.9 percent of never-married men). Results by type of union indicate higher prevalence among women in polygamous unions (4.7 percent) than among those in monogamous unions (2.5 percent). Among men, HIV prevalence is higher in monogamous unions (3.5 percent) than polygamous unions (2.3 percent). HIV prevalence is slightly higher among women who were not pregnant or were unsure at the time of the survey (3.7 percent) than among women who were pregnant (2.2 percent). The data did not suggest a correlation between HIV prevalence and the number of times respondents slept away from home during the past 12 months. 234 | HIV Prevalence and Associated Factors Table 15.6 HIV prevalence by sociodemographic characteristics Percentage HIV positive among women and men age 15-49 who were tested, by sociodemographic characteristics, Rwanda 2005. Women Men Total Sociodemographic characteristic Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number Marital status Never in union 1.6 2,179 0.9 2,164 1.2 4,343 Ever had sex 4.8 421 2.1 826 3.0 1,247 Never had sex 0.8 1,758 0.2 1,338 0.5 3,096 Currently in union 2.8 2,716 3.5 2,091 3.1 4,807 Widowed 15.9 227 * 21 15.8 248 Divorced/separated 10.9 519 5.1 73 10.2 592 Type of union In polygynous union 4.7 325 2.3 101 4.2 427 Not in polygynous union 2.5 2,368 3.5 1,987 3.0 4,355 Not currently in union 4.3 2,925 1.2 2,257 3.0 5,183 Currently pregnant Pregnant 2.2 431 na na na na Not pregnant/not sure 3.7 5,224 na na na na Circumcision status Circumcised na na 3.8 418 na na Not circumcised na na 2.1 3,909 na na Number of times slept away None 3.2 4,378 2.2 3,225 2.8 7,603 1-2 4.6 946 2.2 662 3.6 1,608 3-4 6.6 214 3.0 237 4.7 451 5+ 3.3 97 2.4 208 2.7 305 Away for more than one month Away for more than 1 month 3.6 216 1.9 342 2.6 559 Away always for < 1 month 5.0 1,039 2.7 738 4.0 1,776 Never away 3.2 4,378 2.2 3,225 2.8 7,603 Birth in the past 3 years No birth 3.9 3,364 na na na na Birth and antenatal care 2.8 2,162 na na na na Birth, no antenatal care 8.8 130 na na na na Total1 3.6 5,656 2.3 4,361 3.0 10,016 Note: An asterisk indicates than a figure is based on fewer than 25 unweighted cases and has been suppressed. na = Not applicable 1 Includes women and men with missing information 15.3.3 HIV Prevalence by Sexual Behavior Characteristics Overall, HIV prevalence among respondents who have ever had sexual intercourse is estimated at 4.2 percent; 3.3 percent among women and 4.9 percent among men (Table 15.7). There is no clear correlation between HIV prevalence and age of first sexual intercourse, whether respondent is male or female. Those who had sex before age 16 have the lowest prevalence (4.2 percent for women and 1.4 percent for men) and those whose first intercourse was at age 16-17 have the highest prevalence (5.2 percent for women and 4.6 percent for men). HIV Prevalence and Associated Factors | 235 Table 15.7 HIV prevalence by sexual behavior characteristics Percentage HIV positive among women and men age 15-49 who ever had sex and were tested for HIV, by sexual behavior characteristics, Rwanda 2005. Women Men Total Sexual behavior characteristic Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number Age at first sex < 15 4.2 423 1.4 549 2.6 973 15-17 5.2 680 4.6 400 5.0 1,080 18-19 4.7 991 3.9 608 4.4 1,600 20+ 4.9 1,675 3.2 1,442 4.1 3,117 Missing 6.4 128 * 22 7.0 150 Higher-risk sex1 in past 12 months Had higher-risk sex 8.2 251 2.7 379 4.9 630 Had sex, not higher-risk sex 3.0 2,650 3.5 1,980 3.2 4,630 No sex in past 12 months 8.9 997 2.9 663 6.5 1,660 Number of lifetime sexual partners 1 3.0 2,694 1.2 1,154 2.4 3,848 2 8.1 835 2.9 768 5.6 1,603 3-4 12.1 302 4.2 750 6.4 1,052 5-9 9.1 39 7.8 233 8.0 272 10+ * 9 11.7 97 11.7 106 Number of partners in past 12 months 0 8.9 997 2.9 663 6.5 1,660 1 3.5 2,882 3.3 2,238 3.4 5120 2+ * 19 4.1 121 4.6 140 Number of higher-risk sexual partners in past 12 months 0 4.6 3,647 3.3 2,643 4.1 6,290 1 8.1 239 2.6 356 4.8 595 2+ * 13 * 22 (7.1) 35 Paid for sex in past 12 months Yes na na (6.3) 38 na na No na na 3.2 2,984 na na Any condom use Ever used condom 15.5 157 7.5 543 9.3 700 Never used condom 4.4 3,741 2.3 2,479 3.6 6,220 Condom use at last sex in past 12 months Used a condom 23.4 88 12.8 140 16.9 228 Did not use a condom 2.9 2,813 2.8 2,219 2.8 5,032 Condom use at last higher-risk sex in past 12 months Used a condom 15.9 56 4.2 142 7.5 198 Did not use a condom 6.0 195 1.7 236 3.7 431 Total 4.9 3,898 3.3 3,022 4.2 6,920 Note: An asterisk indicates that an figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. Total includes respondents with missing information on sexual behavior. na = Not applicable 1 Sex with a person who is neither married to nor lives with the respondent Table 15.7 also shows prevalence rates by whether the respondent engaged in higher-risk sexual intercourse. Paradoxically, it is not only women who have engaged in higher-risk sex, but also those who had no intercourse in the past 12 months that prevalence rates are the highest (8.2 percent and 8.9 percent, respectively). Among male respondents, prevalence is slightly higher among those who engaged in sex but not higher-risk sex (3.5 percent) in the past 12 months, compared with men who had higher-risk sex or no sex at all (less than 3 percent). 236 | HIV Prevalence and Associated Factors Generally, HIV prevalence increases with increasing number of lifetime sexual partners. Thus, prevalence varies from 1.2 percent for men who have had only one partner during their life to 4.2 percent for those who have had 3-4 partners, to 11.7 percent for those who have had at least 10 partners. For women, prevalence increases from 3.0 percent for those who have had one sexual partner during their life to 12.1 percent to those who have had 3-4 partners. Paradoxically, HIV prevalence is 8.9 percent among women who have had no sexual partner in the past 12 months and 3.5 percent among those who have had only one partner. In contrast, the prevalence among men who have had two partners during the past 12 months is 4.1 percent, higher than among those who have had a single partner (3.3 percent) or those who have had no partners (2.9 percent). Women who engage in higher-risk sex tend to have higher HIV prevalence: 8.1 percent among women who have had a higher-risk partner during the past 12 months and 4.6 percent among those who have not. Regarding condom use during the past year—whether at the last sexual intercourse or at the last higher-risk sexual intercourse—it can be seen that HIV prevalence is higher among male and female condom users than among those who have not used condoms. It is difficult to establish the exact relationship between condom use and HIV. Condoms could be used by those who are HIV negative to protect themselves from the disease, but they could also be used by those who are seropositive to protect their partners. It is the latter pattern that emerges from the RDHS-III data. 15.3.4 HIV Prevalence among Youth Table 15.8 shows HIV prevalence among youth age 15-24 by sociodemographic and sexual behavioral characteristics. Prevalence among youth gives an indication of the level of recent infections and is an indirect estimate of the number of new cases. HIV prevalence among youth age 15-24 is estimated at 1.0 percent. This figure varies from 1.5 percent among women to 0.4 percent among men, which gives a ratio of infection of 3.8 between women and men. In other words, 380 women in this age group are infected for every 100 men. This ratio is 2.4 times higher than that of the combined 15-49 age group. Overall, the results in Table 15.8 indicate an increase in seroprevalence by age up through 20-22 years, the age group with the highest rate (1.7 percent). Subsequently, rates begin to decrease among the 23-24 age group (1.4 percent). Whatever the age group, prevalence among women is always higher than prevalence among men. It increases less rapidly among young men and never surpasses 1 percent; the highest level is among men age 18-19 (0.8 percent). Among women, prevalence is highest in the 20-22 age group (2.7 percent). The ratio is particularly high in this age group (6.8). HIV prevalence is higher in urban areas than rural areas (2.7 percent versus 1.7 percent). The differences are seen for both sexes. Across regions, seroprevalence among young women ranges from 0.5 percent in the South province to 4.2 percent in the city of Kigali. For young men, HIV prevalence is the highest in the city of Kigali (1.4 percent). Note that in the North province, seroprevalence is higher among young men (1.1 percent) than young women in the same age group (0.8 percent). By marital status, the highest prevalence is among women who are separated, divorced, or widowed (3.8 percent versus 1.2 percent for married women and 1.7 percent of never-married women). Noteworthy is the 1.6 percent of young never-married women who reported never having had sex but are nonetheless HIV positive. The finding indicates that they were infected by another means or they falsely reported not having had sex. HIV Prevalence and Associated Factors | 237 Table 15.8 HIV prevalence among young people Percentage HIV positive among women and men age 15-24 who were tested for HIV, by background characteristics, Rwanda 2005 Women Men Total Background characteristic Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number Age 15-17 0.3 826 0.2 691 0.3 1,517 18-19 1.2 490 0.8 396 1.0 887 20-22 2.7 720 0.4 611 1.7 1,331 23-24 2.2 421 0.5 328 1.4 749 Residence Urban 3.9 431 1.1 348 2.7 779 Rural 1.0 2,027 0.3 1,678 0.7 3,705 Province Kigali city 4.2 271 1.4 224 2.9 495 South 0.5 616 0.0 544 0.3 1,161 West 2.2 656 0.4 487 1.4 1,143 North 0.8 409 1.1 344 0.9 754 East 0.9 505 0.0 427 0.5 932 Marital status Never married 1.7 1,145 0.4 1,850 0.9 2,995 Ever had sex 1.9 429 0.9 621 1.3 1,050 Never had sex 1.6 716 0.2 1,229 0.7 1,946 Currently in union 1.2 1,251 0.5 166 1.1 1,418 Divorced/separated/widowed 3.8 61 * 10 3.8 71 Relative age of first sexual partner 10+ years older (10.4) 38 na na na na <10 years older/same age/younger/don’t know 1.4 2,419 na na na na Higher-risk intercourse in past 12 months Had higher-risk intercourse 3.3 108 1.5 171 2.2 279 Had intercourse, not higher risk 2.9 514 0.5 164 2.3 679 No sexual intercourse in last 12 months 1.0 1,835 0.3 1,691 0.7 3,526 Number of sexual partners in past 12 months 0 1.0 1,835 0.3 1,691 0.7 3,526 1 3.0 618 1.1 320 2.3 938 2+ * 5 * 15 * 20 Number of higher-risk partners in past 12 months 0 1.4 2,349 0.3 1,855 0.9 4,204 1 3.4 105 1.6 160 2.3 265 2+ * 4 * 11 * 15 Condom use Ever used a condom 7.3 51 2.1 163 3.4 214 Never used a condom 1.4 2,407 0.3 1,863 0.9 4,270 Condom use at last sex in past 12 months Used condom at last sex (11.7) 38 1.5 63 5.3 102 Did not use condom 2.4 584 0.9 272 1.9 856 Condom use at first sex Used a condom 5.9 54 1.4 92 3.0 146 Did not use a condom 1.4 2,403 0.4 1,934 1.0 4,338 Total 1.5 2,458 0.4 2,026 1.0 4,484 Note: An asterisk indicates that an figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. Seroprevalence is higher among respondents who engaged in higher-risk sex, especially women (3.3 percent versus 1.5 percent for men). At the same time, it should be noted that among both men and women, prevalence is higher for those using condoms than for those not using condoms; this difference is greater among young women (7.3 percent using condoms versus 1.4 percent not using condoms) than among young men (2.1 percent versus 0.3 percent). 238 | HIV Prevalence and Associated Factors 15.3.5 HIV Prevalence and Other Risk Factors Table 15.9 shows STI prevalence for women and men who have ever had and whether the respondent was tested for HIV before the survey. HIV prevalence is markedly higher among those who reported they already had an STI or symptoms of an STI. Among women who reported having an STI or symptoms of an STI in the past 12 months, HIV prevalence is 18.1 percent compared with 4.2 percent among those who reported that they did not have an STI or symptoms of an STI. For men who reported having an STI or symptoms of an STI in the past 12 months, prevalence is 9.9 percent versus 3.0 percent for those who have not had an STI or symptoms of an infection. Table 15.9 HIV prevalence by other characteristics Percentage HIV positive among women and men age 15-49 who ever had sex and who were tested, by other characteristics, Rwanda 2005 Women Men Total Characteristic Percentage HIV positive Number Percentage HIV positive Number Percentage HIV positive Number Sexually transmitted infection in past 12 months Had STI or STI symptom 18.1 204 9.9 83 15.7 287 No STI, no symptoms 4.2 3,646 3.0 2,907 3.7 6,554 HIV testing status Ever tested 9.5 665 4.3 826 6.6 1,492 Received results 10.0 603 4.2 759 6.8 1,362 Did not receive results 4.5 62 5.5 67 5.0 129 Never tested 3.9 2,690 2.8 2,183 3.4 4,873 Total1 4.9 3,898 3.3 3,022 4.2 6,920 1 Includes men and women with missing information. HIV prevalence among men who had never been tested previously for HIV is lower than among women (2.8 percent and 3.9 percent, respectively). Table 15.10 provides additional information about the relation between a previous HIV test and the respondent’s HIV status. This is useful for measuring infected respondents’ knowledge of their HIV status prior to the HIV test done during the RDHS-III. Among seropositive women, more than half (56.2 percent) did not know their status because they had never been tested for HIV before the survey. Among seropositive men, 66 percent did not know their status, either because they had never been tested (62 percent), or, if they had been, had never received their results (3.7 percent). Although the proportion of women and men who are aware of their HIV status is higher among seropositive respondents (31.3 percent and 31.6 percent, respectively) than among the HIV negative respondents (12.3 percent for women and 19.5 percent for men), a large proportion of those infected with HIV do not know they carry the virus and should therefore take the necessary measures to avoid transmitting the infection. HIV Prevalence and Associated Factors | 239 Table 15.10 Prior HIV testing by HIV status Percent distribution of women and men age 15-49 who were tested for HIV by whether they were tested prior to the survey, and whether they received the test results, according to HIV status (positive or negative), Rwanda 2005 Women Men Total HIV testing prior to the survey HIV positive HIV negative HIV positive HIV negative HIV positive HIV negative Previously tested and received results of last test 31.3 12.3 31.6 19.5 31.4 15.4 Previously tested and did not receive results of last test 0.0 0.0 3.7 1.4 1.2 0.6 Not previously tested 56.2 76.8 62.4 78.5 58.3 77.5 Missing 12.5 11.0 2.3 0.6 9.1 6.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Number 204 5,451 101 4,260 305 9,711 15.3.6 HIV Prevalence and Male Circumcision The RDHS-III included questions on whether men had been circumcised. These data can be used to examine possible relationships between HIV prevalence and male circum- cision. Among men age 15 to 59 who were tested for HIV, 9 percent had been circumcised. Table 15.11 indicates higher prevalence of HIV among circumcised males (3.5 percent) than among uncircumcised males (2.1 percent). This pattern is found for all sociodemographic variables, except urban residence, where preva- lence among circumcised men (5.0 percent) is slightly lower than among uncircumcised men (5.7 percent). 15.3.7 HIV Prevalence among Couples Table 15.12 presents HIV prevalence rates for couples living together, and in which both partners were tested. HIV status was obtained of both partners in a total of 2,231 couples. In 96.0 percent of couples both spouses were HIV negative and in 1.7 percent of couples both spouses were positive. The per- centage of couples in which both partners tested positive is especially high in urban areas (5.2 percent), in Kigali (4.5 percent) among couples having at least a secondary education (5.0 percent), and among couples in the wealthiest quintile (4.4 percent). Table 15.11 HIV prevalence by male circumcision Among men age 15-59 who were tested for HIV, the percentage HIV positive by whether circumcised, according to background characteristics, Rwanda 2005 Circumcised Uncircumcised Background characteristic Percentage HIV positive Number Percentage HIV positive Number Age 15-19 2.1 82 0.1 994 20-24 0.0 82 0.5 849 25-29 4.9 76 1.8 548 30-34 3.1 61 4.2 432 35-39 (0.0) 39 2.5 391 40-44 (19.6) 39 5.7 359 45-49 (2.0) 39 5.7 336 50-54 * 20 1.9 238 55-59 * 10 0.9 132 Education No education (5.6) 45 2.4 781 Primary 1.7 222 1.8 2,679 Secondary or higher 5.2 182 2.6 818 Religion Catholic 4.7 181 2.1 2,222 Protestant 4.3 142 2.1 1,406 Adventist 0.0 52 2.1 521 Muslim 2.2 65 (0.0) 25 Other/missing * 7 1.1 105 Residence Urban 5.0 210 5.7 609 Rural 2.2 239 1.5 3,669 Total 15-59 3.5 449 2.1 4,278 Note: An asterisk indicates that an figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 240 | HIV Prevalence and Associated Factors Table 15.12 HIV prevalence among couples Percent distribution of couples living in the same household, both of whom were tested for HIV, by HIV status, according to background characteristics, Rwanda 2005 Background characteristic Both HIV positive Man HIV positive, woman HIV negative Woman HIV positive, man HIV negative Both HIV negative Total Number Woman’s age 15-19 (0.0) (0.0) (0.0) (100.0) (100.0) 25 20-29 1.7 1.1 0.7 96.6 100.0 908 30-39 2.2 1.7 0.9 95.3 100.0 800 40-49 1.3 1.7 1.1 95.9 100.0 498 Man’s age 15-19 * * * * * 2 20-29 1.3 0.4 0.5 97.8 100.0 538 30-39 1.3 1.1 0.5 97.1 100.0 792 40-49 3.2 2.7 0.6 93.5 100.0 681 50-59 0.0 0.8 3.8 95.3 100.0 218 Age difference between partners Woman older 2.0 1.8 0.8 95.4 100.0 278 Same age/man older by 0-4 years 1.3 0.9 0.5 97.2 100.0 992 Man older by 5-9 years 1.4 1.4 0.7 96.5 100.0 586 Man older by 10-14 years 3.4 2.5 0.4 93.8 100.0 225 Man older by 15+ years 2.8 2.2 4.3 90.6 100.0 149 Marital status Married 1.7 1.2 0.7 96.4 100.0 1,363 Living together 1.8 1.7 1.1 95.4 100.0 868 Type of union Monogamous 1.7 1.3 0.7 96.2 100.0 1,995 Polygynous 1.4 1.3 2.0 95.3 100.0 223 Residence Urban 5.2 3.7 2.5 88.7 100.0 285 Rural 1.2 1.1 0.6 97.1 100.0 1,946 Province Kigali city 4.5 3.9 1.9 89.7 100.0 145 South 2.0 1.3 0.4 96.3 100.0 569 West 2.2 1.5 0.7 95.7 100.0 597 North 0.8 0.2 0.2 98.8 100.0 426 East 0.9 1.8 1.7 95.5 100.0 493 Woman’s education None 1.2 1.1 1.1 96.7 100.0 637 Primary 1.3 1.2 0.8 96.7 100.0 1,135 Secondary or higher 3.6 2.4 0.5 93.5 100.0 459 Man’s education None 0.3 0.2 0.8 98.7 100.0 400 Primary 1.6 2.0 0.9 95.5 100.0 1,555 Secondary or higher 5.0 0.0 1.0 94.1 100.0 216 Wealth quintile Lowest 0.9 0.8 0.2 98.2 100.0 449 Second 1.7 0.4 0.2 97.6 100.0 465 Middle 1.0 1.4 0.7 97.0 100.0 459 Fourth 1.3 2.0 1.5 95.2 100.0 499 Highest 4.4 2.7 1.8 91.1 100.0 359 Total1 1.7 1.4 0.8 96.0 100.0 2,231 Note: An asterisk indicates that an figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 1 Includes men and women with missing information In 2.2 percent of cases, only one of the partners was seropositive. In some of these discordant couples the woman was seropositive (0.8 percent), but in most cases it was the man who was seropositive (1.4 percent). HIV Prevalence and Associated Factors | 241 15.4 SENTINEL SURVEILLANCE SYSTEM AND RDHS-III In 2003, data from the national sentinel surveillance system indicated that HIV prevalence in Rwanda ranged from 6.9 percent to 8.3 percent in urban areas. This does not differ greatly from the rates observed in 2002, which ranged from 7.0 percent to 8.5 percent. These estimates are also close to the results found in the RDHS-III, where HIV prevalence in urban areas was 7.3 percent (with a 95 percent confidence interval between 6.0 and 8.6 percent). According to the national sentinel surveillance system, HIV prevalence in rural areas ranged from 2.6 percent to 3.6 percent in 2002 and from 2.7 percent to 3.6 percent in 2003. The RDHS-III estimate for HIV prevalence in rural areas is lower at 2.2 percent (95 percent confidence interval between 1.8 and 2.6 percent). The difference between the sentinel surveillance data and the RDHS-III data for rural residents can be explained primarily by the distribution of the sentinel sites in rural areas. Orphaned and Vulnerable Children | 243 ORPHANED AND VULNERABLE CHILDREN 16 One of the most devastating impacts of the HIV/AIDS epidemic is the dramatic increase in the number of children orphaned and made vulnerable by the death or chronic illness of one or more of the adults in their household. Deprived of the protection of these adults, such children are at increased risk of violence, exploitation, and other forms of abuse. With the spread of the HIV/AIDS epidemic, it is urgent that national strategies be adapted to strengthen governmental, family, and community capacities to support and protect these children. In June 2001, a special session of the United Nations General Assembly issued a Declaration of Commitment on HIV/AIDS (United Nations, 2001) signed by 189 member states that focused special attention on children orphaned and made vulnerable by HIV/AIDS. Numerous goals were established aimed at developing policies and strategies to support orphans by ensuring their access to education, proper nutrition, and health and social services. To assess progress in meeting this commitment, a series of indicators was developed to “monitor and evaluate the national response to orphans and children made vulnerable by HIV/AIDS” (UNICEF, 2005). The third DHS survey in Rwanda gathered data for use in estimating a number of these indicators. The results are presented in this chapter. 16.1 ORPHANHOOD AND CHILDREN’S LIVING ARRANGEMENTS Because the family is the primary safety net for children, any strategy aimed at protecting children must place a high priority on strengthening family capacities to care for children. It is therefore essential to identify orphaned children and find out whether those who have one or both parents living are living with either or both surviving parents. Table 16.1 presents these two types information for children under age 18, according to background characteristics. The data show that 60 percent of Rwandan children under the age of 18 live with both their parents. This proportion declines steadily with age, from a high of 82 percent at age 0-1 year and 63 percent at age 5 to 9 years, to a low of 38 percent at age 15 to 17 years. The results show practically no difference according to the child’s sex. The proportion of children living with their parents is higher in rural areas (61 percent) than in urban areas (54 percent). The lowest proportion of children living with both parents is in the City of Kigali (50 percent); the highest proportion is in the West and North provinces (64 percent for both). Twenty-three percent of children under age 18 live with their mother only, whether their father is alive (12 percent) or deceased (11 percent), and 3 percent live with their father only. Thirteen percent (13 percent) do not live with either parent. Overall, 21 percent of children under age 18 have lost their father and/or mother: 4 percent have lost both parents, 13 percent have lost their father, and 3 percent have lost their mother. Because a parent’s risk of dying increases with time, the proportion of children who have lost their father and/or mother increases significantly with the age of the child, from 2 percent at age 0 to 1 year, to 6 percent at age 2 to 4 years, to 16 percent at age 5 to 9 years. These proportions jump to very high levels among children age 10 to 14 (36 percent) and 15 to 17 (41 percent), largely due to the effects of the 1994 genocide. 244 | Orphaned and Vulnerable Children Table 16.1 Children’s living arrangements and orphanhood Percent distribution of de jure children under age 18 by children’s living arrangements and survival status of parents, and the percentage of children with one or both parents dead, according to background characteristics, Rwanda 2005 Living with mother but not with father Living with father but not with mother Not living with either parent Background characteristic Living with both parents Father alive Father dead Mother alive Mother dead Both alive Only mother alive Only father alive Both dead Missing infor- mation on father or mother Total Percent- age with one or both parents dead Number of children Age 0-1 81.7 14.9 1.5 0.3 0.1 0.8 0.1 0.0 0.1 0.6 100.0 1.9 3,411 2-4 74.1 14.2 3.7 0.9 0.5 4.5 0.4 0.3 0.7 0.9 100.0 5.5 4,711 5-9 62.8 11.4 9.8 1.3 1.3 6.6 1.3 1.8 1.9 1.8 100.0 16.4 7,168 10-14 44.0 10.0 18.6 1.3 2.8 7.0 2.7 4.0 7.4 2.1 100.0 35.9 6,341 15-17 38.0 9.3 20.7 1.1 3.0 7.7 3.0 4.4 9.5 3.3 100.0 41.0 3,235 Sex Male 60.2 11.8 11.1 1.3 1.5 5.2 1.5 2.1 3.6 1.7 100.0 20.0 12,406 Female 58.8 11.8 11.2 0.8 1.7 6.0 1.6 2.3 4.0 1.7 100.0 21.0 12,460 Residence Urban 53.8 13.0 12.6 1.2 1.1 5.9 1.8 2.8 5.1 2.6 100.0 23.8 3,548 Rural 60.5 11.6 10.9 1.1 1.7 5.6 1.5 2.1 3.6 1.6 100.0 20.0 21,319 Province City of Kigali 50.2 13.0 14.7 1.4 1.3 6.9 1.7 2.8 6.0 2.1 100.0 26.6 1,774 South 55.9 14.0 11.1 1.1 2.1 6.1 1.6 2.1 3.9 2.0 100.0 21.1 6,343 West 63.5 9.3 10.7 0.7 1.4 4.5 1.6 2.0 4.2 2.0 100.0 20.2 6,663 North 63.7 10.2 11.4 0.9 1.5 5.4 1.0 1.9 2.9 1.1 100.0 18.9 4,953 East 57.9 13.3 10.4 1.5 1.5 6.2 1.7 2.6 3.4 1.4 100.0 19.8 5,135 Total < 15 years 62.7 12.2 9.7 1.1 1.4 5.3 1.3 1.9 3.0 1.5 100.0 17.5 21,632 Total <18 years 59.5 11.8 11.2 1.1 1.6 5.6 1.5 2.2 3.8 1.7 100.0 20.5 24,867 Table 16.2 shows the percentage of children who are orphans and vulnerable children (OVC). Children are considered vulnerable (UNICEF, 2005) if they are under age 18 and: 1. Have lost one or both parents (21 percent); 2. One or both parents have been chronically ill for at least three of the past 12 months (8 percent); 3. Live in a household in which at least one adult age 18 to 59 has been chronically ill for at least three of the past 12 months (10 percent); 4. Live in a household in which at least one adult age 18 to 59 has died during the past 12 months after being chronically ill for at least three months (1 percent).1 Overall, 11 percent of children are considered vulnerable by virtue of being in categories 2, 3 and/or 4. When the data for orphans are added, 29 percent of children under age 18 are considered to be OVC. 1 Children deprived of family protection, i.e., living in an institution or on the street, are also considered vulnerable. However, these children are not included here because, by definition, they are not identifiable within the scope of a household survey. Orphaned and Vulnerable Children | 245 The proportion of OVC increases steadily with age, from 11 percent at age 0 to 1 year, to 25 percent at 5 to 9 years; at age 15 to 17 years, 48 percent of children are OVC. The proportion of OVC shows no variation by sex; however, OVC are more common in urban areas (33 percent) than in rural areas (28 percent). The highest proportion of OVC is in the City of Kigali (35 percent); the lowest proportion is in the North province (25 percent). The proportion of OVC is higher in the poorest households (33 percent) than in the richest households (28 percent). Table 16.2 Orphans and vulnerable children (OVC) Percentage of children under age 18 years who are orphans or made vulnerable due to illness among adult household members, according to background characteristics, Rwanda 2005 Percentage of children who are vulnerable because Background characteristic Percentage of children with one or both parents dead (orphans) Have a chronically ill parent1 Live in a household where at least 1 adult2 was chronically ill in the past 12 months Live in a household where at least 1 adult2 died in the past 12 months and had been chronically ill before he/she died Have a chronically ill parent OR live in a household where an adult was chronically ill OR died in the past 12 months (vulnerable) Percentage of children who are orphans and/or vulnerable (OVC) Number of children Age 0-1 1.9 8.0 8.6 0.3 9.1 10.7 3,411 2-4 5.5 8.3 8.6 0.4 9.6 14.4 4,711 5-9 16.4 8.0 9.1 0.5 10.2 24.7 7,168 10-14 35.9 8.6 10.3 0.8 11.9 43.0 6,341 15-17 41.0 9.2 11.3 1.1 13.3 48.2 3,235 Sex Male 20.0 8.5 9.6 0.6 10.9 28.2 12,406 Female 21.0 8.3 9.5 0.6 10.7 28.9 12,460 Residence Urban 23.8 9.7 11.7 0.8 13.4 33.1 3,548 Rural 20.0 8.2 9.2 0.6 10.4 27.8 21,319 Province City of Kigali 26.6 8.9 10.8 1.1 12.8 34.8 1,774 South 21.1 9.6 11.1 0.4 12.4 30.1 6,343 West 20.2 8.2 9.5 0.5 10.5 28.1 6,663 North 18.9 6.1 6.8 0.6 7.8 24.7 4,953 East 19.8 9.2 10.0 0.7 11.4 28.7 5,135 Wealth quintile Lowest 24.0 9.2 9.9 0.7 11.3 32.6 5,237 Second 20.6 7.3 8.3 0.4 9.0 26.6 4,871 Middle 20.0 8.7 10.1 0.5 11.4 28.4 5,143 Fourth 17.6 8.8 10.6 0.6 11.7 27.0 4,917 Highest 20.3 7.9 8.8 0.9 10.4 27.8 4,699 Total <15 years 17.5 8.3 9.3 0.5 10.4 25.6 21,632 Total <18 years 20.5 8.4 9.5 0.6 10.8 28.6 24,867 Note: Table is based on de jure household members, i.e., usual household members. Chronically ill means person was too sick to work or do normal activities. 1 Whether or not lives in same household as child. 2 Person age 18 to 59 years. 246 | Orphaned and Vulnerable Children 16.2 ACCESS TO ESSENTIAL SERVICES Access to education is considered an “essential service” and is included among the key components of national responses to guarantee OVC access to services on an equal basis with other children. To assess whether OVC are educationally disadvantaged in relation to other children, an indicator was devised to compare school attendance among OVC and non-OVC. The results are presented in Table 16.3 for children age 10 to 14, the age group in which school attendance is generally assumed for all children. The data show a clear relationship between parent survivorship and school attendance of children age 10 to 14. Whereas 91 percent of children whose parents are both alive and who are living with one of their parents attend school, only 75 percent of children who have lost both parents attend school. The ratio of school attendance for orphaned and nonorphaned children is less than 1 (0.82), indicating an educational disadvantage for orphans. The results also show that 82 percent of OVC attend school, compared with 89 percent of non-OVC. The ratio of OVC to non-OVC is 0.92. These results indicate that orphans and OVC are educationally disadvantaged in relation to other Rwandan children. Table 16.3 School attendance by survivorship of parents and by OVC status For children age 10-14, the percentage attending school by parental survival and by OVC status, and the ratios of the percentages attending school by parental survival and OVC status, according to background characteristics, Rwanda 2005 Percentage attending school by survivorship of parents Percentage attending school by OVC status Background characteristic Both parents deceased Number Both parents alive and living with at least one parent Number Ratio1 OVC Number Non OVC Number Ratio2 Sex Male 70.1 223 90.7 1,741 0.77 81.3 1,333 88.1 1,760 0.92 Female 78.8 245 91.6 1,770 0.86 83.3 1,394 90.1 1,854 0.92 Residence Urban 80.1 90 94.8 414 0.85 85.2 455 90.0 431 0.95 Rural 73.3 379 90.7 3,096 0.81 81.7 2,272 89.0 3,184 0.92 Province City of Kigali 76.0 48 97.5 178 0.78 82.8 238 90.8 198 0.91 South 71.6 136 89.3 848 0.80 78.4 732 85.8 892 0.91 West 74.5 136 92.2 1,024 0.81 82.3 701 91.9 1,037 0.90 North 75.4 57 91.0 727 0.83 83.4 471 89.5 739 0.93 East 78.1 92 90.5 735 0.86 86.2 584 88.4 748 0.97 Wealth quintile Lowest 75.8 72 87.6 709 0.86 81.8 629 87.6 701 0.93 Second 72.3 83 91.7 702 0.79 83.6 504 91.2 698 0.92 Middle 71.2 97 90.3 767 0.79 79.9 567 88.9 775 0.90 Fourth 71.2 96 90.9 718 0.78 83.8 504 87.6 757 0.96 Highest 81.1 120 96.0 615 0.84 83.1 523 90.3 685 0.92 Total 74.6 468 91.2 3,511 0.82 82.3 2,727 89.1 3,615 0.92 Note: Table is based on de jure household members, i.e., usual household members. 1 Ratio of the percentage with both parents deceased to the percentage with both parents alive and living with a parent 2 Ratio of the percentage OVC to the percentage not OVC Orphaned and Vulnerable Children | 247 16.3 STRENGTHENING FAMILY CAPACITIES TO SUPPORT AND PROTECT CHILDREN 16.3.1 Malnutrition The death or illness of a parent or other household member often leads to economic hardship for children and increases their risk of falling short of basic nutritional needs. Table 16.4 shows the proportion of children under age five who are underweight, for all children and by OVC status, according to background characteristics. The ratio of malnutrition among OVC to malnutrition among non-OVC is also shown. Table 16.4 Underweight orphans and vulnerable children Percentage of de facto children under age five years who are underweight, percentage of OVC and non-OVC who are underweight, and ratio of malnutrition (OVC to non-OVC), according to background characteristics, Rwanda 2005 Children under age 5 years OVC Non-OVC Background characteristic Percentage underweight1 Number of children Percentage underweight1 Number of OVC Percentage underweight1 Number of non-OVC Ratio2 Age < 1 year 11.1 774 9.8 73 11.3 701 0.87 1-2 years 30.9 1,652 27.8 180 31.3 1,472 0.89 3-4 years 18.6 1,388 18.5 211 18.6 1,177 0.99 Sex Male 22.9 1,878 20.2 220 23.2 1,658 0.87 Female 22.0 1,936 21.3 244 22.1 1,692 0.97 Residence Urban 16.0 536 21.2 80 15.1 456 1.40 Rural 23.5 3,278 20.7 385 23.8 2,894 0.87 Province City of Kigali 14.3 247 11.3 50 15.0 197 0.75 South 27.5 972 22.6 151 28.4 821 0.79 West 20.2 994 20.4 127 20.2 867 1.01 North 23.7 789 24.2 52 23.7 737 1.02 East 20.2 813 21.6 85 20.0 728 1.08 Wealth quintile Lowest 30.5 786 31.3 120 30.4 666 1.03 Second 25.8 815 27.7 87 25.5 729 1.09 Middle 22.2 798 15.6 79 22.9 719 0.68 Fourth 21.6 785 11.6 98 23.0 687 0.50 Highest 9.3 630 13.8 81 8.7 549 1.59 Total 22.4 3,814 20.8 464 22.6 3,350 0.92 Note: Table is based on de facto household members, persons who slept in household the night preceding the interview. 1 More than two standard deviations below the mean of the WHO/CDC/NCHS reference standard for weight-for-age. 2 Ratio of the percentage OVC to the percentage not OVC The results show that in Rwanda a little more than one in five children (22 percent) are underweight. This form of malnutrition affects 21 percent of OVC, compared with 23 percent of non- OVC. The ratio of OVC to non-OVC is less than 1 (0.92), indicating that non-OVC are slightly more undernourished than OVC. This result is confirmed regardless of child’s age or sex. However, OVC in the poorest households (ratio of 1.03), and also in the richest households (ratio of 1.59), appear to be less well-nourished than their non-OVC counterparts. Similarly, in urban areas, OVC appear to be less well- nourished than non-OVC (ratio of 1.4), while the opposite is true in rural areas (ratio of 0.87). 248 | Orphaned and Vulnerable Children 16.3.2 Early Sexual Intercourse Deprived of family protection, OVC are more exposed than other children to risky sexual encounters. It is therefore important to assess the “prevalence of early sexual activity among orphans and vulnerable children and other children between the age of 15 and 17” (UNICEF, 2005). Table 16.5 shows the proportion of youth who have had sexual intercourse before exact age 15, according to OVC status. This table also shows the ratio of OVC to non-OVC age 15 to 17 who have had sexual intercourse before exact age 15. Early sexual intercourse is much more frequent among men (14 percent) than women (5 percent). Moreover, it appears that early sexual intercourse is slightly more frequent among OVC (6 percent of girls, 15 percent of boys) than among non-OVC (5 percent of girls, 14 percent of boys); this difference translates into a ratio of greater than 1 (1.22 for girls, 1.08 for boys). Table 16.5 Sexual intercourse before age 15 among orphans and vulnerable children Percentage of children age 15-17 who had sexual intercourse before exact age 15, by OVC status and ratio of OVC to non-OVC sexual intercourse before age 15, Rwanda 2005 Women 15-17 Men 15-17 OVC status Percentage who had sexual intercourse before exact age 15 Number of women Percentage who had sexual intercourse before exact age 15 Number of men OVC 6.0 759 14.7 317 Non-OVC 4.9 829 13.6 369 Total 5.4 1,588 14.1 687 Ratio1 1.22 na 1.08 na Note: Table is based on de facto household members, persons who slept in household the night before the survey. na = Not applicable 1 Ratio of the percentage OVC to the percentage not OVC 16.3.3 Succession Planning Strengthening family capacities to support and protect orphans is essential. Identifying someone who will care for a child if his caregiver dies or falls ill is one way to ensure a better future for children. In Rwanda, 8 percent of women and men reported being primary caregivers to children under age 18, regardless of whether these children were their own (Table 16.6). The proportion increases significantly with the age of the respondent, from 8 percent among respondents age 20 to 29, to 14 percent among respondents age 40 to 49. The proportions are highest among those with the highest educational levels (14 percent), those living in urban areas (13 percent) and in the City of Kigali (13 percent), and those living in the richest households (13 percent). The proportions are almost the same for men (9 percent) and women (8 percent). Orphaned and Vulnerable Children | 249 Among these primary caregivers, only 19 percent said that they had made arrangements to have someone care for the children in the event of their own illness or death. The proportion of caregivers who have made succession arrangements is higher among men (25 percent) than women (17 percent). It is also higher in rural areas (20 percent) than in urban areas (17 percent), and higher in the North province (32 percent) than elsewhere. The percentage of caregivers who have made succession arrangements is higher for those with the highest education (24 percent for those with a secondary education or higher) and those in the richest households (23 percent). Table 16.6 Succession planning Percentage of de facto women and men age 15-49 who are the primary caregivers of children under age 18 years, and among the primary caregivers, the percentage who have made arrangements for someone else to care for the children in the event of their own inability to do so because of illness or death, by background characteristics, Rwanda 2005 Background characteristic Percentage of women and men who are primary caregivers Number of women and men age 15-49 Percentage of caregivers who have made succession arrangements Number of primary caregivers Age 15-19 1.5 3,687 16.0 56 20-29 8.0 5,669 20.9 454 30-39 11.3 3,550 19.4 400 40-49 14.2 2,828 18.1 403 Sex Male 9.1 4,413 25.2 402 Female 8.0 11,321 16.8 911 Residence Urban 13.1 2,705 17.4 356 Rural 7.4 13,029 20.1 958 Province City of Kigali 12.9 1,622 18.5 209 South 8.7 4,097 15.7 356 West 8.5 3,890 15.9 330 North 6.4 2,840 31.7 183 East 7.2 3,285 21.0 236 Education No education 8.1 3,364 11.7 272 Primary 7.5 10,724 20.7 807 Secondary or higher 14.3 1,646 23.7 235 Wealth quintile Lowest 6.9 3,220 13.6 222 Second 7.3 3,119 18.3 226 Middle 7.5 2,991 16.5 225 Fourth 6.8 3,033 22.0 205 Highest 12.9 3,371 23.1 434 Total 8.3 15,734 19.4 1,313 Note: Table is based on de facto household members, persons who slept in household the night before the survey. 16.4 PROTECTION OF VULNERABLE CHILDREN Dispossession of property can worsen the vulnerability of both people who care for children and the children themselves. It is therefore important to improve inheritance laws, including enforcement mechanisms, to ensure the right of women and children to inherit property after the death of a husband or father (UNICEF, 2005). For this reason, an indicator was devised to estimate the proportion of women who were dispossessed of property after the death of a spouse. 250 | Orphaned and Vulnerable Children Table 16.7 shows the proportion of women who were or are widows and the percentage of widowed women who were dispossessed of their property after the death of their spouse. Altogether, 7 percent of the women surveyed have been widowed. This proportion increases with the age of the woman, from 2 percent at age 20 to 29, to 21 percent at age 40 to 49. Similarly, the proportion of widows increases with the age of the child. The results according to other background characteristics show no major differentials. One-third of ever-widowed women said they had been dispossessed of their property. This proportion is much higher for women age 20 to 29 (67 percent) and women age 30 to 39 (42 percent) than for older women (21 percent at age 40 to 49). More- over, it appears that those most often disin- herited are women with a primary education (36 percent), women in urban areas (37 per- cent), and women in the South (37 percent) and West (36 percent) provinces. 16.5 CARE AND SUPPORT 16.5.1 Care and Support of the Chronically Ill When an adult member of a house- hold dies or falls chronically ill, it can have a devastating effect on the remaining members of the household, particularly children. In such cases, household survival often depends on external assistance or support. For this reason, the survey asked households in which someone age 18 to 59 had been chronically ill for three of the past 12 months, or had died after a chronic illness in the past 12 months, whether the household had received free medical, emotional, or material support to care for these persons in the past year. The results are presented in Table 16.8. Table 16.7 Widows dispossessed of property Percentage of de facto women age 15-49 who have been widowed, and the percentage of widowed women who have been dispossessed of property, by s background characteristics, Rwanda 2005 Background characteristic Percentage of women who have been widowed Number of women Percentage who were dispossessed of property1 Number of ever- widowed women Age 15-19 0.0 2,585 na 0 20-29 1.8 4,092 66.5 73 30-39 12.1 2,600 41.9 314 40-49 20.9 2,045 21.2 428 Age of youngest child No children 0.3 4,234 * 12 <18 years 7.9 5,167 44.9 406 18+ years 20.6 1,921 * 396 Residence Urban 7.6 1,921 36.5 145 Rural 7.1 9,400 32.5 669 Province City of Kigali 8.4 1,127 34.8 94 South 7.7 2,958 36.5 228 West 7.1 2,824 36.0 200 North 6.8 2,063 22.4 141 East 6.4 2,348 33.6 150 Education No education 10.9 2,646 30.3 289 Primary 5.8 7,591 36.2 437 Secondary or higher 8.0 1,084 28.1 87 Wealth quintile Lowest 8.2 2,421 29.1 198 Second 7.7 2,325 35.1 178 Middle 7.7 2,099 34.5 162 Fourth 6.0 2,133 36.3 128 Highest 6.3 2,342 32.4 148 Total 7.2 11,321 33.2 814 Note: Table is based on de facto household members, persons who slept in household the night before the survey. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. na = Not applicable 1 Dispossessed of property indicates that most of late husband’s property went to another wife, to the husband’s family (not including respondent or children), or to another person. Orphaned and Vulnerable Children | 251 Table 16.8 External support for chronically ill persons For persons age 18 to 59 chronically ill for at least 3 of the past 12 months or who died within the past 12 months after being chronically ill for at least 3 months, the percentage whose household received some type of free basic external support in the past year to care for them, by background characteristics, Rwanda 2005 Percentage of chronically ill persons whose households received: Background characteristic Medical support at least once a month during illness Emotional support1 in the past 30 days3 Social/ material, support2 in the past 30 days3 At least one type of support in the past 30 days3 All three types of support in the past 30 days3 None of the three types of support Number of persons Age 18-29 2.2 10.0 1.6 12.6 0.2 87.4 338 30-39 5.0 7.2 3.0 11.8 0.3 88.2 244 40-49 2.8 6.5 3.0 10.4 0.0 89.6 305 50-59 4.5 9.4 3.2 14.9 0.3 85.1 226 Sex Male 2.5 8.5 2.2 11.5 0.2 88.5 428 Female 4.0 8.2 2.8 12.8 0.2 87.2 685 Residence Urban 3.8 6.4 1.5 10.2 0.3 89.8 209 Rural 3.4 8.8 2.9 12.8 0.1 87.2 904 Province City of Kigali 4.7 8.2 1.8 14.7 0.0 85.3 114 South 4.0 5.7 2.2 9.8 0.2 90.2 320 West 2.3 11.9 4.8 16.0 0.0 84.0 280 North 2.7 13.5 2.6 15.6 0.4 84.4 148 East 3.8 4.6 1.0 8.3 0.2 91.7 251 Wealth quintile Lowest 2.9 7.5 2.2 10.3 0.0 89.7 252 Second 1.6 7.1 1.7 8.7 0.0 91.3 196 Middle 2.8 8.0 2.7 11.1 0.0 88.9 233 Fourth 5.1 8.9 3.8 15.7 0.2 84.3 243 Highest 4.7 10.5 2.3 15.8 0.7 84.2 187 Total 3.4 8.3 2.6 12.3 0.2 87.7 1,113 Note: Table is based on de jure household members, i.e., usual household members, who were chronically ill in the past 12 months or who died of a chronic illness in the past 12 months. 1 Support such as companionship, counseling from a trained counselor or spiritual support for which there was no payment 2 Support such as help with household work, training for a caregiver, legal services, clothing, food or financial support for which there was no payment. 3 In the past 30 days for living persons and in the 30 days preceding death for dead persons The data show that very few households had received assistance to care for a chronically ill member. The great majority of households caring for chronically ill persons, or that had lost a member to chronic illness, in the past 12 months, had received no support at all (88 percent). When assistance was received, it was generally in the form of emotional support in the past 30 days (8 percent). Only a small proportion of households caring for chronically ill persons received any other type of support, be it medical (3 percent) or material/social (3 percent). Altogether, 12 percent of households with chronically ill members received a single type of support; less than 1 percent of households received all three types of support. 16.5.2 Care and Support of OVC OVC are generally cared for by their families, which, in turn, often depend on community assistance to survive. Strengthening family and community capacities to protect OVC and ensure their basic needs is therefore a key component of OVC support. For all households supporting an OVC under age 18, the RDHS-III asked if the household had received free assistance to care for the OVC in the form of one of the external supports covered by the survey. The indicator presented in Table 16.9 estimates the level of free external support received by families to care for OVC. 252 | Orphaned and Vulnerable Children As for households caring for chronically ill, the majority of households supporting OVC (87 percent) received no external support to assist in their care. When support was received, it was generally in the form of school-related assistance (9 percent). Only a small proportion of OVC received any other type of support, be it medical (3 percent), emotional (2 percent), or material/social (2 percent). Altogether, 13 percent of OVC households received a single type of support; less than 1 percent of households received all three types of support. Table 16.9 External support for orphans and vulnerable children Percentage of orphans and vulnerable children under age 18 years whose household received some type of free basic external support to care for the child in the past 12 months, by background characteristics, Rwanda 2005 Percentage of orphans and vulnerable children whose households received: Background characteristic Medical support1 in the past 12 months Emotional support2 in the past 3 months Social/ material support3 in the past 3 months School- related assistance4 in the past 12 months At least one type of support5 All of the types of support5 None of the types of support Number of OVC Age 0-4 1.7 2.4 1.2 na 4.4 0.0 95.6 1,001 5-9 3.3 2.6 2.5 7.9 12.2 0.0 87.8 1,728 10-14 4.5 1.8 2.1 13.2 16.7 0.3 83.3 2,676 15-17 2.7 1.8 1.8 8.5 11.2 0.2 88.8 1,531 Sex Male 3.2 2.1 2.0 8.3 11.8 0.2 88.2 3,427 Female 3.6 2.1 2.1 9.6 13.4 0.1 86.6 3,509 Residence Urban 6.2 2.6 3.3 9.2 14.4 0.3 85.6 1,161 Rural 2.8 2.0 1.8 8.9 12.2 0.1 87.8 5,775 Province City of Kigali 4.0 3.3 1.2 6.6 11.0 0.4 89.0 616 South 4.7 1.2 2.0 7.7 10.8 0.1 89.2 1,897 West 2.8 2.7 1.3 13.8 17.5 0.1 82.5 1,826 North 3.4 1.9 3.8 9.6 14.3 0.0 85.7 1,177 East 2.3 2.2 1.9 4.9 8.0 0.3 92.0 1,420 Wealth quintile Lowest 3.3 1.5 1.7 8.5 11.5 0.1 88.5 1,657 Second 3.6 2.4 2.4 11.3 15.3 0.3 84.7 1,247 Middle 3.3 1.4 2.0 7.8 11.0 0.0 89.0 1,416 Fourth 3.4 2.0 1.9 8.4 13.0 0.1 87.0 1,322 Highest 3.5 3.4 2.3 9.0 12.6 0.3 87.4 1,295 Total 3.4 2.1 2.0 8.9 12.6 0.2 87.4 6,936 Note: Table is based on de jure household members, i.e., usual household members. 1 Medical care, supplies or medicine 2 Companionship, counseling from a trained counselor, or spiritual support for which there was no payment. 3 Help with household work, training for a caregiver, legal services, clothing, food, or financial support for which there was no payment. 4 Allowance, free admission, books, or supplies for which there as no payment. Percentage calculated for age 5-17 years. 5 Four types of support for those age 5-17, three types of support (i.e. excluding school support) received by those age 0-4. References | 253 REFERENCES ACC/SCN. 2000. Fourth report on the world nutrition situation. Geneva: CC/SCN in collaboration with IFPRI. Attaran, A., K.I. Barnes, C. Curtis et al. 2004. WHO, the Global Fund, and medical malpractice in malaria treatment. Lancet 363(9404): 237-40. Barrère, B., J. Schoemaker, M. Barrère, T. Habiyakare, A. Kabagwira, and M. Ngendakumana. 1994. Enquête Démographique et de Santé, Rwanda 1992. Kigali, Rwanda and Calverton, Maryland, USA: Office National de la Population [Rwanda] and Macro International Inc. Boerma, T. 1988. Monitoring and evaluation of health interventions : Age- and cause-specific mortality and morbidity in childhood. In Research and intervention issues concerning infant and child mortality and health, 195-218. Proceedings of the East Africa Workshop, International Development Research Centre, Ottawa, Canada. Centers for Disease Control and Prevention (CDC). 1998. Recommendations to prevent and control iron deficiency in the United States. Morbidity and Mortality Weekly Report 47 (RR-3): 1-29. Delpeuch, F. 1991. Indices et indicateurs anthropométriques : choix, interprétation, présentation et utilisation. In Atelier sur la surveillance nutritionnelle en Afrique de l’Ouest : méthodologie des enquêtes nutritionnelles. Working paper. Dakar, Sénégal: ORANA, ORSTOM and OMS. DeMaeyer, E.M. 1989. Preventing and controlling iron deficiency anaemia through primary health care : A guide for health administrators and programme managers. E.M. DeMaeyer with the collaboration of P. Dallman et al. Geneva: World Health Organization. Department of Statistics [Rwanda]. 2004. Rwanda Development Indicators, 2004. Kigali, Rwanda: Ministry of Finances and Economic Planning. Direction des Statistiques [Rwanda]. 1998. Enquête Socio-démographique (ESD, 1996). Kigali, Rwanda: Ministère des Finances et de la Planification Économiques. Direction des Statistiques [Rwanda]. 1997. L’Enquête Intégrale sur les Conditions de Vie (EICV),1997. Kigali, Rwanda: Ministère du Plan. Fall, I.S. 2003. Analyse de situation pour la lutte contre le paludisme dans le cadre de l’initiative « Roll Back Malaria » au Rwanda. Gwatkin, D.R., S. Rutstein, K. Johnson, R.P. Pande and A. Wagstaff. 2000. Socio-economic differences in health, nutrition and poverty. HNP/Poverty Thematic Group of the World Bank, Washington, D.C.: The World Bank. Hinde, A. 1998. Demography methods. New York: Oxford University Press, Inc. Ivorra, C.V. 1967. Paludisme. In Santé et maladies au Rwanda. Bruxelles: AGCD. 427-447. 254 | References Manga, L. 1997. Mise en œuvre accélérée de la lutte contre le paludisme en Afrique en 1997. Mission d’appui au programme national de lutte contre le paludisme au Rwanda. Final Report. Meyus, H., M. Lips, and H. Caubergh. 1962. L’état actuel des problèmes de paludisme d’altitude au Ruanda-Urundi. Annales de la Société belge de médecine tropicale 42(5): 771-782. Ministry of Local Government, Community Development and Social Affairs (MINALOC) [Rwanda]. 2001. Rapport national sur le suivi du sommet mondial de 1990 pour les enfants. Kigali: MINALOC. National Census Bureau [Rwanda]. 2005. Third Rwandan General Population and Housing Census – August 15, 2002. Kigali, Rwanda: National Census Bureau. Office National de la Population (ONAPO) [Rwanda]. 1984. National Fertility Survey, 1983, Vol. 1, Analyse des résultats. Kigali, Rwanda: Office National de la Population. Office National de la Population (ONAPO) [Rwanda] and Macro International Inc. 2001. Enquête Démographique et de Santé Rwanda 2000. Kigali, Rwanda and Calverton, Maryland, USA: Office National de la Population and Macro International Inc. Population Reference Bureau (PRB). 2005. International data sheet. http://www.prb.org/ Programme National Intégré de Lutte contre le Paludisme (PNILP). 2001. Rapport de l’atelier sur la nouvelle approche thérapeutique. PNILP. Rusanganwa, A. 1999. Epidemiologic Microstratification of paludism: Index plasmodisques and its determinants in two basic medical zones of Rwanda. Work of end of studies of the DEA in sciences of health: specialization in statistical epidemiology, Université Libre de Bruxelles. Rutstein, S.O., and G. Rojas. 2003. Guide to DHS statistics. Calverton, Maryland, USA: ORC Macro. Strauss, M.A. 1990. Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. In Physical violence in American families: Risk factors and adaptations to violence in 8,145 families, ed. M.A. Strauss and R.J. Gelles, 39-47. New Brunswick: Transation Publishers. Sullivan, J.M., S.O. Rutstein, and G.T. Bicego. 1994. Infant and child mortality. DHS Comparative Studies No. 15. Calverton, Maryland, USA: ORC Macro. Trussell, J. and G. Rodriguez. 1990. A note on the sisterhood estimate of maternal mortality. Studies in Family Planning 21(6): 344-346. U.S. Census Bureau. 2005. International data base. http://www.census.gov/ipc UNICEF. 2005. Guide to monitoring and evaluation of the national response for children orphaned and made vulnerable by HIV/AIDS. New York: UNICEF. United Nations. 1973. The determinants and consequences of population trends. Vol. 1. New York: United Nations. United Nations. 1982. Model Life Tables for developing countries. New York: United Nations. United Nations. 2001. Declaration of Commitment on HIV/AIDS. New York: United Nations. References | 255 United Nations General Assembly. 1993. Declaration on the Elimination of Violence against Women. Secretary General’s Report. New York: United Nations. Vermylen, M. 1967. Répartition des Anophèles de la République du Rwanda et Burundi. Rivista di Malariologia 46(1): 13-22. World Health Organization (WHO). 2003. Position of WHO’s Roll Back Malaria Department on malaria treatment policy. Geneva: World Health Organization. Yamey, G. 2003. Malaria researchers say global fund is buying “useless drug.” British Medical Journal 327(7425): 1188. Yip, R. 1994. Changes in iron metabolism with age. In Iron metabolism in health and disease, ed. J.H. Brock, J. Halliday and L. Powell. London: W.B. Sanders. 427-448. Appendix A | 257 SAMPLE DESIGN APPENDIX A A.1 INTRODUCTION The third Demographic and Health Survey in Rwanda (2005 RDHS-III) followed those conducted in 1992 and 2000. It is composed of a nationally representative sample of approximately 10,500 households. All women age 15-49 who were usual residents of the household or who were present in the sampled households on the night before the survey were eligible to be interviewed. In addition, a subsample of 50 percent of all households selected for the women’s questionnaire was selected for the men’s questionnaire. In this subsample of households, all men age 15-59 were eligible to be interviewed and, in addition, all eligible men and women were asked to consent to an HIV test. As with the prior two surveys, the primary goal of the survey was to collect data on fertility, knowledge and use of contraception, maternal and childhood mortality, and sexually-transmitted infections and HIV/AIDS. The data were representative at the national level, for urban-rural residence, and for each of the five provinces. The sample was designed to be representative for each of the 12 old provinces, and is therefore representative at the level of the five new provinces, since these represent a regrouping of the 12 old provinces. A.2 SAMPLE FRAME The National Census Service possesses a computer file of 7,727 enumeration areas (EAs) created for the 2002 General Census of Population and Housing (RGPH, 2002). In that file, each EA is listed with all of its identifiers (province, district, and identification code), its population size, number of households, and classification as urban or rural. The boundaries for each EA are clearly identifiable on the cartographic maps created for the 2002 RGPH. The distribution of EAs and of households among the 12 old provinces and according to urban-rural residence is shown in Table A.1. Table A.1 Distribution of households and enumeration areas (EAs) by old province and according to residence (RGPH, 2002) Number of households Number of EAs Old province Urban Rural Total Urban Rural Total Ville de Kigali 124,964 0 124,964 565 0 565 Kigali Ngali 11,513 160,967 172,480 41 694 735 Gitarama 27,205 157,108 184,313 116 698 814 Butare 27,117 137,526 164,643 113 568 681 Gikongoro 6,258 100,833 107,091 28 465 493 Cyangugu 9,284 111,267 120,551 42 559 601 Kibuye 9,654 92,747 102,401 40 432 472 Gisenyi 12,360 174,853 187,213 51 761 812 Ruhengeri 14,474 178,686 193,160 61 779 840 Byumba 12,294 139,645 151,939 50 615 665 Umutara 1,843 89,817 91,660 7 393 400 Kibungo 16,015 140,996 157,011 64 585 649 Total 272,981 1,484,445 1,757,426 1,178 6,549 7,727 258 | Appendix A A.3 SAMPLE SELECTION The sample for the RDHS-III used a stratified, two-stage cluster selection. The primary sampling unit is the EA as defined in the 2002 census. Each province is separated into urban and rural areas to create the sampling strata and the sample was drawn independently in each stratum. There were therefore 23 strata in total, because the City of Kigali comprised one strata, as it had no rural component. In the first stage, 432 EAs were selected with probability proportional to size, the size being the number of households in the EA. An updating operation listed all the households in each selected EA and this list was used to select the households for the second stage. Before this updating of the households, the larger EAs were divided into segments, of which only one was selected for the survey. In the second stage, in each EA selected in the first stage, a fixed number of households (20 households in each urban cluster, 24 households in each rural cluster) were selected using a systematic selection based on the new list of households created during the household listing. In all, 10,644 households were selected for the women’s interview. All members of each selected household were listed in the Household Questionnaire. Every woman age 15-49 in the household was interviewed using the Women’s Questionnaire. Half of the households selected for the women’s interview were also selected for the men’s interview. In this subsample of house- holds all men age 15-59 were interviewed. All men age 15-59 and all women age 15-49 in this subsample of households were also asked to consent to an HIV test. Table A.2 shows the sample allocation by old province and according to urban-rural residence. In all, 462 EAs were selected (111 in urban areas and 351 in rural areas) and 10,644 households were selected (2,220 in urban areas and 8,424 in rural areas.) Table A.2 Sample allocation by old province and according to residence Number of households Number of EAs Old province Urban Rural Total Urban Rural Total Expected number of interviewed women Ville de Kigali 880 0 880 44 0 44 899 Kigali Ngali 100 792 892 5 33 38 911 Gitarama 180 696 876 9 29 38 894 Butare 200 672 872 10 28 38 890 Gikongoro 100 792 892 5 33 38 911 Cyangugu 120 768 888 6 32 38 907 Kibuye 120 768 888 6 32 38 907 Gisenyi 100 792 892 5 33 38 911 Ruhengeri 120 768 888 6 32 38 907 Byumba 120 768 888 6 32 38 907 Umutara 40 864 904 2 36 38 923 Kibungo 140 744 884 7 31 38 903 Total 2,220 8,424 10,644 111 351 462 10,868 Appendix A | 259 A.4 SAMPLING PROBABILITY The sampling probabilities were calculated separately for each sampling stage and for each stratum. For each stratum h, the following notations are used: P1hi: first-stage sampling probability of EA i. P2hi: second-stage sampling probability of households in EA i. Let ah be the number of clusters selected in stratum h, Mhi the number of households of the ith EA in stratum h, and Mh the total number of households in stratum h. In the first stage, the probability of inclusion of the ith EA in the sample is calculated as follows: h hih hi M Ma P × =1 In the second stage, a number of bhi households is selected from the number Lhi households found during the household listing in the ith EA. We then have: hi hi hi L b P =2 Because of the non-proportional distribution of the sample between strata, sampling weights are used to insure that the sample is representative at the national level. Sampling weights for individuals of cluster i in strata h are calculated as follows: hihi hi PP W 21 1 = with a correction for non-response and normalization. A.5 SURVEY RESULTS Tables A.3 and A.4 present the detailed results of the household interviews and the women’s and men’s interviews, according to urban-rural residence and the five provinces. Tables A.5 and A.6 present the coverage of HIV testing among women and men, respectively, by background characteristics. Tables A.7 and A.8 present the coverage of HIV testing among women and men, respectively, according to characteristics related to risk status. 260 | Appendix A Table A.3 Sample implementation: women Percent distribution of households and eligible women by results of the household and individual interviews, and household, eligible women and overall response rates, according to urban-rural residence and province, Rwanda 2005 Residence Province Result Urban Rural City of Kigali South West North East Total Selected households Completed (1) 94.9 96.9 93.7 96.9 96.9 97.4 96.4 96.5 Household present but no respondent at home (2) 0.4 0.2 0.6 0.3 0.1 0.2 0.1 0.2 Household absent (3) 0.5 0.4 0.1 0.0 0.0 0.0 0.0 0.5 Postponed (4) 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 Refused (5) 0.2 0.0 0.1 0.0 0.1 0.0 0.0 0.1 Dwelling vacant/address not a dwelling (6) 3.0 1.8 0.5 0.3 0.3 0.3 0.8 2.1 Dwelling destroyed (7) 0.9 0.6 3.1 2.2 1.9 1.7 2.0 0.6 Dwelling not found (8) 0.0 0.0 1.4 0.4 0.6 0.4 0.7 0.0 Other (9) 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of sampled households 2,220 8,424 1,100 2,640 2,764 1,752 2,388 10,644 Household response rate (HRR) 1 99.3 99.8 98.7 99.7 99.7 99.8 99.9 99.7 Eligible women Completed (a) 97.3 98.4 96.3 98.7 97.6 98.0 99.1 98.1 Not at home (b) 1.6 1.0 2.2 0.8 1.6 1.3 0.3 1.2 Postponed (c) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Refused (d) 0.3 0.0 0.5 0.0 0.1 0.1 0.0 0.1 Partly completed (e) 0.1 0.0 0.1 0.0 0.0 0.1 0.0 0.1 Incapacitated (f) 0.7 0.5 0.8 0.5 0.6 0.5 0.5 0.5 Other (g) 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of women 2,689 8,850 1,380 2,796 3,043 1,858 2,462 11,539 Eligible women response rate (EWRR) 2 97.3 98.4 96.3 98.7 97.6 98.0 99.1 98.1 Overall response rate (ORR) 3 96.6 98.1 95.0 98.4 97.4 97.8 99.0 97.8 1 Using the number of households falling into specific response categories, the household response rate (HRR) is calculated as: 100 * (1) _____________________________________ (1) + (2) + (4) + (5) + (8) 2 Using the number of eligible women falling into specific response categories, the eligible woman response rate (EWRR) is calculated as: 100 * (a) ____________________________________________________ (a) + (b) + (c) + (d) + (e) + (f) + (g) 3 The overall response rate (ORR) is calculated as: ORR = HRR * EWRR/100 Appendix A | 261 Table A.4 Sample implementation: men Percent distribution of households and eligible men by results of the household and individual interviews, and household, eligible men and overall response rates, according to urban-rural residence and province, Rwanda 2005 Residence Province Result Urban Rural City of Kigali South West North East Total Selected households Completed (1) 94.9 96.9 93.8 96.5 96.9 97.7 96.4 96.5 Household present but no respondent at home (2) 0.5 0.2 0.7 0.4 0.2 0.1 0.2 0.3 Household absent (3) 0.6 0.4 0.2 0.0 0.0 0.0 0.0 0.5 Postponed (4) 0.1 0.0 0.4 0.0 0.1 0.0 0.0 0.0 Refused (5) 0.1 0.0 0.2 0.0 0.0 0.0 0.0 0.1 Dwelling vacant/address not a dwelling (6) 2.8 1.8 0.7 0.2 0.4 0.5 0.6 2.0 Dwelling destroyed (7) 1.0 0.5 2.4 2.4 1.7 1.4 2.2 0.6 Dwelling not found (8) 0.1 0.0 1.6 0.5 0.7 0.2 0.7 0.0 Other (9) 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of sampled households 1,110 4,212 550 1,320 1,382 876 1,194 5,322 Household response rate (HRR) 1 99.2 99.7 98.5 99.6 99.7 99.9 99.8 99.6 Eligible men Completed (a) 95.5 97.7 93.9 97.6 97.1 97.0 99.0 97.2 Not at home (b) 3.1 1.3 4.1 1.1 2.2 1.4 0.6 1.7 Postponed (c) 0.1 0.0 0.2 0.0 0.0 0.0 0.1 0.0 Refused (d) 0.4 0.1 0.5 0.1 0.2 0.4 0.0 0.2 Partly completed (e) 0.3 0.1 0.5 0.2 0.1 0.1 0.1 0.2 Incapacitated (f) 0.6 0.7 0.9 1.0 0.4 1.0 0.2 0.7 Other (g) 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number of men 1,183 3,776 659 1,180 1,274 769 1,077 4,959 Eligible men response rate (EMRR) 2 95.5 97.7 93.9 97.6 97.1 97.0 99.0 97.2 Overall response rate (ORR) 3 94.8 97.4 92.5 97.2 96.8 96.9 98.8 96.8 1 Using the number of households falling into specific response categories, the household response rate (HRR) is calculated as: 100 * (1) _____________________________________ (1) + (2) + (4) + (5) + (8) 2 Using the number of eligible men falling into specific response categories, the eligible man response rate (EMRR) is calculated as: 100 * (a) ____________________________________________________ (a) + (b) + (c) + (d) + (e) + (f) + (g) 3 The overall response rate (ORR) is calculated as: ORR = HRR * EMRR/100 262 | Appendix A Table A.5 Coverage of HIV testing among interviewed women by background characteristics Percent distribution of interviewed women age 15-49 by testing status, according to background characteristics (unweighted), Rwanda 2005 Background characteristic Tested Refused Absent for testing Other/ missing Total Unweighted number Marital status Currently married/in union 99.0 0.8 0.1 0.1 100.0 2,737 Widowed 97.8 2.2 0.0 0.0 100.0 229 Divorced/separated 99.4 0.4 0.0 0.2 100.0 522 Never married 98.6 1.0 0.3 0.1 100.0 2,241 Ever had sex 99.3 0.2 0.2 0.2 100.0 435 Never had sex 98.4 1.2 0.3 0.1 100.0 1,806 Type of union In union, polygynous 99.1 0.9 0.0 0.0 100.0 325 In union, not polygynous 99.0 0.8 0.1 0.1 100.0 2,387 Not in union 98.7 1.0 0.2 0.1 100.0 2,992 Missing 96.0 4.0 0.0 0.0 100.0 25 Ever had sexual intercourse Yes 99.0 0.8 0.1 0.1 100.0 3,923 No 98.4 1.2 0.3 0.1 100.0 1,806 Currently pregnant Yes 99.5 0.5 0.0 0.0 100.0 434 Not pregnant/not sure 98.8 0.9 0.2 0.1 100.0 5,295 Times slept away from home in past 12 months Never 98.9 0.9 0.2 0.1 100.0 4,420 1-2 99.2 0.7 0.1 0.0 100.0 967 3-4 98.2 1.3 0.0 0.4 100.0 223 5+ 96.5 2.7 0.0 0.9 100.0 113 Missing 100.0 0.0 0.0 0.0 100.0 6 Whether away for more than one month in past 12 months Away for more than 1 month 98.3 0.8 0.4 0.4 100.0 239 Away for less than 1 month 98.9 1.0 0.0 0.1 100.0 1,062 Never away 98.9 0.9 0.2 0.1 100.0 4,420 Missing 100.0 0.0 0.0 0.0 100.0 8 Religion Catholic 99.0 0.7 0.2 0.1 100.0 2,536 Protestant 98.7 0.9 0.2 0.1 100.0 2,224 Adventist 98.6 1.4 0.0 0.0 100.0 720 Muslim 99.2 0.8 0.0 0.0 100.0 119 Other/missing 99.2 0.8 0.0 0.0 100.0 130 Total 98.8 0.9 0.2 0.1 100.0 5,729 Appendix A | 263 Table A.6 Coverage of HIV testing among interviewed men by background characteristics Percent distribution of interviewed men age 15-59 by testing status, according to background characteristics (unweighted), Rwanda 2005 Background characteristic Tested Refused Absent for testing Other/ missing Total Unweighted number Marital status Currently married/in union 98.2 1.5 0.2 0.0 100.0 2,478 Widowed 97.2 0.0 0.0 2.8 100.0 36 Divorced/separated 95.7 2.2 2.2 0.0 100.0 92 Never married 98.1 1.7 0.2 0.1 100.0 2,214 Ever had sex 97.8 2.1 0.1 0.0 100.0 858 Never had sex 98.2 1.4 0.2 0.1 100.0 1,356 Type of union In union, polygynous 98.5 1.5 0.0 0.0 100.0 134 In union, not polygynous 98.2 1.5 0.2 0.0 100.0 2,341 Not in union 98.0 1.7 0.3 0.1 100.0 2,342 Missing 100.0 0.0 0.0 0.0 100.0 3 Ever had sexual intercourse Yes 98.0 1.7 0.2 0.1 100.0 3,463 No 98.2 1.4 0.2 0.1 100.0 1,357 Circumcision status Circumcised 95.1 4.5 0.4 0.0 100.0 532 Not circumcised 98.5 1.2 0.2 0.1 100.0 4,261 Missing 92.6 7.4 0.0 0.0 100.0 27 Times slept away from home in past 12 months Never 98.1 1.6 0.3 0.1 100.0 3,592 1-2 99.0 1.0 0.0 0.0 100.0 704 3-4 97.4 1.9 0.4 0.4 100.0 265 5+ 97.1 2.5 0.4 0.0 100.0 238 Missing 90.5 9.5 0.0 0.0 100.0 21 Whether away for more than one month in past 12 months Away for more than 1 month 98.1 1.6 0.0 0.3 100.0 373 Away for less than 1 month 98.1 1.6 0.2 0.0 100.0 809 Never away 98.1 1.6 0.3 0.1 100.0 3,592 Missing 97.8 2.2 0.0 0.0 100.0 46 Religion Catholic 98.5 1.2 0.2 0.1 100.0 2,416 Protestant 97.9 1.6 0.3 0.1 100.0 1,586 Adventist 97.9 2.1 0.0 0.0 100.0 585 Muslim 95.5 4.5 0.0 0.0 100.0 112 Other/missing 95.0 5.0 0.0 0.0 100.0 121 Total 98.1 1.6 0.2 0.1 100.0 4,820 264 | Appendix A Table A.7 Coverage of HIV testing among women who ever had sex by risk status variables Percent distribution of women age 15-49 who ever had sex by testing status, according to characteristics relating to risk status (unweighted), Rwanda 2005 Risk status characteristic Tested Refused Absent for testing Other/ missing Total Unweighted number Age at first sex < 16 100.0 0.0 0.0 0.0 100.0 433 16-17 99.9 0.0 0.0 0.1 100.0 683 18-19 99.3 0.6 0.1 0.0 100.0 986 20 or older 98.5 1.2 0.1 0.1 100.0 1,701 Missing 95.8 2.5 0.8 0.8 100.0 120 Higher-risk sex in past 12 months Had higher-risk sex 99.3 0.0 0.4 0.4 100.0 269 Had sex, not higher-risk sex 99.0 0.8 0.1 0.1 100.0 2,673 No sex in past 12 months 99.1 0.8 0.0 0.1 100.0 981 Number of partners in past 12 months 0 99.1 0.8 0.0 0.1 100.0 981 1 99.0 0.8 0.1 0.1 100.0 2,923 2 or more 100.0 0.0 0.0 0.0 100.0 19 Number of higher-risk sexual partners in past 12 months 0 99.0 0.8 0.1 0.1 100.0 3,654 1 99.2 0.0 0.4 0.4 100.0 256 2 or more 100.0 0.0 0.0 0.0 100.0 13 Any condom use (FP, other) Used condom at any time 98.3 0.6 0.6 0.6 100.0 175 Never used condom 99.1 0.8 0.1 0.1 100.0 3,748 Condom use at last sex in past 12 months Used condom at last sex 97.1 0.0 1.0 1.9 100.0 104 No condom at last sex 99.1 0.8 0.1 0.0 100.0 2,838 Condom use at last higher-risk sex in past 12 months Used condom at last higher-risk sex 96.8 0.0 1.6 1.6 100.0 63 No condom at last higher-risk sex 100.0 0.0 0.0 0.0 100.0 206 Condom use at first sex Used condom at first sex 98.2 0.0 1.8 0.0 100.0 57 No condom at first sex 99.0 0.8 0.1 0.1 100.0 3,866 Number of lifetime sexual partners 1 98.9 0.8 0.1 0.1 100.0 2,721 2 99.3 0.6 0.0 0.1 100.0 845 3-4 99.7 0.3 0.0 0.0 100.0 303 5-9 97.4 2.6 0.0 0.0 100.0 39 10 or more 100.0 0.0 0.0 0.0 100.0 10 Missing 80.0 20.0 0.0 0.0 100.0 5 HIV testing status Ever tested and knows results of last test 98.4 1.1 0.2 0.3 100.0 1,137 Ever tested, does not results 98.0 2.0 0.0 0.0 100.0 152 Never tested 99.4 0.5 0.1 0.0 100.0 2,618 Missing 100.0 0.0 0.0 0.0 100.0 16 Total 99.0 0.8 0.1 0.1 100.0 3,923 Appendix A | 265 Table A.8 Coverage of HIV testing among men who ever had sex by risk status variables Percent distribution of men age 15-59 who ever had sex by testing status, according to characteristics relating to risk status (unweighted), Rwanda 2005 Risk status characteristic Tested Refused Absent for testing Other/ missing Total Unweighted number Age at first sex < 16 99.2 0.8 0.0 0.0 100.0 597 16-17 97.6 2.2 0.2 0.0 100.0 465 18-19 97.8 1.8 0.3 0.1 100.0 730 20 or older 97.8 1.8 0.3 0.1 100.0 1,661 Missing 100.0 0.0 0.0 0.0 100.0 10 Higher-risk sex in past 12 months Had higher-risk sex 96.7 3.1 0.2 0.0 100.0 425 Had sex, not higher-risk sex 98.3 1.5 0.2 0.0 100.0 2,344 No sex in past 12 months 98.0 1.6 0.3 0.1 100.0 694 Number of partners in past 12 months 0 98.0 1.6 0.3 0.1 100.0 694 1 98.0 1.7 0.2 0.0 100.0 2,635 2 or more 99.3 0.7 0.0 0.0 100.0 134 Number of higher-risk sexual partners in past 12 months 0 98.2 1.5 0.2 0.1 100.0 3,038 1 96.5 3.3 0.3 0.0 100.0 400 2 or more 100.0 0.0 0.0 0.0 100.0 25 Paid for sex in the past 12 months Yes 97.6 2.4 0.0 0.0 100.0 42 No 98.0 1.7 0.2 0.1 100.0 3,421 Any condom use (FP, other) Used condom at any time 95.1 4.4 0.3 0.2 100.0 653 Never used condom 98.7 1.0 0.2 0.0 100.0 2,810 Condom use at last sex in past 12 months Used condom at last sex 97.0 2.4 0.6 0.0 100.0 169 No condom at last sex 98.1 1.7 0.2 0.0 100.0 2,600 Condom use at last higher-risk sex in past 12 months Used condom at last higher-risk sex 96.9 3.1 0.0 0.0 100.0 160 No condom at last higher-risk sex 96.6 3.0 0.4 0.0 100.0 265 Condom use at last paid sexual encounter in past 12 months Used condom at last sex 100.0 0.0 0.0 0.0 100.0 27 No condom at last sex 93.3 6.7 0.0 0.0 100.0 15 Condom use at first sex Used condom at first sex 97.2 2.8 0.0 0.0 100.0 106 No condom at first sex 98.1 1.6 0.2 0.1 100.0 3,357 Number of lifetime sexual partners 1 98.5 1.2 0.2 0.0 100.0 1,233 2 98.2 1.6 0.1 0.1 100.0 871 3-4 97.7 2.1 0.2 0.0 100.0 898 5-9 98.4 1.0 0.6 0.0 100.0 314 10 or more 95.6 3.7 0.0 0.7 100.0 136 Missing 81.8 18.2 0.0 0.0 100.0 11 HIV testing status Ever tested and knows results of last test 96.2 3.5 0.3 0.0 100.0 858 Ever tested, does not results 100.0 0.0 0.0 0.0 100.0 82 Never tested 98.6 1.1 0.2 0.1 100.0 2,522 Missing 100.0 0.0 0.0 0.0 100.0 1 Total 98.0 1.7 0.2 0.1 100.0 3,463 Appendix B – Estimates of Sampling Errors | 267 ESTIMATES OF SAMPLING ERRORS APPENDIX B The estimates from a sample survey are affected by two types of errors: (1) nonsampling errors, and (2) sampling errors. Non-sampling errors are the results of mistakes made in implementing data collection and data processing, such as failure to locate and interview the correct household, misunderstanding of the questions on the part of either the interviewer or the respondent, and data entry errors. Although numerous efforts were made during the implementation of the 2005 RDHS to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically. Sampling errors, on the other hand, can be evaluated statistically. The sample of respondents selected in the 2005 RDHS is only one of many samples that could have been selected from the same population, using the same design and expected size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results. A sampling error is usually measured in terms of the standard error for a particular statistic (mean, percentage, etc.), which is the square root of the variance. The standard error can be used to calculate confidence intervals within which the true value for the population can reasonably be assumed to fall. For example, for any given statistic calculated from a sample survey, the value of that statistic will fall within a range of plus or minus two times the standard error of that statistic in 95 percent of all possible samples of identical size and design. If the sample of respondents had been selected as a simple random sample, it would have been possible to use straightforward formulas for calculating sampling errors. However, the 2005 RDHS-III sample is the result of a multistage stratified design, and, consequently, it was necessary to use more complex formula. The computer software used to calculate sampling errors for the 2005 RDHS-III is the ISSA Sampling Error Module. This module used the Taylor linearization method of variance estimation for survey estimates that are means or proportions. The Jackknife repeated replication method is used for variance estimation of more complex statistics such as fertility and mortality rates. The Taylor linearization method treats any percentage or average as a ratio estimate, r = y/x, where y represents the total sample value for variable y, and x represents the total number of cases in the group or subgroup under consideration. The variance of r is computed using the formula given below, with the standard error being the square root of the variance: ∑ ∑ = = ⎥⎦ ⎤⎢⎣ ⎡ ⎟⎟⎠ ⎞ ⎜⎜⎝ ⎛ − − − == H h h h m i hi h h m zz m m x frvarrSE h 1 2 1 2 2 2 1 1)()( in which hihihi rxyz −= , and hhh rxyz −= 268 | Appendix B – Estimates of Sampling Errors where h represents the stratum which varies from 1 to H, mh is the total number of clusters selected in the hth stratum, yhi is the sum of the weighted values of variable y in the ith cluster in the hth stratum, xhi is the sum of the weighted number of cases in the ith cluster in the hth stratum, and f is the overall sampling fraction, which is so small that it is ignored. The Jackknife repeated replication method derives estimates of complex rates from each of several replications of the parent sample, and calculates standard errors for these estimates using simple formulae. Each replication considers all but one clusters in the calculation of the estimates. Pseudo- independent replications are thus created. In the 2005 RDHS-III, there were 462 non-empty clusters. Hence, 461 replications were created. The variance of a rate r is calculated as follows: SE r var r k k r r i k i 2 1 21 1 ( ) ( ) ( ) ( )= = − − = ∑ in which )()1( ii rkkrr −−= where r is the estimate computed from the full sample of 462 clusters, r(i) is the estimate computed from the reduced sample of 461 clusters (ith cluster excluded), and k is the total number of clusters. In addition to the standard error, ISSA computes the design effect (DEFT) for each estimate, which is defined as the ratio between the standard error using the given sample design and the standard error that would result if a simple random sample had been used. A DEFT value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a value greater than 1.0 indicates the increase in the sampling error due to the use of a more complex and less statistically efficient design. ISSA also computes the relative error and confidence limits for the estimates. Sampling errors for the 2005 RDHS-III are calculated for selected variables considered to be of primary interest for woman’s survey and for man’s surveys, respectively. The results are presented in this appendix for the country as a whole, for urban and rural areas, and for each of the five provinces. For each variable, the type of statistic (mean, proportion, or rate) and the base population are given in Table B.1. Tables B.2 to B.9 present the value of the statistic (R), its standard error (SE), the number of unweighted (N) and weighted (WN) cases, the design effect (DEFT), the relative standard error (SE/R), and the 95 percent confidence limits (R±2SE), for each variable. The DEFT is considered undefined when the standard error considering simple random sample is zero (when the estimate is close to 0 or 1). In the case of the total fertility rate, the number of unweighted cases is not relevant, as there is no known unweighted value for woman-years of exposure to child-bearing. The confidence interval (e.g., as calculated for children surviving) can be interpreted as follows: the overall average from the national sample is 2.141 and its standard error is 0.022. Therefore, to obtain the 95 percent confidence limits, one adds and subtracts twice the standard error to the sample estimate, i.e., 2.141±2×0.022. There is a high probability (95 percent) that the true average number of children surviving is between 2.141 - 2×0.022 and 2.141 + 2×0.022, that is, between 2.097 and 2.185. Appendix B – Estimates of Sampling Errors | 269 Sampling errors are analyzed for the national woman sample and for two separate groups of estimates: (1) means and proportions, and (2) complex demographic rates. The relative standard errors (SE/R) for the means and proportions range between 0.3 percent and 17.3 percent with an average of 3.8 percent; the highest relative standard errors are for estimates of very low values (e.g., women currently using IUD). If estimates of very low values (less than 10 percent) were removed, then the average drops to 2.6 percent. So in general, the relative standard error for most estimates for the country as a whole is small, except for estimates of very small proportions. The relative standard error for the total fertility rate is small, 1.6 percent. However, for the mortality rates, the averaged relative standard error is much higher, 5.2 percent. There are differentials in the relative standard error for the estimates of sub-populations. For example, for the variable Children ever born to women 40-49, the relative standard errors as a percent of the estimated mean for the whole country and for rural and urban areas are 1.0 percent, 1.1 percent and 2.4 percent, respectively. For the total sample, the value of the design effect (DEFT), averaged over all variables, is 1.22 which means that, due to multi-stage clustering of the sample, the average standard error is increased by a factor of 1.22 over that in an equivalent simple random sample. 270 | Appendix B – Estimates of Sampling Errors Table B.1 List of selected variables for sampling errors, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Variable Estimate Base Population –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence Proportion All women 15-49 Literate Proportion All women 15-49 No education Proportion All women 15-49 Secondary education or higher Proportion All women 15-49 Never married/in union Proportion All women 15-49 Currently married/in union Proportion All women 15-49 Married before age 20 Proportion Women 20-49 Currently pregnant Proportion All women 15-49 Children ever born Mean All women 15-49 Children ever born to women 40-49 Mean Women 40-49 Children surviving Mean All women 15-49 Knows any contraceptive method Proportion Currently married women 15-49 Ever used any contraceptive method Proportion Currently married women 15-49 Currently using any contraceptive method Proportion Currently married women 15-49 Currently using pill Proportion Currently married women 15-49 Currently using condom Proportion Currently married women 15-49 Currently using female sterilization Proportion Currently married women 15-49 Currently using periodic abstinence Proportion Currently married women 15-49 Obtained method from public sector source Proportion Current users of modern methods Want no more children Proportion Currently married women 15-49 Want to delay birth at least two years Proportion Currently married women 15-49 Ideal number of children Mean All women 15-49 Mothers received tetanus injection for last birth Proportion Most recent births in the last 5 years Mothers received medical assistance at delivery Proportion Births in the last 5 years Child had diarrhea in the 2 weeks prior to survey Proportion Children under 5 Treated with oral rehydration salts (ORS) Proportion Children with diarrhea in two weeks before interview Taken to a health provider Proportion Children with diarrhea in two weeks before interview Vaccination card seen Proportion Children age 12-23 months Received BCG Proportion Children age 12-23 months Received DPT (3 doses) Proportion Children age 12-23 months Received polio (3 doses) Proportion Children age 12-23 months Received measles Proportion Children age 12-23 months Fully immunized Proportion Children age 12-23 months Weight-for-height (below -2SD) Proportion Children under 5 who were measured Height-for-age (below -2SD) Proportion Children under 5 who were measured Weight-for-age (below -2SD) Proportion Children under 5 who were measured Anemia among children Proportion Children age 6-59 months Anemia among women Proportion All women 15-49 BMI <18.5 Proportion All women 15-49 Total Fertility Rate (0-3 years) Rate All women Neonatal mortality¹ Rate Number of births in past 5 (10) years Postneonatal mortality¹ Rate Number of births in past 5 (10) years Infant mortality¹ Rate Number of births in past 5 (10) years Child mortality¹ Rate Number of births in past 5 (10) years Under-five mortality¹ Rate Number of births in past 5 (10) years Maternal mortality (0-9 years)² Rate Number of births in past 10 years HIV prevalence Proportion All women 15-49 tested for HIV –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence Proportion All men 15-59 Literate Proportion All men 15-59 No education Proportion All men 15-59 Secondary education or higher Proportion All men 15-59 Never married/in union Proportion All men 15-59 Currently married/in union Proportion All men 15-59 HIV prevalence (15-49) Proportion All men 15-49 tested for HIV HIV prevalence (15-59) Proportion All men 15-59 tested for HIV –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) Proportion All women and men 15-49 tested for HIV –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– ¹ Past 5 years for national-level rate and past 10 years for residence- and regional-level rates ² Maternal mortality rate is only calculated at the national level. Appendix B – Estimates of Sampling Errors | 271 Table B.2 Sampling errors – National sample, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.170 0.005 11,321 11,321 1.420 0.030 0.160 0.180 Literate 0.703 0.006 11,321 11,321 1.426 0.009 0.691 0.715 No education 0.234 0.006 11,321 11,321 1.448 0.025 0.222 0.245 Secondary education or higher 0.096 0.004 11,321 11,321 1.585 0.046 0.087 0.105 Never married/in union 0.377 0.006 11,321 11,321 1.209 0.015 0.366 0.388 Currently married/in union 0.487 0.006 11,321 11,321 1.225 0.012 0.475 0.498 Married before age 20 0.419 0.007 6,370 6,383 1.178 0.017 0.404 0.433 Currently pregnant 0.080 0.003 11,321 11,321 1.103 0.035 0.074 0.085 Children ever born 2.683 0.028 11,321 11,321 1.025 0.011 2.627 2.740 Children ever born to women 40-49 6.565 0.067 2,032 2,045 1.132 0.010 6.431 6.699 Children surviving 2.141 0.022 11,321 11,321 0.985 0.010 2.097 2.185 Knows any contraceptive method 0.979 0.003 5,458 5,510 1.305 0.003 0.974 0.984 Ever used any contraceptive method 0.346 0.008 5,458 5,510 1.206 0.022 0.330 0.361 Currently using any contraceptive method 0.174 0.006 5,458 5,510 1.124 0.033 0.162 0.185 Currently using pill 0.024 0.003 5,458 5,510 1.384 0.118 0.019 0.030 Currently using condom 0.009 0.001 5,458 5,510 0.955 0.136 0.006 0.011 Currently using female sterilization 0.005 0.001 5,458 5,510 0.942 0.173 0.004 0.007 Currently using periodic abstinence 0.042 0.003 5,458 5,510 1.140 0.074 0.035 0.048 Obtained method from public sector source 0.726 0.021 621 592 1.164 0.029 0.684 0.768 Want no more children 0.427 0.007 5,458 5,510 1.077 0.017 0.413 0.442 Want to delay birth at least two years 0.388 0.007 5,458 5,510 1.025 0.017 0.375 0.402 Ideal number of children 4.283 0.022 10,937 10,899 1.404 0.005 4.240 4.327 Mothers received tetanus injection for last birth 0.634 0.007 5,393 5,425 1.078 0.011 0.620 0.648 Mothers received medical assistance at delivery 0.386 0.009 8,649 8,715 1.424 0.024 0.368 0.405 Child had diarrhea in the 2 weeks prior to survey 0.141 0.005 7,752 7,797 1.088 0.032 0.132 0.151 Treated with oral rehydration salts (ORS) 0.116 0.011 1,096 1,103 1.046 0.092 0.094 0.137 Taken to a health provider 0.141 0.011 1,096 1,103 1.007 0.078 0.119 0.163 Vaccination card seen 0.759 0.014 1,624 1,626 1.304 0.018 0.731 0.787 Received BCG 0.965 0.008 1,624 1,626 1.718 0.008 0.949 0.981 Received DPT (3 doses) 0.870 0.011 1,624 1,626 1.314 0.013 0.848 0.892 Received polio (3 doses) 0.843 0.012 1,624 1,626 1.331 0.014 0.819 0.867 Received measles 0.856 0.012 1,624 1,626 1.337 0.014 0.833 0.880 Fully immunized 0.752 0.014 1,624 1,626 1.287 0.018 0.724 0.780 Weight-for-height (below -2SD) 0.039 0.003 3,874 3,859 1.042 0.086 0.032 0.046 Height-for-age (below -2SD) 0.453 0.009 3,874 3,859 1.084 0.020 0.435 0.472 Weight-for-age (below -2SD) 0.225 0.008 3,874 3,859 1.106 0.035 0.209 0.240 Anemia among children 0.563 0.012 3,554 3,537 1.363 0.022 0.539 0.587 Anemia among women 0.328 0.012 5,638 5,657 1.898 0.036 0.304 0.352 BMI <18.5 0.098 0.004 5,083 5,100 0.960 0.041 0.090 0.106 Total Fertility Rate (0-3 years) 6.076 0.095 na 31,571 1.308 0.016 5.885 6.266 Neonatal mortality (0-4 years) 36.975 2.348 8,714 8,774 1.091 0.063 32.279 41.670 Postneonatal mortality (0-4 years) 49.144 3.002 8,751 8,808 1.233 0.061 43.140 55.147 Infant mortality (0-4 years) 86.118 3.976 8,757 8,815 1.245 0.046 78.166 94.071 Child mortality (0-4 years) 72.294 3.736 8,933 9,005 1.171 0.052 64.822 79.767 Under-five mortality (0-4 years) 152.187 5.410 8,982 9,052 1.303 0.036 141.366 163.007 Maternal mortality (0-9 years) 750 79 na na na 0.105 592 908 HIV prevalence 0.036 0.003 5,677 5,656 1.070 0.073 0.031 0.041 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.174 0.005 4,820 4,820 1.006 0.032 0.163 0.185 Literate 0.775 0.008 4,820 4,820 1.256 0.010 0.760 0.791 No education 0.174 0.007 4,820 4,820 1.267 0.040 0.160 0.188 Secondary education or higher 0.123 0.006 4,820 4,820 1.355 0.052 0.110 0.136 Never married/in union 0.456 0.008 4,820 4,820 1.180 0.019 0.439 0.473 Currently married/in union 0.519 0.008 4,820 4,820 1.166 0.016 0.502 0.535 HIV prevalence (15-49) 0.023 0.002 4,340 4,361 1.044 0.103 0.018 0.028 HIV prevalence (15-59) 0.022 0.002 4,742 4,763 1.049 0.101 0.018 0.027 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.030 0.002 10,017 10,016 1.186 0.067 0.026 0.035 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable 272 | Appendix B – Estimates of Sampling Errors Table B.3 Sampling errors – Urban sample, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 1.000 0.000 2,616 1,921 na 0.000 1.000 1.000 Literate 0.838 0.007 2,616 1,921 1.021 0.009 0.823 0.852 No education 0.135 0.007 2,616 1,921 1.111 0.055 0.120 0.149 Secondary education or higher 0.277 0.015 2,616 1,921 1.730 0.055 0.246 0.307 Never married/in union 0.470 0.012 2,616 1,921 1.186 0.025 0.447 0.493 Currently married/in union 0.387 0.012 2,616 1,921 1.213 0.030 0.364 0.410 Married before age 20 0.353 0.016 1,381 1,011 1.232 0.045 0.321 0.384 Currently pregnant 0.063 0.005 2,616 1,921 1.013 0.077 0.053 0.072 Children ever born 2.103 0.059 2,616 1,921 1.182 0.028 1.985 2.221 Children ever born to women 40-49 5.806 0.137 362 264 0.991 0.024 5.532 6.080 Children surviving 1.792 0.047 2,616 1,921 1.100 0.026 1.698 1.886 Knows any contraceptive method 0.993 0.003 1,026 744 1.040 0.003 0.987 0.998 Ever used any contraceptive method 0.529 0.019 1,026 744 1.191 0.035 0.492 0.566 Currently using any contraceptive method 0.316 0.021 1,026 744 1.458 0.067 0.273 0.358 Currently using pill 0.042 0.008 1,026 744 1.217 0.181 0.027 0.057 Currently using condom 0.040 0.006 1,026 744 1.031 0.159 0.027 0.052 Currently using female sterilization 0.011 0.003 1,026 744 1.061 0.321 0.004 0.017 Currently using periodic abstinence 0.069 0.012 1,026 744 1.512 0.173 0.045 0.093 Obtained method from public sector source 0.482 0.038 238 175 1.184 0.080 0.406 0.559 Want no more children 0.493 0.019 1,026 744 1.195 0.038 0.456 0.530 Want to delay birth at least two years 0.338 0.015 1,026 744 1.034 0.045 0.307 0.368 Ideal number of children 3.818 0.035 2,540 1,864 1.254 0.009 3.748 3.887 Mothers received tetanus injection for last birth 0.713 0.015 1,063 774 1.089 0.021 0.683 0.743 Mothers received medical assistance at delivery 0.631 0.021 1,701 1,228 1.439 0.033 0.589 0.673 Child had diarrhea in the 2 weeks prior to survey 0.127 0.010 1,582 1,144 1.189 0.083 0.106 0.148 Treated with oral rehydration salts (ORS) 0.146 0.028 203 145 1.075 0.190 0.090 0.202 Taken to a health provider 0.162 0.028 203 145 1.028 0.170 0.107 0.218 Vaccination card seen 0.693 0.038 308 214 1.387 0.054 0.618 0.769 Received BCG 0.976 0.009 308 214 1.044 0.010 0.958 0.995 Received DPT (3 doses) 0.849 0.026 308 214 1.195 0.030 0.797 0.901 Received polio (3 doses) 0.810 0.030 308 214 1.271 0.037 0.750 0.869 Received measles 0.896 0.019 308 214 1.032 0.021 0.859 0.933 Fully immunized 0.710 0.032 308 214 1.172 0.045 0.646 0.773 Weight-for-height (below -2SD) 0.038 0.007 780 543 0.925 0.196 0.023 0.052 Height-for-age (below -2SD) 0.331 0.021 780 543 1.116 0.062 0.289 0.372 Weight-for-age (below -2SD) 0.162 0.013 780 543 0.948 0.081 0.136 0.188 Anemia among children 0.543 0.022 718 495 1.095 0.041 0.498 0.587 Anemia among women 0.333 0.021 1,272 938 1.620 0.064 0.291 0.376 BMI <18.5 0.099 0.010 1,165 862 1.122 0.099 0.079 0.118 Total Fertility Rate (0-3 years) 4.908 0.168 na 5,289 1.074 0.034 4.571 5.244 Neonatal mortality (0-9 years) 31.849 3.421 3,218 2,335 0.970 0.107 25.006 38.692 Postneonatal mortality (0-9 years) 37.198 3.504 3,223 2,339 0.946 0.094 30.190 44.206 Infant mortality (0-9 years) 69.047 4.823 3,224 2,340 0.963 0.070 59.400 78.694 Child mortality (0-9 years) 57.386 5.895 3,249 2,355 1.132 0.103 45.595 69.176 Under-five mortality (0-9 years) 122.470 7.766 3,256 2,360 1.082 0.063 106.938 138.003 HIV prevalence 0.086 0.009 1,283 946 1.092 0.099 0.069 0.103 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 1.000 0.000 1,130 840 na 0.000 1.000 1.000 Literate 0.860 0.013 1,130 840 1.252 0.015 0.834 0.886 No education 0.095 0.009 1,130 840 0.999 0.092 0.077 0.112 Secondary education or higher 0.310 0.023 1,130 840 1.669 0.074 0.264 0.356 Never married/in union 0.547 0.015 1,130 840 1.003 0.027 0.517 0.576 Currently married/in union 0.420 0.014 1,130 840 0.940 0.033 0.392 0.447 HIV prevalence (15-49) 0.058 0.008 1,004 774 1.040 0.133 0.042 0.073 HIV prevalence (15-59) 0.056 0.007 1,077 830 1.066 0.134 0.041 0.070 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.073 0.006 2,287 1,720 1.192 0.089 0.060 0.086 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable Appendix B – Estimates of Sampling Errors | 273 Table B.4 Sampling errors – Rural sample, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.000 0.000 8,705 9,400 na na 0.000 0.000 Literate 0.676 0.007 8,705 9,400 1.431 0.011 0.661 0.690 No education 0.254 0.007 8,705 9,400 1.441 0.026 0.241 0.267 Secondary education or higher 0.059 0.004 8,705 9,400 1.529 0.066 0.051 0.067 Never married/in union 0.358 0.006 8,705 9,400 1.204 0.017 0.345 0.370 Currently married/in union 0.507 0.006 8,705 9,400 1.212 0.013 0.494 0.520 Married before age 20 0.431 0.008 4,989 5,372 1.165 0.019 0.415 0.448 Currently pregnant 0.083 0.003 8,705 9,400 1.092 0.039 0.077 0.089 Children ever born 2.802 0.032 8,705 9,400 0.991 0.011 2.738 2.865 Children ever born to women 40-49 6.678 0.074 1,670 1,781 1.141 0.011 6.530 6.826 Children surviving 2.212 0.025 8,705 9,400 0.958 0.011 2.163 2.261 Knows any contraceptive method 0.977 0.003 4,432 4,766 1.280 0.003 0.971 0.983 Ever used any contraceptive method 0.317 0.008 4,432 4,766 1.188 0.026 0.300 0.334 Currently using any contraceptive method 0.152 0.006 4,432 4,766 1.038 0.037 0.140 0.163 Currently using pill 0.022 0.003 4,432 4,766 1.429 0.144 0.015 0.028 Currently using condom 0.004 0.001 4,432 4,766 1.040 0.243 0.002 0.006 Currently using female sterilization 0.005 0.001 4,432 4,766 0.924 0.205 0.003 0.006 Currently using periodic abstinence 0.037 0.003 4,432 4,766 1.060 0.081 0.031 0.043 Obtained method from public sector source 0.828 0.023 383 417 1.196 0.028 0.782 0.874 Want no more children 0.417 0.008 4,432 4,766 1.051 0.019 0.401 0.433 Want to delay birth at least two years 0.396 0.007 4,432 4,766 1.012 0.019 0.381 0.411 Ideal number of children 4.379 0.025 8,397 9,035 1.398 0.006 4.329 4.429 Mothers received tetanus injection for last birth 0.621 0.008 4,330 4,651 1.053 0.013 0.605 0.636 Mothers received medical assistance at delivery 0.346 0.010 6,948 7,487 1.421 0.029 0.326 0.366 Child had diarrhea in the 2 weeks prior to survey 0.144 0.005 6,170 6,653 1.056 0.035 0.134 0.154 Treated with oral rehydration salts (ORS) 0.111 0.012 893 958 1.029 0.104 0.088 0.134 Taken to a health provider 0.138 0.012 893 958 0.986 0.086 0.114 0.162 Vaccination card seen 0.769 0.015 1,316 1,412 1.277 0.019 0.739 0.799 Received BCG 0.963 0.009 1,316 1,412 1.708 0.009 0.945 0.981 Received DPT (3 doses) 0.873 0.012 1,316 1,412 1.312 0.014 0.849 0.897 Received polio (3 doses) 0.848 0.013 1,316 1,412 1.322 0.016 0.822 0.874 Received measles 0.850 0.013 1,316 1,412 1.326 0.016 0.824 0.877 Fully immunized 0.758 0.015 1,316 412 1.282 0.020 0.728 0.789 Weight-for-height (below -2SD) 0.039 0.004 3,094 3,316 1.049 0.095 0.032 0.047 Height-for-age (below -2SD) 0.473 0.010 3,094 3,316 1.066 0.021 0.453 0.494 Weight-for-age (below -2SD) 0.235 0.009 3,094 3,316 1.096 0.038 0.217 0.253 Anemia among children 0.566 0.014 2,836 3,042 1.372 0.024 0.539 0.593 Anemia among women 0.327 0.014 4,366 4,719 1.910 0.041 0.300 0.354 BMI <18.5 0.098 0.004 3,918 4,238 0.924 0.045 0.089 0.107 Total Fertility Rate (0-3 years) 6.306 0.105 na 25,961 1.300 0.017 6.095 6.516 Neonatal mortality (0-9 years) 46.080 2.071 13,351 14,380 1.002 0.045 41.939 50.221 Postneonatal mortality (0-9 years) 61.948 2.830 13,372 14,402 1.267 0.046 56.288 67.608 Infant mortality (0-9 years) 108.028 3.598 13,377 14,408 1.207 0.033 100.833 115.223 Child mortality (0-9 years) 94.199 3.906 13,557 14,598 1.204 0.041 86.387 102.010 Under-five mortality (0-9 years) 192.051 5.370 13,588 14,632 1.321 0.028 181.311 202.790 HIV prevalence 0.026 0.003 4,394 4,710 1.089 0.100 0.021 0.031 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.000 0.000 3,690 3,980 na na 0.000 0.000 Literate 0.758 0.009 3,690 3,980 1.229 0.011 0.740 0.775 No education 0.191 0.008 3,690 3,980 1.263 0.043 0.174 0.207 Secondary education or higher 0.083 0.006 3,690 3,980 1.260 0.069 0.072 0.095 Never married/in union 0.436 0.010 3,690 3,980 1.204 0.023 0.417 0.456 Currently married/in union 0.539 0.010 3,690 3,980 1.196 0.018 0.520 0.559 HIV prevalence (15-49) 0.016 0.002 3,336 3,587 1.105 0.151 0.011 0.021 HIV prevalence (15-59) 0.015 0.002 3,665 3,934 1.100 0.145 0.011 0.020 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.022 0.002 7,730 8,297 1.222 0.094 0.018 0.026 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable 274 | Appendix B – Estimates of Sampling Errors Table B.5 Sampling errors – City of Kigali, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.854 0.014 1,329 1,127 1.486 0.017 0.825 0.883 Literate 0.850 0.013 1,329 1,127 1.344 0.015 0.824 0.876 No education 0.113 0.015 1,329 1,127 1.669 0.128 0.084 0.142 Secondary education or higher 0.301 0.020 1,329 1,127 1.619 0.068 0.260 0.342 Never married/in union 0.482 0.017 1,329 1,127 1.226 0.035 0.448 0.515 Currently married/in union 0.361 0.016 1,329 1,127 1.241 0.045 0.328 0.393 Married before age 20 0.356 0.021 668 573 1.150 0.060 0.314 0.399 Currently pregnant 0.069 0.007 1,329 1,127 1.006 0.101 0.055 0.083 Children ever born 1.894 0.067 1,329 1,127 1.026 0.035 1.760 2.027 Children ever born to women 40-49 5.914 0.185 147 132 0.950 0.031 5.543 6.284 Children surviving 1.611 0.058 1,329 1,127 1.046 0.036 1.495 1.727 Knows any contraceptive method 0.995 0.003 481 407 1.038 0.003 0.989 1.000 Ever used any contraceptive method 0.590 0.023 481 407 1.031 0.039 0.544 0.636 Currently using any contraceptive method 0.355 0.031 481 407 1.440 0.089 0.292 0.418 Currently using pill 0.042 0.011 481 407 1.203 0.262 0.020 0.064 Currently using condom 0.052 0.011 481 407 1.064 0.208 0.030 0.073 Currently using female sterilization 0.013 0.005 481 407 1.019 0.401 0.003 0.024 Currently using periodic abstinence 0.079 0.019 481 407 1.521 0.237 0.042 0.117 Obtained method from public sector source 0.332 0.046 132 105 1.128 0.140 0.239 0.425 Want no more children 0.520 0.027 481 407 1.182 0.052 0.466 0.574 Want to delay birth at least two years 0.321 0.018 481 407 0.864 0.057 0.284 0.358 Ideal number of children 3.694 0.046 1,295 1,096 1.207 0.012 3.603 3.785 Mothers received tetanus injection for last birth 0.760 0.018 502 427 0.953 0.024 0.724 0.796 Mothers received medical assistance at delivery 0.618 0.031 772 655 1.425 0.050 0.556 0.679 Child had diarrhea in the 2 weeks prior to survey 0.112 0.015 711 599 1.231 0.132 0.082 0.141 Treated with oral rehydration salts (ORS) 0.203 0.050 78 67 1.106 0.248 0.103 0.304 Taken to a health provider 0.186 0.044 78 67 0.997 0.237 0.098 0.274 Vaccination card seen 0.690 0.056 127 103 1.321 0.081 0.578 0.801 Received BCG 0.974 0.015 127 103 1.042 0.015 0.944 1.000 Received DPT (3 doses) 0.806 0.037 127 103 1.031 0.046 0.732 0.881 Received polio (3 doses) 0.764 0.046 127 103 1.179 0.060 0.673 0.855 Received measles 0.854 0.033 127 103 1.027 0.039 0.787 0.920 Fully immunized 0.617 0.048 127 103 1.090 0.079 0.520 0.714 Weight-for-height (below -2SD) 0.075 0.015 312 250 0.798 0.196 0.046 0.105 Height-for-age (below -2SD) 0.292 0.030 312 250 1.063 0.103 0.232 0.352 Weight-for-age (below -2SD) 0.144 0.020 312 250 0.916 0.139 0.104 0.185 Anemia among children 0.696 0.033 286 226 1.056 0.048 0.629 0.762 Anemia among women 0.459 0.043 640 547 2.212 0.095 0.372 0.545 BMI <18.5 0.097 0.013 576 493 1.072 0.135 0.071 0.124 Total Fertility Rate (0-3 years) 4.301 0.230 na 3,110 1.148 0.053 3.842 4.760 Neonatal mortality (0-9 years) 27.489 4.567 1,393 1,197 0.977 0.166 18.356 36.622 Postneonatal mortality (0-9 years) 40.187 6.168 1,394 1,198 1.121 0.153 27.850 52.523 Infant mortality (0-9 years) 67.675 8.229 1,395 1,199 1.151 0.122 51.217 84.134 Child mortality (0-9 years) 60.261 9.950 1,398 1,202 1.253 0.165 40.360 80.162 Under-five mortality (0-9 years) 123.858 15.460 1,401 1,204 1.498 0.125 92.939 154.778 HIV prevalence 0.080 0.012 647 556 1.156 0.154 0.055 0.104 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.881 0.017 619 523 1.314 0.019 0.847 0.915 Literate 0.860 0.018 619 523 1.293 0.021 0.824 0.896 No education 0.099 0.012 619 523 1.000 0.122 0.075 0.123 Secondary education or higher 0.338 0.029 619 523 1.522 0.086 0.280 0.396 Never married/in union 0.592 0.020 619 523 1.009 0.034 0.552 0.632 Currently married/in union 0.378 0.018 619 523 0.914 0.047 0.342 0.413 HIV prevalence (15-49) 0.052 0.010 542 487 1.064 0.196 0.031 0.072 HIV prevalence (15-59) 0.049 0.010 575 517 1.080 0.199 0.029 0.068 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.067 0.009 1,189 1,043 1.254 0.136 0.048 0.085 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable Appendix B – Estimates of Sampling Errors | 275 Table B.6 Sampling errors – South Province, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.139 0.005 2,760 2,958 0.808 0.038 0.129 0.150 Literate 0.728 0.011 2,760 2,958 1.283 0.015 0.706 0.750 No education 0.203 0.010 2,760 2,958 1.334 0.050 0.183 0.224 Secondary education or higher 0.083 0.008 2,760 2,958 1.486 0.094 0.068 0.099 Never married/in union 0.385 0.009 2,760 2,958 0.950 0.023 0.367 0.402 Currently married/in union 0.477 0.010 2,760 2,958 1.045 0.021 0.457 0.497 Married before age 20 0.300 0.013 1,611 1,728 1.116 0.042 0.275 0.326 Currently pregnant 0.076 0.005 2,760 2,958 1.083 0.072 0.065 0.087 Children ever born 2.574 0.047 2,760 2,958 0.878 0.018 2.480 2.669 Children ever born to women 40-49 6.116 0.129 554 605 1.172 0.021 5.857 6.375 Children surviving 2.059 0.038 2,760 2,958 0.879 0.019 1.982 2.135 Knows any contraceptive method 0.985 0.003 1,327 1,411 0.754 0.003 0.980 0.990 Ever used any contraceptive method 0.344 0.014 1,327 1,411 1.073 0.041 0.316 0.372 Currently using any contraceptive method 0.148 0.011 1,327 1,411 1.079 0.071 0.127 0.169 Currently using pill 0.017 0.006 1,327 1,411 1.696 0.356 0.005 0.029 Currently using condom 0.007 0.002 1,327 1,411 0.922 0.294 0.003 0.012 Currently using female sterilization 0.004 0.002 1,327 1,411 1.089 0.460 0.000 0.008 Currently using periodic abstinence 0.034 0.005 1,327 1,411 0.957 0.140 0.025 0.044 Obtained method from public sector source 0.791 0.037 109 120 0.952 0.047 0.716 0.865 Want no more children 0.407 0.015 1,327 1,411 1.110 0.037 0.377 0.437 Want to delay birth at least two years 0.420 0.013 1,327 1,411 0.958 0.031 0.394 0.446 Ideal number of children 4.275 0.036 2,711 2,902 1.251 0.008 4.204 4.347 Mothers received tetanus injection for last birth 0.644 0.014 1,284 1,357 1.029 0.022 0.616 0.672 Mothers received medical assistance at delivery 0.399 0.013 2,020 2,122 0.950 0.032 0.374 0.424 Child had diarrhea in the 2 weeks prior to survey 0.145 0.008 1,821 1,909 0.896 0.055 0.129 0.161 Treated with oral rehydration salts (ORS) 0.066 0.017 277 277 1.061 0.256 0.032 0.100 Taken to a health provider 0.109 0.018 277 277 0.929 0.169 0.072 0.146 Vaccination card seen 0.764 0.025 384 393 1.117 0.033 0.714 0.814 Received BCG 0.983 0.006 384 393 0.909 0.006 0.971 0.995 Received DPT (3 doses) 0.925 0.013 384 393 0.921 0.014 0.899 0.950 Received polio (3 doses) 0.888 0.016 384 393 0.973 0.018 0.856 0.920 Received measles 0.941 0.013 384 393 1.050 0.014 0.915 0.967 Fully immunized 0.843 0.019 384 393 0.999 0.023 0.805 0.881 Weight-for-height (below -2SD) 0.050 0.009 938 987 1.167 0.171 0.033 0.067 Height-for-age (below -2SD) 0.448 0.019 938 987 1.108 0.042 0.410 0.487 Weight-for-age (below -2SD) 0.276 0.018 938 987 1.155 0.066 0.240 0.312 Anemia among children 0.472 0.020 864 908 1.126 0.043 0.431 0.512 Anemia among women 0.280 0.016 1,405 1,518 1.341 0.057 0.248 0.312 BMI <18.5 0.131 0.007 1,268 1,367 0.791 0.057 0.116 0.146 Total Fertility Rate (0-3 years) 5.646 0.172 na 8,251 1.222 0.030 5.302 5.989 Neonatal mortality (0-9 years) 47.648 3.762 3,907 4,130 0.942 0.079 40.124 55.172 Postneonatal mortality (0-9 years) 59.047 5.005 3,912 4,136 1.289 0.085 49.037 69.056 Infant mortality (0-9 years) 106.694 6.352 3,914 4,138 1.159 0.060 93.991 119.397 Child mortality (0-9 years) 79.591 6.497 3,969 4,196 1.176 0.082 66.596 92.585 Under-five mortality (0-9 years) 177.793 9.152 3,978 4,207 1.233 0.051 159.489 196.097 HIV prevalence 0.031 0.005 1,408 1,501 1.020 0.151 0.022 0.041 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.131 0.007 1,152 1,250 0.753 0.057 0.116 0.146 Literate 0.746 0.015 1,152 1,250 1.156 0.020 0.716 0.775 No education 0.164 0.014 1,152 1,250 1.314 0.088 0.135 0.192 Secondary education or higher 0.105 0.011 1,152 1,250 1.262 0.109 0.082 0.128 Never married/in union 0.468 0.014 1,152 1,250 0.980 0.031 0.439 0.497 Currently married/in union 0.505 0.014 1,152 1,250 0.962 0.028 0.477 0.533 HIV prevalence (15-49) 0.020 0.004 1,045 1,126 0.829 0.179 0.013 0.027 HIV prevalence (15-59) 0.018 0.003 1,141 1,235 0.830 0.180 0.012 0.025 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.027 0.004 2,453 2,627 1.099 0.134 0.019 0.034 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable 276 | Appendix B – Estimates of Sampling Errors Table B.7 Sampling errors – West Province, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.079 0.009 2,971 2,824 1.805 0.113 0.061 0.096 Literate 0.658 0.014 2,971 2,824 1.641 0.022 0.629 0.687 No education 0.281 0.014 2,971 2,824 1.656 0.049 0.253 0.308 Secondary education or higher 0.062 0.008 2,971 2,824 1.789 0.128 0.046 0.078 Never married/in union 0.377 0.012 2,971 2,824 1.332 0.031 0.353 0.401 Currently married/in union 0.505 0.012 2,971 2,824 1.283 0.023 0.482 0.529 Married before age 20 0.447 0.014 1,639 1,550 1.112 0.031 0.420 0.475 Currently pregnant 0.079 0.006 2,971 2,824 1.162 0.073 0.068 0.091 Children ever born 2.834 0.061 2,971 2,824 1.061 0.021 2.713 2.955 Children ever born to women 40-49 7.110 0.114 561 521 1.032 0.016 6.882 7.338 Children surviving 2.264 0.047 2,971 2,824 1.016 0.021 2.170 2.357 Knows any contraceptive method 0.960 0.007 1,478 1,427 1.383 0.007 0.946 0.974 Ever used any contraceptive method 0.276 0.016 1,478 1,427 1.386 0.058 0.244 0.308 Currently using any contraceptive method 0.145 0.010 1,478 1,427 1.134 0.072 0.124 0.166 Currently using pill 0.022 0.003 1,478 1,427 0.867 0.150 0.015 0.029 Currently using condom 0.004 0.002 1,478 1,427 1.020 0.415 0.001 0.007 Currently using female sterilization 0.010 0.002 1,478 1,427 0.834 0.221 0.005 0.014 Currently using periodic abstinence 0.023 0.005 1,478 1,427 1.221 0.206 0.014 0.033 Obtained method from public sector source 0.873 0.027 153 139 0.999 0.031 0.819 0.927 Want no more children 0.395 0.015 1,478 1,427 1.158 0.037 0.365 0.424 Want to delay birth at least two years 0.374 0.015 1,478 1,427 1.162 0.039 0.344 0.403 Ideal number of children 4.574 0.053 2,821 2,673 1.542 0.012 4.469 4.680 Mothers received tetanus injection for last birth 0.598 0.012 1,442 1,395 0.932 0.020 0.574 0.622 Mothers received medical assistance at delivery 0.344 0.017 2,352 2,290 1.443 0.051 0.309 0.379 Child had diarrhea in the 2 weeks prior to survey 0.137 0.009 2,133 2,075 1.177 0.067 0.119 0.155 Treated with oral rehydration salts (ORS) 0.170 0.028 276 284 1.216 0.165 0.114 0.226 Taken to a health provider 0.132 0.021 276 284 1.009 0.156 0.091 0.174 Vaccination card seen 0.760 0.025 454 440 1.274 0.033 0.709 0.810 Received BCG 0.967 0.010 454 440 1.213 0.010 0.947 0.987 Received DPT (3 doses) 0.844 0.018 454 440 1.034 0.021 0.809 0.879 Received polio (3 doses) 0.826 0.021 454 440 1.180 0.025 0.785 0.868 Received measles 0.825 0.020 454 440 1.131 0.024 0.785 0.865 Fully immunized 0.720 0.026 454 440 1.221 0.035 0.669 0.772 Weight-for-height (below -2SD) 0.028 0.005 1,044 999 0.994 0.178 0.018 0.039 Height-for-age (below -2SD) 0.469 0.018 1,044 999 1.132 0.038 0.433 0.505 Weight-for-age (below -2SD) 0.203 0.016 1,044 999 1.221 0.079 0.171 0.235 Anemia among children 0.593 0.023 973 933 1.390 0.039 0.547 0.638 Anemia among women 0.262 0.016 1,466 1,397 1.412 0.062 0.230 0.295 BMI <18.5 0.081 0.007 1,345 1,280 1.008 0.093 0.066 0.096 Total Fertility Rate (0-3 years) 6.638 0.200 na 7,726 1.406 0.030 6.239 7.037 Neonatal mortality (0-9 years) 43.194 3.389 4,526 4,387 1.041 0.078 36.415 49.973 Postneonatal mortality (0-9 years) 57.251 5.362 4,530 4,392 1.420 0.094 46.526 67.975 Infant mortality (0-9 years) 100.445 5.948 4,532 4,393 1.236 0.059 88.549 112.341 Child mortality (0-9 years) 87.219 7.300 4,590 4,447 1.368 0.084 72.618 101.819 Under-five mortality (0-9 years) 178.903 8.845 4,598 4,455 1.315 0.049 161.213 196.592 HIV prevalence 0.037 0.005 1,475 1,406 1.080 0.143 0.026 0.048 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.078 0.008 1,237 1,185 1.094 0.107 0.061 0.095 Literate 0.773 0.015 1,237 1,185 1.282 0.020 0.743 0.804 No education 0.178 0.013 1,237 1,185 1.211 0.074 0.151 0.204 Secondary education or higher 0.106 0.012 1,237 1,185 1.372 0.113 0.082 0.130 Never married/in union 0.414 0.019 1,237 1,185 1.337 0.045 0.377 0.452 Currently married/in union 0.560 0.019 1,237 1,185 1.354 0.034 0.522 0.598 HIV prevalence (15-49) 0.024 0.005 1,096 1,051 0.986 0.189 0.015 0.034 HIV prevalence (15-59) 0.023 0.004 1,220 1,169 0.980 0.182 0.015 0.032 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.032 0.004 2,571 2,458 1.209 0.132 0.023 0.040 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable Appendix B – Estimates of Sampling Errors | 277 Table B.8 Sampling errors – North Province, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.098 0.011 1,821 2,063 1.622 0.115 0.076 0.121 Literate 0.695 0.016 1,821 2,063 1.441 0.022 0.664 0.726 No education 0.254 0.014 1,821 2,063 1.359 0.055 0.227 0.282 Secondary education or higher 0.094 0.012 1,821 2,063 1.781 0.130 0.070 0.118 Never married/in union 0.348 0.013 1,821 2,063 1.179 0.038 0.321 0.374 Currently married/in union 0.513 0.012 1,821 2,063 1.064 0.024 0.488 0.538 Married before age 20 0.475 0.020 1,055 1,205 1.296 0.042 0.435 0.515 Currently pregnant 0.079 0.006 1,821 2,063 1.026 0.082 0.066 0.092 Children ever born 2.936 0.075 1,821 2,063 1.052 0.026 2.786 3.086 Children ever born to women 40-49 6.745 0.166 349 397 1.124 0.025 6.414 7.077 Children surviving 2.394 0.054 1,821 2,063 0.938 0.023 2.285 2.502 Knows any contraceptive method 0.984 0.005 921 1,058 1.247 0.005 0.973 0.994 Ever used any contraceptive method 0.315 0.021 921 1,058 1.361 0.066 0.273 0.356 Currently using any contraceptive method 0.160 0.013 921 1,058 1.066 0.081 0.134 0.185 Currently using pill 0.028 0.007 921 1,058 1.274 0.248 0.014 0.042 Currently using condom 0.007 0.002 921 1,058 0.791 0.320 0.002 0.011 Currently using female sterilization 0.002 0.002 921 1,058 1.004 0.700 0.000 0.005 Currently using periodic abstinence 0.035 0.007 921 1,058 1.083 0.188 0.022 0.048 Obtained method from public sector source 0.809 0.049 119 121 1.367 0.061 0.710 0.908 Want no more children 0.441 0.017 921 1,058 1.040 0.039 0.407 0.475 Want to delay birth at least two years 0.395 0.016 921 1,058 1.006 0.041 0.363 0.428 Ideal number of children 4.291 0.056 1,721 1,941 1.382 0.013 4.179 4.403 Mothers received tetanus injection for last birth 0.610 0.020 921 1,052 1.232 0.032 0.571 0.650 Mothers received medical assistance at delivery 0.341 0.026 1,493 1,716 1.736 0.076 0.289 0.393 Child had diarrhea in the 2 weeks prior to survey 0.145 0.012 1,366 1,571 1.152 0.080 0.121 0.168 Treated with oral rehydration salts (ORS) 0.147 0.021 202 227 0.748 0.140 0.106 0.188 Taken to a health provider 0.225 0.033 202 227 1.059 0.147 0.159 0.291 Vaccination card seen 0.766 0.030 295 340 1.208 0.039 0.706 0.825 Received BCG 0.990 0.006 295 340 0.981 0.006 0.978 1.000 Received DPT (3 doses) 0.903 0.018 295 340 1.016 0.020 0.868 0.939 Received polio (3 doses) 0.866 0.021 295 340 1.055 0.024 0.824 0.908 Received measles 0.921 0.017 295 340 1.034 0.018 0.888 0.954 Fully immunized 0.812 0.025 295 340 1.095 0.031 0.761 0.862 Weight-for-height (below -2SD) 0.029 0.007 709 793 1.065 0.233 0.015 0.042 Height-for-age (below -2SD) 0.522 0.022 709 793 1.102 0.043 0.477 0.567 Weight-for-age (below -2SD) 0.236 0.018 709 793 1.017 0.075 0.201 0.272 Anemia among children 0.562 0.034 655 729 1.633 0.061 0.494 0.630 Anemia among women 0.316 0.044 905 1,020 2.842 0.139 0.228 0.404 BMI <18.5 0.066 0.010 799 905 1.137 0.152 0.046 0.086 Total Fertility Rate (0-3 years) 6.353 0.224 na 5,702 1.173 0.035 5.904 6.802 Neonatal mortality (0-9 years) 42.308 4.265 2,924 3,358 1.004 0.101 33.779 50.837 Postneonatal mortality (0-9 years) 46.764 4.864 2,928 3,363 1.178 0.104 37.037 56.492 Infant mortality (0-9 years) 89.072 6.439 2,928 3,363 1.122 0.072 76.195 101.950 Child mortality (0-9 years) 77.446 7.002 2,965 3,404 1.130 0.090 63.441 91.451 Under-five mortality (0-9 years) 159.620 10.422 2,969 3,409 1.322 0.065 138.775 180.464 HIV prevalence 0.026 0.006 907 1,019 1.143 0.233 0.014 0.038 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.089 0.010 746 845 0.972 0.114 0.069 0.110 Literate 0.760 0.020 746 845 1.293 0.027 0.720 0.801 No education 0.201 0.021 746 845 1.434 0.105 0.159 0.243 Secondary education or higher 0.097 0.017 746 845 1.554 0.174 0.063 0.131 Never married/in union 0.420 0.022 746 845 1.209 0.052 0.376 0.464 Currently married/in union 0.561 0.023 746 845 1.278 0.041 0.514 0.607 HIV prevalence (15-49) 0.011 0.004 682 773 1.035 0.372 0.003 0.020 HIV prevalence (15-59) 0.012 0.004 743 840 1.029 0.348 0.004 0.020 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.020 0.004 1,589 1,792 1.243 0.221 0.011 0.028 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable 278 | Appendix B – Estimates of Sampling Errors Table B.9 Sampling errors – East Province, Rwanda 2005 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Number of cases ––––––––––––––––– Confidence Standard Un- Design Relative limits Value error weighted Weighted effect error –––––––––––––– Variable (R) (SE) (N) (WN) (DEFT) (SE/R) R-2SE R+2SE –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.051 0.004 2,440 2,348 0.817 0.071 0.044 0.059 Literate 0.662 0.012 2,440 2,348 1.271 0.018 0.638 0.687 No education 0.255 0.012 2,440 2,348 1.304 0.045 0.232 0.278 Secondary education or higher 0.055 0.005 2,440 2,348 1.124 0.094 0.045 0.066 Never married/in union 0.341 0.013 2,440 2,348 1.375 0.039 0.315 0.367 Currently married/in union 0.515 0.015 2,440 2,348 1.462 0.029 0.485 0.544 Married before age 20 0.516 0.017 1,397 1,327 1.265 0.033 0.482 0.550 Currently pregnant 0.090 0.007 2,440 2,348 1.151 0.074 0.077 0.104 Children ever born 2.796 0.064 2,440 2,348 1.076 0.023 2.668 2.923 Children ever born to women 40-49 6.571 0.145 421 391 1.116 0.022 6.280 6.861 Children surviving 2.129 0.048 2,440 2,348 1.044 0.023 2.033 2.225 Knows any contraceptive method 0.985 0.006 1,251 1,208 1.669 0.006 0.973 0.996 Ever used any contraceptive method 0.375 0.014 1,251 1,208 0.989 0.036 0.348 0.402 Currently using any contraceptive method 0.189 0.011 1,251 1,208 1.017 0.060 0.167 0.212 Currently using pill 0.027 0.008 1,251 1,208 1.678 0.283 0.012 0.043 Currently using condom 0.004 0.002 1,251 1,208 1.037 0.462 0.000 0.008 Currently using female sterilization 0.002 0.001 1,251 1,208 0.987 0.632 0.000 0.004 Currently using periodic abstinence 0.065 0.008 1,251 1,208 1.095 0.118 0.049 0.080 Obtained method from public sector source 0.752 0.053 108 107 1.267 0.070 0.646 0.858 Want no more children 0.446 0.013 1,251 1,208 0.925 0.029 0.420 0.472 Want to delay birth at least two years 0.384 0.014 1,251 1,208 1.027 0.037 0.356 0.412 Ideal number of children 4.228 0.045 2,389 2,288 1.475 0.011 4.137 4.319 Mothers received tetanus injection for last birth 0.640 0.015 1,244 1,194 1.088 0.023 0.610 0.669 Mothers received medical assistance at delivery 0.385 0.021 2,012 1,932 1.593 0.055 0.343 0.428 Child had diarrhea in the 2 weeks prior to survey 0.151 0.010 1,721 1,644 1.094 0.066 0.131 0.171 Treated with oral rehydration salts (ORS) 0.056 0.016 263 248 1.092 0.280 0.025 0.088 Taken to a health provider 0.099 0.019 263 248 1.027 0.193 0.061 0.137 Vaccination card seen 0.766 0.035 364 350 1.576 0.046 0.696 0.837 Received BCG 0.914 0.032 364 350 2.137 0.035 0.850 0.979 Received DPT (3 doses) 0.826 0.038 364 350 1.902 0.046 0.749 0.902 Received polio (3 doses) 0.814 0.038 364 350 1.863 0.047 0.737 0.891 Received measles 0.739 0.039 364 350 1.694 0.053 0.661 0.818 Fully immunized 0.670 0.041 364 350 1.645 0.061 0.588 0.752 Weight-for-height (below -2SD) 0.038 0.007 871 831 0.993 0.172 0.025 0.051 Height-for-age (below -2SD) 0.424 0.017 871 831 0.990 0.041 0.389 0.459 Weight-for-age (below -2SD) 0.202 0.013 871 831 0.953 0.064 0.177 0.228 Anemia among children 0.596 0.025 776 741 1.331 0.042 0.546 0.647 Anemia among women 0.416 0.024 1,222 1,175 1.694 0.057 0.369 0.464 BMI <18.5 0.105 0.009 1,095 1,055 1.023 0.090 0.086 0.124 Total Fertility Rate (0-3 years) 6.491 0.176 na 6,459 1.118 0.027 6.139 6.844 Neonatal mortality (0-9 years) 48.246 4.339 3,819 3,643 1.071 0.090 39.567 56.925 Postneonatal mortality (0-9 years) 76.487 5.517 3,831 3,653 1.179 0.072 65.452 87.521 Infant mortality (0-9 years) 124.732 7.870 3,832 3,654 1.283 0.063 108.991 140.473 Child mortality (0-9 years) 123.291 7.614 3,884 3,705 1.103 0.062 108.064 138.518 Under-five mortality (0-9 years) 232.645 10.723 3,898 3,716 1.323 0.046 211.199 254.090 HIV prevalence 0.029 0.005 1,240 1,173 0.971 0.159 0.020 0.039 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban residence 0.047 0.006 1,066 1,017 0.874 0.121 0.035 0.058 Literate 0.784 0.015 1,066 1,017 1.219 0.020 0.753 0.815 No education 0.198 0.014 1,066 1,017 1.118 0.069 0.171 0.226 Secondary education or higher 0.075 0.008 1,066 1,017 0.958 0.103 0.060 0.091 Never married/in union 0.448 0.020 1,066 1,017 1.286 0.044 0.408 0.487 Currently married/in union 0.525 0.018 1,066 1,017 1.182 0.034 0.488 0.561 HIV prevalence (15-49) 0.021 0.006 975 923 1.398 0.308 0.008 0.033 HIV prevalence (15-59) 0.022 0.006 1,063 1,002 1.383 0.283 0.009 0.034 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– WOMEN AND MEN –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIV prevalence (15-49) 0.025 0.004 2,215 2,096 1.195 0.157 0.017 0.033 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– na = Not applicable Appendix C – Data Quality Tables | 279 DATA QUALITY TABLES APPENDIX C Table C.1 Household age distribution Single-year distribution of the de facto household population by sex (weighted), Rwanda 2005 Females Males Females Males Age Number Percent Number Percent Age Number Percent Number Percent 0 871 3.5 897 4.1 36 220 0.9 167 0.8 1 794 3.2 855 3.9 37 250 1.0 182 0.8 2 897 3.6 895 4.1 38 183 0.7 160 0.7 3 735 3.0 705 3.2 39 209 0.8 156 0.7 4 711 2.9 745 3.4 40 254 1.0 196 0.9 5 792 3.2 775 3.6 41 194 0.8 180 0.8 6 617 2.5 643 3.0 42 231 0.9 171 0.8 7 712 2.9 738 3.4 43 267 1.1 157 0.7 8 734 3.0 741 3.4 44 191 0.8 143 0.7 9 683 2.8 712 3.3 45 213 0.9 204 0.9 10 624 2.5 592 2.7 46 226 0.9 147 0.7 11 612 2.5 595 2.7 47 174 0.7 146 0.7 12 672 2.7 649 3.0 48 162 0.7 131 0.6 13 693 2.8 602 2.8 49 147 0.6 114 0.5 14 629 2.5 614 2.8 50 148 0.6 131 0.6 15 544 2.2 522 2.4 51 145 0.6 102 0.5 16 609 2.5 545 2.5 52 175 0.7 106 0.5 17 508 2.1 499 2.3 53 160 0.6 114 0.5 18 547 2.2 518 2.4 54 133 0.5 72 0.3 19 438 1.8 405 1.9 55 122 0.5 99 0.5 20 475 1.9 460 2.1 56 94 0.4 83 0.4 21 464 1.9 389 1.8 57 68 0.3 65 0.3 22 508 2.1 417 1.9 58 66 0.3 44 0.2 23 491 2.0 384 1.8 59 67 0.3 44 0.2 24 445 1.8 317 1.5 60 135 0.5 72 0.3 25 389 1.6 388 1.8 61 66 0.3 38 0.2 26 374 1.5 276 1.3 62 72 0.3 39 0.2 27 363 1.5 280 1.3 63 79 0.3 48 0.2 28 314 1.3 229 1.1 64 51 0.2 28 0.1 29 318 1.3 203 0.9 65 108 0.4 52 0.2 30 312 1.3 281 1.3 66 39 0.2 24 0.1 31 283 1.1 185 0.9 67 57 0.2 40 0.2 32 318 1.3 218 1.0 68 68 0.3 34 0.2 33 265 1.1 178 0.8 69 76 0.3 32 0.1 34 287 1.2 174 0.8 70+ 565 2.3 413 1.9 35 278 1.1 196 0.9 Don’t know/ missing 4 0.0 4 0.0 Total 24,727 100.0 21,762 100.0 280 | Appendix C – Data Quality Tables Table C.2.1 Age distribution of eligible and interviewed women De facto household population of women age 10-54, interviewed women age 15-49, and percentage of eligible women who were interviewed (weighted), by five-year age groups, Rwanda 2005 Interviewed women age 15-49 Age group Household population of women 10-54 Number Percent Percentage of eligible women interviewed 10-14 3,232 na na na 15-19 2,647 2,562 22.8 96.8 20-24 2,382 2,330 20.8 97.8 25-29 1,759 1,727 15.4 98.2 30-34 1,464 1,449 12.9 99.0 25-39 1,141 1,129 10.1 99.0 40-44 1,136 1,125 10.0 99.1 45-49 921 904 8.1 98.1 50-54 762 na na na 15-49 11,449 11,226 100.0 98.1 Note: The de facto population includes all residents and nonresidents who stayed in the household the night before the interview. Weights for both household population of women and interviewed women are household weights. Age is based on the household schedule. na = Not applicable Table C.2.2 Age distribution of eligible and interviewed men De facto household population of men age 10-59, interviewed men age 15- 54, and percentage of eligible men who were interviewed (weighted), by five-year age groups, Rwanda 2005 Interviewed men age 15-59 Age group Household population of men 10-64 Number Percent Percentage of eligible men interviewed 10-14 1,543 na na na 15-19 1,124 1,088 22.8 96.8 20-24 960 937 19.7 97.7 25-29 643 624 13.1 97.0 30-34 521 503 10.6 96.5 25-39 444 433 9.1 97.5 40-44 416 401 8.4 96.3 45-49 381 376 7.9 98.8 50-54 260 258 5.4 99.2 55-59 148 145 3.0 98.2 60-64 109 na na na 15-59 4,896 4,764 100.0 97.3 Note: The de facto population includes all residents and nonresidents who stayed in the household the night before the interview. Weights for both household population of men and interviewed men are household weights. Age is based on the household schedule. na = Not applicable Appendix C – Data Quality Tables | 281 Table C.3 Completeness of reporting Percentage of observations missing information for selected demographic and health questions (weighted), Rwanda 2005 Subject Reference group Percentage with missing information Number of cases Birth date Births in the 15 years preceding the survey Month only 2.4 22,458 Month and year 0.1 22,458 Age at death Deceased children born in the 15 years preceding the survey 0.2 4,114 Age/date at first union 1 Ever-married women age 15-49 0.1 7,058 Respondent's education All women age 15-49 0.1 11,321 Diarrhea in past 2 weeks Living children age 0-59 months 1.5 7,797 Anthropometry 2 Living children age 0-59 months (from the household questionnaire) Height 1.2 4,099 Weight 1.7 4,099 Height or weight 1.8 4,099 Anemia 3 Anemia – children Living children age 6-59 months (from the household questionnaire) 3.1 3,649 Anemia – women All women age 15-49 (from the household questionnaire) 3.3 5,818 1 Both year and age missing 2 Child not measured 3 Not tested Table C.4 Births by calendar years Number of births, percentage with complete birth date, sex ratio at birth, and calendar year ratio by calendar year, according to living, dead, and total children (weighted), Rwanda 2005 Number of births Percentage with a complete birth date 1 Sex ratio at birth 2 Calendar year ratio 3 Year Living Dead Total Living Dead Total Living Dead Total Living Dead Total 2005 743 36 779 100.0 100.0 100.0 109.2 93.0 108.4 na na na 2004 1,711 129 1,840 100.0 100.0 100.0 90.8 74.6 89.5 na na na 2003 1,595 184 1,779 100.0 99.5 99.9 89.8 128.8 93.2 92.3 101.3 93.1 2002 1,746 234 1,980 100.0 99.0 99.9 101.0 83.3 98.7 122.4 125.4 122.7 2001 1,259 190 1,448 100.0 97.8 99.7 104.1 97.2 103.2 79.1 75.4 78.6 2000 1,437 268 1,705 100.0 98.7 99.8 96.1 106.4 97.7 108.3 100.9 107.0 1999 1,395 342 1,737 98.5 92.9 97.4 99.6 89.7 97.6 110.9 113.1 111.4 1998 1,078 337 1,414 97.6 92.4 96.3 92.0 84.5 90.1 83.0 100.2 86.5 1997 1,201 330 1,531 97.4 92.4 96.3 99.6 71.4 92.8 102.0 92.7 99.9 1996 1,277 375 1,652 97.7 92.9 96.6 93.6 97.1 94.4 112.5 110.1 112.0 2001-2005 7,053 773 7,827 100.0 99.1 99.9 97.1 94.7 96.9 na na na 1996-2000 6,388 1,651 8,039 98.3 93.7 97.4 96.3 88.7 94.7 na na na 1991-1994 4,483 1,513 5,996 96.4 90.0 94.8 101.8 90.2 98.8 na na na 1986-1990 3,475 1,069 4,544 94.0 89.4 92.9 103.0 92.4 100.4 na na na < 1986 2,839 1,132 3,971 92.1 86.7 90.5 105.9 81.3 98.2 na na na Total 24,237 6,139 30,376 97.1 91.4 96.0 99.6 89.0 97.4 na na na 1 Both year and month of birth given 2 (Bm/Bf)*100, where Bm and Bf are the numbers of male and female births, respectively 3 [2Bx/(Bx-1+Bx+1)]*100, where Bx is the number of births in calendar year x na = Not applicable 282 | Appendix C – Data Quality Tables Table C.5 Reporting of age at death in days Distribution of reported deaths under one month of age by age at death in days and the percentage of neonatal deaths reported to occur at age 0-6 days, for five-year periods preceding the survey (weighted), Rwanda 2005 Number of years preceding the survey Age at death in days 0-4 5-9 10-14 15-19 Total 0-19 <1 135 165 118 77 495 1 25 36 28 9 98 2 21 25 25 11 82 3 21 31 34 9 96 4 10 11 10 3 35 5 9 11 6 8 34 6 6 9 8 7 29 7 33 53 33 27 146 8 4 9 12 5 29 9 3 6 2 1 12 10 1 6 4 3 13 11 1 1 0 2 4 12 1 4 1 0 5 13 1 1 1 0 3 14 20 12 26 7 65 15 10 10 6 2 29 16 0 0 0 1 1 18 1 3 1 0 4 19 0 0 0 1 1 20 5 6 1 1 13 21 2 5 3 4 13 23 1 0 1 0 2 24 0 1 2 0 3 25 0 1 1 1 2 26 2 0 0 0 2 27 1 0 0 1 2 28 1 2 2 1 5 29 0 2 0 0 2 30 6 3 10 6 25 Missing 0 0 1 0 1 Total 0-30 321 413 333 185 1 252 Percent early neonatal1 71.1 69.7 68.8 66.7 69.4 1 0-6 days/0-30 days Appendix C – Data Quality Tables | 283 Table C.6 Reporting of age at death in months Distribution of reported deaths under two years of age by age at death in months and the percentage of infant deaths reported to occur at age under one month, for five-year periods preceding the survey, Rwanda 2005 Number of years preceding the survey Age at death in months 0-4 5-9 10-14 15-19 Total 0-19 <1 month1 321 413 333 185 1,253 1 56 60 36 15 166 2 51 70 51 28 200 3 27 57 43 23 151 4 28 46 38 14 127 5 24 46 27 5 103 6 27 45 45 16 133 7 35 56 30 7 128 8 24 36 20 14 95 9 53 91 37 24 206 10 15 10 16 3 43 11 19 31 15 15 80 12 41 108 75 41 265 13 18 17 14 9 59 14 14 19 18 6 58 15 10 22 21 3 55 16 13 8 4 4 29 17 2 13 12 8 34 18 21 55 33 21 129 19 8 14 8 1 31 20 4 8 7 4 24 21 5 4 1 3 14 22 2 6 2 1 11 23 1 3 3 1 8 24 or more 0 0 1 0 1 1 year 2 2 2 3 9 Total 0-11 681 961 691 350 2,683 Percent neonatal2 47.2 42.9 48.3 52.9 46.7 1 Includes deaths under one month reported in days 2 Under one month/under one year Appendix D – Results According to Old Provinces | 285 RESULTS ACCORDING TO OLD PROVINCES APPENDIX D Table D.2.3 Educational attainment of household population Percent distribution of the de facto household population age six and over by highest level of education attended or completed, according to old province, Rwanda 2005 Old province No education Primary Primary complete1 Secondary incomplete Secondary complete2 Superior Total Number WOMEN Kigali 14.7 45.9 11.7 16.0 6.4 4.0 100.0 1,298 Kigali Ngali 28.0 57.9 9.3 3.8 0.8 0.0 100.0 1,885 Gitarama 23.0 61.3 9.7 4.0 1.4 0.1 100.0 2,164 Butare 29.4 57.9 7.4 3.6 1.2 0.1 100.0 1,870 Gikongoro 31.6 58.6 5.6 3.0 0.5 0.1 100.0 1,227 Cyangugu 29.8 58.4 6.6 2.7 0.9 0.0 100.0 1,512 Kibuye 30.1 59.8 7.4 1.9 0.5 0.0 100.0 1,175 Gisenyi 33.9 57.9 4.2 2.9 0.7 0.3 100.0 2,218 Ruhengeri 31.2 58.2 5.4 3.6 1.2 0.1 100.0 2,131 Byumba 31.1 62.2 1.7 4.1 0.7 0.0 100.0 1,586 Umutara 40.5 49.5 6.2 2.7 0.7 0.1 100.0 1,001 Kibungo 28.1 61.9 6.1 2.9 0.5 0.0 100.0 1,860 MEN Kigali 13.3 42.6 11.5 18.3 6.5 5.8 100.0 1,211 Kigali Ngali 20.1 62.3 11.8 4.2 0.7 0.3 100.0 1,584 Gitarama 17.5 65.3 10.1 5.1 1.2 0.6 100.0 1,856 Butare 23.7 65.1 5.6 4.3 0.8 0.3 100.0 1,568 Gikongoro 22.1 64.5 8.2 3.6 1.0 0.3 100.0 1,013 Cyangugu 20.7 63.9 9.7 3.1 1.3 0.4 100.0 1,288 Kibuye 23.3 63.5 8.3 3.5 1.1 0.1 100.0 959 Gisenyi 22.5 63.1 6.2 5.3 1.7 0.7 100.0 1,745 Ruhengeri 20.1 63.9 7.1 5.9 1.7 0.4 100.0 1,726 Byumba 23.3 67.9 2.5 5.0 1.0 0.2 100.0 1,429 Umutara 32.8 53.4 8.4 4.5 0.6 0.0 100.0 947 Kibungo 22.4 65.8 7.7 3.2 0.6 0.1 100.0 1,565 1 Completed 6 grades at the primary level 2 Completed 6 grades at the secondary level 286 | Appendix D – Results According to Old Provinces Table D.2.4 School attendance ratios Net attendance ratios (NAR) and gross attendance ratios (GAR) for the de jure household population by level of schooling and sex, according to old province, Rwanda 2005 Net attendance ratio 1 Gross attendance ratio2 Old province Male Female Total Male Female Total Gender parity index3 PRIMARY SCHOOL Kigali 81.2 81.7 81.4 132.7 130.1 131.5 0.98 Kigali Ngali 74.2 77.1 75.7 134.2 140.8 137.5 1.05 Gitarama 79.6 82.1 80.8 141.3 142.6 142.0 1.01 Butare 64.7 70.0 67.3 116.1 122.2 119.0 1.05 Gikongoro 74.8 72.7 73.8 128.9 131.1 130.0 1.02 Cyangugu 71.4 70.0 70.7 140.1 135.7 137.9 0.97 Kibuye 71.0 78.5 74.7 133.0 150.1 141.4 1.13 Gisenyi 76.5 78.6 77.6 134.6 141.1 138.0 1.05 Ruhengeri 78.0 80.9 79.4 138.3 139.1 138.7 1.01 Byumba 72.5 76.2 74.3 115.3 120.6 117.9 1.05 Umutara 67.2 73.7 70.4 134.0 131.9 133.0 0.98 Kibungo 71.9 74.7 73.3 145.6 158.1 151.8 1.09 SECONDARY SCHOOL Kigali 16.2 16.9 16.6 29.7 30.2 30.0 1.02 Kigali Ngali 3.6 3.2 3.4 5.1 5.2 5.2 1.02 Gitarama 2.2 3.5 2.8 4.0 5.9 4.9 1.46 Butare 4.3 4.8 4.6 6.9 6.1 6.5 0.88 Gikongoro 2.1 1.5 1.8 3.0 2.9 3.0 0.96 Cyangugu 2.5 1.3 1.9 5.1 3.2 4.2 0.62 Kibuye 2.7 1.6 2.1 4.8 2.7 3.7 0.56 Gisenyi 3.8 2.9 3.3 9.5 4.4 6.7 0.47 Ruhengeri 5.6 3.7 4.6 8.8 4.7 6.6 0.53 Byumba 5.3 2.8 4.1 9.8 4.0 6.9 0.40 Umutara 4.4 3.0 3.7 6.0 4.8 5.4 0.81 Kibungo 3.7 2.3 3.0 5.6 4.9 5.2 0.87 1 The NAR for primary school is the percentage of the primary-school-age (6-11 years) population that is attending primary school. The NAR for secondary school is the percentage of the secondary-school-age (12-18 years) population that is attending secondary school. By definition the NAR cannot exceed 100 percent. 2 The GAR for primary school is the total number of primary school students, expressed as a percentage of the official primary-school-age population. The GAR for secondary school is the total number of secondary school students, expressed as a percentage of the official secondary-school-age population. If there are significant numbers of over-age and underage students at a given level of schooling, the GAR can exceed 100 percent. 3 The Gender Parity Index for primary school is the ratio of the primary school GAR for females to the GAR for males. The Gender Parity Index for secondary school is the ratio of the secondary school GAR for females to the GAR for males. Table D.2.7 Wealth quintiles Percent distribution of households by wealth quintiles, according to old province, Rwanda 2005 Wealth quintile Old province Lowest Second Middle Fourth Highest Total Number Kigali 3.0 2.6 4.9 6.5 83.0 100.0 664 Kigali Ngali 21.3 20.1 28.5 16.6 13.6 100.0 1,023 Gitarama 15.4 20.5 20.3 23.3 20.6 100.0 1,100 Butare 22.6 20.3 19.1 21.7 16.4 100.0 988 Gikongoro 29.9 22.2 16.5 21.9 9.5 100.0 633 Cyangugu 22.0 8.7 20.7 31.1 17.5 100.0 726 Kibuye 28.3 20.4 21.7 19.1 10.6 100.0 598 Gisenyi 21.3 17.2 22.2 22.3 16.9 100.0 1,071 Ruhengeri 25.6 15.5 22.0 24.0 12.8 100.0 1,081 Byumba 30.4 24.9 22.9 13.2 8.6 100.0 867 Umutara 14.0 19.5 17.1 30.8 18.6 100.0 550 Kibungo 23.0 26.5 23.7 16.5 10.3 100.0 970 Appendix D – Results According to Old Provinces | 287 Table D.2.8 Birth registration of children under age five Percentage of de jure children under five years of age whose births are registered with the civil authorities, according to old province, Rwanda 2005 Percentage of children whose births are registered: Old province Had a birth certificate Did not have a birth certificate Total registered Number of children Kigali 72.3 6.2 78.5 457 Kigali Ngali 79.9 2.7 82.6 745 Gitarama 82.2 1.0 83.2 740 Butare 77.9 4.6 82.5 758 Gikongoro 86.5 4.2 90.7 516 Cyangugu 73.1 2.7 75.7 593 Kibuye 82.9 1.4 84.3 474 Gisenyi 69.4 8.2 77.6 980 Ruhengeri 82.1 6.2 88.3 958 Byumba 79.2 5.4 84.5 728 Umutara 67.6 2.9 70.5 446 Kibungo 80.3 5.5 85.8 727 Table D.3.3 Educational attainment Percent distribution of women and men by highest level of schooling attended or completed, according to old province, Rwanda 2005 Highest level of schooling attended or completed Old province No education Primary Secondary More than secondary Total Number WOMEN Kigali 8.8 55.5 30.6 5.1 100.0 900 Kigali Ngali 22.2 70.5 7.3 0.0 100.0 1,118 Gitarama 17.3 73.9 8.5 0.3 100.0 1,219 Butare 21.1 70.0 8.5 0.4 100.0 1,090 Gikongoro 24.7 68.9 6.3 0.1 100.0 650 Cyangugu 23.3 70.3 6.4 0.0 100.0 852 Kibuye 24.5 71.3 4.0 0.1 100.0 649 Gisenyi 32.8 60.2 6.4 0.6 100.0 1,179 Ruhengeri 28.1 63.5 8.2 0.2 100.0 1,180 Byumba 27.4 63.7 8.8 0.0 100.0 873 Umutara 30.1 63.7 5.9 0.2 100.0 554 Kibungo 22.1 72.1 5.8 0.0 100.0 1,057 MEN Kigali 8.9 52.2 29.9 8.9 100.0 426 Kigali Ngali 18.4 73.6 7.6 0.4 100.0 449 Gitarama 12.2 74.2 12.7 0.9 100.0 522 Butare 19.5 72.5 7.5 0.5 100.0 452 Gikongoro 19.1 72.3 8.3 0.3 100.0 275 Cyangugu 17.4 73.6 7.9 1.1 100.0 386 Kibuye 14.3 77.4 8.1 0.3 100.0 244 Gisenyi 21.4 66.2 11.6 0.8 100.0 488 Ruhengeri 16.4 72.0 10.5 1.1 100.0 478 Byumba 20.8 70.0 8.6 0.5 100.0 395 Umutara 21.1 70.9 7.7 0.3 100.0 271 Kibungo 20.5 72.3 6.7 0.4 100.0 433 288 | Appendix D – Results According to Old Provinces Table D.3.4 Literacy Percent distribution of women and men by level of schooling attended and by level of literacy, and percent literate, according to old province, Rwanda 2005 No schooling or primary school Old province Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all Total1 Number Percent literate2 WOMEN Kigali 35.7 46.2 6.6 11.0 100.0 900 88.6 Kigali Ngali 7.3 52.6 10.8 29.3 100.0 1,118 70.7 Gitarama 8.8 61.8 9.5 19.7 100.0 1,219 80.1 Butare 8.9 50.7 9.4 30.1 100.0 1,090 69.1 Gikongoro 6.4 53.5 5.6 34.5 100.0 650 65.5 Cyangugu 6.4 51.9 9.0 32.0 100.0 852 67.2 Kibuye 4.1 51.8 14.9 29.2 100.0 649 70.8 Gisenyi 6.9 41.4 14.1 37.3 100.0 1,179 62.4 Ruhengeri 8.4 39.9 17.9 33.5 100.0 1,180 66.2 Byumba 8.9 48.5 10.5 31.9 100.0 873 67.9 Umutara 6.2 50.6 9.5 33.6 100.0 554 66.2 Kibungo 5.8 48.5 12.9 31.9 100.0 1,057 67.2 MEN Kigali 38.8 43.6 6.7 10.3 100.0 426 89.2 Kigali Ngali 8.0 52.5 14.9 24.3 100.0 449 75.4 Gitarama 13.6 53.9 9.5 22.1 100.0 522 77.0 Butare 8.0 54.2 11.3 26.2 100.0 452 73.5 Gikongoro 8.6 52.6 10.6 28.0 100.0 275 71.7 Cyangugu 8.9 58.9 10.2 21.2 100.0 386 78.0 Kibuye 8.4 58.3 10.9 21.9 100.0 244 77.6 Gisenyi 12.4 51.4 10.9 25.0 100.0 488 74.7 Ruhengeri 11.6 57.2 12.6 18.3 100.0 478 81.4 Byumba 9.2 53.8 11.7 25.2 100.0 395 74.8 Umutara 8.0 56.8 12.6 22.4 100.0 271 77.3 Kibungo 7.1 64.3 6.7 21.7 100.0 433 78.1 1 Includes those with missing information 2 Refers to women and men who attended secondary school or higher and women who can read a whole sentence or part of a sentence. Table D.3.5 Exposure to mass media Percentage of women and men who usually read a newspaper at least once a week, watch television at least once a week, and listen to the radio at least once a week, by old province, Rwanda 2005 Old province Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media No media Number WOMEN Kigali 17.2 36.8 81.8 9.2 14.0 900 Kigali Ngali 4.9 1.8 49.6 0.4 49.2 1,118 Gitarama 14.6 3.1 56.5 1.6 40.9 1,219 Butare 5.5 3.0 62.7 1.2 36.4 1,090 Gikongoro 3.0 1.9 50.6 0.4 48.4 650 Cyangugu 16.7 3.4 50.5 1.0 43.3 852 Kibuye 4.4 1.2 38.3 0.5 60.0 649 Gisenyi 9.4 3.2 39.5 1.2 57.2 1,179 Ruhengeri 7.7 3.7 50.6 1.1 47.0 1,180 Byumba 7.8 1.2 45.7 0.1 52.0 873 Umutara 2.3 1.0 57.3 0.0 42.3 554 Kibungo 1.9 1.8 64.1 0.2 34.9 1,057 MEN Kigali 39.5 56.9 90.7 30.8 7.8 426 Kigali Ngali 4.9 4.6 79.1 1.2 18.9 449 Gitarama 7.3 10.1 77.4 3.2 22.6 522 Butare 5.2 6.2 84.4 1.5 15.3 452 Gikongoro 5.4 4.3 61.0 0.7 36.7 275 Cyangugu 9.7 9.2 75.9 2.8 22.1 386 Kibuye 11.2 3.2 77.5 0.3 19.9 244 Gisenyi 6.2 4.8 69.3 1.1 30.2 488 Ruhengeri 9.9 7.1 84.2 2.4 15.7 478 Byumba 3.6 4.3 79.7 1.5 18.9 395 Umutara 18.2 10.6 66.7 4.7 30.6 271 Kibungo 3.1 4.5 98.1 0.5 1.0 433 Appendix D – Results According to Old Provinces | 289 Table D.3.6 Employment status Percent distribution of women and men by employment status, according to old province, Rwanda 2005 Employed in the 12 months preceding the survey Old province Currently employed Not currently employed Not employed in the 12 months preceding the survey Total Number WOMEN Kigali 45.5 12.0 42.2 100.0 900 Kigali Ngali 61.3 1.9 36.8 100.0 1,118 Gitarama 54.9 26.5 18.3 100.0 1,219 Butare 79.8 0.9 19.3 100.0 1,090 Gikongoro 89.9 0.8 9.3 100.0 650 Cyangugu 41.9 7.2 51.0 100.0 852 Kibuye 76.1 7.7 16.2 100.0 649 Gisenyi 62.4 7.3 30.3 100.0 1,179 Ruhengeri 49.9 18.3 31.8 100.0 1,180 Byumba 71.4 0.4 28.3 100.0 873 Umutara 61.3 21.1 17.6 100.0 554 Kibungo 85.4 0.7 13.9 100.0 1,057 MEN Kigali 62.1 11.3 26.4 100.0 426 Kigali Ngali 25.1 5.0 69.6 100.0 449 Gitarama 35.4 10.5 53.9 100.0 522 Butare 63.7 3.6 32.4 100.0 452 Gikongoro 23.4 2.3 74.3 100.0 275 Cyangugu 37.8 6.0 56.2 100.0 386 Kibuye 57.2 7.3 35.2 100.0 244 Gisenyi 64.5 9.9 25.1 100.0 488 Ruhengeri 37.9 2.3 59.6 100.0 478 Byumba 84.8 0.4 13.9 100.0 395 Umutara 33.7 3.6 62.2 100.0 271 Kibungo 84.9 0.6 14.3 100.0 433 Table D.3.6 Occupation Percent distribution of women and men employed in the 12 months preceding the survey by occupation, according to old province, Rwanda 2005 Old province Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agri- culture Missing Total Number WOMEN Kigali 16.1 6.1 20.9 5.5 31.0 17.4 3.1 100.0 518 Kigali Ngali 1.6 0.5 1.4 0.7 2.5 92.9 0.3 100.0 707 Gitarama 2.4 0.6 3.1 0.9 3.1 89.4 0.5 100.0 992 Butare 1.7 0.5 2.0 0.8 3.1 91.8 0.1 100.0 879 Gikongoro 1.3 0.3 0.8 0.8 1.4 93.5 1.9 100.0 589 Cyangugu 2.6 0.4 13.5 0.7 3.8 77.0 1.9 100.0 418 Kibuye 1.0 0.1 1.4 0.1 0.8 96.2 0.4 100.0 544 Gisenyi 2.0 0.2 7.0 1.0 4.0 85.8 0.0 100.0 822 Ruhengeri 2.8 0.6 6.1 1.5 2.6 86.3 0.2 100.0 805 Byumba 1.7 0.1 3.7 0.7 4.3 89.0 0.5 100.0 626 Umutara 1.4 0.1 3.5 0.8 1.6 92.4 0.1 100.0 457 Kibungo 1.8 0.3 2.1 0.4 1.6 93.6 0.3 100.0 910 MEN Kigali 18.0 3.6 13.0 26.9 31.2 4.5 2.9 100.0 313 Kigali Ngali 3.7 0.9 6.4 10.5 24.7 52.8 1.0 100.0 135 Gitarama 6.9 0.4 11.4 13.4 14.1 52.0 1.8 100.0 240 Butare 2.5 0.0 1.1 10.4 6.8 78.1 1.1 100.0 304 Gikongoro 7.9 1.1 5.4 9.8 35.6 40.2 0.0 100.0 71 Cyangugu 9.5 1.3 6.8 10.1 15.2 56.6 0.5 100.0 169 Kibuye 5.1 0.0 10.3 12.1 3.8 67.3 1.4 100.0 158 Gisenyi 3.8 0.4 3.4 6.1 10.1 76.1 0.0 100.0 363 Ruhengeri 9.7 0.7 6.6 23.7 20.6 37.9 0.7 100.0 192 Byumba 1.7 0.3 2.1 4.7 6.6 84.7 0.0 100.0 336 Umutara 4.6 0.0 3.2 7.5 27.5 56.4 0.8 100.0 101 Kibungo 1.8 0.7 2.9 2.0 4.7 87.2 0.6 100.0 369 290 | Appendix D – Results According to Old Provinces Table D.4.2 Fertility by old province Total fertility rate for the three years preceding the survey, percentage of women 15-49 currently pregnant, and mean number of children ever born to women age 40-49 years, by old province, Rwanda 2005 Old province Total fertility rate1 Percentage currently pregnant1 Mean number of children ever born to women age 40-49 Kigali 4.0 6.0 5.5 Kigali Ngali 5.3 7.4 6.3 Gitarama 4.8 6.8 6.2 Butare 5.9 7.2 5.9 Gikongoro 6.8 9.8 6.3 Cyangugu 6.2 7.8 7.1 Kibuye 6.3 7.6 6.4 Gisenyi 7.1 7.9 7.7 Ruhengeri 6.9 8.4 7.1 Byumba 7.1 9.5 6.4 Umutara 6.8 9.3 6.8 Kibungo 6.2 9.2 6.6 1 Women age 15-49 years Table D.4.6 Birth Intervals Percent distribution of non-first births in the five years preceding the survey by number of months since preceding birth, by old province, Rwanda 2005 Months since preceding birth Old province 7-17 18-23 24-35 36-47 48+ Total Number of non-first births Median number of months since preceding birth Kigali 13.5 16.2 30.9 15.3 24.0 100.0 371 29.8 Kigali Ngali 7.3 14.1 40.2 21.4 16.9 100.0 643 32.5 Gitarama 5.6 14.1 39.4 22.4 18.5 100.0 618 33.6 Butare 8.6 15.5 38.5 19.1 18.2 100.0 637 31.9 Gikongoro 6.3 15.0 39.5 22.2 17.0 100.0 453 31.5 Cyangugu 10.8 17.5 40.3 16.9 14.5 100.0 502 29.2 Kibuye 4.9 14.0 43.9 23.1 14.1 100.0 394 31.8 Gisenyi 7.1 17.5 40.8 19.3 15.3 100.0 869 30.2 Ruhengeri 8.3 14.4 47.6 16.5 13.2 100.0 869 30.4 Byumba 7.0 15.2 43.2 17.9 16.7 100.0 654 31.5 Umutara 8.1 14.8 42.1 15.9 19.1 100.0 398 30.2 Kibungo 10.4 13.4 36.8 21.7 17.7 100.0 666 31.9 Note: First-order births are excluded. The interval for multiple births is the number of months since the preceding pregnancy that ended in a live birth. Table D.4.8 Median age at first birth Median age at first birth among women age 25-49 years, by current age and old province, Rwanda 2005 Current age Old province 25-29 30-34 35-39 40-44 45-49 Women age 25-49 Kigali 22.4 22.7 23.7 21.7 21.1 22.5 Kigali Ngali 22.0 21.9 22.1 22.1 20.8 21.8 Gitarama 23.9 23.4 22.9 22.0 21.8 22.8 Butare 23.6 23.0 24.1 23.6 23.0 23.5 Gikongoro 21.2 22.3 23.0 22.7 23.3 22.5 Cyangugu 22.6 22.6 21.8 21.5 21.1 22.0 Kibuye 21.8 22.5 21.8 21.5 22.6 22.0 Gisenyi 20.4 21.6 21.2 21.6 22.7 21.3 Ruhengeri 20.7 21.2 21.1 21.2 21.4 21.1 Byumba 21.0 21.7 22.6 22.1 23.0 21.9 Umutara 21.0 21.6 21.2 21.4 21.4 21.3 Kibungo 21.0 21.0 21.2 21.6 21.1 21.2 Appendix D – Results According to Old Provinces | 291 Table D.4.9 Teenage pregnancy and motherhood Percentage of women age 15-19 who are mothers or pregnant with their first child, by old province, Rwanda 2005 Percentage who are: Old province Mothers Pregnant with first child Percentage who have begun childbearing Number of women Kigali 5.3 0.7 6.0 226 Kigali Ngali 4.8 0.7 5.5 225 Gitarama 3.3 0.9 4.2 258 Butare 0.5 1.9 2.4 242 Gikongoro 4.7 0.5 5.2 147 Cyangugu 2.8 0.4 3.3 202 Kibuye 3.4 0.5 3.9 155 Gisenyi 2.8 1.4 4.2 300 Ruhengeri 1.9 0.5 2.3 291 Byumba 2.5 0.9 3.4 171 Umutara 4.3 2.3 6.6 106 Kibungo 4.6 0.0 4.6 262 Table D.5.4 Current use of contraception by background characteristics Percent distribution of currently married women by contraceptive method currently used, by old province, Rwanda 2005 Modern method Traditional method Old province Any method Any modern method Female sterili- zation Pill Inject- ables Male condom LAM Standard days method/ beads Other modern methods Any trad- itional method Periodic abstinence With- drawal Not currently using Total Number Kigali 42.3 28.0 1.1 5.0 7.9 6.3 3.0 1.7 2.9 14.3 9.6 4.7 57.7 100.0 309 Kigali Ngali 13.9 8.2 0.4 2.5 4.4 0.7 0.0 0.0 0.2 5.7 3.6 2.1 86.1 100.0 532 Gitarama 19.0 12.9 0.3 3.5 5.7 0.9 1.0 1.0 0.5 6.1 3.2 2.9 81.0 100.0 540 Butare 13.2 6.4 0.9 0.7 3.5 0.3 0.2 0.7 0.2 6.7 4.4 2.4 86.8 100.0 513 Gikongoro 10.9 4.3 0.0 0.3 1.9 1.0 0.6 0.1 0.3 6.6 2.4 4.2 89.1 100.0 358 Cyangugu 17.4 13.1 2.6 1.7 8.1 0.5 0.0 0.0 0.2 4.2 2.3 2.0 82.6 100.0 413 Kibuye 9.7 8.2 0.2 2.4 5.0 0.3 0.0 0.0 0.2 1.6 0.7 0.9 90.3 100.0 319 Gisenyi 15.2 10.1 0.1 2.2 2.9 0.4 2.9 1.2 0.3 5.1 3.2 1.9 84.8 100.0 622 Ruhengeri 13.9 8.7 0.4 2.4 4.6 0.4 0.1 0.5 0.3 5.3 2.0 3.3 86.1 100.0 597 Byumba 15.9 9.6 0.2 4.3 4.4 0.5 0.0 0.0 0.3 6.3 3.5 2.8 84.1 100.0 494 Umutara 15.2 8.2 0.4 0.8 5.0 0.2 1.4 0.2 0.1 7.0 3.9 3.1 84.8 100.0 298 Kibungo 27.2 10.0 0.2 3.1 4.5 0.8 0.7 0.0 0.6 17.2 11.4 5.7 72.8 100.0 515 Note: If more than one method is used, only the most effective method is considered in this tabulation. LAM = Lactational amenorrhea method 292 | Appendix D – Results According to Old Provinces Table D.5.11 Exposure to family planning messages Percentage of women and men who heard or saw a family planning message on the radio or television, or in a newspaper/magazine in the past few months, by old province, Rwanda 2005 Old province Radio Television Newspaper/magazine None of these three media sources Number WOMEN Kigali 65.5 19.8 16.9 33.4 900 Kigali Ngali 21.4 1.3 1.6 78.5 1,118 Gitarama 32.6 3.8 7.4 67.1 1,219 Butare 41.0 2.1 3.8 58.8 1,090 Gikongoro 40.8 1.1 2.1 59.1 650 Cyangugu 36.4 6.8 8.6 63.0 852 Kibuye 32.6 1.1 2.7 67.2 649 Gisenyi 28.1 1.8 2.3 71.7 1,179 Ruhengeri 52.3 1.8 3.4 47.5 1,180 Byumba 49.5 0.8 2.0 50.2 873 Umutara 44.2 1.4 1.8 55.8 554 Kibungo 50.3 1.1 1.9 49.6 1,057 MEN Kigali 77.8 31.5 33.0 20.4 426 Kigali Ngali 68.1 3.4 8.1 31.4 449 Gitarama 50.2 4.5 7.6 49.4 522 Butare 81.5 6.7 17.6 18.3 452 Gikongoro 52.4 2.7 9.1 47.5 275 Cyangugu 63.9 3.3 11.4 35.4 386 Kibuye 49.2 3.0 12.1 49.3 244 Gisenyi 41.8 2.2 6.7 58.0 488 Ruhengeri 64.6 5.4 12.3 35.4 478 Byumba 57.0 1.7 9.7 43.0 395 Umutara 71.1 10.3 23.2 27.7 271 Kibungo 50.4 1.4 1.3 49.6 433 Table D.6.2 Number of co-wives and wives Percent distribution of currently married women by number of co-wives and percent distribution of currently married men by number of wives, by old province, Rwanda 2005 Women Men Old province 0 1 2+ Total Number 1 2 3+ Total Number Kigali 90.1 0.9 9.0 100.0 309 95.8 3.7 0.0 100.0 155 Kigali Ngali 90.2 0.0 9.8 100.0 532 93.7 5.7 0.5 100.0 236 Gitarama 92.4 0.0 7.3 100.0 540 98.3 1.7 0.0 100.0 238 Butare 87.7 0.2 12.0 100.0 513 96.8 3.2 0.0 100.0 239 Gikongoro 85.8 0.2 14.0 100.0 358 93.2 4.1 2.7 100.0 154 Cyangugu 91.7 0.0 8.1 100.0 413 94.9 4.8 0.0 100.0 201 Kibuye 91.9 0.0 7.6 100.0 319 94.0 6.0 0.0 100.0 142 Gisenyi 80.9 0.0 18.9 100.0 622 91.9 8.1 0.0 100.0 288 Ruhengeri 86.9 0.2 12.9 100.0 597 96.9 3.1 0.0 100.0 277 Byumba 92.5 0.0 7.5 100.0 494 93.6 6.0 0.0 100.0 213 Umutara 87.1 0.0 12.9 100.0 298 92.1 6.5 1.4 100.0 139 Kibungo 84.1 0.0 15.0 100.0 515 93.0 6.5 0.6 100.0 217 Appendix D – Results According to Old Provinces | 293 Table D.6.4 Median age at first marriage Median age at first marriage among women age 25-49 and men age 30-59, by current age and old province, Rwanda 2005 Current age Old province 25-29 30-34 35-39 40-44 45-49 Women 25-49 Men 30-59 Kigali 22.9 22.5 23.2 20.3 20.1 21.9 27.7 Kigali Ngali 20.5 21.4 21.2 20.6 19.5 20.7 24.7 Gitarama 23.1 22.3 22.1 20.7 20.4 21.7 25.5 Butare 22.4 22.3 22.6 22.1 21.4 22.2 25.5 Gikongoro 20.1 21.1 21.9 21.0 21.7 21.1 24.7 Cyangugu 21.5 21.4 20.7 20.0 19.6 20.6 24.0 Kibuye 21.1 21.8 20.9 20.2 20.6 21.0 23.7 Gisenyi 18.9 20.3 20.1 20.5 21.0 20.0 22.8 Ruhengeri 19.6 20.0 19.3 19.5 19.6 19.6 23.5 Byumba 19.9 20.6 20.6 20.9 21.2 20.5 24.5 Umutara 19.9 20.3 20.4 19.9 20.1 20.1 24.5 Kibungo 19.8 19.8 20.0 19.9 19.2 19.8 24.0 Table D.6.6 Median age at first sexual intercourse Median age at first sexual intercourse among women age 25-49 and men age 25-59, by current age and old province, Rwanda 2005 Current age Old province 25-29 30-34 35-39 40-44 45-49 Women 25-49 Men 30-59 Kigali 20.4 21.6 22.7 20.5 20.2 20.9 20.6 Kigali Ngali 20.2 20.6 20.4 20.1 19.2 20.2 21.4 Gitarama 21.7 21.4 21.0 20.4 20.3 21.0 21.1 Butare 21.7 21.8 21.7 21.5 20.8 21.5 21.1 Gikongoro 19.7 20.7 21.3 20.9 21.5 20.7 22.0 Cyangugu 20.9 20.9 20.0 19.7 19.6 20.2 21.0 Kibuye 20.5 21.3 20.6 19.9 21.0 20.7 20.5 Gisenyi 18.7 19.9 19.8 20.2 20.3 19.5 20.0 Ruhengeri 19.3 20.2 19.6 19.0 19.6 19.5 20.9 Byumba 19.7 20.2 20.2 20.7 21.0 20.2 21.4 Umutara 19.7 20.0 20.2 19.9 20.2 19.9 20.4 Kibungo 19.3 19.3 19.4 19.6 18.8 19.3 19.3 294 | Appendix D – Results According to Old Provinces Table D.6.7 Recent sexual activity Percent distribution of women and men by timing of last sexual intercourse, by old province, Rwanda 2005 Timing of last sexual intercourse Old province Within the past 4 weeks Within 1 year1 One or more years Missing Never had sexual intercourse Total Number WOMEN Kigali 29.9 11.0 20.0 3.4 35.7 100.0 900 Kigali Ngali 42.9 7.5 16.2 2.6 30.9 100.0 1,118 Gitarama 39.0 6.9 20.4 2.7 31.0 100.0 1,219 Butare 40.2 8.9 16.2 3.2 31.5 100.0 1,090 Gikongoro 49.4 7.8 12.1 1.0 29.7 100.0 650 Cyangugu 43.4 6.4 13.8 2.4 34.0 100.0 852 Kibuye 45.3 6.6 11.3 2.7 34.1 100.0 649 Gisenyi 49.3 6.8 10.6 1.6 31.8 100.0 1,179 Ruhengeri 45.9 8.2 13.1 3.5 29.3 100.0 1,180 Byumba 53.4 7.8 11.7 1.7 25.3 100.0 873 Umutara 49.1 7.9 13.8 2.5 26.8 100.0 554 Kibungo 42.0 11.2 14.3 2.0 30.5 100.0 1,057 MEN Kigali 31.0 20.9 25.7 0.2 22.2 100.0 426 Kigali Ngali 48.0 8.2 7.9 0.0 35.8 100.0 449 Gitarama 42.3 10.5 21.7 0.0 25.5 100.0 522 Butare 49.4 8.3 13.5 0.0 28.9 100.0 452 Gikongoro 50.8 7.5 7.7 0.0 34.1 100.0 275 Cyangugu 49.1 7.9 12.1 0.2 30.6 100.0 386 Kibuye 53.9 9.7 13.1 0.0 23.3 100.0 244 Gisenyi 57.1 4.2 11.0 0.0 27.7 100.0 488 Ruhengeri 50.8 11.1 11.8 0.0 26.3 100.0 478 Byumba 49.1 11.2 7.5 0.3 31.9 100.0 395 Umutara 46.3 9.7 16.9 0.0 27.1 100.0 271 Kibungo 46.7 9.3 18.3 0.0 25.7 100.0 433 1 Excludes women and men who had sexual intercourse within the past 4 weeks Table D.6.9 Median duration of postpartum insusceptibility by background characteristics Median number of months of postpartum amenorrhea, postpartum abstinence, and postpartum insusceptibility following births in the three years preceding the survey, by old province, Rwanda 2005 Old province Postpartum amenorrhea Postpartum abstinence Postpartum insusceptibility Number of births Kigali 8.8 2.4 10.0 310 Kigali Ngali 13.5 0.6 16.5 517 Gitarama 16.0 0.7 17.5 477 Butare 15.2 0.6 16.6 485 Gikongoro 16.4 0.6 16.4 344 Cyangugu 14.9 0.6 15.1 396 Kibuye 16.2 0.6 17.1 310 Gisenyi 16.4 0.6 16.9 644 Ruhengeri 13.5 0.5 14.0 645 Byumba 13.4 0.6 15.0 517 Umutara 12.3 1.1 13.1 305 Kibungo 13.1 0.6 13.7 519 Note: Medians are based on current status. Appendix D – Results According to Old Provinces | 295 Table D.7.2 Desire to limit childbearing Percentage of currently married women who want no more children, by number of living children and the percentage of currently married women and currently married men who want no more children by old province, Rwanda 2005 Number of living children1 Old province 0 1 2 3 4 5 6+ Women Men Kigali * 15.1 34.2 57.8 78.2 (88.7) (92.4) 52.3 52.5 Kigali Ngali (0.0) 4.7 22.8 41.0 54.5 (60.7) 88.9 40.9 37.8 Gitarama * 4.0 19.6 46.1 60.7 73.4 92.3 47.2 49.8 Butare * 1.1 18.1 27.1 60.6 (74.3) 91.6 37.4 34.9 Gikongoro (0.0) 1.6 8.8 25.4 40.3 66.1 79.6 35.7 41.1 Cyangugu * 7.1 14.1 39.0 68.3 61.9 73.3 46.7 52.2 Kibuye * 9.7 28.7 26.2 55.8 (58.5) 74.6 41.8 44.7 Gisenyi * 1.5 20.1 23.0 46.2 46.9 64.7 34.7 39.8 Ruhengeri * 7.7 26.8 25.3 47.8 (44.8) 79.9 41.6 44.3 Byumba * 1.4 23.2 41.4 58.9 72.0 90.8 45.3 48.5 Umutara * 6.8 26.8 40.7 56.4 61.3 82.0 44.6 44.6 Kibungo * 15.9 18.3 39.2 68.8 83.5 92.7 49.1 39.3 Note: Women and men who have been sterilized are considered to want no more children. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Figures in parentheses are based on 25-49 unweighted cases. 1 Includes current pregnancy Table D.7.3 Need for family planning among currently married women Percentage of currently married women with unmet need for family planning, and with met need for family planning, and the total demand for family planning, by old province, Rwanda 2005 Unmet need for family planning Met need for family planning (currently using) Total demand for family planning Old province For spacing For limiting Total For spacing For limiting Total For spacing For limiting Total Percentage of demand satisfied Number of women Kigali 17.2 12.0 29.2 16.5 25.8 42.3 33.8 37.7 71.5 59.1 309 Kigali Ngali 21.0 14.8 35.8 5.9 8.0 13.9 26.8 22.8 49.6 27.9 532 Gitarama 20.8 15.5 36.3 7.7 11.3 19.0 28.4 26.9 55.3 34.4 540 Butare 28.1 10.2 38.2 6.2 7.0 13.2 34.2 17.2 51.4 25.6 513 Gikongoro 25.2 13.2 38.5 6.6 4.3 10.9 31.8 17.5 49.3 22.1 358 Cyangugu 23.2 15.0 38.1 7.0 10.4 17.4 30.2 25.4 55.5 31.3 413 Kibuye 23.5 15.6 39.2 4.0 5.7 9.7 27.6 21.3 48.9 19.9 319 Gisenyi 26.8 11.5 38.3 8.1 7.0 15.2 34.9 18.6 53.5 28.4 622 Ruhengeri 32.1 12.2 44.2 6.5 7.5 13.9 38.5 19.6 58.2 24.0 597 Byumba 22.8 14.0 36.8 6.1 9.8 15.9 28.9 23.8 52.7 30.1 494 Umutara 21.8 18.5 40.3 5.1 10.1 15.2 26.9 28.7 55.5 27.4 298 Kibungo 25.5 11.5 37.0 10.8 16.4 27.2 36.3 27.9 64.2 42.4 515 Table D.7.5 Mean ideal number of children Mean ideal number of children for all women and men, by age and background characteristics, Rwanda 2005 Age Old province 15-19 20-24 25-29 30-34 35-39 40-44 45-49 All women All men Kigali 3.6 3.5 3.4 3.7 3.7 4.3 (4.0) 3.6 3.5 Kigali Ngali 4.0 3.9 4.0 4.1 4.4 4.4 4.4 4.1 2.8 Gitarama 4.0 3.7 3.6 4.1 4.0 4.0 4.5 4.0 3.9 Butare 4.3 4.5 4.1 4.3 4.4 4.3 4.3 4.3 4.4 Gikongoro 4.8 4.7 4.7 4.6 4.7 5.0 5.2 4.8 4.3 Cyangugu 4.5 4.5 4.3 4.1 4.4 4.3 3.8 4.3 4.2 Kibuye 4.4 4.5 4.3 4.6 4.5 4.5 5.1 4.5 4.2 Gisenyi 4.7 4.5 4.6 5.1 5.1 5.3 5.4 4.8 4.6 Ruhengeri 4.6 4.3 4.4 4.5 4.7 5.1 4.7 4.6 4.2 Byumba 4.1 3.9 4.0 4.1 4.4 4.4 4.9 4.1 3.9 Umutara 4.1 4.0 4.2 4.1 4.2 4.3 4.0 4.1 3.8 Kibungo 4.2 4.3 4.2 4.5 4.2 4.1 4.1 4.3 3.9 Note: Figures in parentheses are based on 25-49 unweighted cases. 296 | Appendix D – Results According to Old Provinces Table D.7.7 Wanted fertility rates Total wanted fertility rates and total fertility rates for the three years preceding the survey, by old province, Rwanda 2005 Old province Total wanted fertility rate Total fertility rate Kigali 3.2 4.0 Kigali Ngali 4.2 5.3 Gitarama 3.6 4.8 Butare 4.7 5.9 Gikongoro 5.6 6.8 Cyangugu 4.2 6.2 Kibuye 4.8 6.3 Gisenyi 5.4 7.1 Ruhengeri 5.4 6.9 Byumba 5.1 7.1 Umutara 4.5 6.8 Kibungo 4.7 6.2 Note: Rates are calculated based on births to women age 15-49 in the period 1-36 months preceding the survey. The total fertility rates are the same as those presented in table D.4.2. Table D.8.1 Antenatal care Percent distribution of women who had a live birth in the five years preceding the survey by antenatal care (ANC) provider during pregnancy for the most recent birth, according to old province, Rwanda 2005 Old province Doctor Nurse/midwife/ auxiliary nurse/ midwife/trained traditional birth attendant Trained personnel Untrained traditional birth attendant/other No one Total1 Number Kigali 23.4 68.8 92.2 0.3 7.5 100.0 329 Kigali Ngali 2.4 89.2 91.5 0.0 8.5 100.0 507 Gitarama 1.4 95.3 96.8 0.0 3.2 100.0 528 Butare 14.5 79.7 94.2 0.0 5.8 100.0 490 Gikongoro 3.4 89.9 93.4 0.0 6.6 100.0 339 Cyangugu 5.5 87.1 92.6 0.0 6.9 100.0 392 Kibuye 30.2 63.4 93.6 0.0 6.4 100.0 309 Gisenyi 2.3 90.6 93.0 0.1 6.7 100.0 616 Ruhengeri 1.3 94.4 95.7 0.0 2.9 100.0 602 Byumba 1.6 94.3 95.9 0.0 4.1 100.0 505 Umutara 2.0 93.8 95.9 0.0 4.1 100.0 303 Kibungo 8.0 89.1 97.1 0.0 2.9 100.0 504 1 Includes those with missing information Note: If more than one source of ANC was mentioned, only the provider with the highest qualifications is considered in this tabulation. Appendix D – Results According to Old Provinces | 297 Table D.8.3 Components of antenatal care Percentage of women with a live birth in the five years preceding the survey who received antenatal care for the most recent birth, by content of antenatal care, and percentage of women with a live birth in the five years preceding the survey who received iron tablets or syrup or anti-malarial drugs for the most recent birth, according to old province, Rwanda 2005 Among women who received antenatal care Old province Informed of signs of pregnancy complications Weight measured Height measured Blood pressure measured Urine sample taken Blood sample taken Number of women who received antenatal care Received iron tablets or syrup Received anti-malarial drugs Number of women Kigali 10.4 96.3 61.9 92.9 33.2 73.9 304 38.2 9.9 329 Kigali Ngali 2.8 97.0 51.8 64.3 4.9 16.0 464 13.6 5.1 507 Gitarama 7.1 96.1 61.4 84.8 10.7 24.1 511 32.5 12.5 528 Butare 6.3 95.6 66.6 90.0 7.1 21.6 462 44.9 9.0 490 Gikongoro 8.7 97.1 64.7 83.1 3.1 27.8 316 29.1 0.8 339 Cyangugu 9.7 91.6 65.1 71.4 14.3 35.8 363 40.7 6.1 392 Kibuye 4.5 94.9 55.7 74.7 2.2 28.8 289 44.6 2.9 309 Gisenyi 9.2 90.3 36.5 61.1 4.7 22.4 573 23.0 3.5 616 Ruhengeri 3.7 91.5 39.5 61.1 3.8 17.5 577 26.3 3.3 602 Byumba 4.2 95.2 51.7 55.8 2.4 14.8 484 13.8 0.9 505 Umutara 5.5 90.9 73.5 79.3 9.5 17.3 291 29.3 13.9 303 Kibungo 4.0 90.4 59.0 57.3 4.5 17.2 489 17.2 4.3 504 Table D.8.4 Tetanus toxoid injections Percent distribution of women who had a live birth in the five years preceding the survey by number of tetanus toxoid injections received during pregnancy for the most recent birth, according to background characteristics, Rwanda 2005 Old province None One injection Two or more injections Don’t know/missing Total Number Kigali 20.4 41.6 35.3 2.8 100.0 329 Kigali Ngali 37.8 45.6 15.3 1.3 100.0 507 Gitarama 38.1 44.2 17.2 0.5 100.0 528 Butare 32.5 50.1 17.0 0.4 100.0 490 Gikongoro 34.4 44.6 20.5 0.6 100.0 339 Cyangugu 38.3 37.2 22.6 2.0 100.0 392 Kibuye 31.5 48.5 19.3 0.7 100.0 309 Gisenyi 42.6 35.3 20.0 2.0 100.0 616 Ruhengeri 37.2 38.6 23.0 1.2 100.0 602 Byumba 39.0 36.6 23.9 0.5 100.0 505 Umutara 42.3 32.9 24.8 0.0 100.0 303 Kibungo 26.9 40.3 32.6 0.2 100.0 504 Table D.8.5 Place of delivery Percent distribution of live births in the five years preceding the survey by place of delivery, according to old province, Rwanda 2005 Health facility Old province Public sector Private sector Home Total1 Number of births Kigali 56.1 9,4 33,4 100,0 500 Kigali Ngali 25.4 1,1 72,8 100,0 809 Gitarama 37.9 0,9 60,0 100,0 776 Butare 27.2 0,8 70,6 100,0 802 Gikongoro 13.0 0,3 86,0 100,0 544 Cyangugu 36.0 0,1 61,6 100,0 632 Kibuye 25.7 0,0 72,9 100,0 489 Gisenyi 17.4 1,0 79,3 100,0 1,029 Ruhengeri 27.6 1,4 69,8 100,0 1,032 Byumba 22.9 1,4 74,7 100,0 798 Umutara 28.3 0,5 70,9 100,0 488 Kibungo 17.2 0,2 80,8 100,0 816 1 Includes those with missing information 298 | Appendix D – Results According to Old Provinces Table D.8.6 Assistance during delivery Percent distribution of live births in the five years preceding the survey by person providing assistance during delivery, according to old province, Rwanda 2005 Old province Doctor Nurse/midwife/ auxiliary midwife/trained traditional birth attendant Trained personnel Untrained traditional birth attendant Relative/other No one Total1 Number of births Kigali 19.5 50.9 70.3 21.0 1.6 6.7 100.0 500 Kigali Ngali 0.9 30.3 31.2 46.0 0.0 22.6 100.0 809 Gitarama 7.6 41.0 48.7 36.2 0.2 15.0 100.0 776 Butare 7.6 38.3 45.9 36.9 0.2 16.6 100.0 802 Gikongoro 2.7 15.9 18.6 61.5 0.0 19.9 100.0 544 Cyangugu 5.3 47.3 52.6 39.4 0.1 7.6 100.0 632 Kibuye 11.4 17.3 28.8 50.8 2.1 18.2 100.0 489 Gisenyi 2.8 25.8 28.6 43.3 0.2 27.1 100.0 1,029 Ruhengeri 2.3 30.0 32.3 50.1 1.0 16.4 100.0 1,032 Byumba 1.9 28.7 30.6 59.4 0.4 9.4 100.0 798 Umutara 2.8 37.6 40.4 37.7 0.1 21.8 100.0 488 Kibungo 3.3 42.7 46.0 32.7 0.0 21.0 100.0 816 1 Includes those with missing information Note: If the respondent mentioned more than one person attending during delivery, only the most qualified person is considered in this tabulation. Table D.8.7 Delivery characteristics Percentage of live births in the five years preceding the survey delivered by caesarean section, and percent distribution by birth weight and by mother’s estimate of baby’s size at birth, according to old province, Rwanda 2005 Birth weight Size of child at birth Old province Delivery by C-section Not weighed Less than 2.5 kg 2.5 kg or more Total1 Very small Smaller than average Average or larger Total1 Number of births Kigali 11.5 23.2 3.6 70.6 100.0 4.1 8.6 86.6 100.0 500 Kigali Ngali 0.8 72.4 1.8 25.0 100.0 3.0 10.0 86.5 100.0 809 Gitarama 5.3 58.8 2.7 37.0 100.0 4.9 11.7 83.4 100.0 776 Butare 2.6 71.0 1.8 25.9 100.0 4.7 7.4 87.8 100.0 802 Gikongoro 1.7 87.6 1.1 10.4 100.0 4.3 8.7 87.0 100.0 544 Cyangugu 5.3 64.4 1.5 30.5 100.0 4.4 9.9 85.6 100.0 632 Kibuye 2.0 74.7 0.9 23.6 100.0 3.4 11.7 84.9 100.0 489 Gisenyi 1.8 78.8 0.7 18.5 100.0 1.0 9.6 88.2 100.0 1,029 Ruhengeri 1.7 67.2 1.0 29.5 100.0 3.1 10.6 85.8 100.0 1,032 Byumba 1.7 75.4 1.2 21.4 100.0 2.2 4.4 92.8 100.0 798 Umutara 2.1 67.9 1.4 29.7 100.0 2.4 7.6 89.5 100.0 488 Kibungo 2.2 63.7 3.0 31.8 100.0 3.9 13.7 81.9 100.0 816 1 Includes those with missing information Appendix D – Results According to Old Provinces | 299 Table D.8.8 Postnatal care Percentage of live births in the five years preceding the survey for which the mother delivered in a health facility and percent distribution of women whose last live birth in the five years preceding the survey occurred outside a health facility by timing of postnatal care, according to old province, Rwanda 2005 Timing of first postnatal checkup for births occurring outside a health facility Old province Delivered in a health facility Number of births 0-2 days after delivery 3-6 days after delivery 7-41 days after delivery Don’t know/ missing Did not receive postnatal checkup1 Total Number of births occurring outside a health facility Kigali 65.7 329 8.3 1.4 1.5 1.0 87.8 100.0 113 Kigali Ngali 26.8 507 1.6 0.0 0.0 0.3 98.0 100.0 371 Gitarama 38.3 528 7.8 0.4 1.3 0.4 90.0 100.0 325 Butare 29.6 490 2.1 0.0 0.0 0.2 97.6 100.0 345 Gikongoro 14.1 339 1.7 0.0 0.2 0.0 98.1 100.0 291 Cyangugu 37.8 392 3.7 0.0 0.7 0.0 95.6 100.0 244 Kibuye 25.2 309 0.8 0.0 0.0 0.0 99.2 100.0 231 Gisenyi 20.0 616 6.9 0.5 0.8 0.9 91.0 100.0 493 Ruhengeri 29.7 602 2.0 0.9 0.0 1.0 96.1 100.0 424 Byumba 26.7 505 5.2 0.0 1.4 0.0 93.4 100.0 370 Umutara 29.7 303 4.1 0.3 0.1 0.0 95.4 100.0 213 Kibungo 17.5 504 1.8 0.0 0.4 0.0 97.8 100.0 416 1 Includes women who received the first postnatal checkup after 41 days Table D.8.10 Vaccinations Percentage of children age 12-23 months who received specific vaccines at any time before the survey (according to a vaccination card or the mother’s report), and percentage with a vaccination card, by old province, Rwanda 2005 DPT Polio1 Old province BCG 1 2 3 0 1 2 3 Measles All2 No vacci- nations Per- centage with a vacci- nation card seen Number of children Kigali 96.8 95.2 86.8 80.0 85.9 97.9 92.0 74.8 85.5 58.6 2.1 64.7 82 Kigali Ngali 91.0 90.3 89.5 83.4 78.5 90.3 88.0 81.1 80.7 73.9 9.0 76.1 145 Gitarama 98.9 97.8 97.8 90.6 67.9 97.9 94.7 88.8 94.0 83.9 1.1 69.1 131 Butare 98.3 98.6 96.2 92.2 81.8 97.4 93.7 87.1 94.9 83.7 0.9 78.7 148 Gikongoro 97.6 97.6 96.5 95.0 68.4 97.9 96.2 90.9 93.0 85.6 1.3 81.8 114 Cyangugu 97.7 97.7 95.7 79.8 84.9 96.5 88.1 73.4 87.5 70.0 2.3 72.5 122 Kibuye 99.0 99.5 97.2 96.3 82.5 99.5 97.1 95.5 92.4 89.6 0.5 94.0 94 Gisenyi 94.5 98.1 87.6 79.5 59.6 96.0 90.8 79.9 74.4 63.4 1.9 71.9 198 Ruhengeri 98.2 98.3 96.0 93.1 71.1 98.0 98.0 88.8 93.0 83.9 0.4 69.9 196 Byumba 95.9 95.6 91.9 81.5 80.0 95.9 93.2 80.2 85.7 73.7 4.1 76.9 161 Umutara 93.6 93.2 91.8 89.7 56.7 93.2 91.9 87.8 74.6 71.4 5.1 80.7 95 Kibungo 96.9 97.8 94.7 85.6 77.1 97.8 92.3 84.9 73.9 64.2 2.2 81.3 139 1 Polio 0 is the polio vaccination given at birth. 2 BCG, measles and three doses each of DPT and polio vaccine (excluding polio vaccine given at birth) 300 | Appendix D – Results According to Old Provinces Table D.8.11 Prevalence and treatment of symptoms of ARI and fever Percentage of children under five years who had a cough accompanied by short, rapid breathing (symptoms of ARI), and percentage of children who had fever in the two weeks preceding the survey, and percentage of children with symptoms of ARI and/or fever for whom treatment was sought from a health facility or provider, by old province, Rwanda 2005 Old province Percentage of children with symptoms of ARI Percentage of children with fever Number of children Among children with symptoms of ARI and/or fever, percentage for whom treatment was sought from a health facility/provider1 Number of children Kigali 20.4 26.6 467 46.7 158 Kigali Ngali 6.1 13.9 722 26.8 119 Gitarama 13.1 22.1 709 39.1 180 Butare 21.3 37.9 707 28.5 308 Gikongoro 19.2 28.0 493 15.3 164 Cyangugu 21.0 28.0 562 20.5 207 Kibuye 8.8 12.5 448 17.6 65 Gisenyi 14.9 26.2 944 20.0 282 Ruhengeri 13.4 27.7 938 29.2 293 Byumba 24.9 22.2 701 31.4 228 Umutara 16.6 28.6 430 23.8 148 Kibungo 27.3 38.5 675 22.5 299 1 Excludes pharmacy, shop, and traditional practitioner Table D.8.12 Prevalence of diarrhea Percentage of children under five years with diarrhea in the two weeks preceding the survey, by old province, Rwanda 2005 Old province Diarrhea in the two weeks preceding the survey Number of children Kigali 12.3 467 Kigali Ngali 7.5 722 Gitarama 7.2 709 Butare 19.7 707 Gikongoro 17.6 493 Cyangugu 16.9 562 Kibuye 7.5 448 Gisenyi 13.8 944 Ruhengeri 16.6 938 Byumba 16.2 701 Umutara 14.5 430 Kibungo 18.5 675 Table D.8.13 Knowledge of ORS packets Percentage of mothers with births in the five years preceding the survey who know about ORS packets for treatment of diarrhea, by old province, Rwanda 2005 Old province Percentage of mothers who know about ORS packets Number of mothers Kigali 88.9 329 Kigali Ngali 86.3 507 Gitarama 94.3 528 Butare 92.6 490 Gikongoro 70.8 339 Cyangugu 74.2 392 Kibuye 90.9 309 Gisenyi 84.3 616 Ruhengeri 86.6 602 Byumba 93.4 505 Umutara 89.8 303 Kibungo 85.9 504 ORS = Oral rehydration salts Appendix D – Results According to Old Provinces | 301 Table D.8.14 Diarrhea treatment Percentage of children under five years who had diarrhea in the two weeks preceding the survey taken for treatment to a health provider, percentage who received oral rehydration therapy (ORT), and percentage given other treatments, according to old province, Rwanda 2005 Oral rehydration therapy (ORT) Other treatments Old province Percentage taken to a health provider1 ORS packets RHF Either ORS or RHF Increased fluids ORS, RHF, or increased fluids Pill/syrup Injection Home remedy/ other No treatment Number of children Kigali 20.3 23.4 10.3 31.8 27.8 44.1 23.5 1.5 26.2 28.9 57 Kigali Ngali (17.0) (11.6) (5.9) (17.5) (16.9) (28.5) (32.9) (0.0) (30.4) (24.9) 54 Gitarama (19.0) (2.4) (18.4) (18.4) (24.2) (40.2) (16.3) (0.0) (21.8) (40.5) 51 Butare 8.7 6.7 15.4 22.1 19.7 33.7 13.8 1.3 30.1 32.7 139 Gikongoro 9.7 9.0 2.2 10.3 32.9 36.9 14.2 0.4 31.3 35.6 87 Cyangugu 8.7 16.1 4.7 19.2 15.3 31.7 15.9 0.0 14.5 47.1 95 Kibuye (14.0) (13.9) (23.1) (25.8) (20.6) (34.8) (7.1) (0.0) (33.1) (40.3) 34 Gisenyi 16.7 18.3 7.7 25.1 13.1 33.8 23.0 1.9 34.1 28.3 130 Ruhengeri 18.8 16.5 4.6 19.8 12.1 29.1 20.7 0.0 26.8 36.9 156 Byumba 22.3 12.1 5.9 15.4 15.9 24.6 21.6 1.8 36.8 28.5 114 Umutara 6.7 3.6 2.8 5.8 26.2 30.2 15.2 0.0 46.3 32.9 62 Kibungo 9.1 3.3 11.5 13.5 16.1 27.5 14.4 1.3 50.2 28.0 125 Note: ORT includes solution prepared from oral rehydration salt (ORS) packets, recommended home fluids (RHF), or increased fluids. The figures in parentheses are based on 25-49 unweighted cases. 1 Excludes pharmacy, shop and traditional practitioner Table D.8.16 Problems in accessing health care Percentage of women who reported they have big problems in accessing health care for themselves when they are sick, by type of problem and old province, Rwanda 2005 Problems in accessing health care Old province Knowing where to go for treatment Getting permission to go for treatment Getting money for treatment Distance to health facility Having to take transport Not wanting to go alone Concern there may not be a female provider Any of the specified problems Number of women Kigali 5.7 3.9 55.6 29.0 30.5 17.5 10.0 67.5 900 Kigali Ngali 2.5 1.6 76.0 41.4 36.2 12.9 3.6 79.9 1,118 Gitarama 6.0 4.5 65.5 50.4 50.7 19.0 10.0 81.9 1,219 Butare 1.6 1.4 69.4 41.8 42.6 19.3 5.0 83.4 1,090 Gikongoro 1.9 0.6 82.2 37.2 33.5 10.1 6.4 87.2 650 Cyangugu 5.7 4.7 77.8 40.1 37.2 23.9 9.3 87.7 852 Kibuye 3.1 7.6 88.7 54.2 52.6 19.4 3.7 93.4 649 Gisenyi 5.6 2.9 80.7 40.5 42.4 14.8 19.1 88.7 1,179 Ruhengeri 6.4 1.8 62.0 26.4 30.6 17.2 9.5 71.2 1,180 Byumba 2.0 1.2 60.1 19.9 18.1 5.4 2.1 65.6 873 Umutara 2.5 1.3 75.6 51.9 36.4 10.0 4.4 84.4 554 Kibungo 9.8 5.2 67.6 51.6 52.9 29.1 20.5 84.4 1,057 302 | Appendix D – Results According to Old Provinces Table D.9.1 Household possession of mosquito nets Percentage of households with at least one and more than one mosquito net (treated or untreated), ever treated mosquito net and insecticide treated net1 (ITN), and the average number of nets per household, by old province, Rwanda 2005 Any type mosquito net Ever treated mosquito net 1 Insecticide treated mosquito nets (ITNs)2 Old province Percentage with at least one Percentage with more than one Average number per household Percentage with at least one Percentage with more than one Average number per household Percentage with at least one Percentage with more than one Average number per household Number of households Kigali 49.5 26.3 0.9 48.9 25.7 0.9 39.9 19.8 0.7 664 Kigali Ngali 11.6 4.1 0.2 11.5 4.1 0.2 8.2 3.3 0.1 1,023 Gitarama 23.4 6.0 0.3 23.3 6.0 0.3 19.2 4.2 0.3 1,100 Butare 23.7 5.0 0.3 23.5 5.0 0.3 18.8 3.5 0.2 988 Gikongoro 7.5 1.4 0.1 7.5 1.4 0.1 6.2 1.0 0.1 633 Cyangugu 26.7 8.1 0.4 26.6 8.1 0.4 22.9 6.2 0.3 726 Kibuye 18.4 5.3 0.2 18.3 5.3 0.2 14.7 3.9 0.2 598 Gisenyi 10.4 2.9 0.2 10.3 2.9 0.1 8.8 2.6 0.1 1,071 Ruhengeri 7.3 2.6 0.1 7.0 2.3 0.1 5.5 1.8 0.1 1,081 Byumba 16.9 3.1 0.2 16.9 3.1 0.2 14.1 2.6 0.2 867 Umutara 19.1 4.6 0.2 19.1 4.6 0.2 16.3 3.8 0.2 550 Kibungo 14.6 3.9 0.2 14.2 3.4 0.2 10.7 2.4 0.1 970 1 An ever-treated net is (1) a pretreated net or (2) a non-pretreated which has subsequently been soaked with insecticide at any time. 2 An insecticide treated net (ITN) is (1) a factory treated net that does not require any further treatment or (2) a pretreated net obtained within the past 12 months or (3) a net that has been soaked with insecticide within the past 12 months. Table D.9.2 Use of mosquito nets by children Percentage of children under five years of age who slept under a mosquito net (treated or untreated), an ever-treated mosquito net1, and an insecticide treated net2 (ITN) the night before the survey, by old province, Rwanda 2005 Old province Percentage who slept under any net the preceding night Percentage who slept under an ever-treated net1 the preceding night Percentage who slept under an ITN2 the preceding night Number of children Kigali 37.7 36.2 29.2 419 Kigali Ngali 9.3 9.3 7.0 692 Gitarama 24.3 24.3 20.1 682 Butare 27.0 27.0 20.9 692 Gikongoro 4.6 4.6 3.9 490 Cyangugu 24.7 24.7 20.9 557 Kibuye 17.3 17.3 14.7 430 Gisenyi 8.0 8.0 7.5 911 Ruhengeri 5.1 5.1 4.5 894 Byumba 17.2 17.2 15.0 696 Umutara 17.3 17.3 15.4 414 Kibungo 10.7 10.7 8.1 655 1 An ever-treated net is (1) a pretreated net or (2) a non-pretreated which has subsequently been soaked with insecticide at any time. 2 An insecticide treated net (ITN) is (1) a factory treated net that does not require any further treatment or (2) a pretreated net obtained within the past 12 months or (3) a net that has been soaked with insecticide within the past 12 months. Appendix D – Results According to Old Provinces | 303 Table D.9.3 Use of mosquito nets by women Percentage of all women age 15-49 and pregnant women age 15-49 who slept under a mosquito net (treated or untreated), an ever-treated mosquito net1, and an Insecticide Treated Net2 (ITN) the night before the survey, by old province, Rwanda 2005 Percentage of all women age 15-49 who Percentage of pregnant women age 15-49 who Old province Slept under any net the preceding night Slept under an ever-treated net1 the preceding night Slept under an ITN2 the preceding night Number of women Slept under any net the preceding night Slept under an ever-treated net1 the preceding night Slept under an ITN2 the preceding night Number of women Kigali 30.6 30.4 25.4 879 32.7 32.7 30.4 53 Kigali Ngali 8.7 8.7 6.6 1,116 9.5 9.5 4.9 83 Gitarama 17.3 17.3 13.9 1,211 25.6 25.6 22.7 81 Butare 17.6 17.4 13.8 1,097 31.8 31.8 27.2 79 Gikongoro 4.2 4.2 3.7 652 4.6 4.6 4.6 64 Cyangugu 18.0 17.9 14.9 855 35.4 34.3 31.5 67 Kibuye 12.8 12.8 10.6 645 (14.7) (14.7) (13.2) 49 Gisenyi 6.8 6.6 5.9 1,162 10.1 10.1 7.9 92 Ruhengeri 4.4 4.4 3.6 1,168 12.2 12.2 11.3 96 Byumba 13.6 13.6 11.9 876 19.9 19.9 16.3 83 Umutara 13.8 13.8 12.5 560 29.3 29.3 27.0 52 Kibungo 8.4 8.4 6.0 1,059 22.5 22.5 18.7 97 Note: Figures in parentheses are based on 25-49 unweighted cases. Table D.9.4 Use of Intermittent Preventive Treatment by women during pregnancy Percentages of women who took any antimalarial drugs for prevention, who took SP/Fansidar, and who received Intermittent Preventive Treatment (IPT), during the pregnancy for their last live birth in the five years preceding the survey, by old province, Rwanda 2005 Old province Percentage of women who took any antimalarial drug to prevent or treat malaria during an ANC visit during the last pregnancy Percentage of women who received an Intermittent Preventive Treatment during an ANC visit1 Number of last-born children born in the five years preceding the survey Kigali 9.9 1.1 329 Kigali Ngali 5.1 0.0 507 Gitarama 12.5 1.0 528 Butare 9.0 0.0 490 Gikongoro 0.8 0.0 339 Cyangugu 6.1 0.0 392 Kibuye 2.9 0.0 309 Gisenyi 3.5 0.0 616 Ruhengeri 3.3 0.3 602 Byumba 0.9 0.0 505 Umutara 13.9 0.8 303 Kibungo 4.3 0.0 504 1 Intermittent Preventive Treatment is preventive intermittent treatment with at least two doses of SP/Fansidar during an antenatal care (ANC) visit. 304 | Appendix D – Results According to Old Provinces Table D.9.6 Prevalence and prompt treatment of children with fever Percentage of children under age five with fever in the two weeks preceding the survey, and among children with fever, the percentage who took antimalarial drugs and the percentage who took the drugs the same or next day following the onset of fever, by old province, Rwanda 2005 Among children under age five: Among children under age five with fever: Old province Percentage with fever in the two weeks preceding the survey Number of children Percentage who took antimalarial drugs Percentage who took antimalarial drugs same or next day Number of children Kigali 26.6 467 9.0 0.6 124 Kigali Ngali 13.9 722 22.1 2.2 100 Gitarama 22.1 709 31.8 5.9 157 Butare 37.9 707 14.1 3.7 268 Gikongoro 28.0 493 2.8 0.0 138 Cyangugu 28.0 562 15.8 2.9 157 Kibuye 12.5 448 1.5 0.0 56 Gisenyi 26.2 944 2.5 1.1 247 Ruhengeri 27.7 938 5.1 1.1 259 Byumba 22.2 701 8.1 2.3 156 Umutara 28.6 430 16.7 3.7 123 Kibungo 38.5 675 18.5 3.9 260 Table D.9.7 Type and timing of antimalarial drugs taken by children with fever Among children under age five with fever in the two weeks preceding the survey, the percentage who took specific antimalarial drugs and the percentage who took each type of drug the same or next day after developing fever, by old province, Rwanda 2005 Percentage of children who took drug: Percentage or children who took drug the same or next day: Old province SP/Fansidar Amodiaquine Quinine SP/Fansidar Amodiaquine Quinine Number of children with fever Kigali 0.9 3.6 5.4 0.0 0.0 0.6 124 Kigali Ngali 7.6 13.8 7.1 0.0 1.1 1.1 100 Gitarama 12.3 12.0 15.0 4.1 1.8 1.5 157 Butare 0.8 7.8 5.9 0.3 2.4 1.0 268 Gikongoro 0.5 0.5 1.7 0.0 0.0 0.0 138 Cyangugu 6.5 12.4 4.5 0.5 2.1 0.8 157 Kibuye 0.0 0.0 1.5 0.0 0.0 0.0 56 Gisenyi 0.6 0.6 1.4 0.0 0.6 0.5 247 Ruhengeri 1.6 2.5 2.6 0.0 0.0 1.1 259 Byumba 2.7 5.6 3.9 0.7 2.3 0.7 156 Umutara 5.0 6.5 8.8 0.5 1.6 2.2 123 Kibungo 8.6 9.5 6.5 2.7 3.0 0.4 260 Table D.10.1 Initial breastfeeding Percentage of children born in the five years preceding the survey who were ever breastfed, and for the last children born in the five years preceding the survey ever breastfed, the percentage who started breastfeeding within one hour and within one day of birth and the percentage who received a prelacteal feed, by old province, Rwanda 2005 Old province Percentage ever breastfed Number of children Percentage who started breastfeeding within 1 hour of birth Percentage who started breastfeeding within 1 day of birth1 Percentage who received a prelacteal feed2 Number of breastfed children Kigali 96.3 500 43.9 59.9 25.8 481 Kigali Ngali 97.8 809 50.9 58.7 11.9 792 Gitarama 96.0 776 49.1 64.8 19.4 744 Butare 98.0 802 43.5 58.9 26.6 786 Gikongoro 97.6 544 44.9 56.6 27.4 531 Cyangugu 96.3 632 38.6 56.9 17.0 609 Kibuye 97.3 489 37.7 57.2 20.3 476 Gisenyi 97.7 1,029 33.2 47.0 28.1 1,006 Ruhengeri 96.8 1,032 31.9 45.8 26.3 999 Byumba 97.7 798 47.4 58.9 26.1 780 Umutara 96.9 488 41.4 56.8 26.0 472 Kibungo 96.6 816 34.8 60.3 29.7 788 Note: Table is based on all births whether the children are living or dead at the time of interview. 1 Includes children who started breastfeeding within one hour of birth. 2 Children given something other than breast milk during the first three days of life before the mother started breastfeeding regularly. Appendix D – Results According to Old Provinces | 305 Table D.10.3 Median duration and frequency of breastfeeding Median duration of any breastfeeding, exclusive breastfeeding, and predominant breastfeeding among children born in the three years preceding the survey, percentage of breastfeeding children under six months of age living with the mother who were breastfed six or more times in the 24 hours preceding the survey, and mean number of feeds (day/night), by old province, Rwanda 2005 Median duration (months) of breastfeeding Frequency of breastfeeding among children under six months of age Old province Any breastfeeding Exclusive breastfeeding Predominant breastfeeding Number of children Percentage breastfed 6+ times in past 24 hours Mean number of day feeds Mean number of night feeds Number of children Kigali 20.7 3.7 3.9 316 97.6 6.4 6.0 44 Kigali Ngali 26.9 5.8 6.3 524 100.0 8.4 5.4 84 Gitarama 28.4 5.1 5.2 479 98.7 8.0 5.1 70 Butare 26.3 5.8 5.9 497 97.0 8.1 4.2 77 Gikongoro 27.1 6.8 6.9 348 98.6 7.7 4.9 53 Cyangugu 25.3 4.7 5.7 399 98.5 7.3 5.3 59 Kibuye 28.0 6.1 6.3 315 96.8 7.2 6.4 49 Gisenyi 25.3 6.0 6.0 646 97.8 7.2 6.1 115 Ruhengeri 23.5 6.3 6.3 652 98.8 7.7 6.8 111 Byumba 24.8 6.1 6.3 525 98.8 10.5 7.1 80 Umutara 21.8 5.7 5.8 309 94.3 7.9 4.5 59 Kibungo 21.5 5.2 5.4 525 99.0 7.9 7.5 72 Note: Median and mean durations are based on current status. Table D.10.5 Iodization of household salt Percent distribution of households with salt tested for iodine content, by level of iodine in salt (parts per million), percentage of households tested, and percentage of households with no salt, according to old province, Rwanda 2005 Iodine content among households tested Old province None (0 ppm) Inadequate (<15 ppm) Adequate (15+ ppm) Total Number of households Percentage of households tested Percentage of households with no salt Number of households Kigali 0.4 12.6 87.0 100.0 540 81.5 8.5 664 Kigali Ngali 0.4 6.4 93.2 100.0 944 92.3 6.4 1,023 Gitarama 0.8 1.6 97.6 100.0 983 89.4 8.2 1,100 Butare 0.1 9.4 90.5 100.0 879 89.0 8.8 988 Gikongoro 0.1 5.2 94.6 100.0 548 86.5 11.8 633 Cyangugu 0.3 67.4 32.4 100.0 622 85.6 11.7 726 Kibuye 8.9 13.4 77.7 100.0 542 90.6 7.1 598 Gisenyi 1.3 5.7 93.0 100.0 866 80.9 13.2 1,071 Ruhengeri 2.0 7.1 90.9 100.0 872 80.7 16.2 1,081 Byumba 0.1 1.5 98.4 100.0 797 91.9 6.9 867 Umutara 1.0 11.0 88.0 100.0 472 85.8 13.5 550 Kibungo 0.4 7.1 92.5 100.0 849 87.5 6.9 970 Table D.10.6 Micronutrient intake among children Percentage of youngest children under age three living with the mother who consumed fruits and vegetables rich in vitamin A in the seven days preceding the survey, percentage of children age 6-59 months who received vitamin A supplements in the six months preceding the survey, and percentage of children under five living in households using adequately iodized salt, by old province, Rwanda 2005 Old province Consumed fruits and vegetables rich in vitamin A1 Number of children Consumed vitamin A supplements Number of children Living in households using adequately iodized salt2 Number of children Kigali 63,9 240 82,0 422 85,7 416 Kigali Ngali 52,2 423 83,6 637 92,8 696 Gitarama 71,7 392 88,1 639 97,0 665 Butare 55,2 380 87,4 628 91,1 650 Gikongoro 52,4 283 93,3 440 95,3 443 Cyangugu 55,5 310 82,3 500 31,3 520 Kibuye 61,6 251 82,7 398 76,4 420 Gisenyi 54,0 505 67,8 828 93,2 803 Ruhengeri 60,5 507 86,7 826 91,2 766 Byumba 63,7 412 93,7 620 98,1 665 Umutara 47,3 239 82,0 370 88,3 396 Kibungo 57,4 391 84,3 599 91,7 614 Note: Information on vitamin A supplements is based on mother’s recall. na = Not applicable 1 Includes pumpkin, red or yellow yams or squash, carrots, red sweet potatoes, green leafy vegetables, mango, papaya, and other locally grown fruits and vegetables t