Nigeria - Demographic and Health Survey -1989

Publication date: 1989

ONDO STATE, NIGERIA DEMOGRAPHIC AND HEALTH SURVEY 1986 Ondo State, Nigeria Demographic and Health Survey 1986 Medical/Preventive Health Division Ministry of Health, Akure Ondo State, Nigeria Institute for Resource Development/Macro Systems, Inc. Columbia, Maryland April 1989 This report presents the findings of the Ondo State Demographic and Health Survey (ODHS), implemented by the Government of Ondo State, through the Medical/Preventive Health Division of the Ministry of Health in 1986. The survey is part of the worldwide Demographic and Health Surveys (DHS) Program, which is designed to collect data on fertility, family planning, and maternal and child health. Additional information on this survey can be obtained from the Medical/Preventive Health Division, Ministry of Health, State Secretariat, PMB 712, Akure, Ondo State, Nigeria. The Ondo State Demographic and Health Survey was carried out with the assistance of the Institute for Resource Development (IRD), a Macro Systems company with headquarters in Columbia, Maryland. Funding for the survey was provided by the U.S. Agency for International Development (Contract No. DPE-3023-C-00-4083-00). Additional information about the DHS Program can be obtained by writing to: DHS Program, IRD/Macro Systems, Inc., 8850 Stanford Boulevard, Suite 4000, Columbia, Maryland 21045, USA (Telephone: 301-290-2800, FAX: 301-290-2999, Telex: 87775). CONTENTS Page CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii L IST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v L IST OF F IGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv MAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii 1. BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 Geography and History of Nigeria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Geography and Climate of Ondo State . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Population and the Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Health Priorities and Programmes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Population and Family Planning Policies and Programmes . . . . . . . . . . . . . 2 Objectives of the Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Selection of Ondo State for a DHS Survey . . . . . . . . . . . . . . . . . . . . . . . 3 Organization of the Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Population Statistics for Ondo State . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Background Characteristics of ODHS Respondents . . . . . . . . . . . . . . . . . . 5 2. MARRIAGE AND OTHER DETERMINANTS OF FERTIL ITY . . . . . . . . . . . . . . . 11 2.1 2.2 2.3 2.4 Current Marital Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Polygyny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Age at First Marriage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Breastfeeding, Postpartum Amenorrhoea, and Abstinence . . . . . . . . . . . . . 15 3. FERTIL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.1 3.2 3.3 3.4 3.5 Fertility Data in the ODHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Levels, Differentials and Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Age-Specific Fertility Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Children Ever Born . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Age at First Birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 4. CONTRACEPT IVE KNOWLEDGE AND USE . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1 4.2 Contraceptive Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Ever Use of Contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 iii Page 4.3 4.4 4.5 4.6 4.7 4.8 Current Use of Contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Trends in Family Planning Knowledge and Use . . . . . . . . . . . . . . . . . . . 36 Sources of Contraceptive Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Attitudes Abort Pregnancy and Rea,~ons for Nonuse . . . . . . . . . . . . . . . . 38 Intention to U,,e Contraception in the Future . . . . . . . . . . . . . . . . . . . . . 40 Approval c f Family Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 5. FERTIL ITY PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.1 5.2 Future Fertility Preferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Ideal Number of Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 6. MORTAL ITY AND HEALTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 6.1 6.2 6.3 6A Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Maternity Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Child Health ?'ndicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Anthropometric Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 APPENDIX A SURVEY DEHGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 A.1 A.2 A.3 AA Sample Design and Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Questionnaire Design and Pretest . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Main Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Data Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 APPENDIX B SAMPL ING ERRORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 APPENDIX C QUESTIONNAIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 iv LIST OF TABLES Page Table 1.1 Schedule of Activities for the Ondo State Demographic and Health Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Table 1.2 Number of Selected Primary Sampling Units by Local Government Area and Urban-Rural Residence, ODHS, 1986 . . . . . . . . . . . . . . . . . . 5 Table 1.3 Percent Distribution of Women Age 15-49 by Background Characteristics, 1980 NDSS and 1986 ODHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Table 1.4 Percent Distribution of Women by Education, According to Age and Area of Residence, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Table 1.5 Percent Distribution of Respondents by Housing Characteristics and Percentage Owning Certain Household Possessions, by Residence, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 2.1 Percent Distribution of All Women by Current Marital Status, According to Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Table 2.2 Percentage of Currently Married Women in Polygynous Unions, by Age, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . 12 Table 2.3 Percent Distribution of Women by Age at First Union and Median Age at First Union, According to Current Age, ODHS, 1986 . . . . . . . . . . . . . . . . . 13 Table 2.4 Median Age at First Union Among Women Aged 25-49 Years, by Current Age and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . 14 Table 2.5 Percentage of Births Whose Mothers Are Still Breastfeeding, Postpartum Amenorrhoeic, Abstaining, and Insusceptible, by Number of Months Since Birth, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Table 2.6 Mean Durations of Breastfeeding, Postpartum Amenorrhoea, Abstinence, and Insusceptibility by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 3.1 Total Fertility Rates for 1981-83, 1984-86, and the Five-Year Period Prior to the Survey, and Mean Number of Children Ever Born to Women Aged 40-49, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Table 3.2 Age-Specific Fertility Rates (Per 1000 Women) for the Five-Year Period Prior to the Survey, by Area of Residence, ODHS, 1986 . . . . . . . . . . . . . . . 22 Page Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8 Percent Distribution of All Women and Currently Married Women by Number of Children Ever Bom, According to Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Mean Number of Children Ever Bom to Ever-Married Women, by Age at First Marriage and Years Since First Marriage, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Percent Distribution of All Women by Age at First Bin.h, According to Current Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Median Age at First Birth Among Women Aged 25-49 Years, by Current Age and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . 25 Percentage Knowing Any Method, Knowing Any Modem Method, and Knowing Specific Methods, Among All and Currently Married Women by Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Percentage of Currently Married Women Knowing At Least One Modem Method, by Number of Living Children and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Percent Distribution of Women Who Have Ever Heard of a Contraceptive Method by the Main Problem Perceived in Using the Method, According to Method, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Percent Distribution of Women Aged 15-49 Knowing a Contraceptive Method by Source Where They Would Obtain the Method, According to Method, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Percentage of All and Currently Married Women Who Have Ever Used Family Planning Methods, by Method and Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Percent Distribution of All Women and Currently Married Women by Contraceptive Method Currently Used, According to Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Percent Distribution of Currently Married Women by Contraceptive Method Currently Used, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Percent Distribution of All Women and Women Who Have Ever Used Periodic Abstinence by Knowledge of the Fertile Period During the Ovulatory Cycle, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 vi Page Table 4.9 Table 4.10 Table 4.11 Table 4.12 Table 4.13 Table 4.14 Table 4.15 Table 4.16 Table 4.17 Table 4.18 Knowledge, Ever Use, and Current Use of Specific Contraceptive Methods Among All Women 15-49, Nigerian Fertility Survey (Ondo State), 1981/82 and ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Percent Distribution of Current Contraceptive Users by the Most Recent Source of Supply, According to the Method Used, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Among Women Who Are Not Pregnant, Not Using Contraception, and Are Sexually Active, the Percent Distribution by Attitude Toward Becoming Pregnant in the Next Few Weeks, According to Number of Living Children, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Among Women Who Are Not Pregnant, Not Using Contraception, and Who Would Not Be Happy If They Became Pregnant, the Percent Distribution By the Main Reason for Nonuse, According to Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Percent Distribution of Currently Married Nonusers by Intention to Use in the Future, According to Number of Living Children, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 For Currently Married Nonusers Who Intend to Use a Contraceptive Method in the Future, Percent Distribution by Preferred Method, According to Intention to Use in Next Twelve Months or Later, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Percentage of All Women Who Believe That it is Acceptable to Have Family Planning Messages on the Radio or Television, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Percentage of Currently Married Women Knowing at Least One Contraceptive Method, by Husband's and Wife's Attitudes Towards Family Planning Use, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Percentage of Currently Married Women Knowing at Least One Contraceptive Method Who Think That Their Husband Approves of Family Planning Use, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Percentage of Currently Married Women Knowing at Least One Contraceptive Method, Who Have Discussed Family Planning With Their Husband at Least Once in the Past Year, by Woman's Age and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . 44 vii Page Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Percent Distribution of Currently Married Women 15-49 by Desire for More Children According to Number of Living Children, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Percent Distribution of Currently Married Women by Desire for More Children, According to Age, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Percentage of Currently Married Women 15-49 Who Want No More Children by Number of Living Children and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Among Currently Married Women, the Percent Who Are In Need of Family Planning, and the Percent Who Are In Need and Plan to Use a Contraceptive Method in the Future, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Percent Distribution of All Women by Ideal Number of Children and Mean Ideal Number of Children for All Women and Currently Married Women, According to Number Living Children, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Mean Ideal Number of Children for All Women by Age and Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Infant and Child Mortality Estimates 1981-1986, ODHS, 1986 . . . . . . . . . . . . . . . 55 Infant and Child Mortality 1981-1986, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Mean Number of Children Ever Born, Surviving, and Dead, and Proportion of Children Dead Among Those Ever Born, by Age of Women, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Percent Distribution of Births in the Last Five Years by Type of Prenatal Care for the Mother and Percentage of Births Before Which the Mother Received a Tetanus Toxoid Injection, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Percent Distribution of Births in the Last Five Years by Type of Assistance at Delivery, by Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 viii Page Table 6.6 Among All Children Under Five, the Percentage With Health Cards, the Percentage Who Are Recorded as Immunised on the Health Card or Who Are Reported by the Mother as Having Been Immunised, and Among Children With Health Cards, the Percentage for Whom BCG, DPT, Polio and Measles Immunisations are Recorded on the Health Card, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Table 6.7 Among Children Under Five Years, the Percentage Reported by the Mother as Having Had Diarrhoea in the Two Weeks Preceding the Survey and, Among Children With Diarrhoea, the Percentage Receiving Various Treatments, According to Background Characteristics of Child and Mother, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . 62 Table 6.8 Among Mothers of Children Under Five Years, the Percentage Who Know About ORT by Education, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Table 6.9 Among Children Under Five Years, the Percentage Who Are Reported by the Mother as Having Had Fever in the Past Four Weeks and, Among Children Who Had Fever, the Percentage Receiving Various Treatments, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Table 6.10 Among Children Under Five Years, the Percentage Who Are Reported by the Mother as Having Suffered from Severe Cough and/or Difficult Breathing in the Past Four Weeks and, Among Children Who Suffered from Severe Cough and/or Difficult Breathing, the Percentage Receiving Various Treatments, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . 65 Table 6.11 Percent Distribution of Children Aged 6-36 Months by Standard Deviation Category of Height-for-Age Using the Intemational NCHS/CDC/WHO Reference, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Table 6.12 Percent Distribution of Children Aged 6-36 Months by Standard Deviation Category of Weight-for-Height Using the International NCHS/CDC/WHO Reference, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Table 6.13 Percent Distribution of Children Aged 6-36 Months by Standard Deviation Category of Weight-for-Age Using the International NCHS/CDC/WHO Reference, According to Background Characteristics, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 ix Page APPENDIX A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Table A.1 Population Estimates for Ondo State, by Local Government Area . . . . . . . . . . . . . 78 Table A.2 Population Estimates in Selected Enumeration Areas (EAs), EA Demarcation Exercise and DHS Household Listing . . . . . . . . . . . . . . . . . . . . . . . . 79 Table A.3 Household Response Rate and Eligible Women Response Rate by Residence, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 APPENDIX B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Table B.1 Table B.2 Table B.2 Table B.2 Table B.2 Table B.2 Table B.2 List of Variables for Which Sampling Errors Were Calculated, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Sampling Errors: Ondo State, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Sampling Errors: Urban Areas, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Sampling Errors: Rural Areas, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Sampling Errors: Women Aged 15-24, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . 93 Sampling Errors: Women Aged 25-34, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . 94 Sampling Errors: Women Aged 35-49, ODHS, 1986 . . . . . . . . . . . . . . . . . . . . . . . 95 LIST OF FIGURES Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 5.1 Figure 5.2 Figure 6.1 Figure 6.2 Figure 6.3 Page Family Planning Knowledge and Use, Currently Married Women 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Current Use of Family Planning by Method, Currently Married Women 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Current Use of Family Planning by Education and Residence, Currently Married Women 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Source of Family Planning Supply, Current Users . . . . . . . . . . . . . . . . . . 38 Fertility Preferences, Currently Married Women 15-49 . . . . . . . . . . . . . . . 46 Fertility Preferences by Number of Living Children and AU Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Age Distribution of Measured Children and An Children . . . . . . . . . . . . . 66 Cross-tabulation of Weight-for-Height and Height-for-Age . . . . . . . . . . . . . 70 Nutritional Status of Children 6-36 Months . . . . . . . . . . . . . . . . . . . . . . 72 xi PREFACE The Ondo State Demographic and Health Survey (ODHS) was conducted in one of the twenty- one states of the Federal Republic of Nigeria. The ODHS was carded out in all of the seventeen Local Government Areas of the State. The publication of the final report of the ODHS has come at a time when the Ondo State Government is contemplating an overhaul of the health and welfare system which serves the riverine areas of the state. I therefore hope that results of the ODHS will assist the govemment in making decisions. It is also expected that the findings for the riverine population of Ondo State, as presented in this report, will be of value to other states of the Federation which face the problem of providing health services to such population groups. Fieldwork for the Ondo State DHS was carried out from September 1986 to January 1987 by the Government of Ondo State through the Medical/Preventive Health Division of the Ministry of Health. The survey was jointly financed by the Ondo State Government and the United States Agency for International Development (USAID). Technical assistance for the survey was provided by the Institute for Resource Development, Inc. The Nigeria National Population Bureau (NPB) contributed senior project staff for technical support. The United Nations Children's Fund (UNICEF), Nigeria provided vehicles during the fieldwork phase of the survey. The achievements recorded in this undertaking would not have been possible without the efforts and dedication of many administrative officials of the Federal Ministry of Health and both the Ondo State Ministries of Health and Finance and Economic Planning with the support of intematinnal agencies. I commend the efforts of those people and organizations who contributed to the success of the Ondo State Demographic and Health Survey. I wish to express my gratitude to the following people for their efforts: Dr. Femi Fajewonyomi, former Honourable Commissioner of Health, Ondo State; Mr. Omolorun Modupe, Secretary to the Military Government and Head of Service; Dr. H.T. Aladesawe, ODHS Survey Director and Permanent Secretary, Ministry of Health, Ondo State; Dr. A. Adetunji, former Principal, School of Health Technology, Ondo State; Mrs. C. Adekunle, Technical Director of ODHS, National Population Bureau, Lagos; Mr. A. Akinyande, ODHS Field Supervisor, Ministry of Health, Ondo State; Mrs. J.A. Babalola, ODHS Data Processing Supervisor and Health Statistician, Ministry of Health, Ondo State; Dr. Ronald Schoenmaeckers, Ms. Anne R. Cross, Mr. Brian Taaffe, Ms. Jeanne Cushing, and Dr. Jeremiah M. Sullivan, DHS Programme Staff, Institute for Resource Development; Mr. Irwin Shorr and Dr. Christopher Scott, consultants to the DHS Programme. Dr. T.A. Adebayo Honourable Commissioner of Health Akure, Ondo State, Nigeria xii i SUMMARY The Ondo State Demographic and Health Survey (ODHS) was conducted by the Ondo State Ministry of Health (MOH) as part of the worldwide Demographic and Health Survey programme. The primary objective of the survey was to provide information about fertility, family planning, and maternal and child health to the MOH. The ODHS was conducted from September 1986 to January 1987 and collected data from a representative sample of 4213 women aged 15-49. The overwhelming majority of survey respondents are Yoruba (88 percent); 84 percent am Christian; 40 percent live in urban areas, 52 percent in mral areas, and 8 percent in rlverine areas. Fertility Fertility in Ondo State is high. The total fertility rate (TFR), which indicates the number of children that a woman would have in her lifetime if she experienced the fertility rates of a particular time period, was 6.0. The TFR was approximately the same in urban and rural areas (5.9 and 6.0). However, it was substantially higher for women with no education (6.7) or a primary education (7.1) than for women with a secondary or higher education (5.4). The ODHS found evidence of recent declines in fertility. The evidence comes from a comparison of the l r r t with the average number of children ever born to women who are at the end of their childbearing years. The "I'I~R for Ondo State (6.0) was lower by about one child than the number of children ever born to women 40-49 (6.9). Marriage is almost universal among the women of Ondo State, although the median age at first marriage is high (20 years for women age 25-49). In addition to the pattern of late marriage, ODHS data suggest recent increases in the age at first marriage. The proportion of women that reported being married by age 20 was lower for women in the age group 20-24 (38 percent) than for women age 25-29 (48 percent) or women age 30-34 (60 percent). Polygyny is common; among currently married women, 46 percent reported that their husbands had other wives. The women of Ondo State reported relatively long durations of breastfeeding, amenorrhoea and sexual abstinence following the birth of a child. On average, women breasffeed 18 months, are amenorrhoeic 14 months, and practice abstinence for 23 months. The long period of abstinence following a birth is of particular importance in determining birth intervals for women. Family Planning Forty-eight percent of all women reported having knowledge of a contraceptive method. Knowledge of modem methods was about the same, 47 percent. There are clear differentials in knowledge by area of residence and education; urban women and more educated women have the highest levels of contraceptive knowledge. Respondents most frequently reported knowledge of the pill (35 percent) and injection (35 percent); for the IUD, condom, and female sterilization, about 20 percent of respondents reported knowledge of each method. Contraceptive prevalence is low in Ondo State. Ever-use was reported by only 15 percent of women, current use by 9 percent. Current users rely primarily on the pill and periodic abstinence (3 XV percent of women using each) and the condom (2 percent). Current use is higher for urban women, more educated women, and women with many living children. An interesting finding was that the level of current use for all women is higher than that of married women (9 versus 6 percent). With respect to sources of supply, private sector sources (pharmacies, shops and physicians) supplied 38 percent of current users and public sector sources (government hospitals and health centers) provided supplies to 37 percent of current users. The ODHS contained a number of questions which were intended to measure the level of approval of family planning. In response to a question about family planning messages on radio and television, 85 percent of respondents indicated that this was acceptable. Almost one-quarter (23 percent) of married women wanted no more children, 38 percent wanted another child after two years, 20 percent wanted another child within two years and 18 percent were undecided about having another child or the preferred timing of another child. The fact that 61 percent of married women either wanted no more children or wanted to wait at least two years before having another child indicates a need for family planning services. Infant and Child Mortality The infant mortality rate (IMR) was 56 per 1,000 live births and the under five mortality rate was 108 per 1,000 for the period 1981-86. A strong mortality differential was found by the length of the birth interval. The IMR was almost twice as high for a birth following an interval of less than two years (80 per 1,000) than for a bi~a following an interval of two or more years (46 per 1,000). Maternal and Child Health Care The ODHS collected information on the health care received by women during pregnancy for births in the five years preceding the survey. Respondents reported receiving a prenatal checkup by a trained practitioner (doctor, nurse or midwife) for 80 percent of these births; women received a tetanus injection for 71 percent of births. ODHS respondents were asked if they could show a health card for their children under age five and, if they could, information about immunisations was recorded on the survey questionnaire. For the children in the age group 12-23 months, mothers were able to show health cards for 37 percent of children. All of these children had received at least one immunisation and 70 percent were fully immurdsed (i.e., BCG, measles and three doses of DPT and polio). Nutritional Status of Children Height and weight data were collected for 1,387 children and used to calculate indices of nutritional status. The height-for-age index measures linear growth and is used to detect chronic undemutrition. A child who is two or more standard deviations below the reference mean for this index is considered short or stunted. In the ODHS, 32 percent of children were found to be stunted. The weight-for-height index is used to detect recent or acute undemutrition. A child who is two or more standard deviations below the reference mean on this index is referred to as wasted. In the ODHS, 7 percent of children were found to be wasted. A cross-tabulation of these two indices found that 2 percent of children were both stunted and wasted, that is to say, severely undemourisbed. xvi NIGERIA ONDO STATE, NIGERIA North | Ondo Ekiti Central Ekiti . East Ekiti South eAkure Akure Idanre/Ifedore Akoko North South Owo Ifesowapo Ikale Ilaje/Eseodo N r • State Capital Local Government Area xviii 1. BACKGROUND 1.1 Geography and History of Nigeria The Federal Republic of Nigeria is located on the coast of West Africa. It has a land area of 923,766 square kilometres. The officially accepted population figure from the 1963 census is 55.7 million. Estimates for 1986 indicate a population of about 100 million, making Nigeria the most populous country in Africa. Nigeria gained independence from Britain in 1960 and the parliamentary system of govemment was adopted. In 1963, Nigeria was declared a Federal Republic. A major event in the post-independence period was a three-year civil war from which the Federation emerged intact in 1970. Nigeria was divided into 12 states in 1967 and redivided into 19 States in 1976. The later date was the year that Ondo State was created out of the former Western State. Two additional states were created in 1987 so that Nigeria now consists of 22 administrative divisions: 21 states and Abuja, the federal capital territory. 1.2 Geography and Climate of Ondo State Ondo State is located on the coast in the southwestern part of Nigeria (see map). It has two main vegetation belts: 1) the rain forest with tall evergreen trees and thick undergrowth, which extends from the south coast to about 50 miles inland (the riverine area of Ondo State) and 2) the deciduous forest, which is characterised by tall trees and thin undergrowth. There are two seasons: the rainy season from April to October and the dry season from November to March. Ondo State experiences heavy rainfall during the rainy season and high temperatures throughout the year--between 25 and 33 degrees Celsius. 1.3 Population and the Economy Population estimates for Ondo State must be viewed with caution because there has been no official accepted population census in Nigeria since 1963 and because there is some controversy over the accuracy of the figures reported by that census (Ekanem, I.I., 1972). Nevertheless, according to the census figures, the population of the area now comprising Ondo State was 2.7 million in 1963. In developing the sample for the ODHS, the population of Ondo State was estimated by various procedures. The evidence available suggests that the population of Ondo State in 1986 was about 3.3 million. The rate of natural population increase in Nigeria was 2.5 percent per annum in the 1960s, increasing to more than 3.0 percent by the early 1980s (Federal Ministry of Health, 1988). The rate for Ondo State was probably the same. Such rates would lead to a population characterised by a young age structure and a high dependency ratio. Fertility and mortality rates are not known for Ondo State. However, crude birth and death rates for all Nigeria are estimated to be 45-48 per 1,000 and 13-16 per 1,000 respectively. There is little reason to think that the rates for Ondo State differ greatly (National Population Bureau, 1984). The population of Ondo State is predominately made up of Yombas (85 to 90 percent) who are native to southwestern Nigeria. The population is also predominately Christian (85 to 90 percent). Overall, the settlement pattern of Ondo is about 40 percent urban, 55 percent rural and 5 percent riverine. The urban population resides in seven major urban centers each with a population in excess of 50,000 and twenty smaller towns each with a population over 20,000. Commerce is a thriving activity in the urban areas where there is also some industry; the major industries being textiles, palm-oil and cocoa products, and building materials. In most of the state, farming is the main occupation. Ondo Stale is divided into 17 Local Government Areas (see map), one of which, llaje Eseodo, is a riverine area. This is the low-lying area adjacent to the coast where the land is continually dissected by frequent branching of the Oluwa River. In the habitable area of ljero Eseodo, somewhat inland from the mangrove swamps of the coast, fishing villages are located along river banks. Access to many of these villages is mainly by boat. 1.4 Health Priorities and Programmes The Ondo State Ministry of Health (MOH) is responsible for the health of all members of the population. The main strategy of the MOH has been the provision of primary health care with emphasis on maternal and child health care services. The MOH programme consists of prenatal, postnatal and child welfare clinics, which are held in health facilities on specific days of the week. Basic health education is an important activity at these clinics and an effort is made to promote breastfeeding, to instruct in the use of oral rehydration therapy and to leach good nutrition and food hygiene practices. Since 1983, a top priority has been the Expanded Programme of Immunisation in which children are immunised against six major childhood diseases: diphtheria, whooping cough, tetanus, tuberculosis, poliomyelitis and measles. Basic health services are provided through facilities operated by the MOH (24 hospitals and 52 health centers), the Local Government Area Councils (191 malemity centers and 229 dispensaries), missionary groups (6 hospitals and 6 malemity centers) and private sector institutions (53 small hospitals, 70 health and maternity centers and numerous pharmacies). The capacity of the public sector and missionary facilities, in terms of in-patient beds, is as follows: hospitals, 2,200 beds; health centers, 500 beds; and maternity centers, 1,500 beds. While these facilities tend to be located in the urban areas, they are nevertheless well distributed throughout the state. Only the riverine area suffers from a relative shortage of health facilities. 1.5 Population and Family Planning Policies and Programmes During the decade of the 1980s, there has been increasing concern over the high rate of population growth in Nigeria. Thus, in 1988, a policy fostering the provision of family planning services came into being with the adoption of the National Policy on Population for Development, Unity, Progress and Self-Reliance. An underlying principle of this policy is that "all couples and individuals have the basic right to decide freely and responsibly on the number and spacing of their children and to have the information, education and means to do so." Notwithstanding the recent genesis of a national population policy, family planning services have been available in Ondo State for a number of years. Family planning activities were initiated by the Planned Parenthood Federation of Nigeria (PPFN) in 1969. By 1985, PPFN was providing family planning services free-of-charge through eight clinics in urban areas. In 1981, the provision of family planning services in a MOH hospital was begun by a project sponsored by the United Nations Fund for Population Activities. In 1984, the International Training in Health Programme (INTRAH) held a workshop to develop the capacity of the MOH to train staff to provide family planning services. By 1986, 150 nurses and midwives were trained and providing services throughout the state. Although the reporting of family planning service statistics in Ondo State is incomplete, the available statistics indicate that by 1986, PPFN was providing services to about 4,000 new acceptors per year and the MOH was providing services to about 8,000 new acceptors per year. 1.6 Objectives of the Survey The primary objective of the ODHS is to provide policymakers and planners with information on fertility, family planning, maternal and child health, and infant and child mortality. The data will be used to evaluate existing health and family planning programmes and to aid in the design of new programmes. Another objective of the survey is to test the feasibility of conducting a population-based survey in Ondo State where, as with the rest of Nigeria, a sampling frame is available for only part of the State. A third objective is to document recent changes in population characteristics and fertility and family planning practices in Ondo State. 1.7 Selection of Ondo State for a DHS Survey The ODHS differs from most other surveys in the DHS Programme which are national in scope. Ondo State was chosen for a DHS survey for two reasons. First, the Ondo State MOH expressed a need for, and interest in, the type of data collected in DHS surveys. Second, the efforts of the National Population Board to provide a population-based sampling frame for Nigeria through the Enumeration Area Demarcation Exercise had progressed furtJaer in Ondo State than in any other state of Nigeria. Without the Enumeration Area Demarcation Exercise, it would have been difficult to implement the ODHS. In 1985, when the survey was designed, it had been 22 years since the 1963 Population Census, the last officially accepted census of Nigeria. A census was attempted in 1973 but abandoned because of data irregularities. By 1980 there was considerable uncertainty about the size and distribution of the national population. To rectify this situation, the National Population Bureau began the Enumeration Area Demarcation Exercise in 1983. By 1985, this project had not progressed very far, except in Ondo State, where new enumeration areas had been created in 13 of the state's 17 Local Government Areas. 1.8 Organization of the Survey The ODHS was funded by the United States Agency for International Development and implemented by the Ondo State Ministry of Health. Technical assistance was provided by the Institute for Resource Development, located in Columbia, Maryland. The National Population Bureau participated in the survey by providing one senior technical staff member from its Lagos office and the manpower for sample implementation from its Ondo State office. The United Nations Children's Fund provided vehicles during the fieldwork phase of the survey. The ODHS survey activities took place over a 38 month period, from February 1986 to March 1989. The schedule of activities is shown in Table 1.1 and described in Appendix A. Two survey instruments were used in the ODHS: a household schedule and an individual woman's questionnaire. Both were adapted from the Model "B" Core Questionnaires of the DHS Programme. The English version of the questionnaires is reproduced in Appendix C. 3 Table 1.1 Schedule of Activities for the Ondo State Demographic and Health Survey Activity Time Period Survey Design and Questionnaire Development Feb-April 1986 Preparation of Sampling Frame March-June 1986 Pretest June-July 1986 Household Listing in Sample PSUs July 1986 Printing Questionnaires August 1986 Training for Main Survey August-Sept 1986 Fieldwork for Main Survey Sept 1986-Jan 1987 Data Entry and Editing Sept 1986-Mar 1987 Tabulations for Preliminary Report May 1987 Publication of Preliminary Report August 1987 Special Tabulations of Owe LGA (for UNICEF) September 1987 Tabulations for First Country Report August-Nov 1987 Analysis and Report Preparation Dec 1987-Mar 1989 Publication of First Country Report April 1989 The sample was a stratified, self-weighting probability sample, representative of the entire state. The primary sampling units were enumeration areas created by the National Population Bureau and the second stage units were households. Table 1.2 shows the distribution of the 90 selected primary sampling units by Local Govemment Area and type of area (urban, rural or riverine). A household listing operation was carried out in these sampling units and household selection was done at the survey office in Akure using systematic random sampling. Details of the sample design are presented in Appendix A. Eligibility for the woman's questionnaire was on a de facto basis: all women aged 15-49 who had stayed in a sample household during the previous night were eligible respondents. The sample design specified a target of 3,600 completed interviews from female respondents. For the survey, a total of 4,213 completed interviews were obtained. 1.9 Population Statistics for Ondo State Population statistics for Ondo State are available from two national-level surveys of Nigeria: the 1980 National Demographic Sample Survey (NDSS) and the 1981-82 Nigeria Fertility Survey (NFS). The latter was conducted as part of the World Fertility Survey Programme. Three relevant publications from these surveys are National Demographic Sample Survey, Ondo State (National Population Bureau and Institute for Resource Development/Westinghouse, 1988), Principal Report of the Nigeria Fertility Survey (National Population Bureau, 1984) and Nigeria Fertility Survey, Ondo State Report (National Population Bureau and Federal Ministry of Health, 1986). Unfortunately, the last report is based on only 389 female respondents, so the information available is limited. When appropriate, statistics from the above three sources are cited in this report. 4 Table 1.2 Number of Selected Pr imary Sampl ing Units by Local Government Area and Urban-Rural Residence, ODHS, 1986 No. of Pr imary Sampling Units Local Government Area Urban Rural River ine Total i Akure 8 1 9 2 Akoko North 3 4 7 3 Ekit i East 2 2 4 4 Ere 2 2 4 5 Ekiti South 3 1 4 6 Ekiti South-west 1 1 2 7 Ekit i Central 2 3 5 8 lJero 2 1 3 9 Ekit i West 1 2 3 I0 Akoko South 1 3 4 ]i Ekiti North 2 4 6 12 Owe 3 5 8 13 Idanre Ifedore 2 3 5 14 Ondo 2 7 9 15 I fesowapo 2 2 16 Ilaje Eseodo 9 9 ]7 Ikale 2 4 6 Total 36 45 9 90 1.10 Background Character i s t i cs o f ODHS Respondents This section presents information on the demographic characteristics of ODHS respondents. Table 1.3 shows the distribution of respondents by selected characteristics along with Ondo State data from the 1980 National Demographic Sample Survey (NDSS). Age Data on age were obtained in the ODHS by asking respondents two questions, "In what month and year were you bom?" and "How old were you at your last birthday?" Interviewers were instructed that, in cases where respondents could not provide a year of birth, it was essential to obtain information on age and they were trained in techniques of probing for age. As a last resort, interviewers were instructed to estimate the respondent's age. The year of birth was recorded for eighty-five percent of respondents: both month and year for 66 percent and year only for 19 percent. Fifteen percent of respondents did not provide a year of birth. Age was recorded for all 4,213 respondents. The age distribution of respondents in the ODHS and the NDSS, by five-year age groups, is shown in Table 1.3. Both distributions conform to the pattem characteristic of a high fertility population, i.e., more women in the younger than in the older age groups. However, the ODHS distribution is somewhat irregular; the percentage of women is high in the age group 15-19 (26 percent), declines to 13 percent for the age group 20-24, remains constant for the two following age groups, declines to 11 percent for the age group 35-39 and remains constant again for the two following age groups. This contrasts with the NDSS distribution where, above age 25, the percentages in each successive age group gradually decline. In terms of broad age groups, the distribution of respondents is about the same in the ODHS and the NDSS with the percentage under 30 being 53 and 56 percent, respectively. Moreover, the substantial decline in the percentage of women between age groups 15-19 and 20-24 in the ODHS is consistent with the enumeration of twice as many women in the age group 10-14 as in the age group 15-19 in the 1980 NDSS (National Population Bureau and Institute for Resource Development/Westinghouse [1988]). Thus, the ODHS survey found the same size difference between adjacent age groups as was found six years earlier in the NDSS. Nevertheless, peculiarities of the ODHS age distribution, particularly the fiat spots over the broad ages from 20 to 34 and 35 to 49, suggest errors in age reporting and those errors may affect some of the survey results. Table 1.3 Percent Distr ibut ion of Women Age 15-49 by Background Characterist ics, 1980 NDSS for Ondo State and 1986 ODDS Background 1980 1986 Background 1980 1986 Character ist ic NDSS ODHS Character ist ic NDSS ODHS Age Rel ig ion f 15-19 21.6 26.3 Cathol ic |80.6 10.3 20-24 16.6 13.4 Protestant [ 74.4 25-29 17.6 13.3 Moslem 13.0 13.4 30-34 14.9 13.0 Tradit ional 5.3 1.0 35-39 13.2 11.4 None 1.2 0.9 40-44 9.7 11.3 45-49 6.3 11.3 Ethnicity Yoruba 87.8 85.8 Residence Other 12.2 14.2 Urban 31.7 40.2 Rural 168.3 52.0 River ine 7.8 Educat ion None 53.5 36.9 Pr imary 20.5 23.8 Secomdary 19.2 33.4 Post-Secondary 6.7 5.8 Total Percent I00.0 100.0 Number of Women 8867 4213 Source; Nat ional Populat ion Bureau and Institute for Resource Development/ Westinghouse [1988]: Tables I-2, I-3, I-4, 11-2. Residence The distribution of respondents by residence area was as follows: 40 percent urban, 52 percent rural, and 8 percent riverine. This represents a somewhat greater concentration of the population in urban areas than was found in the 1980 NDSS (32 percent) which suggests a net rural-to-urban migration flow in recent years. Religion and Ethnicity The majority of ODHS respondents reported that they were Christian (74 percent Protestant and 10 percent Catholic), another 13 percent reported that they were Muslim, and 2 percent followed either a traditional religion or bad no religious affiliation. In terms of ethnicity, the women of Ondo State were overwhelming Yoruba (86 percent). Education All women were asked if they had ever attended school. Respondents who had attended school were asked the highest level of school attended (primary, secondary or pest-secondary) and the highest class or year completed at that level. Thirty-seven percent of respondents reported no education; 24 percent reported completing between one and six years of primary education; 33 percent, between one and four years of secondary education and 6 percent reported post-secondary education. Relative to 1980 NDSS results, the ODHS shows a pronounced increase in secondary education. This reflects the success of the government's efforts to promote education since 1976 when the provision of free primary education was started. Table 1.4 shows percent distributions of women by education, five-year age groups and urban- rural residence. The table indicates strong associations between educational attainment and age and between educational attainment and area of residence. Younger women have much higher levels of education than older women. For example, 80 percent of women aged 15-19 have a secondary or higher education while only 3 percent of women aged 45-49 attained that level. The association between education and area of residence, while significant, is not as pronounced as in the case of age. The proportion of women with a secondary or higher education increases from 22 percent in riverine areas to 48 percent in urban areas. Table 1.4 Percent Distr ibut ion of Women by Education, According to Age and Arsa of Residence, ODES, 1986 Educat ion Background Number Character~ of istic None Pr imary Secondary + Total Women Age 15-19 4.4 15.1 80.5 100.0 1109 20-24 13.0 23.5 63.5 100.0 563 25-29 30.7 30.4 38.9 I00.0 560 30-34 49.8 36.3 13.9 I00.0 548 35-39 56.1 33.5 10.4 100.0 478 40-44 67.2 24.3 8.6 I00.0 478 45-49 83.7 13.0 3.3 100.0 477 Residence Urban 29.1 22.8 48.1 I00.0 1695 Rural 40.4 24.8 34.8 100.0 2192 River ine 54.0 23.6 22.4 I00.0 326 Total 36.9 23.9 39.2 I00.0 4213 7 Housing Characteristics The ODHS collected a limited amount of information on the dwelling units in which respondents lived (Table 1.5). The flooring material of the dwelling units was either cement (73 percent), packed earth (21 percent), or wood planking (6 percent). Cement flooring was most common in urban areas (95 percent). A majority of respondents in rural areas (65 percent) also had cement flooring, while wood planking was most prevalent in riverine areas (63 percent) where homes are built on stilts over swampy terrain. Table 1.5 Percent Distr ibut ion of Respondents by Housing Character ist ics and Percentage Owning Certain Household Possessions, by Residence, ODHS, 1986 Residence Housing Charac- ter ist ic /Pos- session Urban Rural Riverlne Total Mater ia l of F loor Cement 95.1 65.4 I0.i 73.1 Earth 4.2 33.2 27,3 21.1 Wood Planks 0.7 1.4 62.6 5.8 Drinking Water Source Piped to Home 10.6 2.0 0.0 9.3 Publ ic Tap 55.1 17.6 0.0 31.3 Well 16.6 5.3 5.8 9.9 R iver /Spr ing 16.6 73.5 86.8 81.7 Tanker Truck 1.1 1,6 7.4 1.8 Source of Water for Household Use Piped to Home 10.4 1.8 0.0 5.2 Publ ic Tap 39.4 12.4 0.O 22.3 Well 29.3 8.0 4,6 16.4 River /Spr ing 19.7 77.2 89.0 54.9 Tanker Truck 1.2 0.6 6.4 1.2 Total i00.0 i00.0 I00.0 I00.0 Household Possess ions E lectr ic i ty 78.1 23.3 7.1 44,1 Televis ion 40.1 9.1 6.1 21.4 Refr igerator 28.5 6.4 8.0 14.8 Radio 73,1 56.9 35.0 61.7 Bicycle 12.8 17.6 6.4 14.5 Motorcycle 26.4 27.8 8.3 25.7 Car 20.1 10.3 1.8 13.6 Tractor 0.4 i.i 1.2 0.9 Sources of Water Information was also collected on the sources of water for drinking and for general household use. Sources of water differ considerably by area of residence. In urban areas, piped water is the primary source of drinking water--ll percent of respondents have water piped to their homes and another 55 percent obtain water from a public tap. In rural and riverine areas, water from rivers and springs is the main source of drinking water (74 and 87 percent, respectively). In both urban and rural areas, well water is relied on to a greater extent for general household use than for drinking. This reflects the fact that wells are more accessible than public taps throughout Ondo State. Household Amenities/Possessions Respondents were also asked about household amenities (electricity, television, refrigerator and radio) and about the ownership of specific vehicles of transportation by a household member (bicycle, motorcycle, car, and tractor). Electricity is available to the majority of women in urban areas (78 percent) but to a minority in rural (23 percent) and riverine (7 percent) areas. Television and refrigerators are common in urban areas (40 percent and 29 percent of respondents, respectively) but not in rural and riverine areas (less than 10 percent of respondents). Household possession of a radio is common in all areas, but more prevalent in urban (73 percent) and rural (57 percent) than in riverine (35 percent) areas. In terms of vehicles, 20 percent of urban women and 10 percent of rural women resided in households in which a member owned a car, 25 percent of urban and rural women lived in households in which a member owned a motorcycle. These percentages were much lower in riverine areas. 2. MARRIAGE AND OTHER DETERMINANTS OF FERTILITY 2.1 Current Marital Status In the ODHS, marriage is def'med as including all stable unions regardless of the degree of formal recognition. Respondents were first asked if they had ever been married or lived with a man, and if yes, whether they were currently married, living with a man, widowed, divorced or separated. In all age groups, less than one percent of respondents reported themselves as living with a man; for that reason the categories "married" and "living with a man" are combined and presented as "married" throughout this report. Table 2.1 shows the marital status of respondents at the time of the survey. Of all women aged 15-49, 30 percent had never married, 67 percent were currently married and 3 percent were widowed, divorced or separated. These findings are similar to those of the National Demographic Sample Survey where the comparable statistics were 27, 69 and 4 percent, respectively. Table 2.1 Percent Distr ibut ion of All Women by Current Mar i ta l Status, Accord ing to Age, ODHS, 1986 Age of Never L iv ing Total No. of Women Marr ied Marr ied Together Widowed Divorced Separated Percent Women 15 - 19 88.6 9.7 0.6 O.0 0.3 0.8 I00.8 1109 20 - 24 40.i 57.2 0.7 0.0 8.0 2.0 100.0 563 25 - 29 7.3 89.8 0.9 0.7 0.4 0.9 100.0 560 30 - 34 0.0 98.2 0.0 1.3 0.2 0.4 100.0 548 35 - 39 0.0 96.9 0.2 2.1 8.6 0.2 100.0 478 40 - 44 0.0 95.2 0.2 4.4 0.0 0.2 I00.0 478 45 - 49 0.0 88.9 0.2 9.6 0.2 I.I 100.0 477 Total 29.7 66.8 0.5 2.1 0.2 0.8 I00.0 4213 The marital status distributions by age group show that the percentage of never married women is high for women aged 15-19 (89 percent), decreases rapidly in age groups 20-24 and 25-29 (40 and 7 percent, respectively) and is nil for women above age 30. The percentage of currently married women is low for women aged 15-19 (10 percent), increases rapidly, reaching 90 percent for women aged 25-29 and remains high for women at older ages. At every age, a high percentage of women are either never married or currently married, so the percentage widowed, divorced or separated is low (less than 5 percent) for all age groups except 45-49. 2.2 Polygyny Polygyny, the marriage of more than one woman to the same man, is common in Nigeria. Data on this practice were collected by asking currently married women if their husbands had other wives. Table 2.2 shows that overall, 46 percent of currently married women are in a polygynous union, which is 11 consistent with the comparable statistic of 41 percent for southwestem Nigeria reported by the 1981-82 Nigeria Fertility Survey (National Population Bureau, 1984, Table 4.12). Polygyny is associated with a distinct age pattern. The percentage of women in a polygynous union is lowest for women 15-19 (17 percent), increases at subsequent ages and reaches a maximum for women 45-49 (62 percent). The similarity of the ODHS statistics with those found by the 1981-82 Nigerian Fertility Survey for women of southwestem Nigeria (National Population Bureau, 1984) and by the 1980 NDSS for women of Ondo State (National Population Bureau and Federal Ministry of Health, 1986) suggests that the age pattern associated with polygyny is a life-cycle phenomenon. Table 2.2 Percentage of Currently Married Women in Polygynous Unions, by Age, According to Background Characteristics, ODHS, 1986 Age Background Characteristic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 17.4 22.1 30.8 41.7 43.2 53.9 59.5 39.6 Rural ii.I 26.8 36.8 49.4 57.1 59.6 62.0 49.6 Riverine 29.2 26.5 59.4 57.6 69.4 76.0 69.7 55.8 Education None 29.0 33.3 44.4 49.4 58.6 61.8 62.6 55.2 Primary 16.1 26.4 39.6 50.8 51.6 55.9 56.4 45.2 Secondary + 11.3 19.3 22.5 23.7 25.0 41.0 64.3 23.3 Rel igion Cathol ic 5.9 17.4 31.1 42.2 41.3 64.3 72.7 42.1 Protestant 16.4 22.9 31.8 43.9 51.2 57.1 58.4 43.9 Musl im 25.0 37.2 57.1 57.6 63.9 56.9 69.8 56.9 Other/None 40.0 50.0 44.4 53.9 66.7 64.6 69.2 61.8 Total 17.4 24.5 35.4 46.3 52.9 58.6 61.9 46.1 Table 2.2 also contains statistics on the prevalence of polygyny by area of residence, education and religious affiliation. The percentage of women in polygynous unions is lower in urban areas (40 percent) than in rural (50 percent) or riverine (56 percent) areas. Polygyny is also inversely related to education, being lowest among women with secondary or more education (23 percent) and highest for women with no education (55 percent). Religious affiliation shows a less pronounced effect, with the percentage for Catholics (42 percent) and Protestants (44 percent) being somewhat lower than that for Muslims (57 percent). The relatively modest differentials in polygyny, except in the case of education, testify to its general acceptance throughout the population. 12 2.3 Age at First Marriage Women who reported ever being in union were asked the month and year in which they began living with their first partner. If they could not report the year, they were asked the age when they began their first cohabiting relationship. Sixty-two percent of respondents reported the year and the month of first marriage, 20 percent reported the year only, and 18 percent reported the age at first marriage. From these data, statistics on age at first marriage were calculated. However, caution should be exercised when drawing conclusions, since some respondents, particularly older women, may have had difficulty recalling the date of first marriage or their age at first marriage, and interviewers had to probe for the information. Table 2.3 shows the distribution of women in age groups by age at first union. For each age group starting at age 25, the median age at first marriage is also shown. (The median age at first marriage is not shown for age groups under age 25 because less than 50 percent of respondents had married (women 15-19) or some single members of the age group were younger than the median age at marriage of those who had married and, if they married, could lower the median age (women 20-24).) Table 2.3 Percent D is t r ibut ion of Women by Age at First Union and Median Age at First Union, Accord ing to Current Age, ODHS, 1986 Age at First Union Number Median Current Never Total of Age at Age Marr ied < 15 15-17 18-19 20-21 22-24 25 + Percent Women ist union* 15 - 19 88.6 2.0 7.3 2.2 0.0 0.0 0.0 I00.0 1109 20 - 24 40.1 4.6 14.2 19.5 15.1 6.4 0.0 i00.0 563 28 - 29 7.3 7.0 22.1 18.9 19.8 19.3 5.8 i00.0 560 20.2 30 - 34 0.O 10.4 29.2 20.4 17.5 15.3 7.1 i00.0 548 18.9 35 - 39 0.0 5.4 25.1 23.2 22.4 14.6 9.2 100.0 478 19.7 40 - 44 0.0 8.2 24.9 20.5 20.1 16.1 10.3 100.0 478 19.7 45 - 49 0.O 2.5 20.8 28.5 21.8 15.7 10.7 100.0 477 19.9 Total 29.7 8.3 18.6 16.8 14.2 I0.7 5.I 100.0 4213 * Def ined as the age by which one-ha l f of women have ever marr ied The median age at first marriage for women 25 to 49 is about 20 and shows little variation between five-year age groups. This suggests that age at first marriage was stable in Ondo State from about 1960 to 1980. However, for the period since 1980 there is evidence that age at first marriage has increased. The evidence comes from statistics on the proportion single at age 20 among women 20-24 and 25-29. Table 2.3 indicates that 62 percent of women 20-24 were single at age 20 while only 52 percent of women 25-29 were single at age 20. Additional evidence of a recent change in the age at first marriage can be obtained by comparing statistics for Ondo State from the 1980 NDSS and the 1986 ODHS. The proportion of single women aged 15-19 increased from 80 percent in the NDSS to 89 percent in the ODHS, among women aged 20- 24, 40 percent were single in both surveys (National Population Bureau and Institute for Resource Development/Westinghouse [1988]: Table III-1). Given the problem with age reporting in Nigeria, these 13 results must be interpreted with caution; however, they suggest recent increases in the age at first marriage in Ondo State. Table 2.4 shows differentials in the median age at first marriage for women age 25-49 by area of residence, education and religious affthation. Differences in the median age at marriage exist by area of residence and religious affiliation, but they are small and not always consistent across age groups. The finding of similar values of the median age at first marriage for Muslim (19.7) and Protestant women (19.8) in the ODHS is consistent with the findings of the 1981-82 Nigeria Fertility Survey for Ondo State (National Population Bureau and Federal Ministry of Health, 1986, Table 4). This is distinctly different from the findings of that survey at the national level where the median age at first marriage for Muslim women (15.3) is much lower than for Christian women (19.0) (National Population Bureau, 1984, Table 4.4). The findings for Ondo State are probably a case of the small Muslim minority adopting the norm of the overwhelming majority of the population. Table 2.4 Median Age at First Union Among Women Aged 25-49 Years, by Current Age and Background Character- istics, ODHS, 1986 Current Age Background Characteristic 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 20.5 19.3 20.2 20.1 20.1 20.1 Rural 20.2 18.4 19.3 19.5 19.7 19.5 Riverine 17.6 19.9 20.1 19.0 20.2 19.5 Education None 18.7 18.5 19.2 19.4 19.8 19.3 Primary 19.0 18.7 28.2 20.5 19.7 19.5 Secondary + 22.8 22.4 20.3 19.9 22.0 22.2 Rel ig ion Cathol ic 19.2 19.0 19.3 19.2 19.2 19.2 Protestant 20.4 19.0 19.7 19.8 19.9 19.8 Musl im 19.4 18.7 20.2 19.7 20.3 19.7 Other/None 20.7 18.5 19.2 18.3 20.0 19.2 Total 20.2 18.9 19.7 19.7 19.9 19.7 Pronounced differences in the median age at first marriage are associated with education. For women in the age group 25-49, the median age at first marriage is higher for women with secondary or more education (22.2) than for women with primary (19.5) or no education (19.3). The difference between women with secondary or more education and women with no education amounts to four years for women aged 25-29 (22.8 versus 18.7). 14 2.4 Breastfeeding, Postpartum Amenorrhoea, and Abst inence Fertility is affected by a number of factors other than marriage, including breastfeeding, postpartum amenorrhoea and sexual abstinence. In the ODHS, information on these factors was obtained for all live births in the five-years preceding the survey. In this section information is reported for births within 36 months of the survey. Table 2.5 presents the proportion of babies who are still breastfeeding, the proportion of women still amenorrhoeic and the proportion of women abstaining from sexual intercourse by months since the birth. The proportion of women who are insusceptible to the risk of pregnancy (either amenorrhoeic or abstaining) is also presented. Table 2.5 shows that the practice of bmastfeeding is nearly universal in Ondo State and that it continues for a prolonged period. Three months after a birth, over 90 percent of mothers are breastfeeding and, a year after a birth, approximately 75 percent are still breastfeeding. Overall, the median duration of breastfeeding is 17 months. This extended period of breastfeeding is consistent with the estimate of 18 months from the 1981-82 Nigeria Fertility Survey for women of Ondo State (National Population Bureau and Federal Ministry of Health, 1986) and with recent median estimates for sub- Saharan countries such as Liberia, 17 months (Chieh-Johnson, D., et al., 1988) and Senegal, 19 months (Ndiaye, S., et al., 1988). Table 2.5 Percentage of Births Whose Mothers Are Sti l l Breastfeeding, Postpar tum Amenorrhoe ic , Absta in ing , and Insusceptible, by Number of Months S ince B i r th , ODHS, 1986 Propor t ion of B i r ths Whose Mothers Are: Months Number Since Breast- ~a~enor- Absta in - Insus - of Bi r th feed ing rhoe ic ing cept ib le B i r ths* Less than 2 94.9 90.7 96.9 97.9 97 2 - 3 92.5 86.8 94.3 97.2 106 4 - 5 94.4 80.8 94.4 97.6 125 6 - 7 83.3 67.6 84.3 87.0 108 8 - 9 88.0 63.9 81.2 84.2 133 10 - ii 84 .4 59.4 76 .6 81.5 128 12 - 13 71.7 48.3 69.2 74.2 120 14 - 15 74.2 45.0 68.0 71 .9 128 16 - 17 56.1 31.8 68.2 71 .9 107 18 - 19 34.3 20.4 47.2 50.0 108 20 - 21 35.2 13.6 47.7 47.7 88 22 - 23 14.5 7.2 44 .6 45.8 83 24 - 25 12.9 7.1 29.4 31.8 85 26 - 27 3.3 3.3 29.7 29.7 91 28 - 29 1.0 2.0 17.7 17.7 102 30 - 31 5 .6 3.7 21.3 22.2 108 32 - 33 4 .6 2.3 23 .9 23 .9 68 34 - 35 3.3 2.2 21.1 21.1 90 .Tota l Med ian 17.1 12.2 18.2 18.5 * Inc ludes b i r ths occur r ing in the per iod 0-35 months pr io r to the survey. 1895 15 Following the birth of a child, there is a period of time during which a woman is infecund-- physiologically incapable of conception. This period can be approximately measured by the time from pregnancy termination to the retum of the woman's menstrual cycle. The duration of postpartum amenorrboea depends on a number of factors, such as the general health and nutritional status of the woman and her breastfeeding practices. Among healthy, well-nourished women who do not breastfeed at all, postpartum amenorrhoea usually lasts two to three months. However, intensive breast:feeding which is continued well aRer the birth of a child can extend this period to a year or more. The median duration of the postpartum amenorrhoea for women in Ondo State is 12 months. This is consistent with the estimate of 13 months from the 1981-82 NFS for women of Ondo State (National Population Bureau and Federal Ministr) of Health, 1986). In Ondo State the practice of sexual abstinence is almost universal following the birth of a child (Table 2.5). The median duration of abstinence is 18 months. The 1981-82 NFS reported a mean duration of abstinence of 21 months for women of Ondo State (National Population Bureau and Federal Ministry of Health, 1986). The ODHS finding that the median duration of sexual abstinence is slightly longer than the median duration of breastfeeding, but substantially longer than the median duration of postpartum amenorrhoea implies that sexual abstinence plays a crucial role in determining pregnancy intervals. The relatively long period of insusceptibility characteristic of women in the Ondo State (median duration 19 months) is primarily due to the practice of sexual abstinence. Table 2.6 presents the mean durations for breastfeeding, postpartum amenorrhoea, sexual abstinence, and insusceptibility for subgroups of the population. The mean durations were calculated by the "current status" method in which the number of women who had a birth in the last 36 months and were still breastfeeding (amenorrhoeic, abstaining or insusceptible) is divided by the average number of births per month in the last 36 months. While there are important differences in the mean durations of breastfeeding and postpartum abstinence, the discussion here focuses on abstinence because it is the primary determinant of the length of insusceptibility for women in Ondo State. The mean duration of sexual abstinence is shorter for women under age 30 (20 months) than for women age 30 and above (25 months) and the insusceptible period is similarly shorter for younger women (21 months) than older (26 months) women. In terms of area of residence, abstinence is shorter for urban women (20 months) than for rural women (26 months) and shortest for riverine women (19 months). As a result, insusceptibility is about the same for urban and riverine women (22 and 21 months, respectively) but longer for rural women (27 months). In the case of religious affiliation, Catholic and Protestant women have shorter periods of abstinence (22 months) than Muslim women (27 months) and similar differences in insusceptibility (23 versus 28 months). Education is by far the most important background characteristic affecting the period of postpartum insusceptibility for women of Ondo State. The durations of bmastfeeding, postpartum amenorrhoea, and abstinence are longest for women with no education (20, 16 and 27 months, respectively), decrease by about 3 months for women with a primary education (18, 14 and 21 months, respectively) and by another 3 months for women with a secondary education (14, 11 and 18 months, respectively). The net result is a steady and substantial decrease in the length of insusceptibility between women with no education (29 months), women with a primary education (22 months) and women with a secondary education (19 months). All other factors being equal, this would mean that better educated women would have greater need for family planning methods to space their births. 16 Table 2.6 Mean Durat ions of Breastfeeding, Postpartum Amenorrhoea, Abstinence, and Insusceptibi l i ty by Background Character- istics, ODHS, 1986 Background Number Character- Breast- Amenor- Insuscep- of istlc feeding rhoea Abst inence t ibi l i ty Births Age <30 18.1 13.4 20.2 21.3 858 30 + 18.4 14.5 24.9 26.3 1067 Residence Urban 16.3 12.5 20.2 21.6 841 Rural 19.8 15.1 25.8 26.8 932 River ine 19.4 15.4 18.5 21.1 152 Educat ion None 20.4 16.3 26.7 26.6 803 Pr imary 18.4 13.6 21.4 22.4 656 Secondary + 14.4 10.6 18.0 18.6 466 Rel igion Cathol ic 17.9 13.9 21.8 23.2 233 Protestant 17.6 13.4 22.0 23.4 1362 Musl im 21.0 16.4 27.0 27.8 297 Other/None 21.8 18.6 22.9 25.1 33 Total 18.2 14.0 22.8 24.1 1925 Note: Means are current status est imates based on births occurr ing in the per iod 0-35 months pr ior to the survey. 17 3. FERTIL ITY 3.1 Fertility Data in the ODHS This chapter presents fertility indices describing different aspects of childbearing among women in Ondo State. First, statistics depicting fertility levels during the period 1981-86 are presented--total fertility rates and age-specific fertility rates. Then, statistics on the total number of children ever born are presented by age group. Finally, statistics on the median age at first birth are presented by age group and background characteristics. Data on the childbearing experience of respondents were collected in several ways. First, women were asked their aggregate fertility in a series of six questions: the number of sons and daughters that live at home, the number of sons and daughters that live elsewhere and the number that had died. Second, data pertaining to specific births were collected with a truncated birth history for the six year period from 1981 to the survey date (i.e., respondents were asked to report their most recent live birth and each preceding live birth extending back in time to the birth immediately preceding 1981). For all reported births, information was obtained on month and year of birth, sex, name, survival status and, if dead, age at death. Third, respondents were asked the month and year of their first birth, if that birth occurred prior to the births enumerated in the truncated birth history. The truncated birth history is an established procedure for the collection of retrospective fertility data. This approach is designed to provide data suitable for the estimation of fertility levels for a period immediately prior to a survey--in the case of the ODHS, for the period 1981-86. As with any retrospective procedure, the truncated birth history can suffer from event underreporting (in particular, the underreporting of children who die in early infancy) and misreporting of the dates of birth. The former problem is potentially serious for infant mortality estimation, while the latter problem primarily concerns fertility estimation. For example, when the truncated birth history approach is used, misreported birth dates can transfer events across the reference boundaries and bias period-specific fertility rates. In order to limit the effect of such misreporting, fertility data should be collected for a longer retrospective period than that for which fertility rates are actually calculated. In the ODHS, fertility data were collected for all births since 1981 and for the immediately preceding birth, while estimation of fertility rates is limited to the period since 1981. 3.2 Levels, Differentials and Trends in Fertility Estimates of fertility levels are presented in terms of the total fertility rate (TFR) and the mean number of children ever born. The "1"1~1¢. is a period fertility index which indicates the number of children that a woman would have if she experienced throughout her lifetime the age-specific fertility rates of a particular time period. The mean number of children ever born is a retrospective fertility index which indicates the actual number of children women have had. Fertility trends can be determined by comparing the "lrK for recent time periods with the number of children ever bom to women aged 40-49. In Table 3.1, TFRs for women up to age 45 are presented for two three-year calendar periods preceding the survey (1981-83 and 1984-86) and one five-year period preceding the survey (which, because the fieldwork for the survey was primarily done in late 1986, approximates 1982-86). The pair of three-year estimates suggest a substantial decline in fertility from 6.3 children per woman in 1981-83 to 5.6 children per woman in 1984-86; a decline of 0.7 children per woman. However, caution must be exercised in interpreting these three-year estimates. In the ODHS, interviewers were trained to probe for 19 the age of the child and the season of its birth when the respondent could not report a date of birth. This information was used by the interviewer to calculate the date of birth. Respondent digit preference in reporting the age of a child could have resulted in an erroneous concentration of births in the fourth and fifth years before the survey and an erroneous appearance of a fertility decline between 1981-83 and 1984-86. Final resolution of this issue is beyond the scope of this report. It is possible that the TFR did not decline by as much as 0.7 children per woman between 1981-83 and 1984-86. In these circumstances the most conservative approach is to rely on the intermediate estimate of the I~R for the longer time period: i.e., 6.0 for the five-year period preceding the survey. Table 3.1 Total Fert i l i ty Rates for 1981-83, 1984-86, and the Five- Year Per iod Pr ior to the Survey, and Mean Number of Chi ldren Ever Born to Women Aged 40-49, According to Background Characterist ics, ODHS, 1986 Total Fert i l i ty Rates* Mean No. of Chi ldren Background 0-4 Years Ever Born to Character ist ic 1981-83 1984-86 Before Women 40-49 Residence Urban 5.9 5.7 5.9 6.4 Rural 6.6 5.5 6.0 7.1 River ine 6.3 6.1 6.4 7.7 Educat ion None 6.9 6.3 6.7 7.0 Pr imary 7.3 6.8 7.1 7.0 Secondary + 5.9 5.1 5.4 6.1 Rel ig ion Cathol ic 6.7 6.4 6.6 7.5 Protestant 6.2 5.5 5.8 6.9 Musl im 6.6 6.0 6.4 6.7 Other/None 8.4 3.8 6.3 6.9 Total 6.3 5.6 6.0 6.9 * For women aged 15-44 Table 3.1 provides other evidence that fertility has declined over the last twenty years. The retrospective measure of fertility, the number of children ever born to women aged 40-49 (6.9), exceeds the Tt'K for the recent period (6.0). This indicates that fertility has declined about one child per woman over the last two decades. With respect to area of residence, differences between the TI-R and children ever bom are apparent in all areas, although somewhat smaller in urban (5.9 versus 6.4) than in rural (6.0 versus 7.1) and riverine areas (6.4 versus 7.7). With respect to religious affdiation, the difference between the TH, t and children ever born is less pronounced for Muslim women (6.4 versus 6.7) than for Catholic (6.6 versus 7.5) and Protestant women (5.8 versus 6.9). It appears that fertility change in Ondo State is being experienced to about the same extent by Catholic and Protestant women and to a much lesser extent by Muslim women. With respect to 20 educational attainment, differences between the TFR and children ever born are minor for women with no education (6.7 versus 7.0) and primary education (7.1 versus 7.0), but more substantial for women with secondary or more education (5.4 versus 6.1). This suggests that education has played an important role in bringing about the decline in fertility in Ondo State over the last twenty years. If this is so, the fact that the percentage of women with secondary or more education is much higher among younger than among older women (80 percent versus 3 percent, Table 1.4) suggests that further declines in fertility may occur. Evidence for fertility decline in southwest Nigeria is recent. The Nigerian Fertility Survey found a TFR for southwestem Nigeria for the period 1977 to 1981 of 6.6 and a mean number of children ever born to women 45-49 of 6.0 (National Population Bureau, 1984, Tables 5.11 and 5.12). The relatively low estimate for the number of children ever boru may have been due to underreporting of births by older women in the NFS or may reflect low fertility in the late 1960s during the Nigerian Civil War (when women 45-49 in 1981-82 were in their peak childbearing years). Notwithstanding the absence of a definitive explanation for the pattem of estimates from the Nigerian Fertility Survey, the ODHS estimates clearly indicate a fertility decline. Differentials in fertility for population subgroups are also shown in Table 3.1. For the five-year period preceding the survey, there is little difference between the TFR for urban (5.9) and rural (6.0) areas and, while the TFR for riverine areas (6.4) is somewhat higher, it is based on relatively few woman-years of exposure. Differentials by religious affiliation indicate that fertility is higher for Catholic (6.6) and Muslim women (6.4) than for Protestant women (5.8). The largest fertility differentials occur between education groups; the TFR is highest for women with a primary education (7.1), lowest for women with a secondary education (5.4) and intermediate for women with no education (6.7). The finding of higher fertility for women with a primary education than for women with no education is not uncommon in sub-Saharan countries and was found in the Liberia DHS survey (Chieh-Johnson, et al., 1988). In the case of Ondo State, this differential is not due to a difference in the age at which childbearing begins, since age at first marriage and at first birth are about the same for women with no education and women with a primary education (Tables 2.4 and 3.6). Instead, the difference probably arises as a result of longer intervals between births for women with no education, who practice sexual abstinence for a longer period following the birth of a child than do women with a primary education (27 versus 21 months, Table 2.6). 3.3 Age-Specific Fertility Rates Age-specific fertility rates for the five-year period immediately preceding the ODHS are shown in Table 3.2. Fertility is quite low for the age group 15-19 (61 per 1,000), rises to a peak for the 25-29 age group (301 per 1,000), remains high for women 30-34 (293 per 1,000) and declines thereafter. This pattem of fertility is similar to the pattern for southwestern Nigeria reported by the 1981-82 NFS--i.e., a broad-peaked fertility schedule with little variation in fertility between the ages of 20 and 35 (National Population Bureau, 1984, Table 5.12). Age-specific rates for urban and rural areas are almost identical. The primary difference is the more rapid decline in fertility in urban areas for women 40-44 and 45-49. This sharper decline of fertility rates for older women in urban areas was found throughout Nigeria in the 1981-82 NFS (National Population Bureau, 1984, Table 5.12). 21 Table 3.2 Age-Speci f ic Fert i l i ty Rates (Per i000 Women) for the F ive-Year Period Pr ior to the Survey, by Area of Residence, ODHS, 1986 Age-Speci f ic Fert i l i ty Rates Age of Women Ondo State* Urban Rural 15-19 61 60 53 20-24 238 234 246 25-29 301 300 302 30-34 293 290 291 35-39 194 200 188 40-44 llO 87 120 45-49 (40) (28) (45) TFR (10-44) 6.0 5.9 6.0 TFR (15-49) 6.2 6.0 6.2 Note: Numbers in parentheses represent part ia l ly t runcated rates. * Includes riverine areas 3.4 Children Ever Born Table 3.3 shows the mean number of children ever born--the cumulative childbearing experience--for women 15-49 by age group. Among all women, the mean number of children ever born is low for women 15-19 (0.1), remains low for women 20-24 (0.8) (reflecting the late age of marriage in Ondo State) and increases rapidly for older age groups so that the completed family size of women 45-49 is quite high (7.3). Among currently married women, the number of children ever bom for women 15-19 (0.6) is higher than among all women, rises with age and reaches the same peak value in the age group 45-49 (7.3) as for all women, reflecting the near universality of marriage and the low rates of widowhood and divorce in Ondo State. It is worth noting that in many surveys, statistics on children ever bom rise steadily up to the age groups 35-40 or 40-44, at which point they decline. This pattern generally reflects underreporting of events by the oldest cohorts of women--most probably underreporting of children who have died or who have moved out of the home. The ODHS statistics show a steady rise in the number of children ever bom with increasing age of women and a substantial increase between women aged 40-44 (6.5) and aged 45- 49 (7.3). This suggests relatively complete reporting of data on children ever bom. Table 3.3 also shows the percent distribution of women by the number of children they have had. Fewer than 7 percent of women in the age group 15-19 have had a live birth, an indication of the late age at marriage and late onset of childbearing among the women of Ondo State. These distributions also indicate that the incidence of primary infertility, as measured by the proportion of women in the older ages who have had no births, is low--on the order of 1 percent. Finally, the source of the difference in the number of children ever born to women aged 40-44 (6.5) and 45-49 (7.3) is evident from the 22 distributions. The proportion with ten or more births is substantially lower among women 40-44 (10 percent) than among women 45-49 (17 percent). Table 3.3 Percent Distr ibution of All Women and Currently Marr ied Women by Number of Chi ldren Ever Born, According to Age t ODHS, 1986 Number of Children Ever Born (CEB) Age 0 1 2 3 4 5 6 7 8 9 10+ Total NO. Mean of No.of Women CEB ALL WOMEN 15-19 93.1 6.1 0.7 0.i 0.0 0.0 0.0 0.0 0.0 0.0 0.0 i00.0 1109 0,i 20-24 50.6 27.7 14.2 5.0 2.1 0.2 0.2 0.0 0.0 0,0 0.0 i00.0 563 0.8 25-29 Ii.i 18.6 21.6 19°i 13.7 8.4 4.3 2.3 0.9 0.0 0.0 I00.0 560 2.7 30-34 0.7 2.9 9.3 15.3 20.0 19.4 16.5 9.3 4.0 1.3 1.3 I00.0 548 4.6 35-39 0.4 2.3 2.1 7.3 10.7 18.9 21.5 16.9 9.6 4.8 4.4 100.0 478 5.9 40-44 0.4 2.3 1.5 2.3 7.5 20.5 17.8 17.4 11.5 9.2 9.6 100.0 478 6.5 45-49 1.3 1.0 0.8 3.1 5.9 i0.I 17.6 14.0 17.8 10.9 17.4 100.0 477 7°3 Total 33.1 8.8 6.7 6.7 7.4 9.4 9.2 7.0 5.1 3.0 3.7 100.0 4213 3.3 CURRENTLY MARRIED WOMEN 15-19 45.2 47.0 7.0 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 I00.0 115 0.6 20-24 19.0 43.6 24.5 8.6 3.7 0.3 0.3 0.0 0.0 0.0 0.0 ]00.0 326 1.4 25-29 4.1 19.7 23,4 20.9 14.4 9.3 4.7 2.6 1.0 0.0 0.0 i00o0 508 2.9 30-34 0.7 2.8 9.5 15.1 20.3 19.7 16.2 9.3 3.9 1.3 i.i 100.0 538 6.6 35-39 0.4 2.4 2,2 7.5 10.6 19.0 21.3 17.2 8.9 5.0 4.5 i00.0 464 5.9 40-44 0,4 2.2 1.3 2.2 7.0 20.0 18.2 18.2 11.4 9.4 9.6 i00.0 456 6,6 45-49 1.2 0.9 0.9 3.3 5.4 10.4 17.4 14.6 17.2 ii.i 17.6 100.0 425 7.3 Total 5.2 11.9 9.8 9.7 10.5 13.3 13.0 I0.0 7.0 4.2 5.2 i00.0 2832 4.7 In many populations, age at first marriage is strongly associated with subsequent childbearing performance--women marrying at younger ages typically have more children than women marrying at older ages. Table 3.4 shows data on the cumulative fertility (number of children ever born) to ever- married women by age at first marriage and duration since first marriage. Among all ever-married women, the mean number of children ever born is one for women married 0-4 years and increases to 6.7 for women married 20-24 years. Among women first married between the ages of 15 and 25, differences in childbearing performance are minimal and the number of children ever bom by marriage duration are about the same as for all women. Among these women, the number of children ever born at durations 25- 29 and 30+ years increases to seven and eight, although the relationship between age at first marriage and children ever bern is inconsistent for longer unions. Among women first marrying under age 15 or at age 25 and above, childbearing performance differs from that of all ever-married women. Women marrying under age 15 have fewer children at all durations of marriage than all ever-married women. On the other hand, women marrying at age 25 and above have more children ever born at durations 0-4 and 5-9 and about the same number of children ever born at durations 15-19 and 20-24 as all ever-married women. It appears that the relatively high fertility of these women immediately following marriage is compensated for by lower than average fertility at subsequent durations so that, at durations 15-24 years, their cumulative childbearing performance is about the same as for all ever-married women. This slowing of 23 the pace of childbearing probably results, in part, because of a decrease in the fecundability of these women by the time they reach marriage durations of 15 years and longer. For women first marrying at age 25 and above, the mean number of children ever born at durations 20-24 years can be considered their completed family size since they are unlikely to have significantly more children. This number (6.6) is significantly lower than the number of children ever born at marriage durations 25-29 and 30+ of women marrying between the ages of 15 and 25 and is an indication of the fertility inhibiting effect of late age at marriage. Table 3.4 Mean Number of Chi ldren Ever Born to Ever -Marr ied Women, by Age at First Marr iage and Years Since First Marriage, ODHS, 1986 Age at First Marr iage Years Since First Marr iage <15 15-17 18-19 28-21 22-24 25+ Total 0-4 0.9 0.9 0.9 i.i I.i 1.3 1.0 5-9 2.7 3.3 3.0 2.9 2.9 3.8 3.1 10-14 4.6 4.8 4.5 4.8 4.9 4.9 4.7 15-19 5.6 6.2 6.1 5.8 6.3 5.7 6.0 20-24 6.0 7.1 6.3 6.6 7~0 6.6 6.7 25-29 6.8 7.0 7.3 7.0 7.9 7.1 30+ 6.9 8.7 8.0 8.2 Total 5.1 5.2 4.8 4.4 4.2 4.0 4.7 - Outs ide the age range of the survey 3.5 Age at. First Birth The age at onset of childbearing is an important demographic and health indicator for a population. From a demographic perspective, it indicates the extent to which the childbearing potential of women is being achieved; changes in age at first birth may foreshadow changes in fertility levels in a population. From a health perspective, early childbearing is associated with high levels of maternal and child morbidity and mortality. Table 3.5 presents statistics on the distribution of women by age at first birth for all age groups and on the median age at first birth for age groups above age 25. The value of the latter statistic is not shown for women under age 25 because less than half the respondents in those age groups have had a live birth. In Ondo State, the median age at first birth is oldest for women 25-29 (21.1 years) and ranges between 20.2 and 20.5 years in subsequent age groups. The older age at first birth reported by women 25-29 indicates a trend toward later childbearing in recent years. The existence of a trend to a later age at onset of childbearing can be further investigated by reference to the statistics on the distribution of women by a~e at first birth. Table 3.5 indicates that the proportion of women who had a live birth by age 20 increases from 28 percent for women 20-24 to 39 and 47 percent for women 25-29 and 30-34, respectively. This is evidence that the shift to later childbearing evidenced by women 25-29 is being experienced, to an even greater extent, by women 20-24. 24 Table 3.5 Percent Distr ibut ion of All Women by Age at First Birth, According to Current Age t ODHS r 1986 Age at F irst Birth Median No. Age at Current No Total of First Aqe Births <15 15-17 18-19 20-21 22-24 25+ Percent Women Birth 15-19 93.1 0.9 4.1 2.0 0.0 0.0 0.0 1OO.0 1109 20-24 50.8 3.4 9.4 15.5 15.3 5.7 0.0 100.0 563 25-29 ii.i 4.6 18.4 16.4 18.8 23.7 7.8 100.0 560 21.1 30-34 0.9 5.8 20.4 21.2 20.6 19.5 Ii.5 I00.0 548 20.2 35-39 0.6 3.8 15.9 23.4 21.3 18.6 16.3 100.0 478 20.5 40-44 0.4 6.9 19.9 17.4 17.8 19.9 17.8 i00.0 478 20.5 45-49 1.3 1.5 15.9 24.7 21.4 19.3 15.9 i00.0 477 20.5 Total 33.1 3.4 13.3 15.0 14.1 13.0 8.1 100.0 4213 Table 3.6 presents statistics on the median age at first birth for women aged 25-49 by background characteristics of women. The age at first birth is later in urban areas (20.9 years) than in rural or riverine areas (both 20.3 years) and later for Protestant (20.6 years) and Muslim (20.8) women than for Catholics (20.0 years). However, the greatest difference in the median age at first birth occurs in education. Among women 25-49, the median age is the same for women with no education as for women with a primary education (20.2) but much later for women with secondary or higher education (23.0). The difference is substantial and amounts to almost three years. Table 3.6 Median Age at First Birth Among Women Aged 25-49 Years, by Current Age and Background Characterist ics, ODHS, 1986 Current Age Background Character ist ic 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 21.5 20.6 21.0 20.7 20.7 20.9 Rural 21.1 19.7 20.2 20.4 20.4 20.3 River ine 18.2 21.4 20.5 20.5 20.6 20.3 Education None 19.9 19.9 20.3 20.3 20.5 20.2 Pr imary 19.7 19.8 20.7 21.5 20.4 20.2 Secondary + 23.4 23.2 22.0 20.5 24.0 23.0 Rel ig ion Cathol ic 20.5 19.8 20.3 20.2 19.4 20.0 Protestant 21.4 20.3 20.4 20.6 20.6 20.6 Musl im 20.7 20.2 21.4 21.1 21.0 20.8 Other/Nons 20.0 20.5 19.6 18.0 20.0 19.8 Total 21.1 20.2 20.5 20.5 20.5 20.6 25 4. CONTRACEPTIVE KNOWLEDGE AND USE 4.1 Contraceptive Knowledge Knowledge of contraceptive methods and of places where they can be obtained are preconditions for their use. A basic objective of the ODHS was to determine the level of knowledge of methods. Data on knowledge of family planning methods were collected first by asking respondents to name the ways that a man or woman could keep a woman from getting pregnant. If a respondent did not spontaneously mention a particular method, the method was described by the interviewer and the respondent was asked if she recognized the method. Descriptions were included in the questionnaire for seven modem methods (pill, IUD, injection, condom, vaginal methods (diaphragm, foam, and jelly), male and female sterilization, and two traditional methods--periodic abstinence (rhythm) and withdrawal. In addition, any other methods mentioned by the respondent, e.g., herbs, charms, were recorded. Finally, for any modem method that she recognized, the respondent was asked if she knew about a place or a person from which she could obtain the method. If she reported knowing about periodic abstinence, she was also asked if she knew a place or person from whom she could get information about the method. Survey results indicate that only 48 percent of women in Ondo State know of a contraceptive method (Table 4.1). Injection is the most widely recognized method (36 percent), followed closely by the pill (35 percent). Female sterilization, the IUD, and condom are each recognized by about 20 percent of women, while smaller proportions have heard about periodic abstinence (12 percent), withdrawal (I1 percent), vaginal methods (8 percent), and male sterilization (5 percent). Knowledge of contraceptive methods is highest among women in their 20s and early 30s and lowest among teenagers and women over the age of 45. It is also slightly higher among married women than among all women, with the patterns of knowledge by method and by age being similar for married and all women. Contraceptive knowledge varies considerably according to backgmund characteristics of the woman, as is shown in Table 4.2. A total of 61 percent of married urban women have heard of at least one modem method, compared to 47 percent of married rural women and only 13 percent of married women living in riverine areas of the state. Education, too, seems to have an effect on contraceptive knowledge. Only 34 percent of married women with no education have heard of a modem method, compared to 59 percent of women with primary education and 80 percent of women with secondary education. Differentials by religious affiliation are not so large. Knowledge is somewhat lower among Muslim women (42 percent) than Protestant women (53 percent), with Catholic women intermediate (47 percent). The number of children a woman has does not appear to affect her knowledge of contraceptives, except that childless women are slightly less likely to have heard of a modem method of contraception than women with one or more children. Women who had heard of specific contraceptive methods were asked what they thought was the main problem, if any, with using the method. Results are given in Table 4.3 by method. For all of the modem methods, half or more of respondents answered "don't know", which implies that many women may have heard of methods without knowing much about them. About one-quarter of women cited health reasons as the main problem with using the pill, IUD, injectable contraceptives, and male and female sterilization. One-quarter to one-third of women reported no problem for the condom, periodic abstinence, and withdrawal, however a greater proportion of women reported ineffectiveness as a problem for these methods than for other methods. Problems of access/availability, cost, disapproval of partner, and inconvenience were reported by only small proportions of women. 27 oo Table 4.1 Percentage Knowing Any Method, Knowing Any Modern Method, and Knowing Specific Methods, Among All and Currently Marr ied Women by Age, ODHS, 1986 Percentage of Women Who Know: A~y Diaphragm, Periodic No. ~y Modern InJec- Foam, Female Male ~bstl- With- of Age Method Method* Pill IUD tion Jelly Condom Steril. Steril. hence drawal Other Women ALL WOMEN 15-19 38.0 36.8 24.6 7.7 21.4 3.2 16.3 13.3 1.8 ii.I 7.9 1.4 1109 20-24 55.8 54.9 42.3 20.4 43.0 10.7 27.9 27.5 7.1 20.1 20.6 3.2 563 25-29 59.8 58.7 45.5 32.3 48.4 13.7 29.6 27.0 8.2 18.8 18.2 5.9 560 30-34 56.2 55.5 43.8 28.6 46.7 i0.0 22.4 27.4 6.4 12.8 11.7 5.3 548 35-39 53.3 52.5 38.9 27.6 45.0 9.6 17.8 25.9 5.9 10.5 8.8 4.0 478 40-44 47.1 46.2 32.0 21.1 37.7 7.9 12.6 23.4 4.2 7.1 6.3 4.0 478 45-49 36.9 34.2 22.6 13.2 24.3 2.5 7.1 17.0 2.7 5.9 4.0 5.5 477 All Ages 48.3 47.1 34.5 19.8 36.0 7.7 19.1 21.9 4.8 12.4 10.9 3.8 4213 CURRENTLY MARRIED WOMEN 15-19 40.9 38.3 24.3 11.3 26.1 5.2 11.3 20.9 4.3 10.4 6.1 3.5 115 20-24 53.1 51.8 37.1 19.9 42.6 8.3 21.8 25.8 5.5 13.5 16.0 3.4 326 25-29 58.7 57.7 43.9 32.1 48.0 12.8 27.6 26.4 7.5 16.9 16.3 6.1 508 30-34 57.2 56.5 44.6 29.2 47.6 10.2 22.9 27.9 6.5 13.0 11.9 5.4 538 35-39 53.7 53.0 39.2 27.6 45.3 9.9 17.5 25.9 5.8 9.7 8.4 3.7 464 40-44 47.1 46.5 32.5 21.5 37.9 8.1 13.2 23.9 3.9 7.5 6.6 3.9 456 45-49 37.2 34.6 23.3 13.2 24.9 2.6 6.4 16.9 2.6 4.7 3.5 4.9 425 All Ages 51.1 50.0 36.8 24.0 40.9 8.7 18.2 24.5 5.4 11.0 10.2 4.6 2832 * Modern methods include pill, IUD, injection, diaphragm, foam, Jelly, condom, and male and female steril ization. Table 4.2 Percentage of Current ly Marr ied Women Knowing At Least One Modern Method, by Number of Living Chi ldren and Background Characterist ics, ODH5, 1986 Number of L iving Chi ldren Background Character ist ic 0 1 2 3 4 5 6+ Total Residence Urban 53.7 58.0 71.1 60.2 56.7 58.6 65.2 60.8 Rural 36.9 52.0 43.0 47.8 46.5 47.7 47.6 47.0 Riverine (16.7) 12.8 9.5 14.3 17.4 12.5 II.I 12.9 Education None I0.0 17.6 31.0 37.3 34.4 35.7 38.0 33.6 Pr imary 38.8 52.7 52.9 55.2 62.8 66.7 66.2 59.2 Secondary + 68.4 72.5 80.4 90.2 86.0 90.2 92.2 79.6 Rel ig ion Cathol ic (44.4) 50.0 47.2 47.2 39.1 44.6 54.5 47.4 Protestant 49.6 55.9 58.2 55.7 52.2 51.3 49.1 52.8 Musl im 21.7 34.5 38.2 45.5 42.9 42.4 63.0 42.4 Other/None (0.0) (0.0) (40.0) (7.7) (30.0) (60.0) (28.6) 23.5 Total 43.8 50.7 53.4 51.4 49.0 49.5 49.8 50.0 Note: Numbers in parentheses represent fewer than 20 cases. Table 4.3 Percent Distribution of Women Who Have Ever Heard of a Contraceptive Method by the Main Problem Perceived in Using the Method, ODHS, 1986 Contraceptive Method Diaphragm, Female Male Periodic Main InJec- Foam, Sterlll- Sterill- Abstl- With- Problem Pill IUD tion Jelly Condom zatlon zation nence drawal No Problem 15.7 6.7 8.2 12.1 24.1 5.0 5.0 39.8 32.0 Access/Avallabillt y 0.6 1.7 0.2 0.6 0.i 0.0 0.0 0.0 0.7 Costs TOO Much 0.7 0.8 1.5 0.0 0.2 2.0 4.5 0.0 0.0 Health Concerns 25.1 23.0 25.0 16.1 6.1 30.1 26.7 2.3 2.6 Not Effective 1.4 2.9 i.I 4.0 13.0 0.0 0.0 14.8 19.3 Partner Disapproves 0.7 2.3 1.5 0.6 2.5 0.I 1.5 1.3 3.0 Inconvenient to Use 2.1 3.7 1.4 4.3 5.0 1.2 4.0 5.5 9.8 Other I.i 0.6 0.7 0.3 0.2 1.0 2.5 0.0 0.9 Don't Know, Unstated 52.6 58.3 60.6 61.9 48.7 60.7 55.9 37.1 31.0 Total i00.0 i00.0 i00.0 100.0 I00.0 100.0 100.0 i00.0 100.0 NO. of Women 1453 834 1517 323 806 921 202 523 461 29 Women who had heard of methods were also asked where they would go to get the method if they wanted to use it. By far the most common response was a government hospital (Table 4.4), where half to three-quarters of women knowing a modem method said they would go to obtain the method. Government health centers and private doctors were also cited frequently as sources for most modem methods, while pharmacies and shops were mentioned as sources for the pill and condom. In order to get information about using periodic abstinence, the majority knowing the method said they would go to a government hospital (36 percent), to a private doctor (15 percent), or to nowhere at all (26 percent). It is interesting to note that almost all women who have heard of a method also know a place to obtain it-- generally, only five percent or less said they didn't know where to go to get the method. Table 4.4 Percent Distr ibution of Women Aged 15-49 Knowing a Contraceptive Method by Source Where They Would Obtain the Method, ODHS, 1986 Contraceptive Method Diaphragm, Female Male Periodic InJec- Foam, Sterll i- Sterl l i- Absti- Source Pil l IUD tion Jelly Condom zation zatlon nonce Private Doctor 4.7 5.4 4.9 5.0 3.5 6.9 9.9 14.5 Govt. Hospital 64.6 75.3 75.3 70.3 47.1 82.3 79.2 36.3 Govt. Health Center 10.2 13.1 12.9 12.7 6.9 4.1 6.4 4.4 Mobile Clinic 0.1 0.7 0.6 0.3 0.0 0.0 0.0 0.4 Private Hospital 1.2 1.2 0.8 1.2 1.7 0.I 0.0 0.6 Pharmacy/Shop 13.8 0.6 0.3 5.9 33.0 0.1 1.0 0.4 Field Worker 0.I 0.0 0.1 0.0 0.1 0.0 O.0 1.3 None 0.2 0.I 0.I 0.3 0.6 0.2 0.5 25.6 Other 0.6 0.5 0.9 0.6 1.I 1.1 0.0 11.5 Don't Know 4.5 3.1 4.2 3.7 5.8 5.I 3.0 5.0 Total 100.0 i00.0 I00.0 I00.0 i00.0 i00.0 100.0 I00.0 No. of Women 1453 834 1517 323 806 921 202 523 4.2 Ever Use of Contraception Each respondent who had heard of a contraceptive method was asked if she had ever used it. As shown in Table 4.5, only 15 percent of all women and 13 percent of married women in Ondo State have ever used a contraceptive method. The pill (6 percent), periodic abstinence (5 percent), and condom (4 percent) are the methods used most commonly. Withdrawal has been used by 3 percent, injection by 2 percent, the IUD by 1 percent, and vaginal methods and sterilization by less than one percent of all women. For most methods, ever use is highest among women in their 20s. It is lower among married women than all women, however, the pattern by age and method is similar for both groups. 30 Table 4.5 Percentage of All and Currently Married Women Who Have Ever Used Family Planning Methods, by Method and Age, ODHS, 1986 Contraceptive Method Any Diaphragm, Female Periodic No. Any Modern InJec- Foam, Sterlli- Absti- With- of Age Method Method* Pill IUD tlon Jelly Condom zatlon hence drawal Other Women ALL WOMEN 15-19 14.3 10.7 6.0 0.0 I.I 0.3 8.0 0.0 8.5 2.7 0.4 1109 20-24 21.5 14.6 10.8 0.2 I.I 0.5 7.5 0.0 8.5 6.6 0.7 563 25-29 21.4 14.5 8.4 1.3 1.8 i.i 7.0 0.0 9.1 6.3 0.7 560 30-34 14.2 9.5 5.3 1.8 2.4 0.2 2.0 0.4 5.3 2.2 0.4 548 35-39 13.6 11.7 6.1 3.6 4.4 0.2 2.5 0.0 2.5 2.3 0.4 478 40-44 9.4 7.7 4.0 1.7 2.9 0.4 1.7 0.2 2.3 0.4 0.4 478 45-49 8.6 5.9 2.3 0.8 3.6 0.2 1.5 0.0 2.3 0.8 1.0 477 All Ages 14.9 10.8 6.2 i.i 2.2 0.4 4.4 0.i 5.0 3.1 0.5 4213 CURRENTLY MARRIED WOMEN 15-19 11.3 7.0 4.3 0.0 0.9 0.0 1.7 0.0 4.3 1.7 1.7 115 20-24 14.1 9.2 5.8 0.3 0.9 0.8 4.6 0.0 4.3 4.3 0.9 326 25-29 18.5 12.2 6.7 1.4 1.4 1.0 5.9 0.0 7.1 4.7 0.6 508 30-34 14.5 9.7 5.4 1.9 2.4 0.2 2.0 0.4 5.4 2.2 0.4 538 35-39 13.1 11.4 5.8 3.4 4.3 0.2 2.6 0.0 2.2 2.2 0.2 464 40-44 9.4 7.7 4.2 1.8 2.6 0.4 1.8 0.2 2.4 0.4 0.4 456 45-49 8.0 5.6 2.4 0.9 3.8 0.2 1.2 0.0 1.9 0.9 0.9 425 All Ages 13.0 9.3 5.0 1.6 2.5 0.4 2.9 0.i 4.0 2.4 0.6 2832 * Modern methods include pill, IDD, injection, dlaphra~, foam, Jelly, condom, and male and female sterilization. 4.3 Current Use of Contraception Survey data indicate that nine percent of all women 15-49 in Ondo State are currently using a method of family planning (Table 4.6). About three percent of women (one-third of users) are using periodic abstinence and another three percent are using the pill. The condom is the next most widely used method, with just over one percent of women relying on it. Other methods such as injection, IUD, withdrawal, and vaginal methods are relied on by fewer than one percent of women. Figure 4.1 shows the rapid drop-off in knowledge of methods, knowledge of a source, ever use, and current use of family planning among married women. 31 Table 4.6 Percent Distribution of All Women and Currently Married Women by Contraceptive Method Currently Used r According to Age r ODHSf 1986 Contraceptive Method Currently Used Not Diaphragm, Periodic Cur- Any InJec- Foam, Abstl- With- rently Age Method Pill IUD tlon Jelly Condom nence drawal Other Using Total No. of Wo~en ALL WOMEN 15-19 11.9 4.1 0.0 0.4 0.I 3.5 3.2 0.6 0.1 88.i i00,0 1109 20-24 13.0 4.4 0.2 0.O 0.0 2.7 4.4 0.9 0.4 87.0 100.0 563 25-29 10.4 9.0 0.4 0.9 0.2 0.4 4.5 0.9 0,2 89,6 100.0 560 30-94 7.3 1.6 0.9 I.i 0.0 0.5 2.7 0.0 0.4 92.7 100.0 548 35-39 7.1 1.9 1.5 1.5 0.0 8.4 1.7 0.0 0.0 92.9 100.0 478 40-44 4.8 0.8 0.8 1.9 0.2 0.2 0.6 0.0 0.2 98.2 100.0 478 45-49 4.6 0.8 0.2 1.9 0.0 0.2 1.3 0.0 0.2 95.4 100.0 477 All Ages 9.1 2.7 0.5 0.9 0.1 1.5 2.8 0.4 0.2 90.9 100.0 4213 CURRENTLY MARRIED WOMEN 15-19 2.8 1.7 0.0 0.0 0.0 0.0 0.9 0.0 0.0 97.4 1OO.0 115 20-24 4.6 0.6 0.3 0.0 0.0 1.2 1.5 0.6 0.3 95.4 100.0 326 25-29 7.3 1.8 0.4 0.6 0.0 0.2 3.5 0.8 0.0 92.7 100.0 508 30-34 7.4 1.7 0.9 I.I 0.0 0.6 2.8 0.0 0.4 92.6 100.0 598 35-39 6.9 1.9 1.5 1.5 0.0 0.4 1.3 0.0 0.0 93.1 100.0 464 40-44 4.8 0.9 0.9 1.8 0.2 0.2 0.7 0.0 0.2 95.2 100.0 456 45-49 4.9 0.9 0.2 2.1 0.0 0.2 1.2 0.0 0.2 95.1 100.0 425 All Ages 6.0 1.4 0.7 1.2 0.0 0.4 1.9 0.2 0.2 94.0 100.0 2832 / 60 50 40 30 20 10 / 0 - - Figure 4.1 Family Planning Knowledge and Use Percent Currently Married Women 15-49 51 Know Know Ever Used Currently Using Method Source F.P. F.P. Ondo State DHS 1986 32 Contraceptive use is lower among married women than all women (6 vs. 9 percent). This is primarily due to the fact that younger married women are less likely to be using contraception than their unmarried counterparts. While levels of use among all women generally fall with age, they first rise and then fall among currently married women. Such a pattern implies that single women may be using contraception to avoid a premarital pregnancy, and that when women marry they want to start their families. It is also likely that a greater proportion of young married women are either pregnant or have recently delivered and thus are not at risk of getting pregnant again. These issues will be investigated further in later sections of this report. The pattern of use by method (Figure 4.2) is similar for married women and all women, except that married women are less likely to rely on the condom. Figure 4.2 Current Use of Family Planning by Method Currently Married Women 15-49 Not Using 94% Periodic Abstinence 2% Pill 1% Injection 1% Other 2% Ondo State DHS 1986 The pattern of contraceptive use by method and by background characteristics is presented in Table 4.7 for married women. The data show that urban women are almost twice as likely (9 percent) to be using a family planning method as rural women (5 percent), and nine times as likely as women in riverine areas (I percent). Contraceptive use is higher among educated than uneducated women; 15 percent of married women with secondary education are using, compared to 6 percent of women with primary education and 3 percent of those with no education (Figure 4.3). Religious afftliation also appears to influence contraceplive use with Protestant women having a higher level of use than women of other religious affiliations. As expected, family planning use increases with the number of living children a woman has, from less than two percent of women with no children, to seven percent of women with four or more children. 33 Table 4.7 Percent Distribution of Currently Married Women by Contraceptive Method Currently Used r According to Background Characteristics, ODHS, 1986 Contraceptive Method Currently Used Not Background Diaphragm, Periodic Cur- NO. Character- Any InJec- Foam, Absti- With- rently of Istlc Method Pill IDD tlon Jelly Condom hence drawal Other Dsing Total Women Residence Urban 8.9 2.2 1.2 1.2 0.I 0.6 2.8 0.3 0.3 91.1 100.0 1144 Rural 4.5 0.9 0.4 1.3 0.0 0.3 1,4 0.i 0.i 95.5 I00.0 1471 Rlverlne 0.9 0.5 0.0 O.0 0,0 0.0 0.5 O.0 0.O 99.1 100.0 217 Education None 2.9 0.6 0.2 1.0 0.0 O.l 1.0 0.0 0.I 97.1 100.0 1454 Primary 5.5 1.2 0.6 1.2 0.0 0.5 1.5 0.2 0.2 94,5 100.0 825 Secondary + 15.0 3.8 2.2 1.6 0.2 i.i 4.9 0.7 0.4 85.0 100.0 553 Rellglon Catholic 4.3 0.7 0.3 1,0 0.0 0,3 2.0 0.8 0.0 95.7 i00.0 304 Protestant 7.0 1.6 0.8 1.3 0.0 0.5 2.1 0.3 0.2 93.0 lOO.0 2038 Muslim 2.8 0.9 0.2 0,7 0.0 O.O 0.9 0.0 0.0 97.2 i00.0 422 Other/None 2.9 0.0 1.5 1.5 0.0 O.O 0.0 0.0 0.0 97.1 100.0 6B Living Children 0 1.7 1.7 0.O 0.0 0.0 0.0 0.0 0.0 0.0 98.3 100.0 178 1 5.5 1.0 0,3 0.5 0.0 0,8 2.3 0.5 0.0 94.5 100.0 383 2 5.7 0.6 0.6 0.0 O.O 0.9 3.0 0.3 0.3 94.3 100.0 335 3 4.8 1.0 0.5 0.5 0.0 0.5 1.3 0.8 0.3 95.2 100.0 399 4+ 7.0 1.7 l.O 1.9 0.1 0.3 1.9 0.0 0.2 93.0 100.0 1535 Total 6.0 1.4 0.7 1.2 0.0 0.4 1.9 0.2 0.2 94.0 100.D 2832 20 15 10 Percent No Education Figure 4.3 Current Use of Family Planning by Education and Residence Currently Married Women 15-49 Primary Secondary and Higher 9 Urban Rural Rlvertne Ondo State DHS 1986 34 Periodic abstinence is the most widely used method among current users of family planning and the second most widely used method among women who have ever used a family planning method. Successful practice of this method depends on knowledge of when, in the monthly ovulatory cycle, a woman is most likely to become pregnant if exposed to sexual intercourse. In the ODHS, all respondents were asked when in the ovulatory cycle a woman is most at risk to becoming pregnant. These data can provide an indication of whether or not women have sufficient knowledge to use periodic abstinence effectively. Table 4.8 presents the distribution of responses for all women and for the small number of women who had ever used periodic abstinence. A broad interpretation of response categories would consider as correct either "in the middle of the cycle" or "right after the period ends." Under this definition, almost 65 percent of all women in Ondo State and about 85 percent of those who have ever used periodic abstinence would be considered sufficiently knowledgeable to practice the method. While it is encouraging that women who have used periodic abstinence are more knowledgeable about the ovulatory cycle'than women in general, it must be noted that one in seven of these women lacks the knowledge to effectively use periodic abstinence and that figure would be considerably higher if the criteria for determining knowledge were stricter and limited to the response "in the middle of the cycle." Table 4.8 Percent Distr ibut ion of All Women and Women Who Have Ever Used Per iodic Abst inence by Knowledge of the Fert i le Per iod Dur ing the Ovulatory Cycle, ODHS, 1986 Ever Users of Fert i le Al l Periodic Per iod Women Abst inence Dur ing Menstrual Period 0.5 0.0 Right After Per iod Ends 43.6 48.6 In Middle of Cycle 20.7 37.3 Just Before Per iod Begins 2.8 1.4 At Any Time ]9.7 10.4 Other 13.6 2.4 Total 100.O 100.0 Number of Women 4213 212 35 4.4 Trends in Fami ly P lann ing Knowledge and Use Table 4.9 presents data on contraceptive knowledge and use from the 1981-82 Nigeria Fertility Survey (NFS) and the ODHS. Although the NFS was a national-level survey, the data presented here are restricted to Ondo State and are from a relatively small sample (389 weighted respondents). Also, in the NFS, respondents who had never menstruated or initiated sexual relations were not asked questions concerning family planning. Thus, caution should be taken in comparing the two surveys. The data indicate that knowledge of modem methods of family planning has increased in Ondo State since 1981-82. Knowledge of traditional methods has either remained the same or decreased. Ever use of modem methods has increased substantially in the five years between surveys, from 3.5 percent in 1981-82 to 10.8 percent in 1986. There was a small decline in ever use of traditonal methods. Since respondents in the ODHS were not asked about prolonged abstinence, the two surveys are not fully comparable. Current use of contraceptives has increased from 1 to 9 percent of women of reproductive age. The increase includes both modem and traditional methods. Tab le 4.9 Knowledge, Ever Use, and Current Use of Specif ic Contracept ive Methods Among All Women 15-49, Niger ian Fert i l i ty Survey (Ondo State), 1981-82 and ODHS t 1986 Percent Percent Percent Knowing: Ever Using: Current ly Using: NFS (OS) ODHS NFS (OS) ODHS NFS (OS) ODHS Method 1981-82 1986 1981-82 1986 1981-82 1986 Any Method 42.7 48.3 23.1 14.9 1.0 9.1 Any Modern Method 32.7 47.1 3.5 10.8 1.0 8.7 Pi l l 18.1 34.0 2.2 6.2 0.7 2.7 IUD 5.4 19.8 0.0 1.1 0.0 8.5 Injectables 17.0 36.0 0.3 2.2 0.3 8.9 Diaphragm, Foam, Jelly 3.2 7.7 0,O 0.4 0.0 0.I Condom 7.5 19.1 0.9 4.4 0.0 1.5 Female Ster i l i zat ion 8.8 21.9 0.0 0.i 0.0 0.O Male Ster i l izat ion 3.7 4.8 0.0 0.0 0.O 0.0 Any Tradit ional Method 9.9 19.6 O.0 3.4 Per iodic Abst inence 16.9 12.4 7.6 5.0 0.0 2.8 Withdrawal ll.O 10.9 4.5 3.1 O.0 0.4 Pro longed Abst inence 23.7 - 15.5 Other 3.8 0.5 0.2 Number of Women 329* 4213 329* 4213 329* 4213 Note : Figures on ever -use f rom the NFS include pro longed abstinence, while f igures on cur rent use exclude it, so as to be comparable to the ODHS. Excludes women who had never menst ruated or had sexual relations, Not ava i lab le Source : National Populat ion Bureau and Federal Min istry of Health, 1986: Tables 15 and 17. 36 4.5 Sources of Contraceptive Methods Information on the sources for contraceptive methods was obtained by asking current users where they had obtained their methods the last time. Overall, the most frequently cited source was government hospitals, which supply 31 percent of users (Table 4.10 and Figure 4.4). With another 6 percent of users relying on government health centers, a total of 37 percent rely on public sources for services. Private sector sources provide services to 38 percent of users, with pharmacies and shops accounting for the largest share (19 percent). Almost one-quarter of users (22 percent)--primarily those relying on periodic abstinence--said that they did not go anywhere to obtain their contraceptive method. The sources relied on by users vary with the method used. Most pill users obtain their supplies either from government hospitals or pharmacies and shops, while the majority of condom users obtain supplies from pharmacies or shops. As expected, most users of the IUD and injection cite govemment hospitals as their source. Table 4.10 Percent Distr ibut ion of Current Contracept ive Users by the Most Recent Source of Supply, According to the Method Used, ODHS, 1986 Supply Methods Cl inic Methods Per iodic Source of Injec- Absti- Supply Pi l l Condom Total* IUD tion Total nence Total* Private Doctor 15.9 ii.i 14.5 5.0 12.5 10.0 24.8 17.1 Government Hospital 37.2 15.9 30.2 90.0 60.0 70.0 13.7 31.5 Govt. Health Center 4.4 3.2 3.9 5.0 25.0 18.3 1.7 5.6 Mobi le Cl inic O.0 0.0 0.0 0.0 0.0 0.0 0.9 0.3 Private Hospital 4.4 3.2 3.9 0.0 2.5 1.7 0.9 2.5 Pharmacy/Shop 29.2 59.0 37.4 0.0 O.0 0.0 0.0 18.8 Field Worker 0.9 6.3 2.8 0.0 0.0 0.0 3.4 2.5 None 8.0 6.3 7.3 0.0 0.0 0.0 54.7 21.6 Total i00.0 100.0 I00.0 i00.0 i00.0 i00.0 i00.0 I00.0 Number of Women 113 63 179 20 40 60 117 356 *Includes three users of vaginal methods (diaphragm, foam, or jelly). 37 Figure 4.4 Source of Family Planning Supply Current Users Government Hospita l 31% ~er 3% Govt. Hea l th Center 6% Private Doctor 17 c, None 22% Pr ivate Hospi ta l 2% Pharmacy/Shop 19% Ondo State DHS 1986 4.6 Attitudes About Pregnancy and Reasons for Nonuse All respondents who were not pregnant, not using family planning, and who reported being sexually active during the month prior to interview, were asked if they would mind if they became pregnant in the next few weeks. Two-thirds of the women said they would not mind, while one-third said they would mind (Table 4.11). There is no definite pattern according to the number of living children a woman has. Table 4.11 Among Women Who Are Not Pregnant, Not Us ing Contra- ception, and Are Sexual ly Active, the Percent Distr ibut ion by Att i tude Toward Becoming Pregnant in the Next Few Weeks, According to Number of Livinq Chi ldren r ODHS~ 1986 Number of Would Number L iv ing Not Would of Chi ldren Mind Mind Total Women 0 65.8 34.2 1O0°0 174 1 74.0 26.0 100.O 73 2 63.2 36.8 I00.8 76 3 77.8 22.2 I00.0 72 4 + 61.9 38.1 i00.0 281 Total 66.1 33.9 10O.0 676 38 Women who stated that they did not want to get pregnant in the next few weeks were asked why they were not using a method to avoid pregnancy. As is shown in Table 4.12, health concerns are the most common mason given for nonuse (27 percent); they are also the most common reason given when women are asked about the main problem with each method (Fable 4.3 above). Lack of knowledge of a method or of a source for obtaining a method (16 percent) is the next most common reason given. Disapproval of either the woman or her parmer was cited in 13 percent of cases. Less than 5 percent of women gave reasons such as cost, religion, inconvenience, infrequent sexual activity, menopause, etc. Altogether, half of the women gave a reason for nonuse which could be addressed by family planning programmes, such as health concerns, lack of knowledge, cost, access, etc. There is little difference in the masons given by women under age 30 and those age 30 and over, except that older women are less likely to say that they do not know a method or a source and more likely to say they or their partners are opposed to family planning. Table 4.12 ~ong Women Who Are Not Pregnant, Not Using Contraception, and Who Would Not Be Happy If They Became Pregnant, the Percent Distr ibut ion By the Main Reason for Nonuse, According to Age, ODHS, 1986 Under 30 Reason for 30 or Non~se Years More Total Health Concerns 27.3 27.0 27.1 Lacks Knowledge or Source 20.5 13.5 16.2 Access/Avai labi l i ty I.i 0.7 0.9 Costs Too Much 2.3 0.7 1.3 Inconvenient to Use 4.5 4.3 4.4 Opposed to Family P lanning 5.7 9.9 8.3 Partner Disapproves 3.4 6.4 5.2 Rel ig ion 4.5 4.3 4.4 Fatal ist ic 0.0 2.1 1.3 Infrequent Sex 4.5 2.1 3.1 Postpartum/Breast feedlng 2.3 0.0 0.9 Menopausal /Subfecund 0.0 5.0 3.1 Other 8.0 13.5 11.4 Don't Know 15.9 10.6 12.7 Total i00.0 i00.0 i00.0 Number 88 141 229 39 4.7 Intention to Use Contraception in the Future Women who were not using a contraceptive method at the time of the ODHS interview were asked if they thought that they would do something to keep from getting pregnant at any time in the future. Those who answered affirmatively, were asked which method they preferred to use and whether they intended to use in the next twelve months or not. Table 4.13 shows that 32 percent of currently married nonusers intend m use at some time in the future, 53 percent do not intend to use, and 15 percent are unsure. Of those who intend to use, the majority said that they thought that they would not use within the next 12 months. The proportion of married nonusers who say they intend to use in the future does not vary much according to the number of children a woman has, however, of those who say they plan to use, the proportion who say they intend to use in the next 12 months increases as the number of children increases. Table 4.13 Percent Distr ibut ion of Current ly Marr ied Nonusers by Intent ion to Use in the Future, Accord ing to Number of L iv ing Children, ODHS, 1986 Number of L iv ing Chi ldren * Intent ion to Use in Future 0 1 2 3 4+ Total Use in Next 12 Months 3.4 3.3 2.8 6.9 I2.2 8.5 Use Later 32.0 25.I 30.4 21.6 16.9 21.3 Use, Unsure When 0.6 0.8 0.9 2.6 3.2 2.3 Unsure if Wil l Use 17.7 18.2 16.8 17.9 13.3 15.3 Does Not Intend to Use 46.3 52.5 49.1 50.9 54.5 52.5 Total i00.0 100.0 i00.0 I00.0 i00.0 i00.0 Number of Women 175 362 316 379 1425 2659 * Includes current pregnancy 40 Injection is the most popular method among nonusers who intend to use a method in the future (39 percent), with the pill and the IUD (13 percent each) next most favored (Table 4.14). Method preference does not vary much by whether respondents plan to use in the next 12 months or later. Table 4.14 For Current ly Marr ied Nonusers Who Intend to Use a Contracept ive Method in the Future, Percent Distr ibut ion by Preferred Method, According to Intention to Use in Next Twelve Months or Later, ODHS, 1986 Use in Next 12 Use Method Months Later Total Pil l 16.3 11.1 12.6 IUD 13.7 12.9 13.1 Injection 35.7 40.9 39.4 Diaphragm, Foam, Jel ly 1.3 0.2 0.5 Condom 2.2 0.5 1.0 Female Ster i l izat ion 1.8 2.8 2.5 Periodic Abst inence 4.8 1.1 2.1 Withdrawal 0.4 0.2 0.3 Other 7.5 9.3 8.8 Not Sure 16.3 21.0 19.6 Total i00.0 100.0 100.0 Number of Women 227 567 794 4.8 Approva l o f Fami ly P lann ing Respondents in the ODHS were asked a number of questions which were intended to measure their level of approval of family planning. All women were asked if they thought it was acceptable for family planning information to be provided on radio or television. They were also asked if they approved of couples using a method to avoid a pregnancy. Finally, currently married women were asked if they thought that their husbands/parmers approved of couples using family planning and how often they had talked to their husbands about family planning in the past year. Table 4.15 shows that 85 percent of women in Ondo State find family planning messages on radio and television to be acceptable. Only 8 percent of women think such messages are unacceptable, and 7 percent do not have an opinion. Acceptance does not vary appreciably by age of woman or by urban-rural residence. Women in riverine.areas, women with no education, and Catholic women tend to be somewhat less accepting of family planning messages than other women. 41 Table 4.16 presents data on wives' and husbands' attitudes about use of family planning. Overall, three-quarters of married women who have heard of a contraceptive method approve of couples using family planning; one-quarter do not approve. Just over 40 percent of these women say that their husband approves of family planning use, another 18 percent say their husband disapproves and 40 percent say they don't know their husband's attitude. According to the wives, only 41 percent of couples jointly approve of family planning, while 8 percent jointly disapprove. Table 4.15 Percentage of All Women Who Bel ieve That it is Acceptable to Have Family P lanning Messages on the Radio or Television, by Background Characterist ics, ODHS, 1986 Background Background .Characterist ic Percentage Character ist ic Percentage Age Education 15-19 84.7 None 80.1 20-24 87.4 Pr imary 84.1 25-29 85.9 Secondary + 90.6 30-34 82.1 35-39 86.0 Rel ig ion 40-44 84.5 Cathol ic 79.6 45-49 86.4 Protestant 86.9 Musl im 82.6 Residence Other/None 66.3 Urban Rural R iver ine 86.5 86.4 69.9 Total 85.2 Table 4.16 Percentage of Current ly Marr ied Women Know- ing at Least One Contracept ive Method, by Husband's and Wife's Att i tudes Towards Family P lanning Use, ODHS, 1986 Husband's Att i tude Wife 's Doesn't Att i tude Approves Disapproves know Total Approves 40.6 9.9 26.4 76.9 Disapproves 1.0 8.3 13.8 23.1 Total 41.6 18.2 40.2 I00.0 42 Table 4.17 shows the percent of married women knowing a family planning method who think that their husband approves of family planning according to background characteristics. Differences in approval by urban-rural residence, religion and age am minor; however, better-educated women are more likely to say that their husband approves. Table 4.18 presents data on the proportion of married women who have discussed family planning with their husband in the year prior to the survey. Overall, two out of five women have discussed family planning with their husband at least once. Better-educated women are more likely to discuss family planning than less educated women. Table 4.17 Percentage of Currently Married Women Knowing at Least One Contraceptive Method Who Think That Their Husband Approves of Family Planning Use, by Background Characterist ics, ODHS, 1986 Background Background Character ist ic Percentage Character ist ic Percentage Age Education 15-19 36.2 None 29.0 20-24 45.1 Pr imary 40.2 25-29 48.3 Secondary + 57.2 30-34 38.0 35-39 41.4 Rel ig ion 4G-44 40.5 Cathol ic 42.6 45-49 34.8 Protestant 43.1 Musl im 34.1 Residence Other/None (Ii.I) Urban 44.9 Rural 37.9 Rivor ine 44.8 Total 41.5 Note: Numbers in parentheses represent fewer than 20 cases. 43 Table 4.18 Percentage of Current ly Marr ied Women Knowing at Least One Contra- ceptive Method, Who Have Discussed Family P lanning With Their Husband at Least Once in the Past Year, by Woman's Age and Back- ground Characterist ics, ODHS, 1986 Age of Woman Background Character is t ic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 20.8 41.5 48.8 40.9 46.2 41.2 52.8 43.9 Rural (26.3) 44.0 43.9 28.5 47.2 41.6 34.0 39.4 River ine (75.0) (0.0) (0.0) (71.4) (20.0) (60.0) (50.0) 44.8 Educat ion None (0.0) (23.4) 24.5 25.5 36.5 37.9 37.4 32.9 Pr imary (16.7) 35.8 42.3 38.5 48.5 44.8 40.0 41.4 secondary + 35.5 46.7 55.9 49.3 63.6 50.0 (69.2) 52.0 Rel ig ion Cathol ic (33.3) 37.5 48.1 30.3 (55.6) 50.0 35.0 41.2 Protestant 28.6 43.4 46.5 40.1 45.5 42.2 38.9 42.7 Musl im (16.7) (36.8) 42.9 25.6 47.5 30.4 46.2 37.3 Other/None (0 .0 ) (0.0) (0.0) (50.0) (75.0) (33.3) Total 27.7 41.6 46.3 36.4 46.2 41.9 40.5 41.7 Note: Numbers in parentheses represent fewer than 20 cases; dashes (-) indicate no cases, 44 5. FERTIL ITY PREFERENCES 5.1 Future Fertility Preferences In the ODHS interview, all currently married women were asked whether they wanted to have another child, and, if so, how long they wanted to wait before having the next child. Married women who were pregnant at the time of the interview were asked if they wanted a another child after the one they were expecting. The purpose of these questions was to investigate the fertility desires of women of childbearing age, with a view to assessing the need for family planning services. Women who want to cease childbearing or postpone their next child can be thought of as being in need of contraception. Table 5.1 and Figure 5.1 show the percent distribution of married women by whether they want another child and, if so, how long they want to wait to have their next child. Almost one-quarter (23 percent) of married women in Ondo State do not want to have any more children, 11 percent are undecided about having another child, and 65 percent want another child. Among those wanting another child, more than half want to wait two or more years before having the child. Table 5.1 Percent Distr ibut ion of Current ly Marr ied Women 15-49 by Desire for More Chi ldren According to Number of L iving Children, ODHS, 1986 Number of L iving Chi ldren * Desire For More Chi ldren O 1 2 3 4 5 6+ Total Want No More 1.3 1.7 Want Another Within 2 Yrs. 73.4 25.7 Want Another After 2+ Yrs. 13.9 88.3 Want Another, Don't Know When i0.i 12.6 Undecided if Want Another 1.3 1.7 5.3 11.3 22.3 34.0 49.2 23.3 31.8 20.2 20.9 11.7 8.7 19.8 51.6 49.0 37.8 30.4 20.2 38.0 9.5 9.9 7.6 3.8 3.6 7.4 1.8 9.6 11.6 20.2 18.4 11.4 Total i00.0 100.0 I00.0 100.0 100.0 i00.0 i00.0 100.0 No. of Women 79 420 337 406 448 488 665 2832 * Includes current pregnancy 45 Figure 5.1 Fertility Preferences Currently Married Women 15-49 Want to Space (2 or more yre) 38 knt NO More 23% Undecided if Went More 12% Went Soon (within 2 yrs) 20% ent Another, 1decided When 7% Ondo State DHS 1986 100% Figure 5,2 Fertility Preferences by Number of Living Children Currently Married Women 15-49 80% 60% 40% 20% 0%0 5 2 3 4 Number of Living Children 6* m Went No More Want Soon Went to Speoe Undeolded Ondo State DHS 1986 46 The fact. that 65 percent of women want another child reflects the high value placed on children in Ondo State; however, the fact that 61 percent of women either want no more children or want to wait at least two years before having another child, suggests a need for family planning services. If those who are undecided about either whether or when to have another child are included, the proportion of married women who are potential candidates for family planning services rises to 80 percent. Since future fertility preferences are significantly influenced by the number of children a woman already has, Table 5.1 and Figure 5.2 show the data on fertility preferences by the number of living children women have, including the current pregnancy for all pregnant women. The proportion of women wanting no more children is only 5 percent or less for women with less than three children, but it increases steeply to almost 50 percent of women with six or more children. The proportion of women who are undecided about having another child is also quite high for women with five or more children (about 20 percent). Equally revealing is the high proportion of women who want to space their next child. Over half of women with 1-3 children want to wait at least two years before having their next child. These women constitute a sizable group who are potentially in need of family planning services for spacing purposes. Fertility preferences are shown by age group of women in Table 5.2. Since age and number of children are positively correlated, it is not surprising that similar trends are evident in Tables 5.2 and 5.3. As is true of women with 1-3 children, the proportion of women aged 15-19, 20-24, and 25-29 who want to wait two or more years before having their next child is high (60 percent), again suggesting a need for family planning services for spacing purposes among younger married women. The desire to stop childbearing rises with age, from less than one percent of women under age 25 to 72 percent of women aged 45-49. Table 5.2 Percent Distr ibut ion of Current ly Marr ied Women by Desire for More Children, According to Age, ODHS, 1986 Current Age Desire For More Chi ldren 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Want No More 0.0 0.9 2.4 9.1 19.8 44.1 71.5 23.3 Want Another Within 2 Yrs. 21.7 24.2 23.2 23.2 20.9 17.1 9.4 19.B Want Another After 2+ Yrs. 61.7 59.2 59.3 47.0 34.7 14.7 7.3 38.0 Want Another, Don't Know When 13.9 15.3 9.6 8.0 5.2 4.4 1.6 7.4 Undecided if Want Another 2.6 0.3 5.5 12.6 19.4 19.7 10.1 11.4 Total I00.0 i00.0 I00.0 I00.0 I00.0 100.0 I00.0 I00.0 No. of Women 115 326 508 538 464 456 425 2832 47 In order to examine fertility preferences by background characteristics, a single index--the percentage of women who want no more children--is presented in Table 5.3 by number of living children. The proportion of women who want no more children varies considerably by area of residence, with relatively lower proportions of women in riverine areas wanting to stop childbearing. This is as expected, and probably reflects the greater isolation and more traditional thinking of these women with respect to childbearing. However, the fact that more rural women want no more children than urban women (26 vs. 22 percent) is surprising. This is due to the fact that women in rural areas generally have more children than urban women; as shown in Table 5.3, there is little difference between the proportions of urban and rural women wanting no more children when the number of living children is taken into account. Table 5.3 Percentage of Current ly Marr ied Women 15-49 Who Want No More Chi ldren by Number of Living Chi ldren and Background Character - is t ics , ODHS, 1986 Number of Living Chi ldren * Background Character - istic 0 1 2 3 4 5 6+ Total Residence Urban 2,4 1.0 4.6 12.6 20.8 35.1 50.9 21.5 Rural 0.0 2.7 6.3 11.2 25.4 34.3 51.9 26.3 River ine (0.0) 0.0 4.0 0.0 3.7 22.2 29.0 12.9 Educat ion None 3.2 4.1 12.0 19.5 27.7 41.3 50.4 32.3 Pr imary 0.0 0.9 2,9 3.9 9.5 20.0 40.7 15.0 Secondary + 0.0 0.5 0,0 3.7 33.3 31.0 67.3 12.3 Rel ig ion Cathol ic 0.0 2.0 5.7 10.3 20.0 26.7 37.7 19.1 Protestant 1.7 1.7 4.5 11.9 24.3 36.1 52.2 25.0 Musl im 0.0 1.7 9,4 9.0 12.7 27.9 42.2 18.5 Other/None (0.01 (0.0) (0.0) (18.2) (25.0) (50.0) (43.8} 23.5 Total 1.3 1,7 5.3 11.3 22.3 34.0 49.2 23.3 Note: Numbers in parentheses are based on fewer than 20 cases . * Includes current pregnancy The characteristic with the greatest influence on fertility preferences is educational level. Surprisingly, married women with no education are generally more likely to want to stop childbearing than women with primary or secondary education, regardless of the number of children they already have. Among women with fewer than 4 children, those with primary education are more likely to want to stop childbearing than those with secondary education, however, among women with 4 or more children, a greater proportion of those with secondary education want no more children than those with primary education only. As for differences in fertility preferences by religion, Protestant women are generally more likely to want to stop childbearing than either Catholic or Muslim women, regardless of the number of children they already have. 48 Table 5.4 looks at the proportion of married women who are in need of family planning services, that is, they either want no more children or they want to delay having their next child for at least two years, and they are not using family planning. Because contraceptive use is so low in Ondo State, the figures given in Table 5.4 are almost identical to those presented in previous tables for the proportions of married women wanting no more children or wanting to postpone their next child. In other words, almost all the women who want to limit or delay childbearing are in need of family planning services, because such a small proportion of them are using contraception. Of course, some of them may not be in immediate need of family planning services because some of them may be pregnant, amenorrheic, or not sexually active. The table shows that 75 percent of married women in Ondo State are in need of family planning services, roughly one-third because they want no more children, and two-thirds because they want to space their next birth. Not only do these figures represent a high overall level of need for services, but it is also noteworthy that the large majority are in need of methods for spacing births as opposed to limiting the number of births. Need is greatest among women with no education, and least among women with secondary or more education. Table 5.4 Among Current ly Marr ied Women, the Percent Who Are In Need of Family Planning, and the Percent Who Are In Need and Plan to Use a Contra- ceptive Method in the Future, by Background Characterist ics, ODHS 1986 In Need and Intend In Need of Family Planning I To Use Contracept ion Total in Background Want No Want to Total Want No Want to Need and Character- More Postpone/ In More Postpone/ Intending istic Chi ldren Undecided 2 Need Chi ldren Undecided' To Use Residence Urban 18.1 54.3 72.4 5.2 21.1 26.3 Rural 24.6 52.1 76.8 5.5 17.1 22.6 River ine 12.9 63.6 76.5 2.8 8,3 ii.I Education None 30.7 48.8 79.4 5.0 11.7 16.7 Pr imary 13.2 60.2 73.5 6.7 21.6 28.2 Secondary + 7.6 57.9 65.5 3.4 29.3 32.7 Rel ig ion Cathol ic 17.8 58.9 76.6 3.9 17.4 21.4 Protestant 22.4 52.4 74.8 5.5 18.5 24.0 Musl im 17.3 58.1 75.4 4.7 18.2 23.0 Other/None 20.6 50.0 70.6 2.9 4.4 7.4 Total 21,1 53.9 75.0 5.2 18.0 23.2 Women in need are def ined as those who are not current ly us ing family planning and who want no more births or want to postpone the next b i r th for at least two or more years. Includes women undecided about whether to have another b i r th or about t iming for the next birth. 49 Table 5.4 also indicates that less than one-third of women in need of family planning services intend to use a method sometime in the future (23 percent out of 75 percent). A somewhat larger proportion of women who are in need because they want to postpone their next birth intend to use in the future (18 out of 54 percent) than of those in need because they want to stop childbearing altogether (5 out of 21 percent). 5.2 Ideal Number o f Ch i ld ren In order to assess fertility preferences in Ondo State, all ODHS respondents were asked: "If you could (go back to the time you did not have any children, and could) choose exactly the number of children to have in your whole life, how many would that be?" Women with children were asked the entire question, while those with no children were asked the question excluding the phrase in parentheses. The question measures two things--first, among women who have just started childbearing, the data give an idea of the total number of children these women hope to have in the future; secondl, among older women with more children, the data provide an idea of the level of unwanted fertility. It is important to note that some women have difficulty in answering such a hypothetical question, presumably more so in cultures in which control over fertility is a new concept. The fact that two-fifths of women in Ondo State gave a non-numerical response ("As many as God gives me", "Don't know", etc.) is evidence of this difficulty. Also, it is usually assumed with this question that some women report their actual number of children as their ideal number, since they find it difficult to admit that they would not want some of their children if they could choose again. Tab le 5.5 Percent Distr ibution of All Women by Ideal Number of Children and Mean Ideal Number of Children for All Women and Currently Married Women, According to Number Living Children, ODHS, 1986 Number of Living Children * Ideal Number of All Chi ldren 0 1 2 3 4 5 6+ Women 0 0 .0 0 ,0 0 .0 0 .0 0 .0 0 .0 0 ,0 0 .0 1 0 .0 0 ,0 0 .0 0 .0 0 .0 0 .0 0 .0 0 .0 2 0.4 0,9 0.6 0.0 0.4 0,4 0.4 0.4 3 4.7 2.4 2.3 1.7 0.6 1 .2 0.9 2.4 4 22.2 17.3 11.2 9.7 7.9 3.6 4.8 12.8 5 22.2 19.7 20.6 15.2 ii.4 12.2 5.5 16.0 6+ 23.4 25.2 20.6 29.5 32.4 33.6 41.2 29.1 Non-Numerlc Rsp. 27.2 34.6 44.7 43.9 47.2 49.0 47.3 39.2 Tota l Percent i00.0 I00.0 I00.0 i00.0 I00.0 I00.0 100.0 i00.0 No. of Women 1326 437 349 421 466 500 694 4213 Mean Ideal No., All Women 5.2 5.4 5.3 5.9 6.1 6.3 7.0 5.7 Mean Ideal No., Marr ied Women 5.9 5.4 5.3 5.9 6.1 6.3 7.1 6.1 * Includes cur rent p regnancy 50 The distribution of women by ideal and actual number of children is presented in Table 5.5. Aside from the 39 percent of respondents who did not give a numerical response, the bulk of women said they would want to have 6 or more children (29 percent). Almost no one stated an ideal of fewer than 4 children. The mean ideal family size is 5.7 children among all women and 6.1 children for married women. Women who already have several children state higher ideal family sizes than women with fewer children--in fact, the mean ideal number of children rises from 5.3 among women with 2 children to 7.0 among women with 6 or more children. This may be due either to the fact that women who want more children actually end up having them, or to the phenomenon mentioned above, that women rationalize the number of children that they already have. Despite the generally pronatalist attitude of women in Ondo State, there is some evidence of unwanted fertility in that 12 percent of women with 6 or more children report lower ideal numbers of children. Table 5.6 indicates that there is variation in mean ideal numbers of children by age and background characteristics of women. The most outstanding difference is by area of residence; while both urban and rural women report an ideal family size of about six children, women in riverine areas favor having well over 7 children on average. This pronatalist attitude of riverine women holds across all age groups. Table 5.6 Mean Ideal Number of Chi ldren for All Women by Age and Background Characterist ics, ODHS, 1986 Age Background Character- All istic 15-19 26-24 25-29 30-34 35-39 40-44 45~49 Ages Residence Urban Rural River ine Education None Primary Secondary + Rel igion 5.3 4.9 5.3 6.0 6.5 6.1 6.2 5.6 5.1 5.2 5.3 6.5 6.3 6.4 6.6 5.7 7.4 7.1 7.0 7.2 8.4 7.8 7.6 7.5 6.? 5.4 6.5 6.6 6.8 6.5 6,6 6.6 6,0 5.6 5.5 6.3 6,5 6.4 6.8 6.1 5.2 5.0 4.8 5.4 5.6 5.3 (5,7) 5.1 Cathol ic 5.5 5.2 5.5 6.5 6.1 6.3 7.8 5.9 Protestant 5.3 5.1 5.2 6.1 6.6 6.3 6.4 8.7 Musl im 5.7 5.3 6.1 6.6 6.3 6.7 6,6 6.0 Other/None (7.8) (8.0) (5.0) (7.3) (7.3) (7.0) (7.4) 7.2 Total 5.4 5.2 5.3 6.3 6.5 6.3 6.6 5.7 Note: Numbers in parentheses represent fewer than 20 c~ses. Women with secondary education prefer smaller families than less educated women, and women with primary education report higher ideal family sizes than women with no education in most age groups. With regard to religion, Muslim women in almost every age group desire larger families than Catholic or Protestant women. 51 6. MORTALITY AND HEALTH 6.1 Mortality In the ODHS, mortality data were collected primarily for the purpose of estimating infant and child mortality rates. In this section mortality rates are calculated, using direct estimation procedures, for: Infant mortality, the probability of dying between birth and exact age one; Child mortality, the probability of dying between age one and exact age five; Under five mortality, the probability of dying between birth and exact age five. Rates are calculated on a period basis (i.e., utilizing infonnation on deaths and exposure to mortality during a specific time period) rather than on a birth cohort basis. A complete description of the methodology for computing period-specific mortality probabilities is given elsewhere (Rutstein, 1984). Birth History Survivorship Data The data for the estimation of mortality rates were collected in the reproduction section of the individual woman's questionnaire. The data were obtained in the form of a truncated birth history in which questions were asked about the sex, date of birth, survivorship status and, if appropriate, age at death of the respondent's live births. The truncated birth history collected information on all births which occurred to respondents during the time period 1981-86. As a result of this procedure, the observed person-years of exposure to mortality are less for the older childhood ages (ages 3 and 4) than for the younger childhood ages (ages 1 and 2). The decline in the number of persons exposed to mortality should not substantially increase the sampling variance of the estimated child mortality rates because older children contribute relatively little to the overall child mortality rate. Nevertheless, in the tables of this chapter, any reported mortality rate which is based on fewer than 500 person-years of exposure is enclosed in parentheses. Data Quality The truncated birth history is susceptible to the same types of data collection errors as are other retrospective procedures; namely, underreporting of events, misreporting of age at death, and misreporting of date of birth. Event underreporting and age at death misreporting are the more serious sources of error for mortality estimation. The ODHS data were investigated with respect to these two sources of error by testing their internal consistency. However, it should be stated that the power of intemal consistency checks for detecting error is quite limited so that, while they can detect gross defects, they cannot detect less serious data problems and cannot defmitively establish the accuracy of the data collected. Underreporting of deaths is most likely in the case of babies who die in early infancy. In the ODHS, age at death was recorded in one of three units: days, for deaths in the first month of life; months, for deaths under two years of age; and years, for deaths at age two and above. A test to detect underreporting of early infant deaths was made by forming the ratio of deaths under seven days to all deaths in the first month of life. Since mortality is known to decline steeply with age throughout early infancy, the value of this ratio should exceed 0.25. For the period 1981-86, the values of this ratio from 53 the ODHS are well in excess of 0.25, indicating that gross underreporting of babies who died in early infancy is not a problem: Both Males Females Sexes 0.73 0.64 .70 Age at death misreporting can result in a net transfer of events between infancy and early childhood and can bias mortality estimates. The possibility of such misreporting was investigated by looking for heaping of deaths at 12 months in the distribution of deaths by age. The dis~bution is as follows: Deaths bv Age During 1981-86 Age in Both Months Males Females Sexes 6 10 10 20 7 4 3 7 8 5 4 9 9 4 9 13 10 1 5 6 11 3 4 7 12 6 5 11 13 3 1 4 14 1 1 2 15 4 2 6 16 2 0 2 17 0 1 1 18 8 9 17 There is some indication of heaping of deaths at 12 months of age 7 at 11 months and 4 at 13 months) but it is not significant relative to reported for the 1981-86 period. (11 deaths at 12 months versus the total of 214 infant deaths Mortality Levels 1981-1986 Table 6.1 displays infant and child mortality rates for Ondo State for the period 1981-1986. For Ondo State, the infant mortality rate is 56 deaths per 1,000 live births and the child mortality rate is about the same, 55 per 1,000. The overall probability of dying between birth and age five is 108 per 1,000 (i.e., about one in every ten children dies before reaching five years of age). Sex-specific rates are similar in magnitude with male rates being higher than female rates -- a differential found in most populations. The rates by area of residence indicate somewhat lower infant and child mortality in urban areas (54 and 49 per 1,000, respectively) than in rural areas (57 and 61 per 1,000, respectively). These differences are not great and may be due to sampling variance. The rates for riverine areas are, on the other hand, decidedly higher for infants (70 per 1,1300) hut lower for children age 1-4 (38 per 1,000). However, these rates are based on fewer than 500 person-years of exposure and should be viewed with caution. 54 Table 6.1 Infant and Chi ld Mortal i ty Est imatss 1981-1986, ODHS, 1986 Infant Chi ld Under F ive Background Mortal i ty Mortal i ty Morta l i ty Character ist ic (lq0) (4ql) (Sq0) Sex of Chi ld Male 59 58 114 Female 53 51 I01 Residence Urban 54 49 100 Rural 57 61 115 River ine (70) (38) (105) Total 56 55 108 Note: Rates presented include deaths and exposure for 1986 through ths calendar month preceding the month of interview. Note: Rates in parenthesis are based on fewer than 500 person-years of exposure. Mortality Differentials 1981-1986 Mortality differentials by education, mother's age at birth, birth order, and previous birth interval are presented in Table 6.2. The rates by education indicate lower infant mortality for women with no education (54 per 1,000) than for women with a primary education (64 per 1,000), although the difference is not great and could reflect sampling variance rather than a true differential. The child mortality rates are the same for women with no education and with a primary education (57 per 1,000). On the other hand, for women with a secondary or higher education, the infant and child mortality rates are decidedly lower (40 and 15 per 1,000, respectively). Differentials also exist in the mortality rates by demographic characteristics of the mother. With respect to age, children born to women under age 20 are more likely to die in infancy (93 per 1,000) and early childhood (74 per 1,000) than children bom to women 20 to 34 (about 51 per 1,000 for both infants and children age 1-4). Elevated rates are also apparent for births to women age 35 and over (61 per 1,000 for infants and 58 per 1,000 for children age 1-4). In terms of birth order, infant mortality rates are elevated for first births (78 per 1,000) and births of order 7+ (63 per 1,000) relative to births of orders 2 through 6 (about 50 per 1,000). With respect to the preceding birth interval, substantial infant mortality differentials exist between births occurring within two years of a previous birth (80 per 1,0(30) and births occurring after an interval of two years or more (46 per 1,000). 55 Table 6.2 Infant and Chi ld Mortal i ty 1981-1986, by Back- ground Characterist ics, ODHS, 1986 Infant Chi ld Under Five Background Mortal i ty Morta l i ty Mortal i ty Characteristic (lqO) (4ql) (SqO) Education None 54 57 109 Pr imary 64 57 117 Secondary + (48) (15) (54) Mother 's Age at Birth Less than 20 93 74 160 20-29 51 53 102 30-34 51 48 97 35+ 61 58 115 Birth Order I 78 (54) 128 2-3 48 47 93 4-6 51 59 107 7+ 63 (56) (116) Birth Interval* Less than 2 yrs, 80 (35) (112) 2-3 years 47 60 205 4 or more years 45 (68) (110) Total 56 55 108 Note : Rates presented include deaths and exposure for 1986 through the calendar month preceding the month of interview. Note : Rates in parenthesis are based on fewer than 500 person-years of exposure. * Based On births of order two and higher. Another perspective on infant and child mortality can be obtained by calculating statistics on the proportion dead of children ever born. Overall, the proportion dead of children ever born to women 15- 49 is .20 (Table 6.3). In other words, one in five children born to women 15-49 have died. As expected, this proportion varies considerably by age of women. Fewer than one in 12 children born to women 15- 19 have died while women 45-49 have lost over one-quarter of their children. The higher proportion dead of children ever born among older women reflects the fact that their children were bom longer ago and have been exposed longer to the risk of mortality. 56 Table 6.3 Mean Number of Chi ldren Ever Born, Surviving, and Dead, and Proport ion of Chi ldren Dead Among Those Ever Born, by Age of Women, ODHS, 1986 Mean Number of Children: Pro- Number Age of Ever Sur- port ion of Women Born viving Dead Dead Women 15-19 0.08 0.07 0.01 0.080 1109 20-24 0.82 0.71 0.I0 0.124 563 25-29 2.65 2.30 0.34 0.129 560 30-34 4.65 3.84 0.80 0.173 548 35-39 5.86 4.78 1.08 0.184 478 40-44 6.53 5.06 1.47 0.225 478 45-49 7.24 5.26 1.98 0.274 477 Total 3.31 2.63 0.68 0.205 4213 6.2 Maternity Care In order to obtain information on the type of prenatal care received by Ondo State women, respondents were asked if they saw anyone for a checkup during each pregnancy (resulting in a live birth) during the five years preceding the survey. Respondents were also asked if anyone assisted with the delivery of each birth. If a prenatal checkup or assistance at delivery was received, they were asked who rendered the care. In cases where maternity care was rendered by more than one individual, the most qualified of the providers was recorded by the interviewer. Nevertheless, it is pertinent to remark that in Ondo State, relatives who render maternity services to their relations might not differ appreciably from traditional birth attendants (TBAs); relatives are unlikely to give birthing assistance unless they have received either '~formal" or "informal" training in such. Neonatal tetanus has been identified as a major cause of infant deaths in developing countries. In the ODHS, mothers were asked if they received an injection during pregnancy to prevent the baby from getting tetanus ("jerking"). The responses rely on the mother's recall of events during the pregnancy and on her ability to distinguish a tetanus toxoid vaccination from other injections she might have received. Still, the proportion of women receiving a tetanus toxoid vaccination during pregnancy provides a useful measure for assessing the coverage of a community's maternal and child health services. It also provides a yardstick for estimating the number of pregnant women who attend maternity health centres and for whom health education in other facets of primary health care could be delivered. 57 Table 6.4 shows the percent distribution of births in the last five years by type of prenatal care received by the mother and the percentage of births whose mothers received tetanus toxoid injections during pregnancy. The figures indicate that prenatal care is widespread in Ondo State. For 80 percent of births the mother received prenatal care from a doctor, nurse or midwife; only 15 percent of mothers did not receive a prenatal checkup. For seventy-one percent of births, the mothers were vaccinated against tetanus. However, there is one segment of the population which receives relatively little prenatal health care. In the riverine areas, for 70 percent of births, the mothers did not receive a prenatal checkup and for only 13 percent of births were they vaccinated against tetanus during pregnancy. These data point to the need for improved health services for the riverine people of the State. Table 6.4 Percent Distribution of Births in the Last Five Yeara by Type of Erenatal Care for the Mother and Percentage of Births Before Which the Mother Received a Tetanus Toxo~d Injection, by Background Characteristics, ODHS, 1986 Type of Prenatal Care PerceRt Nurse Birth Receiving Background or Attend- Rela- No Total Tetanus NO. of Characteristic Doctor Midwife ant tlve Other One Missing Percent Injection Births Age <30 13.7 67.6 1.9 1.3 1.6 13.5 0.3 100.0 72.7 1230 30 + 11.9 67.2 1.3 1.6 2.5 15.3 0.2 i00.0 70.3 2019 Residence Urban 16.5 72.4 1.2 0.3 2.5 6.7 0.4 100.0 80.4 1360 Rural 10.8 71.6 0.9 2.0 2.2 12.5 0.1 100.0 72.6 1635 Riverlne 3.9 13.4 7.5 4.7 0.4 70.1 0.0 100.0 I3.0 254 Education None 8.1 64.4 2.3 2.3 2.6 20.0 0.3 100.0 61.9 1473 Primary 13.6 69.0 0.7 I.i 2.4 13.1 0.2 i00.0 73.8 1105 Secondary + 21.0 71.2 1.0 0.3 1.0 5.1 0.3 100.0 87.6 671 Religion Catholic 11.5 72.2 1.3 i.i 0.3 13.1 0.5 100.0 76.7 374 Protestant I2.4 66.5 1.4 1.6 2.9 15.0 0.2 100.0 70.9 2292 Muslim 15.4 69.5 1.8 1.2 0.6 11.2 0.4 100.0 70.5 508 Other/None 6.7 54.7 4.0 1.3 0.0 33.3 0.0 100.0 57.3 75 Total 12.6 67.4 1.5 1.5 2.2 14.6 0.3 I00.0 71.2 3249 While for most births, the women of Ondo State received assistance at delivery from either a doctor (3 percent), nurse or midwife (56 percen0, birth attendant or relative (29 percent), there are noticeable differences between population subgroups with respect to the type of attendant at delivery (Fable 6.5). Older women, women in the riverine areas, women with no education and women in the other or no religion category are less likely to be assisted at delivery by either a doctor, nurse or a midwife. 58 Table 6.5 • Percent Distr ibut ion of Births in the Last Five Years by Type of Assistance at Delivery, by Background Characterist ics, ODHS, 1986 Type of Assistance at Del ivery Nurse Birth Background or Attend- Rela- No Miss- Total No. of Character ist ic Doctor Midwife ant tire Other One ing Percent Births Age <30 3.8 57.1 3.9 23.4 6.5 5.2 0.I 100.0 1230 30 + 2.6 54.7 2.9 27.3 5.9 6.6 0.0 100.0 2019 Residence Urban 3.5 64.6 2.9 18.2 8.8 2.1 0.1 100.0 1360 Rural 3.1 55.5 2.6 27.1 4.7 7.0 0.0 100.0 1635 River ine 0.4 8.7 9.8 58.7 1.6 20.9 0.0 100.0 254 Educat ion None 1.7 44.7 4.4 35.4 5.3 8.5 0.0 100.0 1473 Pr imary 3.6 57.5 2.5 23.2 8.0 5.2 0.I 100.0 1105 Secondary + 5.0 76.5 1.9 9.2 5.1 2.2 0.0 100.0 671 Rel igion Cathol ic 2.4 62.8 4.0 24.9 2.4 3.5 0.0 100.0 374 Protestant 2.7 56.5 2.9 24.2 7.4 6.2 0.0 100.0 2292 Musl im 5.5 49.4 3.7 30.3 4.1 6.9 0.0 100.0 508 Other/None 8.0 33.3 9.3 49.3 0.0 8.0 0.O i00.0 75 Total 3.1 55.6 3.3 25.8 6.2 6.1 0.0 100.0 3249 59 6.3 Child Health Indicators The ODHS Survey obtained information on immunisation coverage and the incidence and treatment of diarrhoea, fever and respiratory ailments among children under five. Data collection was limited to the children of women interviewed in the survey. Therefore, information is unavailable for children whose mothers were dead, living out of the state, institutionalized, or who, for any other reason, were not interviewed in the survey. Although the immunisation status and the morbidity experiences of children excluded from the survey probably differs from that of children whose mothers were interviewed, the number of such children is small. Thus, the results presented here are considered to describe the health status of children under five in Ondo State. Immunisation of Children In the survey, women who had children under five years of age were asked if these children had health cards. If the answer was affirmative, the date of each immunisation was copied from the card onto the questionnaire by the interviewer. The immunisation data were collected for tuberculosis (BCG), diphtheria, whooping cough (pertussis) and tetanus (DPT), poliomyelitis and measles. If a child had no card, or the interviewer could not examine the card, the mother was asked if the child had ever received a vaccination. For these children, information about specific immunisations was not obtained. Table 6.'6 indicates that health cards were seen for only about 25 percent of children under the age of five. All children with cards had received at least one vaccination. An additional 46 percent of children did not have a card available but were reported by their mothers to have been immunlsed. Thus, about 70 percent of children under the age of five in Ondo State may be presumed to have received at least one immunisation. Information on specific immunisations is shown in Table 6.6. The World Health Organisation's (WHO) recommended schedule for childhood immunisation is given below (Sherris et al., 1986). According to this schedule children should be fully immunised by one year of age. Age Immunisations Birth BCG 6 weeks DPT, Polio 10 weeks DPT, Polio 14 weeks DPT, Polio 9 months Measles The ODHS indicates that among children age 12-23 months with immunisation cards (an age group that according to WHO standards should be fully immunised), almost 100 percent had received a BCG vaccination and the first dose of DPT and polio vaccine. About 75 percent had been vaccinated against measles. The proportions decrease between the first and third dose for both DPT and polio. Of children 12-23 months with immunisation cards, 67 percent were fully immunised (i.e., had received BCG, three doses each of DPT and polio, and measles). 60 For the investigation of differentials in the proportion of children who were fully immunised, children aged 12-59 months were selected as the base, since children under one year had not had the oppommity to be fully vaccinated. As seen in Table 6.6, both residence and educational status of the mother are associated with immunisation coverage. The highest rates of full immunisation are reported for children of urban women (67 percent) and children of women with secondary or more education (77 percent). Catholic mothers are somewhat more likely to possess health cards for their children while women in the other or no religion category are least likely to possess health cards. Table 6.6 Among All Chi ldren Under Five, the Percentage With Health Cards, the Percentage Who Are Recorded as I,ununised on the Health Card or Who Are Reported by the Mother as Having Been In~nunlsed, and ~ong Chi ldren With Health Cards, the Percentage for Whom BCG, DPT, Pol io and Measles I r~unlsat lons Are Recorded on the Health Card t Accord ing to Background Character - istics, ODN8, 1986 Percent of Children: Percent of Chi ldren With Health Cards Who Have Received: Immunlsed Immunlsed DPT Pol io No. Background With as Recorded as Ful ly of Character- Health on Health Reported Mea- Im/nu- Chi l - ist lc Cards Cards by Mother BCG 1 2 3 1 2 3 sles nlsed dren ALL CHILDREN UNDER 5 Age in Months < 6 38.3 38.3 19.0 99.1 77.1 36.7 9.2 77.1 36.7 9.2 0.9 0.0 285 6 - 11 39.8 39.8 31.4 99.5 98.5 84.7 61.3 98.5 84.7 62.8 26.3 23.4 344 12 - 23 36.9 36.9 36.4 99.5 98.2 87.2 75.2 98.6 87.2 75.2 75.2 67.4 591 24 - 35 24.5 24.5 53.3 98.4 19.2 86.5 74.6 98.4 89.7 74.6 78.1 70.6 514 36 - 47 11.7 11.7 53.7 94.7 98.7 84.0 58.7 98.7 84.0 58.7 81.3 56.0 644 48 - 59 9.7 9.7 61.0 100.0 94.9 89.8 67.8 96.6 89.8 67.8 69.5 55.9 608 Total 24.3 24.3 45.8 98.8 95.0 78.9 60.2 95.2 79.4 60.5 57.2 47.4 2986 CHILDREN 12-59 MONTHS Residence Urban 25.7 25.7 55.8 99.2 87.3 86.4 72.8 97.7 87.6 72.4 78.6 67.3 1000 Rural 18.6 18.6 53.1 97.7 99.1 87.3 70.1 99.1 87.8 70.6 79.2 62.4 1186 River lne 0.0 0.0 10.5 171 Educat ion None 15.6 15.6 46.8 98.3 97.7 86.1 67.4 97.1 87.2 67.4 76.7 61.6 1103 Primary 22.2 22.2 51.9 98.9 97.8 83.2 66.9 98.9 83.7 66.3 75.3 60.1 803 Secondary + 28.4 28.4 60.5 98.4 99.2 93.0 83.6 99.2 93.8 84.4 86.7 76.6 451 Rel ig ion Cathol ic 23.8 23.8 47.6 98.5 95.4 86.2 72.3 96.9 87.7 72.3 80.0 66.2 273 Protestant 19.5 19.5 50.9 98.1 99.1 87.6 71.1 99.1 88.2 71.1 80.8 65.2 1654 Mus l im 21.5 21.5 56.1 i00.0 98.8 87.7 74.1 98.8 88.9 74.1 72.8 64.2 376 Other /None 18.5 18.5 44.4 100.0 80.0 60.0 60.0 80.0 60.0 60.0 60.0 60.0 54 Total 20.3 20.3 51,2 98.5 98.1 86.8 71.6 98.3 87.7 71.6 78.9 65.1 2357 61 Child Morbidity and Treatment Information was collected for all children under age five on the prevalence of diarrhoea in the two weeks preceding the survey and on the prevalence of fever and respiratory ailments in the four weeks before the survey. The types of treatment given were also recorded. The data on diarrhoea, fever and respiratory illness are adequate for measuring the period prevalence of each illness (i.e., the percentage of children under 5 years whose mothers reported that they suffered from the illness under investigation during the reference period). Table 6.7 Among Children Under Five Years, the Percentage Reported by the Mother as Having Bad Diarrhoea in the Two Weeks Preceding the Survey and, Among Chi ldren With Diarrhoea, the Percentage Receiving Various Treatments, According to Background Characterist ics of Chi ld and Mother, ODHS, 1986 Among Children wlth Diarrhoea, Percent Receiving Various Treatments Percent Number Having No. of Background Diarrhoea of Consult ORT Other No Children Character- in Last Child- Medical ORS Home Treat- Treat- With istic 2 Weeks ren Faci l ity Packets Solution ment ment Diarrhoea Age of Child < 6 mos; 4.9 285 64.3 0.0 42.9 71.4 0.0 14 6 - ii mos. 7.9 344 25.9 3.7 29.6 70.4 II.i 27 12 - 23 mos. 8.3 590 38.8 0.0 18.4 71.4 10.2 49 24 - 35 mos. 4.7 514 58.3 4.2 16.7 70.8 12.5 24 36 - 47 mos. 4.2 695 29.6 0,0 18.5 85.2 14.8 27 48 - 59 mos. 1.8 688 54.6 0.0 27.3 45.5 27.3 11 Sex Boy 5.6 1524 37.2 0.0 26.7 74.4 10.5 86 Girl 4.5 1462 47.0 3.0 18.2 68.2 13.6 66 Residence Urban 4.8 1263 51.7 0.O 28.3 70.0 8.3 60 Rural 5.1 1495 38.2 2.6 23.7 72.4 13.2 76 Riverlne 7.0 228 18.8 0.0 0.0 75.0 18.8 16 Education None 4.9 1357 43.9 1.5 13.6 63.6 15.2 66 Primary 6.4 1014 43.1 1.5 24.6 75.4 12.3 65 Secondary + 3.4 615 28.6 0.0 47.6 85.7 0.0 21 Rel igion Cathol ic 5.2 347 55.6 0.0 16.7 72.2 5.6 18 Protestant 5.5 2102 39.1 1.7 24.4 73.0 11.3 115 Musl im 3.8 474 38.9 0.0 22.2 61.1 22.2 18 Other/None 1.6 63 100.0 0.0 0.0 i00.0 0.0 1 Total 5.1 2986 41.5 1.3 23.0 71.7 11.8 152 62 In considering the morbidity information, it is important to bear in mind that the subjective evaluation of the mother on the health condition of her child is reflected in the data collected. For example, in some homes, mothers believe that diarrhoea is no illness but a "natural cleaning mechanism for the body." Thus, diarrhoea may not be regarded as "running stomach" or "stooling" until a child becomes severely dehydrated, develops a temperature, and is very ill. Morbidity measures are also affected by the accuracy of the mother's recall concerning when an illness occurred. Both the failure to report an illness which occurred within the reference period and the reporting of an episode which occurred prior to the reference period would affect the accuracy of a prevalence estimate. Diarrhoea Table 6.7 shows the percentage of children under age five who had diarrhoea in the two weeks preceding the survey. As reported by their mothers, 1 out of 20 children in this age group suffered at least one bout of diarrhoea during the reference period. Diarrhoea prevalence varied with the age of the child; the highest rate was for children between the ages of 6 and 23 months. Prevalence rates did not differ substantially by the sex of the child. Regarding background characteristics, the percentage of children with diarrhoea was highest in the riverine area and for children whose mothers had a primary education. Table 6.7 indicates what treatment, if any, was used by mothers to treat the diarrhoea. Forty-two percent were taken to a medical facility, 23 percent were given a homemade salt and sugar solution (oral rehydration therapy), 1.3 percent were treated with a solution prepared from ORS packets, and 12 percent received no treatment at all. However, differences in the treatment of diarrhoea did not follow the expected pattern. For example, a smaller proportion of children whose mothers have a secondary or more education (29 percent) consulted a medical facility when their children had diarrhoea than mothers with either primary (43 percent) or no education (44 percent). However, caution should be exercised when interpreting this finding because of the small number of cases of diarrhoea reported by mothers with secondary or more education and the relatively high sampling variance of the proportion consulting a doctor for treatment. Among mothers of children under age five, the percemage who know about oral rehydration therapy (ORT)--either commercially prepared packets of salts (ORS) or home solution--increases with the level of education (Table 6.8). Almost 70 percent of women with secondary education know about ORT, compared to 40 percent of women with primary education and less than 30 percent of uneducated mothers. Also, knowledge of ORT was greater among urban than rural women while women in riverine areas were least likely to know about ORT. Kowledge of ORT based on religon was about the same for Catholic, Protestant and Muslim women, but decidely lower among the small number of women in the other/none religious category. Table 6.9 Among Mothers of Children Under Five Years, the Percentage Who Know About ORT by Educatlont According to Background Characteristics I ODHS I 1986 Background Education Characterlstlc None Primary Secondary + Total Res idence Urban 41.1 49.0 74.8 93.6 Rural 22.2 34.8 54.8 31.0 Riverlne 4.2 i0.0 0.O 5,8 Rellqlon Cathollc 30.0 36.0 60.0 40.8 Protestant 25.5 99.9 66.5 40.0 Musllm 31.5 44.7 64.3 37.5 Other/None 11.8 29.0 100.0 17.4 Total 26.9 39.6 67.9 39.2 63 Fever Table 6.9 shows the percentage of children under age five who had fever in the four weeks preceding the survey. It is worth noting that malaria is endemic to Ondo State and the reporting of fever could well indicate an episode of malaria. Twenty-two percent of children under five were reported as having had fever during the four weeks preceding the survey. Age was related to the occurrence of fever, with the greatest prevalence (26 percent) occurring among children 12 to 23 months. The prevalence of fever showed little variation based on background characteristics; but surprisingly, it was lowest in the riverine area. Over 40 percent of children with fever in the four weeks preceding the survey were trea~xl with antimalarial medicine. Only a small percent (2 percent) of children with fever received no treatment. Table 6.9 ~ong Chi ldren Under Five Years, the Percentage Who Are Reported by the Mother as Having Had Fever in the Past Four Weeks and, ~ong Chi ldren Who Had Fever, the Percentage Receiv ing Various Treatments, According to Background Characterist ics, ODES, 1986 Percent Among Chi ldren With Fever, Percent Receiv ing Number Having Various Treatments of Fever Chil- Background in Consult No dren Character- Last 4 Medical Anti- Anti- Other Treat- Under istic Weeks Faci l i ty malar ia l b iot ic Medic ine ment Five Age of Chi ld < 6 mos. 12.3 37.1 28.6 0.0 85,7 0.0 285 6 - II mos. 24.1 36,1 50.6 0.0 89,2 2.4 344 12 - 23 mos. 26.4 35.3 45.5 0.0 87,2 1.3 590 24 - 35 mos. 26.1 40.3 32.8 0.0 91.8 0.0 514 36 - 47 mos. 21.7 26.4 39.3 0.0 94.3 2.9 645 48 - 59 mos. 19.2 34.2 47.9 0.0 92.3 1.7 608 Sex of Chi ld Boy 22.6 36.2 43.5 0.0 90.7 1.5 1524 Girl 21.9 32.5 40.0 0.0 90.6 1.6 1462 Residence Urban 22.3 42.2 43.3 0.0 88.3 0.4 1263 Rural 24.3 29.7 40.1 0.0 94.2 1.4 1495 River ine 8.3 10.5 52.6 0.0 57.9 21.1 228 Educat ion None 20.9 29.7 36.0 0.0 89.1 3.2 1357 Primary 24.6 34.5 45.0 0.0 93.6 0.4 1014 Secondary + 21.6 44.4 48.1 0.0 88.7 0.0 615 Rel ig ion Cathol ic 19.9 43.5 40.6 0.0 89.9 1.5 347 Protestant 22.4 34.6 41.8 0.0 91.5 1.7 2102 Musl im 24.5 31.0 42.2 0.0 87.1 0.9 474 Other/None 14.3 0.0 44.4 0.0 100.0 0.0 63 Tota l 22.3 34.4 41.8 0.0 90.7 1.5 2986 64 Cough/Difficult Breathing The ODHS collected information on the prevalence of respiratory illness by inquiring from the mothers of children under age five whether the child had severe coughing and/or difficult breathing in the four weeks preceding the survey. Table 6.10 shows that seven percent of children under age five suffered from severe cough and/or difficult breathing in the month before the survey. Age was related to the respiratory-related ailment; the highest prevalence (11 percent) occurring among children age 6-11 months. The data indicate little difference in the prevalence of severe coughing and/or difficult breathing by sex of child, residence, education, and religion of mother. About 67 percent of children who had a respiratory problem received cough syrup, 35 percent were taken to a medical facility, 23 percent received antibiotics, and 31 percent received other medicine. Only 4 percent received no treatment. Table 6.10 Among Chi ldren Under Five Years, the Percentage Who Are Reported by the Mother as Having Suffered from Severe Cough and~or Diff icult Breathing in the Past Four Weeks and, Among Chi ldren Who Suffered from Severe Cough and/or Diff icult Breathing, the Percentage Receiv ing Various Treatments, According to Background Characteristics, ODHS, 1986 Background Character- istic Percent ~mong Chi ldren With Cough, Percent Receiv ing Number Having Various Treatments of Cough Chll- in Consult No dren Last 4 Medical Anti- Cough Other Treat- Under Weeks Faci l i ty biot ic syrup Medicine ment Five Age of Child < 6 7.7 36.4 22.7 81.8 31.8 4.5 285 6 - ii 10.8 27.0 21.6 78.4 24.3 0.0 344 12 - 23 10.2 33.3 36.7 65.0 20.0 6.7 590 24 - 35 5.3 48.2 18.5 63.0 44.4 7.4 514 36 - 47 3.7 20.8 12.5 54.2 29.2 4.2 645 48 - 59 5.8 45.7 11.4 60.0 48.6 2.9 608 Sex of Chi ld Boy Girl 6.4 35.1 26.8 71.1 28.9 3.1 1524 7.4 35.2 19.4 63.0 33.3 5.6 1462 Residence Urban 6.7 38.1 17.9 63.1 25.0 6.0 1263 Rural 2.4 34.2 25.2 72.1 34.2 3.6 1495 River ine 4.4 20.0 40.0 40.0 50.0 0.0 228 Education None 7.2 30.9 16.5 62.9 29.9 6.2 1357 Primary 7.3 40.5 25.7 67.6 31.1 4.1 1014 Secondary + 5.5 35.3 35.3 76.5 35.3 0.0 615 Rel igion Cathol ic 7.5 42.3 23.1 50.0 26.9 11.5 347 Protestant 7.0 34.7 23.1 72.1 30.6 3.4 2102 Musl im 5.7 33.3 22.2 59.3 33.3 3.7 474 Other/None 7.9 20.0 20.0 40.0 60.0 0.0 63 Total 6.9 35.1 22.9 66.8 31.2 4.4 2986 65 6.4 Anthropometric Data The assessment of the nutritional status of children in the ODHS is based on height and weight data and age (in months). Procedures for converting such information into nutritional indices are well- established. The data necessary for calculating these indices were collected for children age 6-36 months of women interviewed in the survey. The interviewers responsible for collecting the height and weight data were provided with standardized equipment and special training to ensure that they would be able to collect accurate data (See Appendix A.3). While age in years is sufficient for most demographic analyses, age in months is required for anthropometric assessment. This is because a child can be misclassified as severely undernourished or overnourished if his/her reported age is in error by just a few months. In Ondo State it is common for mothers not to know the precise ages of their children. When this occurred, the interviewer was instructed to aid the respondent to estimate the birth date of the child on the basis of a local events calendar. The degree to which there are errors in the ODHS age data can, to some extent, be assessed by examining the data for age heaping. Figure 6.1 shows the number of children age 6 to 36 months of respondents in the ODHS survey. To the extent that heaping occurs in a data set, it usually occurs at months 6, 12, 18, 24, 30 and 36. There is no discernible pattern of heaping in Figure 6.1. In the ODHS, 1504 children between the ages of 6 and 36 months were identified and 1402 (93 percent) were weighed and measured. No children were missing information on the month and year of birth. However, fifteen cases were excluded from the analysis because the anthropometric indices were improbably too high or too low, suggesting an error in the recording of information. Analysis was carded out on 1387 cases. Figure 6.1 Age Distribution of Measured Children and All Children Number of Children 80 70 60 50 40 30 20 10 0 0 + + + + 4 -++ + ++ ++ + 4. + ÷ 4 -++ 4- + + ÷ 5 10 15 20 25 30 35 40 Age in Months Measured Children + All Children / Ondo State DHS 1986 66 Nutritional Status Nutritional status assessment is based on the concept that in a well-nourished population, there will be a distribution of children of a given age with respect to height and weight. In terms of a particular index (say, height-for-age), the distribution will approximate the normal curve. About 68.2 percent of children will have a height-for-age within 1 standard deviation of the mean. About 13.6 percent will be between +1 and +2 standard deviations from the mean and another 13.6 percent between -1 and -2 standard deviations from the mean. Finally, about 2.3 percent will be more than +2 standard deviations from the mean and another 2.3 percent will be more than -2 standard deviations from the mean. The nutritional indices presented in this report are based on the WHO/CDC/NCHS reference population (U.S. Department of Health, Education and Welfare, 1976). Thus, the children in Ondo State are classified into nutrition status categories in terms of the mean and standard deviation values of that reference population. Four nutritional indices are presented in this report. Height-for-age. A child who is 2 or more standard deviations below the mean of the reference population is considered short for his/her age which could reflect the cumulative effect of chronic malnutrition. Such a child is referred to as "stunted". Weight-for-height. A child who is 2 or more standard deviations below the mean of the reference population is considered thin which could reflect a recent episode of illness resulting in acute malnutrition. Such a child is referred to as "wasted". Height-for-age by weight-for-height. This index is a cross tabulation of the above two indices and can identify a child who is both chronically and acutely malnourished. A child who is 2 or more standard deviations below the mean of the reference population on both indices is considered severely malnourished. Weight-for-age. A child who is 2 or more standard deviations below the mean of the reference population could reflect chronic malnutrition, a recent acute episode of malnutrition or both. Thus, this index provides less precise information than the previously described indices. Nevertheless, weight-for-age is reported because it may be useful for comparison with other data on the nutritional status of children in Ondo State. Height-for-Age Table 6.11 shows the percent of children 6-36 months by various standard deviation categories from the mean of the reference population in terms of height-for-age. To make interpretation of anthropometric data easier, the World Health Organisation has classified children whose height-for-age is between 2 and 3 standard deviations below the reference mean as moderately stunted and children whose beight-for-age is 3 or more standard deviations below the reference mean as severely stunted. Table 6.11 indicates that 19.5 percent are moderately stunted and 12.9 percent are severely stunted. 67 Stunting is evident in equal proportions among males and females, but by other background characteristics there are some important differentials. Stunting increases considerably when progressing from younger to older children. The percent that are severely stuntr~d is lowest for children age 6-11 months (2 percent), increases sharply for children age 12-23 months (11 percent) and is still greater for children age 24-36 months (22 percent). Twins are highly likely to be stunted: 38 percent are severely stunted. By area of residence there is little difference between children living in riverine areas and in rural areas but somewhat less stunting among urban children. There is little difference in the degree of stunting by education of mother. Table 6.11 Percent Distr ibut ion of Children Aged 6-36 Months by Standard Deviat ion Category of Height- for-Age Using the In£ernational NCHS/CDC/WHO Reference, According to Background Characterist ics, ODHS, 1986 Standard Deviat ions from NCHS/CDC/WHO Reference Background -3.00 -2.00 -I .00 -0.99 +i.00 +2.00 Character- or to to to to or Total istic more -2.99 -1.99 +0.99 +1.99 more Percent International Reference 0.6 1.7 13.6 68.2 13.6 2.3 i00.0 Number of Chi ldren 6-36 Months Sex of Chi ld Male 12.7 20.4 32.5 31.1 2.5 1.0 100.0 727 Female 13.2 18.6 29.1 35.8 1.7 1.7 100.0 660 Age of Chi ld 6-11 mos. 2.2 8.9 29.4 52.2 5.1 2.2 100.0 316 12-23 mos. II.I 21.3 33.3 31.9 1.4 1.6 i00.0 559 24-36 mos. 21.5 24.2 29.1 23.8 1.0 0.4 i00.0 512 Birth Interval* First Births 16.5 19.9 32.5 30.7 0,0 0.4 100.0 231 < 2 years 13.6 17.6 33.5 33.0 1.7 0.6 100.0 176 2-3 years 12.4 19.9 29.6 33.4 2.9 1.7 100.0 749 4 years + 10.4 19.5 31.2 35.9 1.7 1.3 100.0 231 Twins 37.5 17.2 26.6 17.2 1.6 0.0 100.0 64 Residerce Urban 10.2 15.3 34.7 36.5 2.1 I.i 100.0 619 Rural 15.0 23.4 27.0 31.6 1.9 1.2 i00.0 675 Riverine 16.1 19.4 33.3 24.7 3.2 3.2 I00.0 93 Educat ion None 12.8 21.1 28.9 34.1 2.0 1.2 100.0 596 Pr imary 14.5 17.1 34.0 30.6 2.4 1.5 100.0 468 Secondary + 10.8 20.1 30.0 35.9 1.9 1.2 100.0 329 All Chi ldren 12.9 19.5 30.9 33.3 2.1 1.3 100.0 1387 * Twins are included in the preceding birth interval statistic; both twins have the s~u~e interval. Twins are also presented as a separate category. 68 Weight-for-Height Weight-for-height is a measure of undemutrition of recent onset. About 7 percent of the children have a weight-for-height index 2 or more standard deviations below the reference mean and are classified as wasted (Table 6.12). Table 6.12 Percent Distr ibut ion of Chi ldren Aged 6-36 Months by Standard Deviat ion Category of Weight-for-Height Using the International NCHS/CDC/WHO Reference, According to Background Characteristics, ODHS, 1986 Standard Deviations from NCBS/CDC/WHO Reference Background -3.00 -2.00 -I.00 -0.99 +I.09 +2.00 Character- or to to to to or Total iatic more -2.99 -1.99 +0.99 +1.99 more Percent International Reference 0.6 1.7 13.6 68.2 13.6 2.3 100.0 Number of Chi ldren 6-36 Months Sex of Child Male 0.8 5.9 24.3 64,2 4.5 0.I 100.0 727 Female 0.5 5.6 27.4 62.3 3.5 0.8 100.0 660 Age of Child 6-11 mos. 0.3 3.8 26.3 62.0 6.6 0.9 100.0 316 12-23 mos. 0.5 8.1 29.9 58.3 2.9 0.4 100.0 559 24-36 mos. l.O 4.5 21.1 69.5 3.7 9.2 i00.0 512 Birth Interval* First Births 1.3 7.4 27.3 60.2 3.5 0.4 100.0 231 < 2 years 0.6 5.1 27.3 60.8 6.3 0.0 100.0 176 2-3 years 0.5 5.6 25.0 64.4 3.9 0.7 100.0 749 4 years + 0.4 5.2 26.0 64.9 3.5 0.0 100.0 231 Twins 4.7 9.4 32.8 48.4 4.7 0.O 180.0 64 Residence Urban 0.3 4.2 23.3 66.9 4.8 0.5 10O.0 619 Rural 0.7 5.8 26.7 63.0 3.6 9.3 100.0 675 Riverine 2.2 16.1 36.6 41.9 2.2 1.1 100.0 93 Education None 0.8 5.5 28.0 61.2 4.0 0.3 1O0.0 596 Primary 0,4 6.6 25.2 63.0 4.1 0.6 100.0 468 Secondary + 0.6 5.0 22.6 67.5 4.0 0.3 100.0 323 Recent Morbidity** Diarrhoea I.i 17.0 27.7 48.9 5.3 0.0 100.0 94 Fever 0.3 5.5 26.8 63.0 4.1 0.3 100.0 365 Cough 0.8 7.3 27.6 60.2 3.3 0.8 100.0 123 All Children 0.7 6.1 26.3 62.4 4.1 0.4 1O0.0 1387 * Twins are included in the preceding birth interval statistic; both twins have the same interval. Twins are also presented as a separate category. **The reference periods were 2 weeks for diarrhoea and 4 weeks for fever and cough. 69 Differentials in wasting for population subgroups are also shown in Table 6.12. No differential is found in the case of sex. Significant differentials are found in the case of age. Wasting increases when progressing from children age 6-11 months (4 percent) to children 12-23 months (9 percent), but decreases for children age 24-36 months (6 percent). Wasting is only weakly associated with the length of the birth interval, although, a greater percentage of first births suffer from wasting than second and higher order births. As is found for the stunting index, the prevalence of wasting is much higher for twins (14 percent) than for all children (7 percent). Wasting is also higher for children in the riverine area than for children in the urban and rural areas. Among children with recent diarrhoea, 18 percent are wasted, reflecting the importance of diarrhoea in contributing to malnutrition. Height-for Age by Weight-for-Height The relationship between stunting and wasting, or chronic and acute undemutrition, is shown in Figure 6.2. The figure depicts children classified according to their status with respect to both height-for- age and weight-for-height. It reveals that approximately 2 percent of children age 6 to 36 months are both stunted and wasted (i.e., fall 2 standard deviations or more below the mean of the reference population in terms of height-for-age and weight-for-age). They are the most severely undernourished children in the population. Figure 6.2 is also useful for demonstrating the amount of hidden chronic undemutrition in the childhood population. A child who is stunted but not wasted will not be recognized as being undemourished by the casual observer: a child who appears to be a healthy two year old, may in fact be a stunted three year old. The survey found that 21 percent of children were moderately or severely stunted and normal with respect to weight-for-height. 30 25 20 15 10 5 0 Figure 6.2 Cross-tabulation of Weight-for-Height and Height-for-Age Percent of Children / Moderat Stuntin Mod. a Say wasting Weight-for-Height 21 I ~25~22~ ~ Height-for-Age ; and Severe 10 ! an dWaeting ~ 7 Mod., ssv. s,un,,°, M,,Os,on,lng J ' Normal / Mild Wasting Normal Ondo State DHS 1986 70 Weight-for-Age Table 6.13 shows the percent of children by various standard deviation categories from the mean of the reference population in terms of weight-for-age. Because weight-for-age is a composite index which reflects long term chronic undemutrition and recent acute undcmutrition, it does not provide information beyond that already presented in the tables on height-for-age and wcight-for-hcight. Table 6.13 Percent Distr ibut ion of Chi ldren Aged 6-36 Months by Standard Deviat ion Category of Weight-for-Age Using the International NCHS/CDC/WHO Reference, According to Background Characterist ics, ODHS, 1986 Standard Deviations from NCHS/CDC/WHO Reference No. of Background -3.00 -2.00 -1.00 -0.99 +1.00 +2.00 Chi ldren Character- or to to to to or Total 6-36 istic more -2.99 -1.99 +0.99 +1.99 more Percent Months International Reference 0.6 1.7 13.6 68.2 13.6 2.3 100.0 Sex of Chi ld Male 5.9 22.3 34.7 35.2 1.8 0.1 100.0 727 Female 5.3 22.7 37.1 32.9 1.7 0.3 100.0 660 Age of Child 6-11 mos. 3.5 13.9 37.0 40.5 4.7 0.3 100.0 316 12-23 mos. 6.1 22.5 38.6 31.1 1.3 0.4 100.0 559 24-36 mos. 6.4 27.7 32.8 33.4 0.4 0.0 i00.0 512 Birth Interval* First Births 10.4 26.0 31.6 31.6 0.0 0.4 I00.0 231 < 2 years 5.7 21.6 35.2 36.4 I.i 0.0 100.0 176 2-3 years 4.3 23.5 35.4 34.4 2.3 0.I i00.0 749 4 years + 5.2 16.5 42.0 33.8 2.2 0.4 I00.0 231 Twins 14.1 46.9 23.4 14.1 1.6 0.0 100.0 64 Residence Urban 3.4 17.9 36.2 40.4 1.8 0.3 100.0 619 Rural 7.0 25.8 34.8 30.7 1.6 0.I I00.0 675 Riverine 10.8 29.0 40.9 17.2 2.2 0.0 100.0 93 Education None 4.9 23.2 38.4 32.2 1.0 0.3 I00.0 596 Pr imary 6.6 22.2 34.2 33.5 2.8 0.2 i00.0 468 Secondary + 5.6 21.2 33.4 38.4 1.5 0.0 100.0 323 Recent Morbidity** Diarrhoea 7.4 30.9 29.8 30.9 i.I 0.0 100.0 94 Fever 5.7 20.3 41.5 29.3 3.3 0.0 100.0 365 Cough 4.4 24.4 34.2 33.7 2.5 0.8 100.0 123 All Children 5.6 22.5 35.8 34.1 1.7 0.2 i00.8 1387 * Twins are included in the preceding birth interval statistic; both twins have the same interval. Twins are also presented as a separate category. **The reference periods were 2 weeks for diarrhoea and 4 weeks for fever and cough. 71 Summary of Nutritional Status by Age Figure 6.3 provides a summary of the nutritional status of children in Ondo State by age. The figure shows the mean standard deviation score of all children from the mean of the reference population in terms of height-for-age, weight-for-age and weight-for-height. The height-for-age index indicates a mean value which is about -0.6 standard deviations from the reference mean at six months and about -2.0 standard deviations from the reference mean at thirty-six months. The weight-for-age index shows a similar trend with age being about -0.6 standard deviation from the reference mean at six months of age and about -1.5 standard deviations at thirty-six months. On the other hand, the weight-for-height index is only -0.2 standard deviations from the reference mean at six months, falls to -0.8 standard deviations at fourteen months and is back at about -0.3 standard deviations at thirty-six months. Each of these indices presents a different perspective on the nutritional problems of children in Ondo State. The weight-for-height index shows that undemutrition of recent onset is most common in children 12-18 months of age. The height-for-age index shows that undemutrition of longer standing (stunting) is most pronounced in children 30-36 months. However, at this age the weight-for-height index is close to the reference population mean, and the children do not appear undernourished. 0 -1 -2 -3 0 Figure 6.3 Nutritional Status of Children 6-36 Months Compared to International Reference SD's from International Reference 6 12 18 24 30 36 Age in Months Weight-for-Age ~' Height-for-Age ~ Weight-for-Height I Ondo State DHS 1986 72 REFERENCES Cttieh-Johnson, D., Cross, A.R., Way, A.A., and Sullivan, J.M. 1988. Liberia Demographic and Health Survey 1986, Columbia, Maryland: Ministry of Planning and Economic Affairs and Institute for Resource Developmenl/Westinghouse. Ekanem, I.I. 1972. The 1963 Nigerian Census: A CriticalAppraisal. Benin City: Ethiope Press. Federal Ministry of Health, Federal Republic of Nigeria. 1988. National Policy on Population for Development, Unity, Progress and Self-Reliance. Lagos, Nigeria. Institute for Resource Development. 1987. Model "B" Questionnaire With Commentary, For Low Contraceptive Prevalence Countries. Basic Documentation No. 2. Columbia, Maryland. National Population Bureau (NPB). 1984. The Nigeria Fertility Survey 1981-82. Lagos, Nigeria: NPB and International Statistical Institute. National Population Bureau and Federal Ministry of Health. 1986. The Nigeria Fertility Survey 1981-82, State Level Report, Ondo State. Lagos, Nigeria. National Population Bureau (NPB) and Institute for Resource Development/Westinghouse (IRD). [1988]. National Demographic Sample Survey: Ondo State Report, 1980. Columbia, Maryland: NPB and IRD. Ndiaye, S., Sarr, I., and Ayad, M. 1988. Enqudte Ddmographique et de Santd au Sdndgal 1986. Columbia, Maryland: Ministere de l'Economie et des Finances and Institute for Resource Development/Westinghouse. Rutstein, S.O. 1984. Infant and Child Mortality: Levels, Trends and Demographic Differentials. Comparative Studies, Cross-National Summaries No. 43. Voorburg, Netherlands: International Statistical Institute (World Fertility Survey). Sherris, J., Blackburn, R., Moore, S., and Mehta, S. 1986. Population Reports, Series L, No. 5. Baltimore, Maryland: Information Program. "Immunizing the World's Children." Johns Hopkins University, Population Singh, S. and Ferry, B. 1984. Biological and Traditional Factors that Influence Fertility: Results from WFS Surveys. • Comparative Studies, Cross-National Summaries No. 40. Voorburg, Netherlands: International Statistical Institute (World Fertility Survey). United Nations, Department of Technical Co-operation for Development and Statistical Office. 1986. How to Weigh and Measure Children: Assessing the Nutritional Status of Young Children in Household Surveys. New York. (National Household Survey Capability Program). United States, Department of Health, Education and Welfare, National Center for Health Statistics. 1976. "NCHS Growth Charts, 1976." Monthly Vital Statistical Report, 25 (June 22; Suppl.), No. 3. 73 APPENDIX A SURVEY DESIGN APPENDIX A SURVEY DESIGN A.1 Sample Design and Implementation The sample specifications for the ODHS called for a self-weighting sample of approximately 3,600 women 15-49 representative of the entire state. Considering the cultural homogeneity of Ondo State, it was decided that an efficient design would be achieved with 90 primary sampling units (PSUs). Thus, the overall sampling fraction for the ODHS was F 3960 (Population of Ondo State) (.22) where 3960 is the target sample size increased by 10 percent for nonresponse, and .22 is an estimate of the proportion of women 15-49 in the population (National Population Bureau, 1984). A two-stage sampling procedure was used. In the first stage, census enumeration areas (EAs), were selected with sampling fraction f(1). In the second stage, households were selected with sampling fraction f(2). Within selected households, eligibility for the woman's questionnaire was on a de facto basis: all women 15-49 who stayed in the household the previous night were eligible respondents. Sampling Frame The sampling frame for PSU selection consisted of two lists of enumeration areas. One list covered 13 of the 17 Local Govemment Areas (LGAs) of Ondo State and was created between 1984 and 1986 as part of Enumeration Area Demarcation Exercise of the National Population Bureau. The other list, covering 4 LGAs, consisted of the enumeration areas created for the 1973 Population Census and updated for the ODHS. EAs in the sampling frame showed little variation in measure of size, so that it was unnecessary to consider selection of the first stage with probability proportional to size. First Stage Selection Within each Local Government Area, the EAs were listed with urban areas first in order to achieve some implicit stratification. A systematic sample of 90 EAs was then selected using the following procedure. The total of 7,638 EAs was divided by 90 to give 85 (f (1) = .0118); a random number was selected between 1 and 85; then EAs were selected from the list at the fixed interval of 85, starting with the random number. Household Listing The next step was to list households in the selected EAs. During the listing operation, the number of persons residing in each household was recorded. At this stage, it was observed that the EA 77 populations obtained from the listing were systematically smaller by about 46 percent (Table A.2) than those from the Demarcation Exercise. This discrepancy suggested that the Demarcation Exercise was producing serious overestimates of the population. An adjustment was therefore introduced for this in the next step. Second Stage Selection The relationship between sample selection fractions is as follows: f(2) = F/f(1). Thus, before f(2) could be calculated, a value for the overaU sampling fraction, F, was needed and this required an estimate of the population of Ondo State. At the time of second stage sample selection (August 1986), three sources of population information were available: the 1963 Population Census, the Enumeration Area Demarcation Exercise and the household listing operation of the ODHS survey. As shown in Table A.1, the population estimate for the 13 LGAs covered by the Enumeration Area Demarcation Exercise (2,674,734) indicates an increase of 22 percent over the 1963 Census Population for those LGAs (2,193,603). Assuming equal growth in the 4 remaining LGAs, a total population estimate for Ondo State would be 3,329,856 (2,729,390* 1.22). Table A.I Populat ion Est imates for Ondo State, by Local Government Area Populat ion Est imate 1963 EA Demarcat ion Exercise Local Government Area Census 1984-86 01 Akure 129,415 400,435 02 Akoko North 153,311 282,165 03 Ekit i East 123,649 156,720 04 Ero 224,088 173,277 05 Ekit i South 220,076 151,619 06 Ekit i Southwest 124,044 87,304 07 Ekit i Central 262,337 212,410 08 lJero 133,963 117,490 09 Ekit i West 171,347 143,555 I0 Akoko South 131,508 165,735 Ii Ekit i North 181,455 253,111 12 Owo 189,847 322,273 13 Idanre Ifedore 148,596 208,640 14 Ondo* 148,734 -- 15" Ifesowapo* 109,330 -- 16 Ilaje Eseodo* 93,926 -- 17 Ikale* 183,797 -- TOTALS - All LGAs 2,729,390 -- LGAs 01-13 2,193,603 2,674,734 *The EA Demarcat ion Exercise was not completed in four LGAs at the time of sample selection. 78 However, as shown in Table A.2, the results from the ODHS household listing operation indicate that the estimates from the Enumeration Area Demarcation Exercise were 46 percent too high. Thus, for sampling purposes, an appropriate population estimate would be 1,797,090 (3,327,945*.54). This estimate was used to calculate t'(2) and I(2), the second stage sampling interval. I(2) = l/f(2)= 1/.8488 = 1.18. Table A.2 Populat ion Est imates in Selected Enumerat ion Areast EA Demarcat ion Exercise and ODHS Household L ist ing Est imated Populat ion in Selected EAs Local Government Area Number EA Demarcat ion DHS Household of EAs Exercise List ing in sample 1984-86 1986 Ratio 01 Akure 9 4,956 3,401 .69 02 Akoke North 7 3,472 1,807 .52 83 Ekit i East 4 1,696 821 .48 04 Ere 4 1,821 520 .29 05 Ekit i South 4 1,688 775 .46 06 Ekit l Southwest 2 1,190 547 .46 07 Ekit i Central 5 2,687 ],749 .65 08 Ijero 3 1,307 522 .40 89 Ekit i West 3 1,890 1,247 .74 i0 Akoko South 4 2,122 1,309 .62 ii Ekiti North 6 3,130 1,234 .39 12 Owe 8 3,759 1,740 .46 13 Idanre Ifedore 5 2,283 1,177 .52 14 Ondo* 9 . . . . . . 15 Ifesowapo* 2 . . . . . . 16 IlaJe Eseodo* 9 . . . . . . 17 Ikale* 6 . . . . . . TOTAL 90 31,800 16,849 .54 * Populat ion est imates not avai lable at time of sample selection. Est imates were obtained while the DHS data col lect ion was underway dur ing a combined updat ing/ l i s t ing exercise. This exercise was carr ied out by NPB/Akure and was restr icted to EAs selected in the DHS sample. The household l ist ing was conducted using DHS forms and procedures. A.2 Quest ionna i re Des ign and Pretest Questionnaires Two questionnaires were used in the ODHS: a household schedule and an individual questionnaire for women. Both were adapted from the model questionnaires of the DHS Programme (Institute for Resource Development, 1987). 79 The household schedule collected basic information on household members and visitors who slept in the household the night preceding the survey, including name, whether a usual resident or a visitor, sex, age and, for children 15 years and below, presence of natural parents in the household. The individual questionnaire contained eight sections and collected data on: • Fertility - including a truncated birth history covering the six years preceding the survey and questions on desired number of children, and future childbearing intentions; • Fertility regulation - including knowledge and use of family planning, sources of family planning methods, and reasons for nonuse of family planning; • Maternal and child health - including prenatal care, breastfeeding, weaning practices, incidence of childhood diseases (such as fever, diarrhoea, and respiratory illness), immunisation status for children, and height and weight of children aged 6-36 months. A significant difference between the ODHS and other DHS surveys is the use of a truncated birth history rather than a full birth history. The questionnaires were printed in Yoruba, the first language of over 85 percent of the population of Ondo State. English versions of the questionnaires are reproduced in Appendix C. Pretest The pretest was conducted in June and July 1986. Pretest training consisted of one week of classroom instruction and one week of anthropometric measurement training and practice interviewing. A total of 16 people were trained. Trainers for the pretest consisted of the senior survey staff and two DHS staff members, including a specialist in anthropometric measurement. Pretest fieldwork took place in three locations, lasted two weeks and covered 250 urban and rural households. Two teams conducted interviews: each consisted of a supervisor and five interviewers. Completed questionnaires were edited in the field by the senior survey staff and returned to the survey office for manual tabulation of results. A.3 Main Survey Training for the Main Survey Training for the main survey took place in August 1986. As the ODHS interviewers were responsible for collecting data on the height and weight of children, anthropometric training was included in the training schedule. The four week training period was divided into one week of classroom instruction on the survey questionnaires, one week of practice interviewing with village women, one week of training in anthropometric measurement techniques and a final week of practice interviewing. Anthropometric training and subsequent fieldwork were conducted with standardized equipment: hanging spring scales for weighing children and portable wooden measuring boards for measuring their recumbent length. Trainees were taught to measure in teams of two and to follow the procedures specified in the manual "How to Weigh and Measure Children" (United Nations, 1986). At the end of the training period, all interviewers were tested on the accuracy with which they measured children. 80 A total of 32 field staff participated in the training: four supervisors (one female and three males), four female editors, sixteen female interviewers and eight data entry clerks. About one-third of these had participated in the pretest. In addition, two tutors from the School of Health Technology, Ondo State, were trained in anthropometric measurement and worked with the interviewing teams throughout fieldwork. Training was conducted by the senior survey staff and three staff members from DHS headquarters, including a specialist in anthropometric measurement. Fieldwork Fieldwork began September 5, 1986 and continued into January 1987. Data collection was accomplished by four teams each consisting of a supervisor, a field editor, four interviewers and a driver. Based on experience with the pretest, it was decided that anthropometric measurement of children would be done in respondents' homes rather than at a central location in each sample cluster. Since the DHS protocol requires that anthropometric measurements be made by two trained persons (a measurer and an assistant), this required that two measuring boards and two scales be provided to each field team and that interviewers work in pairs when taking these measurements. The task of measurer was assigned to the interviewer who conducted the interview and could identify the children to be measured; the task of assistant was assigned to the other interviewer. Response Rates The number of households selected for the ODHS sample was 3836. Of these, 3521 households were located in the field and 3437 completed questionnaires were obtained (household response rate of 98 percent). The completed household questionnaires identified 4239 eligible respondents. Completed interviews were obtained for 4213 of these (eligible women's response rate of 99 percent). The overall survey response rate, the product of the ho

View the publication