Reproductive Health Survey Romania, 1999 Final Report
Publication date: 2001
Final ReportARSPMS., , . Reproductive Health Survey Romani , 1999 REPRODUCTIVE HEALTH SURVEY ROMANIA, 1999 FINAL REPORT Edited by: Florina Serbanescu, MD Leo Morris, PhD Mona Marín, MD Romanian Association of Public Health and Health Management (ARSPMS) School of Public Health, University of Medicine and Pharmacy "Carol Davila" National Commission for Statistics (CNS) BUCHAREST, ROMANIA Division of Reproductive Health, Centers for Disease Control and Prevention (DRH/CDC) ATLANTA, GEORGIA, USA United States Agency for International Development (USAID) United Nations Population Fund (UNFPA) United Nations Children's Fund (UNICEF) September, 2001 Additional information about the 99RRHS English Final Report may be obtained from: Division of Reproductive Health, Centers for Disease control and Prevention (DRH/CDC), Mailstop K-35, 4770 Buford Highway, N.E., Atlanta, Georgia 30341-3724, USA. Fax (770) 488-6242, phone (770) 488-6200, Efirstname.lastname@example.org. The male questionnaire (English) is also available upon request addressed to DRH/CDC. Additional information about the 99RRHS Romanian Final Report may be obtained from: Department of Public Health and Health Management, University of Medicine and Pharmacy "Carol Davila ", 1-3 Dr. Leonte Street, Bucharest, 76256, Romania. Fax 401-224-3950, phone 401-2126297, E-mail email@example.com. Female and male questionnares (Romanian) are also available upon request. TABLE OF CONTENTS PREFACE . i ACKNOWLEDGMENTS . iii I. INTRODUCTION . 1 (Dan Enachescu, Fiorina Serbanescu, Leo Morris, Mihai Horga) II. METHODOLOGY . 5 (Leo Morris,Florina Serbanescu) 2.1 Sampling Design. 5 2.2 Data Collection. 6 2.3 Response Rates. 7 2.4 Sampling Weights. 9 III. CHARACTERISTICS OF THE SAMPLE. 11 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 3.1 Household Characteristics . 11 3.2 Characteristics of Eligible Women and Men . 17 TV. FERTILITY AND PREGNANCY EXPERIENCE . 25 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 4.1 Fertility Levels . 25 4.2 Fertility Differentials . 31 4.3 Nuptiality . 33 4.4 Age at First Sexual Intercourse, Union, and Birth .36 4.5 Recent Sexual Activity .40 4.6 Planning Status of the Last Pregnancy .42 4.7 Future Fertility Preferences.47 INDUCED ABORTION .55 (Fiorina Serbanescu, Leo Morris) 5.1 Induced Abortion Levels.58 5.2 Induced Abortion Differentials .60 5.3 Abortion Services .64 5.4 Reasons for Abortion . 70 5.5 Abortion Complications . 72 MATERNAL AND INFANT HEALTH . 75 (Fiorina Serbanescu, Gabriel Banceanu, Paul Stupp, Carmen Cruceanu) 6.1 Prenatal Care. 76 6.2 Intrapartum Care . 86 6.3 Postnatal Care . 92 6.4 Smoking and Drinking During Pregnancy . 94 6.5 Pregnancy and Postpartum Complications . 97 6.6 Poor Birth Outcomes . 100 6.7 Breastfeeding . 100 6.8 Infant and Child Mortality . 106 CONTRACEPTIVE AWARENESS AND KNOWLEDGE OF USE . 111 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 7.1 Contraceptive Awareness and Knowledge of Use . 111 7.2 First Source of Information About Contraception . 122 7.3 Knowledge About Contraceptive Effectiveness. 127 7.4 Young Adults Knowledge about Condoms' Effectiveness in Preventing STDs . . 130 CURRENT AND PAST CONTRACEPTIVE USE . 133 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 8.1 Current Contraceptive Prevalence . 133 8.2 Source of Contraception . 149 8.3 Dissatisfaction with the Current Method and Preference for Other Methods . 152 8.4 Users of Non-Supplied Methods . 157 8.5 Reasons for Not Using Contraception . 162 8.6 Intention to Use Contraception among Nonusers. 164 8.7 Recent Trends in Contraceptive Use . 167 8.8 Contraceptive Failure and Discontinuation. 170 IX. NEED FOR CONTRACEPTIVE SERVICES . 175 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 9.1 Potential Demand and Unmet Need for Contraception . 175 9.2 Potential Demand For Family Planning Services According to Fertility Preferences . 185 X. CONTRACEPTIVE COUNSELING . 193 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 10.1 Communication with Family Planning Providers .193 10.2 Satisfaction with Counseling Services . 196 10.3 Post-abortion and Post-Partum Counseling . 198 XI. ATTITUDES AND OPINIONS ABOUT CONTRACEPTION AND ABORTION . . 201 (Mihail Gr. Marcu, Aurelia Marcu, Gabriela Cristisor, Jay Friedman) 11.1 Interest in Information on Contraception .201 11.2 Opinions Regarding Reliable Sources of Information About Contraception . 203 11.3 Opinions Regarding Advantages and Disadvantages of the Pill and IUD .207 11.4 Opinions on Risks to Women's Health Due to Contraceptive Use .212 11.5 Opinions on Risks to Women's Health Due to Abortion .220 XII. REPRODUCTIVE HEALTH KNOWLEDGE AND ATTITUDES .225 (Adriana Galan, Silvia Gabriela Scintee, Adriana Vasile, Jay Friedman) 12.1 Ideal Family Size .225 12.2 Knowledge of the Menstrual Cycle .225 12.3 Knowledge of the Fertility Effect of Breastfeeding .230 12.4 Opinions About Abortion.233 12.5 Attitudes Toward Family and Reproductive Norms .240 XIII. HEALTH BEHAVIORS .249 (Fiorina Serbanescu, Jennifer Ballentine) 13.1 Cigarette Smoking .250 13.2 Alcohol Use.257 13.3 Prevalence of Routine Gynecologic Visits .261 13.4 Breast Self-Exam .265 13.5 Cervical Cancer Screening.267 13.6 Prevalence of Selected Health Problems .271 XIV. SEX EDUCATION.275 (Silvia Florescu, Jay Friedman, Valentina Mihaila, Dana Chitica, lulia Constantinescu) 14.1 Opinions about Sex Education In School .276 14.2 Discussions About Sex Education Topics with Parents .283 14.3 Sex Education Instruction in School.288 14.4 Most Important Source of Information About Sexual Matters .295 14.5 Impact on Knowledge About Fertility Issues and Contraception.298 XV. SEXUAL AND CONTRACEPTIVE EXPERIENCE OF YOUNG ADULTS .303 (Fiorina Serbanescu, Leo Morris, Jay Friedman) 15.1 Sexual Experience of Young Adults.303 15.2 Current Sexual Activity .312 15.3 Contraceptive Use at First Sexual Intercourse .315 15.4 Reasons for Not Using Contraception at Time of First Sexual Intercourse .321 15.5 Use of Contraception at Most Recent Sexual Intercourse .324 XVI. KNOWLEDGE AND EXPERIENCE OF SEXUALLY TRANSMITTED DISEASES . 327 (Fiorina Serbanescu, Leo Morris, Silvia Florescu) 16.1 Awareness of AIDS and Other STDs .330 16.2 Self-Reported STD Testing and Diagnosis.340 16.3 Self-Reported STD Symptoms .346 16.4 Number of Current Sexual Partners and Condom Use Patterns .348 16.5 Self Perceived Risk of STDs.352 XVII. KNOWLEDGE OF AIDS TRANSMISSION AND PREVENTION.355 (Mona Marin, Carmen Moga, Bogdan Pana) 17.1 Knowledge About HIV/AIDS .356 17.2 Knowledge About HIV/AIDS Transmission .360 17.3 Knowledge About HIV/AIDS Prevention . 366 17.4 Beliefs About Risk of HIV/AIDS and Self-Perceived Risk of HIV/AIDS …. 373 XVIII PHYSICAL AND SEXUAL ABUSE .385 (Fiorina Serbanescu, Leo Morris) 18.1 Comparative Findings on Domestic Violence in Eastern Europe . 386 18.2 History of Witnessing or Experiencing Parental Physical Abuse.387 18.3 Verbal, Physical and Sexual Abuse by a Partner or Ex-Partner . 389 18.4 Physical Consequences of Intimate Partner Violence . 399 18.5 Prevalence of Forced Sexual Intercourse . 405 REFERENCES . 409 ANNEX A: PILOT JUDETS . Al ANNEX B: SAMPLING ERROR ESTIMATES . Bl ANNEX C: INSTITUTIONS AND PERSONS INVOLVED IN 99RRHS . C1 SURVEY QUESTIONNAIRE (FEMALE). Ql Preface During the early 1990s Romania was faced with the reproductive health consequences of an aberrant pronatalist policy enforced for several decades by the Ceausescu's regime. Health policy makers tried to rapidly respond to these consequences by adopting new health strategies to reduce maternal and infant mortality. These strategies included development of the first national family planning program; introduction of new technologies in neonatal and maternal health services; implementation of active measurements to control the HIV/AIDS epidemic; and development of social programs for abandoned, institutionalized, and drug-using children and for domestic violence. Such a rapidly changing array of critical reproductive health issues could not have been documented and addressed with only the help of vital records. More information was needed to assess the reproductive health status of the Romanian population during a period of rapid change in health care that influenced the health of women and children. In 1993, the Romanian Ministry of Health, with technical assistance provided by the Division of Reproductive Health of the Centers for Disease Control and Prevention (DRH/CDC), conducted the first national population-based survey of women's reproductive health (93RRHS). The survey was designed to provide the Ministry of Health, international agencies, and nongovernmental organizations active in women's and children's health with essential information on fertility, women's reproductive practices, maternal care, maternal and child mortality, health behaviors, and attitudes toward selected reproductive health issues. The 93RRHS was instrumental in developing, evaluating, and fine-tuning the national family planning program and other reproductive health policies. In 1996, a representative sample survey of women and men aged 15-24 was implemented to document young adult's sex education, attitudes, sexual behavior and use of contraception. Such survey had never before been carried out in Eastern Europe. Survey results were used to plan effective information campaigns, policies and programs targeting young people, and to monitor and evaluate the impact of programs already in place. In 1999, a new nationwide reproductive health survey was designed and implemented in Romania (99RRHS) using the same methodology to allow for the study of reproductive health trends among the women aged 15-44 and to document the reproductive health of men aged 15-49. The surveys employed two separate probability samples to allow independent estimates for males and females. This final report improves the already impressive contribution of the previous two studies because: a) documents reproductive health aspects among both women and men of reproductive age (men were selected from different households than women); and b) by oversampling three target i judet (Constanta, Iasi and Cluj) documents the impact of region-wide interventions, implemented with USAID support, that consists of the establishment of modern women's health clinics, training of health professionals, development of IEC messages, social marketing, and provision of high- quality contraceptive supplies. In conclusion, the results of these large nationwide cross-sectional studies implemented in 1993 (sample size of 4,861 women aged 15-44), 1996 (sample size of 2025 women and 2047 men aged 15-24), and 1999 (sample size of 6,888 women aged 15-44 and 2,434 men aged 15-49), allow for generalizing the results to the entire reproductive age population of Romania. Although the surveys did not interview the same households, by applying similar questionnaires, the same sampling and field work methodology, they allow for a) a longitudinal examination of reproductive health issues among women, b) a detailed image of specific aspects of reproductive and sexual behaviors among men and c) a programmatic evaluation of reproductive health services in three regions. The results presented in this report should inspire as to several reflections: the impact of consensual unions on reproductive and contraceptive behaviors, the levels of induced abortion as an indication of failed family planning efforts, the role of pregnancy intervals and pregnancy intendedness in achieving the desired family size, the reproductive health differentials among various subgroups, particularly the differences between urban and rural population, the need for integrated family planning services and personnel training. I cannot end this preface without thanking the organizations and individuals who help design, implement and analyze this study—the Division of Reproductive Health of the Centers for Disease Control and Prevention (DRH/CDC)—to provide funding—the United States Agency for International Development (USAID), the United Nations Population Fund (UNFPA), the United Nations Children's Fund (UNICEF)—and to carry out the field work activities—the Romanian Association of Public Health and Health Management (ARSPMS). Prof. Dr. Dan Enachescu 99RRHS National Survey Director School of Public Health University of Medicine and Pharmacy, Bucharest, Romania ii Acknowledgments The 1999 Romanian Reproductive Health Survey (99RRHS) was conducted by the Romanian Association of Public Health and Health Management (ARSPMS) in collaboration with the National Commission for Statistics (CNS) and the School of Public Health, University of Medicine and Pharmacy "Carol Davila". Technical assistance in survey design, sampling, questionnaire development, training, data processing, and report writing was provided by the Division of Reproductive Health (DRH) of the United States Centers for Disease Control and Prevention (CDC). Principal investigators of this study were Prof. Dr. Dan Enachescu, national director of the 99RRHS, and Dr. Fiorina Serbanescu and Dr. Leo Morris DRH/CDC principal investigators. Most of the funding for the 99RRHS was provided by the United States Agency for International Development (US AID PASA DPE-3038-X-HC-1015-00), the United Nations Population Fund, and the United Nations Children's Fund (UNICEF). We wish to thank the 6,888 women and 2,434 men who made such a major contribution to our knowledge on women's and men's reproductive health in Romania by their participation in the 99RRHS. We thank our dedicated interviewers and supervisors for their commitment and discipline. Many thanks are extended to the survey headquarters team—Prof. Dr. Dan Enachescu, Scientific Survey Director, Aurora Dragomiristeanu and Mona Marin, Executive Survey Directors, and Silvia Florescu, Project Manager, Lucia Branga and Bogdan Barta, Field Work Coordinators, Dr. Carmen Cruceanu, Training Consultant, Doina Apostol, Data Entry Supervisor, Victor Dinculescu, President of the National Commission for Statistics (NCS), Radu Halus, Senior Advisor or NCS, Doina Gheorghe, Sampling Consultant. We also appreciate Dr. Mihai Horga, Director, Department of Woman's and Child Care of the Romanian Ministry of Health, who reviewed several chapters of the report. Also, many thanks to Adriana Galan and Gabriela Scintee for their assistance in preparing the Romanian translation and to Rose Pecorraro, DRH/CDC Graphics, for her contribution to the cover design. Special thanks are also extended to the USAID staff in Romania—Susan Monaghan, Senior Health Adviser and Randal Thompson, General Development Officer—and to the UNFPA Romania—Elin Ranneberg-Nilsen, UNFPA Representative and Country Director, Rodica Furnica, Program Officer, and Camelia Ieremia, Financial Administrator—for their assistance in design, planning and financial management. Many thanks to Mary Ann Micka, Mary Jo Lazear, and Willa Pressman, USAID/Washington, for their continued support of the survey. iii iv CHAPTER I INTRODUCTION The status of women's health in Romania is strongly influenced by cultural, historical and socioeconomic factors. The pronatalist policies of the Ceausescu regime (1964-1989) had a particularly profound impact on women and their reproductive health. During that period, among European countries Romania had one of the highest rates of infant mortality and the highest maternal mortality rate, over 80% of which was attributable to unsafe abortion. In 1989, similar to other former communist countries in Eastern Europe, Romania experienced a major change in government and entered a long period of transition in which major reforms of different sectors have been planned and carried out. During the early 1990s Romania experienced major socioeconomic and political changes, including access to legal abortion services and contraception. Because use of modern contraceptives has remained low, abortion has played a considerably larger role than contraception in fertility control. Induced abortion reached unprecedented high rates in the early 1990s, surpassing the abortion estimates reported by any other country in the region, including those reported from Russia. In 1996, Romania continued to report the highest rates of abortion in Europe and twice as many abortions as live births among women aged 15 to 44. Despite the progress made during the last decade, the legacy of the past, compounded by the present lack of resources, continues to place Romania far behind other European countries in family planning and reproductive health services. However, many changes have occurred in Romania since the 93RRHS was conducted, including expansion of public and private family planning services and dissemination of educational materials. In 2000, contraceptive consultations were for the first time included on the list of activities financed by the health insurance system, and contraceptives were procured using government funds and distributed throughout the network of family planning clinics. A new system of dispensing these contraceptives was put in place, allowing several categories of disadvantaged women to have access to free contraceptives and, in the same time, creating a revolving fund for contraceptive procurement at local levels. Since 1991, US AID has worked with the Romanian government, other international agencies, and local NGOs to increase access to family planning. Most of the work focused on designing client-centered family planning and reproductive health policies and programs, training physicians and other medical professionals, organizing public information campaigns, and developing a nationwide system for delivery of contraceptive supplies. As a direct result of these and other efforts 1 by USAID, the unmet need for family planning services among Romanian couples has gradually declined. A nationwide reproductive health survey conducted in Romania in 1993 (93RRHS), the first survey of reproductive age women since the 1978 World Fertility Survey, showed that the use of modern contraceptives was very low and reliance on traditional methods, which are prone to high failure rates and subsequent unintended pregnancies, was high. Women reported frequent use of traditional methods (withdrawal, calendar), high rates of abortion, general lack of awareness and poor quality of information about reproductive health issues, and a high level of mistrust of some modern contraception methods. However, many changes have occurred in Romania since the 93RRHS was conducted, including dissemination of educational materials and expansion of public and private family planning services. To assess the impact of new programs and provide planning data for upcoming women's reproductive health projects and information, education, and communication campaigns, the United States Agency For International Development (USAID) and other international donors sponsored two additional national reproductive health surveys in Romania: a Young Adult Reproductive Health Survey conducted in 1996 (96YARHS) and the 1999 Romanian Reproductive Health Survey (99RRHS); both are similar to the 93RRHS in design and content, but they also include a sample of men. These surveys were specifically designed to meet the following objectives: -to assess the current situation in Romania concerning fertility, abortion, contraception and various other reproductive health issues; -to enable policy makers, program managers, and researchers to evaluate and improve existing programs and to develop new strategies (a good example is use of 96YARHS data to provide data needed to develop sex education and health promotion programs); -to measure changes in fertility and contraceptive prevalence rates and study factors that affect these changes, such as geographic and socio-demographic factors, breast-feeding patterns, use of induced abortion, and availability of family planning services; -to measure the impact of public and private sector services over the past 6 years. -to obtain data about knowledge, attitudes, and behavior of young adults 15-24 years of age; -to provide data on the level of knowledge about transmission and prevention of AIDS; -to identify high-risk groups and focus additional reproductive health studies toward them. 2 In addition, the 99RRHS provides judet (county)-specific information for three USAID priority judets, Cluj, Constanta, and Iasi, which were oversampled to provide baseline data for project activities in these areas. In all three surveys, the questionnaire covered a broad array of reproductive health topics, including a pregnancy history, abortion, childbearing, contraceptive use, maternal and child health, health behaviors and attitudes. These surveys had a similar design and methodology; however, in 1996 and 1999, the surveys employed two separate probability samples to allow independent estimates for males and females. The interviews were conducted by trained interviewers in a face-to-face manner at the homes of randomly selected respondents; households were selected by a multi-stage cluster design using Census enumeration districts as the sampling frame. In all three instances response rates among women were high: 92% in 1993, 93% in 1996, and 90% in 1999. The response rate was slightly lower for male respondents (87%). The Division of Reproductive Health (DRH), Centers for Disease Control and Prevention (CDC), Atlanta, Georgia was responsible for coordinating survey activities for all three surveys and provided technical assistance to the Romanian counterpart. For the 99RRHS, the Romanian Association of Public Health and Health Management (ARSPMS) was the counterpart. 3 4 CHAPTER II METHODOLOGY 2.1 Sampling Design The 99RRHS was designed to collect information from a representative sample of women and men of reproductive age throughout Romania. Respondents were selected from the universe of all females aged 15-44 years and all males aged 15-49 years, regardless of marital status, who were living in Romania when the survey was conducted. The desired sample for females was 6,500, including an oversample of women in the three US AID priority judets (Cluj, Constanta, and Iasi). The desired sample size for males was 2,500. The female and male samples were selected independently. The survey used a three-stage sampling design, which allows independent estimates for the female and male samples. An updated master sampling frame (EMZOT), based on the 1992 census enumeration areas, was used as the sampling frame (National Commission for Statistics, 1996). The EMZOT master sample represents 3% of the population in each judet. In the female sample, the US AID priority judets were oversampled in both urban and rural areas to allow for independent estimates with adequate precision for women's health behaviors in these judets. Except for the three oversampled judets (in which all available census sectors in the sample were retained), the first stage of the sample design was a selection of census sectors with probability proportional to the number of households recorded in the EMZOT. This step was accomplished by using a systematic sample with a random start for the female sample. A 50% subsample of the census sectors selected in the female sample (not including the oversample in the priority judets) constituted the first stage of the male sample. Thus, the first-stage selection included 317 sectors for the female sample and 128 sectors for the male sample. In the second stage of sampling, clusters of households were randomly selected in each census sector chosen in the first stage (separate households were selected for the female and male samples). Finally, in each of the households in the female sample, one woman aged 15—44 years was selected at random for interviewing and in the male sample one man aged 15-49 years was randomly selected in each household. Because only one woman was selected from each household with women of reproductive age, and one male was selected from households with men of reproductive age, all results have been weighted to compensate for the fact that some households included more than one eligible female or male respondent. Survey results were also weighted to adjust for oversampling of households in the three US AID priority judets, and two more weights were added to adjust for non-response and 5 for urban-rural distribution of the population (see below). Except for Tables 2.1A and 2. 1B, all tables in this report present weighted results. Table 2.2, however, presents results weighted only for oversampling of households in the pilot judets and for selection of a single respondent per household. All other results were also adjusted for non-response and for unequal urban-rural distribution. The unweighted number of cases, used for variance estimation, are shown in each table. Cluster size was determined based on the number of households required to obtain an average of 20 completed interviews per cluster. The number of households in each cluster took into account estimates of unoccupied households, average number of women aged 15-44 per household (men aged 15-49 for the male sample), the interview of only one respondent per household, and an estimated response rate of 90% in urban areas and 92% in rural areas for women and of 85% overall for men. Cluster size was determined to be 51 households in urban areas and 59 households in rural areas for the female sample and 49 and 55 households, respectively, for the male sample. 2.2 Data Collection Data collection for the 99RRHS was carried out by 30 female interviewers for the women's sample and 10 male interviewers for the male sample, most of whom had experience conducting interviews in other household surveys, including the two previous national reproductive health surveys. Fieldwork was managed by staff of the Romanian Association of Public Health and Management (ARSPMS). Interviewer training was managed by the ARSPMS, with the involvement of Prof. Dan Enachescu and Dr. Aurora Dragomiristeanu, ARSPMS survey directors; Dr. Silvia Florescu and Dr. Mona Marin, ARSPMS survey deputy directors; Dr. Carmen Cruceanu, training expert; and the Centers for Disease Control and Prevention team (Dr. Fiorina Serbanescu, medical epidemiologist; Dr. Leo Morris, demographer; and Jay Friedman, program analyst) who had also provided technical assistance in the other two Romanian reproductive health surveys. Interviewer training took place at the ARSPMS headquarters just before data collection began and consisted of one week of classroom training in fieldwork procedures and proper administration of the questionnaire, and one week of practical training in the field with close monitoring by the trainers. At the end of training, six female teams and two male teams were selected, each consisting of four interviewers and one supervisor. The overall fieldwork implementation was supervised by two fieldwork coordinators (Lucia Branga and Dr. Barta Bogdan). Fieldwork lasted from July through October 1999. Each team was assigned to visit a number of census sectors in all regions of the country. Interviews were conducted at the homes of respondents and lasted an average of 60 minutes for women and 32 minutes for men. Completed questionnaires were reviewed in the field by team supervisors, then taken by the fieldwork coordinators to the Romania's National Commission of Statistics (CNS) headquarters for data processing. 6 2.3 Response Rates Of the 17,349 households selected in the female sample and 6,310 households selected in the male sample, 7,645 and 2,812 included at least one eligible respondent (a woman aged 15-44 or a man aged 15-49). Of these, 6,888 women and 2,434 men were successfully interviewed, yielding response rates of 90% and 87%, respectively (Tables 2.1 A and 2.1B). As many as four visits were placed to each household with eligible respondents who were not at home during the initial household approach. 7 Almost all respondents who were selected to participate and who could be reached agreed to be interviewed. Only 2% of respondents (regardless of gender) refused to be interviewed, and 7% of women and 11% of men could not be located. Response rates were not significantly different by residence, except for Bucharest, where the participation rate was slightly lower. Even though the overall response rate was similar in urban and rural areas, eligible respondents in urban areas were somewhat more likely to refuse to be interviewed; in rural areas eligible respondents were more likely to not be found at home. 8 2.4 Sampling Weights Because the sample design is not self-weighting, it was necessary to weight observations for data analysis. The initial two of four weights represented the differential selection of households in each judet and the selection of one eligible respondent per household. The number of households in each judet selected in the survey was compared with the number of households per judet estimated in the 1998 census projections (National Commission for Statistics, 1999). The ratio between the estimate of all existent households and sample-selected households in each judet represents the judet-specific household expansion factor. This factor was later normalized so that the sum of the household weights was equal to the total number of households in the sample. The variation in this weight reflects variation in the selection probability of the households within each judet. These two weights (judet-specific household weight and one-respondent selection weight) were used to compare demographic characteristics of respondents with completed interviews with the 1999 population projections for Romania by age group, sex, and place of residence (Table 2.2). The age distribution of the 99RRHS sample closely reflected that of the female population as a whole but overrepresented adolescent women (15-19 years old) and underrepresented women aged 25-29 years residing in urban areas. The age distribution of the male sample overrepresented adolescent men (15-19 years old) in urban areas and men 45-49 years of age in both urban and rural areas, whereas 20- to 24-year-old men in rural areas and 35- to 44-year-old men in urban areas were underrepresented (National Commission for Statistics, 1999). If percent distributions of the respondents in the two samples were calculated by urban-rural residence (not shown), both the female and male samples underrepresented the urban population of reproductive age according to the 1999 population projections estimated by CNS. Thus, an adjustment factor for non-response and a post-survey adjustment for the urban-rural distribution of the population were added to the two weights mentioned above. The non-response adjustment weight was based on information gathered in the household questionnaire concerning background characteristics (residence, age, education and marital status) of those who refused to participate or who could not be found in up to four separate household visits. Because information on education and marital status was not available for a substantial number of potential respondents, non-response rates we calculated by taking into account only age, sex, and residence. The non-response weight was the ratio of the proportion of all respondents selected in the sample and grouped in sub-classes (composed of five-year age groups and residence for each sample) to the proportion of respondents in each sub-class who completed interviews. The post-survey adjustment for the urban-rural distribution was based on the 1999 CNS projections of the Romanian population by sex, age, and residence (National Commission for 9 Statistics, 1999). For each sub-class, the post-survey adjustment factor was the ratio of the known national value to the sample estimate of that value. Thus, the final survey weight is the product of four weights: a household weight, a one- respondent-per-household weight, a non-response weight, and a post-stratification weight. Beginning with Table 3.2.1 A, all survey results are based on this final weight. 10 CHAPTER III CHARACTERISTICS OF THE SAMPLE 3.1 Household Characteristics Similar to the definition used in the 93RRHS and 96YARHS, a household was defined as a person or group of persons who shared the dwelling and the household expenses. Visitors were not counted in the household composition and were not included in the number of eligible respondents. After all eligible respondents in the household were listed, only one woman aged 15-44 years (in the female sample) or man aged 15-49 years (in the male sample) was randomly selected for the individual interview. Table 3.1.1 presents the percent distribution and average number of persons per household for households which contain at least one eligible respondent. Most of the households with eligible respondents (60% in the female sample, 64% in the male sample) had three or four persons. 11 One- or two-person households (presumably childless couples) were not common (12% and 17%, respectively); these types of households are more frequent in urban areas (17% and 20%) than in rural areas (7% and 14%). In Bucharest one in five households with eligible respondents contained only one or two persons (data not shown). Overall, households with six or more persons were also uncommon; they were the least prevalent in urban areas (6%) and the most prevalent in rural areas (21%). A typical household containing an eligible respondent was composed of almost four persons. Households in urban areas contained fewer persons (3.6 per household) than did rural households (4.3 per household). The larger household size in rural areas can be partially explained by higher fertility levels (see Chapter TV). The mean household size was lowest in Bucharest (data not shown), where a higher proportion of women or men of reproductive age lived in single households and 12 fertility was the lowest in the country (TFR=1.0 child per woman). Socio-economic well-being is an important determinant of reproductive health status. The 99RRHS collected various information on household amenities (electricity, flush toilet, telephone line, and central heat) and ownership of various goods or properties (television, refrigerator, private car, video recorder, mobile phone, vacation home, and vegetable garden or orchard or vineyard). Response options to each of these items were "yes" and "no". In addition, information on the average number of hours of electricity per day and on household crowding were obtained for each respondent. Crowding was determined by the total number of persons living in the household divided by the total number of rooms in the house (not including the kitchen or bathroom) being 13 greater than one; respondents were classified as living in crowded conditions (more than one person per room) or not living in crowded conditions (one or fewer person per room). In 1999 virtually all households in Romania were supplied with electricity 24 hours per day (Tables 3.1.2A and 3.1.2B and Figure 3.1.1). On average, almost two thirds of respondents lived in households with flush toilets and about half had a telephone line and central heating at home. The proportion of households with such amenities varied significantly by residence. For example, urban women were 13 times as likely as rural residents to have central heating, 5 times as likely to have flush toilets, and 3 times as likely to have a telephone. These differences were slightly narrower in the male sample. Bucharest had by far the highest prevalence of households with basic amenities: the majority of households had flush toilets, more than two thirds had central heating, and telephone coverage was the highest in the country (65%-67%). In Moldova and Vallahia regions, more rural than other regions, households were less likely to have flush toilet, central heating, or telephone coverage. Since 1993, households with telephone lines increased by 40% (from 38% to 53%), but there were no substantive changes in other household amenities. 14 As shown in Figure 3.1.2, among durable consumer goods, television was available in almost every household with women of reproductive age (95%), with higher coverage in urban areas (98%) than in rural areas (91%). Similarly, 96% of households with men of reproductive age had a television. Almost all households had refrigerators (88%-90%), especially in urban areas (96%- 97%) but less frequently in Moldova (78%). As expected, almost all households in rural areas with women and men of reproductive age had a vegetable garden, orchard, or vineyard (91% and 86%), whereas only one in three and one in four urban households, respectively, had such gardens. The proportion of households with reproductive-age women and men with automobiles was fairly low (41%o and 43%). Families living in Bucharest and other urban areas were more likely to own a car. Video recorders were not very widespread in Romania: one in three households in urban areas and 15%-20%) in rural areas owned a video recorder. Also, very few families owned a vacation home or a secondary residence (15%—21%); respondents in urban areas were significantly more likely than rural residents to own an additional residence. The use of mobile phones was low (14% of women and 13% of men reported they had one) and was concentrated in urban areas. Interestingly, they were owned mostly by households that also had a telephone line whereas only 6% of households without phone lines had a mobile phone (data not shown). 15 Level of household crowding is another important indicator of housing conditions. More than two thirds (69%) of reproductive-age women and 60% of reproductive-age men lived in crowded conditions. Crowding was not substantially different in urban households than in rural households, although the average number of persons per household was lower in urban areas than in rural areas. The most crowded households were in Moldova—71% of women and 66% of men reported living in households with more than one person per room. The least crowded households for women were in Bucharest (65%) whereas for men they were in Transylvania (55%). All of these household amenities and goods, including living in uncrowded conditions and having electricity 24 hours per day, were summed to create a score to classify the socio-economic status (SES) of each household. Equal values were assigned for possession of each amenity or good. For each household this inventory yielded a score whose reliability was assessed using the Cronbach coefficient alpha. On the basis of this initial evaluation only 10 items were selected for use in the SES score (alpha coefficients.72 for female sample and 0.70 for male sample); possession of a vegetable garden, orchard, or vineyard and having electricity 24 hours a day were not included in the final score. These items were excluded because the score, as in the 93RRHS, was based exclusively 16 on possession of items associated with higher SES (possession of a garden was inversely correlated with SES and 24-hour-per-day electricity in Romanian households was practically universal). Figure 3.1.3 shows the percent distribution of households by their SES score; the score ranged from 0 to 10, where 0 represented the lower end (no amenities and goods included in the score) and 10 represented the higher end (all 10 items included in the score). The score was further divided into terciles to create three levels for the socio-economic status variable. Respondents with a score of 0-3 amenities were classified as low SES; those with a score of 4-6 as middle SES; and those with a score of 7-10 as high SES. 3.2 Characteristics of Eligible Women and Men General characteristics of respondents with completed interviews, by residence, are shown in Tables 3.2.1A and 3.2.1B. It should be emphasized that all results presented in these tables and all the following tables in this report are weighted to adjust for sampling design, non-response, and unequal distribution between urban and rural respondents within each age group and gender (post- stratification weights), as described in the preceding section. Overall, 36% of the female sample and 32% of the male sample were young adults (15-24 years of age) (Tables 3.2.1A and 3.2.1B). The age distribution was slightly younger in rural areas, where young adults represented 41% of the women and 34% of the men, compared with 34% and 31% in urban areas. Romanians tended to be well educated, as evidenced by the fact that only 17% of female and 14% of male respondents did not have any secondary education (Figure 3.2.1). Most of the respondents who did not complete secondary education were older respondents or were very young respondents, still in secondary school. The proportion who had received formal education beyond the secondary level was similar between women and men of reproductive age (17% - 18%). Respondents residing in urban areas were more likely to be better educated than those in rural areas. The urban-rural difference was most pronounced at the postsecondary level, where women and men of reproductive age were three and four times more likely, respectively, to have completed technical college (23% vs. 7%) or university training (25% vs. 6%) than their rural counterparts. Romania is a low-fertility country, with a total fertility rate under the replacement level of two children per woman (1.2 births per woman in 1997-1998, according to CNS). In the 99RRHS, 39% of women were childless, 25% had only one child, 25% had two children, and 11% had three or more children. Fertility reported by male respondents was comparable to that reported by females except for a higher percentage of childless men (46%), which is consistent with a later age of marriage. Fertility was higher in rural areas, where only 34% of women were childless (compared 17 with 42%o in urban areas) and 16%. reported three or more children (twice as many as in urban areas; 8%). A slight majority of women and men were legally married (59% and 57%); additionally, a small proportion (6% of women and 3% of men) were in consensual unions (unregistered marriages or living with a partner "as husband and wife" but not legally married). Respondents in these two categories constitute currently married or in union respondents. Women residing in rural areas were somewhat more likely to be in a legal or consensual union (67%) than were women living in urban areas (63%), but this urban-rural difference in marital status was not evident among men. Divorce and separation appeared uncommon: only 7% of women and 4% of men reported that they were previously married. More than one of four women (29%) and more than one of three men (37%) had never been married or lived with a partner. The dominant religion among survey respondents was Eastern Orthodox (87% of women and 89% of men stated they belong to this religious denomination. Most other respondents were 18 19 20 Protestant (6% of women, 3% of men) or Catholics (4% for both); less than 1% of respondents said they had no religious affiliation. When asked their nationality, 87% of female and 90% of male respondents reported they were Romanian, 6% Hungarian, 4% of females and 5% of males Roma, and 1% other ethnic descent. More respondents of Hungarian or Roma ethnic background live in rural areas than in urban areas. Almost one third of respondents lived in households classified as low SES; almost half lived in middle-SES households, and one of four lived in a high-SES household. The urban-rural differentials in SES were striking for both female and male respondents; the percentage of respondents living in low-SES households was six times higher among rural residents than among urban residents. At the same time, only 5% of female and 9% of male rural residents were classified as living in high-SES households, but 35% and 38% of those living in urban areas were classified as high-SES. Slightly more than half (55%) of women and 43% of men reported that they did not work outside the house (even part time). Because of lower job availability, rural female and male respondents were less likely to work outside the house, which likely contributed to the urban-rural differences in SES. Important differences existed in marital experience between females and males and between urban and rural residents, regardless of gender. Almost one of two (46%) women aged 20-24 years, but only 17% of men, were legally married or in a consensual union and an additional 4% and 6% were previously married (Tables 3.2.2A and 3.2.2B). By age 34, the difference in marital experience between females and males tended to disappear; the proportion of women and men currently or ever married both increased to 94% and 90%. Women in urban areas were much more likely to postpone marriage, probably because they delayed marriage until after they completed their desired educational level; for example, 15% of rural women aged 15-19 and 64% of rural women aged 20-24 but only 7% and 40% of urban women had marital experience. Except for 20-29-year-olds, there were no significant differences in marital experience by residence among men. Marriage dissolution among older respondents was not significantly different in urban and rural areas. Comparison with the 93RRHS (data not shown) showed that fewer young adult women had ever been married (46% in 1993 vs. 58% in 1999), probably because of higher educational attainment among younger women than previously. This trend was noticeable among both urban and rural women but more obvious among urban residents. Table 3.2.3 presents the percent distribution of female and male respondents by the highest level of education attained, according to age and residence. Younger women were more likely than older women to have had a postsecondary education. Women in urban areas were much better educated (high school completed or higher education level) in each age group; 75% of women aged 20-24 years residing in urban areas, but 46% of those in rural areas had completed secondary school. Also, the proportion of women with a university education was three times higher in urban areas (44%) than in rural areas (14%). The urban-rural disparity in education showed a similar pattern 21 among older residents. Likewise, younger men tended to be better educated than older cohorts. Urban residents, regardless of their age, were almost twice as likely as rural residents to have completed high school or a higher level of education. Compared with the 93RRHS, in the 99RRHS fewer women reported only primary or lower education level (17% vs. 26%), whereas the proportion reporting postsecondary education rose (from 11% to 17%). 22 23 24 CHAPTER IV FERTILITY AND PREGNANCY EXPERIENCE One of the objectives of the 99RRHS was to assess the current levels and trends of reproductive behaviors and to identify factors that might change such behaviors. The findings presented here are particularly useful in assisting policy makers and program managers design programs that respond to the reproductive behavior of the Romanian reproductive-age population and to tailor programs to meet the needs of key subgroups. To obtain information about reproductive patterns, the survey questionnaire included a series of questions about marriage, divorce, sexual activity, contraceptive use, childbearing, induced abortion, infertility, desired family size, planning status of all pregnancies in the last five years, and information about prenatal care for all births during the past five years. Information about pregnancies and their outcomes (birth, abortion, fetal loss) was collected through a complete pregnancy history for each woman up to the time of the interview. This survey also collected information about male reproductive behavior in Romania, which included a history of live births, but not of female partners' pregnancies. These data included sexual activity status, current marital status, the numbers and dates of live births to each male respondent's female partners in the past five years, whether the children are still alive (and if not, their age at death), whether the pregnancy was intended by the male respondent, and whether the children live with the male respondent. 99RRHS data represent an important addition to vital statistics routinely compiled at the local and national level, because the survey included many background characteristics not included on birth certificates and abortion registries. In addition, the survey explored in depth the circumstances surrounding each abortion or birth within the past five years, documenting use of prenatal care and abortion services and the prevalence of pregnancy-related morbidity. 4.1 Fertility Levels Current fertility levels were estimated using age-specific fertility rates. The total fertility rate (TFR) was computed by accumulating the age-specific fertility rates and multiplying the sum by 25 five. The TFR is thus defined as the average number of births a woman would have during her reproductive lifetime (15-44 years of age) if she experienced the currently observed age-specific fertility rates. Numerators for the age-specific fertility rates were calculated by selecting births that occurred during the 36 months preceding the survey and grouping them (in five-year age groups) by the age of the mother at the time of pregnancy outcome (calculated from the mothers' reported date of birth). The denominators for the rates represent the number of woman-years lived in each specified five-year age group divided by those mothers during the three years preceding the survey. TFR for the three years preceding the survey (July 1996-June 1999) was 1.3 births per woman and the general fertility rate was 43.8 births per 1,000 women aged 15-44 years (data not shown), which was consistent with the recent fertility decline and the most recent vital statistics estimates. According to CNS data, TFR=1.2 births per woman aged 15-49 years in 1996-1998 and the general fertility rate was 40.6 births per 1,000 women this age (Comisia Nationala pentru Statistica, 1999). A comparative analysis of results from recent national reproductive health surveys in Eastern Europe and Newly Independent States (Romania, Czech Republic, Russia, Ukraine, 26 Georgia, and Moldova; Table 4.1.1) showed a fertility pattern in Romania similar to these other former Soviet-bloc countries (Goldberg et al., 1993; VCIOM and CDC, 1998,2000; KIIS and CDC, 2000; Serbanescu et al. 1994, 1998, 2000). All have low fertility rates that have declined substantially in recent years and high rates of induced abortion (see Chapter V). Similar to other countries in eastern Europe fertility in Romania exhibits an early peak in the age pattern, with the highest level among 20-24-year-old women, then among 25-29-year-olds (Table 4.1.2, Figure 4.1.1). Notably, fertility among adolescent women is the third highest (36 births per 1,000 women aged 15-19). As a result, 52% of the TFR was contributed by women aged 15-24 and 83% by women less than 30 years old. Women aged 35-39 and 40-44 years contributed minimally to total fertility; their age-specific fertility rates accounted for only 5% and 1% of overall fertility. Total fertility among married women was more than twice as high as for all women, which 27 implies that extramarital fertility plays a minor role in overall fertility. Young married women had much higher age-specific fertility rates than all young adult women (280 vs. 36 births per 1,000 women aged 15-19 years and 178 vs. 100 births per 1,000 women aged 20-24). The fertility rate of married women and that of all women differed little after 30 years or age, since almost all Romanian women had marital experience by that age (see Section 3 in this Chapter). The decline in fertility in Romania that began in the late 1980s (documented by the findings of the 93RRHS) appeared to continue in more recent years but at a slower pace; TFR decreased by almost 20% between the 93RRHS and 99RRHS, whereas three-year fertility rates calculated in 1993 showed a 30% decline between June 1987-May 1990 and June 1990-May 1993. The decline was almost entirely due to lower fertility among women aged 15-24 years; the fertility rate of women aged 25 and older remained basically unchanged. Fertility patterns, as documented through age-specific fertility rates, were consistent with the cumulative past fertility of women interviewed in the 99RRHS (calculated as the percent distribution 28 29 30 of women by number of live births and stratified by current age of each woman at the time of the interview). Overall, 39% of all women aged 15-44 were childless at the time of the interview, but only 14% of women currently in union had not had their first child (Table 4.1.3 A). Although few women reported birth before age 20, by age 29 69% of all women had given birth. Very few women remained childless at the end of their reproductive-age years (5%). Among currently married women, one of two adolescents have already had her first child, two of three 20-24 year-olds have given birth, and over 90% of women 30 years of age have had their first child. Only 3% remained childless by 40-44 years of age. A minority of women had three or more children (12% of all women and 17% of currently married women). Compared with women, a slightly higher proportion of all men aged 15-49 (46%) were childless at the time of the interview, which reflected the later age of marriage for men (Table 4.1.3B). At age 20-29, the highest fertility years for women, a considerably lower proportion of men report that they had fathered a live birth. Whereas by 24 years of age 32% of women had had a live birth, this was true of only 9% of men. The corresponding percentages by age 29 were 69% and 47%, respectively. Although these differences narrowed after age 30, in all age groups a lower proportion of all men compared with all women reported a live birth although, like women, very few men had not reported a live birth by age 49 (7%). Among men and women in union these differences are much less, indicating that fewer men not in union ever had a live birth or possibly were unaware of some that they fathered. Tables 4.1.3A and 4.1.3B also show an obvious two-child family size pattern with only a minority of women and men having three or more children: among all respondents of both genders and among those who were currently married, 10%—12% and 16%—17%, respectively, had three or more children. For men, the mean number of children reported by 45-49-year-olds was 2.1 (2.2 for men in union). 4.2 Fertility Differentials Table 4.2 shows the age-specific fertility rates and total fertility rates among different subgroups. Urban-rural residence is an important determinant of fertility. Women who lived in urban areas had, on average, almost one child less than rural women in the three-year period preceding the 99RRHS interview. All age-specific fertility rates were higher among rural residents; the differences are particularly important among younger women (15-19 and 20-24 years of age), whose age-specific fertility rates were three and two times higher, respectively, in rural areas then in urban areas. Women living in Bucharest reported the lowest level of fertility (1.0 birth per woman) whereas women living in Moldova—which had higher a percentage of reproductive age women living in rural areas than other regions—had the highest fertility rate (1.6 births per woman). Again, the most prominent differences in age-specific fertility rates by region were among young adults. In Bucharest, the fertility rates of young women were the lowest in the country (29 per 1,000 among 15-19-year-olds and 51 per 1,000 among 20-24-year-olds). 31 32 Fertility and education were strongly inversely correlated, with less educated women (primary education or less) reporting much higher fertility rates (Table 4.2). Women with only a primary education had, on average, 1.6 births more than women with a postsecondary education. Fertility differences according to education were very wide among younger women and diminished among older women. Socio-economic status (SES) was also inversely related to fertility level. Women with low SES had, on average, 2.0 births per woman and women with high SES had 0.7 births per woman. Among various ethnic groups, TFR and age-specific fertility rates among Roma young adults were the highest in the country. 4.3 Nuptiality Because most pregnancies occur among women and men who are married or in a consensual union, reproductive health behaviors are greatly influenced by marital status. At the time the 99RRHS was carried out, about two thirds (65%) of women aged 15-44 were currently married (59%) or living in a consensual union (6%) (Table 4.3A). The corresponding percentages for men were 57% and 3% (Table 4.3B). Seven percent of women and 4% of men were previously married (widowed, divorced, or separated) from a spouse, or a partner in a consensual union. More than one of four women (29%) and more than one of three men (37%) had never been married or lived with a partner. The proportion of all women who were currently married started at 5% among 15-19 year olds, increased rapidly to 37% among women aged 20-24 and to 72% among 25-29-year-olds; it reached a maximum of 83% for women aged 35-39. Consensual unions were slightly more prevalent among women in their twenties (7%-9%) than at other ages. Widowhood, divorce, and separation increased with age, peaking at one of eight women aged 40—44. The proportion of never-married women decreased abruptly with age, from 90% among 15-19-year-olds, to 50% among women 20-24 years of age, 13% among women aged 25-29, and 6% among women aged 30-34. Practically all 40-44-year-old women had ever been married. Until age 30, lower proportions of men than women were in a legal or consensual union. Only 17% of men aged 20-24 and 61% of men aged 25-29 were in a union. Consensual unions were slightly more prevalent among men 25-29 years of age (7%) than among men of other ages. As is the case for women, the proportion of never-married men decreased abruptly with age, from being almost universal among 15-19-year-olds to 81% among men aged 20-24, 32% among men aged 25-29, and 10% among men aged 30-34. Practically all 45-49-year-old men had ever been married or in a consensual union. The proportion of women married or in union was significantly lower among women with a postsecondary education (52%) than among women with a primary education (68%) or a secondary 33 34 35 education (64%-71%), which suggested that women tended to delay marriage until after they completed their education. Consensual unions were much more prevalent among women with a primary education (15%), those with a low SES (10%), and Roma women (31 %). Women who were employed at the time of the survey were more likely to have ever been in union than those who were not working, presumably because unemployed women were younger and wanted to delay marriage to complete their education. 4.4 Age at First Sexual Intercourse, Union, and Birth Age at first union and age at first sexual intercourse could play an important role in determining fertility. Delays in these events decrease the number of reproductive years that a woman spends at risk of getting pregnant and increase the likelihood of having fewer children. Age at first birth also has a direct impact on the overall fertility, since postponing the first birth may contribute to the decline of the total fertility rate. As shown in Figure 4.4, men reported a much younger age 36 37 of sexual initiation, whereas their age at first union was considerably higher than for female respondents. Information on age at first sexual intercourse, first union and first live birth for all women (Table 4.4A) and information about first sexual intercourse and first union for all men (Table 4.4B) are presented by age of the respondent at the time of interview. The left side of each table shows the proportion of respondents within each age cohort (five-year age group) who have ever had sexual intercourse (top panel), ever been in formal or consensual marriage (middle panel), and ever had a 38 live birth (bottom panel) before reaching specific ages. The overall median age (age by which 50% of women aged 15-44 or 50% or men aged 15-49 have experienced the event) and the median age within each age group are also displayed for each event. By comparing respondents categorized by their current ages it is possible to detect whether the age of occurrence of each event has been changing over time. For example, the proportion of women who had sexual intercourse before age 20 had increased from 49% among 40-44 year olds to 58% among 20-24 year-olds whereas the proportion who started their first union before age 20 had decreased from 39% to 30% between these two cohorts. In Romania, sexual abstinence among females before marriage was historically common but, as documented in the 93RRHS, young women aged 20-24 became increasingly sexually experienced prior to their first union, signaling a transition in sexual behaviors. Similar to the pattern documented in 1993, the 99RRHS showed a further disparity between older cohorts (e.g. 40-44 years old) and young cohorts (e.g. 20-24 year olds). Apparently, while traditional norms are weakening, the forces of modernization-urbanization, rising educational attainment, more exposure to the mass media, and changes in the status of women have altered every aspect of life, including the age patterns of sexual activity, marriage, and motherhood. In addition to a slight decrease in age at first intercourse (median age is half a year lower for 20-24 year olds than for 40-44 years old) between these two cohorts, age at first union had gradually increased from 20.7 to 22.1. Consequently, the median age at first birth was delayed by almost three years, from the median age of 22.3 among women aged 40 - 44 to a median age of 24.9 among those aged 20-24. These findings suggest that although more women became sexually experienced prior to their first union, premarital births were rare. Moreover, younger cohorts tended to wait longer than older cohorts to have their first child. For example, women aged 20-24 had their first birth on average 2.8 years later than their first union, whereas those aged 40-44 had their first birth on average 1.6 years later than their first union. However, among all reproductive-age women, 93% had already had their first union by age 30 but only 84% already had had their first live birth (data not shown). Similarly, data for men show a one-year decrease in the median age at first intercourse of men aged 20-24 compared with men aged 40-44, parallel with a delay in age at first union (Table 4.4B). Age at first union was typically much later for men than for women, but did not change that much over time among men cohorts. Median age at first union among the younger cohorts was less than a year later than for older cohorts (24.1 among men aged 35-44 compared to 24.5 among men aged 20-24). 39 4.5 Recent Sexual Activity Information about current sexual activity is crucial for estimating the proportion of women who are at risk of having an unintended pregnancy and therefore need contraceptive services. This information also suggests contraceptive methods that best suit the reproductive behavior and fertility preferences of population subgroups. Detailed information about the proportion of women in need of family planning services and their contraceptive choices is presented in Chapter IX. Overall, 82% of women aged 15-44 years and 90% of men aged 15 - 49 years who were interviewed in the 99RRHS reported they had ever had sexual intercourse (Table 4.5 and Figure 4.5). Not all women and men who were sexually experienced were currently sexually active (within the month preceding the interview), however: only 63% of all women reported sexual intercourse within the last month and 7% reported intercourse one to three months before the interview. The corresponding percentages for men were 71% and 12%. If respondents who had never had intercourse were excluded, 77% of sexually experienced women and 92% of sexually experienced men were currently sexually active (data not shown). 40 In Table 4.5 information on sexual activity status is also presented by marital status and by current age. Among women who were married or living with a partner, 86% reported having 41 intercourse at least once within the past month and 6% had intercourse two or three months previously. These women constituted the majority (87%) of those classified as currently sexually active (data not shown). Only 27% of previously married women were in a current sexual relationship; most (57%) had their last sexual intercourse more than three months ago. Conversely, although only 36% of never-married women had ever had sexual intercourse, more than three of four of those who were sexually experienced had their last sexual encounter within the preceding three months. Almost 4% of all women (slightly more among women currently in union) were either pregnant or in postpartum abstinence at the time of the interview. Of those men who were married or living with a partner, 95% reported having intercourse at least once within the past month and 4% had intercourse two or three months ago. Men in union constituted 71% of all men classified as currently sexually active. In addition, 73% of never-married men were sexually experienced and more than three quarters of them had their last sexual encounter within the past three months. Although only 53% of young adult women (aged 15-24 years) had had sexual intercourse, 67% of those who were sexually experienced reported their last sexual encounter within the past 30 days and 12% reported having had intercourse one to three months before the time of the interview. About 5% were currently pregnant or nursing. Among sexually active women aged 25 years or older, over 80% reported current sexual activity and 7%-8% had had intercourse one to three months ago. A greater proportion (71%) of young adult men than young women were sexually experienced and more than half of these young men reported their last sexual encounter was within the past 30 days; 29% had their last sexual encounter one to three months before the interview. (For more details on young adults, see Chapter XV). Among sexually active women aged 25 years or older, more than 90% reported sexual activity in the past one to three months. 4.6 Planning Status of the Last Pregnancy Similar to the 93RRHS, all 99RRHS respondents with pregnancies within the five years prior to the interview were asked about the planning status of every pregnancy ending in the five-year period preceding the survey. Each pregnancy was classified as either intended (wanted at the time it occurred), mistimed (occurring earlier than intended), unwanted (the respondent wanted no children or no more children), or not sure. Mistimed and unwanted pregnancies together constituted unintended pregnancies. As shown in Figure 4.6.1, more than two thirds of all pregnancies between 1990 and 1993 were unintended and the vast majority of them were unwanted. A similar pattern was documented in 1999. Although the prevalence of unintended pregnancies decreased by 12%, from 68% to 60%, unwanted conceptions continued to prevail. In both surveys, about five times as many conceptions were unwanted as mistimed. The most likely explanation for this pattern may be the early start and completion of childbearing, which leaves later pregnancies at risk of being unwanted rather than mistimed. Both the 93RRHS and the 99RRHS documented that almost 9 out of 10 42 pregnancies reported as mistimed or unwanted were aborted and only 6%-8% of unintended conceptions were carried out to term (data not shown). Data on pregnancy intendedness should be interpreted with caution. One common problem in collecting these data is that induced abortions are not always reported; abortion underreporting necessarily implies that unintended pregnancies will be underreported to the extent that abortions are underreported. Abortion underreporting does not appear to be a major concern in these surveys, however, because abortion rates calculated from both 93RRHS and 99RRHS exceeded recent officially reported levels. Another problem that might occur for pregnancies that end in live births is postpartum rationalization. Women are asked to report retrospectively their thoughts about the planning status of their pregnancies at conception. Some of them change their feelings after the child is born and may be reluctant to admit that it was an unintended pregnancy at conception. Therefore, the planning status of the last pregnancy almost certainly represents an underestimate of mistimed and, particularly, unwanted conceptions that ended in live births. Thus, data shown here represent conservative estimates of the true levels of unintended pregnancy. 43 Tables 4.6A and 4.6B present the percent distribution of women and men according to the reported planning status of the last pregnancy in the past five years, by selected characteristics. Despite the potential under-reporting of unintended conceptions, the figures show some important differences in the level of pregnancy intendedness among various subgroups. These data may underscore the need to address the risk of unintended pregnancy differently for various subgroups. Only 44% of women of childbearing age, regardless of their marital status, said their most recent pregnancy was intended at the time of conception, whereas 9% reported it as mistimed and 47% as unwanted. Thus, more than one of two women reported their last pregnancy as unintended and most of them (84%) reported that the unintended pregnancy was unwanted rather than mistimed. 44 45 As shown in Figure 4.6.2, a pregnancy's outcome and its planning status were strongly correlated. All but a small proportion of women whose last pregnancies resulted in a live birth said those births resulted from intended conceptions. By implication, very few unintended pregnancies (particularly unwanted ones) resulted in a live birth. Conversely, almost all women whose last pregnancy ended in induced abortion reported that the conception was unintended. It should be 46 noted that a relatively high proportion (21%) of women whose last pregnancy ended in miscarriage or stillbirth reported that it was an unwanted conception; this rate was almost six times the proportion of women with a live birth who reported an unwanted pregnancy. Although some of this difference may underscore the negative influence of unintendedness on pregnancy outcome, it is also plausible that some of these outcomes may have been induced abortions that were reported as spontaneous abortions or stillbirths. No consistent relationship was apparent between the educational level of respondents and the planning status of their pregnancies. Planning status of the last pregnancy varied by residence, however: women in urban areas were somewhat more likely than women in rural areas to experience an unwanted pregnancy (50% vs. 41%). In addition, young adults were less likely to report an unintended pregnancy (40%-42%) than women aged 25-29 (51%), 30-34 (70%), or 35-44 (83%). Among 15-19- and 20-24-year-olds, many unintended pregnancies were mistimed, with an unwanted-to-mistimed ratio of 1.4:1 and 2.0:1. This ratio increased abruptly after age 24 as spacing failure was replaced by the desire to terminate childbearing, from 5.0:1 among 25-29-year-olds to 22:1 among 29-34-year-olds; after age 34, virtually all unplanned pregnancies were unwanted. The same pattern was seen when the planning status of the last pregnancy was examined by the number of living children. Women with no or one child were less likely to report that their last pregnancy was unwanted than were women with two or more live births. Although young women and childless women reported slightly more mistimed pregnancies, the relatively high proportion of unwanted pregnancies among these subgroups may reflect poor understanding of the survey question, conflicting or ambivalent feelings about the last pregnancy, or indecision about childbearing. Pregnancy intendedness among men was measured differently: men were asked whether the pregnancy leading to the last live birth of their partner within the past five years was intended. Relatively few men (7%) reported that the pregnancy in question was unintended (Table 4.6B). This proportion was similar to the 12% of women reporting that their last live birth was unintended. As was the case for women, the proportion of unintended pregnancies was positively correlated with age and number of living children, but to a much less degree than among women. 4.7 Future Fertility Preferences One of the important factors that health care providers should consider in their efforts to help couples avoid unintended pregnancies, particularly those unwanted, is fertility expectations, which may vary among different subgroups. The preference among women for keeping family size small is reflected not only in low fertility and high abortion rates but also in the stated desires for additional children. In the 99RRHS, 47 only 22% of women currently in union said they intended to have a child in the future, including 8% who wanted a child right away (within one year) and 13% who wanted to wait at least one year before having a or another child (Table 4.7.1 A and Figure 4.7.1). Fifty-eight percent of women in legal or consensual marriage do not want to have any more children. An additional 5% were unsure if they wanted to have more, and a substantial proportion (16%) said they could not have any more children. The intention to have any (more) children decreased rapidly with increasing number of living children after a two-child family size was achieved. Among those with no living children, about two of three women wanted more children; this proportion dropped to one of three among one-child women and to less than 4% among women with two or more children. Of the vast majority of women who want more children, regardless of their parity, about half wanted to have another child(ren) within the next two years. Regardless of demographic or socioeconomic characteristics, most women had one or two children (see also Chapter HI) and had little desire to have more. 48 49 50 Younger women were much more likely than older women to want more children. The intention to have more children decreased from 70% among the youngest age group (15-19) to 44% for women aged 20-24,36% among women aged 25-29,16% among women aged 30-34, and fewer than 4% for women aged 35 and older. Among those who desired additional children, however, very young women were more likely than their older counterparts to want to wait one or more years to have a child. Only 24% of 15-19-year-olds wanted a child within a year, but 36%-37% of those aged 20-29 and 60% of those aged 30-34 did; very few women aged 35 and older wanted any more children, but most of those who did wanted to get pregnant within a year. These findings indicate that family planning programs should consider spacing methods for younger women and long-term or permanent birth limit methods for older women. Men wanted to limit their future fertility to a slightly lesser extent than women did: 13% of men in legal or consensual marriage but 8% of women want a child within one year; 52% of men but 58% of women did not want to have any more children (Table 4.7.1 .B). Fifteen percent of men want to wait at least one more year before having a child and an additional 5% were unsure if they wanted to have more. As is the case for women, a substantial proportion (14%) said they or their partners cannot have any more children. The intention of men to have any(more) children with 51 increasing number of living children, was similar to that of women. Among men with no living children (the top panel of the table) almost two of three wanted a child within the next two years; this proportion dropped to one of three among one-child men and to less than 10% among men with two or more children. Like women, of those men who wanted more children, regardless of their parity, almost half wanted to have another child(ren) within the next two years. Like women, most men, regardless of demographic or socioeconomic characteristics, had one or two children (see also Chapter III) and seemed to have little desire to have more than that. 52 Among fecund women in union in 1999, 69% reported that they wanted to have no more children, but 73% did so in 1993 (Table 4.7.2A and Figure 4.7.2). About half of those with only one living child (53%) wanted no more children. In both surveys, virtually all women (over 92%) were ready to end childbearing by the time they had two children. Rural women, women younger than 35 years of age, and women with less education were slightly less likely report that they wanted to end childbearing at lower parity levels. Among fecund men in union, 61 % reported that they wanted to have no more children (Table 4.7.2.B). By the time men had two children, most (over 84%) wanted no more. Rural men and men younger than 35 years of age were less likely to report that they want no (more) children at lower parity. 53 54 CHAPTER V INDUCED ABORTION In 1957, following the lead of the Soviet Union, Romania legalized abortion on request mainly to discourage unsafe abortion practices. For the next decade, abortion was the main method of fertility control; maternal mortality due to abortion was as low as 17 deaths per 100,000. In October 1966, however, Romania became the notable exception to the trend of readily available abortion in Eastern Europe when the Ceausescu government restricted access to abortion (Decree 770, October 1966) and contraception and instituted various surveillance measures (such as mandatory pelvic exams in the workplace and the presence of security police personnel in maternity hospitals) to ensure compliance. In an effort to stop the fertility decline, the government outlawed any means of modern contraception, introduced pronatalist incentives, and imposed harsh penalties on providers and seekers of induced abortions. Despite these penalties, the ban on family planning resulted in widespread use of illegal abortions, most of them self-induced or induced by lay persons. Romania soon became the country with the highest maternal mortality in Europe. The maternal mortality ratio (MMR) rose from 86 to 170 maternal deaths per 100,000 live births (Figure 5.0), which translated into almost 10,000 abortion-related maternal deaths between 1966 and 1989. In addition, it is estimated that at least 100,000 children were placed in state institutions by families who could not afford to raise them. Yet, in the long run, the pro-birth policy affected little the overall fertility rate. In December 1989, during the democratic revolution and after the fall of the Ceausescu government, the restrictions on abortion and contraception were repealed. The effect of replacing the use of illegal, unsafe abortions with modern contraception and legal abortions was dramatically reflected in the prompt decline in maternal mortality. After 23 years of high rates of maternal mortality, more than 85% abortion related, the MMR decreased in one year by more than 50% (from 170 to 84 deaths per 100,000 births) and has continued to decrease due to the decline in the abortion- related deaths (21 abortion-related deaths per 100,000 live births in 1997). In the early 1990s, women relied more on abortion than on contraception to control fertility. This situation has been attributed to health care providers' unfamiliarity with contraceptive methods after more than 20 years of medical isolation, to lack of available contraceptives, and to the legacy of public distrust of the medical establishment (as health care providers had been put in the roles of collaborators and informers for the previous regime). In such an atmosphere of distrust, women are more likely to seek an urgently needed procedure, such as abortion, than ongoing preventive care such as contraceptive services. 55 In the early 1990s, perhaps in response to the recent liberalization of the abortion laws, Romania had the highest abortion rate in Europe and probably in the world. Since then, official statistics indicate that abortion rates have gradually declined (from 182 abortions per 1,000 women aged 15-49 in 1990 to 93 per 1,000 in 1994 and to 47 per 1,000 in 1998); at the same time, the abortion-to-live-birth ratio decreased from about three abortions for each live birth in 1990 to 2.1:1 in 1994, and to 1.1:1 in 1998. 56 In recent years, Romanians continued to report widespread use of induced abortion; among other Eastern European countries and Newly Independent States with adequate population-based data, these rates are comparable with those in the Russian Federation and higher than most other former Soviet-block countries, but substantially lower than those reported by women in the Republic of Georgia, whose total induced abortion rate is more than 50% higher than in Romania (Table 5.0). 57 5.1 Induced Abortion Levels In the 99RRHS, the total abortion rate (similar to the total fertility rate describing the number of abortions a woman would have in her lifetime under the current age-specific abortion rates) was 70% higher than the total fertility rate (2.2 vs. 1.3). The ratio of induced abortions to live births was 1.6 abortions for each live birth, according to the survey estimates for the past three-year period (Figure 5.1.1). Because official data for 1999 were not yet available, annual abortion-to-live-birth ratios (data not shown) from the survey data were compared with official statistics for the most recent years (Center for Health Statistics and Information, Ministry of Health, 1999). Survey estimates were slightly higher than the official ratio published in 1997 (1.7 vs. 1.6) but 27% higher in 1998 (1.4 vs. 1.1). Age-specific induced abortion rates represent the proportion of women in a specific age group who terminated pregnancy by induced abortion instead of giving birth within the three-year period preceding the survey. These rates were calculated by using the age of the woman at the time of pregnancy termination. Similar to the fertility pattern, the age pattern of induced abortions in Romania is concentrated at younger ages. The highest rate occurred among women aged 25-29 58 (119 per 1,000), followed by rates of 105 per 1,000 among 30-34-year-olds and 101 per 1,000 among 20-24-year-olds (Table 5.1). These age groups alone constituted 76% of the total induced abortion rate. Although the abortion rate decreased with increasing age, abortion rates were higher than fertility rates for women over age 30. These findings suggest that Romanian women completed their desired family size at younger ages, after which most pregnancies were unintended and were intentionally terminated. The benefit of permanent methods of contraception for these women is obvious, but fewer than 3% of all women were using these methods (see Chapter VIII), indicating that an information campaign is needed to explain the advantages of permanent methods. Induced abortion rates for married women were higher than those for all women and, by implication, higher than those for unmarried women. Since most women are married by age 29, among women aged 30 years and above, abortion rates differed little between married women and all women. The difference between these rates was greatest for young adults (15-24 years of age). 59 Since 1993, the total induced abortion rate (TIAR) declined by one third, from 3.4 to 2.2 abortions per woman and from 4.4 to 3.1 among married women. This decrease was consistent with the increased use of modern contraception (see Chapter VIII). The decline was notable in all age groups but occurred mostly among women 20-34 years of age. Because age-specific abortion data from the official reporting system were not available for comparison, survey data on induced abortion were compared with the general abortion rate (number of abortions within a period of time to 1,000 women aged 15-49) published in the 1998 Annual Health Statistics Report (Center for Health Statistics and Information, Ministry of Health, 1999). The weighted sum of the age-specific abortion rates (Table 5.1) yielded a general abortion rate of 74 abortions per 1,000 women aged 15-44 during 1996-1999; this rate was 19% higher than the rate reported by official statistics (62 per 1,000 women aged 15-49 for 1996-1998). 5.2 Induced Abortion Differentials Abortion rates between urban and rural residents differed slightly (2.0 vs. 2.0 abortions per woman), but it was less pronounced than the urban-rural fertility difference (Table 5.2.1). Women residing in Bucharest and the southeastern part of the country (Vallahia) had almost two lifetime abortions more than Transylvania residents and one abortion more than Moldovan residents. The TIAR was inversely correlated with education level; on average, women with a primary education had a TIAR almost three times higher than women with a postsecondary education. The age-specific induced abortion rate (ASIAR) among adolescent women with only a primary education was strikingly higher than that for 15-19-year-olds with more education. Specifically, the ASIAR among adolescent women in the lowest education category was four to five times higher than women with any secondary education and 40 times higher than women with a postsecondary education. Likewise, induced abortion rates were twice as high among low-SES women than among women with high SES (2.9 vs. 1.5), and the difference was more pronounced among 15-19-year-olds. Recourse to abortion was much higher among Roma women (TIAR=4.6) regardless of the age group (103 abortions per 1,000 women aged 15-19,218/1,000 among women aged 20-24, 206 per 1,000 among women aged 25-29, 231 per 1,000 for women aged 30-34, 112 per 1,000 for women aged 35-40, and 49 per 1,000 for those aged 40-44 years). Despite the substantial decline in abortion levels in recent years, the current levels of induced abortion are proof that abortion continues to play a considerably larger role than contraception in reducing fertility. One means to reduce unintended pregnancies resulting in abortion is to provide comprehensive family planning services. Not surprisingly, a larger share of the potential demand is among subgroups of women who have also reported higher rates of induced abortion (rural women, those less educated, women with two or more children, Roma women) indicating that access to services is not equal and that the family planning program needs to expand its reach. 60 61 62 Meeting the unmet need for modern contraception will require a substantial increase in programmatic and financial support compared with current levels of effort (see Chapter XI). Table 5.2.2 shows that slightly more than one third (39%) of all women of reproductive age reported having had at least one induced abortion. The likelihood of having an abortion was positively associated with age (that is, as exposure to pregnancy, particularly unintended pregnancy, increased with age). Although very few adolescents reported induced abortion (2%), the percentage rose to 16% by ages 20-24 and to over 50% among 25-34-year-olds and women aged 35 and older (data not shown). The likelihood that a pregnancy would end in abortion varied directly with the number of living children, which was also correlated with age and a strong predictor of unintendedness, because Romanian women achieved their desired family size of one or two children fairly rapidly (see Chapter IV). It was also inversely correlated with education (from 39% and 43%, respectively, among less educated women to 31% among women with the highest education level). Having at least one abortion was more common among women living in Bucharest and Vallahia and among Roma women. As shown in Figure 5.2, the use of abortion was also heavily influenced by pregnancy order (pregnancy order refers to all prior pregnancies, including live births, induced abortions, miscarriages, or other outcomes). Women with no prior pregnancies were the least likely to have pregnancies ending in abortion (26%) and the most likely to have a live birth (63%). The likelihood of abortion increased rapidly if a woman had any prior pregnancies; from women with one prior pregnancy, whose likelihood of abortion was lower than that of having a live birth, to women with two prior pregnancies, who had an almost equal likelihood to resort to abortion or to keep the pregnancy, to women with three or more prior pregnancies, who were substantially more likely to end their pregnancies in abortion than in a live birth. Thus, the induced abortion to live birth ratio was directly correlated with pregnancy order, increasing from 0.4:1 among women with no prior pregnancy to about 1.0:1.0 among women with two prior pregnancies, to about 3.0:1.0 among women with four prior pregnancies, and 5.0:1.0 among those with five or more prior pregnancies. Because not all women were exposed to the risk of an unplanned pregnancy and subsequent abortion, in the right panel of Table 5.2.2 we restricted the denominator to include only women who ever had an abortion. Among these women, more than one of three (36%) reported they had only one abortion, 25% had two, 16% had three, and 12% four or five. One of 10 Romanian women reported six or more induced abortions in their lifetime. Women who reported multiple abortions were more likely to be older, less educated, of higher parity, have a low SES and Roma. 63 5.3 Abortion Services As is the case with all of the former Soviet-block countries, Romania was subject to the liberal abortion legislation and regulations issued by the former U.S.S.R. until 1966 when Ceausescu completely banned abortion on request (Decree 770 of Oct. 1966). In December 1999, when Decree 770 was repealed through a popular referendum, abortion was again made available on request within the first 12 weeks of gestation, as the former abortion legislation issued in 1957 was re- instated. Several additions and modifications were issued to introduce vacuum aspiration for early abortion, to permit induced abortion during the first 28 weeks of gestation on medical, genetic, judicial, and social grounds, and to regulate abortions performed by private practitioners. Under the current law, induced abortion, performed by either vacuum aspiration or sharp curettage, is performed either in government facilities as an inpatient procedure (with admission and discharge in the same day of the abortion procedure, if no complications occur) or in private 64 obstetric/gynecology clinics or cabinets (normally, as an ambulatory procedure). The 99RRHS collected information on the last four abortions performed since January 1994 in a detailed abortion history which included questions about the reason for abortion, place where the procedure was performed, payment, use of local or general anesthesia and antibiotics prescription, number of nights (if any) spent in the hospital after the procedure (abortion patients are released in the same day of the intervention if they do not have postabortion complications), and the presence or absence of early and late abortion complications. In an attempt to minimize recall bias, data were collected starting with the most recent procedure. Of 3,107 abortions reported to have occurred in 1994-1999, 95% were recorded in the abortion history. Almost all aborted pregnancies (89%) were reported to be terminated in the first trimester of gestation (data not shown). Women's reports on this issue are subject to several possible biases, however, including irregular menses, problems in recalling the event, and reluctance to admit abortions beyond the legal gestational limit. Almost two thirds of all abortions (64%) were reported to be performed between 7 and 9 weeks of gestation, 21% under 7 weeks, 3% at 10-12 weeks, and 11% were reported at 13 weeks or more (late abortion). Numbers are too small to draw any statistical conclusions, but late abortion was reported more often by rural women and Roma and were inversely correlated to a woman's education and socioeconomic status. By law, all abortions should be performed in hospitals or private clinics or cabinets by obstetric-gynecologists. As shown in Table 5.3.1, the vast majority of induced abortions reported since 1994 were performed in gynecological wards (64%) and about a third were performed in private clinics or cabinets, up from 11% in the 93RRHS. In the 99RRHS, abortions performed in the private sector were more prevalent in urban areas (42%) than in rural areas (26%). There is little variation in the place of abortion by region, but abortions performed in Moldova (more rural than other regions) were the least likely to be performed in the private sector. Private sector abortions increased with education and socioeconomic levels, and were more likely to be early abortions (less than 7 weeks). Roma women were the least likely to report induced abortions in the private sector, probably because of their lower socioeconomic status and their greater likelihood to obtain late abortions. Less than 1% of abortions were reported as being performed outside the health system. Since unsafe abortions (self-induced, performed by lay persons, or performed by doctors outside the health system) are illegal, it was likely that women were reluctant to admit these outcomes, in spite of the interviewer's assurance of anonymity. 65 66 67 Although abortion is an inpatient procedure, patients are generally released within the same day and do not have to spend the night in the hospital. Survey results showed that the majority of women (94%) who had an abortion since 1994 was released the same day of the abortion procedure (Table 5.3.2). Overall, 1% of women had to be hospitalized for one night, 1% for two or three nights, and 5% for four nights or more. The length of hospital stay varied with the woman's characteristics, gestational age, and presence or absence of abortion complications. Hospital stay was slightly longer in Bucharest than in other regions. For all women, abortion hospitalization was directly correlated with a woman's age and inversely correlated with her socioeconomic status. Hospitalization was directly correlated with gestational age, increasing from 5% for early abortions to 14% for abortions performed after 12 weeks of gestation, probably because of the type of abortion procedure and the risk of complications, which is strongly affected by gestational age. Almost half of the abortions with early complications required at least a one-night stay, and one third required hospitalization for four or more nights. At the time of the survey, abortion procedures officially cost about 65,000 lei (about US$:5.00) and were free of charge for women with four or more children. The average amount paid for an abortion was 92,000 lei, ranging from no payment to one million lei; the mean amount paid was higher than the official amount because about one third of abortions were performed in the private sector, where charges are not regulated and are usually higher than in governmental hospitals (Table 5.5.3). Furthermore, the dollar-lei exchange rate changed considerably in the past five years, the cost of living increased substantially, and the value of the local currency declined. Thus, the average abortion cost doubled, from 47,000 lei in 1994-1995 to 80,000 lei in 1996-1997 and 138,000, on average, in 1998-1999. Only 6% of abortions were performed at no charge; 30% of abortion payments were 60,000 lei or less, 18% were between 101,000 and 300,000 lei, and 2% were over 300,000 lei. For about one in four women (28%), the payment was a gift of unknown value or the amount paid was not recalled. Women in rural areas, those living in Transylvania or Moldova, older women, and Roman women (who also were more likely to have four or more children) were more likely to have had a free abortion or to pay less than other women. The cost of abortions performed in private clinics or offices was 40% higher than abortions performed in maternity wards and hospitals (78,000 lei vs. 118,000 lei); the gap between abortion payments in governmental and private facilities widened over time. Between 1994 and 1995 the mean abortion payment in the private sector was only 20% higher than the cost in the governmental sector (54,000 vs. 45,000 lei); in 1998-1999 the difference was over 50% (173,000 vs. 112,000 lei). 68 69 5.4 Reasons for Abortion More than half (53%) of abortions were performed for limiting childbearing, 30% for economic or social reasons (low income, unemployment, fear of losing their job), 11% for partner related reasons, including 6% of abortions to women who had out of wedlock pregnancies or were separated from their partners), 4% for medical reasons (pregnancy was threatening the woman's health), and 3% for birth defects (Table 5.4 and Figure 5.4). The use of abortion for limiting childbearing was mentioned slightly more often by rural women (who had a higher mean number of living children than urban women), women who resided in Vallahia, women over 34 years of age (who also had more children), and by Romanian and Hungarian women; this reason was positively correlated with pregnancy order, from 33% among first-order pregnancies to over 60% for third- or higher-order pregnancies. Socioeconomic reasons were reported by more than one third of women in Bucharest (36%), where life is more expensive and adequate housing is an increasing problem, and reported less by women with postsecondary 70 71 education and the highest SES. This reason was also claimed more often by women residing in Moldova (42%), and by Roma women (36%) and women of "other" ethnic groups. Partner-related reasons were more common among women who were not married by the time they got pregnant, and among those who were pregnant for the first time. Women with higher levels of education were slightly more likely to report partner's opposition to have another child than were women with less than complete secondary education. Health-related reasons were more often reported by residents of Bucharest and women of high SES. Similarly, the risk of birth defects was mentioned more often by urban women (including Bucharest) and residents of Transylvania region, and increased with education and SES. Almost 6% of first pregnancies were terminated because of known risk of fetal malformation. 5.5 Abortion Complications Induced abortions—even legal ones—are associated with a certain risk of postoperative complications, whose incidence and severity are strongly correlated with age of gestation, parity, woman's age, surgical procedure and operator's skills, type of anesthesia, and preexisting pathology (Henshaw, 1990). For example, abortions performed at 7-9 weeks of gestation have significantly fewer complications than those performed before 7 weeks and those performed at 10-14 weeks. Abortions performed by vacuum aspirations, with or without cervical dilatation, have fewer complications compared with the classic sharp curettage. First-trimester abortion complication rates from studies performed in developed countries ranges from 0.9 per 100 abortion procedures in the United States (Hakim-Elahi et al., 1990) to 6.1 per 100 in Denmark (Heisterbeerg and Kringlebach, 1989) but, in the absence of an international definition of abortion morbidity, comparisons between countries should be interpreted with caution. Survey estimates of postabortion complications are usually based on symptoms or conditions reported by respondents and may be less accurate than hospital based statistics. In Romania, 10% of all abortions performed since 1994 were followed by immediate complications (8%) or late sequelae (2%) (Table 5.5.1). This was consistent with the level of postabortion complications documented by the 93RRHS (7% and 2%, respectively). Rural women and those living in Moldova were slightly more likely to report postabortion complications. Early complications were slightly more prevalent among women with low education (data not shown) and low SES, and more prevalent among Roma women (13%) and among women with second- trimester abortions (13%). Abortions with early complications were more likely to be followed by late sequelae (at six months or more after the abortion was performed). 72 73 Most of the early complications involved severe or prolonged bleeding (66%), prolonged pelvic pain (59%), pelvic infection (45%), and high fever (43%); about seven percent of complicated abortions had perforations of the uterus and 10% were accompanied by other complications (Table 5.5.2). Except for uterine perforation, it is difficult to assess how serious the other early complications might have been. An indirect approach to measure severity is to consider early complications as serious when they required overnight hospitalization or were followed by late complications. Almost half of immediate complications (44%) required one or more nights of hospitalization and 14% were associated with late complications. The prevalence of early complications increased directly with gestational age. 74 CHAPTER VI MATERNAL AND INFANT HEALTH Maternal and infant mortality are measures of a nation's health and world-wide indicators of social well-being. As of 1998, the last year for which comparison data were available, Romania had the second-highest maternal mortality ratio (41 deaths per 100,000 live births) among Eastern European countries, after the Russian Federation (50 deaths per 100,000 live births) (World Health Organization, 1999). As of 1998, the infant mortality rate in Romania (20.5 infant deaths per 1,000 live births) ranked the second highest in Central and Eastern Europe (Population Reference Bureau, 2000). In Romania, women's access to perinatal care was free of charge for many years. Currently, under the new health care reform, it is included in the basic health care package. Perinatal care consists of three components: preconception care, prenatal care, and postnatal care. Preconception counseling and prenatal care are generally offered by primary care providers and consists of a wide array of information, including risks associated with pregnancies, health risk factors that can affect the development of the fetus (e.g., tobacco and alcohol), maternal infection (e.g., rubella, toxoplasma, HIV and other sexually transmitted diseases), risks associated with maternal health conditions, and risks associated with genetic conditions. In Romania, preconception counseling is offered only to young couples before marriage, without any follow-up before they plan to start childbearing. Preconception counseling is not provided during routine health care visits in spite of the essential role the primary care provider could play in modifying women's health behaviors (many healthy behaviors must be in place before pregnancy is recognized) and in identifying medical conditions that may require special attention during pregnancy. The use of timely and periodic prenatal care can effectively reduce perinatal mortality and morbidity. The Romanian Ministry of Health recommended number of prenatal care visits for women with uncomplicated pregnancies carried out to term is at least 10 prenatal visits. As part of comprehensive prenatal care, health risk assessment should include, in addition to the medical examination, an initial series of laboratory investigations (blood, urine, vaginal bacteriological exams, screening for sexually transmitted diseases and isoimmunization Rh) that will be repeated periodically. This chapter examines selected aspects of maternal and child care in Romania (e.g., sources of health care, utilization of maternal care services, quality of care), to identify subgroups with specific needs for care and to investigate maternal and child health outcomes that may be related to the availability and quality of maternity care services. All this information can be used to help direct or modify program interventions. 75 6.1 Prenatal Care This section describes the use of prenatal care for all pregnancies carried to term (either live births or still births) since January 1994. Women were asked in what week or month of gestation they had their first visit for prenatal care (not counting a visit that was just for a pregnancy test or just for the delivery) and the number of prenatal care visits during pregnancy. Of the 2,040 births during the five years prior to the 99RRHS, the majority of women (89%) had received some prenatal care but less than two thirds (60%) had received their first prenatal care visit in the first trimester (Table 6.1.1). Approximately one in four women had the first visit during the 2nd trimester and 4% during the third trimester. The level of any prenatal care within different subgroups varied sometimes by a considerable margin (between 70% and 95%). Rural women, residents of Vallahia, those who did not complete secondary education or had a low SES, Roma women, and women who had already had two or more births, were more likely to not have any prenatal care. Similarly, the percentage of infants whose mothers entered prenatal care in the first trimester varied widely, from a low of 45% to a high of 84%. Women living in urban areas were more likely to start prenatal care earlier than women in rural areas (68% vs. 58%). Early entry into prenatal care was higher among women living in Transylvania (71%) than women living in other regions, including Bucharest. The likelihood of early prenatal care was slightly higher among young adults (65%) than among older women. Early entry into prenatal care was highly correlated with mother's education; women who had not completed high school had a lower likelihood of initiating prenatal care early (53%) compared with women with higher education levels (69% and 77%). In addition, 16% of these women had reported no prenatal care, whereas only 5% of women with a postsecondary education had no prenatal care. Similarly, women with low SES had much lower likelihood of initiating prenatal care early. Among various ethnic groups, Hungarian women had the highest rates of early prenatal care (84%) and Roma women had the lowest rate (45%). Births preceded by one or two previous births (birth order three or higher) had the lowest rate of early prenatal care (45%). Low birth weight was positively correlated with prenatal care, probably because these pregnancies were more likely to be associated with complications during pregnancy which required close medical supervision (not shown). Prenatal care should not only start early but also should continue throughout pregnancy, according to recommended standards of periodicity. To assess the adequacy of prenatal care, it is necessary to monitor not only the time of first visit but also the number of prenatal care visits once care has begun. Overall, pregnancies ending in the five years prior to the survey averaged 5 prenatal visits, and ranged from 0 visits to 30 visits (data not shown). Among women with any prenatal care, the 76 77 average number of prenatal care visits was 5.8 visits, much lower than the Romanian Ministry of Health recommendation of 10 visits. About one in four women had only 1-3 visits (27%), the majority of women had 4—9 visits (50%), and only 10% of women had 10 or more prenatal care visits (Table 6.1.1). A small proportion of women (2%) stated they did not remember the number of prenatal care visits. Women who had 10 or more prenatal visits were generally the same women who started prenatal care early, since the number of visits was correlated with the month of initiation of care. As shown in Figure 6.1.1, use of prenatal health services did not improve between the 93RRHS and the 99RRHS. Instead, the proportion of women with no prenatal care has almost doubled (from 6% to 11%) and the proportion of women with 10 or more prenatal care visits decreased by more than 50% (from 23% to 10%). The proportion of women with early prenatal care entry (during the first trimester) did not change significantly between 1993 (57%) and 1999 (60%), however. These findings show that the majority of women began prenatal care during the first 12 weeks of pregnancy but the frequency of prenatal care visits was inadequate. The low prenatal care 78 attendance may be due to access barriers (e.g., distance, cost, waiting time, working hours of the health facility), competing demands on women's time, or miscommunication between clients and prenatal care providers. Compared with international standards, the majority of women do not meet the criteria of adequate prenatal care. In the United States, the adequacy of prenatal care is assessed by using the Adequacy of Prenatal Care Utilization Index (APNCU), also known as the Kotelchuck index. This index assesses the adequacy of initiation of prenatal care (month when prenatal care begins) combined with the adequacy of use of services (percentage of recommended visits received) once care has begun; this last component of the index is calculated by comparing actual use with the recommended number of visits (based on the American College of Obstetricians and Gynecologists recommendations), adjusted for the length of gestational period and the gestational age at initiation of care. These two dimensions are combined into a single index with four levels: inadequate, intermediate, adequate or adequate plus. Inadequate use is defined as either late prenatal care or less than 50% of recommended visits and includes also "no prenatal care." The three remaining levels require early initiation of care (by the fourth month of gestation). Intermediate care requires 50%- 79% of the recommended number of visits, adequate care 80%-109% and adequate plus 110% or more of the recommended visits (Kotelchuck, 1994). 79 80 By applying this index to data from the 99RRHS we found that only 12% of births within the past five years had received adequate or adequate plus care (Table 6.1.2 and Figure 6.1.2). About one in two women (53%) had received inadequate prenatal care. Inadequate prenatal care was more prevalent in rural areas (63%) than in urban areas (43%), in Vallahia (64%), and among women who did not complete a secondary education (62%), those with low SES (66%), Roma women (80%), and those who had two or more other births (68%). Targeting the groups that did not receive prenatal care in the first trimester or who had fewer than recommended visits can help improve both pregnancy and infant outcomes and help Romania lower perinatal mortality and morbidity. Prenatal care in Romania is provided mostly through primary health care centers (urban or rural dispensaries) and polyclinics (only in urban areas), ambulatory centers with specialized multidisciplinary care. Overall, in the 99RRHS, the principal source of prenatal care was a dispensary (50%). The second source of most prenatal visits was a hospital (23%), followed by a polyclinic (17%) (Table 6.1.3). About 1 in 10 women (11%) sought prenatal care in a private clinic or office. Generally, in dispensaries, primary care providers and midwives cover most of prenatal care, whereas in polyclinics and hospitals most care is provided by obstetricians. Dispensaries were the principal source for prenatal care for all pregnancies, irrespective of women's background characteristics, except in Bucharest, were most prenatal care was provided through polyclinics (41%). Between 93RRHS and 99RRHS, the role of polyclinics in providing prenatal care (implicitly the role of Ob/Gyns) had gradually declined; the proportion of respondents with recent births who mentioned a polyclinic as the source of prenatal care decreased to less than a third of the level in 1993 (55% vs. 17%). Under the new health reforms, pregnancy risk assessment at the beginning of prenatal care is increasingly performed by a primary health care provider (at the dispensary level), without a second Ob/Gyn opinion. If no pregnancy risk factors are identified, the general practitioner at the dispensary level will provide most of the care. Dissemination of health messages is an important component of prenatal care visits. In the absence of routine preconception care, the first prenatal visit is a critical opportunity to screen women for behavioral risk factors (e.g. tobacco and alcohol use), medical and genetic risks, and occupational risks and to provide comprehensive counseling. Counseling should include information about maternal behaviors and exposures that may affect the health of the fetus, nutrition, rest, and early signs and symptoms of pregnancy complications. In addition, near the time for delivery, counseling should prepare women for what they will face when giving birth, distribute accurate information regarding labor and delivery, and advise about techniques to reduce pain and anxiety during labor. Also, counseling about breastfeeding and family planning after birth should be initiated during the prenatal period and reinforced during postpartum care. Because the initiation and frequency of prenatal care visits evaluate only one dimension of the prenatal care (i.e., adequacy of use of services), the 99RRHS included additional questions aimed 81 82 83 at assessing information received during the prenatal visits (i.e., adequacy of content of prenatal care). Table 6.1.4 shows the percentage of pregnancies that received some information about specific educational topics during prenatal care. Overall, less than one in two women received some counseling about specific prenatal care topics. Information about nutrition, delivery, and breastfeeding were the most prevalent topics (all 46%-47%), followed by information about the negative effects of smoking and alcohol (44%-45%), about breastfeeding (43%), about postnatal care (36%), and early signs of complications during pregnancy (31%). Only one in four women received information about family planning after birth. Maternal characteristics that appeared to be associated with lower levels of counseling for most topics included rural residence, older age (over age 34), less than complete secondary education, Roma ethnic background, having less than seven prenatal visits, and receiving most of the prenatal care visits at a rural or urban dispensary. The proportion receiving information during prenatal care visits appeared highly correlated with the number of prenatal care visits (Figure 6.1.3). 84 85 Compared with results from the 93RRHS, in the 99RRHS fewer women stated that they received information on any of the specified topics during the prenatal care visits. For example, the prevalence of counseling on the harmful effects of smoking or alcohol use during pregnancy decreased from 61% in 1993 to about 44% in 1999, advice about nutrition during pregnancy from 61% to 47%, and advice about breastfeeding from 56% to 43%. Ultrasound imaging has been increasingly used in perinatal care but debate still exists about routine ultrasound screening. 99RRHS data do not allow differentiation between use for selected specific indications (e.g., confirmation of gestational age; assessment of fetal viability, fetal malformations, fetal growth, fetal presentation, and multiple pregnancy; examination of the placenta; assessment of amniotic fluid) or for routine screening, either during early pregnancy (16-20 weeks) or in late pregnancy (after 20 weeks). Table 6.1.5 shows the prevalence of ultrasound exams during pregnancies carried to term between 1994 and 1999. Overall, about one of two pregnancies had at least one ultrasound exam. Maternal characteristics associated with higher levels of ultrasound exams included: urban residence (67%), residence in Bucharest (73%) or in Transylvania (65%), postsecondary education (83%), high SES (87%), three or more prenatal care visits (82%), and most of prenatal visits in a private clinic (80%). Lower prevalence of ultrasound exams was associated with rural residence (40%), living in Moldova or Vallahia (40%-42%), low SES (36%), and having most prenatal care in an urban or rural dispensary (39%). Slightly more than a half of the ultrasound exams were performed for the first time in the second half of pregnancy, suggesting the use of ultrasound for specific indications rather than for screening (the main reason for starting screening in late pregnancy is to assess fetal growth and abnormal presentations or positions that may benefit from Caesarian delivery). Women in urban areas, those with high educational attainment, those with seven or more prenatal care visits, those who started prenatal care during the first trimester (data not shown), and those whose source of prenatal care was a private consultation clinic were slightly more likely than other women to have their first ultrasound exam during the first 20 weeks of pregnancy. 6.2 Intrapartum Care All births should occur in medical facilities where adequately trained personnel can monitor the progress of labor and delivery. The majority of deliveries in Romania take place in maternities or hospitals with inpatient obstetrical care. Births delivered outside medical facilities are rare and, in the event a home delivery occurs, both the mother and her baby are immediately referred to a hospital or maternity to be supervised for at least five postpartum days. 86 87 In the 99RRHS very few deliveries occurred outside the hospital (Table 6.2.1). The majority of women gave birth in a maternity ward or a hospital obstetrical ward (98%). Almost all other women delivered at home, and very few delivered in a private clinic or a birth house. Although home deliveries were rare, they reached a significant proportion among some subgroups. Home deliveries were relatively high among rural residents (4%), those with low levels of education or low SES (3% and 4%, respectively), those with two or more other births (6%), and those with no prenatal care (4%). Self-reports about onset and duration of labor are not very reliable because of wide individual variation in contraction frequency and in perception of uterine activity. There is often uncertainty about the beginning of labor, particularly of the latent phase. Although the 99RRHS included questions about the duration of labor (defined as the interval between the beginning of periodic contractions every five minutes or less and the time of delivery), respondents' reports were rather at the lower limit for both nulliparous (6.3 hours, on average) and multiparous women (4.7 hours, on average); according to data published in the literature the average duration of labor is 10 hours (for nulliparous women) and 6 hours for multiparous women (Duig, 1975). Because of the limitations of self-reported duration of labor, this report includes data on the duration of the hospital stay prior to delivery as a proxy for the labor duration. Table 6.2.2 shows the time spent in a medical facility prior to delivery and the length of stay after delivery. The average time spent in a medical facility prior to delivery was 12 hours (ranging from less than 1 hour to four days). More than a half of the respondents (53%) were admitted to the hospital only six hours prior to delivery (data not shown), presumably after the onset of labor. The average time spent in the hospital prior to delivery was at least four hours shorter for parous women with at least two prior births and women with no prenatal care. About half of women who gave birth in a medical facility were discharged in the first five days (52%), including 27% who were discharged after four days or less. One in three women (33%) was discharged after six or seven days. Very few women had to stay eight or more days after delivery (15%). Women in rural areas spent, on average, more time in hospital after delivery (data not shown) and were slightly more likely than urban women to be hospitalized for eight days or more (17% vs. 13%). Conversely, residents of Bucharest were the least likely to be discharged after eight or more days (10%) and the most likely to stay four days or less (not shown). As expected, women with low birth weight babies, women with early postpartum complications (data not shown), and those with Caesarian-sections had much longer stays than other new mothers. Almost two thirds (61%) of births delivered in medical facilities were assisted by obstetricians. Deliveries to rural women (51%), women residing in the Moldova region (48%), or those with no or late prenatal care (49% and 45%) were less likely to be assisted by an Ob/Gyn. Conversely, almost all births delivered in Bucharest (90%) were assisted by an Ob/Gyn (data not shown). 88 89 90 The Caesarean section (C-section) rate varies considerably among countries, from about 5% to more than 20% of all deliveries. The optimal rate is not known, but little improvement in birth outcomes has been demonstrated if the rate is higher than 7%. In addition to unequivocal obstetrical indications, C-section is often performed in less clear situations (e.g., prolonged labor) and often if a previous C-section was performed, which is rarely an adequate indication by itself. In Romania, the overall prevalence of C-section deliveries among all deliveries between 1994 and 1999 was 11% (Table 6.2.3). Women residing in urban areas were twice as likely to have this type of delivery as women residing in rural areas. Young adult women and women aged 35 years or older reported C-section rates higher than women aged 25-34 (15% and 17% vs. 9%). The C- section rate increased directly with education and SES, suggesting that financial considerations may sometimes be more important than obstetrical indications for C-section delivery. Women who experienced prolonged labor were more likely to deliver C-section than were women with uncomplicated pregnancies. Births with labor duration of more than 20 hours (more than 14 hours for multiparous women) had an almost five-fold increase in the rate of delivery by C-section than births with shorter duration of labor. The majority of C-sections were performed prior to the onset 91 of labor, however. Three fourths of women with C-section deliveries had the intervention performed prior to the beginning of labor while 22% were performed for deliveries with prolonged labors and five percent for deliveries with labor duration within the normal limits (data not shown). Overall, the most often cited reasons for having had a C-section delivery were that the baby was in an abnormal position (28%), a previous birth was delivered by C-section (19%), and prolonged labor (19%) (Figure 6.2). Other often mentioned reasons were maternal pre-existing health conditions (e.g. cardiovascular problems), cited by 17% of mothers, fetopelvic disproportion (13%), and fetal distress (12%). About one in eight respondents who delivered by C-section reported that they requested this type of delivery, and 7% stated that C-section was performed because of placenta previa. The sum of reasons exceeds 100% because some respondents gave more than one reason. 6.3 Postnatal Care During postnatal care it is important to assess the health of both the mother and her infant and to provide counseling about breast-feeding, nutrition, and family planning. Postnatal care in Romania is initiated soon after the new mother is discharged from the maternity where she delivered and consists mostly of home visit(s) provided by a midwife. The postnatal period is a critical opportunity to evaluate the physical and psychosocial health of a new mother and her infant, to detect and treat postpartum complications, and to provide the counseling and support needed to address any specific problems related to child care and family planning. The 99RRHS provided information about the use of postnatal care and the content of postnatal counseling. Overall, postnatal care was substantially less utilized than prenatal care (32% vs. 89%), in spite of the official recommendations (Table 6.3). Postnatal care was more frequent among urban residents than among rural women (37% vs. 27%). Residents of Bucharest had the highest use of postnatal care (41%) whereas Vallahia residents reported the lowest use (28%). Use was lower among women with less than complete secondary education (26%), those with low SES (23%), and Roma women (25%), but was not influenced by maternal age. Birth order substantially influenced the use of postnatal care, as with the use of prenatal care: women with at least two previous births had the lowest rate of postnatal care (17%). Lower use of postnatal care services among high-parity women has long been recognized and has been explained through greater responsibilities within the household related to child rearing compounded with greater confidence and experience among these women. The C-section deliveries were associated with much higher rates of postnatal care use than were vaginal deliveries (56% vs. 29%). Most women who received postnatal visits were counseled about child immunization (82%), child care (77%), nutrition (70%), breastfeeding (69%) and breast care (69%). However, counseling 92 93 about planning for future pregnancies and methods of birth control was less prevalent (45%). The type of health advice given during postnatal care did not vary significantly by maternal characteristics, except for Roma ethnic background, which was associated with less advice, regardless of the health topic. 6.4 Smoking and Drinking During Pregnancy Use of tobacco or alcohol during pregnancy is a major risk factor for poor pregnancy outcomes. Smoking during pregnancy has been linked to low birth weight babies, preterm deliveries, sudden infant death syndrome, and respiratory problems in newborns (DiFranza and Lew, 1996). The damaging effects of alcohol use during pregnancy include fetal growth retardation, mental retardation, physical abnormalities (especially dysmorphic facial features), and altered neonatal behaviors. Developmental abnormalities occur in approximately 35%-40% of infants born to alcoholic mothers and are associated with consumption of at least two drinks per day during pregnancy (Coles CD, 1993). Overall, 13% of births in the five years preceding the 99RRHS occurred to mothers who were smokers at the time they found out about their pregnancies (Table 6.4). The proportion of women who smoked prior to getting pregnant was higher in urban areas than in rural areas (15% vs. 10%). Women residing in Bucharest were more likely to smoke prior and during pregnancy than were women from other regions. The proportion of smoking mothers was higher among young adults aged 15-24 (17%). Women with high or middle SES reported levels of smoking before and during the pregnancy higher than those with low SES did. Hungarian women reported higher prevalence of tobacco use both before and during pregnancy compared with other ethnic groups (16% and 14%, respectively). The majority of women who smoked prior to getting pregnant (78%) continued to smoke for some time during pregnancy. The same maternal characteristics were associated with tobacco use during pregnancy. Mothers who had a low birth weight baby (less than 2,500 grams) were more likely than mothers of normal weight babies to smoke prior or during pregnancy (18% and 15% vs. 12% and 10%). Drinking alcohol during pregnancy (21%) was much more prevalent than smoking. About one in four women who drank while pregnant did so daily or several times per week (data not shown). Women residing in Bucharest (30%) or Moldova (33%), women older than 34 years (33%), and women with two or more previous births (30%) were more likely to report drinking alcohol during pregnancy. 94 95 96 6.5 Pregnancy and Postpartum Complications As shown in Table 6.5.1, the majority of women who gave birth in 1994-1999 had routine measurement of their blood pressure during pregnancy (91%) and 12% were identified as having high blood pressure (HBP). Only 3% of pregnant women were hospitalized due to HBP. Routine measurement of the blood pressure during pregnancy was less likely to be performed among Roma women (78%) and among women with only one to three prenatal care visits (81%). The prevalence of HBP was slightly higher among women residing in Transylvania (14%), where most of the Hungarian women (also reporting higher HBP prevalence) lived. A higher prevalence of HBP (20%) was reported by women with ten or more prenatal care visits, either because frequent routine measurement of blood pressure increased the likelihood of HBP diagnostics or because these women had early been found to have HBP and were advised to have more prenatal care visits. One in five women with recent births (19%) were hospitalized during pregnancy (Table 6.5.2). The proportion of women who required hospitalization during pregnancy was slightly higher in urban areas than in rural areas (22% vs. 16%) and among women residing in Transylvania (22%). Women aged 15-34 years were more likely than women over 34 years of age to report hospitalization for pregnancy complications. Women with low or middle SES were slightly more likely to report hospitalization during pregnancy than women with high SES (19% and 21% vs. 14%). Women who initiated prenatal care early (i.e., in the first trimester) were almost twice as likely to report hospitalization during pregnancy than women with late prenatal care, since the likelihood of being diagnosed with a pregnancy complication increases with the length of attendance of prenatal care. Hospitalization associated with pregnancy complications, as reported by respondents, ranged from less than 1% to 8%. The highest hospitalization rate was for the risk of preterm labor (8%), followed by pregnancy associated anemia (7%). Four to five percent of pregnancies were hospitalized for urinary tract infection, bleeding (either during the first or the second half of pregnancy), edema, and risk of miscarriage. Hospitalizations for the risk of preterm labor were higher for younger women, lower SES women, and women who had their first prenatal visit during the first trimester. In general, hospitalization rates for most complications were lower for rural women, older women, and women with late prenatal care. One in four women who gave birth in the five years prior to the survey experienced at least one postpartum complication. Except for higher reports of postpartum complications among residents of Bucharest (36%) and Moldova (33%), there was little variation by background characteristics (Table 6.5.3). Generally, women who developed complications during pregnancy were more likely to report postpartum complications. Reported complications ranged from 10% with severe uterine pain to 3% with a breast infection or loss of consciousness. 97 98 99 Severe uterine pain was reported most often by residents of Bucharest or of Moldova region (15%), older women (16%), women with two or more prior births (15%), those who reported pregnancy complications (15%), those who had prolonged labor (15%), and those who had a C-section delivery (14%). 6.6 Poor Birth Outcomes Of all births during the five years prior to the survey, 10.3 per 1,000 were stillbirths (Table 6.6). The stillbirth rate was higher among women living in urban areas than in rural areas (15 vs. 8 per 1,000) among residents of Vallahia and Moldova (both 17 per 1,000), among women with lower educational attendance, and women with low or middle SES. Consistent with data from the literature (DiFranza and Lew, 1996), women who smoked during pregnancy had a higher than average risk of stillbirth (16 per 1,000). Complicated pregnancies that required hospitalization were more likely to have poor birth outcomes, including a higher stillbirth rate (16 per 1,000). Compared with normal labor, prolonged labor (over 20 hours for nulliparous women and over 14 hours for multiparous women) was associated with more than a three times higher prevalence of stillborns (31 vs. 9 per 1,000). The incidence of low birth weight (under 2,500 grams) was 9%. Higher rates were reported by women with low education (12%), women of Roma ethnic background (15%), women with two or more prior births (12%), women who smoked during pregnancy (14%), and women who were hospitalized for pregnancy complications (13%). A major cause of low birth weight is prematurity; the same groups of women in the 99RRHS were more likely to report preterm births. 6.7 Breastfeeding Breast milk is the most complete food an infant can receive during the first few months of life. Breastfeeding is associated with a wide range of benefits for infant health, growth, immunity, and development. These benefits include decreased incidence and severity of diarrhea (Dewey KG et al., 1995; Popkin BM et al.,1990), fewer respiratory and ear infections (Kovar MG et al, 1984; Howie PW et al., 1990), longer birth intervals (by delaying the return of ovulation), and reduced cost to the family. In addition, breastfeeding has been shown to improve maternal health by reducing postpartum bleeding (Chua S et al., 1994), allowing an earlier return to prepregnancy weight (Dewey et al., 1993), and reducing the risks of premenopausal breast cancer (Newcomb PA et al., 1994) and osteoporosis. The 99PvRHS included questions about breastfeeding patterns and duration. As shown in Table 6.7.1, the majority of babies (93%) born during the past five years were breastfed at least for 100 101 short periods of time. The percentage of babies ever breastfed varied little by selected characteristics. Rates of breastfeeding were slightly lower among women living in Bucharest, women older than 34 years of age at the time of delivery (87%), and women with university or postgraduate education (86%). Early initiation of breastfeeding is beneficial for the health of both the infant and the mother. If the mother initiates breastfeeding immediately after she gives birth, the nipple stimulation during suckling triggers the release of oxytocin and uterine contractions that help reduce postpartum bleeding. Sedatives and analgesics given during labor alter the behavior of newborns and can compromise the essential role of the baby in the initiation of lactation. Children who were delivered by C-section had a lower rate of breastfeeding than did those delivered vaginally (87% vs. 94%). Babies with low birth weight were also less likely to be breastfed than those with a birth weight of 2,500 grams or more (75% vs. 95%). According to WHO recommendations, early suckling (within the first hour post-delivery) should be promoted following all spontaneous deliveries. Table 6.7.1 (right panel) also shows the time elapsed between delivery and initiation of breastfeeding. Of infants who were breastfed, only 3% began breastfeeding during the first hour after birth. The majority of children began breastfeeding between 1 hour and the completion of the first day (58%) or during the second day of life (21%). Almost one of six babies (16%) began breastfeeding only after 48 hours. Breastfeeding initiation within the first hour was slightly more prevalent among women living in Bucharest (6%) or in Moldova (5%), among those aged 35 year of age or older (8%), and among Roma women (7%). In terms of babies' characteristics, low birth weight and Caesarean delivery substantially reduced the likelihood of early breastfeeding. For these infants, breastfeeding was more likely initiated after 2 days, if ever. Indeed, 35% of low birth weight babies and 57% of babies delivered by C-section had initiated breastfeeding after 48 hours of life. In Table 6.7.2 the mean duration of breastfeeding is given until the age at which a child was breastfed. An infant is exclusively breastfed if he or she receives only breast milk and almost exclusive or predominantly breastfed if he or she receives breast milk accompanied by water or other liquids (except other types of milk). Children with exclusive or almost exclusive breastfeeding are considered to be fully breastfed (Labbok MH and Krasovec K., 1990). These indicators are recommended by WHO to assess the adequacy of breastfeeding practices in a population and allow for comparisons with findings from other countries. According to the WHO recommendations, "all infants should be fed exclusively on breast milk from birth to 4-6 months of age" and some breast- feeding should be maintained until at least one year of age (World Health Organization, 1991). In the 99RRHS, the proportion of children who were born between 1993 and 1999 who were still breastfed at the time of the interview was calculated by single month of age (0-59 months); the 102 103 104 denominator included all live births in those five years (regardless of survival). These proportions were summed together to calculate the mean duration of breastfeeding. This method is known as the "current status mean" method (World Health Organization, 1991). Durations of exclusive and full breastfeeding were calculated the same way, where babies who did not yet initiate any other liquids or food were classified as exclusively breastfed and those who were either exclusively breastfed or started to receive liquids but no other food were classified as fully breastfed. The mean duration of any breastfeeding was 8.4 months (Table 6.7.2). For most of this time, however, breastfeeding was only partial. The mean duration of exclusive breastfeeding was 0.6 month and did not vary greatly by maternal characteristics. Women who gave birth after the age of 34, Hungarian women, and women with a postsecondary education were less likely to exclusively breastfed. In addition, babies delivered by C-section and those with low birth weight were the least likely to be exclusively breastfed. Although WHO recommends that all children under four months of age should be exclusively breastfed, very few children in Romania were. Mean duration of full breastfeeding was 2.3 months and was shorter for the same groups of women who reported shorter duration of exclusive breastfeeding. Similarly, the duration of any breastfeeding was substantially shorter for these women. 105 The main reasons given by the mother for stopping breastfeeding were that she did not have sufficient milk to breastfed the baby (50%) followed by that the child had reached the age to be weaned (22%) and that the child stopped wanting breast milk (15%) (Table 6.7.3). Insufficient milk was particularly the most common response for children weaned before five months of age (77% and 71%, respectively). For babies aged one year or older, the most common reason was that the child reached the age to be weaned. One in four women stopped breastfeeding when the child was 5-11 months old because the child refused breast milk. 6.8 Infant and Child Mortality One of the principal objectives of the 99RRHS was to estimate levels and trends in infant and child mortality, particularly because infant mortality in Romania is higher than in any other country of the region except Albania (Population Reference Bureau, 2000) (Figure 6.8.1). The survey questionnaire included a series of questions in the pregnancy history, obtaining for each live birth the date of birth, sex of the child, survival status, and for children who had died the age at death. Respondents were asked to report pregnancy outcomes (e.g. stillbirths and live births) according to international definitions. Thus, a live birth was defined as any birth, irrespective of the duration of 106 the pregnancy, that breathes or shows any other signs of life after separation from the mother. Survey data on mortality levels among respondents' children were used to calculate the infant mortality (deaths before the first birthday per 1,000) and child mortality (deaths between 12 and 59 completed months of age per 1,000) rates. Infant mortality was divided into two ranges, neonatal (0-28 days) and post-neonatal (29 days to 11 completed months). Infant and child mortality rates were calculated by means of life tables. The infant mortality rate for the period July 1994—June 1999 was estimated at 31.5 per 1,000, and the mortality rate for under five years was 35 per 1,000 (i.e., 35 of each 1,000 live born children die before their fifth birthday) (Figure 6.8.2 and Table 6.8). In this five-year interval, the neonatal and post-neonatal mortality rates were 20.6 per 1,000 and 10.9 per 1,000, respectively. In this type of survey underestimation of neonatal mortality tends to be greater than underestimation of child mortality at older ages. Some women, especially those without formal education and those who have had many births, do not always consider their births to be live births, especially when the death occurred in the first few days of life. For this reason, the estimated five-year neonatal and infant mortality rates should be considered as minimum values. 107 The infant mortality rate estimated from the 99RRHS is about 40% higher than the infant mortality rate for January 1994-December 1998 reported by Romania to the World Health Organization (Figure 6.8.2). The greatest difference between the survey estimates and the official data was observed in the levels of neonatal mortality rates: for the most recent years of reporting, the official neonatal mortality rate was 39%-46% of the infant mortality rate, ranging from 9.1 per 1,000 in 1994 to 9.3 in 1995, 8.7 in 1996, 9.2 in 1997, and 9.4 in 1998 (WHO, World Health Statistics Annual, 1998 and 2001). Survey data showed that most of the infant deaths during the first year of life (69%) occurred during the first 28 days after birth. Similarly, in western European countries with relatively complete vital records, the ratio between neonatal and postneonatal deaths is typically 60% for comparable infant mortality rates (Demographic Yearbook, 1974). Thus, despite the potential underreporting of early child deaths among survey respondents, the survey estimates of neonatal deaths are substantially higher than the official data; the neonatal death rate of 20.6 per 1,000 in 1994-1998 was about twice as high as the official average rate for the same period of time. Presumably higher underreporting of these deaths exists within the vital records reporting system, either because hospitals use a Soviet-era definition for live birth (e.g., 28 weeks of gestation or weight at birth of 1,000 grams and respiratory movements) or because of underregistration of births within the civil registry system. Differentials in infant and child mortality for the period July 1989-June 1999 by period of exposure (five-year intervals), area of residence, age, education level, and ethnic background of the mother, birth order, birth interval and sex of the child are presented in Table 6.8. The infant mortality rate for the period July 1989-June 1999 was estimated at 29.6 per 1,000, and the mortality rate for under five years was 32.4 per 1,000. Neither infant nor child mortality rates differed significantly by mother's residence or by period of time. Mortality differentials by age of the mother at the time of birth showed that the highest infant and under five years mortality rates are found among births to women aged 30 and older (52 and 54 per 1,000). They had also reported the highest neonatal and postneonatal mortality rates (26.6 and 25.4 per 1,000). Young women under age 20 had the lowest neonatal mortality rate (11.6 per 1,000) but a high postneonatal mortality rate (22.2 per 1,000). Infant and under five years mortality rates were almost two times higher among mothers with primary education than among those with at least completed secondary education (39.7 and 46.1 vs. 22.3 and 23.9 per 1,000). The excess mortality was due to higher probabilities of dying after the first month of life for children born to less educated women, whereas the neonatal mortality rates in these two groups were comparable. Infant mortality was much higher among Roma women than women of other ethnic backgrounds. The Roma infant mortality rate was two times higher than that for Romanian infants. Part of this difference may be due to lower education, earlier initiation of childbearing, higher fertility rate, overcrowding, and lower knowledge, access and use of health services; however, little is known about specific patterns of disease among Roma people and if they are different from those of other groups (e.g., different rates of inherited congenital malformations). These findings are 108 109 consistent with a recent United Nation High Commissioner for Refugees (UNHCR) report on the Roma people of Central and Eastern Europe (Braham M, 1993) that found infant mortality rates four times higher among Roma people than among their non-Roma neighbors. In the 99RRHS, the highest infant and under five years mortality rates were found among births to women with two or more previous births (56.6 and 60.6 per 1,000). The mortality rates were also higher among infants born after short birth intervals (less than two years) or those spaced 4 or more years compared to those spaced 2-3 years. Male infant and under five years mortality rates were slightly higher than the rates for females (32.1 and 34.7 per 1,000 vs. 26.3 and 29.4 per 1,000), reflecting the sex differential in neonatal mortality. 110 CHAPTER VII CONTRACEPTIVE AWARENESS AND KNOWLEDGE OF USE Although the induced abortion rate declined significantly from 1993 to 1999, Romania continues to report more than two abortions for every live birth, owing mostly to low use of effective contraception and a high reliance on traditional methods, public lack of knowledge and mistrust of modern methods, and underutilization of the family planning services recently made available in the country. Despite the recent successes of the national family planning program launched in the early 1990s and the united efforts of non-governmental organizations and donor organizations, much more work is needed to meet the contraceptive needs of all subgroups of the Romanian population. Lack of or misleading information about family planning methods and their side effects, and little knowledge about the places where methods can be obtained, are important barriers to consistent and correct use. An important objective of the 99RRHS was to explore the level of knowledge of family planning methods and their source of supply among women and men of reproductive age in the aftermath of intensified information, education, and communication (IEC) efforts during the 1990s. In reference to 10 modern and traditional contraceptive methods, respondents were asked if they had ever heard about each, if yes from whom, if they knew to use them, and if they knew where they could be obtained. These data were compared with results from the previous reproductive health surveys to examine recent trends in contraceptive knowledge. 7.1 Contraceptive Awareness and Knowledge of Use In 1999 virtually all women had heard of at least one modern method of contraception (99%) and most had heard of a traditional method (93%) (Table 7.1.1 A). Awareness of condoms, pills, and intrauterine devices (IUD) was very high (98%, 93%, and 91%, respectively), followed by awareness of withdrawal and the calendar method (85%). Contraceptive female sterilization (tubal ligation) was known by almost three of four women (72%). The least known methods were those that are seldom available (vasectomy and injectables). Awareness of emergency contraception was also very low, in spite of the relatively good availability of combined oral contraceptives. Although the level of overall awareness of either modern or traditional methods did not vary much by residence, some urban-rural differences were notable in women's awareness about specific contraceptive methods. For example, awareness of pills or the IUD was 12%—13% higher among 111 urban residents than among rural residents, female sterilization 25% higher, and spermicides 71% higher. For some lesser known methods (vasectomy and emergency contraception) the gap was even larger. In 1999, compared with 1993, there was a slight increase in women's overall awareness of both modern (from 94% to 99%) and traditional (from 85% to 93%) methods. Awareness of all modern methods increased . Awareness of female sterilization increased by 30%, spermicides increased by 42%, vasectomy by 230%, and injectables by 63%. Even with these increases, 112 however, fewer than one of two and one of four women, respectively, had heard of vasectomy and injectables. In rural areas, on the other hand, there was a more substantial increase in women's awareness of all modern methods since 1993, including the better known methods such as condoms, pills, and IUDs. Overall, the average number of methods a woman was aware of increased from 5.2 in 1993 to 6.7 in 1999, almost entirely because of a higher average of the number of modern methods known in 1999 (5.0 vs. 3.7 modern methods). Among men, except for universal awareness of the condom, overall awareness of all methods was lower than that of women—men knew, on average, one method less than women—and the differential between the levels of awareness of modern methods in urban and rural areas was equally 113 114 striking (Table 7.1.1B). Comparing the four regions, except for the condom, the levels of awareness of the better known modern methods were higher in Bucharest than in Vallahia, Transylvania, or Moldova. Men's level of awareness of emergency contraception was especially low in Vallahia and Moldova, and their awareness of injectables was low in all four regions. Among women 15-44 years of age, the overall awareness of modern methods was equally high and did not vary with the respondent's age, with the sole exception of female sterilization, which was known by only 56% of the youngest women (Table 7.1.2). The overall awareness of traditional methods was also lower among the youngest women (83% vs. 98%). For women aged 25—44, awareness of the IUD ranked second after condoms and awareness of pills ranked third; for the youngest respondents awareness of condoms and pills were higher than IUD awareness (99% and 91%, respectively, vs. 84%). Virtually all currently married or cohabitating women (i.e., women in union) as well as previously married women had heard of at least one modern method and at least one traditional method. With the exception of never-married women, whose awareness of traditional methods was much lower than that of modern methods (80% vs. 99%), knowledge of modern and traditional methods was equally high. Awareness of some modern methods was lower among never-married than among ever-married respondents (e.g., female sterilization), however. Since marital status is directly correlated with age (see Chapter IV) and never-married women were more likely to be young, the pattern of knowledge of specific methods among unmarried women resembled that for younger women, with higher awareness of condoms (99%) and lower awareness of IUDs (82%) and tubal ligation (55%). Among men the patterns are similar (bottom panel of Table 7.1.2). Except for condoms and pills, knowledge of the more widely available modern methods was substantially lower among the youngest men (15-24 years old). As was the case for women, the youngest men were also somewhat less likely to be aware of any traditional method (86% vs. 96%-97%). The differentials according to marital status were similar to those of women, with fewer never-married men being as aware of all modern methods, except the condom. Overall, the number of modern family planning methods recognized was the lowest among young adults (4.7 among women and 4.0 among men) and unmarried respondents. Compared with previous reproductive health surveys, however, there was a continuous improvement in young adults' awareness of modern methods for both women and men; the improvement was more pronounced between 1996 and 1999 (Figure 7.1.1). Among women, the average number of methods known increased by approximately 50% between 1993 and 1999 (from 2.6 to 4.0 methods among 15-19- year-olds and from 3.6 to 5.3 among 20-24-year-olds) and the increase was more rapid since 1996 (especially among 20-24-year-old women). Similarly, men aged 15-24 showed a more rapid acquisition of awareness of modern methods than 15-19-year-olds between 1996 and 1999. 115 Respondents' overall level of awareness of at least one modern method was not significantly different for better-educated women and men (Table 7.1.3). Awareness of specific methods, with the exception of condom awareness, was lower among women with only a primary education, however. Particularly notable was the much lower awareness of tubal ligation, spermicides, vasectomy, injectables, and emergency contraception among less-educated women. Consequently, the average number of modern methods known was directly correlated with education, ranging from 3.4 modern methods among women with a primary education to 6.6 modern methods among the most educated women. The overall awareness of traditional methods was also positively correlated with education but the variation was less pronounced. The pattern is similar among men; with the exception of condoms, the level of awareness of all modern methods is positively correlated with the level of respondents' education. Men with postsecondary education know, on average, three modern methods more than those with only primary complete education. 116 117 Respondents who reported that they were aware of (i.e., "have heard of) a contraceptive method were asked whether they knew how to use the method. The proportion of respondents who know how each method or procedure is used is usually substantially lower than the proportion aware of each method or procedure (Figure 7.1.2). Among women, knowledge of use of any modern or traditional method was lower than the corresponding awareness (86% vs. 99% and 85% vs. 93%, respectively) (Table 7.1.4). For the most widely known modern contraceptive methods (condoms, pills, and IUDs), there was a serious gap between awareness of the method and knowledge of how they are used (Figure 7.1.2). Although awareness of condoms was universal, only 78% of women said they actually knew how condoms are used. Additionally, although 93% and 91% have heard of the pill or IUD, only 54% and 56%, respectively, knew how the methods are used. A similar gap in knowledge was obvious for tubal ligation, spermicides, and injectables, further narrowing the proportion of women who could start using these methods. The gap between awareness and knowledge of use was also present for the calendar method and, to a lesser extent, for withdrawal. 118 119 The difference between awareness and knowledge of use diminished with increased education. For example, the proportion of women who did not know how condoms are used decreased from 48% among women with a primary education, to 24% among women with a less than complete secondary education, to 13% of those with a complete secondary education and to only 5% among those with a university education. Similarly, the proportion of women who did not know how the IUD works decreases from 68% to 22% between the lowest and highest levels of education, whereas the percentage of women who did not know how to use the pill decreased from 71%> to 23%. Similarly, knowledge of the use of periodic abstinence more than doubles among women with a postsecondary education compared with women with primary education. With the exception of condoms and withdrawal, as was the case for women, the level of knowledge among men about how specific contraceptives are used was also substantially lower than their awareness of these methods (bottom panel of Table 7.1.4). To an even greater extent than among women, this knowledge was also positively correlated with men's level of education. For example, between five and six times as many men in the highest education group compared with the lowest said that they know how pills, IUDs, and tubal ligation are used. 120 Another indicator commonly used to evaluate Information, Education and Communication efforts is knowledge of source(s) of contraception. The 99RRHS found that 93% of women could name at least one source for supplied methods of contraception (Table 7.1.5A). Knowledge about contraceptive source increased 10 percentage points from six years ago (83% in 1993). Respondents were more likely to know a source for commonly used methods. For instance, 89% of women knew where to obtain condoms, 81% knew a source for pills, and 75% knew a source for IUDs, but very few knew where vasectomies are performed or where to get injectables or emergency contraception (Figure 7.1.3). Compared with 1993, knowledge of a source for condoms, pills and IUDs increased markedly, whereas knowledge about sources for the least used methods was much less widespread. The data on knowledge of sources of modern contraceptive methods were similar for men. Not surprisingly, a slightly higher proportion of men (96%) than women (89%) were aware of a source for condoms, although fewer men than women knew of a source of all other modern methods (Table 7.1.5B). As was the case for women, among men knowledge of sources of contraception was lower in rural areas and is higher in Bucharest than in the other three regions. 121 7.2 First Source of Information About Contraception The 99RRHS found that among women 15-44 years of age, the main source of information about birth control methods was a friend or acquaintance (41%), followed by mass-media (9% audio- visual media, 11% print media and 3% books and a physician (11%) (Table 7.2.1 A and Figure 7.2.1). Young women (15-24 years of age) reported similar first sources of information as older women (Figure 7.2.2). More than one in three (38%) young women found out about contraception in discussions with a friend or acquaintance, 25% from mass-media or books, and 7% from a health care provider. Only 9% of the young women surveyed said that they had first heard about contraception from one of their parents (7% from their mothers). Only 4% of young women cited the school as their first source of contraceptive information. Mass media was a first source of information for more young women than all women aged 15-44 years. The first source of information for contraception did not change much over time. Generally, mass media and medical practitioners continued to play a limited role in contraceptive educational efforts, even though 1 in 5 and 1 in 10 women, respectively, mentioned them as the first source of information. Similarly, there was little change in contraceptive information among young adult women. The 96YARHS showed that the first source of information for young women was a friend 122 or acquaintance (42%), followed by mass media (17%) and a physician (11%) (Serbanescu F. and Morris L., 1998). Although the 96YARHS documented a slight increase in the contribution of health providers (from 6% to 11%) and young women's mothers (from 7% to 10%) in spreading contraceptive information, the 99RRHS showed a slight decrease in the prevalence of these sources between 1996 and 1999. Furthermore, there was no significant change in the role played by schools as a first source of contraceptive information, presumably because courses including information on contraception are taught late, after youths are exposed to other sources of information (see Chapter XIV). These findings explain, in part, the poor quality of contraceptive information, often acquired through rumors, and argue for increasing the public health efforts in educating youth through official channels (school, mass-media, health providers) about the benefits of contraception and the availability of family planning products and services. 123 124 Overall, contraceptive information among men was more likely to be acquired for the first time from a friend, peers or colleague (47%), a girlfriend or partner (11%), and mass-media (21%) (Table 7.2.1B and Figure 7.2.1). Physicians (6%), parents (1%), and school (1%) seldom contribute to first contraceptive information among men. The source of first contraceptive information varied only slightly by age: contraceptive information from a friend (52%) or mass media (24%) were mentioned more frequently by young adult men than their older counterparts, whereas information from a medical provider were the least likely among 15-24 year olds (2%). Furthermore, there has been very little change in the source of contraceptive information among young men between 1996 and 1999 (Figure 7.2.2). Regarding the first source of information for specific methods, more women than men mentioned they heard about IUD and tubal ligation from a medical health provider (22%), about periodic abstinence from their mothers (19%), and about condom and withdrawal from their partners (51% and 48%, respectively) (Table 7.2.2). Except for condoms, men were more likely than women to mention mass media as the first source of information for modern methods. Twenty eight percent of men mentioned mass media as their first source of information on pills, 23% for tubal ligation and 23% for the IUD, whereas the percentages for women were 20%, 13% and 10%, respectively. Among men, the first source of information about periodic abstinence was most often a girlfriend or a partner (41%), or a friend (35%), whereas 70% first heard about withdrawal from a friend. 125 126 7.3 Knowledge About Contraceptive Effectiveness Correct information about contraceptive effectiveness can greatly influence a couple's decision about how to prevent unintended pregnancies. It is not realistic to expect individuals to make informed decisions if they have gaps in their knowledge about all possible contraceptives available and if adequate access to comprehensive family-planning services is lacking. Women's lack of knowledge about contraceptive effectiveness is an indirect indicator of the failure of adequate counseling and information/education programs. The 99RRHS included a series of questions in which each respondent was asked to indicate whether specific contraceptive methods (shown on a card) was very effective, effective or not effective in preventing pregnancy when used consistently and correctly. Answers to these questions are presented in Table 7.3, where contraceptive methods are listed in descending order of effectiveness (Hatcher et al., 1998). This ranking is based on studies of unintended pregnancies among users of various family-planning methods in the first 12 months of using that method (method failure), with the exception of emergency contraception for which such analysis does not apply. According to these studies, vasectomy and Norplant (whose specific effectiveness was not explored in the 99RRHS because they are largely unavailable in Romania) are the most effective methods, with a failure rate at one year of use of only 0.1 pregnancies per 100 women. They are followed by injectables, female sterilization, and IUDs, with rates of failure between 0.3 and 0.6 pregnancy per 100 women. Combined oral contraceptives have theoretical failure rates comparable to Norplant and vasectomy (0.1 pregnancy per 100 women), but their actual failure rate, as commonly used, is much higher (6-8 pregnancies per 100 women). Condoms and other barrier methods are considered to be of moderate effectiveness, with failure rates of 3%-6% during correct use and 14%-26% as commonly used. The calendar method can be moderately effective if used correctly. Finally, withdrawal is listed as less effective than all other methods. Overall, no modern method was recognized as very effective by a majority of women, partly because substantial numbers of women lacked awareness of modern methods (Table 7.3, upper panel). Even when women who had never heard of a specific method were excluded, very few effective methods were correctly recognized as highly or very effective. For example, if those who never heard of female and male sterilization are excluded, both these methods were correctly identified as being very effective by 61% and 56%, respectively. However, only 10% of women who had heard of injectables qualified this method as very effective, and most could not assess its effectiveness. Moreover, only 29% of women who were aware of IUDs and 20% of women who were aware of pills considered those methods very effective. Most of those women believed the methods are somewhat effective and about one in five did not know if those methods were reliable. Surprisingly, almost half of women who had heard of emergency hormonal contraception qualified the method as highly effective. Both withdrawal and periodic abstinence were qualified as less or 127 not effective by most of those who had heard of them, whereas one in three respondents believed they were very effective or effective (36% and 33%, respectively). The pattern for men differed slightly because, as shown previously, fewer men than women were aware of most methods, and, of those who were aware of specific methods, smaller proportions thought they are effective (bottom panel of Table 7.3). Among modern methods, condoms are an exception as 78% of men said they are "very effective" or "effective," compared with 60% of women. Also, more than half (52%) of men thought withdrawal is "very effective" or "effective," compared with 31% of women. Compared to the 93RRHS, Romanian women seemed to have acquired more trust regarding the reliability of modern methods in preventing pregnancy: confidence in contraceptive effectiveness of both the pill and IUD doubled (from 33% to 64% for IUD and from 26% to 60% for pills) and confidence in condoms grew by 10%(Figure 7.3). As a result, more women correctly had more confidence in the IUD's effectiveness than in the effectiveness of condoms or pills. Most of the change was the result of better knowledge of these methods, since the proportion of those who could not venture an opinion on the effectiveness of pills and IUDs declined from 45% and 49%, respectively, to 27%-28% (data not shown). 128 129 7.4 Young Adults' Knowledge about Condoms' Effectiveness in Preventing STDs Used correctly, condoms can help prevent both pregnancy and sexually transmitted diseases (STDs). Although the method-specific contraceptive effectiveness is lower than that for other modern methods, condoms are very effective in preventing STDs. To be highly effective, they must be used at each intercourse. Even one unprotected intercourse with an infected partner has a risk of STD transmission, ranging from 1% for HIV to 30% for genital herpes, 40% for chlamydia, and 50% for gonorrhea (Harlap S. Et al, 1991). Studies on all users, including those who used condoms inconsistently or incorrectly, show that condoms reduce by at least 60% the risk of contracting HIV (Davis KR and Weller SC, 1999) and by a third the risk of other STDs (gonorrhea, chlamydia, trichomoniasis). Consistent and correct users have minimal risk of contracting STDs, including HIV. In addition to respondents' knowledge about the condom's contraceptive effectiveness, the survey explored their knowledge on the condom's role in protection against STD transmission. Table 7.4 and Figure 7.4 show the percentage distribution of young adult women and men by their knowledge of the condom's effectiveness in protection against STD transmission. 130 131 Overall, two thirds of young women and three fourths of young men responded that condoms are effective in preventing STDs, including 35% and 43%, respectively, who said they are very effective (Table 7.4). A very small proportion of women and men (5% and 3%, respectively) said that condoms are not at all effective, while 14% of women and 6% of men did not have enough knowledge to assess whether they are effective or not. Rural residence, secondary incomplete education, and lack of experience with modern contraception were associated with lack of knowledge or little confidence about the efficacy of condoms in protecting against STDs. Compared with the 1996 survey, young women's perceptions of the condom's effectiveness in preventing STDs improved slightly—far fewer women in 1999 had no knowledge of condom's effectiveness and more women believed that it is very effective or effective in preventing STDs (Figure 7.4). Among young men, however, the changes were less pronounced. 132 CHAPTER VIII CURRENT AND PAST CONTRACEPTIVE USE The 93RRHS showed that the use of modern contraceptives was low (14%) and reliance on traditional methods, which are prone to high failure rates and subsequent unintended pregnancies, was high (43%). The reasons given for limited use of modern contraception included lack of access, shortages and uneven distribution of contraceptive supplies, little knowledge about modern methods, concerns among both family planning clients and providers about the health risks associated with certain methods, and the easy access to and low cost of obtaining induced abortions. An important objective of the 99RRHS was to assess the current levels of contraceptive practices among different subgroups of women and men and the trend for women in the six years since 1993. 8.1 Current Contraceptive Prevalence The contraceptive prevalence rate for currently married and in-union women increased from 57% to 64% in the six years from 1993 to 1999. More importantly, use of modern methods doubled from 14% to 30% (see Table 8.1.1A and Figure 8.1.1). Although this section focuses on women and men in legal and consensual marriages because they represent 87% of currently sexually active women (within the past 30 days); because they have greater frequency of intercourse, higher fertility and more unintended pregnancies; and because they constitute the common denominator for most national and international studies of contraceptive prevalence, it is important to document the contraceptive behaviors of all women and men. Many previously married or never married women who had ever had intercourse were not currently sexually active and therefore not in need of contraception (see Section 5.6), so not surprisingly, only one of five is currently using contraception (Table 8.1.1A and Figure 8.1.1). There was no substantial difference in the overall use of either modern or traditional methods between previously married and never married women, although previously married women depended to a greater degree on long-term methods. The ratio of modern to traditional methods for women not currently in a consensual or legal union was 1.6:1. Withdrawal, followed by female sterilization and the IUD, were the most widely used methods among previously married women (5% and 4%, respectively), whereas never married women were more likely to use condoms (7%), withdrawal (6%), or pills (5%). 133 134 Among previously married or never married men, a larger proportion (29%) used a contraceptive method in the past 30 days than did unmarried women (20%) (Table 8.1.1B and Figure 8.1.1). However, much of this difference was due to a higher proportion of these men using traditional methods, particularly withdrawal. Indeed, the ratio of modern to traditional method use for men not currently in a consensual or legal union was 1.2:1, compared with 1.6:1 for women. Condoms, followed by withdrawal, were the most widely used methods among previously and never married men (11%-13% and 10%-12%, respectively). 135 The prevalence of contraceptive use among women currently in legal or formal unions was very high (64%) but only 30%, fewer than a half of users, used modern methods (Table 8.1.2A). The proportion of women in union currently using any form of contraception ranged from 38% (among childless women) to 73% (women with two living children). For the entire country, the proportion of all contraceptive users who used a modern method was 46%, ranging between 27% and 70% for those with the lowest and highest levels of educational attainment. The proportion of women currently in union who used any contraceptive method was slightly higher in urban areas than in rural areas, among 25-34-year-olds, and among those with one or two children; the proportion increased directly with educational level. Modern contraceptive use was significantly lower in rural than in urban areas (21% vs. 35%), among young adults (15-24 years) and women aged 35 or over than among women aged 25-34 (26% and 24%, respectively, vs. 37%), among women who had not completed a secondary education (14% for primary or less and 25% for secondary incomplete), among women living in households with a low socioeconomic level (18%), 136 137 138 among women with three or more children (17%), childless women (24%), and among Roma women (16%). Among only very few subgroups did the use of modern methods surpass the use of traditional methods by a considerable margin (Bucharest residents, childless women, women with the highest level of education, and high SES women). The data for men in union were similar to the corresponding female data (Table 8.1.2B). Sixty-six percent of men in union (or their partner) were using any contraceptive method, but similar to women, only 27% (41% of current users) used a modern method. As was the case for women, the proportion of men currently in union using any contraceptive method was slightly higher in urban than in rural areas, among 25-34-year-olds, and among those with one or two children; the proportion increased directly with educational and socio-economic levels. Somewhat similar to women, modern contraceptive use among men was lower in rural than in urban areas (20% vs. 32%), among men in lower educational categories (12% for primary or less and 23% for secondary incomplete), among men living in households with a low socioeconomic level (14%), among men with three or more children and among Roma men (17%). Unlike women, the use of modern methods did not significantly surpass use of traditional methods for any subgroup of men. Contraceptive use (any method) by women in union has increased from 57% in 1993 to 64% in 1999, an increase of 11% (Table 8.1.3 and Figure 8.1.1). More importantly, overall use effectiveness has increased, as the proportion of users who used a modern method almost doubled, from 24% to 46%. The proportion of users who employed a modern method more than doubled among women in Vallahia and Moldova, among the youngest women, among women with no children, those with secondary incomplete or postsecondary education and women in the middle SES category. By far the most prevalent method in use among women in union was withdrawal (29%), which accounts for 45% of contraceptive prevalence (Table 8.1.4A and Figure 8.1.3). Condoms, which were used by 9% of women in union, pills (8%), and IUDs (7%) were the next most used methods and accounted for 80% of modern methods used. Tubal ligation, despite an overwhelming desire by most women to have no more children (see Section IV), was used by only 3% of women currently in union. The calendar method was used by 6% of women in union. 139 140 The overall proportion of women in union currently using a method varied slightly by background characteristics (except childless women, who had a much lower prevalence of contraceptive use), but the choice of a specific method sometimes differed by a considerable margin. The use of condoms was higher than average in Bucharest (13%) and other urban areas (11%), among 25-34 year-olds (11%), among those with a postsecondary education (23%) or with high socioeconomic status (14%), and among women of Hungarian ethnicity (12%). Condom use was very low in rural areas (4%), among less educated women (3%), among those with low SES (3%) and women of Roma descent (2%). The use of other modern methods varied less by background characteristics. The use of withdrawal was significantly higher among rural residents (37%), in Transylvania (36%), among women with less education (data not shown), and among women with low SES (38%). 141 142 143 The proportion of men in union using each method was similar. Withdrawal was by far the most prevalent method (28%), accounting for 42% of all contraceptive use (Table 8.1.4B and Figure 8.1.3). Condoms, used by 9% of men in union, pills (8%), and IUDs (6%) are the next most used methods and accounted for almost 90% of all modern methods used by men or their partners. The calendar method was reported by a greater proportion of men (11%) than women (6%) in union. The overall proportion of men currently using a method varied only slightly by background characteristics, but the choice of a specific method sometimes differed by a considerable margin. Male condom use was lower than average in rural areas (5%), in the Vallahia and Moldova regions (6% and 8%, respectively), among those in lower education and socioeconomic groups, and among men of Roma ethnicity (1%). Pills and IUD use were directly associated with employment, education level, and socioeconomic index. The use of withdrawal was significantly higher among residents of regions other than Bucharest (29%-30%) and among men with less education (data not shown) and in lower socioeconomic categories (30%-34%). 144 Data collected in both the 93RRHS and the 99RRHS demonstrated heavy reliance on traditional methods, mostly withdrawal (Figure 8.1.4). Although withdrawal was the leading method in both surveys, its prevalence among users declined 35% to 29%. At the same time, the proportion of women currently married or in consensual union using modern contraception more than doubled, from 14% in 1993 to 30% in 1999. Almost all of the increase in use was the result of the increased popularity of pills and condoms, whose prevalence more than doubled (from 3% to 8% and from 4% to 9%, respectively). The increase in IUD use (from 4% to 7%) also contributed to the overall increase. There were no noticeable changes in the use of other modern methods of contraception. Use of the IUD, the only long-term method widely available, was very limited among childless women (1%) and increased among women with one (7%) or two (10%) children, although among women with three or more children IUD use decreased again to 7% (Table 8.1.5A). On the other hand, condom and oral contraceptive use were inversely correlated with the number of 145 children. Pills, especially, were used to a lesser extent by women with three or more children. As expected, tubal ligation is mostly used by women with at least two children, but even in this subgroup its prevalence was quite low (4%). Withdrawal use was higher among women with any living children, whereas the calendar method did not show a clear pattern. 146 Condom use by men and pill use by their partners decreased slightly as the number of living children increased, whereas use of more permanent methods (IUDs and tubal ligation) generally increased with the number of living children (Table 8.1.5B). Male use of traditional methods was higher among men in union with living children. The prevalence of contraceptive use among women and men in union increased with educational attainment (from 51 % to 71 % among women and from 46% to 75 % among men) (Table 8.1.6). Among both men and women in union, the use of almost all specific modern methods increased substantially with education. The most striking finding was the eightfold increase in condom use among women (from 3% to 23%). Use of the pill and IUDs as well as all major methods for men increased two to four times between the lowest and highest education groups. 147 Withdrawal use was lowest and calendar use was highest among those with a postsecondary education. Within each education category, there was no significant difference between men and women. 148 8.2 Source of Contraception To assess sources of contraceptive methods for men and women currently in union, the 99RRHS included questions about where current users of supplied contraceptive methods obtained their contraceptives. Commercial sales in general are the largest source of contraception in Romania. Pharmacies, in particular, were the most important source of contraception for men and women in union, supplying 53% of current male users and 48% of current female users (Tables 8.2A and 8.2B). Because pharmacies were the subject of a rapid process of privatization, it is very difficult to differentiate between public, private, and mixed ownership status. Other commercial sales outlets (stores or street markets) were the source of contraception for 16% of male users of modern methods, mostly condoms, but for fewer than 4% of women. The public medical sector was the second most important source for both men (20%) and women (32%). Hospitals with gynecologic wards supplied 13% of men (or their female partners) and 20% of women currently in union with their current method of contraception. Additionally, family planning clinics or offices supplied 5% of men and 9% of women, whereas polyclinics and dispensaries supplied only 2% of men and 3% of women. Private medical clinics or doctors constituted an emerging source of contraception, particularly for IUDs. Other sources, such as partners, friends, and relatives, supplied 4% of male users and 8% of female users. Sources varied greatly according to the contraceptive method used. Pharmacies were the principal provider of condoms, pills, and spermicides, supplying more than 60% of condoms and pills for both men and women and almost all spermicides (data for male spermicide users not shown). Pharmacies also supplied 16% of the IUDs used by female partners of males and 10% of IUDs for female users (with a prescription issued by the OB/Gyn), but the IUD must be inserted at a medical facility. Public hospitals were the primary source of IUDs (38% of male users partners and 42% of female users). Family planning clinics were the second most common source of pills, supplying 18% of women and 13% of female partners of men. Not surprisingly, partners constituted the second source for condoms for women (23% of users), and the second source of condoms for men was a kiosk or store (31% of users). Very few men or women reported obtaining condoms in a family planning clinic. Virtually all contraceptive sterilization procedures took place in maternity hospitals. Among women currently in union, the most noticeable change in source of supplied methods for current users consisted of a complete abandonment of street market contraceptive purchases (Figure 8.2). In 1993 17% of current users reported a street vendor as their main source of supplied methods (particularly condoms and pills); virtually no women used this source of contraception in 149 1999. This finding may explain, in part, the dramatic decrease in the condom's failure rate reported in 1999 compared with 1993 (see also Section 8.8). 150 151 8.3 Dissatisfaction with the Current Method and Preference for Other Methods The percentage of men and women who reported having problems or concerns about their current method of contraception was considerably lower than the percentage who wanted to switch to a different method. Overall, about 1 of 10 female current users and 1 of 8 male current users said they had problems or concerns about their current method of contraception. Condoms, withdrawal, and the calendar method were those with which both male and female respondents were the least satisfied (Table 8.3.1 and Figure 8.3). The main reason for dissatisfaction with traditional methods was their low use-effectiveness. Both male and female condom users reported that the main reason for dissatisfaction was related to difficulty or unpleasantness when using the method. Among male and female pill and IUD users, side effects and health concerns accounted for most of the complaints about the method. 152 153 To assess method acceptability, all current users of contraception were asked if they preferred to be using some other method of preventing pregnancy. Overall, about one of four male and female users answered positively (Table 8.3.2). However, the percentages differed considerably depending on the method used. The calendar, condom, and withdrawal methods were those with which respondents of both genders were the least satisfied; about one of three women who were using any of these three methods and 27%-28% of men. Among women, the only methods with low proportions who preferred other methods were the IUD and female sterilization; among men, only few reported problems with these methods or with pills. Virtually none of the small percentage of women and few men whose partners had been sterilized preferred another method. Among women, the IUD and the pill are the most preferred methods (accounting for 39% and 33%, respectively, of the preferred methods), especially among users of condoms and of traditional methods. Only 6% of women who wanted to switch to another method indicated that they preferred female sterilization. Notably, 3% of all users (representing 12% of the 26% who prefer to use another method) were unable to name a desired method, which indicates the need for renewing IEC efforts among current users. 154 155 156 For men, the pill as well as the condom and the IUD were the most preferred methods (accounting for 41%, 23% and 20%, respectively, of the preferred methods). This was especially true for condom and traditional method users. Only 4% of those men who wanted to switch to another method indicated they preferred their female partner be sterilized. On the other hand, 74% of women and 77% of men were satisfied with the method they are using. Overall, one-third of women and 22% of men who wanted to use another method were still thinking about switching to their preferred method (Table 8.3.3). One-fourth of female respondents and 19% of male respondents who wanted to use another method were concerned about potential side effects associated with the preferred method, principally clinical methods. A relatively high proportion of men (18%) and women (13%) said the cost associated with their preferred method was the most important barrier to switching. Lack of availability of the preferred method was mentioned by 10% of female respondents who wanted to switch to another method. A majority of women preferring the IUD, the method preferred to the greatest extent by female respondents, said they were not using it either because they still were thinking about it or because they feared side effects. The same was true of 48% of men who wanted their partners to use the pill, the method most male respondents preferred. Cost/access (22%) and lack of a doctor's recommendation (9%) were also mentioned as important reasons by female current users who wanted to use the IUD. For those women who preferred pills, fear of side effects (38%), indecision (26%), and cost/access (18%) were the most important reasons they had not changed. Tubal ligation was mentioned by a number of women, for whom the most frequent reasons for nonuse were partner opposition (27%), indecision (19%), fear of side effects (18%) and cost/access (31%). Among men, partner's opposition was another reason why their partners had not yet switched to the IUD or pill (about one of five men preferred each method). 8.4 Users of Non-Supplied Methods Every respondent who was currently using any non-supplied method (i.e., calendar method and withdrawal) was asked whether certain factors were "important" or "somewhat important" in their decision not to use a more effective method. Most women stated that fear of side effects (79%), partner preference (57%), lack of knowledge about modern methods (55%), and cost (42%) or availability (36%) of modern methods were the major factors influencing their decision not to use a modern method (Table 8.4.1). Few women (10%) considered their religious beliefs, a friend (8%) or a doctor's advice (7%) as important factors in their decision to use traditional methods. 157 Among users of non-supplied (traditional) methods there was no significant variation by background characteristics in the proportion mentioning that fear of health/side effects was important in their decision to not use a modern method. Lack of knowledge was more often mentioned by women using withdrawal, women in rural areas, 15-24 year-olds, those with less than complete 158 secondary education, and women with low SES. Less than complete secondary education and low socioeconomic status were also more common for those reporting partner preference. The cost and availability of modern methods were mentioned more often by women living in rural areas, older women, women currently married or in consensual union and women with lower education or low SES. Religious beliefs were more important for rural women and women with less than complete secondary education. A friend's or a doctor's recommendation was cited as being barrier by less than 9% of traditional method users. In conclusion, a substantial number of factors mentioned as important in their decision-making by women who chose to use traditional methods could in fact be influenced by adequate contraceptive counseling and improved access to family planning services. There was little change since 1993 in the most important factors that influenced women in their decision to use traditional methods rather than modern methods (Figure 8.4.1). Fear of side effects, partner's preferences, and little knowledge of modern methods continued to affect women's decisions to use traditional methods in 1999. Users of traditional methods in 1999 were more likely to agree that the cost of modern methods was a more important factor than they were in 1993 (43% vs. 34%). A much lower proportion of women in 1999 than in 1993 stated that a doctor's recommendation influenced their decision to use traditional methods (7% vs. 24%). 159 Women using non-supplied (traditional) methods were asked about the effectiveness of their current method relative to "modern methods like the IUD or the pill." More than a half considered their method more effective (22%) or equally effective (34%) compared with modern methods and only a third (33%) recognized that the IUD or the pill are more effective methods in preventing pregnancy (Table 8.4.2). In addition, 11% admitted that they did not know if their method is more or less effective. The traditional-method users' knowledge about contraceptive effectiveness has improved substantially since 1993. The proportion of users who recognized that their method is less effective than the IUD or pill almost doubled (from 19% to 33%) and the proportion of those who believed that traditional methods are more effective declined one-third from 33% to 22% (Figure 8.4.2). Similar to the 93RRHS, in the 99RRHS belief in high relative effectiveness (more or equally effective) of traditional methods was not significantly influenced by education (data not shown). Perceived relative effectiveness was highly associated with the desire to use another method in the future. As expected, women who did not want to change their current traditional method were also more likely to think highly of its effectiveness (63%). Those who said their preference for a future method would be either the IUD or another supplied method were the least likely to believe that their current method is relatively effective (32%-44%). These data indicate that, to increase the use of more effective methods, the national family planning program should concentrate on heightening public awareness of the relative effectiveness of various types of contraception, including contraceptive sterilization, disseminating information about the health effects of various methods, including their health benefits, and improving access to modern methods. 160 161 8.5 Reasons for Not Using Contraception Women and men currently in union mentioned a broad variety of reasons for not currently using contraception. Among women, the most common reasons given were related to fecundity impairment (43%) or pregnancy (17%) (Table 8.5 and Figure 8.5). Female fecundity impairment includes surgical and medical causes which prevent pregnancy and failure to conceive after at least two years of effort (without using contraception). Pregnancy-related reasons were the respondent being currently pregnant or breastfeeding, or the respondent desiring a pregnancy. About one of seven women reported lack of current sexual activity as the most important reason for not using a method. It is noteworthy that very few women reported reasons related to family planning as contributing to their decision not to use a method, such as lack of access to family-planning services (1%), personal or partner opposition to contraceptive methods (2%), and fear of side effects (1%). The reasons men in union reported they were not currently using a contraceptive method were similar to the reasons reported by women. The major reason, reported by 38% of male nonusers, was that their partner could not become pregnant. Similar to women, but more emphatic, 16% of men mentioned they desire a child soon. The other reasons cited by men were similar to those of women, except opposition or dislike of contraception (5% of men and 2% of women). Reasons for not using a method differed sharply by age group. Younger women in union were more likely to be either pregnant or in the postpartum period (44%) or were seeking to become pregnant (27%), whereas women aged 35-14 years were more likely to not be able to get pregnant 162 163 (68%). Among younger men in union, pregnancy-related reasons (presence or desire of pregnancy) were also most often mentioned (84% and 60%, respectively). Subfecundity and infecundity concerns were cited by more than a half of men aged 35 years or older. 8.6 Intention to Use Contraception among Nonusers The 99RRHS asked all women and men who were not using any contraceptive methods at the time of the interview if they planned to use any contraception in the next 12 months or later. Intention to use contraception in the future among non-users has to be taken into account when forecasting potential need for family planning services. Table 8.6.1 presents this intention among fecund women who are currently married or in consensual unions according to the number of living children they have. Intention to use contraception among male nonusers is shown in Table 8.6.2. Overall, two thirds (66%) of fecund women who are currently in union and who are not currently using a contraceptive method plan to use a method in the future, 48% of them within the next 12 months and 18% at a later time. About one in seven women in this group (13 %) were unsure if they wanted to use contraception in the future. Among male nonusers, the desire for future contraception was substantially lower than among women (43% vs. 66%), presumably because of their higher desire for pregnancy and dislike of contraception. 164 Intention to use contraception among women was influenced by the number of living children and future fertility preferences (Table 8.6.2). Nonusers who intended to begin contraceptive use tended to have one or two children (73% and 69%, respectively). However, more than one in two childless nonusers (57%) planned to use contraception in the future, but very few of them wanted to start within the next 12 months (14%). Conversely, most nonusers with one or more children who said that they wanted to use contraception, wanted to start within the next year. Intention to use contraception in the future among fecund female nonusers was slightly influenced by their desire for additional children—69% of those who desired no more children plan to use contraception compared to 62% among those who did not want to terminate fertility (Table 8.6.2). Those who wanted to stop childbearing were twice as likely to plan using contraception within the next 12 months as those who wanted more children (61% vs. 33%). Among male nonusers, there was no strong desire to use contraception among those who did not desire additional children (28%), chiefly because they reported not being currently sexually active or that their wives did not desire to use contraception. However, almost all those who did not want any more children and planned to use a method wanted to start within the next year. 165 Between the 93RRHS and the 99RRHS, there was a three-fold increase in the overall intention to use contraception in the future among fecund women who are non-users of contraception (Figure 8.6). In 1993, only 22% of fecund women in union intended to use contraception in the future, including 18% who planned to use a supplied method (8% intended to use an IUD, 5% the pill, 3% tubal ligation and 2% another modern method). This proportion rose to 66% in 1999, but the contraception method mix did not change substantially. One in two nonusers who desired to start using a method would choose the IUD or pill, whereas one in four would start using a traditional method. Interestingly, preference for a particular method was not influenced by fertility preferences (data not shown). The desire for long-term contraceptive methods (i.e., IUD and tubal ligation) was similar among those who wanted more children and those who did not, partly because among women who want no more children, a significant proportion was over 40 years of age and less inclined to choose a supplied method. Of the 43% of male non-users who stated that they wanted to use a method in the future, 9% would like to use condoms, 8% pills, 7% long-term methods (4% IUD and 3% tubal ligation), 3% other supplied methods, and 15% a traditional method (data not shown). 166 8.7 Recent Trends in Contraceptive Use The 99RRHS questionnaire included a detailed five-year contraceptive "calendar", whereby contraceptive use, pregnancy events, and marital status were recorded monthly starting with January 1994 through the date of the interview. These data were used to compute mid-year contraceptive prevalence rates for the five years preceding the survey year (1994-1999), using the reported prevalence in the month of July in each year. During these five years, there was a steady and relatively strong rise in the overall contraceptive prevalence among all women but very little change among women in formal or consensual unions. Between July 1994 and July 1998, contraceptive prevalence rose from 39% to 50% among all women and ranged between 64% and 67% among women in union (Table 8.7 and Figure 8.7.1). Most of the increase among all women was the result of higher use of modern methods. Contraceptive prevalence of modern methods has rose by 61% from 1994 among all women (from 15% to 24%) while the use of traditional methods remained basically unchanged. However, there were substantial differences in contraceptive method mix by marital status. The use of modern methods rose by 26% among women in union (from 23% to 29%), and use of traditional methods decreased by 13% (from 42% to 36%), translating into little change of the overall prevalence. 167 168 Conversely, among women not in union, prevalence of both modern and traditional methods increased substantially, but the 2.5-fold increase in contraceptive prevalence of modern methods (from 6% to 15%) was far greater than the increase in use of traditional methods (from 5% to 9%). Overall, most of the increase in modern prevalence was due to a net growth in pill and condom use, especially among unmarried women. The pill prevalence among al
Looking for other reproductive health publications?
The Supplies Information Database (SID) is an online reference library with more than 2000 records on the status of reproductive health supplies. The library includes studies, assessments and other publications dating back to 1986, many of which are no longer available even in their country of origin. Explore the database here.