Thailand - Demographic and Health Survey - 1988

Publication date: 1987

Thailand Demographic and Health Survey 1987 B Institute of Population Studies Chulalongkorn University ® DHS Demographic and Health Surveys Institute for Resource Development/Westinghouse f' \ THAILAND DEMOGRAPHIC AND HEALTH SURVEY 1987 by Napaporn Chayovan Peerasit Kamnuansilpa John Knodel Institute of Population Studies Chulalongkorn University Bangkok, Thailand Institute for Resource Development/Westinghouse Columbia, Maryland USA May 1988 Pre face Since its founding in 1966, the Institute of Population Studies (IPS) has been responsible for a number of national surveys focusing on family planning and the demographic and socio-economic situation in Thailand. These surveys included the National Longitudinal Study of Social, Economic and Demographic Change conducted in 1969/70 and again in 1972/73, the Survey of Fertility in Thailand conducted in 1975 as part of the World Fertility Survey, and the National Survey on Family Planning Practices, Fertility and Mortality in 1979. The Thai Demographic and Health Survey (TDHS), conducted in 1987, represents a continuation of this tradition in survey taking at IPS. At the same time, however, the TDH$ has also broadened the Institute's experience in several ways. Not only is it the largest survey in terms of the number of respondents undertaken so far by IPS, but it is the first large scale survey in Thailand to deal in significant detail with health topics including anthropometric measures of children under 3 years of age and their mother~. Because of the inclusion of health topics, the IPS staff has gained new experience and skills which should prove valuable in the future when new surveys are conducted. The purpose of the TDHS is to provide current and accurate data on fertility, mortality, family planning and selected indicators of health status to be used for program assessment and guidance and for scientific analysis to further our understanding of the demographic and health situation in Thailand. We hope that this report makes a significant contribution to this goal. As comprehensive as the report is, however, it represents only a small portion of the potential information and analysis that can be derived from the data collected by the TDHS. In recognition of this, IPS will undertake two broad further analysis projects during the coming year, both funded by the Population Council. One project will focus on demographic and family planning topics while the other will be concerned with health topics. Each project consists of a set of separate analyses dealing with specific subtopics under the two general project rubrics. Together, the two projects will involve many staff members of IPS and will also draw on colleagues at other organizations with expertise in the relevant areas. Thus the current report should be viewed as just the beginning rather than as the final product of our effort to take full advantage of the valuable data collected by the TDHS. AS with any project as large as the TDHS, the skills and efforts of many qualified and dedicated persons had to be mobilized to carry it out successfully. A list of the TDHS staff is provided as an Appendix of this report and therefore there is no need to repeat their names here but rather to acknowledge with gratitude their collective effort. Special recognition, however, is due Dr. Napaporn Chayovan as the one person who on a daily and virtually full-time basis has guided the TDHS through all its stages from initial formulation to the printing of this report. IPS is indeed grateful to her for her tireless and dedicated efforts. Sincere appreciation is also extended to Prof. John Knodel who has provided valuable advice from the initial stage of the project and devoted a great deal of his effort working with Dr. Napaporn Chayovan on every stage of the project including the data analysis of the report. ii Besides the official staff of the TDHS, many people and organizations have been helpful at various stages in providing assistance and advice. The biggest debt is owed to the Institute for Resource Development (IRD), Westinghouse for providing funding and technical assistance without which the TDHS would not have been carried out. A list of consultants, including those provided by IRD, is included in the Appendix along with the TDHS staff. Each consultant not only provided valuable guidance but did so in a professional and friendly way. We have learned much from them. In addition, we would like to thank the National Statistical Office and the Ministry of Interior for providing information necessary for implementing the sample design. The Division of Nutrition, Ministry of Public Health, kindly lent us equipment for weighing the children. A number of individuals deserve mention for help provided in different aspects of the project. The staff of the USAID office in Bangkok (in particular Edwin McKeithen, Karoon Rugvanichje, and Narintr Tima) provided useful advice and encouragement throughout the course of the project. Mr. Art Wichienchareon at ESCAP helped in transferring the data from diskettes to computer tapes. Ansley Coale, Ronald Freedman, Carl Frisen, Robert Hanenberg, Chintana Petcharanonda, and Nicholas Wright provided useful information or comments that assisted in the writing of this report. Last but not least, Ms. Porntip Sopon deserves our gratitude for patiently typing the many versions that this manuscript went through, often coming in on weekends or staying after hours to ensure that the report would be issued on time. Bhassorn Limanonda, Ph.D. Director Institute of Population Studies Chulalongkorn University May 1988 i i i Contents Pre face . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . L i s t of Tab les . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . L i s t o f F igures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Suaaary . Chapter 1 Background 1.1 Country Setting . 1.2 Population . 1.3 Population and Family Planning Policies and Program . 1.4 Health Priorities and Programs . 1.5 Survey Objectives . 1.6 Organization of the Survey . 1.7 Background Characteristics of the Surveyed Women . Nupt ia l i ty and Other Proximate Determinants 2.1 Nupt ia l i ty Pat terns and Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Breastfeeding and Postpartum Insusceptibility . Fer t i l i ty 3.1 Cur rent Fer t i l i ty Leve ls and Trends . . . . . . . . . . . . . . . . . . . . . . . . . 3 .2 Cumulat ive Fer t i l i ty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 .3 Age a t F i r s t B i r th . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Fer t i l i ty Regu lat ion 4.1 Contraceptive Knowledge . 4.2 Contraceptive Use . 4.3 Knowledge of the Fertile Period . 4.4 Timing of Sterilization . 4.5 Source of Contraception . 4.6 Reasons for Discontinuation . . 4.7 Attitude toward Becoming Pregnant . 4.8 Personal Reason for Non-Use . 4.9 Intentions for Future Use of Cont racept ion . . . . . . . . . . . . . . . . . . 4.10 Family Planning Messages on the Radio . Fertility Preferences 5.1 Desire for Additional Children . 5.2 Future Need for Family Planning . 5.3 Preferred Number of Children . 5.4 Fertility Planning Status of Births and Unwanted Fertil ity . Page i i i v v i x iv 1 3 6 7 7 7 8 19 24 28 34 43 46 48 53 60 62 63 66 67 68 69 71 76 81 82 85 iv 6 Mortality and Health 6.1 6.2 6.3 6.4 6.5 6.6 6.7 - 6.8 Infant and Child Mortality . Prenatal Care . Assistance during Delivery . Immunization . Diarrhea Prevalence . Diarrhea Treatment . ORT Knowledge . Anthropometric Measurements of Length and Weights . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendices Appendix A: A.I A.2 A.3 A.4 A.5 A.6 Appendix B: Sample Design and Implementation The Study Popu la t ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample Size and Allocation . The Frame and Sample Selection . Sample Outcome . Weighting of Sample Results . TDHS Sample Provinces . Comparison of Sample Characterist ics with External SOUrCes B.I Age and Sex Distribution . B.2 Marital Status Distribution . B.3 Educational Level . B.4 Fertil ity . Appendix C: Appendix D: D.1 D.2 Appendix E: Estimating Sanplin~ Er rors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Survey Instruments The Household and Individual Questionnaire . The Community Questionnaire . TDHS Sta f f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 89 94 97 98 i0~ 108 ii0 iii 127 129 129 130 133 134 139 140 144 147 149 157 167 202 216 V List of Tables Table 1.1 1.2 1.3 2.1 2.2 2.3 2.4 2.5 2A.1 3.1 3.2 3.3 Number of households and women selected and successfully interviewed, by reporting domain . Percent d i s t r ibut ion of ever -marr ied women accord ing to se lec ted background character i s t i cs . . . . . . . . . . . . . . . . . . . . . Percent distribution of ever-married women according to education, by selected background characteristics . Percent distribution of ever-married and all women according to current marital status, by current age . Percent d i s t r ibut ion of a l l women accord ing to age a t f i r s t marr iage ( inc lud ing those repor ted in household as never marr ied) and median age at f i r s t marr iage , by cur rent age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Median age at first marriage among all women aged 25-49 years (including those reported in the household as never married), by current age and selected background characteristics . Percentage of births in the last 3 years whose mother are still breastfeeding, still postpartum amenorrheic, still abstaining, and insusceptible, by months since birth . Preva lence / inc idence es t imates of mean number of months of b reast feed ing , postpar tum amenorrhea and postpar tum abst inence , by se lec ted background character i s t i cs . . . . . . Percentage of births in the last 3 years whose mothers are still breastfeeding, and still postpartum amenorrheic, abstaining, and insusceptible, by single months since birth . Fertility rates for 12, 24 and 60 months preceding the survey, for all women (including never-married women), by age of women at time of childbirth . Mean number of ch i ld ren ever born to a l l women ( inc lud ing never -marr ied) aged 40-49 and to ta l fe r t i l i ty ra tes fo r se lec ted per iods and for 60 and 24 months preced ing the survey , by se lec ted background character i s t i cs . . . . . . . . . . Age-period fertility rates (per 1,000 women including never-married), by age at time of childbirth . Page 18 20 23 26 27 28 29 32 33 36 38 40 vi 3.4 3.5 3.6 3.7 3.8 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 Percentage decline in fertility rates between successive five year periods prior to the survey and the period 0-4 years prior to the survey, by age at time of childbirth . Percent d i s t r ibut ion of ch i ld ren ever born among a l l women, ever -marr ied women, and cur rent ly marr ied women, by current age . Mean number of ch i ld ren ever born to ever -marr ied women, by age a t f i r s t marr iage and durat ion s ince f i r s t marriage . Percent d i s t r ibut ion of a l l women ( inc lud ing never -marr ied) accord ing to age a t f i r s t b i r th ( inc lud ing the category "no b i r th" ) , by cur rent age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Median age at first birth among all women (including never-married) aged 25-49 years, by current age and selected background characteristics . Percentage knowing any method, knowing any modern and knowing specific contraceptive methods, among ever-married and currently married women, by current age . Percentage of ever-married women aged 15-49 knowing specific methods and any method, by selected background characteristics . Percent distribution according to the main problem perceived in using methods (if any), by method, for women who have ever heard of the method . Percent distribution of women who know a specific method according to supply source named (if any) . Percentage of women who have ever used specific methods among ever-married and currently married women, by current sge . . . ° , ,o . . . . .Q .eo .o t . . . .o . , t , . . . . ,4 . , . . . . . . i ° . * ,e .91 . . Percent distribution of currently married women according to contraceptive method currently used, by current age . Percentage currently practicing specific methods of contraception among currently married women aged 15-44, 1969-87 . Percent distribution of currently married women aged 15-44 according to the contraceptive method currently used, by selected background characteristics . Page 40 44 45 46 47 50 51 52 54 56 56 57 59 v i i 4.9 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 4.19 Percent distribution of ever-married women according to number of living children at time of first use of contraception, by current age . Percent distribution of ever-married women aged 15-49 and women who have ever used periodic abstinence according to knowledge of the fertile period during the ovulatory cycle . Percent distribution of sterilized women according to age at the time of sterilization, by the number of years since the operation . Median age at sterilization for women sterilized before selected ages, by the number of years since the operation . Percent distribution of all current users of supply or clinic methods of contraception according to most recent source for supply, by method . Percent distribution according to type of dissatisfaction with the service (if any) among current users and past users who obtained a method at a source, by type of source last visited . Percent distribution of women who have discontinued a contraceptive method in the last 5 years according to the main reason for last discontinuation, by specific method . Percent distribution of non-pregnant, non-abstaining, non-contracepting, currently married women according to attitude toward becoming pregnant in the next few weeks, by number of living children . Percent distribution of non-pregnant, non-abstaining, non-contracepting currently married women who would be unhappy if they become pregnant according to the main reason for non-use, by current age . Percent distribution according to intentions to use in the future among currently married women not currently using any method, by number of living children (including any current pregnancy) . Percent distribution according to preferred method among currently married women not currently using a contraceptive method but who intend to use in the future, by timing of intended use . Page 60 61 62 63 64 65 66 67 69 70 71 viii 4.2o 4.21 4.AI 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 Percent distribution of women according to whether they have heard a radio message about family planning during the last month, by selected background characteristics . Percentage of women who believe that it is acceptable to have messages about family planning on the radio, by current age and selected background characteristics . Percent distribution of currently married women aged 15-49 according to the contraceptive method currently used, by selected background characteristics . Percent distribution of currently married women according to whether they want more children and the certainly of their preference, by number of living children (including any current pregnancy) . Percent distribution of currently married women according to fertility preferences, by number of living children (including any current pregnancy) . Percent distribution of currently married women according to whether they want more children, by current age . Percentage of currently married women who want no more children (including sterilized) by years since first marriage and, for all currently married women, standardized for years since first marriage, by background characteristics . Percentage of currently married women who want no more children (including sterilized) by number of living children (including current pregnancy), by background characteristics . Percentage of currently married women who are potentially in need of family planning (i.e., who are not contracepting and who want no more births or want to postpone the next birth for 2 or more years) and the percentage who are in need and who intend to use family planning in the future, by background characteristics . Percent distribution of ever-married women according to preferred number of children and mean preferred number of children for ever-married women and currently married women, by number of living children (including any current pregnancy) . Mean preferred number of children for ever-married women, by current age and background characteristics, and for currently married women married less than 5 years, by background characteristics . Page 72 73 74 76 77 78 79 80 82 83 84 ix 5.9 5.10 5.11 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 Percent distribution of all pregnancies resulting in live births (including current pregnancy) in last five years according to contraceptive practice and planning status, by birth order . Percentage of women who had a birth in the last 12 months who wanted a child then, later, or wanted no more children, by birth order . Total wanted fertility rates and total fertility rates based on women 15-44 only, for the five years preceding the survey, by selected background characteristics . Infant and childhood mortality estimates by time period Infant and child mortality estimates, 1977-1987 based on the TDHS and infant mortality, 1985-86 based on the survey of population change (SPC), by selected background characteristics . Demographic differentials in infant and child mortality, 1977-1987 . Hean number of children ever-born, surviving, and dead, and proportion of children dead among those ever-born, among ever-married women, by current age of mother . Percent distribution of births in the last 5 years according to the type of prenatal care for the mother and percentage of births whose mother received a tetanus toxiod injection, by selected background characteristics . Percent distribution of births in the last 5 years according to type of assistance during delivery, by selected background characteristics . Percent distribution of children under 5 years of age according to immunization status and method of reporting immunization status, by age of child . Among all children under 5 years of age with health record cards or booklets, the percentage for whom BCG, DPT, polio and measles immunizations are recorded, by age of child . Percent distribution of children 12 to 59 months of age according to immunization status and method of reporting immunization status, by selected background characteristics . Page 86 87 88 90 91 93 94 95 97 i00 i00 102 6.10 6.11 6.12 6.13 6.14 6.15 6.16 6.17 6.18 6.19 A.I Among children 12 to 59 months of age with health record cards or booklets, the percentage for whom BCG, DPT, polio and measles immunizations are recorded, by selected background characteristics . Percent distribution of children 12-23 months of age according to immunization status and method of reporting immunization status, by selected background characteristics . Among children 12 to 23 months of age with health record cards or booklets, the percentage for whom BCG, DPT polio and measles immunizations are recorded on the health card, by selected background characteristics . Percentage of children under 5 years of age reported by the mother to have had diarrhea in the past 24 hours and the past two weeks, by selected background characteristics . Percentage of children under 5 years of age who had diarrhea in the past two weeks consulting a doctor or nurse and the percentage receiving different treatments as reported by the mother, by selected background characteristics . Percentage of mothers of children under 5 years of age who know about ORT, by education and selected background characteristics . Percentage of children aged 3-36 months in each standard deviation category of height-for-age using the international NCHS/CDC/WHO reference by selected background characteristics . Percentage among children aged 3-36 months in each standard deviation category of weight-for-height using the international NCHS/CDC/WHO reference by selected background characteristics . Percentage among children aged 3-36 months in each height-for-age standard deviation category by each weight-for-height standard deviation category (Waterlow classification) using the NCHS/CDC/WHO international reference . Percentage among children aged 3-36 months in each standard deviation category of weight-for-age using the international NCHS/CDC/WHO reference by selected background characteristics . Number of households and women selected and successfully interviewed, by sampling domain . Page 103 104 105 107 109 111 116 120 121 126 130 xi A.2 A.3 A.4a A.4b B.I B.2 B.3 B.4 B.5 B.6 B.7 B.8 B.9 Outcome of the sample of households . Comparison between sample distribution and "standard" distribution, by region and urban-rural status . Derivation of "standard" distribution for total population . Derivation of "standard" distribution of ever-married women . Percent distribution according to age, by sex, comparison of results from the TDHS household sample, the NESDB projected population, the Survey of Population Change (SPC) and the census . Percent distribution of ever-married women age 15-49 according to age-group, comparison of.results from TDHS, the Survey of Population Change (SPC), and the census . The sex ratio (males per i00 females), by age, comparison of results from the TDHS household sample, the NESDB projected population, the Survey of Population Change (SPC) and the census . Percentage ever-married, by age and sex, comparison of results from TDHS, the Census and the Survey of Population Change (SPC) . Percent distribution of ever-married women aged 15-49 according to marital status, by age, comparison of results from TDHS, the Survey of Population Change (SPC) and the census . Percent distribution according to educational level, by age and sex, comparison of results from TDHS and the Survey of Population Change (SPC) . Mean number of children ever born, and percent surviving for ever-married women, by age of woman, comparison of results from TDHS, the Third Contraceptive Prevalence Survey (CPS3), Survey of Population Change (SPC) and the census . Total fertility rates to selected age, comparison of results from TDHS and selected other sources, 1970-1986 . Mean number of children ever born according to whether alive or dead and whether living in the mother's household or elsewhere, by age of mother, comparison of results from TDHS and CPS3 . Page 135 137 138 138 141 143 143 145 147 148 150 152 155 xii C.l C.2a C.2b C.2c C.2d C.2e C.2f C.2g C.2h C.2i C.2j C.2k List of variables for sampling errors for TDHS . Sampling errors for the total population . Sampling Sampling Sampling Sampling Sampling Sampling Sampling Sampling Sampling Sampling errors for the urban population . errors for the rural population . errors for the north region . errors for the northeast region . errors for the central region . errors for the south region . errors for Bangkok . errors for woman aged 15-24 . errors for woman aged 25-34 . errors for woman aged 35-49 . Page 159 160 160 161 161 162 162 163 163 164 164 165 xiii List of Figure Figure 1.1 1.2 1.3 3.1 6.1 6.2 6.3 6.4 6.5 6.6 6.7 The organizational structure of the survey . Work plan and actual performance schedule . TDH$ sample districts . Comparison of the trend in the TFR based on the TDHS with trends based on data from SOFT, the 1980 census, and vital registration . Reported ages of children weighted and measured . Mean nutritional status by age . Stunting among children by area of residence . Stunting among children by education of mother . Crosstabulation of weight-for-height and height-for-age (Waterlow table) . Percent simultaneously stunted and wasted by background variables . Population height/age and weight/height compared to international reference . Page 8 9 11 42 113 i15 118 119 122 124 125 xiv Summary The Thai Demographic and Health Survey (TDHS) was a nationally representative sample survey conducted from March through June 1988 to collect data on fertility, family planning, and child and maternal health. A total of 9,045 households and 6,775 ever-married women aged 15 to 49 were interviewed. The results indicate that the longer term decline in fertility that started two decades ago has been continuing during recent years. The very low recent total fertility rate of 2.21 estimated by the TDHS for the 24 month period preceding the survey, however, appears to be lower than evidence from most other sources would indicate and could reflect some understatement of births in the survey. The much higher cumulative fertility of women presently at the end of the reproductive ages, averaging 4.42 children ever born to ever-married women aged 40-49, underscores the recent and substantial nature of Thailand's fertility decline. Age at first marriage, particularly among women, has been increasing moderately over the last two decades. The age by which half of all women are married increased from 19.7 to 21.1 between the cohort currently aged 45-49 and the cohort aged 25-29. Breastfeeding is very common in Thailand and lasts on average almost 17 months. Nevertheless, the average duration of postpartum amenorrhea is only 7 months. Both breastfeeding and postpartum amenorrhea are considerably shorter in urban than rural areas. Postpartum abstinence is relatively short and differs little between urban and rural couples. Contraceptive awareness is virtually universal in Thailand and almost every woman knows a source where modern contraceptive methods can be obtained. Over 80 percent of ever-married women have ever practiced contraception. Among currently married women aged 15-44, 67.5 percent were currently using a contraceptive method thus continuing the steady increase in contraceptive prevalence evidenced by previous surveys over the last two decades. Of women currently using contraception, fully 97 percent were practicing a modern method and over 40 percent were either sterilized or had a husband who was. Contraceptive prevalence is highest in the North (75 percent) and lowest in the South (52 percent). The most outstanding differential is by religion, with the Moslem minority characterized by a prevalence rate only half of that of the Buddhist majority. Among current users of modern contraception, at least 82 percent obtained their method from a government source. Given the very high prevalence of contraception, it is not surprising that there does not appear to be a great deal of unmet need for family planning. Most non-pregnant, non-abstaining women who do not want to get pregnant but are not using contraception appear to be at low risk of pregnancy because they are in a state of postpartum amenorrhea, are subfecund or engage only infrequently in sex. Nevertheless, 14 percent of women who gave birth during the 12 months preceding the survey said the birth was unwanted at the time of pregnancy and 16 percent indicated the birth was mistimed (i.e. wanted but at a later time). since ago. The preferred family size has fallen to the lowest level recorded the first national survey collected such information almost two decades Among all currently married women aged 15-49, the average preferred family size is 2.8 children and among recently married women is only 2.3 children. Two thirds of currently married women say they want no more children. Infant mortality during the preceding five years as calculated from the TDHS is only 35 per 1,000 births, down from 55 per 1,000 live births i0 to 14 years prior to the survey. The estimate for the recent period is low compared to other sources and may reflect some under-reporting of infant deaths. Among children under age five, 6 percent experienced diarrhea during the preceding 24 hours and 16 percent experienced diarrhea during the preceding two weeks, of whom 40 percent received oral rehydration therapy. ApproximatelM 85 percent of children aged 1-4 years received at least one immunization. Among the 26 percent of children aged 1-4 years for whom health record cards or booklets with immunization data were available, the vast majority received BCG, the third dose of DPT, and the third dose of polio vaccines. Only about half, however, have been immunized against measles. Mothers of 65 percent of the births during the last five years received tetanus toxoid injections, 77 percent received prenatal care, and 66 percent were assisted by medical professionals at the time of delivery. 2 Chapter 1 Background 1.1 Country Setting* Thailand is a tropical country in the Indo-Chinese peninsula of southeast Asia bordered by Kampuchea and Laos on the east and northeast, by Burma on the west and northwest, and by Malaysia on the south. Thailand includes tropical rain forests, agriculturally rich plains, and forest-clad hills and mountains. The patterns of rivers and mountains divide Thailand into four natural regions: the mountainous north; the northeast, consisting primarily of the Korat plateau; the central region, consisting primarily of the Chao Phraya Basin; and the south, consisting of the long peninsular extension of Thailand south from the Chao Phraya Basin to the Malaysian frontier. Unlike many other developing countries and all its southeast Asian neighbors, Thailand has never been colonized by a foreign power. There have been periodic invasions by Burmese and Khmers in the more distant past and a brief occupation by the Japanese during World War If, but by and large the country has been an independent nation throughout its history. A common religion is one of the most important factors contributing to the relative cultural homogeneity of the Thai population. The large majority of the population (95.9 percent in 1980) professes Buddhism as its religion. Most non- Buddhists adhere to Islam, which is practiced by about 4 percent of the population. Most Thai Muslims, about 80 percent, live in the south, where they constitute the majority of the population in the four southernmost provinces and make up one fourth of the total population of the south, despite their small percentage nationally. About half of the Muslims living outside the south reside in Bangkok and most of the rest are in the central region. Muslims are a negligible proportion of the populations of the north and northeast. In no region do Christians or members of other religions constitute as much as 1 percent of the population. Administratively, Thailand is currently divided into seventy-three provinces (changwat), one of which is the Bangkok metropolis. Each province is further subdivided into districts (amphur), townships (tambol), and villages (muban). Some areas are also designated as municipalities, including all provincial capitals. Economically and politically the country is dominated by Bangkok, the only major urban area. Although it is located geographically within the central region, for most purposes the Bangkok metropolis is usefully considered a distinct region on its own because its population differs considerably in many characteristics from the remainder of the central region. In socioeconomic terms, Thailand's features are typical of the developing world. Like many other Third World nations, Thailand has been experiencing rapid and fundamental social and economic change as it undergoes *This section is based largely on Knodel, Chamratrithirong and Debavalya, 1987, Chapter 3 with some updating of statistical indices based on World Bank, 1987. the process of modernization and development and becomes increasingly enmeshed in the world economic system. GNP per capita, was $800 in 1985, according'to the World Bank, placing it squarely in the middle-range among those developing countries classified as being lower-middle-income. Thailand's rate of economic growth in recent decades, however, has been well above the average for developing countries generally. Despite increasing proportions of the population living in urban areas and engaging in non-agricultural pursuits, the country remains predominately rural and agrarian. According to World Bank statistics, 82 percent of the population lived outside areas classified as urban in 1985 and 71 percent of the labor force was engaged in agriculture in 1980. With respect to several key health indicators, Thailand's situation appears relatively favorable for a developing country. For example, life expectancy at birth for 1985 was estimated as 64 years which is distinctly better than the average for other lower-middle-income developing countries. In this connection it is notable that the health-service system in Thailand is a complex mixture of public and private providers. In urban areas, private health services are very important. In addition, the Ministry of Interior administers a variety of public-health facilities in Bangkok and other municipalities. For the large rural population, however, the major source of service is the Ministry of Public[Health, operating through an extensive network of outlets including regional health centers, provincial and district hospitals, and local health stations at the township level. The public health system has expanded considerably in the last two decades. For example, the number of government health stations, which are virtually all located in rural areas, more than tripled between 1965 and 1985, at which time there were over 7,000 such stations. In addition, the number of government hospitals more than doubled to over 500 units during the same period, with the increase almost entirely at the district level. In the present report, the most important background variables employed in the tabulations are rural-urban residence, region, education and religion. The religious distribution of the population was discussed above. Each of the other three characteristics are now considered in some detail. l.la Rural-Urban Distribution There is no question that Thailand has been and continues to be a predominantly rural society and is relatively so even within the context of the developing world in general. Defining precisely what is to be considered as urban and rural areas, however, is not entirely straightforward. There is no official definition of rural and urban in Thailand. The usual practice is to define the officially designated municipal areas, including the entire Bangkok metropolis, as urban, and the remainder of the country as rural. This definition is increasingly being criticized as unrealistically narrow and most observers agree that it results in an underestimation of the "true" urban population. The basic problem with a definition based only on municipal areas (including Bangkok) is that it is becoming increasingly out of date. There has been almost no change in the number of officially designated municipalities over the last several decades even though the nature of many places in the nonmunicipal category has changed considerably, including places both on the fringe of municipal areas and elsewhere. Instead, localities that achieve a 4 minimum population size and density and develop some urban characteristics are frequently designated as "sanitary districts." As such, they remain in the rural category when rural is defined exclusively in terms of municipalities. (In 1980, 17.0 percent of Thailand's total population lived in municipal areas, including the Bangkok metropolis, 6.6 percent in large sanitary districts, and 2.7 percent in small sanitary districts.) In addition, there has been insufficient redefinition of the boundaries of existing municipal areas to allow for their defacto expansion. One partial remedy is to include officially designated sanitary districts, or at least the larger ones, as urban. In the present report, however, analyses in subsequent chapters utilize the usual, more limited definition of urban based only on municipal areas to maintain comparability with previous studies. l.lb Regional Variation In many important respects, the Thai population is relatively homogeneous. The vast majority adhere to Buddhism, are ethnic Thais, and speak some version of the Thai language. Moreover, the official central Thai language is understood virtually everywhere. There is generally a sense of national identity reinforced by a widespread allegiance to the monarchy, which serves as an effective symbol of national unity. Nevertheless, to varying extents, cultural and socioeconomic differences characterize the four major regions. The most obvious cultural difference relate to regional dialects. Distinctive dialects are spoken in the north, the northeast, and the south, each of which differs from the standard Thai spoken in the central region. In addition, among Moslems in three of the four southernmost provinces Malay is common. Bangkok, with ii percent of Thailand's population is typically in a class of its own with respect to most socio-economic indicators. Of the four major regions excluding Bangkok, the central region, with 21 percent of the population, generally ranks the highest in socioeconomic terms. It is also the cultural center of the nation, closest in physical and psychic distance to the Bangkok metropolis. The central plain is the heartland of rice cash crop in a country where rice is the mainstay of the economy. Substantial parts of the Chao Phraya Basin have benefited recently from a major irrigation project that has opened up wide expanses of land to the possibility of rice double-cropping. The poorest region is the northeast, which contains 35 percent of the total Thai population. It is the driest region and suffers from periodic droughts combined with a lack of a well developed irrigation system. Although lower primary education is close to universal in all regions, discrepancies still exist with respect to the percentage of children who continue their education beyond this level. For example, the northeast ranks lowest in the percentage of young adults who continued beyond primary education. The north is the second poorest region and contains 21 percent of the total population. Because of its mountainous terrain, rice farming in many areas is concentrated in densely settled narrow valleys and involves particularly intensive agricultural practices. Communally run, small-scale water control systems are common and perhaps are part of the reason why social commitment to the structural organization of the valley community is generally judged to be greater in the north than elsewhere. 5 The smallest region in terms of both land area and population is the south, which contains 12 percent of the population and tends to rank higher on most socioeconomic indexes than either the north or the northeast. It is the region of heaviest rainfall and is least dependent on rice as either a subsistence or export crop. Tin mining, rubber planting, and coastal fishing are important contributors to the local economy. l.lc Education Universal compulsory education in Thailand was enacted into law in 1921. Implementation has been a gradual process but by 1980 was virtually complete. Government efforts have focused mainly on primary education, and until recently the highly educated segment of Thai society consisted almost exclusively of a small elite in Bangkok. This has changed to some extent in recent decades, especially since the establishment of a large open-admissions university in Bangkok, and the opening of regional universities. In the last few decades, education has been a vital government activity representing a critical part of the overall effort to accelerate social development. Nevertheless, advancing through the educational system is still a long and difficult task, especially for rural Thais. After finishing primary education (presently six years) in a village school, a student would typically have to enter a secondary school in a district or provincial center located a considerable distance away. After completing grade 9 or 12, depending on whether vocational or university education was sought, a student often would need to move to Bangkok or at least to a regional center to study further. Until recently, school attendance was compulsory only through the first four grades, known as "lower primary" education. During the 1970s, as part of a reform of the educational system, primary education was reduced from seven to six grades and the distinction between lower and upper primary levels eliminated. Compulsory attendance has also been extended and now covers the entire six primary years. Implementation of the increase in the number of years of compulsory education has been an ongoing process rather than a sudden universal change, but by the mid-1980s was largely in effect. Since the change is quite recent, it is only starting to have a major impact on the educational distribution of the adult population. In 1980, the majority (59 percent) of Thais aged 15 or over had exactly a fourth-grade education and only 21 percent had attended more than fourth grade. Among women in the major reproductive ages 20-44, 17 percent had more than a fourth-grade education in 1980 compared to 70 percent who had exactly a fourth-grade education. 1.2 Population According to recent population projections by Thailand's National Economic and Social Development Board (NESDB), Thailand's population was 54 million in 1987. This represents more than a sixfold increase since 1911, when the population was only 8 million according to the first census. As in many developing countries, population growth, particularly since World War II, has been relatively rapid. Although the intercensal rates of growth can be considered as only approximate due to uncertainties about the completeness of the census enumerations, it seems likely that the rate of growth peaked at over 3 percent per year during the 1950s and the early 1960s. By the first half of the 1980s, according to the recent NESDB estimates, the population growth rate 6 had declined to below 2 percent. This reduction in the growth rate reflects a rapid and substantial decline in fertility over the last two decades. 1.3 Population and Family Planning Po l i c ies and Program During most of the present century, Thailand's official stance on population was pronatalist. Following a report by a World Bank economic mission in 1959 recommending that the government seriously consider the adverse effects of high population growth on economic development, officials started to reconsider the government's position. This culminated with the declaration in 1970 by the Thai Cabinet of an official policy to reduce population growth and the National Family Planning Program was formally established under the auspices of the Ministry of Public Health. A number of steps had been taken prior to 1970, however, that in effect constituted the beginning of a government- sponsored program to promote family planning. Since family planning activities under the jurisdiction of the Ministry of Public Health are integrated into child and maternal health services, the program was able to take advantage of the existing extensive infrastructure available for government health services in general. 1.4 Health P r io r i t ies and Programs The Ministry of Public Health is responsible for the provision of health care services, disease prevention and control, and other welfare services related to the health of the population. It has been the policy of the government to expand and provide medical services to cover the population at all levels of administration. The current Sixth Five Year Plan emphasizes the quality of life for all through the fulfillment of basic minimum needs. The targets for meeting these basic minimum needs during the current Five year Plan are: i) Family members consume sufficient nutrition and safe food; 2) Every family member has appropriate shelter and environmental conditions; 3) People have the opportunity to receive basic services essential for daily living; and 4) Seventy-five percent of married women in reproductive years practice family planning and child spacing while the two-child family norm is promoted. The current national health development programs include health administration, health services, community participation in primary health care, technology development for disease control, and health promotion and consumer protection. These programs are designed to achieve the basic minimum need targets, reduce mortality, morbidity and incidence rate of diseases identified as major health problems, reduce the population growth rate to 1.3 percent by 1991, and expand and promote health personnel and infrastructure. Emphasis is also given to lower morbidity of vaccine preventable diseases common among new born babies such as diphtheria, tetanus, pertussis, polio and measles. 1.5 Survey Objectives The Thai Demographic and Health Survey (TDHS) was undertaken for the main purpose of providing data concerning fertility, family planning and maternal and child health to program managers and policy makers to facilitate their evaluation and planning of programs, and to population and health researchers to assist in their efforts to document and analyze the demographic and health situation. It is intended to provide information both on topics for which comparable data is not available from previous nationally representative surveys as well as to update trends with respect to a number of indicators available from previous surveys, in particular the Longitudinal Study of Social Economic and Demographic Change in 1969-73, the Survey of Fertility in Thailand in 1975, the National Survey of Family Planning Practices, Fertility and Mortality in 1979, and the three Contraceptive Prevalence Surveys in 1978/79, 1981 and 1984. 1.6 Organization of the Survey Thai Demographic and Health Survey (TDHS) is carried out by the Institute of Population Studies (IPS) of Chulalongkorn University with the financial support from USAID through the Institute for Resource Development (IRD) at Westinghouse. The Institute of Population Studies was responsible for the overall implementation of the survey including sample design, preparation of field work, data collection and processing, and analysis of data. IPS has made available its personnel and office facilities to the project throughout the project duration. It serves as the headquarters for the survey. Figure 1.1 shows the organizational structure of the survey and Figure 1.2 shows the detailed work plan. 1.1 ~e organizational structure of the survey i i i Techn ica l Staff A0ainistrative Staff ' I I ! I I , I I I~S Fieldwork Data proeessi~E Assistant Project Of f i ce : I Administratur (1) Accountants (2) , I I - Research - Field director(1) Data Processing associates (5) I Supervisors (2) I - Team supervisors(5) S - Research - Te~ assistants Programmers(2) assistants (2) Supervisors (7) i I S - Intervie.ers(40) Editors and keyers (8) Fi4~m 1.2 Work plan and actual performance schedule (1) (2) (3) (4) ~'t/vlti~ P lmmd Actual Preperation of the survey i.I Translation of questionnaires 1.2 Preparation of supervisors' and intervie~rs' manuals 1.3 Sample design 1.4 Printing of the questiamaires 1.5 Recruitment of project assistants Pretest 2.1 Suporvisors and assistants' training 2.2 Pretest on the questionnairss Recruitment of field staf f Training of field staff ~pervisc~'s and assistants' Interviewers (5) Imp]~tat ion of fieldwork (6) Editing, data entry and consistency check (7) Tabulation and analysis (8) Report writing (9) Report preduction (10) Seminar presenting results Nov. 1986- Feb. 1987 Jan. 1987 Feb. 1987 March 2nd- 3rd week 4th week of March-Rid June 1987 July-Nov. 1987 Dec. 1987-Feb. 1988 March-April 1988 May 1988 June 1988 NOv. I~6- Feb. l~7 12-13 and 19-23 Jan. 1987 14-16 and 24-25 Jan. 1987 Feb. 1987 23 Feb.- 6March 1987 7-18 March 1987 19 March-28 June 1987 July 1987-Jan. 1988 Jan-Feb. 1988 Feb.-April 1988 May-June 1988 July 1988 a) Sample Design The TDHS is based on a national sample designed to provide independent estimates for the four major regions of the country plus the Bangkok Metropolitan Area as well as for the urban and rural populations. To achieve this, the population was divided into six separate sampling domains: the Bangkok Metropolitan Area, all provincial urban areas, and the rural areas in each of the four regions. Provincial urban areas are defined as all administratively defined municipal areas outside of Bangkok. The total urban category consists of Bangkok plus provincial urban areas. The sample design and weighting procedures are described in detail in Appendix A. A brief description of it is as follows. In Bangkok, households were selected in two stage. First a systematic sample of 48 blocks was selected with probability proportional to population size (PPS). Thereafter, households within selected blocks were listed just before the survey and selected so as to obtain a sample with a reasonably uniform overall selection rate for households. All ever-married women aged 15- 49 who were in a sample household the night before the interviewer's visit were eligible for the detailed interview ( de facto coverage). In other domains, the sample was selected in three stages: selection of 24 districts per domain with PPS; selection of 2 villages/blocks per district; and finally, listing and systematic selection of households within villages/blocks. Again, the objective was to obtain a sample with reasonably uniform selection probabilities for households within each domain. The selection procedure described yields the total number of 288 ultimate area units in the sample. The sample districts are shown in Figure 1.3. Of these 288 selected sample units, 9,423 households were identified as the target. The target number of households and eligible women by reporting domain are shown in Table i.i (see below under response rates). All estimates from the survey have been computed after appropriately weighting the sample cases reflecting the sampling design used. b) Questionnaire Translation and Modification The DHS core questionnaires (Household, Eligible Women Respondent, and Community) were translated into Thai. A number of modifications were made largely to adapt them for use with an ever- married woman sample and to add a number of questions in areas that are of special interest to the Thai investigators but which were not covered in the standard core. Examples of such modifications included adding marital status and educational attainment to the household schedule, elaboration on questions in the individual questionnaire on educational attainment to take account of changes in the educational system during recent years, elaboration on questions on postnuptial r~sidence, and adaptation of the questionnaire to take into account that only ever-married women are being interviewed rather than all women. More generally, attention was given to the wording of questions in Thai to ensure that the intent of the original English-language version was preserved. The three questionnaires employed in the TDHS (household, individual and community) are reproduced in Appendix D. I0 Figure 1.3 TDHS Sample Districts L~uend North Northers t Centra 1 ] South • Munici~a~ a~eas 11 i. Household questionnaire The household questionnaire was used to list every member of the household who usually lives in the household and as well as visitors who slept in the household the night before the interviewer's visit. Information contained in the household questionnaire are age, sex, marital status, and education for each member (the last two items were asked only to members aged 13 and over). The head of the household or the spouse of the head of the household was the preferred respondent for the household questionnaire. However, if neither was available for interview, any adult member of the household was accepted as the respondent. Information from the household questionnaire was used to identify eligible women for the individual interview. To be eligible, a respondent had to be an ever-married woman aged 15-49 years old who had slept in the household 'the previous night'. Prior evidence has indicated that when asked about current age, Thais are as likely to report age at next birthday as age at last birthday (the usual demographic definition of age). Since the birth date of each household number was not asked in the household questionnaire, it was not possible to calculate age at last birthday from the birthdate. Therefore a special procedure was followed to ensure that eligible women just under the higher boundary for eligible ages (i.e. 49 years old) were not mistakenly excluded from the eligible woman sample because of an overstated age. Ever-married women whose reported age was between 50-52 years old and who slept in the household the night before the visit were also identified in the household questionnaire as potential candidates for the eligible woman sample and interviews were initiated with them. If in the course of the individual interview, which asked about the birthdate of the woman, it was discovered that these women (or any others being interviewed) were not actually within the eligible age range of 15-49, the interview was terminated and the case disqualified. This attempt recovered 69 eligible women who otherwise would have been missed because their reported age was over 50 years old or over. 2. Individual questionnaire The questionnaire administered to eligible women was based on the DHS Model A Questionnaire for high contraceptive prevalence countries. The individual questionnaire has 8 sections: I. Respondent's background 2. Reproduction 3. Contraception 4. Health and breastfeeding 5. Marriage 6. Fertility preference 7. Husband's background and woman's work 8. Heights and weights of children and mothers The questionnaire was modified to suit the Thai context. As noted above, several questions were added to the standard DHS core questionnaire not only to meet the interest of IPS researchers hut also because of their relevance to the current demographic situation in Thailand. The supplemental questions are marked with an asterisk in the individual questionnaire (see Appendix D). 12 Questions concerning the following items were added in the individual questionnaire: Did the respondent ever experience a miscarriage or abortion? If so, how many? Educational attainment and expectations for each of respondent's living children age 6 or above. Did the respondent ever use contraception subsequent to marriage and prior to first pregnancy? If so, how long after marriage did she first use contraception? Information on whether or not users of oral contraceptives forgot to take the pill any time during the last month and if so, how many times. Information on the type and timing of first contraceptive method used since last birth including a probe on whether contraceptive use was initiated prior to or subsequent to the return of menses The place of the respondent's last delivery. Whether the respondent's marriage was registered; whether the marriage was marked by a ceremony. Did the couple live with any set of parents following marriage? If so, with whose parents did the couple reside following marriage? Does the respondent consider a lower high school education sufficient for young people nowadays? Secondary occupation of husband. Information on respondent's current work, employment status and type of payment. Height and weight of mothers of children 3-36 months of age. 3. Community questionnaire TDHS community questionnaire was based on the model DHS community questionnaire. Again it was modified to suit the situation in Thailand. The community survey was conducted in all 192 sample clusters (villages) of rural areas but not in urban areas. The community questionnaire focuses on information on village characteristics, accessibility to health and family planning services, and availability to public services nearest to the cluster. The community was defined according to official administrative boundaries. A group interview was used as the mode of data collection for the community survey. The interview was conducted by the team supervisor. The respondents were a group of community leaders (typically 3-5 persons). Persons qualifying as respondents included current or former village headmen, or their 13 ass i s tants , v i l l age hea l th vo lunteers , v i l l age hea l th communicators, members of ex i s t ing assoc ia t ions (groups) in the v i l l age , and other v i l l age leaders who have been res id ing in the community for f i ve years or more. V i s i t s were a lso made to a l l government hea l th and fami ly planning serv ice out le ts w i th in a 30 k i lometer rad ius from the c lus ter to co l lec t informat ion from the personnel about serv ices . c) Superv isors ' Tra in ing Most team supervisors of TDHS fieldwork were IPS research associates with extensive fieldwork experience. Training of supervisors and assistants was conducted by the field director and project technical staff. The training of supervisor and assistants was divided into 2 phases. The first phase started with a two day briefing which focused on the content of the household and individual questionnaire. Since it was essential for the supervisors and assistants to understand the questionnaires thoroughly, given their role as field editors, after the initial briefing sessions, the supervisors conducted interviews in the field as part of the questionnaire pretest. This was then followed one week later by a special one-day seminar to "discuss lessons from the first pretest and by an additional day of practice interviews in a slum area of Bangkok. The second phase of the t ra in ing took place from February 23 to March 6, 1987 and included f ive days on anthropometr ic measurement. The anthropometr ic t ra in ing was conducted by a spec ia l i s t provided by DHS headquarters . The second phase a lso included a week of add i t iona l t ra in ing concerning the household and ind iv idua l quest ionna i res . At the same time the superv isors were a lso t ra ined to admin is ter the community quest ionna i re . Further t ra in ing of superv isors and ass i s tants concerned f ie ldwork procedures such as the updat ing of l i s t s of households, se lec t ion of sample households, and v i s i t s to hea l th and fami ly p lanning serv ice out le ts . d) P re tes t The dra f t quest ionna i res were pretes ted in both rura l and urban areas of Kanchanaburi province, about 100 k i lometers from Bangkok, and in a slum area in Bangkok. The pretes t was car r ied out by f ive superv isors and the i r ass i s tants . Resul ts from the pretes ts were used as bas is for rev is ing the quest ionna i res . As part of the quest ionna i re pretes t , a separate short quest ionna i re was administered which was designed to i l l uminate the nature of age and b i r th date repor t ing by mothers for young ch i ld ren . Based on the resu l t s , i t was decided to ins t ruct in terv iewers to request to see documentation of b i r th dates of a l l l i ve born ch i ld ren , e i ther in the form of b i r th reg is t ra t ion cer t i f i ca tes or household reg is t ra t ion forms. The pretes t ind icated that substant ia l numbers of mothers would be able to do th i s and that i t would e l iminate most of the ambigu i t ies assoc ia ted with age and date repor t ing that otherwise ar i se . 14 e) Pretest Results Based on the pretest, it was found that there were difficulties with questions 304 and 305. These questions deal with knowledge of sources and potential problems of methods known to the respondent in the core questionnaire. Women who were currently using a contraceptive method (the majority of eligible respondents in Thailand) had particular difficulty answering the questions. These two questions took a long time to ask given that most respondents knew all modern methods and therefore had to be asked about each one. Some respondents showed impatience with being repeatedly asked a question that made little sense to her. It was also obvious from the pretest that question 227 on knowledge of the period of risk of conception during the menstrual cycle was problematic. Nevertheless, on advice from DHS headquarters, these questions were retained. The pretest also made it evident that the weight and height measurement component demanded both great effort and well organized implementation. Pretest results generally indicated that supervisors and assistants would have to make considerable effort and be very efficient in order to complete all the tasks assigned to them. f) Interviewer Recruitment Announcements of positions for interviewers for TDHS were made and over I00 applicants from the student body of Chulalongkorn University were screened. Ability to speak local dialects and fieldwork experience were the two main criteria for selecting the interviewers. A total of 35 interviewers were hired. g) Interviewers' Training The training of interviewers took place during March 7-18. The training consisted of a detailed, item by item explanation of the household and individual questionnaires, role playing, mock interviews, field interview practice and a seminar to discuss experiences and problems. The field interview practice was done in both rural and urban areas. Five villages in Pathum Than/ Province and non-sample blocks of Bangkok were selected for field interview practice. The training went well. Most interviewers showed enthusiasm and competence in their work. h) Fieldwork and Supervision A total of 5 teams were formed for data collection, each consisting of one supervisor, one or two assistant supervisors, seven female interviewers and one driver. The names of the field staff are shown in Appendix E. The teams were formed according to regions, namely north, northeast, central, south, and the Bangkok Metropolis. Interviewers in each regional team were able to speak the major regional dialect. In urban areas, sample blocks were updated by the supervisor and assistants before selecting the sample households using maps provided by the 15 National Statistical Office (NSO). In rural areas, household lists of the sample villages were obtained at the district office. The lists were later updated through consultation with the village headman. In the updating process, supervisors were instructed to probe for structures without a registered number and vacant households. For both urban and rural areas fixed number of sample households for each cluster was systematically selected. The fieldwork was largely carried out between March and June 1987. The data collection was divided into two main phases. The first phase was from March 17 - April i0 and the second phase from April 17 - June 6, 1987. All teams returned to Bangkok after the completion of the first phase of fieldwork. A two-day seminar was held to discuss problems that arose during the fieldwork and solutions were advised. Extension of data collection to the end of June was required for some sampling clusters in the central region and Bangkok Metropolis. At the end of the originally scheduled second phase of the fieldwork a concluding seminar was held to give feedback to the investigators and the IPS technical staff both for improving future surveys and for interpreting results of the TDHS. The interviews usually took place between 7 am. and 7 pm. The average duration of interviews for household and individual questionnaires was 4.5 minutes and 30.9 minutes respectively. All supervisors and assistants were instructed to closely observe and supervise the interviewers particularly during the first few days of the fieldwork. This procedure was enforced strictly so that any misunderstanding in the questionnaires and errors made could be detected and corrected at an early stage. The field director also visited the teams to help with any problems each team had as well as to deliver any supplies each team needed and bring back completed questionnaires. Completed questionnaires were submitted to the supervisor or assistant immediately following interview. The questionnaires were edited in the field to the extent feasible. If possible, inconsistencies and errors were clarified and corrected and re-interviews on the questions for which answers were omitted or inconsistent were made. The interviewers were instructed to make their best attempt to visit and interview the sample households. Usually three call-backs were made for households with no adult or with no one at all at home. To ensure high response rates, sometimes more than three call-backs were made. The task load of supervisors and assistants was very heavy in the fieldwork. They were responsible not only for the overall management of the team, which included making all contacts, assigning the households to the interviewers, editing the questionnaires, and planning daily work, but they were also assigned to do the anthropometric measurements and the community survey including the visits to the health and family planning service outlets. In retrospect, this workload was excessive. To improve fieldwork quality, it would have been advisable to have had a separate team carry out the time consuming community survey component. One result of the this excessive workload was that it became impossible for the supervisor and assistants to fully edit all the completed questionnaires in a timely manner in the field. 16 i) Response Rates Table i.i shows the number of households and women selected and successfully interviewed by region. Although equal sample sizes for each domain were originally intended, due to population growth, particularly in urban areas, the number of households selected varied slightly by region. The total number of target households is highest in the central region followed by Bangkok, the north, south and northeast. In general the response rates of both household and individual interviews in the TDH$ were relatively high. For the country as a whole, 96 percent of the selected households were successfully interviewed. The main reason for non-response in the household survey is that either no one at all or no adult was at home. The household response rates vary by region being highest in the northeast (99 percent) and lowest in Bangkok (92 percent). However the total number of households interviewed was greatest in the central region and lowest in the south. The overall TDHS response rate is 90 percent. As expected Bangkok yielded the lowest success rate while the north and northeast had the highest success rate. The response rate for the eligible woman sample is lower than the household response rate. About 94 percent of eligible women identified were successfully interviewed. The main reasons for non-response in the eligible women survey were that the targeted respondent was not at home and/or refused to be interviewed. Regional differences in the response rates of the individual interviews were similar to the household interviews. The highest response rate for eligible women was in the north (98 percent) and the lowest in Bangkok (87 percent). The generally high response rates for both household and women interviews were due mainly to the strict enforcement of the rule to revisit the originally selected household if no one was at home initially. No substitution of the originally selected households was allowed. Interviewers were instructed to make at least 3 call-backs if contact with the household or eligible woman had not been made or the interview was incomplete. In many instances revisits were made until the team had moved out of the province. The survey indicates a low ratio of the number of eligible women per household. On the average there are about 80 eligible women per I00 households interviewed. This is much lower than found in SOFT, conducted in 1975, where the ratio was 96 per 100 households. At least in part this could be attributable to the increasing age at marriage (see Chapter 2). There is some regional variation in terms of number of eligible women per household. The ratio is highest in the northeast (83 per i00) and lowest in the south (75 per lO0). This lower ratio of number of eligible women per i00 households explains why the total number of eligible women interviewed was lower than the number targeted (6,775 versus 7,000). 17 Table 1.1 Ntmber of house_holds and women selected and successfully interviewed, by reporting domain Noaseholds Eligible Wc~en Reporting Successfully Reslxmse Successfully Reslxnse domain Selected interviewed rate (%) Selected interviewed rate (%) (i) (2) (3)=(2)/(1) (4) (5) (6)=(5)/(4) Overall EligiBle resIx:mse wr'm~n rate(%) per 100 hh. (7)=(3)x(6) (8)=(4)/(2)xi00 GO Bangkok i, 913 1,762 92.1 I, 441 i, 248 86.6 79.8 81.8 North 1,889 1,857 98.3 1,476 1,448 98.1 96.4 79.5 Northeast 1,730 1,708 98.7 1,419 1,384 97.5 96.2 83.1 Central 2,125 2,014 94.7 i, 585 1,469 92.7 87.8 78.7 South i, 766 1,704 96.5 i, 280 I, 226 95.8 92.4 75.1 Total 9,423 9,045 96.0 7,201 6,775 94.1 90.2 79.6 j) Office Editing and Data Entry All completed questionnaires have been sent to IPS for office editing. It was originally planned that the team supervisors and some assistants would be retained as office editors and keyers. Unfortunately, most of the temporary team supervisors and assistants left the project at the end of the fieldwork. Therefore, five new editors and keyers had to be hired. These new editors and keyers are graduates from various universities in Thailand with a bachelor degree in social science or a related field. They received intensive training on the content and logic of the questionnaire. To further improve their ability to edit the questionnaires, they conducted interviews with households of the sample clusters that required revisits in Bangkok and the central region. Office editing of questionnaires was supervised by the field director and two IPS research associates who had also been TDHS team supervisors. The editing was done by the five new editors/keyers, two project assistants, and two IPS permanent research assistants who had also served as team assistants. All questionnaires were given numbers and sorted by sample cluster number. The questionnaires were checked for completeness, internal consistencies and appropriate codes, particularly of the open-ended questions. The data entry of TDHS started in early July, 1988. The data were directly transferred from the questionnaires to micro-computers, using the ISSA program developed by DHS. Two programmers from DHS were sent to IPS to help set up the ISSA program and train IPS data processing staff on how to work with the program. Office editing and data entry were completed by the first week of January 1988. The tabulations for the preliminary and country report were then prepared with the assistance of the DHS programmer. 1.7 Background Character i s t i cs of the Surveyed Women The Thai Demographic and Health Survey interviewed 6,775 eligible women (aged 15-49). The description of the characteristics of the surveyed women provides a background for interpretation of survey findings presented in Chapters 2 to 6. A discussion of the associations among some of these background variables is useful for the understanding of the data. The background characteristics of the ever-married women respondents in the TDHS survey are shown in Tables 1.2 and 1.3. Table 1.2 presents the percent distribution of ever-married women according to selected background characteristics along with the actual and weighted number of eligible women interviewed. The weighting is necessary to compensate for slight differences in the selection probabilities and response rates and to make the regional and rural-urban distribution of the sample correspond to that expected from official sources. The weights are determined in such a way that the total weighted cases equal the total actual cases. Therefore for most of the sample, the weighted number of cases can serve as a rough guide to the actual numbers. The main exceptions arise when the results are tabulated by the criteria used to define the sampling domains, namely region or urban-rural residence, or any characteristics strongly associated with region or urban-rural residence. All results presented in this report are weighted and only weighted number of cases are shown (to allow readers to properly combine categories if so desired). 19 Table 1.2 Percent distribution of ever-married women according to selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weighted Unweighted Background Percentage number number characteristic (weighted) of women of women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Age 15-19 5.0 342 308 20-24 14.8 1,004 1,017 25-29 19.3 1,309 1,320 30-34 19.6 1,328 1,341 35-39 16.4 1,110 1,137 40-44 12.9 877 871 45-49 11.9 805 781 Urban-rural residence Urban 18.2 1,233 2,423 Rural 81.8 5,542 4,352 Region North 20.6 1,396 1,448 Northeast 34.9 2,365 1,384 Central 21.4 1,450 1,469 South 12.3 833 1,226 Bangkok 10.8 732 1,248 Religion Buddhist 92.6 6,275 6,199 Islam 5.3 359 474 Other 2.0 137 97 Not stated 0.i 4 5 Living children 0 10.4 707 771 1 21.6 1,463 1,503 2 26.1 1,768 1,795 3 16.8 1,138 1,149 4 + 25.1 1,698 1,557 Total i00 6,775 6,775 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 The selected background characteristics discussed include age, regional and rural-urban residence, religion and number of living children of the sample women. The age of interviewed eligible women in this study is derived from reported birthdates. For those whose year of birth is not known, age is obtained directly from the stated age. However, most women interviewed in this survey were able to give their birth year and/or birth month. Among all interviewed eligible women, 89 percent could report both a month and year of birth, i0 percent reported year but not month of birth, and only one percent could not report year of birth. This high proportion of respondents knowing their birth year stems from the importance of knowing one's animal year of birth within the Thai cultural context. Thus ages in the TDHS can be calculated relatively accurately. The data show that almost two-fifths of the sample women are in the age-groups 25-29 and 30-34. The low percent of women interviewed in the age-group 15-19 is a result of the fact that the sample covers only ever-married women and a minority of women are married before age 20 in Thailand. As discussed earlier the weighted distribution of sampled women by regional and urban-rural residence conforms to an expected standard distribution (the 1987 projected distribution) used in the calculation of weights. About 82 percent of ever-married women reside in rural areas and 18 percent in urban areas. Of the total sample women, 35 percent are in the northeast, 21 percent each in the north and the central region, 12 percent in the south, and ii percent in Bangkok. The majority of the sampled women (93 percent) are Buddhists. Only 5 percent are Moslems. This closely reflects the national distribution. The other religious category includes mostly Christians but also anamists, those with no religion, and any others. They constitute about 2 percent of the sampled women. Only 0.i percent of the sampled women (or 4 unweighted cases) did not report their religion. Almost half of the sampled women have one or two living children. About i0 percent have no living children and 42 percent have more than three living children. The association between each of the background characteristics and educational attainment is shown in Table 1.3 for the eligible woman sample. As described in section l.lc, the government implemented compulsory education only about 6 decades ago. Before that period, education was largely in the form of Buddhist schooling and restricted to males. 21 It is important to mention that it is not possible to classify a person in terms of educational level in this report by a uniform conversion from number of years of schooling since several changes in the educational system have occurred over the recent past.* Women of reproductive ages can fall into any one of three different systems of education. Each system divided the number of years of schooling constituting the basic levels slightly differently. Moreover some women might fall into two different systems due to the transition. In general, women aged 40 years and over are likely to be under the first system, women between 20 and 40 years of age are mostly under the second system, and women under 20 tend to fall under the present system. In this study, education of women is classified according to the system to which their cohort belonged. Although education of women in Thailand has been increasing, only 12 percent have a secondary or higher education. The majority (79%) of ever- married women in the reproductive ages still have only primary education. The remaining 10 percent have no formal education, although some of them may be able to read and/or write. The percent of women according to education by different age cohorts reflect an increase of education among Thai women. The percent of women with secondary or higher education declines with increasing age except for women in the younger age-groups for which censoring affects the results, i.e. not all of these women have reached the age necessary to complete secondary or higher than secondary educational levels. As expected, urban women are better educated than their rural counterparts. Regional variation in educational level still remains. Although the majority of women in all regions have only a primary education, the proportion is the highest for the northeast. Women in Bangkok stand out in terms the proportion with higher education. *Since the Second World War there have been three systems of formal education before entering college or university in Thailand. The systems differ in terms of the number of primary and secondary school grades involved. These can be represented in terms of three digits in which the first digit refers to the number of years required to complete primary level and the second and the third digits are number of years required to complete lower and upper secondary levels respectively. The first system, 4:6:2, was in effect until 1959. The second system, 7:3:2, was implemented during 1960-1977. The present system, 6:3:3 has been used since 1978. It is important to note that during the second system, the seven years of primary school were divided into lower and upper levels and that it was common to leave school after completing the first 4 years which constituted lower primary school. Graduates of upper secondary school (or their equivalents) usually spend another four years to complete the bachelor's degree from a university or college. Those who have a lower secondary certificate have in the past been qualified to go to vocational colleges such as technical, teachers, and nursing schools, or to police and military academies. However, over time the requirements for entering some of these institutions have been raised. In addition some vocational colleges have been upgraded to university status. 22 Table 1.3 Percent distribution of ever-married wom~ according to education, by selected background characteristics Higher Weighted Background No than ntmber characteristic education Primary Secondary secondary Total of women Aue 15-19 7.5 83.3 9.1 O. 1 100 342 20-24 5.9 79.8 ii. 2 3.1 100 i, 004 25-29 6.9 76.0 ii. 6 5.5 100 i, 309 30-34 7.1 81.3 5.9 5.8 100 1,328 35-39 9.4 80.2 6.1 4.3 100 I,ii0 40-44 13.4 76.5 5.9 4.1 100 877 45-49 20.8 73.8 3.4 2.0 100 805 Urban-rural residence Urban 6.0 58.2 23.3 12.5 100 1,233 Rural i0.5 83.0 4.2 2.3 100 5,542 Region North 18.2 73.7 5.3 2.9 100 i, 396 Northeast 4.8 89.4 3.6 2.2 100 2,365 Central 9.5 77.9 8.4 4.2 100 I, 450 South 12.2 73.0 9.4 5.4 i00 833 Bangkok 6.8 59.9 22.0 11.3 100 732 Religion* Buddhist 8.2 80.1 7.7 4.1 100 6,275 Islam 19.5 69.6 7.8 3.1 100 359 Other 54.0 30.0 6.5 9.5 100 137 Total 9.7 78.5 7.7 4.2 100 6,775 *Excludes a small ntmlber of cases for whom religion is not stated The data also show differences in education by religion. Moslem women have substantially less education than Buddhist women. The percent of women with no education among Moslems is more than double that of Buddhists (20% versus 8%). Buddhist women are more likely to have completed each of the other three educational levels than Moslem women. Educational composition of women in the other religious category reflects the mixed nature of this group. Compared to Buddhist and Moslem women, these women have both a higher percent with no education and a higher percent with more than secondary education. In the remainder of this report, when results are presented according to religion, only Buddhists and Moslems are shown because of the small number and heterogeneous nature of the remainder of the sample. 23 Chapter 2 Nupt ia l i ty and Other Proximate Determinants This chapter is concerned with nuptiality and other key proximate determinants of fertility. While nuptiality is a phenomenon of considerable social interest in itself, its demographic significance derives from the fact that marriage is a primary indication of the exposure of women to the risk of pregnancy and, therefore, is critical for the understanding of fertility. This is particularly true in a country like Thailand where childbearing is largely confined to marital unions. This chapter therefore begins with a consideration of recent nuptiality patterns and trends. Also considered in this chapter are measures of several other proximate determinants of fertility which influence exposure to risk of pregnancy: breastfeeding, postpartum amenorrhea, and postpartum abstinence. 2.1 Nupt ia l i ty Patterns and Trends Data on the marital status of all household members (assuming those under age 13 are all single) were collected through the household questionnaire. The eligible woman questionnaire, from which most of the data presented in this report are based, was administered only to ever-married women aged 15-49. It is useful, however, to include never-married women in the denominator for certain measures presented so that these measures refer to all women even though the information on which the numerators are based come from the eligible woman questionnaire. The number of never-married women listed in the household questionnaire can not be directly added to the number of eligible women respondents to form the denominator of total women for two reasons. First, not all ever-married women in interviewed households were actually interviewed themselves as indicated in the discussion of the response rate in the previous chapter. Thus simply to add all never-married women listed in interviewed households would disproportionately represent those who were never-married. Second, ages as coded in the household and eligible woman files are not strictly comparable. In the household questionnaire, age is available only from direct statements of age and is provided for all household members by whomever was the respondent for the household. Ever-married women interviewed for the eligible woman sample, however, were asked not only directly their own age but were also asked their hirthdate. Whenever possible, ages of eligible women for the purpose of analyses based on the eligible women file are determined from the birthdate. Since in practice stated age in Thailand often to refers to the age at next birthday rather than to the age at last birthday, recorded ages of a substantial proportion of women in the household listing are a year older than their true age at last birthday while ages of women in the eligible women file are generally correct. Despite these problems, it is possible to derive an appropriate multiplication factor based on the household schedule to apply to interviewed ever-married women in order to expand the denominator so that it represents all women. Based on weighted data from the household questionnaire, the ratio of all women (i.e. including never-married) to ever-married women at each single year of age as reported in the household questionnaire has been calculated. If results are to be presented for separate categories of the population (e.g. by region or educational level), the ratio of all women to ever-married women at each single year of age is calculated separately for each reporting category. The denominators for the measures are expanded by multiplying through by these ratios. Thus each ever-married woman respondent, at each single year of age as reported in the household questionnaire, is multiplied by the ratio of all women to ever-married women at that age listed in the sample households in the same reporting category. Results are then reported by corrected age. The numerators of these measures remain as reported by the eligible respondents. Table 2.1 presents the percent distribution of ever-married women and all women according to their current marital status. No distinction is made between couples who legally registered their marriage and those who did not since this is not a socially meaningful distinction in Thailand. In the case of all women, the number of never-married is determined in the manner referred to above and thus the results are not strictly comparable to those based directly on the household sample and presented in Appendix A. As can be seen for the ever-married woman sample, the large majority at all ages are currently married although the percent declines systematically with age. Among ever-married women who are not currently married, divorce and separation account for the majority at the younger reproductive ages while widowhood accounts for the majority at the older ages. For all groups, separation is more common than divorce, in part reflecting the substantial proportion of marriages that were not legally registered in the first place (and thus did not require divorce to terminate). When the marital status distribution is expanded to refer to all women, the proportion who never married is seen to decline "rapidly with age. By the end of the reproductive ages, very few Thai women have never married as indicated by the fact only 4 percent of women aged 45-49 are in this category. Nevertheless, substantial proportions of women in the young reproductive ages remain unmarried: almost half of women aged 20-24 and almost one fourth 6f those aged 25-29 are still single. Cohort trends in age at marriage can be described by comparing the distribution for successive age groups, although the data for the oldest cohorts should be interpreted cautiously. Older women may not recall marriage dates or ages with accuracy particularly when unions are not registered. Indeed, many respondents including younger ones, did not recall with precision their date of marriage and frequently the date of marriage had to he determined indirectly by deducing it from the date of first birth. These caveats notwithstanding, the proportion married at successive ages can he derived by cumulating across age of marriage categories. Based on this information, the median age of marriage, defined here as the exact age by which 50 percent of an entire cohort has experienced marriage, can be calculated. The median is preferred over the mean as a measure of central tendency, because, unlike the mean, it can he estimated for any cohort for which at least half of the women are ever-married at the time of the survey. 25 Table 2.1 Percent distribution of ever-married and all women according to current marital status, by current age Weighted Current Never Currently No Total number of age married married Widowed Divorced Separated answer percent women Ever tarried ,¢mm~ 15-19 - 97.5 0.1 0.0 i. 8 0.6 100 342 20-24 - 95.4 0.5 0.4 3.5 0.2 I00 1,004 25-29 - 95.0 0,8 1.3 2.7 0.i 100 1,309 30-34 - 94.1 1.3 1.2 3.4 0.0 100 1,328 35-39 - 91.8 3.6 1.4 3.0 0.2 100 I,ii0 40-44 - 86.4 7.1 i. 6 4.7 0.2 100 877 45-49 - 83.9 9,3 i. 2 5.4 0.2 100 805 All a~m - 92.0 3,1 1.1 3.6 0.2 100 6,775 15-19 83.2 16.4 0.0 0.0 0.3 0.1 100 - 20-24 47.8 49.8 0.3 0.2 1.8 0.I 100 - 25-29 23.8 72.4 0.6 1.0 2.1 0.I 100 - 30-34 13.3 81.6 I. 1 1.0 2.9 0.0 100 - 35-39 9.1 83.4 3.3 1.2 2.7 0.2 100 - 40-44 6.4 80.9 6.6 1.5 4.4 0.2 100 45-49 3.7 80.9 8.9 i.i 5.2 0.2 I00 Total 33.6 61.1 2.1 0.8 2.4 0.1 100 *Derived by applying a multiplication factor based on the household questionnaire to the eligible women sample and thus differs from age and marital status distribution based only on the household questionnaire as presented in Appendix A. The weighted number of women is not presented for the tabulation referring to all women because it is influenced by this multiplication factor. See text for explanation. The percent distr ibut ion of women by age at first marr iage (including the category "never married") and the median age at first marr iage are presented in Table 2.2 for di f ferent age cohorts. No median age is provided for women aged 15-19 since less than 50 percent have marr ied or for women age 20-24 since the median falls within this age group and thus would be inf luenced by censoring. The results reveal a steady decl ine in the median age at marr iage for each successive age cohort from 25-29 to 45-49 indicat ing a trend towards an increasing age at marr iage during the past several decades. Such a trend is 26 Table 2.2 Percent distribution of all women according to age at first marriage (including those reported in household as never married) and median age at first marriage, by current age Current Never Total age married* <15 15-17 18-19 20-21 22-24 25-27 28-29 30+ percent Median** 15-19 83.2 i. 9 i0.6 4.3 - - 100 20-24 47.8 2.2 18.2 16.7 I0.5 4.6 - - 100 25-29 23.8 2.1 20.5 17.8 17.1 12.7 5.4 0.6 0.0 100 21.1 30-34 13.3 2.3 22.0 19.3 15.9 15.3 8.3 2.1 1.6 100 20.7 35-39 9.1 2.9 22.1 20.5 15.7 14.1 9.1 2.8 3.6 100 20.5 40-44 6.4 3.1 21.4 22.5 17.5 14.1 7.0 2.7 5.2 100 20.3 45-49 3.9 4.3 24.5 25.0 17.6 12.5 7.8 1.5 3.0 100 19.7 *Derived by applying a multiplication factor based on the household questionnaire to the eligible women sample. See text for explanation. **Median in this table is defined as the exact age by which 50 percent of an entire cohort has experienced marriage. consistent with previous analyses of trends in the age of marriage based on censuses and other surveys (Knodel, et al., 1984). Table 2.3 presents the median age at first marriage for age cohorts from ages 25-29 to 45-49 according to urban-rural residence, region and educational level. Age at marriage is distinctly older for urban women compared to rural women. Regional differences are modest except for the distinctly older median age at marriage for Bangkok women. Age at marriage is positively associated with educational level, being ,ine years older on average for'women with higher than a secondary school education compared to women with no education. For most categories of the population shown in Table 2.3, age at marriage has been increasing as indicated by the inverse association between current age and median age at marriage. The major exceptions are women with secondary or with higher than secondary education, for whom age of marriage is relatively late but for whom little trend across age cohorts is evident. 27 Table 2.3 Median age at first marriage among all women aged 25-49 years (including those reported in the household as never married), by current age and selected background characteristics Current age Background . characteristic 25-29 30-34 35-39 40-44 45-49 Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Urban-rural residence Urban 24,5 23.5 23.9 22.9 21.9 23.6 Rural 20,5 20.3 19.9 19.9 19.3 20.0 Reqion North 20.3 19.7 19.1 19.3 18.9 19.6 Northeast 20.4 20.0 20.3 20.1 19.4 ~I Central 21.6 21.7 21.1 20.6 19.8 210 South 21.1 20.3 19.0 19.8 20.2 ~.I Bangkok 25.3 24.3 24.9 23.5 21.9 ~,2 Education No education 20.1 19.1 18.4 18.9 18.3 187 Primary 20.1 20.2 20.3 20.3 19.7 ~I Secondary 23.4 24.1 24.0 23.4 23.6 ~6 Higher - 26.6 27.6 26.4 27.4 ~.9 Total 21.1 20.7 20.5 20.3 19.7 20.5 Note: See definition of median in Table 2.2 2.2 Breastfeeding and Postpartum Insuscept ib i l i ty Postpartum protection from conception can be prolonged by breastfeeding which can lengthen the duration of amenorrhea and/or by the delayed resumption of sexual relations. The percentage of women still breastfeeding, and still postpartum amenorrhea, abstaining, and insusceptible are presented in Table 2.4 and serve as the basis for estimates of the median length of breastfeeding and amenorrhea as well as estimates of the length of postpartum abstinence which are shown at the bottom of the table. The joint impact of amenorrhea and abstinence is the length of postpartum insusceptibility, defined as the elapsed time between birth and resumption of both menstruation and sexual intercourse, or the later of the two events. This definition assumes that the period of postpartum amenorrhea coincides with the duration of anovulation following childbirth. While this is not strictly true, the two are probably quite closely related. 28 The tabulation presented in Table 2.4 is birth-based rather than woman-based, i.e., any woman who within the 3 years preceding the survey had more than two live births (counting twins as a single birth for the purpose of this tabulation) will be included in the table as many times as she had births. The distributions of the proportion of births by the month of birth of the child are analogous to the Ix column of a synthetic life table. Note, however, that only the mother's current status is considered and retrospective information about how long a particular status lasted, if that status has been terminated, is ignored. In any real cohort, the proportions in any particular status (such as breastfeeding or amenorrheic) could only decline with time since birth. However, since the results in Table 2.5 are crossectional rather than representing the experience of any actual cohort and because of fluctuations associated with small numbers of cases, it is possible for irregularities to appear in the association between the percent in a given status and the time since birth. Table 2.4 Percentage of births in the last 3 years whose mothers are still breastfeeding, still postpartum amenorrheic, still abstaining, and insusceptible, by months since birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Months Weighted since number of birth Breastfeeding Amenorrheic Abstaining Insusceptible* births . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Less than 2 89.6 97.6 86.4 99.0 84 2-3 88.0 66.9 41.0 76.9 113 4-5 83.5 56.4 11.2 59.6 107 6-7 75.2 38.3 5.0 41.6 128 8-9 75.5 46.5 1.3 47.8 105 I0-ii 68.3 22.1 10.2 29.2 124 12-13 65.0 13.4 5.9 18.5 131 14-15 50.2 14.8 3.3 18.1 141 16-17 38.7 5.4 1.2 6.6 139 18-19 38.1 7.8 3.3 ii.i 126 20-21 39.5 6.4 0.0 6.4 140 22-23 22.8 3.5 5.0 8.2 112 24-25 30.7 2.7 3.6 6.4 151 26-27 21.3 4.6 4.5 9.2 130 28-29 15.2 1.5 0.0 1.5 129 30-31 13.7 1.7 2.3 4.0 100 32-33 10.3 2.5 0.5 3.1 123 34-35 5.4 0.6 2.5 3.1 84 Total 45.6 19.8 8.9 23.5 2,168 Median** 14.5 5.3 2.1 5.6 - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Note: Women who are pregnant are not counted as amenorrheic regardless of whether or not menses returned since their most recent birth * Either amenorrheic or abstaining ** Calculated from 3 month moving averages based on percentages by single months tabulated 29 For the purpose of providing some stability to the percentages, the birth data are grouped in two month intervals. Even so, some reversals are apparent. For example, the percentage of children still being breastfed among those born 22-23 months prior to the survey is less than the percent still being breastfed among those born 24-25 months prior to the survey. Nevertheless, the percentages still in the various statuses shown generally decline with each successive duration since birth. In order to calculate medians, three month moving averages were computed based on a comparable set of tabulations by single months since birth (see appendix Table 2A.I). For each of the statuses shown in Table 2.4, it was possible to identify a unique median, i.e. a number of exact months by which 50 percent of mothers had terminated the indicated status. The results show a median duration of breastfeeding of 14.5 months, a median duration of postpartum amenorrhea of 5.3 months, and a median duration of abstinence following childbirth of 2.1 months. The median duration of insusceptibility, 5.6 months, is only slightly longer than the median duration of amenorrhea because few couples abstain longer than the amenorrheic period. The TDHS is the first survey to provide systematic evidence on postpartum abstinence. The short median duration of abstention is quite consistent, however, with previous qualitative assessments (Knodel, Havanon, and Pramualrathana, 1984). The large majority of Thai mothers breastfeed their children as evident from the high proportion of children still being hreastfed among those born in the months just prior to the survey. For example, 90 percent of children born less than two months prior to the survey and 88 percent of those born 2 or 3 months prior to the survey were still being breastfed. Considerable proportions are also breastfed for substantial durations as indicated by the fact that almost two-thirds of children born about a year earlier were still being breastfed at the time of the survey. The average duration of postpartum amenorrhea, during which most women are anovulatory and hence not at risk of becoming pregnant, depends largely on the duration and nature of breastfeeding, although a mother's nutritional level and physiological condition may also have some influence. The considerably shorter median duration of postpartum amenorrhea among Thai women in comparison with the duration of breastfeeding may reflect the common practice in Thailand of introducing supplementary food into the diet of breastfed children at a very early age. This could reduce the impact of lactation on suppressing the resumption of ovulation and return of menses associated with it (Knodel, Kamnuansilpa, and Chamratrithirong, 1985). Given the short duration of abstaining from sexual relations following childbirth and the only moderate duration of postpartum amenorrhea, Thai women become exposed to the risk of pregnancy fairly rapidly following childbirth. According to the definition of insusceptibility used in this analysis, almost one fourth of women would be at risk of pregnancy if they did not practice contraception by 2-3 months following childbirth and 80 percent would be at risk by just over one year. 30 An alternative procedure for computing average durations of breastfeeding and postpartum amenorrhea, abstinence and insusceptibility based on current status data is the "prevalence/incidence" method borrowed from epidemiology. In epidemiology, the mean duration of an illness can be estimated by dividing its prevalence by its incidence. In this case, the event of concern is not illness but rather breastfeeding (amenorrhea, etc.). Prevalence Is defined as the number of children whose mothers are breastfeeding (amenorrheic, etc.) at the time of the survey. Ignoring the slight discrepancy caused by multiple births, the number of children being breastfed is the same as the number of breastfeeding mothers. Incidence is defined as the average number of births per month. This average is estimated by summing the number of births over the last 36 months to overcome problems of seasonality and fluctuations associated with small numbers of births during short periods of time. For example, a simple division of the number of mothers breastfeeding, at the time of the survey, by the average number of births per month provides an estimate of the mean duration in months of breastfeeding. One major advantage of the prevalence/incidence method over the calculation of the medians from current status data is that it does not require tabulating data for separate months since birth and hence is not dependent on stability in the monthly estimates of proportions in a given status. Results of the prevalence/incidence estimates of breastfeeding and aspects of postpartum insusceptibility are presented in Table 2.5 according to selected background characteristics. Note that the resulting estimates are means, not medians, as in the previous table. Thus the two sets of estimates are not comparable given the different procedures used to derive them and the different measure of central tendency that they yield. Very little difference in the mean duration of breastfeeding or the mean of the two components of insusceptibility is evident between older and younger mothers. Urban-rural differences, however, are pronounced except in the case of postpartum abstinence. Urban mothers breastfeed considerably less than rural mothers and, not surprisingly, experience substantially shorter postpartum amenorrhea and hence shorter durations of insusceptibility. Regionally, Bangkok stands out in terms of the short durations of breastfeeding, postpartum amenorrhea and insusceptibility. The northeast is characterized by unusually long durations of breastfeeding but not especially long amenorrhea. This finding is consistent with previous surveys and is probably attributable to the very early introduction of supplemental food for infants there (Knodel, Kamnuansilpa and Chamratrithirong, 1985). The duration of breastfeeding shows a strong association with educational level. Women with a primary education or less breastfeed for longer durations on average than women with secondary or higher education. Postpartum amenorrhea also lasts noticeably longer among lesser educated women. Finally, religious differentials are also evident although not expecially pronounced. Moslems appear to breastfeed somewhat longer than Buddhists, experience longer amenorrhea, and abstain for longer periods following a birth. Overall Moslems remain insusceptible for approximately two months longer than do Buddhists. 31 Based on the data presented here, little can he said about recent trends in breastfeeding or the components of postpartum insusceptibility. Although previous surveys have collected data on breastfeeding and postpartum amenorrhea, the procedures used to estimate the average duration are different, thus preventing direct comparisons. Table 2.5 Prevalence/incidence estimates of mean number of months of breastfeeding, postpartum amenorrhea and postpartum abstinence, by selected background characteristics Weighted Background number of characteristic Breastfeeding Amenorrheic Abstaining Insusceptible* births . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Age <30 16.5 7.0 3.3 8.5 1,474 30+ 16.7 7.4 4.0 9.0 737 Urban-rural residence Urban 9.8 4.6 3.6 6.6 388 Rural 18.0 7.7 3.5 9.1 1,823 Region North 14.0 7.7 3.6 9.8 435 Northeast 22.2 7.8 2.9 8.8 766 Central 12.5 6.2 3.6 7.8 414 South 16.9 7.8 4.6 9.4 356 Bangkok 9.8 4.8 3.9 6.7 240 Education No education 18.7 7.3 4.4 9.2 222 Primary 17.9 7.7 3.5 9.1 1,682 Secondary 7.4 3.4 4.1 6.5 202 Higher 7.9 5.0 2.3 5.4 105 Religion** Buddhist 16.2 7.1 3.4 8.6 1,951 Islam 19.1 8.3 4.7 10.5 182 Total 16.6 7.2 3.5 8.7 2,211 Note: Amenorrheic and insusceptible categories exclude pregnant women * Either amenorrheic or abstaining ** Excludes cases whose religion is other than Buddhism or Islam or stated is not 32 Table.1 Percentage of births in the last 3 years whose ~others are still breastfesdi~, and still pcetpart~n ammmrrheic, abstaining, and insusceptible, by single mcmths since birth Weighted Months since n~ber of birth Breastfeeding Amenorrheic Abstaining Insusceptible* births 0 92.5 I00.0 i00.0 i00.0 21 1 88.7 96.8 81.9 98.6 63 2 87.4 66.8 55.4 80.9 64 3 88.8 67.0 22.3 71.8 49 4 84.7 63.2 15.1 67.6 54 5 82.2 49.6 7.3 51.5 53 6 68.9 48.7 2.5 51.1 64 7 81.5 28.1 7.6 32.3 65 8 75.3 52.7 3.0 55.7 46 9 75.6 41.7 0.0 41.7 59 I0 72.8 21.9 12.2 34.1 64 ii 63.6 22.3 8.0 23.9 60 12 67.4 19 • 8 8.5 26 • 8 64 13 62.8 7.2 3.4 10.5 67 14 51.9 17.6 1.7 19.3 73 15 48.4 12.0 4.9 16.9 69 16 29.6 9.3 1.2 7.6 84 17 52.3 4.1 1.2 5.3 56 18 47.1 14.9 6.2 21.1 60 19 30.0 1.4 0.7 2.2 66 20 32.5 3.2 0.0 3.2 68 21 46.0 9.3 0.0 9.3 72 22 30.0 5.6 2.3 7.8 56 23 15.5 1.3 7.7 8.5 56 24 23.2 0.6 9.7 10.3 57 25 35.1 4.0 0.0 4.0 94 26 22.0 7.8 4.2 11.9 56 27 20.9 2.2 4.8 7.0 73 28 19.0 2.8 0.0 2.8 67 29 11,0 0.0 0.0 0.0 62 30 10.4 1.1 2.6 3.7 44 31 16.4 2.2 2.1 4.3 56 32 15.5 1.4 1.0 2.4 63 33 5.0 3.7 0.0 3.7 61 34 5.0 1.2 3.3 4.4 42 35 5.8 0.0 1.7 1.7 42 Total 45.6 19.8 8.9 23.5 2,168 Note: A~enorrheic and insusceptible categories exclude pregnant women. 33 CHAPTER 3 FERTILITY In the TDHS, information on current, past and cumulative fertility was collected. The eligible woman questionnaire contains questions on the total number of live births and surviving children the woman had over her lifetime as well as a detailed birth history. One innovative feature of the TDHS with respect to eliciting the birth history, as noted in Chapter i, was to ask respondents, once all the live births were listed by name, to show documentary evidence in the form of birth certificates or household registration forms in order to improve the accuracy and completeness of the reporting of birth dates by reducing reliance on the respondent's memory for such information. Respondents were able to provide documentation of the birth dates for about half (52 percent) of all the births reported. The percentage for which documentation was provided does not vary much according to the birth year. For example, documentation of birth dates are provided for 52 percent of the births reported as occurring during the first five years preceding the interview compared to 55 percent of the births reported as occurring during the second preceding five year period. For all births occurring during the first five years preceding the survey, both the month and year of birth are known for 97 percent either from documentation or from the mother's report. In only 1 percent of the cases, did both month and year of birth have to be imputed. Although the TDHS collected birth histories only from ever married women, it is possible to calculate fertility measures relating to all women regardless of marital status by assuming that women who were reported as having never married had no children. To the extent non-marital fertility is missed by the survey, however, the assumption of no births to women reported as unmarried will necessarily result in an underestimate of the level of fertility. Unfortunately there is very little systematic evidence on the extent of non-marital childbearing in Thailand. Since marital status is not recorded when births are registered, information on non-marital births is not available from the vital statistics reports and there has been little research on the topic. Nevertheless, while some births undoubtedly occur outside of marital unions, most observers agree that the level of non-marital fertility is likely to be quite low. Moreover, if an unmarried woman is living with her child in a sample household, she might well have been reported as married in the course of eliciting the household listing and be included as an eligible woman. A check of a sample of 500 TDH$ households questionnaires (i00 from each region plus Bangkok) to see if in the listings of household residents there was evidence of children living with unmarried mothers yielded no unambiguous cases of illegitimate children and very few cases which seemed likely to he so. 3.1 Current Fer t i l i ty Levels and Trends Current fertility levels as reflected in the age specific fertility rates and in the summary total fertility rate (TFR) are presented in Table 3.1. Rates are given for three alternative time periods spanning the preceding 12, 24 and 60 months respectively. The longer the period covered, the greater is the amount of fertility experience taken into account and hence the less subject the rates are to random fluctuation. Note should be made of the fact that since these rates are based on retrospective reports of births during the past, and only women up through age 49 were interviewed, the fertility experience of women in the 45-49 age group presented in Table 3.1 is censored to varying degrees depending on the length of the time period covered. For example, births three years prior to the survey to women who were aged 47, 48 or 49 at that time of giving birth will not be available from the birth history data because these women would have been 50 or over at the time of the survey and hence excluded from the sample. For this reason, TFR's are presented both up to age 44 only (since censoring does not affect rates up to this age for the five year period preceding the survey) as well as to age 49 (the more conventional age span covered by the TFR). In any event, given the very low level of fertility of women 45-49 in Thailand, censoring has little effect on the value of the overall TFR for the periods shown. Fertility appears to have continued to decline during the five year period preceding the survey judging from a comparison of the TFR for the three alternative time spans. The 12 month TFR is lower than the 24 month TFR which in turn is lower than the 60 month TFR. Note that the TFR for each successively longer period is inclusive of the preceding shorter period and thus minimizes the appearance of change which is examined more directly in the following two tables. The most striking feature of Table 3.1 is the very low level of recent fertility indicated by the TDHS. For the 12 month period preceding the survey, the TFR indicated is only 2.11 live births per woman. This is below the replacement level for Thailand (which is about 2.25) given current mortality conditions. For the 24 month period preceding the survey, the TFR of 2.21 is just about at the replacement level and for the full 60 month period, the TFR of 2.36 is only slightly above replacement. These rates are low in comparison to other estimates of recent fertility levels, such as from the most recent Contraceptive Prevalence Survey and the Survey of Population Change, and therefore require some comment. A detailed comparison of TDHS fertility rates with those from other sources covering the period between 1970 and 1986 is provided in Appendix B. The general conclusion from the comparison is that the TDHS probably understates the true fertility level by a modest but unknown degree. The most compelling evidence that the recent levels of fertility are probably higher than indicated from the TDHS is provided by a comparison with fertility rates calculated from registered births. It is widely acknowledged that births are underregistered in Thailand. For example, the most recent Survey of Population Change indicates that birth registration is 88 percent complete. Yet if the TFR is calculated from registered births as reported by the Ministry of Public Health without any adjustment for underregistration, the rates for recent years are quite close to those indicated by the TDHS. For example, for the 5 year calendar period from 1982 to 1986, the TFR as indicated by TDHS is only one percent higher than the TFR based on registered births unadjusted for underreqistration. 35 Table 3.1 Fertility rates for 12, 24 and 60 months preceding the survey, for all women (including never-married women), by age of women at time of childbirth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maternal Fertility rates for preceding age at . childbirth 12 months 24 months 60 months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-19 0.049 0.056 0.052 20-24 0.124 0.124 0.132 25-29 0.ii0 0.131 0.129 30-34 0.065 0.070 0.079 35-39 0.042 0.041 0.052 40-44 0.026 0.017 0.022 45-49 0.006 0.004 0.007 15-49 2.11 2.21 2.36 15-44 only 2.08 2.20 2.32 Notes: The preceding time periods to which the fertility rates refer exclude the month of interview. The total for fertility rates represents the total fertility rate (TFR) for women aged 15-49 and 15-44 respectively. Since women aged 45-49 are progressively censored as one moves back in time from the time of interview to five years preceding the survey, total fertility rates are presented both including and excluding women in this age group. Table 3.2 presents recent estimates of fertility for selected periods according to various background characteristics. In addition, the average number of children ever born to women aged 40-49. is shown and serves as a convenient measure of cumulative fertility for women close to the end of the childbearing span. To indicate recent trends in fertility, the TFR (based on women aged 15-44 rather than 15-49 to eliminate the influence of censoring) is shown for the calendar year period 1981-83 and 1984 through the time of the survey in 1987. In addition, for examining differentials in recent fertility levels according to background characteristics, the TFR is shown for the 60 month period preceding the survey (both including and excluding women 45-49) and for the 24 month period prior to the survey (based on women 15-49). 36 The decline in fertility in Thailand over the last two decades is reflected in the large difference for the total sample between current fertility as measured by the TFR and the cumulative fertility of women currently at the end of the childbearing ages, as represented by the mean number of children ever born to women aged 40-49. This latter measure reflects the fertility levels prevailing in the past when these women passed through the reproductive ages. At 4.4 births per woman, cumulative fertility is twice as high as the most recent TFR of 2.2 for the 24 months preceding the survey. The results also indicate that fertility has continued to decline during recent years as evident from the finding that the fertility rate during the 1984-87 period is 16 percent lower than the rate for the 1981-83 period. The recent continuation of a declining trend is confirmed by data on registered births. While the total number of births are likely to be under registered as noted above, there is no evident reason to suspect that the extent of underregistration has deteriorated during the last few years and thus that registered data would indicate a spurious decline. The fact that the TFR based on registered births (with the number of women from the latest NESDB population projections as the denominator), declined by 20 percent between 1981-83 and 1984-86 is supportive evidence that the decline observed in the TDHS data is genuine. A number of differentials in the level and extent of recent decline in the TFR are evident according to selected background characteristics shown. Recent fertility is distinctly lower for urban than for rural women. Lower urban than rural fertility has been a persistent feature of the Thai demographic situation for at least several decades (Knodel, Chamratrithirong, and Debavalya, 1987) and is also indicated by the forthcoming results of the recent Survey of Population Change (SPC) which refers to the period from mid-1985 to mid-1986. However, the TDHS results indicate that during the six years preceding the survey, the extent of decline was greater among rural than urban women suggesting that the urban-rural differential in fertility is narrowing. Regionally, recent fertility is lowest in Bangkok, followed by the central region and then the North. The highest TFR is found in the south followed by the northeast. These regional differentials are similar in ranking to those found in the recent SPC, except that the TFR for the north according to the SPC is lower than that for the central region. Judging from a comparison of the rates for 1981-83 and 1984-87 from the TDHS results, fertility has declined during recent years in all regions, although the decline is quite modest in Bangkok where fertility was already extremely low for the 1981-83 period. The largest absolute decline is found in the northeast, where the TFR (through age 44) declined by almost seven tenths of a child, followed by the central region where the TFR declined by half a child. 37 3.2 Mean number of children ever born to all wcmen (includ/r@ rev~ied) aged 40-49 and total fertility rates for selected periods and for 60 and 24 menths l~reesdi~ the survey, by selected b a ~ characteristics Total fertility rates for wc~en 15-44 Total fertility rates Children for wom~ 15-49 ever born 60 months m~nths prior to survey Background to women prior to dmracterintic 40-49 1984-87" 1981-83 survey 60 24 Urban-rural res ide~ Urban 3.13 1.62 1.73 1.64 1.68 1.65 Rural 4.69 2.42 2.93 2.53 2.57 2.40 Region North 4.49 2.23 2.41 2.27 2.28 2.17 Northeast 4.80 2.47 3.14 2.62 2.65 2.46 Central 4.09 1.90 2.40 1.99 2.04 1.88 &m/th 4.81 3.03 3.43 3.16 3.21 3.06 Bengknk 3.22 1.60 1.68 1.60 1.64 1.65 mucatien No education 5.64 3.66 3.40 3.44 3.52 3.72 Primary 4.40 2.35 2.86 2.47 2.49 2.34 2.51 1.60 1.78 1.65 1.65 1.68 Higher 1.88 1.39 1.51 1.40 1.40 1.39 Total 4.42 2.23 2.66 2.32 2.36 2.21. Notes: Periods to which total fertility rates refer exclude n~mth of interview. Results in this table ere based on all women, including sever-merried women who ere asstmed to have rm births. The number of never-married wcmen is derived by applying a multiplicaticm factor based on the hottsehold questiesnaire to the eligible women sample. *Coverage for 1987 is limited to the months prior to the m~th of interview. 38 Education is also associated with fertility levels. The recent TFR is inversely related to the number of years of schooling of women. Those with no education have by far the highest fertility while those who studied beyond the secondary level have the lowest. It should be borne in mind that the large majority of Thai women currently in the reproductive ages have a primary education and thus the proportions in other educational categories are relatively modest (see Chapter I). Moreover, caution is necessary before interpreting this finding as evidence of a direct educational effect, since educational level is strongly associated with other characteristics which could have important bearing themselves on fertility. For example, women with no education are disproportionately made up of Moslems and ethnic minorities, both of whom are likely to be characterized by high fertility for reasons other than simply educational differences. The pattern of recent change in fertility according to education is irregular: the TFR shows an increase between 1981-83 to 1984-87 in the group with no education and a decline for the other groups. Again the relatively small numbers of women in the categories other than primary education counsel caution in interpreting their fertility trends. With data on complete birth histories such as collected in the TDHS, a more extensive examination of trends is possible than simply a comparison of the TFR over the last few years prior to the survey. Age specific fertility rates are presented in Table 3.3 for successive 5 year periods preceding the survey. Use of birth histories for analysis of trends places a great burden on the quality of data, which should always be interpreted with caution. Possible omission (or even false inclusion) and incorrect dating of events will affect the accuracy of trends. In the case of the TDHS, the problem of misdating of events is minimized because respondents were requested to show documentation of the birth dates of their children whenever possible. The comparison of fertility rates calculated from the TDHS birth history data with estimates of fertility from external sources presented in Appendix B suggests that the overall fertility level may be understated. Nevertheless the evidence does not suggest any greater omission of more distant births than of recent births. Hence the trends reflected in the TDHS birth history data may be relatively accurate. Note that the age-specific schedule of rates are progressively censored as time before survey increases. The bottom diagonal of estimates (enclosed in parentheses) is partially censored. The rates indicate a clear and consistent pattern of fertility decline over at least the last two decades. For virtually every age-group, fertility has declined steadily during the periods for which rates could be calculated. The only minor exception is the 15-19 year old age-group for which a steady fertility decline is evident over the last 20 year period but not for the earlier period. To facilitate an examination of the relative decline in fertility by age-group, the percent decline in age specific fertility rates between each successive 5 year period prior to the survey and the most recent five year period, i.e. 0-4 years prior to the survey, can be calculated based on the rates provided in Table 3.3. The results of such a set of calculations are presented in Table 3.4. By reading down each column, the age pattern of fertility decline is readily apparent. In general, the older the age group, the greater the relative decline in fertility has been between any period in the past and the most recent five year period. For example, the fertility rate for women age 15- 19 declined by 28 percent between the period 10-14 years before the survey and 0-4 years before; in comparison the rate for women aged 30-34 declined by 55 39 Table 3.3 Age-period fertility rates (per 1,000 women including never-married), by age at time of childbirth Maternal Years prior to survey age at time . of birth 0-4 5-9 10-14 15-19 20-24 25-29 30-34 15-19 52 62 72 78 77 81 (51) 20-24 132 172 192 245 258 (261) - 25-29 129 158 219 262 (311) 30-34 79 118 176 (235) - 35-39 52 79 (129) - 40-44 22 (42) 45-49 (7) - Notes : Results in this Table are based on all women, including never-married women, who are assumed to have no births. The number of never-married women is derived by applying a multiplication factor based on the household questionnaire to the eligible women sample. Results in parentheses are based on partially censored observations. Tab le 3.4 Percentage decline in fertility rates between successive five year periods prior to the survey and the period 0-4 years prior to the survey, by age at time of childbirth Maternal Years prior to survey age at time . of birth 5-9 10-14 15-19 20-24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-19 17 28 33 32 20-24 23 31 46 49 25-29 18 41 51 (59) 30-34 33 55 (66) 35-39 34 (60) - 40 -44 (48) - - Note : Based on rates presented in previous table. Figures in parentheses are based on partially censored information. 40 percent between the same the periods. Almost without exception, the older the age-group, the greater is the percent that fertility declined. The results from the TDHS presented so far clearly indicate a substantial and relatively steady decline in fertility during the recent past. Figure 3.1 compares the trend in the TFR based on the TDHS with the trends based on data from the Survey of Fertility in Thailand (SOFT), estimates based on the "own children" technique as applied to the 1980 census, and uncorrected registration data (in combination with population estimates of the base population). In order to make this comparison, the TFR as derived from the TDHS has been adjusted to allow for the effect of censoring of fertility rates at the older ages for periods in the past.* In addition, given that rates for SOFT and TDHS are from sample surveys, they are presented as two year moving averages in order to stabilize the trend they show. The sources are quite consistent in portraying a more or less steady fertility decline over the last two decades. Several other features of the comparison are worth pointing out. First, the series from TDHS fits quite well with the series from SOFT, both in terms of overlapping fairly closely for the several years shown in common and in continuing the trend of decline evident in the earlier SOFT series. Second, while both the SOFT and the TDHS series are * The adjustments were made as follows. Total fertility rates derived directly from the birth histories collected in the TDHS were calculated for successive 12 months periods preceding the survey based on ages 15-49 for the first 3 prior 12 month periods (covering 1984/85-1986/87), ages 15-44 for the next 5 prior 12 month periods (covering 1979/80-1983/84), ages 15-39 for the next 5 prior 12 months periods (covering 1974/75-1978/79), and ages 15-34 for the next 4 prior 12 month periods (covering 1970/71-1973/74). In order to convert the "partial" total fertility rates derived from the TDHS for the years prior to 1984/85 to complete TFRs covering the entire reproductive age span 15- 49, the ratio of the complete to the partial rate was calculated from the age specific fertility rates from the 1980 census based on the "own children" technique and the most recent SPC. Note that the census estimates refer to 12 month periods beginning in April and ending in March of the next year and thus are almost equivalent to the 12 month periods for which rates from the TDHS have been calculated (which refer to periods from approximately May to April). The partial TFR from the TDHS is then multiplied by the appropriate ratio to estimate the complete rate. For example, the ratio of the TFR 15-49 to the TFR for ages 15-34 was calculated directly from the "own children" estimates for 1970/71-1973/74 and applied to the partial TFRs from the TDHS for each of the equivalent twelve month periods to obtain a TFR for ages 15-49 for these years. In like manner, the TFRs for ages 15-39 and 15-44 for subsequent years were converted to complete TFRs for ages 15-49. Note that for the years 1970/71 to 1979/80, the adjustment factors were calculated directly from the age specific rates from the "own children" estimates. However for 1980/81 to 1983/84, the ratio of the TFR for ages 15-49 to the TFR 15-44 was obtained by interpolating between values of the ratio for 1979/80 based on the "own children" estimates and the ratio for 1985/86 based on the most recent SPC. Note that in all cases these inflation factors depend only on the age pattern of fertility and not the level of fertility reported by the sources from which they are derived. 41 Figure 3.1 Comparison of the trend in the TFR based on TDHS with trends based on data from SOFT, the 1980 census, and vital registration < H r~ 8 7 6 5 4 2 %%% %,. TDHS (2 - year moving average) SOFT (2 - year moving average) Own children (1980 Census) . . . . . . . . Birth registration data (unadjusted) k I I L r I i I I I [ I I i I I I I r I I I i i i i '60 '65 '70 '75 '80 '85 42 quite parallel to the "own children" estimates from the 1980 census, they both generally fall below these estimates. Third, not only is the trend from the TDHS estimates parallel to that of the TFRs based on uncorrected registration data for the years shown, but the average level is relatively similar (although higher is some years and lower in others). This last feature suggests that while the trend shown by the TDHS is probably correct, the level of fertility may be underestimated since any correction for underregistration of births would raise the average level of the TFRs calculated from registration data above those from the TDHS. 3.2 Cumulative Fertility In the TDHS questionnaire, the total number of children ever born has been ascertained by a sequence of questions designed to maximize recall. Each woman was first asked about the number of sons and daughters living with her, then about the number living away from home, and finally about any children that died. Experience suggests that by asking in this way about the separate components of children ever born that omissions of births can be kept to a low level. Since life-time fertility reflects the cumulation of births over the past, it has limited direct relevance to the current situation. Nevertheless, such data provides important background information for understanding current fertility. The data in Table 3.5 are perhaps the most common fertility statistics derived from surveys. The number of children ever born is presented here for all women (assuming that never-married women had no births) and for ever-married and for currently married women. Differences in results between all women and ever-married or currently married women is greatest at the younger ages because of the large proportion of women who are still single and presumed to have no births. In contrast, differences between ever-married and currently married women are modest at all ages, although slightly greater at older ages, and reflect the impact of marital dissolution. The overall impact of marital dissolution, however, can not be judged from this comparison since many women whose marriage ends prior to completion of the reproductive age span remarry and hence are currently married at the time of the survey. Since voluntary childlessness is rare in Thailand, the extent of primary sterility can be judged more or less from the percent of married women who are childless at the end of the childbearing ages. Primary sterility is clearly very low in Thailand as indicated by the finding that less than 3 percent of ever-married women aged 40 and over have no children. The much higher fertility rates of the past compared to the present are evident in the average number of children ever born to these same women. Ever-married women aged 40-44 born an average of 4.2 live births while those age 45-49 bore an average of 5.2 births. Among this oldest age-group, over one fourth gave birth to 7 or more children and only 15 percent gave birth to 2 children or less. This is quite a contrast to the low fertility desires younger married women say they wish to have (see Chapters 4 and 5). Given the current widespread practice of contraception, these younger women are likely to limit their actual family sizes to the small desired numbers and within the next two decades cumulative fertility of women at the end of the reproductive years is certain to be far lower than it is today. 43 Table 3.5 Percent distribution of children ever born among all women, ever-married women, and currently married women, by current age Ntmtber of ckildren ever born Weighted Current Total ntm~er of Mean age 0 1 2 3 4 5 6 7 8 9 10+ percent women C~ All ~ - inclucli~ never-married* 15-19 92.5 6.4 i.I 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 i00 0.09 20-24 59.6 24.9 12.1 2.8 0.5 0.i 0.0 0.0 0.0 0.0 0.0 100 0.60 25-29 31.4 22.8 27.9 12.4 4.2 1.3 0.0 0.i 0.0 0.0 0.0 100 1.39 30-34 17.8 13.0 30.2 20.9 11.9 3.9 1.6 0.4 0.3 0.0 0.0 100 2.18 35-39 11.9 7.9 20.4 23.4 16.9 9.4 5.0 2.7 0.8 i.i 0.3 100 3.03 40-44 8.9 5.1 ii.0 19.1 19.6 13.6 i0.I 6.0 3.2 1.8 1.5 100 3.91 45-49 6.1 4.6 8.0 11.5 13.6 14.3 14.2 10.5 7.5 4.2 5.5 100 4.98 Total 40.4 13.5 15.8 11.3 7.5 4.4 2.9 1.8 1.1 0.6 0.6 100 1.82 Ever-marriud m 15-19 55.3 37.9 6.6 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100 342 0.52 20-24 22.7 47.7 23.2 5.3 1.0 0.2 0.0 0.0 0.0 0.0 0.0 i00 1,004 1.15 25-29 9.9 29.9 36.6 16.2 5.5 1.7 0.0 0.1 0.0 0.0 0.0 100 1,309 1.83 30-34 5.1 15.1 34.8 24.2 13.7 4.5 1.8 0.5 0.4 0.0 0.0 I00 1,328 2.52 35-39 3.0 8.7 22.5 25.7 18.6 10.4 5.6 3.0 0.9 1.2 0.4 100 i,ii0 3.34 40-44 2.7 5.5 11.7 20.5 20.9 14.6 10.8 6.4 3.4 1.9 1.6 100 877 4.18 45-49 2.3 4.8 8.3 12.0 14.1 14.9 14.8 10.9 7.9 4.4 5.7 100 805 5.18 Total 10.2 20.4 23.8 17.0 11.3 6.6 4.4 2.7 1.6 1.0 0.9 100 6,775 2.75 Ct~rently married 15-19 56.4 36.5 6.8 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100 334 0.51 20-24 22.6 47.0 23.9 5.4 0.9 0.2 0.0 0.0 0.0 0.0 0.0 100 957 1.16 25-29 I0.I 28.5 37.3 16.5 5.7 1.8 0.0 0.i 0.0 0.0 0.0 100 1,243 1.85 30-34 5.1 13.6 35.1 25.0 14.0 4.5 1.9 0.5 0.4 0.0 0.0 100 1,250 2.55 35-39 2.9 7.3 22.1 26.3 19.3 10.9 5.6 2.9 1.0 1.4 0.4 100 1,019 3.40 40-44 2.7 4.8 10.8 20.2 21.7 13.4 11.6 7.3 3.6 2.2 1.7 100 758 4.28 45-49 1.2 3.7 7.5 11.5 15.6 15.0 15.4 II.0 7.5 4.8 6.8 100 676 5.38 Total 10.5 19.7 24.2 17.1 11.6 6.3 4.4 2.7 1.5 1.0 1.0 100 6,236 2.74 * Never-married women are asmmm~ to have no children. The ntmber of neve~ied women is derived by applying a multiplication factor based on the household questionnaire to the eligible women sample. The weighted ntmbar of all women is not shown because it includes the derived nt~ber of never-married women. 44 Table 3.6 indicates cumulative fertility as measured by children ever born to ever-married women according to marriage duration and age at first marriage. The purpose of this tabulation is to permit an assessment of the relationship between age at marriage and the rate of marital childbearing. Note that beginning at higher durations, the higher age at marriage cells are empty because the upper limit of the age range of the sample (49) is exceeded (e.g., a woman could not be in the sample who married at 25+ and has been married 25-29 years given that she would be at least 50 years old at the time interviewing took place). At marriage durations 0-4, there is little difference in the average number of children ever born according to age at first marriage. As marriage duration increases, an inverse association between age at marriage and cumulative fertility becomes evident, probably reflecting the higher fecundity of earlier marrying women due to their younger age. Table 3.6 Mean nm~er of children ever born to ever-married women, by age at first marriage and duration since first marriage Duration since first marriage ~e at first marriage <15 15-17 18-19 20-21 22-24 25-27 28-29 30+ All ages 0-4 0.9 0.8 0.7 0.7 0.7 0.8 0.7 0.7 0.7 5-9 2.1 1.9 1.8 1.9 1.8 1.8 1.6 1.4 1.8 i~14 2.7 2.7 2.5 2.4 2.6 2.3 2.1 2.2 2.5 15-19 3.8 3.6 3.5 3.3 3.2 3.0 2.7 3.4 20-~ 4.6 4.5 4.7 4.1 4.1 3.3 - 4.4 25-29 5.5 5.5 5.2 5.0 5.0 5.3 30+ 7.1 6.2 6.5 - - - 6.4 All~rations 3.9 3.1 2.9 2.6 2.3 1.9 1.7 1.5 2.7 45 3.3 ~ge at F i r s t B i r th The onset of childbearing is an important demographic indicator. In many countries, postponement of first births, reflecting a rise in age a t marriage, has made a large contribution to the overall fertility decline. In the case of Thailand, the contribution has been modest but not inconsequential (Knodel et al., 1982). The proportion of women who become mothers before the age of 20 is a measure of the magnitude of adolescent fertility, which is a major health and social concern in many countries. Furthermore, early motherhood is associated with higher subsequent fertility. Table 3.7 shows the percent distribution of women by age at first birth according to their current age. The tabulation includes a category for no birth, and refers to all women, including those who have never married (under the assumption that they have had no children). Median ages at first birth are also presented for all cohorts for which at least 50 percent of the women had a first birth (i.e. age groups 25-29 and above). An increase in the median age at first birth of approximately a year and a half is evident between the cohort of women aged 45-49 and cohort aged 25-29. Given that the timing of marriage and first childbearing are closely linked and, as documented in Chapter 2, that the age of marriage has risen, this increase in age at first birth is not surprising. Indeed, the median age at marriage rose by exactly the same amount between these two cohorts. Very few women in Thailand start childbearing before age 15 and the proportion of women who had a first birth before age 20 decreased sharply from 32 percent for women aged 45-49 to 24 percent for women aged 20-24. Table 3.7 Percent distributio~ of all women (including never-married) according to age at first birth (including the category "no birth"), by current age Age at first birth Current No Total age birth <15 15-17 18-19 20-21 22-24 25-27 28-29 30+ percent Median* 15-19 92.5 0.I 5.3 2.1 0.0 0.0 0.0 0.0 0.0 100 - 20-24 59.6 0.8 8.5 14.7 11.8 4.6 0.0 0.0 0.0 100 - 25-29 31.4 0.4 8.9 15.8 18.2 18.5 5.8 1.0 0.0 100 23.0 30-34 17.8 0.7 I0.0 17.4 16.5 21.0 9.6 4.6 2.3 i00 22.7 35-39 11.9 0.9 10.6 18.1 18.0 18.0 12.1 5.0 5.3 i00 22.3 40-44 8.9 0.7 7.6 20.0 19.8 20.8 11.1 3.6 7.4 i00 22.2 45-49 6.1 1.2 10.9 20.0 21.4 20.2 11.4 4.4 4.5 100 21.6 ALl ages 40.4 0.6 8.5 14.1 13.5 12.9 5.8 2.1 2.0 100 Note8: Results in this table are based on all ~n~n, including never-married women, who are assumed to have no births. The number of never-married women is derived by applying a multiplication factor based on the household questio~maire to the eligible women sample * Omitted for ages under 25 and total due to censoring 46 Table 3.8 presents the median age at first birth for different age cohorts according to selected background characteristics. The age at first childbearing has increased more in urban than rural areas. Overall, urban women start reproduction four years later than their rural counterparts. Regionally age at first birth has risen most in Bangkok and the central region. The pattern is more irregular in the other regions showing little tendency to increase in the northeast or the south and showing an increase in the north mainly among the three youngest cohorts. Overall, the age at the start of reproduction is not greatly different among the regions except for Bangkok where women start childbearing considerably later than elsewhere. Educational differentials are quite pronounced indicating a substantial increase in the age at first birth associated with increased level of schooling completed. This association is evident for almost all age cohorts. Interestingly, there is little evidence of a consistent increase in ages at first childbearing for any of the separate educational categories suggesting that the increase observed nationally is largely a product of the increasing educational levels of younger cohorts. Table 3.8 Median age at first birth among all women (including never-married) aged 25-49 years, by current age and selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Current age Background . characteristic 25-29 30-34 35-39 40-44 45-49 Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Urban-rural residence Urban 26.3 25.8 24.8 23.8 25.9 Rural 22.1 22.2 21.7 21.8 21.3 21.9 Region North 22.1 21.3 21.1 21.1 21.0 2]4 Northeast 21.8 22.2 22.2 22.1 21.8 ~.0 Central 23.6 23.3 23.0 22.4 21.2 ~8 South 22.8 22.4 21.2 21.2 22.5 ~.i Bangkok 27.4 27.0 25.3 23.4 ~.8 Education No education 22.3 20.9 20.3 20.6 20.0 ~.7 Primary 21.7 22.1 22.0 22.1 21.7 21.9 Secondary 25.5 26.4 25.6 24.5 26.8 ~.6 Higher 28.9 29.8 28.2 29.5 ~.8 Tota l 23.0 22.7 22.3 22.2 21.6 22.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notes: Results in this table are based on all women, including never-married women, who are assumed to have no births. Median is not shown for categories for which less than 50 percent of the women have had a birth. The number of never-married women is derived by applying a multiplication factor, based on the household questionnaire to the eligible women sample 47 CHAPTER 4 FERTILITY REGULATION This chapter begins with an appraisal of the knowledge, the source of supply and the perceived problems (if any) for different contraceptive methods and then moves on to a consideration of current and past contraceptive practice. For users of periodic abstinence, knowledge of the ovulatory cycle is examined while for those relying on sterilization, the timing of method adoption is reviewed. Special attention is focused on nonuse, reasons for discontinuation, and intention to use in the future. The chapter concludes with tabulations on exposure to and acceptability of media messages about family planning. These topics are of practical use to policy and program staff in several ways. The early sections concern the main pre-conditions to adoption of contraception such as knowledge of methods and sources of supply. Levels of use of contraceptives provide the most obvious and widely accepted criterion of success of any family planning program. The examination of use in relation to need pinpoints segments of the population for whom intensified efforts at service provision are most needed. In Thailand, where most women have tried at least one method, practical problems with particular methods, or in obtaining supplies and advice, are potential obstacles to further advances in the program. Survey findings on these topics can provide guidance to administrators for the improvement of services. One simple framework for understanding the determinants of contraceptive use divides these determinants into two types: demand factors and cost factors. It should be born in mind, however, that, in reality, the two may not be independent of each other. The creation of conducive cost factors may well strengthen demand and vice versa. The TDHS contained questions dealing with a variety of aspects of demand and cost factors. Demand factors consist of the desire of couples to postpone or terminate childbearing. These are treated in the following chapter. Cost factors consist of attributes of contraception and contraceptive services as perceived by actual and potential users. These include: knowledge of methods; acceptance that the regulation of childbearing by contraception is both possible and moral; knowledge of sources of advice and supply; and a belief that at least some methods present no major barriers to use. A further set of cost factors is likely to influence whether initial and often tentative adoption of a method is sustained or discontinued. These include: satisfactory experiences with the method and the source of supply, and ability to use the method effectively. A number of these cost factors are addressed in this chapter. 4.1 Contraceptive Knowledge Knowledge of contraceptive methods and of places where methods can be obtained are preconditions for their use. The TDHS provides information on the level of knowledge of both methods and service providers. Knowledge data was obtained first by asking the respondent to name the ways that can be used to avoid getting pregnant. If a respondent did not spontaneously mention a particular method, the method was described by the interviewer and the respondent was asked if she recognized the method. Descriptions were included in the questionnaire for nine methods: the pill, IUD, injection, condom, vaginal methods (diaphragm, foam and jelly), female sterilization, male sterilization, periodic abstinence (rhythm) and withdrawal. In addition, other methods mentioned by the respondent (e.g., herbs) were recorded. Finally, for any modern method that she recognized, the respondent was asked if she knew about a place or a person from which she could obtain the method and what main problem, if any, was associated with the method. If she reported knowing about periodic abstinence, she was also asked if she knew a place or a person from which she could get information about the method. As shown in Table 4.1, knowledge of at least some method of contraception is practically universal among married Thai women in reproductive ages. Over 99 percent of both ever-married and currently married women are aware of at least one modern contraceptive method. Knowledge of oral contraception, the IUD, injection, and both female and male sterilization are all close to universal with well over 90 percent of respondents either spontaneously mentioning these methods when asked what methods they know or indicating recognition when the method was read out to them by the interviewer. Condoms are also widely known although to a somewhat lesser extent than the other modern methods. In contrast, vaginal methods (diaphragm, foam or jelly) are not widely known. Likewise, familiarity with periodic abstinence and withdrawal is acknowledged by only a minority of respondents. Table 4.2 shows contraceptive knowledge according to selected background characteristics. Knowledge of at least one method is virtually universal among all the subgroups of the population. Likewise over 90 percent of each subgroup knows the pill and injection. Some differences with respect to knowledge of other specific methods, however, is evident. In general, differentials are most pronounced for the lesser known methods. For example, knowledge of vaginal methods, periodic abstinence, and withdrawal is considerably higher among urban than rural women and increases sharply with educational level. Knowledge of withdrawal is far more common in Bangkok and the south and is the only method better known among Moslems (who are concentrated in the south) than Buddhist. Table 4.3 presents the distribution of responses according to the main problem perceived about particular methods among women who knew the method. If this information is reasonably meaningful, it could be useful in identifying obstacles to the use of specific methods and be helpful in guiding educational and publicity campaigns. It should be noted that many respondents had difficulty answering this question, especially if they had never used the method. Thus interviewers often needed to coax respondent to elicit an answer. For a number of the methods, even probing failed to obtain an answer and substantial percentages fall in the "don't know" category. Based on the percentages who explicitly indicated there was no problem, the most problem free methods in the perceptions of respondents were sterilization (both male and female) and withdrawal. However, if the "don't know" category is assumed to represent persons who do not perceive a problem with the method and is combined with the "no problem" category, vaginal methods, the condom and withdrawal are perceived to be the most trouble free methods. It seems likely that the results in Table 4.3 reflect in part how well known a method is rather than just how problematic it is. Quite plausibly, methods that are known by smaller proportions of respondents are not 49 Table 4.1 Percentwe knowing any mthud, }mozlg any modern m M and )mowing specific ccatraceptive methods, mc~ ever-married and currently married women, by curront age F~ml e Male Weighted Any Any modern Vaginal sterili- sterili- Pexicdic with- nt~ber of Age method metIK~* Pill lid Injection methods Cclz](m zation zation abstinence drawal Other women E~r-marriel wum~ 0 15-19 99.5 99.2 98.8 90.5 98.1 15.1 88.5 96.1 95.3 21.8 20.2 18.5 342 20-24 99,5 99.5 99.0 94.6 97.9 13.3 90.7 97.9 95.0 28.7 30,2 18.5 1,004 25-29 99.7 99.7 98.9 95.1 97.5 16.7 91.8 97.8 96.0 33.4 32,3 20.5 1,309 30-34 99.9 99.9 99.8 96.9 98.8 17.4 92.7 99.2 98.4 31.9 29.9 16.0 1,328 35-39 99.6 99.6 98.9 95.7 97.8 19.2 90.0 98.5 96.8 27.9 27.3 14.7 1,110 40-44 99.1 99.1 96.9 93.2 95.0 15.8 82.7 96.6 93.8 26.8 23,2 9.0 877 45-49 98.4 99.3 96.3 90.5 91.6 17.6 72.0 94.4 91.6 17.8 17.3 5.7 805 ~J~es99.4 99.4 96.5 2 .5 96.9 ~.6 ~.8 ~.6 %.6 28.2 ~.1 ~.0 6 ,~ Cuxrent1¥ married 15-19 99.5 99.2 98.8 90.6 98.0 14.7 88.2 96.4 95.5 22.1 20.0 19.0 334 20-24 99.4 99.4 98.9 94.9 97.9 13.5 91.1 98.4 95.2 29.3 30.9 18.9 957 25-29 99.9 99.9 99.2 95.5 97.8 16.9 92.3 98.0 96.2 33.3 31.9 20.6 1,243 30-34 99.9 99.9 99.8 97.1 98.8 17.5 92.9 99.2 98.6 32.2 30.1 16.4 1,250 35-39 99.6 99.6 99.0 95.9 97.8 19.2 90.3 98.5 97.1 28.3 27.7 14.6 1,019 40-44 99.6 99.6 99.6 93.7 95.8 15.9 84.2 97.4 94.4 26.5 23.0 9.0 758 45-49 98.3 98.2 95.9 90.3 92.0 18.2 71.6 94.3 92.5 18.1 18.4 5.8 676 ~es99.6 99.5 96.7 2.7 ~.2 ~.8 96.5 ~.8 96.0 28.6 ~.5 ~.4 6 ,~ * Includes pill, IUD, injections, vaginal me~ (ctia1~hra~n/foam/jelly), female sterilization, and male sterilization Table 4.2 Percentage of ever-married women aged 15-49 }~owing specific methods and any method, by selected 5ack~ charac- teristics FaQale Male Weighted Background Vaginal sterili- sterili- Periodic Any number of Characteristic Pill IUD Injection methods Condum zation zation abstinence Withdrawal Other method wa~en ~n Urbm-nzal I Urhen 98.9 94.4 96.6 25.5 91.9 98.3 96.9 54.7 48.9 16.9 99.6 1,233 Rural 98.4 94.5 96.9 14.6 86.9 97.4 95.4 22.3 22.3 14.6 99.4 5,542 North 98.0 92.0 96.5 12.1 88.1 96.9 94.4 23.7 23.3 15.5 99.0 1,396 Northeast 98.9 97.8 96.7 14.4 88.4 98.3 96.6 16.8 14.3 16.3 99.7 2,365 Central 99.2 93.9 98.0 21.0 84.3 98.3 95.7 33.5 29.4 9.8 99.9 1,450 South 96.9 91.2 96.8 14.9 89.9 95.2 94.3 34.4 47.7 23.1 98.6 833 Bangkok 98.5 93.3 95.6 25.6 90.3 97.8 96.3 56.1 47.9 11.4 99.5 732 ~ t ~ No education 93.3 79.1 90.0 8.4 71.0 88.9 85.5 9.7 10.8 6.9 96.1 657 Pri ~.-~ry 99.0 95.7 97.4 14.3 88.2 98.4 96.3 22.4 22.1 14.0 99.8 5,316 99.5 98.5 99.0 30.7 98.3 99.2 98.9 75.0 67.6 26.2 99.8 521 H/gher i00.0 99.0 98.8 54.1 i00.0 i00.0 100.0 93.7 84.9 32.5 I00.0 281 Buddhist 99.0 95.7 97.1 16.7 88.1 98.1 96.2 28.4 26.4 15.3 99.6 6,275 Islam 92.8 81.1 92.3 13.5 82.7 88.6 86.7 25.1 39.1 11.5 96.9 359 'rota2 ~.5 ~.5 ~.9 ~.6 ~.8 ~.6 ~.6 ~.2 ~.1 ~.0 ~.4 6 ,~ *Excludes cases whose religion is other than Buddb/sm or Islam or is met stated Table 4.3 Percent distribution according to the main problem perceived in u~ing methods (if any), by method, for ~men who have ever heard of the method Fp.ml e Male Main problem Vaginal sterili- sterili- Periodic perceived Pill I~D Injection methods Cor~k~ zatica zation abstinence Withdrawal Other U1 ~o No problem 32.1 29.7 35.1 30.2 43.3 52.7 48.9 41.0 49.8 43.9 Causes infecundity 3.5 1.0 19.2 0.2 0.0 0.4 0.i 0.2 0.0 1.2 Not effective 0.9 17.1 0.6 3.8 6.4 1.7 2.0 21.4 10.7 1.6 disapproves 0.0 0.2 0.0 0.I 1.3 0.0 0.i 0.9 3.9 0.I Health concerns 3.3 9.6 5.1 2.6 0.6 2.6 1.0 0.0 0.3 2.3 Access/availability 0.0 0.I 0.i 0.0 0.I 0.0 0.I 0.0 0.0 0.1 Easy to make mistake 0.5 0.0 0.0 0.2 0.0 0.0 0.0 11.7 0.3 0.0 Inconvenient to use i.i 0.7 0.I 3.8 3.0 0.0 0.0 4.2 4.7 1.2 Emotional/sexual reactions 3.8 1.5 2.2 1.5 0.5 10.6 9.4 0.2 2.2 1.8 Can't work 0.7 0.7 0.4 0.I 0.0 7.2 13.2 0.0 0.0 0.2 Painful 0.6 15.6 0.8 3.5 i.i 4.6 0.7 0.0 0.1 2.6 Weight change 8.0 3.2 10.3 0.5 0.i 5.6 0.7 0.0 0.0 i.i Allergic reaction 38.3 1.4 12.8 0.7 0.5 2.7 0.3 0.i 0.2 1.7 Ikm't }mow* 7.1 19.4 13.2 52.8 43.1 11.9 23.5 20.4 27.8 42.3 Total i~-~ c~t I00 100 100 100 I00 i00 100 i00 I00 i00 Web.ted ma~er of ~ 6,674 6,399 6,562 1,127 5,950 6,611 6,480 1,911 1,839 1,019 *Includes a small number of cases for whom no answer was recorded particularly well known even among those who have heard of the method. Hence, well known and more commonly practiced methods such as the pill, IUD and injection may elicit answers about a problem just because they are better known. This is important to consider because it is may not be so that methods for which few problems are mentioned, such as vaginal methods, condoms or withdrawal, if given more publicity, would necessary have wide appeal simply because respondents who knew of these methods could not cite a problem. Despite these problems with responses to the question about perceived problems associated with different methods, several interesting features emerge from the results. Almost no one mentioned availability or accessibility (which includes cost) as a major problem for any of the methods. Of problems that are more commonly mentioned, quite different ones show up for different methods. By far the most common problem mentioned in connection with the pill is the possibility of an "allergic" reaction, which includes a variety of negative side effects including headaches, dizziness or nausea. Weight change was also mentioned as a problem of the pill by a substantial percent of respondents. In contrast, the IUD is perceived by significant numbers of respondents as not being effective or as being painful while the injection is associated with causing infecundity. The category "health concerns" includes concern about bleeding, which is probably the reason why health concerns are cited most frequently with the IUD and injection. Both male and female sterilization are associated with loss of sexual interest and loss of ability to do heavy work. Finally, periodic abstinence is perceived to be ineffective or susceptible to mistakes in use. Table 4.4 indicates that most women who knew a specific method could also mention a source where the method (or advice about it) could be obtained. Again many respondents found this question confusing, particularly if they were already using another method or had no intention to use the method. Frequently the question had to be repeated several times to obtain an answer. Nevertheless, the pattern of responses conform largely to where specific methods can actually be obtained and, at a minimum, indicate that Thai women are well informed about how to obtain contraceptive methods (especially considering that several of the methods are virtually universally known). This is not surprising given the very high levels of current and ever-use of contraception discussed below. 4.2 Contracept ive Use Thailand has experienced a virtual reproductive revolution over the last two decades during which contraceptive prevalence rose from low levels to levels which are almost as high as in the economically more advanced countries of the West. According to the first national survey providing prevalence levels (Round 1 of the National Longitudinal Study) taken in 1969 (rural) and 1970 (urban), 19 percent of currently married women aged 15-44 had ever practiced contraception and 15 percent were currently practicing a method (Knodel and Debavalya, 1978). By 1984, according to CPS3, 82 percent of ever-married women 15-49 had ever-used contraception and 65 percent of currently married women aged 15-44 were currently practicing (Kamnuansilpa and Chamratrithirong, 1985). Results from the TDHS indicate that ever-use has remained at this extremely high level and that current use has increased even further. 53 Table 4.4 Perc~t distrilmtion of women who know a specific method accordi~ to supply source nvmed (if any) F~,~!e Male Vaginal sterili- sterili- Periodic Source Pill l~ Injection methods Condom ~tion ~tio~ a~stinence Other to ~ t hospital 15.7 58.2 31.6 41.3 16.4 85.0 78.0 20.0 65.2 Govt. health center 59.3 27.8 49.6 25.0 48.7 5.6 7.9 24.8 4.9 Family p1~nni~ clinic 0.3 0.2 0.4 0.6 0.3 0.I 1.2 1.0 0.2 Mobile cli,Hc 0.0 0.2 0.4 0.0 0.2 0.2 2.5 0.2 0.5 Health volunteer 1.0 0.0 0.0 0.0 0.3 0.0 0.0 0.3 0.1 ~ ing 0.0 0.0 0.0 0.I 0.0 0.0 0.0 17.2 0.i Private hospital or clinic 4.9 4.5 11.2 7.5 1.6 4.8 4.2 7.0 4.5 Pharmacy 12.1 0.0 0.5 3.3 17.7 0.0 0.0 0.I 1.6 Shop 0.9 0.0 0.0 0.2 0.9 0,0 0.0 0.0 0.6 M~ center or Ba~kok health center 4.2 4.9 4.6 4.8 2.4 2.7 2.3 3.6 4.6 Friends, relatives 0.I 0.0 0.i 0.3 0.2 0,0 0.0 15.0 1.4 Other 0.3 0.0 0.1 0.0 0.2 0.0 0.0 2.3 1.7 Nowhere 0.0 0.0 0.0 0.1 0.0 0,0 0.0 1.7 0.0 Don't )mow* 1.2 4.1 1.5 16.8 10.9 1.5 3.9 6.5 14.6 Total l~rCmt 100 100 100 100 100 100 100 100 100 We~bted mmber ,~ wmm 6,674 6,399 6,562 1,127 5,950 6,611 6,480 1,911 1,019 * I~cludes a _~11 nt~0er of cases fo~ whom no answer was re~orded The vast majority of either ever-married or currently married women in the reproductive ages interviewed in the TDHS indicate they have used contraception at sometime. As table 4.5 indicates, 82 percent of ever-married women and 84 percent of currently married women aged 15-49 indicate they ever used a contraceptive method, Almost as high percentages indicate that they have ever used at least one modern method. The pill is by far the most common method ever used, with more than half (56 percent) of ever-married women indicating use at sometime. Injection is the second most common method ever used with more than one in four (27 percent) of ever-married women indicating use either at the present time or in the past. Female sterilization is a close third. Only very small proportions of respondents indicate they have ever used periodic abstinence or withdrawal and use of vaginal methods or Norplant, which has only been recently introduced on a pilot project basis are almost entirely absent among Thai women. The TDHS indicates that contraceptive prevalence as measured by current use of a contraceptive method is now higher than ever before, continuing the rapid increase evident from previous surveys. Rates are shown in Table 4.6 both for currently married women aged 15-49 and aged 15-44. Previous studies of contraceptive prevalence in Thailand have typically focused on the 15-44 age range given the very low reproductive potential of women aged 45-49. To maintain comparability, the following discussion of contraceptive prevalence focuses on currently married women aged 15-44. This restriction to women 15-44 is only maintained when discussing prevalence and for other aspects of the analysis, the full 15-49 age range is used. Contraceptive prevalence among currently married women 15-44 has reached 67.5 percent by 1987. This represents an increase over the equivalent prevalence rate of 64.6 for 1984 found by CPS3. Female sterilization is relied on by 22 percent of currently married women 15-44 which is equivalent to one third of all current users and hence is the most common contraceptive method currently practiced. Male sterilization is considerably less common with a prevalence level of 5 percent. Together, a total of 28 percent of married couples in which the wife is aged 15-44 are sterilized. The contraceptive pill, used by 20 percent of currently married women aged 15-44, is the second most common method while injectable contraceptives, used by 9 percent and the IUD, used by 7 percent, take a more distant third and fourth place. Condoms are used relatively rarely as the current method and use of vaginal methods is virtually nonexistent. Likewise, periodic abstinence and withdrawal are quite rare, Thus virtually all contraceptive use among married couples in Thailand is attributable to modern and potentially very efficient methods. Current contraceptive use is high both among younger and older currently married women although a curv/linear relationship between age and overall use is evident and a considerable difference in the choice of method according to age is apparent. The percentage of currently married women practicing contraception rises with age reaching a peak among women in their 30's and then declines. Even among the youngest and oldest age groups, however, current use is substantial. Considering specific methods, the IUD is the only major method that shows little association between use and age. Among women under 25, contraceptive use is overwhelmingly of modern temporary methods. Use of sterilization (male and female combined), however, is substantial among married women aged 25-29, representing about 30 percent of users in that age category. For age-groups 30-34 and beyond, sterilization accounts for the majority of users. 55 Table 4.5 Percentage of ~ who have ever used specific methods among ever-marrled ~ c-aTently married ~q~en, by current age Age Female &11e Weighted Any Any modern Vaginal sterili- ster£1i- Peri~d/c With- number of method method* Pi/l IUO Injecti~ methods Cc~km zation zation abst~ drawal Other wcrae~ 15-19 64.0 62.8 20-24 77.9 76.7 25-29 87.0 86.0 30-34 88.6 87.5 35-39 95.8 64.9 40-44 81.0 79.9 45-49 67.1 65.0 ~ 81.5 80.3 15-19 63.4 62.2 20-24 79.1 78.0 25-29 88.1 87.1 30-34 89.7 88.6 35-39 87.7 86.7 40-44 84.3 83.1 45-49 73.1 71.0 &11 kOe.S 8.1.6 82.4 Zver-mrrled I 48.4 7.4 20.4 1.1 5.7 0.4 0.1 3.9 3.6 0.4 342 61.9 13.8 31.3 0.6 12.2 4.4 1.0 4.6 6.1 0.5 1,004 64.6 18.0 36.3 0.3 18.2 16.8 3.1 7.1 7.9 0.4 1,309 58.8 16.0 31,9 0.5 15.1 32.6 7.2 7.0 5.9 0.4 1,328 57.9 16.6 28.8 0.1 14.2 30.7 8.8 5.9 5.0 0.6 1,110 49.5 16.5 16.3 0.2 9.0 31.6 9.8 5.0 4.4 0.5 877 36.4 11.9 11.4 1.3 3.7 24.3 6.9 3.1 3.1 0.6 8~5 55.8 15.3 27.1 0.5 12.5 22.3 5.7 5.6 5.5 0.5 6,'F/5 (:~rently mrriel ~,en 47.7 7.4 20.1 1.1 5.9 0.4 0.1 4.0 3.7 0.5 334 62.6 14.4 32.1 0.5 12.7 4.4 1.1 4.7 6.4 0.5 957 65.2 18.6 36.7 0.3 18.1 17.2 3.3 7.0 7.8 0.4 1,243 59.1 16.5 32.6 0.5 15.5 33.3 7.7 7.1 6.2 0.5 1,250 88.5 17.3 29.8 0.i 14.7 31.9 9.2 6.3 5.2 0.6 1,019 51.8 17.4 16.3 0.2 8.3 32.8 10.4 5.0 4.2 0.6 758 39.9 12.8 12.5 1.5 4.1 26.3 8.1 3.0 3.4 0.6 676 57.2 15.9 28.1 0.5 12.8 22.9 6.0 5.7 5.7 0.5 6,2.t6 Table 4.6 Percent distr£butlcn of ctrr~tly married ~ according to contraceptlve method c~-rantly t~sed, by cm~rent ~e Weighted CRrre~tly Female Hale number tsi~ any Vaginal sterili- sterili- Perlod/c With- Hot Total of ~e method Pill IL~ I~ectlon methods Coadk]m zatian zatlon Not-plant abstin~ce drawal Other using percent 15-19 43.0 24.7 7.0 7.0 0.0 1.2 0.4 0.i 0.0 1.9 0.7 0.0 57.0 100 334 20-24 56.8 27.6 8.0 13.7 0.0 1.0 4.4 1.0 0.1 0.4 0.6 0.0 43.2 100 %7 25-29 69.1 25.2 8.8 11.3 0.0 1.2 17.2 3.3 0.1 1.2 0.9 0.1 30.9 100 1,243 30-34 75.0 16.7 5.7 8.5 0.i I.i 33.3 7.3 0.0 1.2 1.2 0.0 ~5.0 100 1,250 35-39 73.3 16.0 5.7 7.7 0.0 1.5 31.9 8.4 0.0 1.1 0.8 0.2 26.7 100 1,019 40-44 69.4 10.9 8.0 4.6 0.0 1.2 32.7 10.2 0.1 0.5 1.0 0.2 30.6 1GO 788 45-49 48.4 6.8 4.3 2.3 0.0 0.0 26.3 7.6 0.0 0.3 0.7 0.0 51.6 100 676 15-4.9 b3.5 18.6 6.9 8.5 0.0 1.1 22.8 5.7 0.0 0.9 0.9 0.1 34.5 100 6,234S 15-44 67.$ 20.0 7.2 9.2 0.0 1.2 22.4 5.5 0.0 1.0 0.9 0.1 32.5 100 5,561 56 The striking increase in contraceptive prevalence over the last two decades in Thailand is documented in Table 4.7, which summarizes the results from a series of more or less equivalent national surveys. The dominance of female sterilization as the most common method was evident in the first survey when overall prevalence was low but did not reemerge again as the most common method until 1984. Compared to the 1984 CPS3, there has been a slight decline in female sterilization and a slight increase in male sterilization. Pill use has remained virtually constant. The largest increases are in use of the IUD and injection. Given sampling error and differences in the sample design between the TDHS and CPS3, the small changes evident should be regarded w~th appropriate caution. Table 4.7 Perc~tage currently practicing specific methods of contraception among currently married women aged 15-44, 1969-87 Year Sterilization Survey Fill IUD Male Female Injection Condos Others All methods* 1969/70 I~I 3.8 2.2 2.1 5.5 0.4 0.0 0.7 14.8 1972/73 LS2 10.6 4.7 2.8 6.8 0.9 0.I 0.5 26.4 1975 S(~T 15.2 6.5 2.2 7.5 2.1 0.5 2.8 36.7 1978/79 CPSI** 21.9 4.0 3.5 13.0 4.7 2.2 4.2 53.4 1981 CPS2 20.2 4.2 4.2 18.7 7.1 1.9 2.7 59.0 1984 CPS3 19.8 4.9 4.4 23.5 7.6 1.8 2.6 64.6 1987 TIRS 20.0 7.2 5.5 22.4 9.2 1.2 2.0 67.5 Not~: ~ource: IZ1 and IZ2 refer to ro~Ms 1 and 2 respectively of the National 5oagit,M~nal Study of Social, Eco~c and Demographic Change; SOFT refers to the Survey of Fertility in Thailand; and C?SI, CPS2, and CPS3 refer respectively to the first, second and third Contraceptive Prevalence Surveys. Results for LSI and LS2 are derived by combining separate rural and urban surveys taken one year apart and weighing the results to reflect the different sa~ling fractiors used. Rounding errors, minor coding discrepancies, and users of unspecified methods accoent for the mmll differences between the s~ of the percentages practicing individual methods and the percentage for all methods. Drl~l i~ wovincial urban. Knedel, Chamratrithirong and Debavalya, 1987 (except for TDHS). 57 Contraceptive practice according to selected background characteristics is examined in Table 4.8 based on currently married women aged 15-44. (Parallel results referring to currently married women in the 15-49 age range are presented in appendix table 4A.I to permit comparison with results from other countries participating in the international DHS project.) The association between number of living children and contraceptive practice is curvilinear. Prevalence is highest among couples with 3 children compared to those with either more or less. The lower percentage practicing among couples with 4 or more children compared to those with 3 probably reflects a selection process whereby couples who do not practice contraception are more likely to reach higher family sizes than those who do practice. In addition, higher parity women are likely to be older and higher proportions may be at ages where they no longer perceive a need for contraception. Permanent methods are relatively rare among women with 0-1 children but quite common among women with 2 or more children. There is almost no difference in the prevalence rate between rural and urban women and only minor differences in the mix of methods practiced. Sterilization is somewhat higher among urban women, perhaps reflecting the easier availability of the method in urban areas where hospitals and medical personnel are disproportionately concentrated. Likewise differences in contraceptive practice according to educational attainment are quite modest. Except for women with no education, for whom prevalence is somewhat lower than for the remainder, there is no clear association with educational level. Regional differences in the contraceptive prevalence rate are apparent. The south is clearly characterized by the lowest prevalence level while only modest differences are evident among the remaining regions including Bangkok. Contraceptive practice in the north is extremely high with 75 percent of married women aged 15-44 currently practicing some method. In comparison with results on contraceptive prevalence measured in CPS3 in 1984, the largest regional increase is evident for the northeast where prevalence rose from 61 to 67 percent. The level in both the north and central regions increased by about three percentage points, while in the south the increase amounted to only one percentage point. Finally, Bangkok actually shows a decline in contraceptive prevalence from 72 to 67 percent. Some regional differences are also evident in the method mix practiced. The north is notable for the high prevalence of contraceptive injectables which were popularized there before other regions through the private program of McCormick Hospital. Pill use is also unusually high in the north. The south stands out with respect to the practice of withdrawal which, although at a low absolute level even in the south, is almost totally absent elsewhere. Its use is associated with the large representation of Moslems in south. Indeed, religious differences in contraceptive prevalence are quite pronounced with Moslems characterized by only half the overall rate experienced by Buddhists. With respect to the practice of specific methods, the Moslems exceed the Buddhists only in the practice of withdrawal. It is also notable the prevalence of female sterilization, the most common method nationally, is only one third as high among Moslems as among Buddhists. The much lower contraceptive prevalence among Moslems, their more frequent practice of withdrawal, and the relative avoidance of sterilization are all consistent with previous findings from CPS3 (Kamnuansilpa and Chamratrithirong, 1985). 58 Table 4.8 Percent distribution of currently married wunam aged 15-44 according to the contraceptive method enrramtly used, by selected hac~ characteristics Currently Weighted using Female Male Not n~ber Backgronnd any Vaginal sterili- sterili- Periodic With- currently Total of characteristic method Fill lid Injection methods Condun zation zation Norplant abstinence drawal Other asing percent ~m~en to N~oer of liWa~ ~Mr~ 0 24.8 2O.3 0.0 1.2 0.0 1.0 0.1 0.6 0.0 1.0 0.4 0.2 75.2 100 660 1 57.9 27.3 9.9 13.7 0.I 1.4 1.5 1.5 0.I 1.3 i.i 0.i 42.1 10(3 1,262 2 78.9 21.4 8.6 10.9 0.0 1.4 27.3 7.1 0.0 1.2 1.0 0.0 2/.1 100 1,592 3 84.0 16.4 6.1 9.1 0.0 0.9 41.6 8.1 0.0 0.7 0.9 0.i 16.0 I00 968 4+ 73.3 12.7 7.2 6.6 O.0 1.0 36.1 8.3 0.0 0.5 0.7 0.i 26.7 I00 1,080 t%~an-rural r ~ Urban 68.5 20.3 4.1 6.6 0.i 2.5 25.8 5.7 0.2 2.0 i.I 0.i 31.5 10(3 1,029 Rural 67.3 20.0 7.9 9.8 0.0 0.9 21.6 5.4 0.0 0.7 0.8 0.I 32.7 I00 4,532 North 74.7 27.9 3.4 16.3 0.0 0.7 19.O 6.0 0.0 0.8 0.3 0.I 25.3 I00 1,161 Northeast 66.5 16.5 13.8 6.5 0.0 0.7 25.3 2.6 0.0 0.7 0.3 0.0 33.5 I00 1,943 Central 71.4 21.4 2.7 i0.0 0.I 1.5 25.6 9.0 0.0 0.7 0.4 0.i 28.6 i00 1,165 South 51.8 12.2 4.9 6.8 0.0 2.1 14.1 5.4 0.i 1.6 4.5 0.2 48.2 i00 680 Bangkok 67.4 22.5 4.2 5.6 0.I 2.1 22.8 7.0 O.i 2.1 0.7 0.i 32.6 10(3 611 r~cat~m NO education 59.4 14.7 4.9 9.6 0.0 1.3 19.4 8.8 0.0 0.7 0.0 0.0 40.6 i00 445 Primary 68.6 21.0 7.5 9.8 0.0 0.8 23.0 5.2 O.0 0.4 0.8 0.i 31.4 I00 4,409 Secondary 66.0 19.4 6.0 7.0 0.0 3.6 20.1 5.1 0.3 3.6 1.0 O.0 34.0 100 454 Higher 65.7 14.1 6.9 3.6 0.0 3.9 21.6 6.0 0.2 6.3 2.9 0.3 34.3 1(30 253 Rel/gicQ* Buddhist 69.7 21.0 7.6 9.2 0.0 1.2 23.4 5.4 0.0 0.9 0.8 0.I 30.3 I00 5,154 Islam 35.1 i0.0 1.5 7.7 0.0 0.8 8.2 3.0 0.0 0.7 3.0 0.1 64.9 100 292 Total 67.5 20.0 7.2 9.2 0.0 1.2 22.4 5.5 0.0 1.0 0.9 0.1 32.4 100 5,56J *Excludes cases whos~ religion is other than Buddkism or Islam or is not stated The timing of first contraceptive use relative to the number of living children is of interest when studying the spread of birth control as it can be indicative of when contraception is initiated during the family building process. Results in Table 4.9 show the percent distribution of ever-married women of different age cohorts according to the number of living children at the time of first use and are indicative of the increasing use of contraception for spacing purposes as adoption of birth control became widely accepted over the last two decades. Since the vast majority of Thai women want at least two children (see Chapter 5}, those who use contraception before having two children are almost certainly doing so for spacing purposes. The percent of women who had no child when first using contraception shows a strong and consistent negative correlation with age. Among ever-married women age 15-19, 43 percent first used when they had no children compared to only 1 percent of women aged 45-49. An additional 20 percent of women 15-19 started to use when they had only one child. Likewise among women in their twenties, well over half used contraception when they had no child or only one child. In contrast among women in their forties, only a relatively small percentage used contraception when they had less than two or even three children. This pattern is indicative that a shift has taken place from an initial pattern in which contraceptive use was primarily for the purpose of limiting family size to one in which family planning in the fuller sense of both spacing and limiting took hold. Table 4.9 Percent distribution of ever-married women according to number of living children at time of first use of contraception, by current age Weighted Current Never Total number age used 0 I 2 3 4+ Missing percent of women 15-19 36.0 43.0 19.6 1.4 0.0 0.0 0.0 i00 342 20-24 22.1 39.8 28.6 8.2 i.i 0.i 0.1 I00 1,004 25-29 13.0 29.0 34.2 16.3 5.6 1.9 0.0 i00 1,309 30-34 11.4 15.0 34.3 22.0 10.7 6.6 0.0 i00 1,328 35-39 14.2 8.5 22.4 21.4 14.4 19.0 0.i I00 1,110 40-44 19.0 4.2 12.8 13.7 18.0 32.2 0.i i00 877 45-49 32.9 1.3 4.8 8.9 13.4 38.6 0.0 100 805 All ages 18.5 18.7 24.4 15.1 9.6 13.6 0.0 I00 6,775 4.3 Knowledge of the Fer t i le Period In an attempt to ascertain whether Thai women have sufficient knowledge of reproductive physiology for the successful practice of periodic abstinence, respondents were asked when during the monthly cycle is a woman at greatest risk of becoming pregnant. Results are presented in Table 4.10 for all ever-married women and for the small subgroup who said they had ever practiced periodic abstinence. Perhaps because knowledge of the fertile period is limited among the Thai population, it was difficult to phrase this question in Thai in a way that appeared to make sense to most respondents. In addition, it is more 60 common in Tha i land to th ink in terms of the sa fe per iod ra ther than the per iod of r i sk . Th is c lear ly was a problem for some women when answer ing the quest ion as posed . As a resu l t , the quest ion o f ten appeared to be misunders tood and i t was d i f f i cu l t to c lear ly d i s t ingu ish someone who d id not th ink they knew the answer from those who e i ther gave a wrong answer or an answer that was d i f f i cu l t to in terpret in terms of the quest ion . Hence cons iderab le caut ion i s appropr ia te in in terpret ing the resu l t s . The results suggest that an accurate knowledge of the fertile period is very limited in Thailand and indeed the whole concept of changing probabilities of conception during the monthly cycle is probably largely unfamiliar. Only 13 percent of ever-married women responded correctly, i.e. gave an answer that could be clearly interpreted as indicating the most fertile period is in the middle of the monthly cycle. The large majority either appeared to not know or gave an answer that did not fit the standard precoded categories. Even among those who claimed to have practiced periodic abstinence, only 39 percent responded correctly. It is possible, however, that these results underestimate the true level of knowledge because the question may have been misinterpreted by the respondent. On the other hand, since some women may have guessed and given the right answer by chance, the results could also overestimate the extent of correct knowledge of the fertile period. Tab le 4 .10 Percent d i s t r ibut ion of ever -marr ied women aged 15-49 and women who have ever used per iod ic abst inence accord ing to knowledge of the fe r t i l e per iod dur ing the ovu la tory cyc le Ever users Ever -marr ied of per iod ic Fer t i le per iod women abst inence During menstrual period Right after period has ended In the middle of the cycle Just before period begins At any time Other Don't know* Total percent Weighted number of wonen 0.9 1.0 14.9 12.8 12.8 39.0 3.5 6.4 0.8 1.1 12.3 25.4 54.7 14.4 100 100 6,775 380 * Includes a small number of cases for whom no answer was recorded 61 4.4 Timing of Ster i l i zat ion Given the impor tance of female s ter i l i za t ion as a cont racept ive method in Tha i land , i t i s of in teres t to know the t rend in the adopt ion of the method and in determin ing whether the age a t the t ime of the operat ion i s dec l in ing . In fo rmat ion on the age a t t ime of the operat ion among s ter i l i zed women was co l lec ted in the TDHS and can serve as the bas is fo r such an ana lys i s . In o rder to use these data fo r th i s purpose , however , the prob lem of censor ing must be taken in to account . S ince the e l ig ib le woman sample , fo r whom these data are ava i lab le , exc ludes women above age 49, there i s a decreas ing age l im i t a t which women in the sample can repor t be ing s ter i l i zed the fu r ther back in t ime the operat ion took p lace . For example, the o ldest a women in the sample cou ld be a t the t ime of be ing s ter i l i zed i f the operat ion took p lace 10 years pr io r to the survey would be 39. Women who were s ter i l i zed 10 or more years pr io r to the survey and were aged 40 or over a t that t ime would have been exc luded from the TDHS e l ig ib le women sample because they would have been over 49 years o ld a t the t ime of the survey . Table 4.11 indicates the percent distribution of sterilized women according to the age at the time of sterilization. Results are shown according to the number of years since the operation. These distributions are influenced by the censoring problem referred to above, In order to obtain a summary measure of the age at sterilization that is unbiased by censoring, the median age at sterilization is calculated for women who were under 40 at the time of sterilization and who had the operation within 9 years prior to the survey. These results show very little change in the average age at steril ization over this period of time. Table 4.11 Percant distr ibut ion of ster i l ized wn,~n according to age at the time of ster i l i zat ion, by the number of years since the operation Aqe at the tree of sterilization Weighted Years since Total nm~er operation <25 25-29 30-34 35-39 40-44 45-49 perce~t of ~ Median* <2 17.7 36 .6 25 .9 12.7 4.9 2.2 100 276 28.7 2 - 3 14.2 36 .8 34 .7 10.8 3.6 0.0 i00 226 29.8 4 - 5 22.1 39.3 21 .1 14.0 3.5 0.0 i00 247 28.6 6 - 7 25.8 30.5 24.5 15.9 3.3 -** i00 209 28.7 8 - 9 16.5 37 .7 28.0 17.0 0.7 -** i00 176 29.4 10+ 20.9 38 .0 36.2 4.9 -** -** i00 375 - *** Total 19.7 36.7 29.0 11.6 2.6 0.4 I00 1,510 - **~ * Based on ~ ster i l ized prior to age 40 in order to avoid effect of ** Completely cansored *** Not shown ~ause influenced by censcr~ 62 Results in Table 4.12 attempt to determine trends in the age at sterilization over a somewhat longer period of time. Note that censoring has an increasing impact as the period under consideration extends further back in time. Thus progressively lower ages are used as cut off points when calculating the median ages as comparisons are made over longer time periods. The results suggest that there was little trend in the age at sterilization over the last 20 years. For example, among women sterilized before age 30, the median age at sterilization was practically identical for those sterilized 15-19 years prior to the survey and those sterilized less than 5 years prior to the survey. Likewise little change is evident over the 15 year period preceding the survey in the median age at sterilization among women sterilized before age 35. Table 4.12 Median age at sterilization for women sterilized before selected ages, by the number of years since the operation Median for women sterilized before age Years since . operation 45 40 35 30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0-4 years 29.4 29.1 28.5 26.5 5-9 years - 29.1 27.2 25.9 10-14 years - - 28.9 25.9 15-19 years - - 26.3 Total 29.2 29.0 28.2 26.2 4.5 Source of Contraception The source of cont racept ive supply or serv ice i s examined in Table 4.13 fo r spec i f i c methods based on a l l cur rent users . Those methods for which supp ly or serv ice i s unnecessary are omi t ted . Sources have been categor i zed to the extent poss ib le as to whether they belong to the government or to the pr ivate sector . Since both government and pr ivate agenc ies operate mobi le clinics, they could fit in either category and therefore are treated as indeterminate with respect to the government - private sector dichotomy. Also considered as indeterminate in this respect are respondents whose source is coded as "friends or relatives," "others," "and don't know." All together, only 3 percent of current users stated sources which are ambiguous with respect to belonging to either the government or private sector. The government sector is clearly the major provider of contraception in Thailand. Over four fifths of current users of a method requiring supply or service indicated that a government outlet provided them with their current method. Government hospitals, including MCH centers, are particularly important as a source for female sterilization and, together with health centers, provide 65 Table 4.13 Percent distr ibut ion of a l l current users of supply or c l in ic methods of contraception according to most recent source for supply, by method Source Pill Condom Suppiy methods Cl_inic methods Total Female Male Total Total supply sterili- sterili- clinic all Injection methods lid zation zation methods methods* Gowro~nt Nete~ Government hospital 9.2 12.4 21.4 13.0 65.7 85.6 55.2 77.1 49.3 Health center 53.7 30.8 60.5 54.9 25.2 1.6 9.6 7.3 28.0 Health volunteer 4.9 3.7 0.0 3.4 0.0 0.0 0.0 0.0 1.5 MCH or Bangkok health center 2.2 2.8 3.2 2.5 4.0 4.0 1.0 3.5 3.1 Private sector Family plannir~ clinic 0.5 2.1 0.2 0.5 0.i 0.i 6.3 1.0 0.8 Private hospital or clinic 4.6 3.6 11.5 6.7 3.2 8.0 10.7 7.5 7.1 Pharmacy 20.6 39.9 i.I 15.5 0.0 0.0 0.0 0.0 6.7 Shop 2.3 0.0 0.0 1.5 0.0 0.0 0.0 0.0 0.7 ~mle~te Mobile clinic 0.i I.i 1.5 0.6 1.7 0.3 14.5 2.8 1.8 Friend/relative 1.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.3 Other 0.8 0.8 0.6 0.8 0.0 0,0 0.0 0.0 0.3 Don't }mow** 0.0 2.9 0.0 0.i 0.0 0.4 2.7 0.7 0.5 Total Gove~t 70.0 49.7 85.1 73.8 94.9 91.2 65.8 87.9 81.9 Private 28.0 45.6 12.8 24.2 3.3 8.1 17.0 8.5 15.3 Indet erminant 1.9 4.8 2.1 2.1 1.7 0.7 17.2 3.5 2.9 pe~t 100 100 100 100 100 100 100 100 100 ~ i~ted ~ of wcmm 1,1"/0 67 530 1,767 435 1,511 359 2,306 4,075 * Total includes women who reported using vaginal methods (supply method) (clinic method). ** Includes women for whom no answer was recorded for source of current method or Morplant 64 vi r tua l ly all IUD's. Nevertheless, the pr ivate sector plays a substant ia l role in provid ing several methods, especia l ly the pi l l and the condom. Although the condom is re lat ive ly ins igni f icant in Thai land as a contracept ive method among marr ied couples, the pi l l is of considerable importance and the pr ivate sector, par t icu lar ly through drug stores, is the source of supply for over one fourth of ever -marr ied women who use pill. The share of all pil l users, including women who are single, may be ever higher but can not be determined from the TDHS given the restr ict ion of the sample to women who have ever married. Also of interest is the extent to which cl ients of var ious sources of contracept ive methods encounter problems when seeking services. Current and past users of contracept ion were asked if there was anything they dis l iked about the services they received the last t ime they received contracept ive suppl ies or services. Results are shown in Table 4.14 for major sources. Those who reported their last source as a pharmacy, shop, re lat ives or fr iends and, because of an error in the rout ing in the quest ionnaire, an MCH center or Bangkok Health Center were not asked about problems with services. In general, the vast major i ty of current and past users do report they encountered no problem. The most common problem reported was wait ing time and d iscourteous service in connect ion with government hospitals, but even these problems are reported by only a very small percentage of respondents. Table 4.14 Percent distrilmtion according to type of dissatiefactiun with the service (if any) ameng current users and past users who obtained a method at a source, by type of source last visited Did not Weighted Source No get method Total nm~er of supply problems Wait Dieccurteens Expensive desired Other percent of wcmun Current tMers Government hospital 88.2 3.2 5.7 0.4 0.I 2.4 i00 2,014 Government health center 94.2 0.9 2.6 0.7 0.5 1.2 100 1,147 Mobile clinic 92.6 0.0 0.0 1.8 0.0 5.6 100 75 Family planning clinic 95.8 4.2 0.0 0.0 0.O 0.0 100 33 Health volunteer 98.8 0.0 1.2 0.0 0.0 0.0 i00 60 Total 90.6 2.2 4.1 0.9 0.2 2.0 100 3,622 Government hospital Government health center Past Umers 86.9 7.5 5.6 0.0 0.0 0.0 100 68 96.1 1.3 1.5 0.6 0.0 0.4 100 164 Total* 93.6 2.8 2.4 0.6 0.3 0.3 100 259 * Includes a small n~mber of cases who obtained method from a mobile clin/c, a family planning clinic or a health volunteer but are not shown separately hecaase of their small n~her. 65 4.6 Reasons for Discont inuat ion Table 4.15 provides information on the main reason for discontinuatic9 among those wom~n who have discontinued a method within the last five years. For women who have discontinued more than one method, the last method that was discontinued is considered. Note that this table includes both women who are currently using as well as those who have not resumed contraception after discontinuing. The most common reason for discontinuing a method is to become pregnant. This is true both overall and for most methods shown. The only exception is injection for which the most common reason given for discontinuing is health concerns (including concerns about irregular bleeding). Health concerns were also mentioned frequently for the IUD. Former pill users cite both health concerns and allergic reactions (including headaches and nausea) relatively frequently. Periodic abstinence and withdrawal stand out because of the substantial proportion of former users citing method failure as a reason for discontinuation. ~.e 4.15 Percent alstribution of w~men who have discontinued a contraceptive method in the last 5 years according to the main reason for last discontinuation, by specific method Periodic Reason Pill IUD Injection Condom abstinence Withdrawal Total* To become pregnant 38.2 27.3 26.4 41.0 39.3 35.5 34.3 Method failed 6.0 11.9 1.9 9.3 33.4 34.8 7.8 Spouse disapproved 0.6 0.5 0.i 15.2 1.0 9.1 1.8 Health concerns 12.3 22.6 30.7 0.5 1.2 0.6 16.2 Access/availability/cost 2.6 0.0 7.7 3.1 0.0 0.0 3.5 Inconvenient to use 4.8 3.2 0.9 7.9 9.9 3.0 3.9 Infrequent s~x 4.4 0.0 4.1 2.0 4.3 9.5 3.9 Switch method 4.7 8.7 5.5 ii.0 8.1 2.7 5.8 Infecund 2.9 2.4 1.6 1.5 0.6 0.7 2.2 Divorced, separated 1.3 i.i 1.9 i.i 0.8 0.0 1.4 Allergic reaction 14.5 1.5 9.2 0.9 0.0 0.0 9.9 Other 7.5 19.7 9.4 6.1 1.3 1.9 8.7 Don't know** 0.4 1.0 0.7 0.5 0.0 2.2 0.6 I~'mnt 100 100 100 100 100 100 100 ~ted ~ of w i re 1,222 213 628 188 T/ 66 2,434 * Includes methods with insufficient cases to be shown separately. ** Includes won~n for whom no answer was recorded. 66 4.7 At t i tude toward Becoming Pregnant Table 4.16 shows the response of currently married women who were not pregnant, not abstaining from sex, and not using contraception when asked how they would feel if they were to become pregnant in the next few weeks. The results are presented according to the number of living children the respondent has. Overall, two fifths (40 percent) indicated that they would welcome a pregnancy. Almost one third (32 percent) indicated they would be unhappy to become pregnant, and the remainder (28 percent) said it would not matter one way or the other (28 percent). The proportion who would be happy to become pregnant is by far highest for women with no living children, among whom more than three fourths (77 percent) indicated a positive reaction to the prospect of a pregnancy in the near future and very few (only 6 percent) said they would be unhappy. The more children a woman has, the less likely she is to say that she would be happy to become pregnant and the more likely she is to indicate that she would be unhappy. Thus while women with one child are still far more likely to be happy than unhappy at the prospect of a pregnancy in the near future, among those with two children, those who would be unhappy outnumber slightly those who would be happy. Among women with more than two children, the number who would welcome a pregnancy is far less than those who would be unhappy about becoming pregnant. The number who say it would not matter, and thus appear to be indifferent, is also substantial but except for being distinctly lower among women with no children, does not vary much with the number of living children. Table 4.16 Percent distribution of non-pregnant, non-abstaining, non-contracepting, currently married women according to attitude toward becoming pregnant in the next few weeks, by number of living children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weighted Would number Number of not No Total of living children Happy Unhappy matter answer* percent women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 77.1 6.4 16.1 0.4 I00 347 1 48.6 22.6 28.1 0.7 i00 390 2 32.5 35.8 30.4 1.3 i00 307 3 21.3 43.2 33.0 2.5 I00 172 4+ 18.0 50.0 31.3 0.8 i00 484 Total 40.0 31.6 27.5 1.0 100 1,698 * Includes a small number of women who are coded "don't know" 67 4.8 Persona l Reason fo r Non-use In a country such as Thailand where knowledge of contraceptive methods is practically universal and prevalence is quite high, it is of considerable interest to identify the reasons why the minority of women who are not practicing contraception hut who say they do not want to be pregnant are not using any method. Such information is of potential value to the National Family Planning Program for targeting publicity and special programs for the remaining non-users as they work toward a goal of providing family planning methods to all who have a need for them. To help determine the reasons why some women who appear to potentially need to use contraception are not using any method, those who were not using, were not abstaining from sexual intercourse, were not currently pregnant, and did not say they would be happy to become pregnant were asked their reason for not practicing contraception. Table 4.17 shows the responses according to the woman's age. The results are additionally restricted to women who are currently married and specifically said they would be unhappy to become pregnant. The nature of the reason for non-use among this selected group of women differs among those who are less than 30 and those who are 30 or over. The most common reason stated overall for non-use is that the respondent considers herself to be menopausal or suhfecund (thus not truly at risk of becoming pregnant). Overall over a third (34 percent) of women included in the tabulation gave this as a reason. However, this reason is limited almost entirely to older women. Almost half (47 percent) of women 30 or over give this as their reason for not using compared to only 2 percent of women under 30. The second most common reason overall is that the respondent reported herself to be amenorrheic or to be hreastfeeding. Presumably these women do not feel they are currently at risk of pregnancy. Overall, 16 percent of the selected women cite this as the reason for non-use. However, this is largely a result of responses from women under 30, for whom this is by far the most common reason. Two other reasons are also relatively common: 14 percent indicate health concerns and II percent indicate infrequent sex. The remaining reasons are all relatively unimportant in terms of accounting individually for a significant number of women not practicing family planning. The fact that 2 percent of the respondents said they are not using because they wished to become pregnant even though the tabulation is restricted to women who replied to an earlier question that they would not be happy to become pregnant in the next few weeks serves as a reminder that questions are not always fully understood by respondents in the way intended by the researchers. One possible reason for this apparent inconsistency is that some women may wish to have a child but not look forward to the period of pregnancy. Hence they say they would be unhappy about becoming pregnant even though they still wish to become pregnant. Among the four most common reasons stated for non-use, health concerns would appears to be most relevant for the National Family Planning Program to address. Most women stating that they are menopausal or subfecund, who have infrequent sex, or who are amenorrheic or breastfeeding are undoubted at a substantially reduced risk of pregnancy, although not necessarily a totally negligible one. Combined, these categories associated with relatively low risks of conception account for 61 percent of non-use among the selected women. Efforts could be made to inform women who are amenorrheic or breastfeeding or 68 Table 4.17 Percent distribution of non-pregnant, non-abstain- ing, non-contracepting currently married women who would be unhappy if they become pregnant, according to the main reason for non-use, by current age Current age Reason <30 30+ All ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Seeks pregnancy Lack of knowledge Opposed to family planning Spouse or others disapproves Infrequent sex Postpartum/breastfeeding Menopausal/subfecund Health concerns Access/availability Costs too much Religion Inconvenient to use Other Don't know 4.9 1.0 2.1 0.8 2.4 1.9 2.9 4.3 3.9 0.7 0.8 0.7 17.0 8.3 10.8 38.8 7.2 16.1 2.2 47.0 34.4 11.9 15.2 14.3 0.4 0.8 0.7 0.5 0.3 0.4 2.7 2.7 2.7 1.0 1.2 i.I 14.5 6.5 8.7 1.7 2.2 2.1 Tota l percent Weighted number o f women 100 100 100 150 386 536 who are not totally abstaining that they are still at some risk. If the health concerns cited by respondents about contraceptive use are based on misinformation, however, a more important task would be for the program to disseminate information addressing those misperceptions. This would presumably increase use as a result. By and large, however, it appears that the vast majority of couples who are in need of family planning in Thailand are already practicing contraception, most of which is provided through the government's National Family Planning Program. 4.9 In tent ions fo r Future Use of Cont racept ion Intention to use contraception in the future provides a forecast of potential demand for services and acts as a convenient summary indicator of disposition towards contraception among current nonusers. The results should not be interpreted literally. The distinction between intended use in the next 12 months and later should he helpful in assessing the extent of demand in the near future. In the case of Thailand, where contraceptive prevalence is already quite high, those who are nonusers are a relatively selected group. As the results just presented indicate, nonusers include a substantial proportion who do not feel they are in need of contraception because they do not perceive themselves to be at risk of becoming pregnant. 69 Table 4.18 indicates the intentions concerning future use of contraception among currently married women aged 15-49 who are not currently using any method. Results are presented according to the number of living children. Overall, half of nonusers intend to use at sometime in the future while about half do not intend to use or are unsure. Of those intending to use, over half indicate they intend to do so in the next 12 months. A substantial share of the remaining women who intend to use are unsure about when they would start to use. Intention to use differs according to the number of living children. A substantial majority of women with 2 or fewer children intend to use contraception at sometime while almost three-fourths of women with four or more children do not intend to use. Quite likely many of the women with large numbers of living children are relatively close to the end of the reproductive ages and may perceive they have little need for future use because of low exposure to risk of pregnancy. Table 4.18 Percent distribution according to intentions to use in the future among currently married women not currently using any method, by number of living children (including any current pregnancy) - - - - - - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Number of living children Intention 0 1 2 3 4+ Total Use in next 12 months 24.6 37.7 35.2 23.2 14.7 27.2 Use later 24.6 16.7 10.3 4.5 2.8 12.9 Unsure about when 10.5 10.5 11.5 6.1 3.8 8.6 Unsure about use 8.6 6.6 6.4 I0.0 4.7 6.9 Does not intend to use 31.6 28.3 36.4 56.0 73.9 44.2 No answer 0.i 0.i 0.2 0.0 0.i 0.I Total percent I00 I00 I00 i00 100 I00 Yelghted number of women 504 554 374 192 530 2,153 Table 4.19 provides some indication of women's preferences for the method they might use in the future. This information should be interpreted with caution since there are two conditions implied: intention to use and method preferred if intention is followed. Overall, those intending to use express a preference for three particular methods about equally: the pill, injection, and female sterilization. For those intending to use in the in next 12 months, the pill and injection are preferred more than sterilization while the reverse is true for those who intend to postpone use for more than twelve months. Apparently substantial numbers of those who intend to use in the near future are planning to use for spacing purposes while those who are postponing use are 70 Table 4.19 Percent d i s t r ibut ion accord ing to pre fer red method among cur rent ly marr ied women not cur rent ly us ing a cont racept ive method but who in tend to use in the fu ture , by t iming of in tended use Use in next Unsure Method 12 months about timing Use later Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pill 29.0 23.9 22.4 26.3 IUD 6.6 5.6 5.7 6.2 Injection 33.2 27.0 19.4 28.4 Condom 1.7 1.0 0.3 1.2 Female sterilization 19.7 30.8 38.3 26.6 Male sterilization 2.1 4.2 8.8 4.3 Norplant 1.7 0.5 0.4 1.2 Periodic Abstinence 0.i 0.0 0.2 0.i Withdrawal 0.5 1.0 0.2 0.5 Other 0.3 0.0 0.0 0.2 Unsure 5.1 6.1 4.2 5.0 Tota l percent 100 Weighted number of women 587 I00 i00 I00 186 278 1,051 planning to use largely for limiting purposes. In general, the percent intending to use specific major methods among those who are unsure about when they will use, is intermediate between the percent indicated for women who intend to use in the next 12 months and those who expect to postpone use for at least 12 months. 4.10 Family P lanning Messages on the Radio The National Family Planning Program, composed of both government and private organizations, has been publicizing family planning over the radio for a number of years. The Family Health Division of the Ministry of Public Health has been regularly broadcasting half hour programs over radio stations in Bangkok and all provinces. Their programs consist of music or drama interspersed with spot announcements concerning contraception and family planning concepts. In addition, several private family planning agencies (such as PPAT and ASIN) have sponsored radio programs advocating family planning. Respondents were asked i f they had heard a message about fami ly p lann ing over the rad io dur ing the las t month and i f so, whether they had heard a message more than once. The resu l t s are presented in Table 4.20 accord ing

View the publication