Kenya - Demographic and Health Survey -1989

Publication date: 1989

Demographic and Health Survey 1989 National Council for Population and Development Ministry of Home Affairs and National Heritage ®DHS / Demographic and Health Surveys Institute for Resource Development/Macro Systems. Inc. Kenya Demographic and Health Survey 1989 National Council for Population and Development Ministry of Home Affairs and National Heritage Nairobi, Kenya Institute for Resource Development/Macro Systems, Inc. Columbia, Maryland USA October 1989 This report presents the findings of the Kenya Demographic and Health Survey (KDHS). The survey was a collaborative effort between the National Council for Population and Development and the Institute for Resource Development/Macro Systems, Inc. (IRD). The survey is part of the worldwide Demographic and Health Surveys Program, which is designed to collect data on fertility, family planning, and maternal and child health. Funding for the survey was provided by the U.S. Agency for International Development (Contract No. DPE-3023-C-00-4083-00) and the Government of Kenya. Additional information on the KDHS can be obtained from the Kenya National Council for Population and Development, Ministry of Home Affairs and National Heritage, P.O. Box 30478, Nairobi, Kenya. Additional information about the DHS Program can be obtained by writing to: DHS Program, 1RD/Macro Systems, Inc., 8850 Stanford Blvd., Suite 4000, Columbia, MD 21045, USA (Telephone: 301-290-2800; Telex: 87775; Fax: 301-290-2999). CONTENTS Page CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii L IST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v L IST OF F IGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii SUMMARY OF F INDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix MAP OF KENYA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxii 1. BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Geography, History and Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Populat ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Populat ion and Family Planning Policies and Programmes . . . . . . . . . . . . . 2 Heal th Priorit ies and Programmes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Object ives of the Kenya Demograph ic and Health Survey . . . . . . . . . . . . . 3 Survey Organisat ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Background Characterist ics of Women Respondents . . . . . . . . . . . . . . . . . 4 2. NUPT IAL ITY , BREASTFEEDING AND POSTPARTUM INSUSCEPT IB IL ITY . . . 9 2.1 2.2 2.3 2.4 2.5 Introduct ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Marital Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Polygyny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Age at First Marr iage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Breast feeding and Postpartum Insusceptibil ity . . . . . . . . . . . . . . . . . . . . 14 3. FERT IL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.1 3.2 3.3 3.4 3.5 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Levels and Trends in Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Fertil ity Dif ferentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Cumulat ive Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Age at First Birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4. FERT IL ITY REGULAT ION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.1 4.2 4.3 Contracept ive Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Acceptabi l i ty of Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Knowledge of Supply Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 iii Page 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 Ever Use of Contracept ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Current Contracept ive Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Current Use by Background Characterist ics . . . . . . . . . . . . . . . . . . . . . . 36 Number of Chi ldren at First Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Knowledge of Ferti le Per iod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Sources for Contracept ive Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Att i tude Toward Pregnancy and Reason for Nonuse . . . . . . . . . . . . . . . 42 Intent ion to Use in the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Approval of Family Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5. FERT IL ITY PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.1 5.2 5.3 Des i re for More Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Ideal Number of Chi ldren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Unwanted Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 6. MORTAL ITY AND HEALTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 6.1 6.2 6.3 6.4 Chi ldhood Mortal ity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Materni ty Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Child Health Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Household Sanitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 7. HUSBAND'S SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 Characterist ics of the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Marr iage and Fertil ity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Knowledge and Use of Family Planning . . . . . . . . . . . . . . . . . . . . . . . . 76 Sources for Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Intent ion to Use Family Phmning in the Future . . . . . . . . . . . . . . . . . . 80 Att i tudes Toward Family Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Des i re for More Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Ideal Number of Chi ldrcn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 APPENDIX A SURVEY DES IGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 APPENDIX B EST IMATES OF SAMPL ING ERROR . . . . . . . . . . . . . . . . . . . . . . 101 APPENDIX C NOTE ON AGE REPORTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 APPENDIX D L IST OF PERSONS INVOLVED IN THE KDHS . . . . . . . . . . . . . . . 117 APPENDIX E SURVEY QUEST IONNAIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 iv LIST OF TABLES Page Table 1.1 Table 1.2 Table 1.3 Table 1.4 Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 3.1 Table 3.2 Percent distribution of all women and currently married women by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . 6 Percent distribution of all women by background characteristics, 1977/78 Kenya Fertility Survey, 1984 Kenya Contraceptive Prevalence Survey, and 1989 KDHS . . . . . . . . . . . . . . . . . 7 Percent distribution of women by level of education, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . 8 Percentage of women who live in households with selected amenities, according to urban-rural rcsidcnce, Kenya, 1989 . . . . . . . . . . . . 8 Percent distribution of women by current marital status, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Percentage of womcn who have never married at the time of various surveys and ccnsuses, by age group, Kenya, 1989 . . . . . . . . . . 10 Percentage of currently married women in a polygynous union, by age, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Percent distribution of womcn by age at first marriage and median age at first marriage, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Median age at first marriage among women age 20-49 years, by current age and background characteristics, Kenya, 1989 . . . . . . . . . . 13 Percentage of births whose mothers are still breastfeeding, postpartum amenorrhocic, abstaining, and insusceptible, by number of months since birth, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . 14 Mean number of months of breastfeeding, postpartum amenorrhoea, postpartum abstinence, and postpartum insusceptibility, by background characteristics, Kenya, 1989 . . . . . . . . . . . 15 Age-specific fertility rates from various surveys and censuses, Kenya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Age-period fertility rate by age of woman at birth, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 V Page Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7 Table 3.8 Table 3.9 Table 3.10 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Age-specific fertility rates and total fertility rates for three periods before the survey, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 20 Percentage of all women who are currently pregnant by age, Kenya, 1977/78 KFS, 1984 KCPS and 1989 KDHS . . . . . . . . . . . . . . . . 21 Total fertility rates for calendar year periods and for five years preceding the survey, and mean number of children ever born to women 40-49 years of age, by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Percent distribution of fill women and currently married women by number of children ever born (CEB), according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Mean number of children ever born as reported in various surveys and censuses, by age group, Kenya . . . . . . . . . . . . . . . . . . . . . . 25 Mean number of children ever born to ever-married women, by age at first marriage and years since first marriage, Kenya 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Percent distribution of women by age at first birth, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . 26 Median age at first birth among women aged 20-49 years, by current age and background characteristics, Kenya, 1989 . . . . . . . . . . . . . 27 Percentage of all women and currently married women knowing contraceptive method and knowing a source by specific method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Percentage of currently married women knowing at least one modern method, knowing a source for a modern method, by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . 30 Percent distribution of women who have ever heard of a contraceptive method by main problem perceived in using the method, according to specific method, Kenya, 1989 . . . . . . . . . . . . . 32 Percent distribution of women knowing a contraceptive method by supply source they say they. would use, according to specific method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Percentage of all women and currently married women who have ever used a contraceptive method, by specific method and age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 vi Page Table 4.6 Table 4.7 Table 4.8 Table 4.9 Table 4.10 Table 4.11 Table 4.12 Table 4.13 Table 4.14 Table 4.15 Percent distribution of all women and currently married women, by contraceptive method currently being used, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Percent distribution of currently married women by contraceptive method currently being used, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 37 Percent distribution of ever-married women by number of living children at time of first use of contraception, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . 38 Percent distribution of all women and women who have ever used periodic abstinence by knowledge of the fertile period during the ovulatory cycle, Kenya, 1989 . . . . . . . . . . . . . . . . . . . 39 Percent distribution of current users of modcrn methods by most recent source of supply or information, according to specific method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Percent distribution of current users of modern methods of family planning, nonusers of modern methods, and all women knowing a method, by time to reach source of supply and transport to source, according to urban-rural residence, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Percent distribution of nonpregnant women who are sexually active and who are not using any contraceptive method by attitude toward becoming pregnant in the next few weeks, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . 42 Percent distribution of nonprcgnant women who are sexually active, not using any contraceptive method and who would be unhappy if they became pregnant by main reason for nonuse, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . 43 Percent distribution of currently married women who are not currently using any contraceptive method, by intention to use in the future, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Percent distribution of currently married women who are not using a contraceptive method but who intend to use in the future, by preferred method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . 44 vii Page Table 4.16 Table 4.17 Table 4.18 Table 4.19 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 6.1 Percent distribution of all women by whether they feel it is acceptable to have family planning information presented on the radio, by age and background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Percent distribution of currently married women knowing a contraceptive method by the husband's and wife's attitude toward the use of family planning, Kenya, 1989 . . . . . . . . . . . . . . . . . . . 45 Percentage of currently married women knowing a contraceptive method who approve of family planning and who say their husband approves of family planning by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 46 Percent distribution of currently married women knowing a contraceptive method by number of times discussed family planning with husband, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Percent distribution of currently married women by desire for children, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Percent distribution of currently married women by desire for childrcn, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . 49 Percentage of currently married women who want no more children (including those steriliscd) by number of living children and background charactcristics, Kenya, 1989 . . . . . . . . . . . . . . . 50 Percentage of currently married women who are in need of family planning by background characteristics, Kenya, 1989 . . . . . . . . . . . 51 Percent distribution of all women by ideal number of children and mean ideal number of children for all women and currently married women, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Mean ideal number of children for all women by age and background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 52 Percent distribution of women who had a birth in the last 12 months by fertility planning status according to birth order, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Infant and childhood mortality rates by five-year calendar periods, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 viii Page Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Table 6.8 Table 6.9 Table 6.10 Infant and childhood mortality rates by background characteristics of the mother for the ten-year period preceding the survey, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Infant and childhood mortality rates by selected demographic characteristics, for the ten-year period preceding the survey, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Mean number of children ever born, surviving and dead, and proportion of children dead among those born, by age of women, Kenya, 1989 . . . . . . . . . . " . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Percent distribution of births in the last 5 years by type of ante-natal care for the mother and percentage of births whose mother received a tetanus toxoid injection, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . 61 Percent distribution of births in the last 5 years by type of assistance during delivery, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Percentage of currently married women and births in the 12 months prior to the survey to women who fall in various categories of high health risk, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . 63 Among all children under 5 years o1" age, the percentage with health cards seen by interviewer, the percentage who are immunised as recorded on a health card or as reported by the mother and, among children with health cards, the percentage for whom BCG, DPT, polio and measles immunisations are recorded on the health card, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Among all children aged 12-23 months, the percentage with health cards seen by interviewer, the percentage who arc immunised as recorded on a health card or as reported by the mother and, among children with health cards, the percentage for whom BCG, DPT, polio and measles immunisations are recorded on the health card, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 65 Among children under 5 years of age, the percentage reported by the mother to have had diarrhoea in the past 24 hours and the past two weeks, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 ix Page Table 6.11 Tablc 6.12 Table 6.13 Table 6.14 Table 6.15 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Among children under 5 years of age who had diarrhoea in the past two weeks, the percentage receiving different treatments as reported by the mother, and the percentage not consulting a medical facility and not receiving treatment, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . 67 Among children under 5 years of age, the percentage who are reported by the mother as having had fever in the past four weeks, and, among children under 5 who had fever in the past four weeks, the percentage consulting a medical facility, according to background characteristics, Kenya, 1989 . . . . . . . . . . 68 Among children under 5 years of age, the percentage who are reported by the mother as having suffered from severe cough or difficult or rapid breathing in the past four weeks, and, among children under 5 who suffered from severe cough or difficult brcathing, the percentage consulting a medical facility, the percentage receiving various treatments, and the percentage not consulting a medical facility and not receiving treatment, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Percent distribution of all women by source of water for drinking, washing, and cooking, according to urban-rural residence and province, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Percent distribution of women by type of toilet facility in the household, according to urban-rural residence and province, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Percent distribution of husbands by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Percent distribution of husbands by level of education, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . 74 Percentage of husbands in polygynous union, according to background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 75 Percent distribution of husbands by number of current wives, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Percent distribution of married couples by number of years husband is older than his interviewed wife(ves), according to wife's age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Page Table 7.6 Table 7.7 Table 7.8 Table 7.9 Table 7.10 Table 7.11 Table 7.12 Table 7.13 Table 7.14 Table 7.15 Table 7.16 Percent distribution of husbands by number of living children, according to age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 76 Percentage of husbands who know contraceptive methods, who know a source for methods, who have ever used and who are currently using, by method, Kenya, 1989 . . . . . . . . . . . . . . . . . . 77 Percent distribution of married couples by knowledge of contraception, according to method, Kenya, 1989 . . . . . . . . . . . . . . . . . . 78 Percentage of husbands who are currently using any method and any modern method of contraceptkm, by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~¢9 Percent distribution of husbands who have ever heard of condom, male sterilisation, or withdrawal, by main problem perceived in using the method, according to specific method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Percent distribution of husbands knowing a contraceptive method by supply source they say they would use, according to specific method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Percent distribution of husbands by number of living children at time of first use of contraception, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Percent distribution of husbands who are not currently using any contraceptive method, by intention to use in the future, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . 82 Percent distribution of husbands who are not using a contraceptive method but who intend to use in the future, by preferred method, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Percentage of all husbands who believe it acceptable to have messages about family planning on the radio and percentage of husbands knowing a contraceptive method who approve of family planning, by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Percent distribution o f husbands knowing a contraceptive method by number of times discussed family planning with wife, according to current age, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . 84 xi Table 7.17 Table 7.18 Table 7.19 Table 7.20 Table 7.21 Table 7.22 Table 7.23 Table A.1 Table A.2 Table B.1 Table B.2 Table B.3 Table B.4 Table B.5 Table B.6 Table B.7 Page Percent distribution of married couples by wife's perception of husband's attitude toward family planning, according to husband's actual attitude, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . 84 Percent distribution of husbands by desire for children, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . 85 Percentage of husbands who want no more children by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . 85 Percent distribution of married couples by desire for more children, according to the number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Percent distribution of husbands by ideal number of children and mean ideal number of children, according to number of living children, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Percent distribution of married couples by whether husband's ideal number of children is less than, the same as, or higher than the wife's according to wife's ideal number, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Mean ideal number of children of husbands by background characteristics, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Sampling results for the whole country, Kenya, 1989 . . . . . . . . . . . . . . . . . Response rates for households, eligible women and eligible husbands, by urban-rural residence and province, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 List of selected variables with sampling errors, Kenya, 1989 . . . . . . . . . . 97 Sampling errors for the total population, Kenya, 1989 . . . . . . . . . . . . . 103 Sampling errors for the urban population, Kenya, 1989 . . . . . . . . . . . . 104 Samping errors for the rural population, Kenya, 1989 . . . . . . . . . . . . . . 105 Sampling errors for women in Nairobi, Kenya, 1989 . . . . . . . . . . . . . . . 106 Sampling errors for women in Central Province, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Sampling errors for women in Coast Province, Kenya, 1989 . . . . . . . . . 108 xii Table B.8 Table B.9 Table B.10 Table B.11 Table B.12 Table C.1 Table C.2 Page Sampling errors for women in Eastern Province, Kenya 1989 . . . . . . . . 108 Sampling errors for women in Nyanza Province, Kenya, 1989 . . . . . . . . 109 Sampling errors for women in Rift Valley, Kenya, 1989 . . . . . . . . . . . . 109 Sampling errors for women in Western Province, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Sampling errors for current contraceptive use among women by district, Kenya, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Percent distribution of the de facto population enumerated in various censuses and surveys by age group, according to sex, Kenya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Whipple's indices of age misreporting from various censuses and surveys, by sex, Kenya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 xiii Figure 2.1 Figure 2.2 Figure 2.3 Figure 3.1 Figure 3.2 Figure 3.3 Figure 4.1 Figure 4.2 Figure 4.3 Figure 5.1 Figure 5.2 Figure 6.1 Figure 6.2 Figure 7.1 Figure 7.2 Figure C.1 LIST OF FIGURES Page Marital Status of Women by Current Age . . . . . . . . . . . . . . . . . . . . . 10 Percent of Women Never Married by Age, KFS, KCPS and KDHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Duration of Breastfeeding, Amenorrhoea and Postpartum Abstinence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Age-Specific Fertility Rates, KFS, KCPS and KDHS . . . . . . . . . . . . . . 19 Total Fertility Rate (TFR) and Mean Number of Children Ever Born (CEB) to Women 40-49 by Residence and Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Total Fertility Rates by Province, KFS, KCPS and KDHS . . . . . . . . . . 24 Trends in Contraceptive Use Among Currently Married Women 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Current Contraceptive Use by Province, Currently Married Women 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Source of Family Planning Supply, Current Users of Modern Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Fertility Preferences, Currently Married Women 15-49 . . . . . . . . . . . . . 48 Fertility Preferences by Number of Living Children . . . . . . . . . . . . . . . 49 Trends in Infant and Child Mortality . . . . . . . . . . . . . . . . . . . . . . . . . 56 Infant Mortality by Province and Education . . . . . . . . . . . . . . . . . . . . 58 Family Planning Knowledge and Use Among Husbands . . . . . . . . . . . . 77 Mean Ideal Number of Children Among Husbands . . . . . . . . . . . . . . . 89 Percent Distribution of De Facto Household Population by Single Year of Age and Sex . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 XV FOREWORD The Kenya Demographic and Health Survey is a welcome addition to demographic and health data sources in Kenya. It provides us with a complete set of relevant data to evaluate population, health and family planning programmes and to assess the overall demographic situation in the country. Given the scope and representativeness, it can stand with census and intercensal survey data to provide the National Council for Population and Development, social scientists and other policymakers with a clear picture about Kenya's demographic trends in the recent past and likely directions for the future. The KDHS is an addition to previous surveys that have been conducted by the Central Bureau of Statistics and have utilised the CBS sample survey programme. Demographic surveys that have been conducted by CBS in the past include: the Kenya Fertility Survey (KFS) in 1977/78; the National Demographic Survey I (NDS I) in 1977; NDS II (1978); NDS llI (1983); and the Kenya Contraceptive Prevalence Survey (KCPS) (1984). The Kenya Demographic and Health Survey is the most complex survey to have been undertaken by NCPD. The KDHS findings provide the first evidence of a major decline in fertility and an increase in the use of family phmning. It further reveals that the infant mortality rate has dcclined between 1978 and 1989. I would like to acknowledge assistance by the United States Agency for International Development for financial support, IRD/DHS (Columbia, Maryland, USA) which provided technical assistance, the Central Bureau of Statistics, and the other members of the National Population Council who contributed to the success of the KDHS project. S. W. Ndirangu Director, National Council for Population and Development xvii SUMMARY OF FINDINGS The Kenya Demographic and Health Survey (KDHS) was conducted between December 1988 and May 1989 to collect data regarding fertility, family planning and maternal and child health. The survey covered 7,150 women aged 15-49 and a subsample of 1,116 husbands of these women, selected from a sample covering 95 percent of the population. The purpose of the survey was to provide planners and policymakers with data useful in making informed programme decisions. The survey data can also be used to evaluate Kenya's efforts to reduce fertility and the picture that emerges shows significant strides have been made toward this goal. KDHS data provide the first evidence of a major decline in fertility. If young women continue to have children at current rates, they will have an average of 6.7 births in their lifetime. This is down considerably from the average of 7.5 births for women now at the end of their childbearing years. The fertility rate in 1984 was estimated at 7.7 births per woman. A major cause of the decline in fertility is increased use of family pIanning. Twenty-seven percent of married women in Kenya are currcntly using a contraceptive method, compared to 17 percent in 1984. Although periodic abstinence continues to he the most common method (8 percent), of interest to programme planners is the fact that two-thirds of marricd women using contraception have chosen a modern method--either the pill (5 percent) or female sterilisation (5 percent). Contraccptive use varies by province, with those closest to Nairobi having the highest levels. Further evidence of the success in promoting family planning is the fact that more than 90 percent of married women know at least one modern method of contraception (and where to obtain it), and 45 percent have used a contraceptive method at some time in their life. The survey indicates a high level of knowledge, use and approval of family planning by husbands of interviewed women. Ninety-three percent of husbands know a modern method of family planning. Sixty-five percent of husbands have used a method at some time and almost 49 percent are currently using a method, half of which are modern methods. Husbands in Kenya are strongly supportive of family planning. Ninety-one percent of those surveyed approve of family planning use by couples, compared to 88 percent of married women. If couples are able to realise their childbearing preferences, fertility may continue to decline in the future. One half of married women say that they want no more children; another 26 percent want to wait at least two years before having another child. Husbands report similar views on limiting births--one-half say they want no more children. The desire to limit childbearing appears to be greater in Kenya than in other subSaharan countries. In Botswana and Zimbabwe, for example, only 33 percent of married women want no more children. Another indicator of possible future decline in fertility in Kenya is the decrease in ideal family size. According to the KDHS, the mean ideal family size declined from 5.8 in 1984 to 4.4 in 1989. The KDHS indicates that in the area of health, government programmes have been effective in providing health services for womcn and children. Eight in ten births benefit from ante-natal care from a doctor, nurse, or midwife and one-half of births are assisted at delivery by a doctor, nurse, or midwife. At least 44 percent of children 12-23 months of age are fully immunised against the major childhood diseases, Almost all children benefit from an extended period of breastfeeding. The average duration of breastfeeding is 19 months and the practice does not appear to be waning among either younger women or urban women. Another encouraging xix piece of information is the high level of ORT (oral rehydration therapy) use for treating childhood diarrhoea. Among children under five reported to have had an episode of diarrhoea in the two weeks before the survey, half were treated with a homemade solution and almost one-quarter were given a solution prepared from commercially prepared packets. The survey indicates several areas where there is room for improvement. Although young women are marrying later, many are still having births at young ages. More than 20 percent of teen-age girls have had at least one child and 7 percent were pregnant at the time of the survey. There is also evidence of an unmet need for family planning services. Of the births occurring in the 12 months before the survey, over half were either mistimed or unwanted; one fifth occurred less than 24 months after a previous birth. Hopefully, the data in this report will be useful to those making decisions regarding the future direction of health and family planning programmes. Total Fert i l i ty Rate KFS, KCPS and KDHS Women 15-49 Births per woman JI 7f~ 6.7 1977-78 KFS 1984 KCPS 1989 KDH8 Kenya DH8 1989 Current Contracept ive Use KFS, KCPS and KDHS Currently Marr ied Women 15-49 Percent 1977-78 1984 KFS KCP8 1989 KDHS Kenya DHS 1989 XX Kenya P JFTV~ UGAN ;0MALIA ) tr~CE W~STER I~Y& PROV ~CE F/Is .rAN 21 , NNROBI CE~ PROVIN~ 22 KIambu 26 K1dnyago 23 Muranga 17 Nyanda~o 2# Ny~ COAST pROVINCE 33 KJllfl 31 Kwale 55 Lomu 32 Mombcmo 30 TaRa 34. Tana E/~zcJ~N PROVINCE RIFT VALLEY PROVINCE 27 Embu 15 8adngo 37 lelolo 3 Flgeyo Momkwot 29 KlbJI 20 KaJlado 28 Mochakoa 14 Kedcho 3g Manmblt 16 Lolklpla 2.5 Meru 18 Nakuru 9 Nandl NORTH--~I~Z~N pROVINCE 19 Narok 36 God~a 38 Somburu 41 Monden3 4 Trane Nzo|o 40 Wo]Ir I Turkana 10 UasIn G18hu NYANZA PROVINCE 2 Wemt Pokot 13 Kla~ 11 KJsumu W~ilu~N pROVINCE 7 5[oya 5 Bun~oma 12 Sou~n Nyanza 6 Busla 8 Kakomega xxii 1 BACKGROUND 1.1 Geography, History and Economy Kenya is located in East Africa and lies between a longitude of 34 degrees and 42 degrees east and a latitude of 4 degrees north and 4 degrees south. It covers an area of approximately 582,646 sq. kilometres and is bordered by Ethiopia and Sudan to the north, Somalia and the Indian Ocean to the east, Uganda to the west and Tanzania to the south (see map). Kenya consists of eight areas called provinces. The next lower administrative units are districts, followed by divisions, locations, sublocations, and villages. Altogether, there are 41 districts in the country, in addition to Nairobi. The climate varies throughout the country and is determined by topography, altitude and precipitation. Most of the northern and eastern part of the country is semi-arid and less than one-third of the country is arable. Kenya achieved her independence in 1963 after a bitter and protracted struggle during which the indigenous people regained self determination and control of their destiny from the British colonial administration. Since then, the country has enjoyed political stability. In the past decade, the government began a new era with the establishment of the District Focus for Rural Development, which emphasizes the development of all parts of thc country. The Government has been committed to the provision of equal educational opportunities for all by providing free primary education. In 1985, the government introduced the 8-4-4 system of education (8 years of primary, 4 years of secondary and 4 years of university) that places emphasis on vocational and technical training at all levels. In addition to private universities, Kenya has established four public universities since independcnce. In Kenya, agriculture remains the leading sector in stimulating economic growth. The most important foreign exchange earners are coffee and tea in the agricultural sector and tourism in the non-agricultural sector. The country registered an impressive growth performance over the period 1964-71, with an average gross domestic product (GDP) of 6.5 percent. The oil crisis, that was caused by a steep rise in the price of crude oil, resulted in a drop in the GDP from an average of 6.7 percent in 1972 to 3.1 percent in 1975. In 1975, a severe frost which affected the coffee crop in Brazil led to an unexpected increase in the price of coffee and tea on the world market from 1976 to 1978. However, another oil crisis in 1979 dampened economic growth until 1983. 1.2 Population On the basis of census statistics, Kenya's population increased from 5.4 million in 1948 to 16.1 million in 1979 (Republic of Kenya, 1989). Estimates from the 1979 population census indicated that the population growth rate in Kenya was 3.8 percent per annum (Central Bureau of Statistics, no date). At this rate, the population is expected to increase to 35 million by the year 2000 (Central Bureau of Statistics, 1983). As a result of high fertility and declining mortality, Kenya is characterised by a young population. Almost 50 percent of Kenya's population is less than 15 years of age. The momentum generated by high fertility and declining mortality implies that the population growth rate will remain high for some time. The crude birth rate increased from 50 per thousand in 1948 to 52 per thousand in 1979, whereas the crude death rate decreased from 25 to 14 in the same period. The infant mortality rate decreased from 184 deaths per thousand births in 1948 to 104 in 1979 (Republic of Kenya, 1989). The 1984 Kenya Contraceptive Prevalence Survey (KCPS) showed some evidence of a possible decline in fertility, from a total fertility rate of 8.1 children per woman in 1977/78 (Kenya Fertility Survey) to 7.7 in 1984 (Central Bureau of Statistics, 1984). The population growth rate in the urban areas is more than 7 percent per year (Republic of Kenya, 1989). The population of the capital, Nairobi, has increased from 897,000 in 1980 to an estimated 1,429,000 in 1989. This increase can be attributed in large part to rural-urban migration. 1.3 Population and Family Planning Pofieies and Programmes The Government of Kenya became concerned about the high rate of population growth after the 1962 population census. During the early 1960s, the Family Planning Association of Kenya (FPAK) was established by private individuals, but it was not until 1967 that the official national family phmning programme was launched. Family planning was integrated into the Maternal and Child Health Division of the Ministry of Health. At first, due to lack of an effective health infrastructure and adequate skilled manpower, the Ministry of Health relied mainly on FPAK and expatriate staff for technical assistance. After the 1969 census provided evidence of a high level of fertility, the government decided to launch a five-year (1975-1979) family planning programme. The specific goals of the programme were to reduce the high annual rate of natural population increase from 3.3 percent (in 1975) to 3.0 percent (in 1979) and to improve the health of mothers and their children under the age of five, Initially, however, the family planning component of the Maternal and Child Health Programme had limited success. The 1979 census results indicated a population growth rate of 3.8 percent per annum, which was higher than the projected growth rate of 3.0 percent. This failure in achieving the population growth rate target could be attributed to shortfalls in the assumptions used to arrive at the target. The plan to reduce the growth rate concentrated on the supply side of family planning instead of putting emphasis on programmes aimed at changing family size norms. It was with the realisation of the need to improve on the earlier weaknesses of the family planning programme that the government of Kenya approved the establishment of the National Council for Population and Development (NCPD) in 1982. The Council's mandate is to formulate population policies and strategies and to co-ordinate the activities of government ministries, non- governmental organisations, and donors involved in population, integrated rural health, and family planning programmes. 1.4 Health Priorities and Programmes The 1989-1993 Kenya Development Plan emphasises the government's commitment to developments in the health sector that are geared toward the attainment of "Health for All by the Year 2000". The government encourages an integrated approach to the health system that involves such essential components as appropriate health education, provision of proper nutrition, basic sanitary facilities, and maternal and child health, including family planning and immunisation against major infectious diseases, among others. In 1981, the Ministry of Health started a major programme in preventive health, the Kenya Expanded Programme on Immunisation (KEPI). Several other government programmes aimed at the reduction of diseases, improvement of nutrition, and provision of maternal and child health services have also been launched. 1.5 Objectives of the Kenya Demographic and Health Survey On March 1, 1988, 'on behalf of the Government of Kenya, the National Council for Population and Development (NCPD) signed an agreement with the Institute for Resource Development (IRD) to carry out the Kenya Demographic and Health Survey (KDHS). The KDHS is intended to serve as a source of population and health data for policymakers and for the research community. In general, the objectives of the KDHS are to: assess the overall demographic situation in Kenya, assist in the evaluation of the population and health programmes in Kenya, advance survey methodology, and assist the NCPD strengthen and improve its technical skills to conduct demographic and health surveys. The KDHS was specifically designed to: provide data on the family planning and fertility behaviour of the Kcnyan population to enable the NCPD to evaluate and enhance the National Family Planning Programme, measure changes in fertility and contraceptive prevalence and at the same time study the factors which affect these changes, such as marriage patterns, urban/rural residence, availability of contraception, breastfeeding habits and other socioeconomic factors, and examine the basic indicators of maternal and child health in Kenya. 1.6 Survey Organisation The KDHS was a national survey that was carried out by NCPD in collaboration with the Central Bureau of Statistics (CBS) and the Institute for Resource Development (IRD). Funds for the survey came from three sources--the Government of Kenya, the United States Agency for International Development (USAID) office in Kenya, and IRD, through its contract with USAID/Washington. IRD also provided technical assistance throughout all stages of the survey. The sample for the KDHS is based on thc National Sample Survey and Ewduation Programme (NASSEP) master sample maintained by the CBS. The KDHS sample is national in coverage, with the exclusion of North Eastern Province and four northern districts which together account for only about five percent of Kenya's population. The KDHS sample was designed to produce completed interviews with 7,500 women aged 15-49 and with a subsample of 1,000 husbands of these women. The NASSEP master sample is a two-stage design, stratified by urban-rural residence, and within the rural stratum, by individual district. In the first stage, 1979 census enumeration areas (EAs) were selected with probability proportional to size. The selected EAs were segmented into the expected number of standard-sized clusters, one of which was selected at random to form the NASSEP cluster. The selected clusters were then mapped and listed by CBS field staff. In rural areas, household listings made betwecn 1984 and 1985 were used to select the KDHS households, while KDHS pretest staff were used to relist households in the selected urban clusters. Despite the emphasis on obtaining district-level data for phmning purposes, it was decided that reliable estimates could not be produced from the KDHS for all 32 districts in NASSEP, unless the sample were expanded to an unmanageable size. However, it was felt that reliable estimates of certain variables could be produced lbr the rural areas in the 13 districts that have been initially targeted by the NCPD: Kilifi, Machakos, Meru, Nyeri, Murang'a, Kirinyaga, Kericho, Uasin Gishu, South Nyanza, Kisii, Siaya, Kakamega, and Bungoma. Thus, all 24 rural clusters in the NASSEP were selected for inclusion in the KDHS sample in these 13 districts. About 450 rural households were selected in each of these districts, just over 1000 rural households in other districts, and about 3000 households in urban areas, for a total of almost 10,000 households. Sample weights were used to compensate for the unequal probability of selection between strata, and weighted figures are used throughout the remainder of this report. The KDHS utilised three questionnaires: one to list members of the selected households (household questionnaire); another to record information from all women aged 15-49 who were present in the selected households the night before the interview (woman's questionnaire); and the third to record information from the husbands of interviewed women in a subsample of households (husband's questionnaire). The questionnaires were pretested in August 1988. Copies of lhe final versions appear in Appendix E. The field staff for the KDHS consistcd of nine teams, each of which was fluent in one of the major indigenous languages. The teams were composed of four or five female interviewers, one editor, one supervisor, and a male interviewer. There was a smaller tenth team that had three interviewers for the Narok-Kajiado region. The teams were supervised by the local District Population Officer, the District Statistical Officer, or in some cases, an officer from NCPD headquarters. A more complete description of the survey design appears in Appendix A. Interviewers and data entry staff were recruited in October 1988 and trained in November 1988. The training included practice interviewing both in the classroom and in the field. Data collection began on 1 December and was completed during tile last week of May. The proportion of women interviewed by month was: December 1988 (7 percent); January 1989 (13 percent); February (14 percent); March (24 percent); April (25 percent); and May (17 percent). 1.7 Background Characterist ics of Women Respondents A total of 9,836 households were selected in the Kenya Demographic and Health Survey. Of these, 8,343 were identified as occupied households during the fieldwork and 8,173 were successfully interviewed. Respondents for the individual interview were women aged 15-49 who had spent the night before the interview in the selected household. In the interviewed households, 7,424 eligible women were identified and 7,150 were successfully interviewed. In general, few problems were encountered during the interviewing and the response rate was high--98 percent for households and 96 percent for individual female respondents. In addition, 1,116 husbands were interviewed out of a total of 1,397 eligible, for a response rate of 81 percent. Eligible husbands were defined as those who spent the night before the interview in the selected households and whose wives were successfully interviewed. Every other household was considered eligible for the husband interview. Details on nonresponse appear in Appendix A. This section of the report briefly examines the background characteristics of the female respondents. Knowledge of these characteristics provides a crude measure of the representativeness of KDHS data and facilitates the interpretation of other survey findings. Table 1.1 presents the distribution of all women and currently married women by selected background characteristics. Table 1.2 indicates that the distribution for all women generally fits the pattern established by the 1977/1978 KFS and 1984 KCPS (Central Bureau of Statistics, 1980; Central Bureau of Statistics, 1984). The proportion of the respondents in the 15-19 age group is slightly lower in the KDHS than in the 1977/1978 and 1984 surveys, and there has been a steady increase over time in the proportion living in urban areas to 17 percent in 1989. The distribution of all women by province indicates only minor differences among the three sources of data. For purposes of comparison, in Table 1.2 respondents are classified into 4 educational categories, according to the highest grade attained at each level. These categories are: no education, 1-4 years, 5-8 years, and 9 or more years. 1 The data show a strong increase in the educational attainment of women over time. The proportion of women with no education declined from 44 percent in 1977/78 to 25 percent in 1989. The proportion of women who have 5 to 8 years of education is higher in 1989 (43 percent) than in 1984 (32 percent) and 1977/78 (27 percent). Women interviewed in the survey were classified into five religious groups: Catholic, Protestant, Moslem, Other, and no religion. More than half of the interviewed women are Protestant. The distribution of the respondents according to religion has changed little over time. There are also inter-relationships between various background characteristics. Table 1.3 shows the distribution of the surveyed women by education, according to other selected characteristics. Nearly one-quarter of the women in the KDHS sample have never attended school, about 28 percent have some primary education only, 27 percent graduated from primary school with no further education, and 1 out of 5 women attended secondary school or higher education. Education is inversely related to age; that is, older women are generally less educated than younger women. For examplc, whereas only 5 percent of women aged 15-19 have had no formal education, more than 50 per cent of the women aged 40 and over have never been to school. 1 For the remainder of the report, respondents are classified into the categories: no education, some primary (Standard 1-6), primary complete (Standard 7 or 8), and secondary or higher (Form 1 and above). Although the introduction of the 8-4-4 system has changed the definition of primary complete, the new system came in 1986 and has not had a chance to affect respondents age 15 and above. Tab[e 1.1 Percent d i s t r ibut ion of a l l women and cur rent ly marr ied women by background characteristics, Kenya, 1989 AIL Women Currently Married Women Weighted Unwtd. Weighted Unwtd. Background Weighted no. of no. of Weighted no. of no. of characteristic percent women women percent women women Age 15-19 20.9 1497 1481 5,8 276 300 20-24 18.5 1321 3482 17,3 827 882 25-29 18,7 1334 1357 23.2 1104 1126 30-34 13.7 982 1007 17.5 833 853 35-39 12.6 898 830 16.4 781 720 40-44 9.4 674 646 12.1 576 544 45-49 6.2 445 427 7.7 369 353 NO. Living children 0-2 47.0 3364 3506 29.4 1400 1535 3-4 20,7 1477 1499 27.1 1291 1314 5 or more 32.3 2310 2145 43.5 2075 1929 Residence Urban 17,3 1236 1917 15.7 748 1160 Rural 82.7 5914 5233 84.3 4018 3618 Province Nairobi 7.7 554 859 7.0 335 519 Central 15.7 1120 1281 13.6 648 787 Coast 7.0 498 720 7.3 350 529 Eastern 17.8 1269 898 16,9 804 561 Nyanza 17.0 1218 1268 18.3 872 895 Rift Valley 21.2 1519 1100 22.0 1047 742 Western 13.6 971 1027 14.9 710 745 Education No educat ion 25.1 1797 1702 31.6 1506 1438 Some primary 27,7 1977 1888 30.7 1462 1394 Primary complete 26,7 1910 1938 20.7 987 1026 Secondary + 20.4 1457 1612 16.9 804 914 Missing 0,1 9 10 0.1 6 6 Religion Catholic 34.7 2480 2390 34.8 1656 1589 Protestant 57,4 4107 4075 56.8 2706 2670 Muslim 3.5 253 317 3,5 165 213 Other 1.6 115 104 1.7 79 77 No religion 2.6 184 254 3.2 151 222 Missing 0.2 12 18 0,2 9 7 Total 100.0 7150 7150 100.0 4765 4778 Womcn who reside in the urban areas have considerably more education than those living in the rural areas. In urban areas, the percentage of women who never attended school is lower than in rural areas and the percentage who have secondary or higher education is more than twice as high as in the rural areas. Looking at the data by province, Nairobi, the capital, has the smallest proportion of uneducated women (9 percent), compared to 47 percent in Coast Province, and 13 percent in Central Province. There is little difference among the othcr four provinccs in terms of the 6 Table 1.2 Percent d is t r ibut ion of all women by background characteristics, 1977/78 Kenya Fertility Survey, 1984 Kenya Contraceptive Prev- alence Survey, and 1989 KDHS Background 1977/78 1984 1989 characteristic KFS KCPS KDHS Age 15-19 23.8 25.9 20.9 20-24 17.9 19.6 18.5 25-29 18.4 15.8 18.7 30-34 12.5 12.7 13.7 35-39 11.5 10.5 12.6 40-44 7.7 8.5 9.4 45-49 8.0 7.0 6.2 Residence Urban 12.3 14.0 17.3 Rural 87.7 86.0 82.7 Province Nairobi 5.4 4.7 7.7 Central 15.2 16.3 15.7 Coast 8.4 10.1 7.0 Eastern 17.2 16.7 17.8 Nyanza 21.9 18.8 17.0 Rift Val ley 18.3 20.9 21.2 Western 13.4 12.5 13.6 Education No education 44.2 34.8 25.1 1-4 years 18.4 16.1 13.6 5-8 years 27.4 32.1 42.6 9+ years 9.8 16.8 18.3 Missing 0.2 0.2 0.3 Religion Catholic 36.2 36.5 34.7 Protestant 53.1 52.8 57.4 Muslim 4.8 3.7 3.5 Other 0.4 1.9 1.6 No religion 5.4 5.1 2.6 Missing 0.I 0.1 0.2 I o ta [ 100.0 100.0 100.0 proportion who are uneducated. The KDHS data show that educational achievement of women in Nairobi is highest among all the provinces, with 30 percent having completed primary school, and 43 percent having attained sccondary or higher education. Among the other provinces, Central Province shows the highest level of educational achievement. The proportions of women who have attended secondary or higher education in the other regions are very similar--16 percent in the Coast, 15 percent in Eastern, 17 percent in Nyanza, 16 percent in Rift Valley, and 20 percent in Western Province. Table 1.4 provides information regarding certain household amenities available to women. Overall, 10 percent of women live in households that have electricity and 61 percent have radios. Only 5 percent of women's households have televisions and 4 percent have refrigerators. As for means of transportation, 28 percent of women live in households in which some member owns a bicycle, 7 percent a car, and just over 1 percent either a motorcycle or tractor. Almost 90 percent 7 Table 1.3 Percent distribution of women by level of education, according to background characteristics, Kenya, 1989 Level of education Wtd. number Background Some Primary Second- Miss- of characteristfc None primary complete ary + ing Total women Age 15-19 4.7 23.5 50.4 21.4 0.I 100.0 1497 20-24 8.5 25.6 30,7 35.0 0.1 100.0 1321 25-29 18.2 30.3 23.2 28.1 0.2 100.0 1334 30-34 36.8 28.7 16.9 17.3 0.2 100.0 982 35-39 42.7 28.6 19.0 9.7 0.I 100.0 898 40-44 50.4 34.0 10.9 4.6 0.1 100.0 674 45-49 64.6 25.8 7.2 2.4 0.1 100.0 445 Residence Urban 12.3 18.6 27,6 41.4 0.1 100.0 1236 Rural 27.8 29.5 26.5 16.0 0.1 100.0 5914 Province Nairobi 8.5 18.2 30,2 43.0 0.2 100.0 554 Central 12.8 26.9 33.1 26.7 0.5 100.0 1120 Coast 47.5 15.5 21,5 15.5 0.0 I00.0 498 Eastern 23°6 32.7 28.3 15.3 0.1 100.0 1269 Nyanza 27.4 31.7 23.9 16.9 0.1 100.0 1210 Rift Valley 32.1 20.2 23.4 16.2 0.0 100.0 1519 Western 25.5 27.6 26.6 20.2 0.1 100.0 971 Total 25.1 27.7 26,7 20.4 0.1 100.0 7150 of women live in households with land, 76 percent live in households with cattle, sheep or goats, and 40 percent live in households where cash crops are grown. Only 35 percent live in permanent houses. It should be noted that interviewers usually relied on personal observation of the respondent's house and did not ask whether the house was permanent or not. Thus, the definition of what constitutes a "permanent" house may have varied by interviewer and/or by team. Ownership of household amenities varies tremendously by urban-rural residence. As expected, the proportion of urban women living in households with these amenities is higher than the proportion of rural women for all items except bicycles, land, animals, and cash crops. The urban-rural differential is particularly strong for electricity, televisions, and refrigerators. Table 1.4 Percent of women who live in households with selected amenities, according to urban-rural residence, Kenya, 1989 Household amenity Residence Urban Rural Total Electricity 45.2 2.8 10.1 Radio 77.6 58.0 61.4 Television 22.4 1.5 5.1 Refrigerator 16.4 0.9 3.6 Bicycle 24.8 28.7 28.0 MotorcycLe 3.1 1.1 1.5 Car 18.1 4.7 7.0 Tractor 1.8 1.2 1.3 Land 61.3 92.7 87.3 Cattle, sheep, goats 47.4 81.4 75.5 Cash crops 25.5 43.5 40.4 Permanent house 65.5 28.5 34.9 Number of Women 1236 5914 7150 2 NUPTIALITY, BREASTFEEDING AND POSTPARTUM INSUSCEPTIBILITY 2.1 Introduct ion Fertility levels and trends depend in part on the extent of and age at marriage among women. From past demographic surveys and censuses, Kenya has reliable indices on nuptiality. Such data show that age at first marriage for Kenyan women has been rising. As in other demographic surveys and censuses carried out in Kenya, marriage is defined in the KDHS to include informal unions. This chapter investigates the trends in age at marriage of different cohorts. Other proximate determinants of fertility that render a woman to be at risk of pregnancy, such as breastfeeding, postpartum amenorrhoea and postpartum sexual abstinence are also explored. 2.2 Mar i ta l Status In this report, the terms "living together" and "married" are combined and referred to as "currently married". Women who are currently married, widowed, divorced or no longer living together are referred to as "ever-married". Table 2.1 Percent distribution of women by current marital status, according to age, Kenya, 1989 Current marita[ status Wtd. Never Living no. mar- Mar- to- Widow- Oi- Separ- of Age ric~J ried gether c~d vorced ated Total worn 15-19 79.9 17.2 1.2 0.0 1.1 0.5 100.0 1497 20-24 31.8 58.6 4.0 0.7 2.7 2.3 100.0 1321 25-29 10.7 78.6 4.1 1.5 3.6 1.3 100.0 1334 30-34 5.4 79.6 5.3 2.1 5.4 2.2 100.0 982 35-39 3.2 02.4 4.5 4.7 3.8 1.4 I00.0 898 40-44 1.5 82.4 3.1 8.2 3.0 1.8 100.0 674 45-49 2.4 79.7 3.2 11.0 3.1 0.6 100.0 445 Total 26.0 63.1 3.6 2.7 3.1 1.5 I00.0 7150 Table 2.1 shows that 26 percent of women of childbearing age have never married, 67 percent are currently married and 7 percent are either widowed, divorced, or no longer living together (separated). The proportion never married falls sharply from 80 percent in the age group 15-19 to 11 percent in the age group 25-29 (see Figure 2.1). This proportion declines to 2 percent in the age group 45-49, reinlbrcing the findings of other demographic surveys--that most women in Kenya marry early. The proportion currently married rises steeply in the first two age groups reaching a high of 87 percent among women 35-39. As expected, widowhood increases steadily with age, varying from none for women aged 15-19 years to 11 percent for the age group 45-49. 9 The proport ion divorced or separated increases from 2 percent for women aged 15-19 years to 8 percent for women aged 30-34 years, and dcclines to 4 percent for women in their forties. 100% Figure 2.1 Marital Status of Women by Age 75% 50% 25% 0% 15-19 20-24 25-29 30-34 Age l Never in union Married I 35-39 40-44 45-49 Widowed/div./sep. / Kenya DH8 1989 Table 2.2 Percent of women who have never married at the time of various surveys and censuses, by age group, Kenya, 1989 1962 1969 1977 1977/78 1979 1984 1989 Age Census Census NDS KFS Census KCPS KDHS 15-19 55 64 71 72 71 74 80 20-24 13 18 22 21 25 24 32 25-29 5 6 6 4 9 6 11 30-34 3 4 3 I 5 4 5 35-39 2 3 2 I 3 2 3 40-44 2 3 I I 3 I 2 45-49 2 3 I 0 2 I 2 Source: Centrat Bureau of Statistics, 1984, Tabte 4.4 Table 2.2 shows the trend in the proportion of women reported as never married by age from past censuses and surveys in Kenya. It is evident that the proportion of women under 30 who have never married has been increasing (Figure 2.2). The KDHS data show an increase for every age group over the KCPS data. For example, the proportion of women 15-19 who have never married increased from 74 percent in 1984 to 80 percent in 1989 and the proportion in age group 20-24 rose from 24 percent in 1984 to 32 percent in 1989. There is also a notable increase in the proportion never married for women aged 25-29, from 6 in 1984 to 11 percent in 1989. These observations suggest that age at first marriage in Kenya is increasing. Above age 25, the proportions of women remaining single are too small to discern any trend over time. 10 As has been observed in other surveys, increased enrolment of women in higher education may be the major cause of the increasing proportions of single women aged 15-24 years. Figure 2.2 Percent of Women Never Married by Age KFS, KCPS and KDHS Percent , o is-19 2o-24 2s-29 Age l mm KFS, 1977-78 ~ KCPS, 1984 ~ KDHS, 1989 Kenya DHS 1989 2.3 Polygyny In order to measure the extent of polygyny in Kenya, married women in the KDHS were were asked if their husbands had other wives. Table 2.3 displays the answers to this question by age of the woman. Overall, 23 percent of currently married women arc in polygynous unions. This is a slight decline from the 25 percent in the 1984 KCPS and the 30 percent in the 1977/78 KFS. The table indicates that polygyny is more common among older than younger women, which may reflect a trend away from this traditional practice. Polygyny is more common in the rural areas than in the urban areas. This is true for women in most age groups. Nyanza Province has the highest proportion of women in polygynous unions (37 percent), with Central Province having the lowest (8 percent). There is considerable provincial variation in polygyny according to the age of the woman. Twenty-nine percent of married women aged 15-19 in Nyanza Province are in polygynous unions, compared to only 4 percent of women in Rift Valley. Among women in the age group 40-44, Coast Province shows 54 percent in polygynous marriages, while Nairobi has only 10 percent. Except for Nyanza and Eastern Provinces, polygyny decreases slightly among women aged 45-49 years. 11 As in other surveys carried out in Kenya, the KDHS found that there is a negative relationship between education and polygyny. The percentage in polygynous unions decreases from 35 percent among women with no education to 12 percent among women with secondary education and higher. Table 2.3 Percentage of currently married women in a polygynous union, by abe, according to background characteristics, Kenya, 1989 Age Background characteristic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 14.5 14.6 14.6 22.9 27.6 18.5 17.5 17.7 Rural 11.9 18.4 18.2 28.4 26.1 34.1 32.2 24.4 Province Nairobi 15.1 10.3 14.6 18.0 34.0 10.3 6.3 15.4 Central 5.3 3.3 2.5 13.6 4.7 18.8 14.8 8.3 Coast 16.0 24.0 30.8 35.8 37.7 53.7 45.3 34.1 Eastern 9.3 14.7 13.6 20.4 11.8 34.0 36.2 19.5 Nyanza 28.7 24.9 32.5 44.1 50.0 34.9 47.3 37.4 Rift Valley 4.0 18.8 13.9 23°6 20.0 34.0 20.2 19.8 Western 5.7 22.3 19.3 29.3 41.8 40.2 39.6 28.0 Education No education 16.3 39.2 25.8 37.2 35.7 44.5 30.2 35.3 Some primary 14.2 17.6 21.2 26.9 21.4 22.8 38.3 22.5 Primary con~lete 12.7 16.3 11.2 20.6 17.5 22.4 21.3 15.8 Secondary + 5.8 10.0 13.1 13.0 16.3 10.1 (0.0) 11.9 Total 12.7 17.5 17.6 27.6 26.3 33.0 31.2 23.4 Note: Numbers in parentheses are based on fewer than 20 unweighted cases. 2.4 Age at First Marriage Table 2.4 shows that the proportion of women who marry before age 15 has declined from 25 percent of women 40-44 to only 4 percent of women 15-19. This suggests a rising age at first union in Kenya. As the table shows, 75 percent of women aged 40-44 married before the age of 20, compared to only 52 percent of women aged 20-24. With the exception of women 45-49, the median age at marriage has increase d over time, from 17.3 among women 40-44 to 19.8 for those aged 20-24. That the median age of marriage for women aged 45-49 years is higher than expected (18.5) could be explained by recall lapse. Table 2.5 presents the median age at first marriage by selected background characteristics. Only women aged 20-49 are included in the table since median age at marriage for younger women is influenced by the large proportion that have not yet married. In general, urban women marry later than their rural counterparts. This is true for all age groups, with an overall difference of 1.5 years in the median age at marriage. 12 Tabte 2.4 Percent d i s t r ibut ion of women by age at f i r s t marriage and median age at f i r s t marriage, according to current age, Kenya, 1989 Curr- Age at first marriage Wtd. Med- ent Never no.of Jan* age married <15 15-17 18-19 20-21 22-24 25+ Totat women age 15-19 79.8 3.5 11.8 4.9 0.0 0.0 0.0 100.0 1497 20-24 31.8 5.6 25.9 20.3 12.0 4.4 0.0 100.0 1321 19.8 25°29 10.7 15.7 27.5 22.0 11.3 8.5 4.2 100.0 1334 18.6 30-34 5.4 23.0 27.7 17.2 13.6 8.5 4.7 100.0 982 17.9 35-39 3.2 20.1 31.3 20.0 12.4 7.1 5.9 I00.0 898 17.9 40-44 1.5 25.0 30.3 19.7 11.9 7.1 4.4 I00.0 674 17.3 45-49 2.4 17.7 28.0 20.5 13.4 10.1 7.9 100.0 445 18.5 Total 26.0 13.8 24.7 16.9 9.7 5.8 3.1 100.0 7150 - Some data for women age 15-19 and the median for all women have been omitted, since a substantial proportion of these women have not yet married. * Defined as the exact age by which 50 percent of women have experienced marriage. Table 2.5 Median age at first marriage among women age 20-49 years, by current age and background characteristics, Kenya, 1989 Current age Background Characteristic 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 20.3 19.9 19.6 18.7 18.7 19.5 19.8 Rural 19.7 18.3 17.7 17.8 17.3 18.4 18.3 Province Nairobi 20.5 20.1 19.9 19.5 19.4 22.6 20.2 Central 21.9 20.2 19.3 19.3 18.2 19.1 19.9 Coast 19.5 17.1 16.3 16.2 15.1 16.3 17.0 Eastern 22.5 19.2 20.0 19.1 18.3 18.9 19.5 Nyanza 17.7 17.1 16.4 16.6 16.4 17.1 16.9 Rift Valley 19.3 17.6 17.2 18.3 17.3 20.4 18.1 Western 19.0 18.5 17.7 17.1 16.9 15.4 17.9 Education No education 16.7 16.7 16.2 17.1 16.7 18.6 16.9 Some primary 18.3 17.4 17.5 17.5 17.1 17.6 17.6 Primary con~piete 19.7 18.6 19.3 19.0 19.5 19.9 19.2 Secondary + 22.6 21.0 21.0 21.2 22.6 (22.4) 21.6 Total 19.8 18.6 17.9 17.9 17.3 18.5 18.5 Note: Numbers in parentheses are based on fewer than 20 unweighted cases. Provincial differentials in age at marriage also exist in Kenya. Women in Coast and Nyanza Provinces marry the earliest, with a median age of about 17 years, while women in Nairobi and Central Province marry the latest, with a median age of about 20 years. Differences in age at marriage have perhaps been influenced most by increased education of women. As Table 2.5 shows, the median age at marriage increases with the level of education 13 for every cohort of women. Those with a secondary school or higher education have the highest median age at marriage (21.6). The difference in the median age at marriage between this group and women who have no education is 4.7 years over all ages and is almost 6 years for those aged 20-24. Perhaps Kenya's 8-4-4 system of education will effect further increases in age at marriage for younger women now in school. 2.5 Breastfeeding and Postpartum Insusceptibility Breastfeeding, postpartum amenorrhoea and postpartum sexual abstinence are factors related to the risk of pregnancy. The duration of amenorrhoea (the period following a birth before the return of the menstrual cycle) is directly related to breastfeeding--the longer a woman breastfeeds, the longer she is likely to be amenorrhoeic. Table 2.6 shows that over 95 percent of babics in Kenya are breastfed for at least some time. Over 80 percent are breastfed to their first birthday, and almost three-fifths are breastfed for at least 18 months. Table 2.6 Percentage of births whose mothers are still breast- feeding, postpartum amenorrhoeie, abstaining, and insusceptible, by number of months since birth, Kenya 1989 Months Breast- Amenor- Abstain- Insus- No. of since birth feeding rhoeic ing ceptible* births Less than 2 96.0 95.6 88.6 97.3 194 2-5 94.4 85.7 50.8 89.2 244 4-5 91.9 69.8 22.2 74.7 262 6-7 92.2 64.1 21.6 69.2 252 8-9 87.2 56.2 15.4 60.3 260 10-11 91.0 51.4 15.2 54.5 241 12-13 81.7 42.4 17.0 46.0 246 14-15 68.3 28.9 9.1 31.0 263 16-17 63.4 19.5 5.8 22.2 257 18-19 55.9 11.4 6.0 15.7 246 20-21 42.3 10.0 10.1 17.2 211 22-23 40.9 9.2 9.9 16.8 200 24-25 21.7 6.4 5.4 9.8 256 26-27 14.6 2.0 6.1 8.1 235 28-29 15.7 0.7 5.4 6.1 234 30-31 12.5 3.5 3.3 6.4 276 32-33 4.0 0.5 1.3 1.6 233 34-35 4.1 0.4 2.9 3.3 278 Iota[ 54.0 30.6 15.6 34.5 4387 Median 19.4 10.8 2.6 11.6 Note: Includes births 0-35 months before survey * Either amenorrhoeic or abstaining at the time of the survey More than 85 percent of Kenyan women experience amenorrhoea for at least two months after birth, with this percentage dropping rapidly to about 42 percent still amenorrhoeic one year after giving birth. The proportion of amenorrhoeic women decreases faster than that of the breastfeeding women, reaching 11 percent by the 18-19 months after birth. 14 There is a sharper decline in the practice of sexual abstinence after a birth than the decline of either breastfeeding or postpartum amenorrhoea. Only 51 percent of women are abstaining 2- 3 months after birth, whereas 17 percent abstain for at least one year and 5 percent abstain for two years after birth. Table 2.6 also shows the proportion of women who are insusceptible to pregnancy due to either amenorrhoea or the practice of sexual abstinence. One year after giving birth, 46 percent of the women are insusceptible. Table 2.7 provides estimates of the mean duration 1 in months of the four birth-related variables by selected background characteristics. As the table and Figure 2.3 show, there is not much difference in any of the variables by age of the woman. Rural women have slightly longer mean durations of breastfeeding, postpartum amenorrhoea and sexual abstinence than their urban counterparts. As a result, they have a longer period of insusceptibility to pregnancy. Table 2.7 Mean number of months of breastfeeding, postpartum amenorrhoea, postpartum abstinence, and postpartum insusceptibitity, by background characteristics, Kenya, 1989 Background Breast- Amenor- Abstain- Insus- No. of characteristic feeding rhoeic ing ceptibte* births Age <30 19.4 10.5 6.3 12.5 2760 30+ 19.5 11.6 5.2 12.9 1689 Residence Urban 18.8 9.1 5.2 11.0 612 Rural 19.5 11.2 6.0 12.9 3837 Province Nairobi 19.9 9.1 6.3 11.5 264 Central 18.4 10.7 7.7 13.9 619 Coast 17.7 9.4 2.6 9.9 255 Eastern 20,9 9.3 6.4 11.2 794 Nyanza 19.3 11.5 3.9 13.0 789 Rift Valley 19.1 12.2 8.2 14.1 1007 Western 19.7 11.7 3.4 12.1 721 Education NO education 20.9 13.4 6.8 14.7 1144 Some primary 19.1 11.0 4.7 12.1 1395 Primary complete 19.4 I0.0 6.5 12.4 1088 Secondary + 18.0 8.5 5.7 10.8 819 Total 19.4 I0,9 5.9 12,6 4449 Note: /nciudes births 1-36 months before survey. Estimates are based on prevalence/incidence method (see text). Three women with education level not stated are omitted. * Either amenorrheoic or abstaining at the time of the survey 1 Estimates of mean duration are calculated using the prevalence/incidence method, borrowed from epidemiology. The duration of breastfeeding, for example, is defined as the prevalence (number of children whose mothers are breastfeeding at the time of the survey), divided by the incidence (average number of births per month over the last 36 months). 15 Figure 2.3 Duration of Breastfeeding, Amenorrhoea and Postpartum Abstinence Mean Duration in Months 241 21 18 15 12 9 6 3 0 (30 30* Urban Rural AGE RESIDENCE [ ~Breast feed ing Amenorrhoea None Some PrL Sec.* pri. comp. EDUCATION m Abstinence Kenya DHS 1989 Eastern Province recorded the highest mean duration of breastfeeding (21 months), with Coast Province having the shortest, about 18 months. Mean duration of sexual abstinence is particularly short for women in Coast, Western and Nyanza Provinces. In Coast Province, this may be attributed to Islamic religious practiccs. There is an inverse relationship between education and the mean duration of brcastfeeding, amenorrhoea and insusceptibility. The higher the level of education, the shortcr the mean duration of these variables. This may be attributed to the fact that better educated women are more likely to work in jobs that make brcasffeeding more difficult. 16 3 FERTILITY 3.1 Background Everything affecting the demographic character of a population--its size, rate of increase, geographic distribution, age and sex structure, life expectation and family composition--must work through one of three demographic variables: fertility, mortality and migration. Of these, fertility is the major dynamic element. In most instances it is the prime determinant of age structure, family composition and population growth rates. To understand fertility is, therefore, to understand not only a major portion of all demographic behaviour, but a fundamental element in social structure. The fertility measures presented in this chapter are based on the reported reproductive histories of women aged 15-49 interviewed in the KDHS. Each woman was asked the number of sons and daughters living with her, the number living elsewhere, and the number who had died. She was then asked for a history of all her births, including the month and year each was born, the sex, the name, and if dead, the age at death, and if alive, whether he/she was living with the mother. Based on this information, fertility measures like completed fertility (number of children ever born) and current fertility (total fertility rate, or TFR) are examined. These measures are also analyzed in connection with different background characteristics. Thus, the chapter contains a discussion of levels, trends and differentials in fertility of Kenyan women. It is appropriate to mention that the birth history approach has some limitations and is susceptible to data collection errors. Data on the total number of children ever born may be distorted due to socio-cultural factors. Women are likely to include relatives' children among their own children, due to the extended family system in the country. Also, babies who die very young are more likely to be omitted from reporting. Another source of error in the reported number of children could be the inclusion of stillbirths. Women in older age groups also tend to forget grown children, especially those who have left the household. Finally, misreporting of the dates of birth is common in many cultures. 'So, fertility levels can be affected by underreporting, while misreporting of dates of births can seriously distort estimates of fertility trends. There is no complete solution to the above problems, but the interviewers were instructed to do all they could to facilitate respondents' recall, probe for early infant deaths, and avoid including stillbirths. Furthermore cross-checks were built into the questionnaires. Interviewers were instructed to probe for reasons for longer birth intervals and to compare ages, dates, etc., for inconsistencies. Despite these safeguards, there are indications in the KDHS results that births occurring five and six years prior to the survey were shifted to seven years before the survey, presumably to avoid the necessity of filling in the health section for the children. In order to obtain data for all children under age five, questions in the health section were asked for all children born since January 1, 1983. KDHS data on births by year show that there are roughly 30 percent more births reported as occurring in 1982 than in 1983. Similar displacement of births has been found in other DHS surveys. For the purpose of this report, data on trends in fertility that involve the year 1982 or 1983 should he regarded with caution. However, this problem most likely does not affect the rates for the five-year period prior to the survey. 17 3.2 Levels and Trends in Ferti l i ty The total fertility rate (TFR) for the five-year period prior to the survey is 6.7, which represents the total number of births a woman would have by the time she reached age 50 if she had children at the same rate as women are currently having at each age group. As shown in Table 3.1, the KDHS data are the first evidence of a major decline in fertility in Kenya. The total fertility rate was about 8 children per woman in the late 1970s, and although the 1984 Kenya Contraceptive Prevalence Survey (KCPS) showed some slight evidence of decline (to a TFR of 7.7), the KDHS rate of 6.7 represents a substantial decline in fertility. It should be noted that the estimates are not strictly comparable. For example, the rates from the 1962 and 1969 censuses are based on reported data, without the upward adjustment for underreporting that is common for census data. Data from the censuses, the 1977 NDS and the 1984 KCPS refer to rates in the 12-month period before the survey, while the rates from the KDHS refer to the five- year period prior to the survey. rabte 3.1 Age-specific fertility rates from various surveys and censuses, Kenya 1962 1969 1977 1977/78 1979 1984 1989 Age Census Census NDS KFS Census KCPS KDHS 15-19 83 111 135 177 179 143 152 20-24 207 284 365 369 368 358 314 25-29 223 290 361 356 372 338 303 30-34 203 253 316 284 311 291 255 35-39 163 200 231 216 226 233 183 40-44 109 121 133 132 105 109 99 45-49 63 60 56 51 14 66 35 Total fer- tility rate 5.3 6.6 8.0 7.9 7.9 7.7 6.7 Source: Central Bureau of Statistics, 1984, Table 4.13 and Central Bureau of Statistics, no date, Table 6.15. Data from the 1979 Census have been adjusted for underreportin9 of births. Age-specific fertility rates for the KDHS show that the rate increases from 152 births per 1000 women in the youngest age group to over 300 for women aged 20-29 and then decreases steadily to 35 for women aged 45-49. Figure 3.1 gives a graphical representation of the KDHS age-spccific fertility rates for comparison with other survey results. For every age group except the youngest the age-specific fertility rates recorded in the KDHS are lower than those recorded in the other surveys. Further evidence of a fertility decline appears in Table 3.2, which presents age-specific fertility rates for five-year periods prior to the survey, based on data from the KDHS birth histories. In reading the table, one should notc that the figures in parenthesis represcnt partial fertility rates due to truncation. Women 50 ycars and over were not included in the survey and the further back into time rates are calculated, the more severe is the truncation. For example, rates cannot be calculated for women aged 45-49 for the period 5-9 years before the survey, 18 because those women would have been aged 50-54 at the time of the survey and were not interviewed. Figure 3.1 Age-Specific Fertility Rates KFS, KCPS and KDHS Birth8 per 1,000 women 400 300 200 100 0 15-19 • - 4 5'- 30'-34 20-24 2 29 35-39 40-44 Age 45-49 - - - - KFS 1977-78 +- KCPS 1984 ~ KDHS 19B9 i Kenya DH$ 1989 Tabte 3.2 Age-period fertitity rate by age of wc~nan at birth, Kenya, 1989 Nun~)er of years preceding survey Age at birth 0-4 5-9 I0-14 15-19 20-24 25-29 30-34 15-19 152 189 203 216 190 192 (108) 20-24 314 338 357 334 347 (322) 25-29 303 314 337 332 (329) 30-34 255 Z93 304 (285) 35-39 183 241 (306) 40-44 99 (158) 45-49 (35) Note: Figures in parentheses are partially truncated rates. Not avaitable due to age truncation. 19 The data show a decline in fertility rates at all ages from those prevailing 10-14 years before the survey. It is interesting to note that, if one assumes that the same fertility rate at age 45-49 prevailed 5-9 years before the survey as 0-4 years before the survey, the total fertility rate 5-9 years before the survey (approximately 1979-1984) would be 7.8, which is equivalent to the rates recorded in Table 3.1 for the late 1970s and 1984. The data in Table 3.2 show a peak in fertility 10-14 years prior to the survey for all but the youngest age group. While it is possible that fertility has risen and then fallen, another possible explanation is a shifting of births from the period 15-19 and/or 5-9 years prior to the survey. It is important to note that the decline in fertility is consistent with the increase in age at marriage discussed in Chapter 2, as weI1 as the increase in contraceptive use recorded by the KI)HS (see Chapter 4). Another factor relating to the decline could be the recent extension of a comprehensive health care system, which makes it easy to promote population programmes. The fall in the rate of infant mortality could also have contributed to the fertility decline. Several studies have shown that there is a close relationship between the infant mortality rate and the fertility rate. When the probability of child survival increases, couples need to have only that number of children which they actually desire, especially when childbearing involves both physical and mental strain and childrearing is expensive. By achieving major reductions in infant mortality in Kenya, a similar decline in fertility has bccn possible. Perhaps the largest single factor that has contributed to fertility decline is education, cspccially education of women. Age at marriage increases with education, hence delaying the start of childbearing. Tabte 3.3 Age-specific fe r t i l i ty rates and total fe r t i l i ty rates for three periods before the survey, Kenya, 1989 Months prior to sucvey Age at 12 24 60 birth months months months 15-19 139 148 152 20-26 302 317 314 25-29 305 297 303 30-34 250 235 255 35-39 192 180 183 40-44 95 87 99 45-49 12 21 35 Total fer- titity rates: 15-49 6.5 6.4 6.7 15-44 6.4 6.3 6.5 Table 3.3 presents age-specific fertility rates for the 12-month, 24-month and 60-month periods prior to thc survey. They indicate a slight decline between the 60-month period and the 24-month period prior to the survey and an increase at certain ages and a decline at other ages between the 24-month and 12-month period before the survey. The most that can be concluded from these data is that fertility has probably continued to decline in the few years before the survey. 20 In the KDHS, all women were asked whether or not they were pregnant at the time of the survey. The percentage of women pregnant in each age group is shown in Table 3.4 along with comparable information from the 1977/78 KFS and the 1984 KCPS. Table 3.4 Percentage of art women who are currently pregnant by age, Kenya, 1977/78 KFS, 1984 KCPS and 1989 KDHS 1977/78 1984 1989 NO. of Age KFS KCPS KDHS women 15-19 8 8 6.8 1497 20-24 17 16 13.6 1321 25-29 19 17 10.5 1334 30-34 16 13 I0.9 982 35-39 12 10 8.4 898 40-44 9 6 3.6 674 45-49 3 2 2.2 445 Total 13 11 8.9 7150 The data provide further corroboration of a recent fertility decline. Only 9 percent of women in the KDHS said they were pregnant, compared to 11 percent in 1984 and 13 percent in 1977/78. Moreover, the KDHS results reveal a consistent decline in the proportion pregnant in all age groups. 3.3 Ferti l i ty Differentials Knowledge of differential fertility provides valuable information about the relative contributions of different socio-economic and cultural factors to the overall level of fertility, thus providing an indication of future fertility rates. Only after these differences have been ascertained is it possible to investigate the pattern of causation underlying it. Questions asked in the 1989 KDHS make it possible to study differentials by urban-rural residence, province, and education; these are presented in Table 3.5. Table 3.5 also presents thc total fertility rates for two calendar year periods (1986-88) and (1983-1985) and for the five-year period preceding the survey, as well as the mean number of children ever born to women 40-49 years old, according to background characteristics of women. Caution should be cxercised in comparing these rates, as the average number of births to women 40-49 refers to past or completed fertility, while total fertility in the preceding five years refers to a more current measure of fertility. A.s mentioned above, the rates for 1983-85 are especially suspect, as they include the year 1983, from which a number of births were evidently displaced back to 1982. This has the effect of reducing the apparent decline in fertility between the two 3-year periods. Comparing the last two columns of Table 3.5 reveals that there has been a major decline in fertility, from 7.5 children ever born to women 45-49 to a total fertility rate of 6.7. This is consistent with the other evidence of fertility decline. 21 ~abke 3.5 Total fe r t i l i ty rates for calendar year per iods and for f i ve years preceding the survey, and mean number of ch i ld ren ever born to women 40-49 years of age, by background character i s t i cs , Kenya, 1989 Total fertility rates* Mean number of 0-4 ch i ld ren years ever born Background 1986- 1983- before to women character i s t i c 1988"* 1985 survey age 40-49 Residence Urban 4.5 5,8 4.8 5.1 Rural 7.1 7,1 7.1 7.7 Province Nairobi 4.2 7.0 4.6 4.9 Central 6,0 6,4 6,0 7,3 Coast 5,4 6.0 5.5 7,3 Eastern 7.2 7,0 7.0 7.4 Nyanza 6,9 7.1 7.1 7,9 Ri f t Va l ley 7.0 6.7 7.0 7.4 ~estern 8.1 7.9 8.1 8.2 Education No education 7.5 7.2 7.2 7.4 some primary 7.5 7.5 7,5 8.0 Primary co~otete 6.4 6.4 6.5 7.3 Secondary + 4.8 5,0 4.9 4.7 Total 15-49 6.7 6,8 6.7 7.5 Total 15-44 6.5 6.5 6.5 * Based on women 15-49 ** Includes exposure in 1989 up to the time of interview. There is a considerable difference in fertility betwcen rural and urban areas. Based on births in the five years before the survey, women in urban areas have a total fertility rate of approximately 5 live births, compared to about 7 for rural women. This difference also exists in the mean number of children ever born to women 40-49 (Figure 3.2). Much of the observed urban-rural differences in fertility are probably due to the differential practice of birth control, which spread outward from urban to rural areas. Urban areas usually have the most educated, highest income population, as well as the best medical facilities. The difference could also be attributed to urban-rural differences in three inter-related factors that determine the ability to control fertility: knowledge about birth control, skill in its practice and degree of access to the most effective means. A/though fertility rates are still high in Kenya, there is considerable variation between provinces (Figure 3.3). For the five years prior to the survey, Nairobi had the lowest total fertility rate (4.6), while Western Province had the highest (8.1). This is also consistent with the mean number of children ever born to women 40-49 years old. The observed regional differentials could be due to the highly diverse physical and climatic environment, reflecting diverse modern and traditional systems of land use. The geographic distribution of Kenyan tribes, with particular tribes being concentrated in certain provinces, leads to variation in cultural practices among the provinces. The large difference in the total fertility ratcs for Nairobi for 1983-85 and 1986-88 is most probably 22 due to a combination of sampling error due to small sample sizes in certain age groups of women and to misreporting of dates of birth. Figure 3.2 Total Ferti l ity Rate (TFR) and Mean Number of Children Ever Born (CEB) to Women 40-49 by Residence and Education RESIDENCE Urban Rural EDUCATION None Some Primary Primary Comp. Secondary* i~\\~\\\\\\\\\\\\\\\\\\\\\\~\\\\\\\\\\~ k\~\\\\\\~\\\\\\\\\\~'~k'~ ~\\\\\\\\\\\\\\\\X~ ~\\\\\\\~kk\~kkk\\\t 2 4 6 8 10 Number of Children TFR i CEB Kenya DHS 1989 No other social variable has been as frequently associated with fertility differentials as education. The total fertility rate of women with complete primary education (6.5) is lower than the one for women with no education (7.2). Women with secondary and higher education have the lowest total fertility rate of 4.9 live births. A negative association between schooling and fertility has been widely observed. Schooling may have its own independent effect on fertility, through raising the age at marriage or it may be an indicator of the existence of certain elements that are correlated with lower fertility, such as higher socio-economic status. 3.4 Cumulat ive Ferti l i ty The number of children ever born (cumulative fertility) is one of the basic measures of fertility. As pointed out above, it is subject to possible errors such as omission of births and inclusion of stillbirths and children of relatives. Table 3.6 shows that the level of fertility in Kenya is still high. It should be noted that just as marriage occurs relatively early, childbearing also occurs early with teenage girls (15-19) reporting an average of 0.3 births (last column of Table 3.6). Women in their early twenties have had an average of more than one and half births each. This increases rapidly to 3.5 births among women in their late twenties and 6.5 births for women in their late thirties. By the time women reach the end of their childbearing years (45-49), they have had, on average, 7.6 live births. 23 10 Figure 3.3 Total Fertility Rates by Province, KFS, KCPS and KDHS Births per woman 8 6 4 2 O i Nairobi Central Coast Eastern Nyanza Rift Valley Western Province m KFS 1977-78 ~ KCPS 1984 ~ KDHS 1989 / I Kenya DHS 1989 Table 3.6 Percent distrib~Jtion of all women and currently married women by number of children ever born (CEB}, according to age, Kenya, 1989 Nun~oer of children ever born Wtd. Mean no. of no. Total women CEB Age 0 I 2 3 4 5 6 7 8 9 10+ At[ Women 15-19 78.6 15.9 4.4 0.9 0.2 0.0 0.0 0.0 0.0 0.0 0.0 100.0 1497 0.3 20-24 21.5 30.0 25.4 16.3 5.5 1.0 0.2 0.0 0.0 0.0 0.0 100.0 1321 1.6 25-29 5.3 9.3 13.7 20.3 24.2 14.7 8.9 2.6 0.8 0.0 0.1 100.0 1334 3.5 30-34 2.9 4.1 6.7 10.8 16.0 14.8 19.0 12.9 8.5 3.3 1.2 100.0 982 5.0 35-39 2.2 1.4 4.9 4.7 7.4 12.9 14.3 15.1 16.2 9.0 11.9 100.0 898 6.5 40-44 2.3 1.7 3.2 3.1 5.6 8.4 12.2 10.9 14.8 14.1 23.6 100.0 674 7.4 45-49 2.8 1.9 1.4 4.3 4.8 6.2 11.8 12.2 14.0 13.7 26.7 100.0 445 7.6 Total 22.5 11.6 10.1 9.6 9.5 7.8 8.0 5.9 5.6 3.9 5.6 100.0 7150 3.7 Curently Married Women 15-19 32.8 42.8 19.2 4.3 0.9 0.0 0.0 0.0 0.0 0.0 0.0 100.0 276 1.0 20-24 8.4 25.0 32.6 23.9 8.3 1.5 0.3 0.0 0.0 0.0 0.0 100.0 827 2.0 25-29 3.0 6.7 12.5 21.8 24.7 16.9 10.2 5.0 1.0 0.0 0.1 100.0 1104 3.7 30-34 1.8 3.2 4.9 10.0 16.5 15.3 19.6 14.3 8.9 3.8 1.4 100.0 833 5.2 35-39 1.3 0.9 4.5 3.7 7.1 12.2 14.8 16.1 15.9 10.3 13.1 100.0 781 6.7 40-44 2.1 1.3 3.0 2.6 5.0 6.8 11.4 11.5 15.4 15.5 25.5 100.0 576 7.6 45-49 2.6 0.8 1.4 4.5 4.7 5.6 9.0 10.7 15.7 15.7 29.4 100.0 369 7.9 Total 5.0 9.3 11.7 12.5 12.2 10.1 10.4 8.1 7.5 5.4 7.6 100.0 4765 4.6 24 The distribution of women by number of births reveals that almost 78 percent of women 20-24 have had at least one child. By the time Kenyan women reach the end of childbearing, 27 percent have had ten or more live births. Primary infertility--the proportion of married women aged 45-49 who have ncver had children--is quite low at 3 percent. This confirms the 1984 KCPS finding that only 3 percent of women aged 40 years and above reported never having given birth (Central Bureau of Statistics, 1984, Table 4.8). Information on cumulative fertility from past surveys and censuses can be compared with the KDHS results. Table 3.7 indicates that fertility generally increased between 1962 and 1984 and declined thereafter; however, these trends should be interpreted cautiously because different methods were employed in eliciting information on births. For example, KFS and KDHS questionnaires employed a birth history approach, while the KCPS collected only summary data on the number of children ever born. The data on children ever born from the censuses may be biased downwards due to difficulty in obtaining high quality data on such a large scale. Table 3.7 Mean number of children ever ~rn as reported in various surveys and censuses, by age group, Kenya 1962 1969 1977 1977/78 1979 1984 1989 Age Census Census NDS KFS Census KCPS KDHS 15-19 0.4 0.4 0.3 0.4 0.3 0.4 0.3 20-24 1.7 1.9 1.8 1.8 1.9 2.0 1.6 25-29 3.0 3.7 3.7 3.8 3.7 4.0 3.5 30-34 4.2 5.1 5.6 5.6 5.4 5.7 5.0 35-39 5.1 6.0 6.7 6.8 6.5 7.0 6.5 40-44 5.6 6,4 7.3 7.6 7.0 7.8 7.4 45-49 5.9 63 7.5 7.9 7.2 8.2 7.6 Source: Central Bureau of Statistics, 1984, Table 4.9 Looking only at the data from the surveys, the mean number of children ever born from the 1984 KCPS is higher for each age group than from either the 1977/78 KFS or the 1989 KDHS. The investigation of possible overreporting of fertility in the KCPS or underreporting in the KDHS could be a useful topic for further analysis. Previous methodological research in Kenya (Central Bureau of Statistics, 1975 and 1977) has shown that birth histories rcsult in lower estimates of cumulative fertility than summary data, though the cause is unclear. In any case, the figures from the KDHS are lower than those from the two surveys in the late 1970s, which is consistent with othe'r evidence of a rccent decline in fertility. The mean number of children ever born by age at first marriage and duration of marriage is given in Table 3.8. As expected, the mean number of childrcn born rises with increasing marital duration. The results indicate that irrespective of the age at first marriage, a Kenyan woman would have given birth to an average of 3.3 children during the first 5-9 years of her marriage. At shorter marriage durations, the mean number of children ever born increases with age at marriage, which is unexpected. This could be due to a greater possibility that late-marrying women experience pre-marital conceptions or births, thus artificially raising the tempo of early marital fertility. Another possibility could be that late-marrying women have shorter birth intervals due to shorter breastfeeding durations. At longer durations of marriage, the relationship between 25 Table 3.8 Mean number of ch i ld ren ever born to ever-marr ied women, by age at f i r s t marriage and years s ince f i r s t marr iage, Kenya, 1989 Years Age at first marriage s ince f i r s t marriage <15 15-17 18-19 20-21 22-24 25+ Total 0-4 1.2 1.3 1.3 1.6 1.4 2.9 1.5 5-9 2.7 3.2 3.2 3.5 3.5 3.8 3.3 10-14 4.4 4.6 4.7 4.7 4.9 5.3 4.7 15-19 5.5 6.3 6.0 6.3 6.5 4.6 6.0 20-24 6,9 7.3 7.1 6.8 6.5 16,&) 7.0 25-29 7.5 8.4 7.8 8.2 (5.3) 7.9 30+ 8.2 8.6 (8 .7) 8.3 Total 6.0 4.9 4.3 4.3 3.8 3.9 4.8 ( ) Fewer than 20 unweighted cases. Ho cases, s ine ~ definition these women would ~ age 50 or over. children born and age at marriage is erratic. The data in Table 3.8 may also reflects adolescent subfecundity, in that women who marry when they are under 15 years generally have a lower mean number of children cvcr born. Caution should be exercised in interpreting the data in Table 3.8 because the data on age at first marriage are subject to reporting errors. 3.5 Age at First Birth The onset of childbearing is art important demographic indicator. In many countries, postponement of first births, rcflecting a rise in age at marriage, has made a large contribution to the overall fertility decline. Also, the proportion of women who become mothers in their teenage years, before the age of 20, is a basic indicator of maternal and child health. Table 3.9 shows the distribution of women by age at first birth and current age. Table 3.9 Percent distribution of women by age at first birth, according to current age, Kenya, 1989 Wtd. Median Age at first birth nun~)er age at Current No of first age births <15 15-17 18-19 20-21 22-24 25+ Total w(~en birth* 15-19 78.6 2.3 14.0 5.1 100.0 1497 20-24 21.5 4.0 28.0 26.4 15.3 4.8 100o0 1321 19.3 25-29 5.3 11.1 29.2 27.2 16.4 8.3 2.5 100.0 1334 18.7 3D-34 2.9 15.1 32.0 22.4 14.0 9.4 4.3 100.0 982 18.2 35-39 2.2 11.3 28.7 27.5 16.6 9.8 3.9 100.0 898 18.6 40-44 2.3 14.7 29.2 20.7 15.8 I0.6 6.6 100.0 674 18.6 45-49 2.8 10.2 21.2 22.7 17.7 13.4 11.9 I00.0 445 19.7 Total 22.5 8.8 25.6 20.9 12.5 6.8 2.9 100.0 7150 * Def ined as the exact age by which 50 percent of wo~n have had a b i r th . 26 The data show that 55 percent of Kenyan women become mothers before they reach age 20. This finding has serious health implications, since young mothers suffer more health problems than older mothers, and their children have higher mortality rates. The data imply that the median age at first birth has been relatively constant over time, with younger women having almost the same median ages at first birth as the older women. However, it should be noted that the data are heavily dependent on correct reporting of dates of birth of both the woman and her first birth. Table 3.10 presents data on differentials in median age at first birth among women aged 20-49 years by background characteristics of women. The table reveals that urban women start childbearing late compared to their rural counterparts. Nyanza Province has the lowest median age at first birth (17.8), while Nairobi has the highest median age at first birth (19.9). A~s expected, women with no education report the lowest median age at first birth (18.1), which is almost 3 years earlier than their secondary and higher educated counterparts (20.7). Table 3.10 Median age at first birth among women aged 20-49 years, by current age and background characteristics, Kenya, 1989 Current age Background characteristics 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 19.9 19,9 19.7 19.2 19,9 21.2 19,8 Rural 19.1 18.4 18.0 18.6 18.5 19.6 18.6 Province Nairobi 19.9 19,9 19.4 18.0 20.5 23.2 19.9 Central 19.7 19.1 18.6 19.2 19.1 19.7 19.2 Coast 20.5 18.3 19.3 17,5 17.4 18,7 18.7 Eastern 19,5 19.2 18.5 19,4 19.2 19.8 19.3 Nyanza 18.0 17.8 17.3 17.7 17.5 18.2 17.8 Rift Valley 19.1 18,2 17.9 18.7 18.6 22.0 18.6 Western 19,2 18.7 18.1 10.3 18.7 17.9 18.6 Education No education 17.7 17.3 17.6 18.2 18,1 20.0 18.1 Some primary 18.3 17.8 17.6 18.5 18.6 18.9 18,2 Primary complete 18.9 18.7 16.5 18.8 20.2 20.6 18.8 Secondary + 20.9 20.3 20.8 20.9 22.4 (23.3) 20.7 Total 19.3 10,7 18.2 18.6 18.6 19.7 18.8 Note: Median is defined as the exact age by which 50 percent of women have had a birth. Numbers in parentheses are based on fewer than 20 unweighted cases, 27 4 FERTILITY REGULATION 4.1 Contracept ive Knowledge Determining the level of knowledge of contraceptive methods and services was a major survey objective, since knowledge of contraceptive methods and of places where these methods can be obtained are preconditions for their use. Information about knowledge of contraceptive methods was collected by asking the repondent to name ways by which a couple could delay or avoid pregnancy. If a respondent failed to mention any particular method spontaneously, the method was described by the interviewer and then the respondent was asked if she recognized the method. In the questionnaire, seven modern methods--pill, IUD, injection, condom, .barrier methods (diaphragm, foam and jelly), female sterilisation, and male sterilisation--were described, as well as two traditional methods--periodic abstinence (or rhythm) and withdrawal. Any other methods mentioned by the respondent, such as herbs or breastfeeding, were also recorded. For any method that she recognised, the respondent was asked if she knew of a source or a person from whom she could obtain the method. If she reported knowing about rhythm she was also asked if she knew a place or person from whom she could get information on the method. Table 4.1 Percentage of all women and currently married women knowing a contraceptive method and knowing a source by specific method, Kenya, 1989 Knows Knows Knows Knows method method source source Method AW CMW AW CMW Any method 90.0 92.4 88.1 90.8 Any modern method 88.4 91.3 86.5 89.9 Pill 84.4 88.4 81.6 86.3 IUD 62.0 67.0 60.0 65.1 Injections 76.3 81.9 74.2 79.9 Diaphragm/Foam/Jelly 24.4 26.7 23.2 25.5 Condom 53.4 55.7 49.2 51.7 Female steritisation 68.2 72.5 65.9 70.6 Male sterilisation 19.8 21.7 19.0 21.2 Any traditional method 54.8 55.8 44.6 44.8 Periedic abstinence 50.7 50.8 44.6 44.8 Withdrawal 16.8 18.2 Other 5.1 6.3 AW = All women (7150); CMW = Currently married women (4765) The KDHS results indicate that 90 percent of Kenyan women know at least one contraceptive method (Table 4.1). This is an increase from the levels reported in the 1977/78 Kenya Fertility Survey (88 percent) and the 1984 Kenya Contraceptive Prevalence Survey (81 percent). More women indicate that they know a modern method (88 percent) than a traditional method (55 percent). 29 The pill, recognised by 84 percent of women interviewed, is the most widely known method. In the 1984 KCPS, only 73 percent of women interviewed recognised the pill. Injection is the second most widely known method (76 percent). The other better known methods are female sterilisation (68 percent) and the IUD (62 percent). Fifty-three percent of all women report knowing about the condom and 24 percent know about barrier methods. The least known method is male sterilisation (20 percent). Lack of knowledge of condom and male sterilisation may be attributed to the fact that they are male oriented. Considering the traditional methods included in the questionnaire, periodic abstinence (51 percent) is better known than withdrawal (17 percent) or any other traditional method (5 percent). It should be noted that for all methods knowledge is higher among currently married women than among all women. In order for women to adopt family planning, they need to know about the available methods as well as to be aware of where they can obtain contraceptive information and services. More currently married women (91 percent) know a source for a contraceptive method than do all women (88 percent). While only 45 percent of all the women interviewed know a source for a traditional method, 87 percent know of a source where they can obtain a modern contraceptive method. Most women (82 percent) know a source where they can obtain the pill, 74 percent know where to obtain injections, 66 percent know a source for female sterilisation and 60 percent know a source for the IUD. Forty-nine percent know where to obtain condoms, 45 percent know where to obtain information on periodic abstinence, and less than 25 percent know a source for either the barrier methods or male sterilisation. TabLe 4.2 Percentage of currentLy married women knowing at teast one modern method, knowing a source for a modern method, by back- ground character i s t i cs , Kenya, 1989 Background Knows Wtd. character- modern Knows no. of i s t ie method source women Age 15-19 86.0 84.5 276 20-24 94.5 93.3 827 25-29 93.9 92.4 1104 30-34 92.3 91.1 833 35-39 92.9 92.1 781 40-44 85.4 83.8 576 45-49 83.1 80,6 369 Residence urban 95.2 94.1 748 RuraL 90.5 89.1 4018 Province Rairobi 94.8 93.8 335 CentraL 95.8 95.2 648 Coast 92.3 89.2 350 Eastern 92.7 90.1 804 Nyanza 93.3 91.6 872 Rift VaLLey 84.6 84.0 1047 Western 90.6 89.7 711 Education No education 82.8 80.6 1506 Some primary 92.0 90.9 1462 Primary co(npLete 96.9 95.8 987 Secondary + 98.8 98.1 804 ReLigion CathoLic 90.7 89.2 1656 Protestant 93.1 92.0 2706 MusLim 94.7 92.0 165 other 79.3 79.3 79 No reLigion 65.5 62.0 151 Total 91.3 89.9 4765 ~ote: Excludes a smaLL number of women not stated as to education and reLigion. Some interesting differences are revealed when knowledge of methods and sources is considered in connection with background characteristics of the respondents (Table 4.2). Over 80 percent of currently married women in each age group know at least one modern contraceptive method. Knowledge of modern methods is lowest (83 percent) among women aged 45-49 and highest (95 percent) among women aged 20-24. Similarly, knowledge of a source for contraceptive information or services is lowest among women 45-49 (81 percent) and highest among .women 20- 24 (93 percent). 30 Although urban women are more likely than rural women to know about a method of contraception or a source of information or services, the difference is not pronounced. Ninety- five percent of currently married urban women know at least one modern contraceptive method, compared to 91 percent of their rural counterparts. While 94 percent of urban women could name a source, almost as many (89 percent) rural women could do the same. Provincial variations in contraceptive knowledge are rather small. Central Province has the highest level of contraceptive knowledge (96 percent), followed closely by Nairobi (95 percent), Eastern and Nyanza Provinces (93 percent), Coast Province (92 percent) and Western Province (91 percent). Contraceptive knowledge is lowest in Rift Valley Province (85 percent). Both knowledge of a modern method and knowledge of a source of information or services increase with higher levels of education. While 83 percent of the currently married women with no education know at least one modern contraceptive method, 97 percent of women with some primary education and 99 percent of those who had acquired secondary and higher education know at least one modern method. The same is true with knowledge of a source of contraceptive information or services--knowledge goes up with increased education. Often contraceptive knowledge is associated with religious affiliation. The results from this survey, however, indicate that the differentials in knowledge by religion are small; over 90 percent of Catholic, Protestant and Muslim women know of a modern method and almost as many know of a source. Knowledge of both method and source is lowest among currently married women with no religion. 4.2 Acceptabi l i ty of Methods The women interviewed during the KDHS were asked to report problems in connection with contraceptive methods that they had heard about. Table 4.3 shows that the proportion of women who give "no problem" or "don't know" as answers is high for most methods, which may reflect a lack of depth of knowledge of many methods. "Health concerns" were cited frequently for most modern methods, especially the pill (39 pcrcent), IUD (29 percent) and injcction (26 percent). Substantial proportions reported ineffectiveness as the major problem with periodic abstinence (20 percent), withdrawal (18 percent), barrier methods (17 percent), and the condom (15 percent). It is somewhat surprising that ineffectiveness was also cited by 10 percent of women knowing the IUD. Disapproval of the partner was cited more frequently for the male-oriented methods (withdrawal, condom, and male sterilisation) and inconvenience was cited more frequently for withdrawal, condom, barrier methods, and periodic abstinence. 4.3 Knowledge of Supply Sources Information on knowledge of sources of contraceptive methods was obtained by asking women who know a method about where they could obtain the method if they wanted to use it. Results show that government institutions are perceived as the primary source of contraceptive services for most methods (Table 4.4). Over 85 percent of respondents said they would go to either a government hospital or health center to obtain the pill, IUD, injection, and fcmale and sterilisation, while over 70 percent said they would go to one of these sources to obtain barrier methods and condoms. The Family Planning Association of Kenya (FPAK) was the next most frequently named potential source for most of the modern methods, except condoms, which a 31 Table 4.3 Percent distril~tion of women who have ever heard of a contraceptive method by main problem perceived in using the method, according to specific method, Kenya, 1989 Contraceptive method Main Diaphragm/ Female Male Per iodic problem In jec - foam Con- s te r i [ - s te r i [ - abs t in - With- perceived P i l l IUD t ion je l l y dom i sa t ion i sa t ion ence drawat None 21.9 17.0 24.2 16.7 25.3 34.7 27.0 51.1 22.7 Not effective 2.7 9.7 3.0 16.5 14.8 2.9 0.7 19.8 18.3 Partner disapproves 0.2 0.4 0.3 2.1 5.2 2.0 5.5 1.2 11.0 Community disapprove 0.1 0.2 0.4 0.0 0.0 0.2 0.5 0.0 0.2 Rel ig ion disapproves 0.3 0.3 0.2 0.1 0.0 0.3 0.4 0.1 0.0 Health concerns 38.8 28.9 25.9 6.4 1.7 17.3 10.3 0.6 0.3 Access/Avai t ab i l i ty 0.1 0.2 0.2 0.1 0.0 0.1 0.3 0.1 0.0 Costs too much 0.2 0.0 0.1 0.2 0.1 0.2 0.1 0.1 0.2 Inconvenient to use 1.5 3.7 1.1 5.0 5.5 1.2 1.3 5.3 12.6 other 1.3 1.4 2.0 0.8 0.4 1.4 0.7 0.3 2.3 Don't know 32.6 37.8 42.2 51.7 46.2 39.0 52.4 20.1 28.3 Missing 0.2 0.4 0.5 0.5 0.8 0.8 0.8 1.3 4.1 Total 100.0 100.0 100.0 100.0 100.0 I00,0 100.0 100.0 100.0 Number of women 6032 4436 5459 1747 3815 4874 1413 3624 1202 Table 4.4 Percent distrib~Jtion of women knowing a contraceptive method by supply source they say they would use, according to specific method, Kenya, 1989 Contraceptive method Supply source Diaphragm/ Female Male Per iodic that would In jec - foam Con- s te r i l - s te r i l - abs t in - be used Pill IUO l ion je l l y dom i sa t ion i sa t ion once Nowhere 0.1 0.1 0.1 0.3 0.1 0.2 0.I 7.9 Govt. hospital 56.7 59.5 59.1 48.5 42.8 80.7 82.1 11.3 Govt. health center 29.1 27.9 27.9 29.4 28.4 8.9 7.4 10.4 FPAK* 5.0 5.1 5.4 8.1 6.6 3.1 2.4 5.6 Mobile clinic 1.1 0.4 8.6 1.1 1.2 0.1 0.0 1.5 Field educator 0.2 0.0 0.0 0.I 0.3 0.0 0.0 1.6 Pharmacy/Shop 0.7 0.2 0.1 3.4 8.1 0.1 0.3 0.0 Private hospital 1.4 1.6 1.8 2.0 1.9 1.7 1.6 1.3 Mission hospital/dispen. 1.9 1.4 1.7 1.6 1.1 1.7 0.9 1.9 Emp[oyer,s c l in i c 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.2 Private doctor 0.4 0.5 0.4 0.7 0.5 0.4 0.6 1.7 Trad i t iona l hea ler 0,1 0.0 0.0 0.0 0.0 0.0 0.0 0.2 Husband would oet 0.0 0.0 0.0 0.0 1.0 0.0 0.1 2.4 Friends/Relatives 0.1 0.0 0.0 0.0 0.1 0.0 0.6 43.2 Other 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.8 Don't know 3.0 3.1 2.4 4.4 7.1 2.6 3.1 3.4 Missing 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 6032 4436 5459 1747 3815 4874 1413 3624 * Family Planning Association of Kenya 32 larger proportion of respondents said they would obtain at a pharmacy or shop. Those who know about periodic abstinence were most likely to say they would go to friends or relatives for advice about the method. 4.4 Ever Use of Contracept ion Ever use of contraception is one of the most important items of information in the Kenya Demographic and Health Survey. The survey asked women if they had ever used any of the contraceptive methods they said they knew, Table 4.5 ~ercentage of all women and currently married wome~ who have ever used a contraceptive method, by 8peclfic method and age, Kenya, 1989 Contraceptive method Dla- Peri- Wtd. Any phragm/ Female Male Any odic With- number Any modern In Jec- foam Con- sterll- sterll- trad' 1 abstl- draw- of Age method method Pill IUD tlon Jelly dom isation iaatlon method hence al Other women All Women 15-10 14.9 4.2 2.2 0.6 0.5 0.3 1.5 0.0 0.I 12,1 11.4 1.0 0.9 I~97 20-24 40.5 21.2 14.9 3.6 2.9 1.5 5.8 0.8 0.0 25,9 25.6 2.4 3.0 1321 25-29 47,1 30.6 21.0 7.9 6.~ 1.8 5.1 1.3 0.i 25.6 23.6 3.1 1.4 1334 30-34 50.5 35,9 23.8 11.7 0.9 2.7 4.4 6,3 0.2 24,5 21.9 3.4 3.5 982 35-30 49.8 34.4 18.0 II.4 8.5 3.2 5,0 8.8 '0.4 24,6 20.3 2.9 3.8 898 40-44 43.9 29.2 15.7 9.6 9.8 3.0 2.6 9.1 0.5 23.3 18.6 2.1 4.9 674 45-49 39.4 26.2 15.9 9.2 5.1 1.6 2.3 10.3 0.i 19,6 14.6 2,4 5.9 445 Total 39.1 24.1 15.1 6.8 5.5 1.8 3.6 3.6 0.2 21.9 19.4 2.4 2.7 7150 Currently Marrled Women 15-19 26.2 Ii.i 8.2 1.6 0.5 0.3 4.2 0.0 0.0 19.0 18.4 2.5 0.3 276 20-24 41.7 22.8 16.2 4.1 3.2 1.3 4.4 0.7 0.0 25.7 22.6 2.7 3.6 827 25-29 46.5 30,0 1~.0 8.1 6.7 1.8 5.2 1.4 0.I 25.2 23.3 3.0 1.3 1104 30-34 48.2 35,1 22.6 II.I 9.4 2.5 4.2 6.4 0.5 22.7 19.6 3.7 3.9 835 35-39 49.8 33.7 17.3 11.4 8.2 3.4 5.3 8.9 0.4 25.7 21.2 3.4 3.8 781 40-44 46.3 30,I 16.2 9.4 I0.0 3.i 2.6 9.8 0.0 25.1 20.2 2.2 5.2 576 45-49 ~2.6 27,6 16.9 0.8 4.9 1.8 2,7 i0,0 0.2 20.7 15.3 2.6 4,1 369 Total 45.0 29.0 18.0 8.4 6.7 2.1 4,3 5,0 0.i 24.2 20.9 3.0 3,2 4765 Table 4.5 shows the proportion of all women and currently married women who have had experience with contraceptive methods. The level of ever use of any method among all women is 39 percent, higher than the level of 29 percent reported in both the 1977/78 Kenya Fertility Survey and the 1984 Kenya Contraceptive Prevalence Survey. KDHS data also show that the level of ever-use among all women (39 percent) is lower than that of currently married women (45 percent). Ever use of modern methods of contraception is slightly higher (24 percent) than that of traditional methods (22 percent) among all women and among the currently married women, 29 percent of whom had used at least one modern method and 24 percent of whom had used a traditional method. Three observations are that: 33 ever-use has increased over the past decade, ever-use is higher among currently married women than all women, and ever-use is slightly higher for the modern methods. The KDHS results reveal that the ever-use rate for any method is low for all women aged 15-19 (15 percent), and then it increases for women in the age groups 20-24 (40 percent) and 25- 29 (47 percent). Ever-use is highest among women aged 30-39 (50 percent), but then declines slightly for all women aged 40-44 (44 percent) and 45-49 (39 percent). Periodic abstinence has been used by more women than any other mcthod (19 percent), followed by the pill (15 percent). Seven percent of all women have used the IUD, 6 percent have used injection, and 4 percent have used either female sterilisation or condoms. The proportion of women who have been sterilised increases with age. 4.5 Cur rent Contracept ive Use The level of current use of contraceptives is the most widely used measure of the success of a family planning programme. The KDHS results show that 27 percent of currently married Kenyan women are currently using a contraceptive method (Table 4.6). As in the case of ever- use, the contraceptive prevalence rate among currently married women is higher (27 percent) than among all women (23 percent). Current use of contraceptives is usually presented tbr currently married women, because they are likely to be more consistently exposed to the risk of pregnancy. More currently married women are using modern contraceptives (18 percent) than traditional methods (9 percent). Nevertheless, periodic abstinence is the single most widely used method--used by. 8 percent of currently married women. The next most popular method is the pill, used by 5 percent of married women. Current use for other methods include female sterilisation (5 percent), IUD (4 percent) and injection (3 percent). Less than one percent of married womcn rely on barrier methods, condoms or withdrawal. The 27 percent level of contraceptive use recorded in the KDHS represents an increase of more than 50 percent over the rate from the 1984 KCPS (17 percent) and almost four times the rate from the 1977/78 KFS (7 percent). Use of modern methods has doubled since 1984, from 9 to 18 percent of currently married women (Figure 4.1). Injectible contraceptivcs have shown the biggest gain, from less than 1 percent of married women in 1984 to over 3 percent in 1989. Use of periodic abstinence and female sterilisation has almost doubled since 1984. An inverted U-pattern of prevalence by age was observed for the currently married sample. As in the case of ever-use, the current use rate for any method is low for the currently married women aged 15-19 (13 percent), but rises steadily, reaching 34 percent among those aged 35-39. Current use then falls to 31 percent in the 40-44 age group and 24 percent for women aged 45- 49. Current use is probably lower among younger women because many of them are interested in starting their families and among older women, because some are no longer fecund. 34 Table 4.6 Percent dls~rlbutlon of all women and currently married women, by contraceptive me~hod currently being used, according to age r Kenya t 1989 Contraceptive method Weigh- Dla- Female Perl- Not ted Any phragm/ st~r- Any odic With- curt- number Any modern InJec- foam/ Con- illsa- trad'l abstl- draw- ently of Age method method Pill IUD tlon ~elly dom tion method nence al Other using Total women All Women 15-19 7.5 1,0 1.3 0.3 0.i 0,0 0.i 0.0 5.7 4.0 0.O 0.8 92.5 i00.0 1497 20-24 20.7 11.5 6,6 2.0 1.4 0.3 0.8 0.4 9.2 7.8 O.l 1.2 79.3 i00.0 1321 25-29 27.2 17.5 8,i 3.7 3.4 0.i 0.9 1,2 9.7 8,6 0,3 0.8 72,0 I00.0 1334 30-34 32.1 22.5 5,6 4.7 5.6 0.6 0.i 5.8 9.6 7,0 0,2 2.4 67.9 i00.0 982 35-39 34.1 23.1 3.7 5.7 4.5 0.4 0.4 8.4 I0.9 8.9 0.2 1.8 65.9 1O0.0 899 40-44 27.8 19.7 2.3 3.6 3.6 1.0 0,I 8.5 8.2 6.9 0.3 1.5 72.2 I00.0 674 45-49 22.3 16.7 1.9 2.0 1,3 0.5 0,I I0.i 5.6 4.0 0.0 1.6 77.7 i00,0 445 Total 23.2 14.7 4,6 3.0 2.7 0.3 0.4 3.6 8.5 7,0 0.2 1,3 76.8 100.0 7150 Currently Married Women 15-19 13,0 6.7 5.1 1.3 0.3 0.0 0,0 0.0 6.3 6.1 0.0 0.3 87.0 i00.0 276 20-24 20.1 11.8 6.7 2.2 1,2 0.i I,i 0.6 8.3 7.1 0.2 1.0 79.9 100.0 827 25-29 26.1 16.8 7.4 3.5 3,5 0.2 0.8 1.3 9.3 8.i 0.4 0.7 73.9 i00.0 1104 30-34 51.5 22.2 5.6 4.9 5.1 0.6 0.i 5,8 9.2 6,4 0.2 2.6 69.5 i00.0 833 35-39 34.2 22.9 3.4 5.4 4.8 0.5 0.3 8.4 11.3 9.6 0.2 1.5 65.8 I00.0 781 40-44 30.6 21.2 2.7 3.0 4.2 1.2 0.I 9.1 9.4 7.9 0.3 i.I 69.4 I00.0 576 45-49 23.7 17.5 1.8 3,~ 1.5 0.6 0.2 10.0 6.2 4.7 0.0 1.6 76.3 i00.0 369 Total 26.9 17.9 5.2 3.7 3.1 0.4 0.5 4.7 9.0 7.5 0,2 1.3 73.1 1O0.0 4765 Figure 4.1 Trends in Contraceptive Use Among Currently Married Women 15-49 Any Method Any Modern Method Pill IUD Injection Female Sterilisation Periodic Abstinence ~\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \~ ~\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \~ 0 5 10 15 20 25 30 35 Percent using I mKCPS 1984 ~KDHS 1989 / Kenya DHS 1989 35 4.6 Current Use by Background Characterist ics Table 4.7 presents the relationship between the level of contraceptive use and background characteristics for currently married women. The percent of women using contraceptives is somewhat higher among urban women (31 percent) than rural women (26 percent). However, while use of modern methods is higher for urban women (26 percent) than for rural women (16 percent), the reverse is observed for traditional methods. The percentage using traditional methods among rural women (10 percent) is twice that of urban women (5 percent). As for provincial differentials, Eastern and Central Provinces have the highest level of current use (40 percent) followed by Nairobi (34 percent), Rift Valley (30 percent), and Coast (18 percent). Nyanza and Western Provinces (14 percent) lag behind, creating a three-fold differential in use by province from highest to lowest (Figure 4.2). The mix of methods also varies substantially by province. In most provinces, about 75 percent of current users are using modern methods, however, in Eastern Province, the figure is less than 50 percent. The pill is the most commonly used method in Nairobi, Coast and Western Provinces, the IUD in Central Province, female sterilisation in Nyanza, and periodic abstinence in Eastern and Rift Valley Provinccs. Figures are also presented in Table 4.7 for the rural areas of the 13 individual districts that were targetted in the sample design lbr the survey. The results should be viewed with caution since the number of women interviewed in each district is not large (see discussion of sampling errors in Appendix B). The data show that there is a nine-fold difference in contraceptive use by district, ranging from a high of 52 percent of married rural women in Kirinyaga District to a low of 6 percent of women in South Nyanza District. In addition to Kirinyaga District, Nyeri, Machakos, Meru and Murang'a Districts all show high levels of use, while Siaya, Bungoma, and Kilifi Districts have levels only slightly higher than South Nyanza. The level of use of modern methods in Meru District (34 percent) is identical to the level found for the Chogoria Hospital catchment area in 1985 (Chogoria Hospital, 1987). The method mix also varies; in some districts, periodic abstinence is the most widely used method, while in others it is the pill, female sterilisation, IUD, or injection. For example, it is clear that the high level of use of periodic abstinence in Eastern Province mentioned above is due almost entirely to the prominence of that method in Machakos District, not Meru District, the other targetted district from Eastern Province, where use of the pill and IUD predominate. Also notable is the high level of IUD use in Kirinyaga and Murang'a Districts and the extent of female sterilisation in Nyeri District. Regarding education levels, the major observation is that the percent of currently married women using any method increases directly with education. Use among women with secondary and higher education is more than double (40 percent) that of women with no education (18 percent). Current contraceptive use directly increases with the number of living children that a woman has, ranging from 5 percent among women with no children to 31 percent among those with four or more. Christian women are more likely to be using contraceptives than are Muslim women and women of other religious groups, or those with no religious affiliation. There is, however, slightly more current use among Protestant women (29 percent) than among Catholic women (26 percent). 36 Table 4.7 Percent distribution of currently married won~n by centracQptivemethod currently being used, according to background characterla~Ic$, Kenya, 1989 Contraceptive mQthod currently being used Weigh- Die- Female Peri- Not ted Background ~ny phragm/ star- Any odic With- curt- number character- Any modern InJec- foam/ Con- ilisa- trad'l absti- draw- en~ly of Istlcs method method Pill IUD tion Jelly dom tlon method nence al Oth@r using Total women Residence Urban 30.5 25.5 9.8 8.0 2°8 0.5 8.8 3.6 5.0 4.0 0.4 0.6 69.5 i00.0 748 Rural 26.2 16.4 4.3 2.9 3.4 0.4 0.4 4.9 9.8 8.i 0.2 1.4 73.8 i00.0 4018 Province Nalrobi 33.5 27.9 ii.9 7.9 2.3 1.2 0.4 4.4 5.6 4.0 0.8 0.8 66.5 I00.0 335 Central 39,5 30.8 9.1 i0.0 3.6 0.3 1.3 7.7 8.7 7.1 0.3 1.3 60.5 I00.0 648 Coast 18.1 14.8 5.5 1.7 3.6 0.i 0.3 3.6 3.3 3.0 0.3 0.0 91.9 i00.0 350 Eastern 40.2 19.5 5.9 4.7 3.5 8.4 0.4 4.5 20.8 17.9 0.3 2.5 59.8 I00.0 804 Nyanza 13.8 10.2 2.7 0.9 2.5 0.0 0.3 3.9 3.5 3.0 0.0 0.5 86.2 I00.0 872 Rift Valley 29.6 18.1 3.6 2.3 5.5 1.0 0.5 5.5 11.5 9.0 0.3 2.3 70.4 I00.0 1047 Western 13.7 i0.0 3.8 1.6 1.6 0.2 0.2 2.6 3.7 3.0 0.0 0.7 86.3 i00.0 711 Dis~riot (Rural) Kilifi 9.7 8.5 3.7 0.7 2.3 0.3 0.3 1.0 1.3 1.0 0.3 0.0 90.3 i00.0 500 Machakos 40.4 12,1 5.3 1,4 i.i 0.0 0,7 5.5 28.4 24.5 0.7 3.2 59.6 i00.0 282 Meru 36.3 34.2 12.4 8.3 5.7 1,6 0.5 5.7 2.1 2.1 0.5 0.0 63.7 i00.0 193 Nyerl 41,2 35.3 7.8 9.3 2.9 0.5 0.5 14.2 5.9 4.9 0.5 0.5 58.8 100.0 204 Muranga 31.3 24.2 2.8 i0.0 2.4 0.5 0.9 7.6 7.1 6.6 0.5 0.0 68.7 i00.0 211 Kirinyaga 52.2 44.2 12.4 18.6 8.0 0.4 0.9 4.0 8.0 7.1 0.4 0.4 47.8 i00.0 226 Kerlcho 23.2 15.2 3.4 0.9 5.3 0.4 0.0 5.3 8.0 6.8 0.0 1.1 78.8 i00.0 263 Oasln Cishu 14.9 I0.I 4.3 0.7 4.1 0.0 0.0 1.4 4.7 3.d 0.7 0.7 85.1 I00.0 148 South Nyanza 5.9 5.3 1.8 0.O O.0 0.0 0.0 1.5 2.6 1.8 0.0 0.7 94.1 100.0 272 Kisii 20.2 15.5 1.7 1.7 5.6 0.0 0.4 6.0 4.7 4.3 0.0 0.4 79.8 i00.0 253 siaya 8.8 5.6 0°6 O.0 1.3 0.0 1.3 2.5 5.i 2.5 0.0 0.6 91.2 i00.0 160 Ka kamega 14.5 10.2 3.2 0,3 2.2 0.5 0.3 3.8 4.1 3.9 0.0 0.3 85.7 i00.0 315 Bungoma 8.5 5.0 1.6 6.9 0.6 0.0 0.3 1.6 3.5 1.3 0.0 2.2 91.5 100.0 317 Education No education 18.3 9.7 2.1 1.5 2.2 0.1 0.3 3.7 8.6 6.9 0.0 1.7 81.7 I00.0 1506 Some primary 26.1 17.3 4.5 2.9 4.1 0.3 0.i 5.7 9.8 7.3 0°2 1.3 73.9 100.0 1462 Primary temp. 30.4 22.0 7.2 4.3 4.5 0.7 0.3 4.9 8.4 6.9 0.3 1.2 69.6 i00.0 987 Secondary + d0.4 29.3 10.2 9.3 2,6 1.0 1.7 4.5 II.i 9.8 0,7 0.7 59.6 I00.0 804 NO. of children None 4.7 0.8 0.6 0.0 0.2 0.0 0.0 0.0 5.8 3.4 0.4 0.O 95.3 i00.0 290 1 16.9 8.6 5.2 1.8 8.9 0.0 0.5 0.3 8.3 B.l 0.0 0.2 85.1 I00.0 497 2 24.2 16.0 6.7 4.d 1.6 0.2 1.2 1.9 8.2 7.2 0.i 0.9 75.8 i00.0 613 3 28.5 18.5 9.3 3.3 2.4 0.3 0.5 2.7 9.9 8.4 0.9 0.6 71.5 i00.0 649 d+ 31.4 21.7 4.3 4.4 4.8 0.7 0.4 7.1 9.7 7.7 0.I 1.9 68.6 100.0 2716 Religion Catholic 25.8 14.4 4.4 3.2 2.3 8.7 0.6 3.2 11.4 9.8 0.4 1o2 74.2 100.0 1656 Protestant 29.3 20.9 5.8 4.1 4.3 0.3 0.d 6.0 9.4 6.7 0.i 1.6 70.7 i00.0 2706 Muslim 16,7 13.9 4.6 2.9 2.7 0.0 1.0 2.6 2.8 2.6 0.2 0.0 83.3 I00.0 165 Other 20.8 15.9 5.5 8.1 0,0 0.0 0.8 1.6 4.9 3.3 1.6 0.0 79.2 i00.0 79 No religion 9.8 6.3 2.3 0.9 1.0 0.0 0.0 2.1 3.5 3.5 0.0 0.0 90.2 i00.0 151 To~al 26.9 17.9 5.2 3.7 3.3 0.4 0.5 4.7 9.0 7.5 0.2 1.3 73.1 I00.0 4765 Note: Excludes a few women not stated as to education and religion, since the sample within individual districts was self-welghtlng, n%Imbers of women in each district are unwQighted. 37 60" 50 40 30 20 10 0 / Percent Figure 4.2 Current Contraceptive Use by Province Currently Married Women 15-49 39.5 40.2 Nairobi Centrar Coast Eastern Nyanza Rift V. Western Province L ~ Traditional Methods I I Modern Methods i Kenya DHS 1989 4.7 Number of Children at First Use Table 4.8 shows the number of living children at the time of first use of contraception among ever-married women. Generally, younger women are starting to use contraception at lower parities than the older women did. For example, 19 percent of women 20-24 started using contraception after their first child, compared to only 4 percent of women 45-49. This probably reflects the fact that younger women are more likely to use contraception to space births, while older women use it to limit births. rabte 4.8 Percent distribution of ever-married women by number of living children at time of first use of contraception, ~ccording to current age, Kenya, 1989 Number of living children at time first used Wtd. Never no. of Age used None I 2 3 4+ Missing Total women 15-19 73.6 12.9 11.9 0.8 0.5 0.0 0.3 100.0 302 20-24 58.8 9.3 18.8 7.5 4.4 0.5 0.6 I00.0 901 25-29 53.9 3.7 15.6 10.7 6.9 7.9 1.2 100.0 1191 30-34 50.3 2.5 8.2 10.3 7.0 21 .I 0.5 100.0 928 35-39 49.9 2.7 6.2 6.5 8.0 24.9 1.6 100.0 869 40-44 55.}" 3.5 4.2 6.5 3.4 25.0 0.8 100.0 664 45-49 59.7 1.8 4.0 1.6 3.9 28.3 0.7 100.0 434 rotat 55.3 4.6 10.7 7.6 5.6 15.2 0.9 100.0 5289 38 4.8 Knowledge of Fert i le Per iod An elementary knowledge of reproductive physiology provides a useful background for successful practice of coital-related methods such as withdrawal, condom or barrier methods, but more so for periodic abstinence. Successful practice of periodic abstinence is dependent on a correct understanding of when in the ovulatory cycle a woman is most likely to conceive. Table 4.9 presents the distribution of all respondents and the small number of respondents who had ever used periodic abstinence by knowledge of the period in the ovulatory cycle when a woman is fertile. Table 4.9 Percent distr ibut ion of all women and women who have ever used periodic abstinence by knowledge of the fertile period during the ovulatory cycle, Kenya, 1989 Ever users of Air periodic Fer t i le period women abstinence During her period 1.2 0.9 Right after period has ended 40.8 46.8 Middle of the cycle 22.4 32.7 Just before period begins 9.0 10.5 At any time 5.3 3.1 Other 0.7 0.7 Don't know 20.5 5.2 Missing 0.2 0.1 Total 100.0 100.0 Number of women 7150 1386 Over 40 percent of the women interviewed said a woman is most likely to conceive just after her period has ended, while 21 percent did not know when a woman is likely to conceive and a small number (9 percent) identified the fertile time to be just before the period begins. Only 22 percent gave the "correct" responsc--that a woman was most likely to conceive in the middle of the cycle. Ever-users of periodic abstinence seem to be more knowledgeable about the ovulatory cycle, since 33 percent identificd the fertile time as occurring in the middle of the cycle (between two periods), and only 5 percent said they did not know. It should be noted that the response categories developed for this question are one attempt at dividing the ovulatory cycle into distinct periods. It is possible that women who gave an answer of, say, "one week after her period" were coded in the category "just after her period has ended," instead of in the category "in the middle of her cycle." Thus, women may actually have a more accurate understanding of their fertility cycles than is reflected in Table 4.9. 4.9 Sources for Contracept ive Methods Information on the source for contraceptive methods was obtained by asking women using modern methods where they obtained their methods the last time and by asking women relying on periodic abstinence where they received advice about the method. The results are presented in Table 4.10 and Figure 4.3. 39 Table 4.10 Percent distribution of current users of modern methods by most recent source of supply or information, according to specific method, Kenya, 1989 Total D iaph. / In- Total Female Source of supply foam/ jec - c l in i c s te r i t - Total supply methods P i t t Condom je t ty t ion methods IUD i sa t ion users Govt. hospital 46.1 44.4 27.8 53.4 50.7 67.4 58.9 75.4 55.7 Govt.ctiniclheatth centre 20.5 24.1 26.9 8.4 15.0 7.9 15.7 1.5 14.8 FPAK clinic 13.0 14.6 9.1 3.1 12.1 6.7 11.4 2.8 10.1 Other hospital/clinic 7.5 6.4 4.1 25.5 7.6 5.7 1.5 9.3 6.7 Mobile c l in i c 1.6 1.2 0.0 0.0 2.7 0.6 0.5 0.7 1.1 F ie ld educators 1.3 2.2 0.0 0.0 0.0 0.1 0.3 0.0 0.8 Pr ivate doctor 7.1 4.9 4.5 9.6 10.8 9.9 10.4 9.5 8.3 Pharmacy 1.4 1.0 17.6 0.0 0.0 0.0 0.0 0.0 0.8 Husband obta ins 0,4 0.0 7.7 0.0 0.0 0.1 0.3 0.0 0.3 F r iends / re la t ives " 0.1 0.0 0.0 0.0 0.3 0.7 0.0 0.0 0.4 Other 0.7 0.8 0.0 0.0 0.8 0.7 0.9 0.5 0.7 Missing 0.4 0.4 2.3 0.0 0.0 0.1 0.0 0.3 0.3 Total 100.0 100.0 100,0 100.0 100.0 100.0 100.0 100.0 100.0 Number of users 574 328 29 25 192 475 215 256 1048 Note: Total inc ludes 3 users of male s ter i t i sa t ion . Figure 4.3 Source of Family Planning Supply Current Users of Modern Methods FPAK 10% Pr ivate Source Other Hoep. /C l in i c 8% Other 2% nment 71% • Pr ivate doctor /pharmacy Kenya DHS 1989 According to Table 4.10, the most frequently mentioned source for both supply methods and clinic methods is the government hospital, which supplies 56 percent of all users. This is followed by government clinics and health centres, which supply 15 percent of users, and the Family Planning Association of Kenya (FPAK) clinics, which supply 10 percent of users. Nine percent of users obtain their methods from private doctors or pharmacies, while 7 percent depend on non- 40 governmental hospitals and"clinics, such as those run by private doctors' and church missions. Users of clinic methods, such as sterilisation and the IUD, are more likely to depend on government hospitals than users of supply methods, such as the pill and condom. Each current user of a modern method of family planning was askcd how much time it takes for her to get from her home to the place she obtained her method and whether she walks or uses some means of transport to get there. These same questions were also asked of nonusers and users of traditional methods. The results are shown in Table 4.11. Table 4.11 Percent dfstr~b~tion of current users of modern methods of family planning, nonusers of modern methods, and all women knowing a method, by time to reach source of supply and transport to source, according to urban-rural residence, Kenya, 1989 Current users of Nonusers of • All women who know a Time to source/ modern methods modern methods contraceptive method transport to source Urban Rural Total Urban Rural rotai Urban Rural Total Minutes to source "" 0-14 16.4 2.6 6.1 14.8 3.3 5.2 15.2 3.2 5.3 15-29 32.3 30.9 31.2 35.2 29.8 30.7 34.6 30.0 30.8 30-59 45.2 49.8 48.7 42.1 44.0 43.7 42.8 44.8 44.5 60 or more 3.4 15.8 12.7 3.9 19.4 16.8 3.8 18.9 16.2 Does not know 0.7 0.1 0.2 2.7 1.8 1.9 2.2 1.5 1.6 Not stated 2.0 0.7 1.0 1.2 1.8 1.7 1.4 1.6 1.6 Transport to source Walk 54.0 45.9 47.9 67.7 65.1 65.5 ' '64.5 62.3 62.7 Use transport 43.8 53.3 50.9 29.9 32.3 31.9 33.1 35.4 35.0 Does not know 2.2 0.8 1.1 2.5 2.6 2.6 2.4 2.4 2.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 ~umber of women 264 785 1048 889 4497 5386 1153 5281 6434 The results show that most (45 percent) women knowing a contraceptive method report that they are 30 to 60 minutes from a place they would go to or do go to for family planning services. A sizable proportion (31 percent) are 15 to 30 minutes from a family planning source. As expected, urban women are more likely to be closer to a source than rural women. However, there is surprisingly little difference between users and nonusers of modern methods in terms of distance from a family planning source. Regarding type of transport to reach family planning sources, roughly two-thirds of women knowing a method either walk to the source they use or say they would walk to a source if they were to use a method in the future; one-third say they would use transport. Users of modern methods are much more likely to use transport to get to their source than nonusers. Users are evenly split between those who walk and those who use transport, while nonusers of modern methods are more likely to say they would walk than use transport. 41 4.10 Att i tude Toward Pregnancy and Reason for Nonuse In the KDHS, nonpregnant women who were sexually active and who were not using any contraceptive method were asked their attitude toward becoming pregnant in the next few weeks. Table 4.12 presents information on the attitude toward becoming pregnant among these women. Sixty-two percent of nonusers exposed to the risk of pregnancy report that they would be unhappy if they got pregnant in the next few weeks, 31 percent say they would be happy, and 5 percent say it would not matter. The percentage who say that they would be unhappy increases with the number of living children, ranging from 43 percent among women with no children to 74 percent among those with 4 or more children. Table 4.12 Percent distribution of nonpregnant wo~en who are sexually active and who are not using any contraceptive method by attitude toward becoming pregnant in the next few weeks, according to number of Living children, Kenya, 1989 Attitude toward becoming pregnant in next few weeks Wtd. Nu~nber of WouLd Nun~oer Living not of ch i ld ren Happy Unhappy matter Missing TotaL women Hone 51.0 42.6 4.9 1.6 ,100.0 550 I 49.1 45.0 4.8 1.0 100.0 400 2 33.4 60.5 2.4 3.8 I00.0 403 3 33.7 60.4 4.3 1.6 100.0 378 4+ 17.5 73.8 5.1 3.7 100.0 1614 Table 4.13 examines the reasons for not using family planning Totat 30.5 62.1 4.6 2.8 100.0 3345 given by exposed nonusers who say that they would be unhappy if they became pregnant right away. Twenty-three percent of the women cite lack of knowledge as the primary reason they are not contracepting, 12 percent cite factors relating to access and availability, whereas 11 percent cite infrequent sex as the reason for not using contraceptives. A further 10 percent of these women say they are not using contraception because their husbands disapprove. It is interesting to note that health concerns and religious beliefs do not appear to be major obstacles to use of family planning. Differences by age are not large, except that older women are more likely than younger women to cite inconvenience as the reason for non-use. 4.12 Intent ion to Use in the Future Married women who were not using a contraceptive method at the time of the KDHS interview were asked if they thought that they would do something to keep from getting pregnant at any time in the future. Data obtained from this question are shown in Table 4.14. About 53 percent of nonusers intend to use a contraceptive method in the future, 12 percent are unsure, and 34 percent do not intend to do anything to avoid future pregnancy. The percentage of women intending to use is lowest for those with no children (41 percent), increases for those with one child (53 percent), and is highest for the women with 2 children (61 percent). The percentage decreases again for women who have 3 or more children. Table 4.15 presents information on method preferences for currently married nonusers who say they intend to use in the future. The most popular method is injection (37 percent), followed by the pill (24 percent), and female sterilisation (13 percent). 42 Table 4.13 Percent distribution of non-pregnant women who are sexua l ly ac t ive , not using any contracept ive method and who would be unhappy i f they became pregnant, by main reason for nonuse, according to age, Kenya, 1989 Main reason for nonuse Age <30 30+ Total Lack of knowledge 23.9 21.2 22.5 Opposed to family planning 3.7 3.7 3.7 Husband disapproves I0.9 8.8 9.8 Others disapprove 0.8 0.8 0.8 Infrequent sex 8.9 13.8 11.4 Postpartum/breastfeeding 1.4 0.7 1.0 Menopausal/subfecund 0.1 0+8 0.5 Heatth concerns 1.5 1.8 1.7 Access~availability 14.2 10.2 12.1 Costs too much 1.4 2.3 1.9 Fatalistic 0.8 1.7 1.2 Religion 7.0 3.1 5.0 Inconvenient to use 0.5 15.6 8.3 Other 18.1 9.9 13.9 Don't know 6.2 4.6 5.4 Missing 0.6 1.2 0.9 Total 100.0 100.0 100.0 Number 1011 1075 2086 Table 4.14 Percent distribution of currently married women who are not currently using any contraceptive method, by intention to use in the future, according to number of tiring children, Kenya, 1989 Intention Number of living chitdren* to use in fu ture None 1 2 3 4+ Total PLan fu ture use 41.1 53.2 60.7 55.7 51.4 53.2 Unsure about use 22.0 9.7 9.7 12.7 11.2 11.7 Does not intend 36.9 37.1 29.5 30.8 36.2 34.3 Missing 0.0 0.0 0.1 0.8 1.2 0.8 Total 100.0 I00.0 I00.0 100.0 100.0 100.0 Number 199 385 484 478 1871 3483 *Inctudes current pregnancy 4.13 Approval of Family Planning Table 4,16 shows responses to a question on whether women believe it acceptable to have family planning messages on the radio. The table shows that almost 90 percent of respondents believe that radio messages are acceptable. Over 80 percent of women in each age group find the idea acceptable, though women in their 20s are more likely than those in their 40s to accept radio messages. 43 Table 4.15 Percent d i s t r ibut ion of currentty married women who are not using a contraceptive method but who intend to use in the future, by preferred method, Kenya, 1989 Preferred method Percent Pitt 24.4 IUD 7.1 injections 37.0 Diaphragm/Foam/Jelly 0.4 Condom 0.9 Female steritisation 12.7 Male sterilisation 0.I Periodic abstinence 4.3 Other 1.8 Don't know 11.2 Missing 0.1 Total 100.0 Number 1852 Table 4.16 Percent distribution of all women by whether they feet it is acceptable to have family planning information presented on the radio, by age and background characteristics, Kenya, 1989 Age of woman Background characteristic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Urban 89.5 92.5 92.8 96.6 93.4 91.7 82.5 92.2 Rural 83.7 92.7 91.7 87.0 90.4 81.5 83.9 87.8 Bairobi 91.6 91.7 91.3 97.5 94.7 98.0 77.3 92.7 Central 86.3 95.3 93.6 94.4 91.7 92.8 94.9 92.2 Coast 63.6 86.1 81.8 81.0 71.0 56.2 58.2 74.4 Eastern 86.2 94.3 98.5 95.9 91.7 87.2 89.2 92.0 Nyanza 90.0 95.7 92.4 92.9 95.9 79.5 90.3 91.7 Rift Valley 86.4 91.0 91.4 81.5 89.3 80.3 83.7 87.1 Western 76.9 90.6 86.9 8O.5 95.8 76.5 67.4 83.0 No education 59.8 72.0 81.1 81.3 86.1 76.3 80.4 79.8 Some primary 78.4 93.3 93.2 90.4 92.8 85.3 87.7 88.9 Primary complete 86.2 92.6 94.6 95.6 94.9 95.6 96.1 91.0 Secondary + 93.8 97.3 95.3 94.2 97.1 96.8 (93.9) 95.6 Total 84.8 92.7 91.9 88.6 90.7 82.4 83.8 88.6 Note: Numbers in parentheses are based on fewer than 20 unweighted cases. Urban-rural differentials in the acceptability of family planning messages on radio are small. Variations by province are also small, except that women in Coast Province are less likely to find radio messages acceptable. The proportion who believe that radio messages on family planning are acceptable increases with educational attainment, from 80 percent of women with no education to 44 96 percent of women with secondary education. It is notable that high proportions of women in almost all categories find radio messages on family planning to be acceptable. This fact encourages increased use of radio for family planning messages. To obtain information about attitudes toward family planning, the respondents were asked whether they approved of couples using something to avoid pregnancy. Although all women were asked the question on approval, the 'analysis presented here is focused on currently married women and excludes those women who have never heard of a contraceptive method. Currently married women were further asked whether they thought that their husbands approved of the use of family planning. Table 4.17 presents information obtained from answers to these questions. Tabie 4.17 Percent distribution of currentiy married women knowing a contraceptive method by the husbar~d's and wife's attitude toward the use of family planning, Kenya, 1989 Wife's Husband's attitude toward ramify planning attitude toward fam- ily ptanning Disap- Ap- Don't proves proves know Missing Totat Number Disapproves 4.5 0.9 3.2 0.0 8.7 382 Approves 14.0 57.5 t6.4 0.3 88.2 3885 Missing 0.5 2.1 0.5 0.1 3.1 138 rotat 19.1 60.4 20.1 0.4 100.0 4405 Number 841 2661 888 16 4405 4405 Overall, 88 percent of married women say they approve of family planning, while 9 percent disapprove. Sixty percent say that their husbands approve of family planning, while 19 percent say their husbands disapprove, and 20 percent say they do not know their husband's attitude. According to the wife's report, only 58 percent of couples jointly approve of family planning, while 5 percent jointly disapprove. Approval of family planning is also discussed in Chapter 7, where the responses of husbands are compared with their wives' perceptions of their beliefs. Table 4.18 shows that there are few differentials in approval of family planning by married women or their husbands by age of the wife, urban-rural residence, or province, except that Coast province has the lowest percentage of women and husbands approving of family planning (78 and 45 percent respectively). Approval by women and their husbands increases with education of the woman. A good indication of the acceptability of family planning is the extent to which couples discuss the subject with each other. Table 4.19 indicates that one-third of currently married women had never talked about family planning with their husbands in the year preceding the survey. About one-third said that they discussed the subject once or twice with their husbands in the past year, while another one-third said they had discussed it more often. 45 Table 4.18 Percentage of currently married women knowing a contraceptive method who approve of family planning ar~J who say their husbar~d approves of family planning by background charscte~stics, Kenya, 1989 Background Woman Husband characteristics Approves Approves Total Age 15-19 87.8 53.8 241 20-24 89.3 61.5 784 25-29 90.3 62.1 1044 30-34 88.9 61.2 779 35-39 90,7 62.2 731 40-44 80.7 56.6 508 45-49 83.4 57.2 318 Residence Urban 90.6 65.4 716 Rural 87.7 59.4 3689 Province Nairobi 92.1 68.5 319 Central 92.0 69.9 628 Coast 77.7 44.8 323 Eastern 91.0 68.6 763 Nyanza 93.8 49.8 814 Rift Valley 81.1 63,9 910 Western 87.7 53.7 648 Education No education 81.4 45.4 1289 $om~ pr i~rv 89,7 59.1 1357 Primary co~plete 92.0 68,4 958 Secondary + 92.4 77.6 795 Total 88.2 60.4 4405 Note: Excludes a small number of women with education not stated. Table 4.19 Percent distribution of currently married women knowing a contraceptive method by number of times discussed family planning with husband, according to current age, Kenya, 1989 Nun~er of times discussed Wtd. nu~nber Once or More of Age Never twice often Missing Total women 15-19 42.1 32.0 25.9 0,0 100.0 241 20-24 30.0 36.0 33.4 0.6 100.0 784 25-29 29.0 33,0 37.7 0.3 100.0 1044 30-34 35,0 31.2 33,1 0.7 100.0 779 35-39 31,5 32.5 35.6 0.4 I00,0 T31 40-44 41.4 24.9 33.4 0,2 100.0 508 45-49 44.1 22,4 33.2 0.3 100.0 318 Total 33.9 31.4 34.3 0,4 100.0 4405 46 5 FERTILITY PREFERENCES In the KDHS, women were interviewed about their fertility preferences. The aim of this part of the interview was to establish the extent of unmet need for contraception and the number of unwanted or mistimed births. This information can be used to assist family planning programmes to carry out their services more effectively. The KDHS questionnaire included a number of questions about fertility preferences. All currently married women were asked if they wanted to have another child (after the current pregnancy if the woman was pregnant) and if so, they were asked how long they wanted to wait before having their next child. All women regardless of marital status were asked how many children they would like to have altogether, assuming they could go back to the time when they did not have any children ("ideal" number of children). Also, women with a birth in the five years before the survey were asked if, at the time they got pregnant, they wanted to have that child then, wait till later, or not have the child at all. 5.1 Desire for More Chi ldren Table 5.1 shows the desire for children among currently married women by the number of living children. Almost 50 percent of married women want no more children and 26 percent want another child, but only after two or more years (Figure 5.l). Thus, three-quarters of married women can be considered potential users of contraception for the purpose of either limiting their family size or spacing births. Table 5.1 Percent distribution of currently married women by desire for children, according to number of living children, Kenya, 1989 Number of living children* Desire for more children 0 I 2 3 4 5 6+ Total Want within 2 years 47.8 31.7 16.7 14.3 9.9 7.5 2.4 12.4 Want after 2+ years 14.0 55.6 49.0 42.7 29.8 15.9 6.1 26.4 Want, unsure when 21.5 5,7 3.3 Io7 2.4 0.4 0.9 2.9 Undecided 4.7 2.6 5.5 6.8 6.3 10.2 5.6 6.0 Want no more** 0.9 3.1 23.1 32.6 49.0 63.5 81.7 49.4 Declared infecund 9.3 1.3 2.2 1.1 2.5 1.6 3.3 2.6 Missing 1.8 0.0 0.1 0.8 0.2 0.8 0.1 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100,0 100.0 ~un~ber 213 468 633 663 648 507 1634 4765 * Includes current pregnancy "" Includes sterilised women The desire to limit childbearing appears to be considerably greater in Kenya than in other sub-Saharan countries where DHS surveys have been conducted. For example, the proportion of married women who want no more children is 33 percent in Botswana and Zimbabwe, 23 percent in Ghana and 19 percent in Uganda, compared to 49 percent in Kenya. This suggests that many women in Kenya may be candidates for more long-term methods of family planning, such as sterilisation or the IUD. 47 Figure 5.1 Fertility Preferences Currently Married Women 15-49 Want Wi th in 2 ¥r~ t~ Want, Unsure When ,3 Undec ided 6% In fecund/Miss ing 3% t After 2+ Yrs. 26% Want No More 49% Kenya DHS 1989 The desire for more children declines with the number of living children (see Figure 5.2). While more than 90 percent of marricd women with one child want another, only 9 percent of women with six or more children want another child. Conversely, the percentage of women who want no more children rises from 3 percent for women with one child to 82 percent for women with six or more children. This indicates substantial interest in limiting fertility among married women. The table also points to a desire among women to space births. For instance, 56 percent and 49 percent of women with one and two children respectively want their next births after two years. Table 5.2 shows the percent distribution of currently married women by desire for children according to age. The data show that the proportion of women who want no more children increases with age. Nine percent of the women aged 15-19 years want no more children, compared to 81 percent of the women aged 45-49 years. The proportion who want to delay their next birth declines with age, as does the proportion of women who want the next birth within two years. Table 5.3 shows the percentage of currently married women who want no more children by number of living children and selected background characteristics. In terms of fertility preference measures, the proportion of women who want no more children is the most significant figure. Therefore, it has been selected as an indicator for studying differentials in fertility preference by background characteristics of women. The proportion of women who want no more children is closely correlated with number of living children as well as background characteristics. For instance, overall, a larger proportion of rural than urban women want to stop childbearing, however, when the number of living children is taken into account, the reverse is true, which means that the overall figures result from the fact that a greater percentage of rural women have more children than urban women, since the proportion wanting no more children rises with the number of living children. 48 Percent 100 80 60 40 20 0 o Figure 5.2 Fertility Preferences by Number of Living Children 1 2 3 4 5 6+ Number of living children m Want No More ~ Want to Space ~ Want Soon i I Undecided ~ Infecund/Missing Kenya DHS 1989 Table 5.2 Percent distribution of currently married women by desire for children, according to age, Kenya, 1989 Age Desire for more children 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Want within 2 years 25,4 15.3 14,3 12.9 8.6 6.6 6.6 12.4 Want after 2+ years 53,8 55.2 35,9 17.8 11.8 2.0 0.8 26.4 Wants, unsure when 7.9 4.4 2.6 2.5 1.5 2.0 1.6 2,9 Undecided 3.3 6,1 6.8 8,4 6.5 4.7 1.0 6,0 Want no more* 9.3 18.3 39,3 56.0 67.0 78.4 81.4 49.4 Declared infeeund 0.2 0.4 0.5 2.2 3.9 6.1 8.5 2.6 Missing 0.0 0.2 0,6 0.2 0.6 0.1 0.1 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 I00.0 Number 276 827 1104 833 781 576 369 4765 * Includes sterilised women Central Province has the highest proportion of women who want to stop childbearing; half of the women with three children and 95 percent of the women with six or more children want to have no more. On the other hand, women in Coast Province seem to be the most pronatalist; only 55 percent of those with six or more children say they want to stop. The relationship between education and the desire to stop childbearing is somewhat erratic. The biggest differences are between women with no education and those with some education; the amount of education seems to have little effect on desire to stop childbearing. 49 Table 5.3 Percentage of currently married women who want no more children (including those sterilised) by ntanber of tiring children and background characteristics, Kenya, 1989 Nuldoer of living children* Background characteristic 0 I 2 3 4 5 6+ Total Residence Urban 2.5 5.4 31.1 47.1 62.9 67.4 85.3 39.6 Rural 0.0 2.1 20.7 29.0 46.3 62.9 81.5 51.2 Province Nairobi 6.1 6.1 34.2 55.8 63.5 79.5 89.3 43.7 Central 0.0 7.2 25.8 49.7 72.2 81.2 94.6 67.3 Coast 0.0 1.8 12.8 28.5 29.7 26.4 55.3 28.0 Eastern (0.0) 4.0 24.6 44.6 54.6 81.8 86.9 59.7 Nyanza 0.0 1.1 15.3 18.4 45.1 55.8 77.5 41.7 Rift Valley 0.0 1.2 28.5 24.5 42.2 56.8 81.5 49.7 Western 0.0 3.2 19.9 10.6 32.0 50.7 76.8 43.2 Education No education 1.2 0.9 26.0 25.8 41.4 52.5 77.3 54.4 Some primary 1.1 3.4 21.9 30.1 46.5 70.6 85.1 53.4 Primary cocaplete 0.0 3.7 20.3 41.6 52.2 68.4 86.5 46.2 Secondary + 1.6 3.4 25.2 35.0 58.1 65.2 84.6 36.5 Total 0.9 3.1 23.1 32.6 49.0 63.5 81.7 49.4 Note: Numbers in parentheses are based on fewer than 20 unweighted cases. Includes current pregnancy Table 5.4 examines the need for family planning among currently married women. Women are considered to be in need if they arc not contracepting and either want no more births or want to postpone their next birth. Overall, 60 percent of currently married Kenyan women are in need of family planning. Of these, 32 percent are in need because they do not want another child, while 28 percent are in need because they want to postpone their next birth. The proportion in need is slightly higher for rural women and women in Western Province. Need is also higher among women with less education. 5.2 Ideal Number of Children In order to assess fertility preferences in Kenya, all KDHS rcspondents regardless of marital status were asked: "(If you could go back to the time when you did not have any children) and if you could choose the number of children to have in your whole life. how many would that be?" Women with children were asked the entire question while those with no children were asked the part excluding the phrase in parentheses. This question aimed at two things--first, among women who have just started childbearing, the data will give an idea of the total numbcr of children these women will have in the future (to the extent that women are able to realise their fertility desires); secondly, among older, higher parity women, the data provide an idea of the level of unwanted fertility. It is important to note that some women have difficulty answering a hypothetical question of this type, especially women for whom control over fcrtility is not culturally acceptable. There 50 is also a possibility that some women report their actual number of children as their ideal since they find it difficult to admit that they would not want some of their children if they could choose again. Table 5.4 Percentage of currently married women who are in need of family planning by back- ground chBracteristics, Kenya, 1989 In need* and: Want Want to Number Background no post- of characteristics more pone** Total women Residence Urban 22.4 31.8 54.2 748 Rural 33,7 27.7 61,4 4018 Province Nairob~ 22,9 30.3 53.2 335 Central 37.9 16.5 54.3 648 Coast 18.4 36.8 55.2 350 Eastern 33.5 15.7 49.2 804 Nyanza 34.6 31.5 66.0 872 Rift Volley 31.0 30.5 61.5 1047 ~estern 33.8 41.4 75.1 771 Education No education 40.4 25,5 65.9 1506 Some primary 35.9 26,1 62.0 1462 Primary complete 25.5 32,4 57.9 987 Secondary + 16.9 32,9 49.8 804 Total 31.9 28,3 60.3 4765 * Includes women who are not using contraception and who either want no more children or want to postpone their next birth for 2 or more years ** Want next birth after two or more years Table 5.5 shows the percent distribution of sill women by ideal number of children and mean ideal number of children for all women and currently married women according to the number of living children. Four children is the most commonly reported ideal family size among all women; overall, 40 percent of women state four as their ideal number. This percentage is high, considering that another 30 percent consldcr five or more children as ideal. However, it is encouraging that while women with more living children are likely to state five or more as their ideal number of children, women with fewer children ~lre more likely to state two or three children as ideal. Thus, the mean ideal number of children increases with the number of living children. This may be due to the fact that women who want more children actually end up having them, or to the fact that women rationalise their family size by reporting their actual number of children as their ideal number. The KDHS data show a large decline in ideal family size from the 1984 KCPS results. The mean ideal number of children for all women was 5.8 in 1984, compared to 4.4 in 1989; for currently married women, the figures sire 6.3 in 1984 and 4.8 in 1989 (Central Bureau of Statistics, 1984, p.61). 51 Table 5.6 shows the mean ideal number of children for all women interviewed in the KDHS by age and selected background characteristics. The mean ideal number of children increases with age from 3.7 among women aged 15-19 to 5.5 among women aged 40-44, implying that if younger women succeed in having only those children they want, then fertility rates may fall in future. Table 5,5 Percent d i s t r ibut ion of a l l women by ideal number of ch i ld ren and mean ideal number of ch i ld ren for a l l women and cur rent ly marr ied women, according to number of l i v ing ch i ld ren , Kenya, 1989 Number of l i v ing ch i ld ren* Ideal number of ch i ld ren 0 1 2 3 4 5 6+ Total 0 0.3 0.3 0.0 0.2 0.2 0.0 0.0 0.1 I 1.3 2.1 1.0 0.9 0,5 0.5 0.3 0.9 2 20.3 13.0 15.7 4.7 9.1 7.0 3.2 10.7 3 17.0 20.8 11.4 20.7 5.2 6.8 5.8 ~2.¢ 4 40.6 42.7 52.2 38.6 45.1 28.0 36.3 40.3 5 9.4 9.2 8.0 12.5 12.0 21.7 8.2 10.5 6+ 8.1 8.4 9.5 19.0 24.4 31.8 40.3 21.1 Non-numeric response 3.1 3.5 2.2 3.2 3.6 4.2 6.0 3.9 ~otal 100.0 100.0 100.0 100.0 100.0 100.0 lO0.O 100.0 Nua~ber of women 1566 890 813 773 724 601 1783 7150 Mean (all women) 3.7 3.8 3.9 4.4 4.6 5.1 5.4 4.4 Mean (currently married) 4.4 4.1 4.1 4.5 4.6 5.2 5.5 4.8 Base (all women) 1516 858 796 749 698 576 1677 6870 Base (currently married) 197 443 616 641 624 488 1532 4540 * Includes current pregnancy Table 5.6 Mean ideal number of children for all women by age and background characteristics, Kenya, 1989 Age Background characteristics 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 3.5 3.6 3.8 3.8 4.5 4.2 4.4 3.8 Rural 3.8 4.1 4.6 5.0 5.0 5.6 5.3 4.6 Province Nairobi 3.3 3.5 3.6 3.7 4.1 3.9 4.2 3.6 Central 3.2 3.2 3.8 4.0 4.4 4.7 4.2 3.8 Coast 4.4 4.7 5.6 6.2 6.3 6.4 8.0 5.6 Eastern 3.5 3.7 4.0 4.4 4.7 5.3 4.7 4.2 Nyanza 3.9 4.3 4.6 5.0 4.9 5.5 4.8 4.6 Rift Valley 4.1 4.1 4.9 4.8 5.0 6.0 5.2 4.7 Western 4.0 4.4 4.6 5.2 5.2 5.9 7.5 4.9 Education No education 5.6 5.4 5.4 5.4 5.3 5.8 5.2 5.4 Some primary 3.9 4.1 4.6 4.7 4.9 5.2 5.5 4.6 Primary co~o[ ete 3.7 4.0 4.2 4.3 4.6 5.2 5.4 4.1 Secondary + 3.3 3.4 3.8 3.9 4.0 4.0 3.2 3.6 Total 3.7 3.9 4.4 4.8 4.9 5.5 5.3 4.4 52 The mean ideal number of children is higher lbr rural women than for urban women regardless of age. Coast Province has the highest average ideal family size (5.6), while Nairobi has the lowest (3.6). Also, women with no education have a higher mean ideal family size (5.4) than women with primary (4.1) or secondary or higher education (3.6). 5.3 Unwanted Fertil ity Table 5.7 shows the percent distribution of women who had a birth in the last twelve months by fertility planning status and birth order. Over 50 percent of the births in the last 12 months were either mistimed or unwanted, Forty-two percent of births were wanted at a later time (mistimed), while 11 percent were not wanted (unwanted). This indicates that a substantial proportion of women need family planning services, espccially for spacing births. Table 5.7 Percent distribution of women who had a birth in the Last 12 months by fertility planning status, according to birth order, Kenya, 1989 gir th order PLanning status of birth 1-2 3+ Total Wanted then 52.6 43,5 46.3 Wanted later 43.1 41,7 42.1 got wanted 3.9 14.2 II,0 Not classifiable 0.4 0,6 0.5 Total 100.0 100.0 100o0 Number 477 1062 1539 53 6 MORTALITY AND HEALTH 6.1 Chi ldhood Mortafity The government of Kenya has long been concerned about the high rates of infant and childhood mortality in the country and has made considerable efforts to reduce them. The infant mortality rate, especially, is often cited as a basic indicator of general health and welfare. In the KDHS, data on mortality were collected for the purpose of estimating infant and childhood mortality rates. This focus is a result of the fact that data appropriate for adult mortality estimation require very large samples and are difficult to collect by the retrospective household survey approach. In this section mortality rates are presented for three age intervals: Infant mortality--the probability of dying between birth and exact age one (lq0), Childhood mortality--the probability of dying between age one and age five(4q~), Under 5 mortality--the probability o[" dying between birth and exact age five (No). Mortality rates are calculated on a period basis (i.e., utilising information on deaths and exposure to mortality by age during a specific time period) rather than on a birth cohort basis. The pcriod approach is preferred for two reasons: first, period-specific rates are more appropriate for programme evaluation and second, the data necessary for the calculation of cohort-based childhood mortality rates are only partially available for the five-year period immediately preceding the survey. For a complete description of the methodology for computing period-spccific mortality probabilities, see Rutstein, 1984. Birth History Survivorship Data The data for the estimation of mortality rates were collected in the reproduction section of the individual woman questionnaire. The section began with questions about the aggregate childbearing experience of respondents (i.e., the number of sons and daughters who live in the household, who live elsewhere, and who died). Those questions were followed by a retrospective birth history in which data were obtained on sex, date of birth, survivorship status and current age or age at death of each of the respondents' live births. The data obtained from these questions are used to calculate infant and childhood mortality rates. A retrospective birth history, in which data are collected from respondents aged 15-49 as of the survey date is susceptible to truncation bias and other data collection errors. Truncation bias refers to the fact that for any time period prior to the year of survey, data are not available for women at the oldest ages of childbearing (e.g., for the period 10-15 years prior to the survey, there is no information about births to women aged 40-49). Other data collection errors involve underreporting of events, misreporting of age at death, and misreporting of date of birth. In general, all these data problems are less serious for time periods close to the survey date. 55 Mortality Levels and Trends 1974-1989 Table 6.1 and Figure 6.1 display infant and childhood mortality rates for the five-year period preceding the survey (1984-89) and for two previous five-year time periods (1974-78 and 1979-83). Table 6.1 Infant and childhood mortality rates by five-year calendar periods, Kenya, 1989 Infant Childhood Under 5 mortality mortality mortality rate rate rate Period (lqO) (4qi) (SqO) 1984-1989" 59.6 31.5 89.2 1979-1983 57.6 37.8 93.1 1974-1978 64.1 44.2 105.5 Percent decline 1974-78 to 1984o89 7,0 28.7 15.4 * Includes calendar year 1989 up to the month preceding date of interview. 120 100 80 Figure 6.1 Trends in Infant and Child Mortality Rate per 1,000 64 ~ RQ Infant Mor ta l i ty lm 44 1974-78 Child Mor ta l i ty 1979-83 Under 5 Mor ta l i ty 1984-89 Kenya DHS 1989 56 The infant mortality rate for Kenya for the period 1984-89 is 60 per thousand live births and the childhood mortality rate is 32 per thousand. The overall probability of dying between birth and exact age five is 89 per thousand. While the KDHS rates indicate a decline in mortality, it is important to note that the decline is small. During the ten-year interval between 1974-78 and 1984-89, infant mortality declined by only 7 percent, childhood mortality by 29 percent and the overall probability of dying between birth and age five, by 15 percent. When KDHS rates are compared to data from previous sources, they imply a substantial decline in infant and childhood mortality. For example, the infant mortality rate reported in the 1977/78 KFS was 96 per thousand births (Central Bureau of Statistics, 1980, p.105) and the rate estimated from 1979 census data was 104 per thousand (Central Bureau of Statistics, no date, p.103). This magnitude of decline is large but certainly possible. Howevcr, the fact that the infant mortality rate from the KDHS for the period 1974-78 (64) is also much lower than the rate in either the 1979 census or the 1977/78 KFS, suggests that children who died might have been underreported in the KDHS. An investigation of this possibility is beyond the scope of this report. Mortality DifferentiaLs 1979-89 Mortality differentials by province:, mothcrs' level of cducation and urban-rural residence are presented in Table 6.2. In order to have a sufficient number of births to calcuIate reliable rates for the study of mortality differentials across population sub-groups, period-specific rates are presented for the ten-year period 1979-1989. Table 6.2 Infant and childhood mortality rates by background characteristics of the mother for the ten-year period preceding the survey, Kenya, 1989 Infant Childhood Under 5 mortality mortality mortality rate rate rate Background (lqO) (4qi) (5qO) characteristics 1979-89 1979-89 1979-89 Residence Urban 56.8 34.2 89.0 Rural 58.9 34.3 91.2 Region Nairobi 46.3 35.T 80.4 Central 37.4 10.0 47.0 Coast 107.3 54.5 156.0 Eastern 43.1 22.2 64.3 Nyanza 94.2 60.0 148.5 Rift Valley 34.6 16.9 50.9 Western 74.6 62.9 132.8 Education None 71.7 39.9 108.7 Some prLmary 59.1 38.3 95.2 Primary co~tete 49.3 24.4 72.5 Secondary + 41.8 23.4 64.2 Total 58.6 34.3 90.9 Note: Rates include calendar year 1989 up to the month preceding date of interview. 57 Curiously, mortality is only slightly higher in rural areas than in urban areas. The provincial rates display marked differentials. The infant mortality rate is highest for Coast Province (107 per thousand live births), followed by Nyanza (94), Western (75), Nairobi (46), Eastern (43) and Central (37) (Figure 6.2). Rift Valley has the lowest infant mortality rate (35). Childhood mortality differentials are even larger, with the ratcs in Western and Nyanza Provinces (63 and 60, respectively) being six times the rate in Central Province (10). PROVINCE Nairobi Centra l Coast Eastern Nyanza Rift V. Western EDUCATION None Some Pri. Pri. Comp. See.÷ Figure 6.2 Infant Mortal ity by Province and Education 46 37 107 J 75 ,I 94 ,. ) 49 42 , i 72 ,, 59 0 20 40 60 80 100 120 Rate per 1,000 Kenya DHS 1989 Mortality differentials by mother's level of formal education display expected differentials. Mortality is highest for children whose mothers have no education, declines for children whose mothers have some primary education and is lowest for children whose mothers have attained secondary education and above. Mortality differentials by sex, mother's age at birth, birth order, and length of the previous birth interval are shown in Table 6.3. As expected, mortality rates are lower for females than for males. Infant mortality differentials by age of mother are moderate, but show higher levels for children born to mothers under age 20. Childhood mortality declines steeply as age of the mother increases. Infant mortality estimates by birth order also display the expected differentials. Infant mortality is higher for first births (65 per thousand), declines for second and third births (55) and births 4-6 (50), then rises sharply for births 7 and above (72). The length of birth intervals also has a strong effect on infant and child mortality levels. The infant mortality rate estimates are 76 per thousand for births occurring after intervals of less than 2 years, 48 per thousand for births after intervals of 2-3 years and 36 per thousand for births 58 Table 6.3 Infant and childhood mor ta l i ty rates by selected demographic character is t ics , fo r the ten-year period preceding the survey, Kenya, 1989 Infant Chitdhood Under 5 morta l i ty mor ta l i ty mor ta l i ty rate rate rate Demographic (lqO) (4q l ) (5qO) character i s t i cs 1979-89 1979-89 1979-89 Sex of ch i ld Mate 63.0 35.4 96.1 Female 54.3 33.2 85.7 Age of rm)ther at birth Less than 20 67.5 43.8 108.3 20-29 54.8 35.8 88.6 30-39 60.2 26.4 85.0 40-49 58.3 15.0 72.5 Birth order F i rs t 65.3 37.5 100.3 2-3 54.8 32.5 85.5 4-6 49.7 33.4 ~ 81.5 7+ 71.9 36.4 105.6 Previous b i r th in terva l <2 years 75.6 41.1 113.6 2-3 years 47.7 32.6 78.7 4 years or more 35.9 17.9 53.2 Note: Rates include calendar year 1988 up to the month preceding date of interview. after intervals of 4 years or more. There are also substantial differentials in childhood mortality by length of the preceding birth interval, in the same direction as the infant mortality differentials. These differentials suggest that a change in birth spacing practices would by itself, have a favourable impact on mortality levels. Additional evidence regarding childhood mortality levels in Kenya can be obtained from the proportion of children ever born who have died, tabulated by age of woman (Table 6.4). Just over 10 percent of all children born to women 15-49 have died. The proportion dead by age of mother shows an unusual pattern; it is very high for women 15-19 and falls for women 20-24 and 25-29, before showing the expected increase with age of mother. 6.2 Matern i ty Care The health care that a mother receives during pregnancy and at the time of delivery is important to the survival and well-being of the child as well as the mother. To obtain data on the type of maternity care that Kenyan women receive, KDHS respondents who had given birth in the five years preceding the interview were asked if they had seen anyone for an ante-natal checkup before the birth and if anyone had assisted with delivery of that child. If they had had an ante- natal checkup or received assistance at delivery, they were asked who provided the care.' In cases where the maternity care was received from more than one provider, the most qualified provider was recorded by the interviewer. 59 Table 6.4 Mean number of ch i ld ren ever born, surv iv ing , and dead, and proportion of chitdren dead among those born, by age of wonmn, Kenya, 1989 Mean number of children: ~td. number of wc41)en Propor- Ever Sur- t ion Age born r iv ing Dead dead 15-19 0.28 0.25 0.03 0.117 1497 20-24 1.58 1.44 0.14 0.088 1321 25-29 3.47 3,19 0.28 0.082 1334 30-34 5.01 4.49 0.52 0.104 981 35-39 6.48 5.80 0.67 0.104 898 40-44 7.36 6.53 0.84 0.114 674 45-49 7.63 6.55 1.08 0.142 445 Total 3,67 3.28 0.39 0.106 7150 Since neonatal tetanus has been shown to be a major cause of infant deaths in developing countries like Kenya, mothers were also asked if they had received an injection before the birth to keep the baby from getting tetanus. The responses to this question are affected by the mother's recall of events during pregnancy and, particularly by her ability to distinguish the tetanus toxoid vaccination from other injections she may have received. Moreover, the failure of a respondent to be immunised against tetanus during any particular pregnancy does not necessarily mean that the mother and child were exposed to the risk of tetanus, since protection may have been provided by tetanus toxoid vaccinations before that pregnancy. Despite these drawbacks, the proportion of women receiving a tetanus toxoid vaccination during pregnancy provides an indicator of the success of maternal and child health efforts. Table 6.5 presents data on the type of ante-natal care obtained for births that occurred in the five years before the survey. The results suggest that the majority of mothers in Kenya receive at least some maternity care. For 77 percent of births, mothers had seen a doctor or trained nurse/midwife to check the pregnancy and for 89 percent of births, mothers had had a tetanus toxoid injection. Tetanus toxoid coverage may be overrcported, since it seems doubtful that in 12 percent of cases, mothers received a tetanus injection without obtaining any other ante-natal care, however, the level of 89 percent is close to the rate of 83 percent of women reported in the 1987 national coverage survey for the Kenya Expanded Programme on Immunisation (KEPI) as having a tetanus injection during either of their last two pregnancies (Ministry of Health, 1987). The authors of the KEPI study believed that respondents in their survey had confused other injections for tetanus toxoid. There are few differences in ante-natal care by age or residence of mother. By province, ante-natal care from either a doctor, trained nurse, or midwife is most prevalent for births to women interviewed in Nyanza Province and Nairobi and least prevalent among births to women interviewed in Central and Coast Provinces (69 percent). The rates for Nairobi and Coast Province are in the expected direction, however, that the rate for Nyanza is higher than the rate for Central Province is unusual. About one-quarter of births to women in Coast, Central and Western Provinces do not receive ante-natal care. Better educated women are slightly more likely than less educated women to obtain ante-natal care from trained professionals. 60 Table 6.5 Percent d i s t r ibut ion of b i r ths in the las t 5 years by type of ante -nata l care fo r the mother and percentage of b i r ths whose mother received a tetanus toxoid in jec t ion , according to background character i s t i cs , Kenya, 1989 Type of ante-natal care Percentage receiving tetanus Number toxoid of injection births Trained Trad'l Background 0oc- nurse/ birth characteristics tor midwife attend. Other None Missing Totat Age of mother <30 28.8 49.4 1.7 0.7 18.4 0.9 I00,0 89.3 4081 30+ 27.9 47.8 2.2 0.2 21.0 0.8 I00,0 88,0 2969 Residence Urban 28.5 53.1 0.9 0.8 16.1 0.6 100.0 92.2 979 Rural 28.4 48.1 2.1 0.5 20.0 0.9 100.0 88.2 6072 Province Nairobi 27.5 55.8 0.6 1.5 13.9 0.6 100.0 90.3 417 Central 52,7 16.2 0.4 1.0 28.8 0.8 100.0 89.9 969 Coast 35.6 33.7 0.4 0.1 29.9 0.3 100.0 89.1 423 Eastern 31.5 48.9 0.9 0.7 17.6 0.4 100.0 88.4 1233 Nyanza 22,9 60.6 1.2 0,3 13.4 1.5 100.0 90.7 1283 Rift Valley 25.7 53.6 4.6 0.0 15.6 0.5 100.0 86.4 1593 Western 12.2 59.4 2.4 0.5 24.1 1.4 100.0 88.5 1133 Education NO education 24.3 48.0 4.4 0.2 22.3 0.9 I00.0 84,8 1888 Some primary 26.2 49.9 1,3 0.3 21.4 0,9 100.0 88,9 2234 Primary co(nplete 33.3 46.6 0.8 0.9 17.5 1.0 100.0 90.0 1655 Secondary + 32.3 50.6 0,7 0,9 14.8 0.7 100.0 92.7 1268 Total 28.4 48.8 1.9 0.5 19.5 0.9 100.0 88.7 7050 Table 6.6 presents data on the type of assistance mothers received at delivery for all births in the five years before the survey. Half of the births in the last five years were assisted at delivery by a doctor or trained nurse/midwife and 14 percent by a traditional birth attendant. A substantial proportion of births were assisted by relatives and friends of the mother (21 percent) or by no one (12 percent). Births to older women, rural women, and women with no education are less likely to benefit from assistance at delivery by trained medical personnel. The results also show that Nairobi leads in maternity care, with over 80 percent of births being assisted by a doctor or trained nurse/midwife, followed by Central Province, where 73 percent of births are assisted by professionals. Presumably, this is because Nairobi is an urban area where medical facilities are more available. Also notable are the higher proportion of births in Central Province that are assisted by doctors (35 percent), the higher proportion of births in Coast Province that are assisted by relatives and friends (44 percent), and the higher proportion of births in Western Province that do not benefit from any assistance at delivery (31 percent). It is also important to note that traditional birth attendants play a more significant role in delivering babies in Rift Valley, Eastern, and Nyanza Provinces. An important indicator of maternal and child health is the proportion of Women and recent births that fall into certain high risk categories. It has been shown that the risk of serious illness and/or death for both mother and child is related to the age and parity of the mother, as well as 61 Table 6.6 Percent d i s t r ibut ion of b i r ths in the last 5 years by type of assistance dur ing de l ivery , according to background character is t ics , Kenya, 1989 Type of assistance at de l ivery Trained Trad'[ Rela- Number Background Doc- nurse/ b i r th t i re / Miss- of character is t ics tor midwife attend, f r iend Other None ing Total b i r ths Age of mother <30 17.9 37.5 14.2 20.2 1.7 7.5 0,9 100.0 4081 30+ 14,4 28.3 14,4 22.8 1,9 17.5 0.7 100.0 2969 Residence Urban 23,1 54.4 5.0 9,9 1.0 6.0 0.6 100.0 979 Rural 15.4 30,3 15,8 23.1 1.9 12.7 0.9 100.0 6072 Province Nairobi 19.8 63.4 2.5 8.8 1.1 4.2 0.3 100.0 417 Central 34.9 38.4 5.9 12,0 2.3 5.6 0.9 100.0 969 Coast 13.7 27.2 4.7 44.1 1.2 8.4 0.6 100.0 423 Eastern 12.8 28,0 19.6 29.1 1,8 8.3 0.4 100.0 1233 Nyanza 14.4 39.4 17,4 14.9 1.8 10.6 1.5 100.0 1283 Rift Valley 16.8 27.9 20,8 23,4 2.3 8.0 0.7 I00,0 1593 Western 6.2 28.6 10.8 21.1 0.9 31,3 1.1 100.0 1133 Education No education 9.5 24.0 17.2 27.6 1.6 19,1 1.0 100.0 1888 Some primary 14,5 30.5 15,2 24,5 1.9 12.7 0.7 100.0 2234 Primary complete 19.5 34.9 15.9 18.6 2.4 7.8 0.9 I00,0 1655 Secondary + 26.3 51.9 6.2 9.7 0.9 4.2 0.8 100.0 1268 Tota[ 16.4 33.6 14.3 21.3 1,8 11.7 0.8 100.0 7050 to the interval between births. Risk is higher for births to younger (under age 18) and older (age 35 or over) mothers, those who have had a prior birth recently (within the previous 24 months), and those of higher parity (four or more births). Table 6.7 indicates that 85 percent of currently married women fall into at least one of the high health risk categories and over half fall into two or more categories. Most married women have had 4 or more births, and almost half have had a birth in the past 24 months. With regard to recent births, two-thirds fall into one category and 28 percent fall into two or more categories. Over half (56 percent) of Kenyan births occur to high parity mothers, while one-fifth are born less. than 24 months after a previous sibling. 6.3 Child Health Indicators The KDHS included a series of questions intended to provide information on immunisation coverage and on the occurrence and treatment of diarrhoea, fever and respiratory illness among children under age five. Strictly speaking, these data do not represent all children under five in Kenya, but only those children of women who were interviewed in the KDHS. Thus, no information was obtained for children of women who had died, who were institutionalised, or who, for some other reason were not interviewed in the survey. Although the immunisation status and the morbidity experience of the latter children are likely to differ from that of children whose mothers were interviewed, their numbers are not large, so the results presented below can be considered as generally describing the health status of children under five years of a~e in Kenya. 62 Table 6.7 Percentage of currently married women and births in the 12 months prior to the survey to women who faLL in various categories of high health risk, Kenya, 1989 Currently Births Health risk married in past category woc~en 12 months Under age 18 1.4 2.9 Age 35 oP older 36.2 18.2 Last birth occurred within past 24 months 48.3 20.5 Four b i r ths or more 61.5 55.8 In at Least one category 85.2 66.4 In 2 or more categories 52.1 28.1 Weighted number 4765 1484 lmmunisation of Children In the KDHS, women who had children under the age of five were asked if the children had health cards. If the health card was available, the interviewers copied from the card the dates on which the child had received immunisations against the following diseases: tuberculosis (BCG); diphtheria, whooping cough (pertussis) and tetanus (DPT); polio; and measles. If the child had no card or the interviewer was not able to examine the card, the mother was asked if the child had ever received a vaccination. However, no information was obtained on specific vaccinations for these children because of doubts about the reliability of the mother's rccall. In examining these data, it should be borne in mind that as of January 1986, the Kenya Expanded Programme of Immunisation (KEPI) recommended that children be immunised according to the following schedule (Ministry of Health, 1987, p.20): Age Immunisation Birth BCG, polio 6 weeks DPT, polio 10 weeks DPT, polio 14 weeks DPT, polio 9 months measles The data in Table 6.8 indicate that immunisation cards were seen for 50 percent of all the children under age five. The proportion of children with health cards seen is highest for children 6-11 months of age. Of children with cards, almost all had received at least one immunisation. This is not surprising since one of the major reasons lbr issuing a health card is to record immunisations. Forty-three percent of children did not have a health card available, but were reported by their mothers to have been immunised. The information on specific immunisations collected for children with health cards is also presented in Table 6.8. In interpreting the data in the table, it is important to bear in mind that the figures are based on children whose health cards were seen by the interviewers. Thus, the 63 Table 6.8 Among a l l ch i ld ren under 5 years of age, the percentage with hea l th cards seen by interv iewer, the percentage who are immunised as recorded on a health card or as reported by the mother and, among ch i ldren with hea l th cards, the percentage fo r whom BCG, DPT, po l io and measles immunisations are recorded on the health card, by age, Kenya, 1989 Among ch i ldren under 5 Among ch i ldren under 5 with health cards seen, the percentage wi th : the percentage who have received: Num- Ui tb Some [ramun. per hea l th irnmun- reported of Age in cards isat ion by DPT DPT DPT Pol io Pol io Pol io Meas- ch i t - months seen on card mother BCG I 2 3÷ I 2 3+ tes ALL* dren ,6 52.7 51.7 2a.4 93.3 62.1 55.5 26.9 89.5 67.2 41.5 4.G 1.8 601 6-11 67.5 67.3 26.2 96.7 98.6 91.5 82.6 99.0 94.0 85.8 25.7 23.1 710 12-17 60.3 59.9 35.2 9S.5 98.4 94.3 88.6 99.4 94.3 91.3 77.0 71.0 703 18-23 61.9 61.8 34.9 98.1 99.4 98.1 93.0 99.2 97.4 93.5 79.1 74.8 612 24-59 43.7 43.4 51.5 96.4 98.0 93.0 86.6 97.4 92.8 86.3 81.0 72.9 3889 Total 50.6 50.3 43.3 96.2 96.8 90.0 81.5 97.3 91.2 83.4 64.8 58.8 6514 * BeG, at Least 3 doses of DPI and poLio, and measles results cannot be interpreted as coverage rates for the entire population of children of that age, but rather, should be viewed as providing measures of drop-out rates, since virtually all children with cards received at least one immunisation. The KDHS found that among children aged 1-5 years for whom health cards were available, more than 95 percent had received a BCG vaccination and at least one dose of DPT and polio. Almost all of those who have the first dose of DPT and polio receive the second and third doses, however, only about 80 percent of children aged 1-5 with health cards have been immunised against measles. Since it is customary to report immunisation coverage based on one-year olds, Table 6.9 presents data on the proportion of children 12-23 months with cards who have received specific immunisations, according to selected background characteristics. The data show that there is a slight difference in immunisation coverage for boys and girls--76 percent of girls whose cards were available bad received all immunisations, compared to 70 percent of boys. This differential is due almost entirely to the differential in measles coverage. Rural and urban differentials on immunisation coverage are modest, with the urban children having higher coverage than their rural counterparts. Rural children are also more likely to have health cards available to show the interviewer. There seem to be some marked variations in coverage by province, with Central Province having the highest proportion fully immunised (88 percent), and Western Province the lowest (57 percent). There is also a much steeper drop-out rate between the three doses of DPT and polio among children in Western Province than for children in other provinces. Considering differentials by educational status of the child's mother, full immunisation coverage is much higher among children whose mothers have attained secondary education (86 percent) than for those whose mothers have no education (55 percent). Estimates of coverage for all children, including those whose health cards were not seen, can be derived by multiplying the proportion of children with particular immunisations recorded on health cards by the proportion of children whose health cards were seen. For example, 64 Table 6.9 Among a l l ch i ld ren aged 12-23 months, the percentage with health cards seen by interv iewer, the percentage who are immunised as recorded on a health card or as reported by the mother and, among ch i ldren with health cards, the percentage for whofn BCG, DPT, po l io and measles immunisations are recorded on the health card, by background character is t ics , Kenya, 1989 Among ch i ldren 12-23 Among ch i ldren 12-23 months with health cards seen, months, percent w i th : the percent who have received: Back- With Some Immun. No. ground health immun, reported of character- card on by DPT DPT DPT Pol io Pol io Pol io Meas- ch i t - istics seen card mother BCG I 2 3+ I 2 3+ les ALl* dren Sex Male 63.4 63.2 32.7 96.9 98.7 96.6 90.3 99.4 96.2 92.6 75.8 70.0 653 Female 58.7 58.4 37.5 96.5 99.1 95.4 91.0 99.2 95.3 92.1 80.3 75.9 662 Residence Urban 49.5 48.9 46.9 96.7 98.7 98.7 94.7 98.7 98.0 94.7 86.1 82.1 197 Rural 63.1 62.9 33.0 96.7 98.9 95.7 90.I 99.4 95.5 92.0 76.8 71.5 1118 Province Nairobi 47.9 46.5 47.9 92.8 97.1 97.1 94.2 97.1 95.7 94.2 85.5 79.7 93 Central 61.0 61.0 36.7 95.6 I00.0 99.4 98.2 00.0 100.0 97.9 93.6 87.7 203 Coast 66.2 65.7 32.0 96.4 99.3 97.1 85.6 99.3 97.1 93.6 71.6 68.7 73 Eastern 73.I 73.1 24.1 97.4 100.0 98.6 92.4 00.0 98.6 96.8 82.0 79.4 241 Nyanza 55.0 55.0 39.7 97.8 98.6 95.2 91.7 99.3 96.6 93.3 67.4 64.8 226 Rift V. 62.3 62.3 33.9 97.8 97.8 94.4 89.5 99.8 94.6 90.8 77.2 70.7 290 Western 55.5 54.6 38.4 95.5 98.3 90.8 80.6 97.4 86.5 783 66.2 56.5 189 Education None 53.0 52.9 38.1 96.3 98.8 91.6 77.0 98.7 92.7 86.9 58.4 55.1 306 Some prim. 65.3 64.7 30.8 94.7 98.8 96.3 91.2 99.4 95.5 91.8 76.8 70.5 392 Prm. cornp. 66.5 66.5 32.5 98.0 100.0 98.8 96.0 99.8 97.8 95.0 84.7 79.4 355 Second.+ 56.7 56.5 41.2 98.6 97.4 96.3 95.4 99.0 96.3 95.2 90.7 85.8 259 Total 61.0 60.8 35.1 96.7 98.9 96.1 90.7 99.3 95.8 92.4 78.0 72.8 1315 * BCG, at least 3 doses of OPT and polio, and measles multiplying the 73 percent of children 12-23 months who are fully immunised according to their health cards by the 61 percent who produced health cards for the interviewer gives an estimate of 44 percent of all children 12-23 months who are fully immunised. This compares closely with the estimate of 41 percent fully immunised according to cards from the Kenya Expanded Programme of Immunisation (KEPI) survey (Ministry of Health, 1987). These are minimum estimates of coverage, since they assume that all children without cards have not received any immunisations. If one assumes that all children without cards whose mothers say they have received some immunisation(s) have received the same immunisations as those with cards, the estimate in the KDHS increases to 70 percent fully immunised among children 12-23 months. This is probably on the high side and the true coverage is most likely between 44 and 70 percent. In the KEPI survey, information on specific immunisations received was asked of the mothers of children without cards; using this information, the proportion of children 12-23 months fully immunised was 51 percent. Child Morbidity and Treatment In addition to the immunisation data, information was collected for all children under age five on the occurrence of diarrhoea, fever and respiratory illness in the weeks preceding the interview and treatment provided for children experiencing these illnesses. The data on diarrhoea, 65 lever and respiratory illness cannot be used to measure incidence of these ailments. However, they provide a basis for a period prevalence estimate for each illness, i.e., the percentage of children under 5 years whose mothers report that they had the illness in question during the weeks preceding the survey. In considering the morbidity information, it is important to remember that the measures are influenced by the mother's subjective evaluation of whether the child experienced the illness in question. For example, the question on diarrhoea simply asked the mother if the child had diarrhoea during the last 24 hours or two weeks. The responses to the question are clearly dependent on what the mother understood by the term diarrhoea and thus there may be considerable variation in the length and severity of the diarrhoea episodes reported in response to the question. The morbidity measures are also affected by the reliability of the mother's recall as to when the episode of the illness in question occurred. Both the failure to report illness occurring within the reference period (two weeks for diarrhoea and four weeks for fever and cough) and the reporting of episodes that occurred prior to the period affect the accuracy of the prevalence estimate. In interpreting the morbidity data, it should be kept in mind that the majority of interviews took place during the dry season, when the number of cases of illness in question--diarrhoea, fever and respiratory problems--would be expected to be somewhat lower than at other times of the year. Table 6.10 Among children under 5 years of age, the percentage reported by the mother to have had diarrhoea in the past 24 hours and the past two weeks, by background characteristics, Kenya, 1989 Percent of children under No. 5 with diarrhoea in: of chit - Background Past Past dren character- 24 two under istics hours weeks 5 Age in months <6 11.0 18.0 601 6-11 12.9 25.3 710 12-17 13.6 25.6 703 18-23 9.2 18.3 612 24-59 3.2 6.4 3889 Sex Male 6.8 12.9 3210 Female 6.5 12.6 3305 Residence Urban 5.4 10.8 899 Rural 6.9 13.1 5615 Province Nairobi 7.3 13.0 386 Central 4.7 10.0 927 Coast 4.3 10.1 378 Eastern 7.8 15.1 1174 Nyanza 7.0 15.5 1106 Rift Valley 4.6 7.4 1533 Western 10.5 18.6 1011 Education No education 7.5 13.2 1725 Some primary 6.9 12.9 2043 Primary comp. 6.7 12.7 1546 Secondary + 5,0 11.9 1194 Total 6.7 12.7 6514 Diarrhoea Table 6.10 shows the percentage of childrcn under age five reported as having had diarrhoea in the two weeks preceding the survey, whereas Table 6.11 shows the kind of treatment they received. Seven percent of children under five were reported to have had diarrhoea in the 24 hours before the survey and 13 percent were reported to have had diarrhoea in the two weeks before the survey. Diarrhoea prevalence varies with the age of the child; the rates are greatest for children aged between 6 and 17 months (when weaning usually takes place), whereas it is lowest for children 24 months or older. The sex differential is insignificant--13 percent of both boys and girls are reported as having had diarrhoea in the previous two weeks. Diarrhoea prevalence is slightly higher for rural than urban children. By province, diarrhoea prevalence is 66 highest for Western Province, followed by Nyanza and Eastern Provinces, in that order; it is lowest for Rift Valley Province. There are no substantial differences in prevalence of diarrhoea by education of mother. For the children who had an episode of diarrhoea in the two weeks preceding the survey, Table 6.11 indicates what, if anything, mothers did to treat the diarrhoea. About 47 percent consulted medical personnel, 21 percent used ORS packets, 49 percent used a homemade rehydration solution, and 84 percent used other treatment. It is important to note that only 10 percent did nothing to control the diarrhoea. Table 6.11 Among children under 5 years of age who had diarrhoea in the post two weeks, the percentage consulting a medical fac i l i ty , the percentage receiving dif ferent treatments as reported by the mother, and the percentage not consulting a medical fac i l i ty and not receiving treatment, according to background characteristics, Kenya, 1989 Percent of children with Not con- No.of Percent diarrhoea treated with*: sulting chil- consult- facility dren Background ing a Home Uther and no with character- medical ORS solu- treat- treat- diar- istics facility packets tion ment ment rhoea Age in months ,6 55.1 9.4 55.8 85.7 6.7 108 6-11 49.0 26.1 40.9 84.0 10.2 180 12-17 53.2 24.6 61.3 92.3 3.2 180 18-23 44.6 21.8 44.5 84.3 14.2 112 24-59 38.0 19.6 44.7 76.3 14.8 250 Sex Male 47.0 16.6 46.8 85.5 9.1 413 Female 46.6 25.4 50.9 82.0 11.2 417 Residence Urban 58.7 20.7 52.0 84.0 12.0 97 Rural 45.2 21.1 48.5 83.7 9.9 733 Province Nairobi 66.7 23.1 57.7 87.2 5.1 50 Central 31.9 19.6 71.4 94.7 3.7 92 Coast 58.2 38.0 35.1 86.6 10.6 38 Eastern 48.4 13.8 54.1 87.2 10.2 177 Nyanza 49.8 24.6 41.0 77.1 13.9 171 Rift Valley 35.8 29.0 27.5 69.3 18.6 113 Western 48.9 16.6 53.5 88.4 6.0 188 Education NO education 42.3 18.6 42.6 77.2 16.3 228 some primary 49.2 21.3 51.7 86.0 7.6 264 Primary con~. 49.2 21.2 55.1 86.9 8.0 196 Secondary + 46.3 24.3 45.4 85.6 8.1 142 Total 46.8 21.1 48.9 83.7 10.2 830 * Percents may add to more than I00, since children may receive more than one treatment. 67 The type of treatment given varies somewhat according to background characteristics. For example, small infants are less likely than older children to receive oral rehydration solution made from packets as treatment for diarrhoea, while older children are more likely tb.an younger children not to receive any treatment at all. Urban children are more likely than rural children to consult a medical facility. Children in Nairobi are more likely to be taken for medical consultation than children in other provinces and children in Central Province are more likely to be given home solution when they have diarrhoea, while those in Coast Province are more likely to be given solutions made from ORS packets. Differences in treatment by education of the mother are small, except that children of mothers with no education are more likely to receive no treatment. Fever In Table 6.12, information is presented on the percentage of children under age five reported to have had fever during the four weeks prior to the KDHS interview. Fever is a specific symptom of many infectious diseases, but increased prevalence of fever may indicate a higher prevalence of malaria. Forty-two percent of children under age five had fever during the month before the survey. The age of the child is related to the reported episode of lever, with prevalence peaking at 55 percent among children aged 6-11 months, whereas it is lowest for children aged 24-59 months (37 percent). There is no evidence of strong differentials in fever prevalence by sex, urban-rural residence, province, or education of the mother, except that prevalence is lowest in Rift Valley Province. Table 6.12 further shows that of the reported children with fever, 56 percent consulted a medical facility, which is higher than the percentage of children with diarrhoea who consulted a medical facility. The percentage of children with fever who receive medical consultation varies little by age or sex of child, or mother's education. Children of urban women are more likely to Table 6.12 Among children under 5 years of age, the percentage who are reported by the mother as having had fever in the past four weeks, and, among children under 5 who had fever in the past four weeks, the percentage consulting a medical facility, according to background characteristics, Kenya, 1989 Percentage Percentage Number Background with fever consulting of characteD- in past medical children istics four weeks facility under 5 Age in months <6 45.5 63.6 601 6-11 55.2 56.8 710 12-17 47.7 60.6 703 18-23 50.5 59.6 612 24-59 36.8 51.5 3889 SeK Male 41.4 55.8 3210 Female 42.8 55.3 3305 Residence Urban 41.5 71.5 899 Rural 42.2 53.0 5615 Province Nairobi 45.9 69.8 386 Central 50.2 52.7 927 Coast 44.1 76.9 378 Eastern 43.7 55.9 1174 Nyanza 50.0 56.4 1106 Rift Valley 29.0 49.9 1533 Western 41.6 48.5 1011 Education NO education 38.6 54.2 1725 Some primary 44.2 52.0 2043 Primary comp. 42.2 56.1 1546 Secondary + 43.3 62.7 1194 Total 42.1 55.5 6514 receive medical consultation (72 percent) than children of rural

View the publication