Yemen - Demographic and Health Survey - 1994

Publication date: 1994

Yemen Demographic and Maternal and Child Health Survey 1991/1992 Central Statistical Organization @ Pan Arab Project for Child Development ®DHS Demographic and Health Surveys Macro International Inc. REPUBLIC OF YEMEN Yemen Demographic and Maternal and Child Health Survey 1991/1992 Central Statistical Organization Sana'a, Yemen Pan Arab Project for Child Development Cairo, Egypt Macro International Inc. Calverton, Maryland USA March 1994 J This report presents the findings of the 1991/1992 Yemen Demographic and Maternal and Child Health Survey (YDMCHS) conducted by the Central Statistical Organization, in cooperation with the Ministry of Public Health. Macro International Inc. and Pan Arab Project for Child Development (PAPCHILD) provided technical assistance. Funding was provided by the U.S. Agency for International Development, the Arab Gulf Program for the United Nations Development Organization (AGFUND), UNFPA, UNICEF, and the Republic of Yemen. The YDMCHS is part of the worldwide Demographic and Health Surveys (DHS) program and also the PAPCHILD program which concentrates on the Arab region. Both the DHS and PAPCHILD programs are designed to collect data on fertility, family planning, and maternal and child health. Additional information on the Yemen survey may be obtained from the Central Statistical Organization (CSO), Sana'a, Yemen. Additional information about the DHS program may be obtained by writing to: DHS, Macro International Inc., 11785 Beltsville Drive, Calverton, MD 20705, USA (Telephone 301-572-0200; and Fax 301-572-0999). Additional information about the PAPCHILD program may be obtained by writing to: Pan American Project for Child Development - League of Arab States, 22 A Taha Hussein Street, Zamalek, Cairo, Egypt (Telephone 3404306; and Fax 3401422). Recommended citation: Central Statistical Organization (CSO)[Yemen] and Pan Arab Project for Child Development (PAPCHILD)[Egypt] and Macro International Inc.(Ml). 1994. Yemen Demographic and Maternal and Child Ilealth Survey 1991/1992. Calverton, Maryland: CSO and MI. CONTENTS Page Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Map of Yemen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi CHAPTER 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 CHAPTER 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 CHAPTER3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 THE SETTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 GEOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 H ISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 NATURAL RESOURCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 POPULAT ION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 SOCIOECONOMIC CONDITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 POPULAT ION POLICY AND STRATEGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 OBJECTIVES OF THE SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 ORGANIZAT ION OF THE SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 IMPLEMENTATION OF THE SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 RESULTS OFTHE HOUSEHOLD AND WOMEN'S INTERVIEWS . . . . . . . . . . . . 10 BACKGROUND CHARACTERISTICS: HOUSEHOLDS, RESPONDENTS, AND CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 HOUSING CHARACTERISTICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 PRESENCE OF DURABLE GOODS IN THE HOUSEHOLD . . . . . . . . . . . . . . . . . . 15 HOUSEHOLD POPULATION BY AGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 HOUSEHOLD COMPOSIT ION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 EDUCATIONAL STATUS OF HOUSEHOLD POPULAT ION . . . . . . . . . . . . . . . . . 18 SCHOOL ENROLLMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 BACKGROUND CHARACTERISTICS OF WOMEN . . . . . . . . . . . . . . . . . . . . . . . . 21 CHARACTERISTICS OF CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 FERTILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 LEVELS AND DIFFERENTIAL IN FERTIL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 TRENDS IN FERTIL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 CURRENT PREGNANCY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 CHILDREN EVER BORN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 BIRTH INTERVALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 AGE AT FIRST BIRTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 TEENAGE FERTIL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 iii CHAPTER 4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 CHAPTER 5 5.1 5.2 5.3 5.4 5.5 CHAPTER 6 6.1 6.2 6.3 6.4 CHAPTER 7 7.1 7.2 7.3 CHAPTER 8 8.1 8.2 8.3 8.4 8.5 8.6 Page FAMILY PLANNING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 KNOWLEDGE OF FAMILY PLANNING METHODS AND SOURCES . . . . . . . . . 37 EVER USE OF CONTRACEPT ION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 NUMBER OF CHILDREN AT FIRST USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 REASON FOR FIRST USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 CURRENT USE OF CONTRACEPT ION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 REASON FOR CURRENT USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 PROBLEMS WITH THE CURRENT METHOD OF CONTRACEPT ION . . . . . . . . 45 SOURCE OF CURRENT CONTRACEPT IVE METHOD . . . . . . . . . . . . . . . . . . . . . 46 T IME TO SOURCE OF CONTRACEPT IVE METHOD . . . . . . . . . . . . . . . . . . . . . . . 47 INTENTION TO USE CONTRACEPTION IN THE FUTURE . . . . . . . . . . . . . . . . . . 48 REASON FOR NONUSE OF CONTRACEPTION . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 SPOUSAL COMMUNICAT ION AND APPROVAL OF FAMILY PLANNING . . . 51 NUPTIALITY AND EXPOSURE TO RISK OF PREGNANCY . . . . . . . . . . . . . . 53 CURRENT MARITAL STATUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 AGE AT FIRST MARRIAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 POLYGYNY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 POSTPARTUM AMENORRHEA AND INSUSCEPTIB IL ITY . . . . . . . . . . . . . . . . . 61 TERMINAT ION OF EXPOSURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 FERTILITY PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 DESIRE FOR CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 SEX PREFERENCES FOR CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 IDEAL NUMBER OF CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 WANTED FERTIL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 MATERNAL HEALTH CARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 ANTENATAL CARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 DEL IVERY CARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 POSTNATAL CARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 CHILD FEEDING AND WEANING PRACTICES . . . . . . . . . . . . . . . . . . . . . . . . 91 PREVALENCE OF BREASTFEED1NG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 REASONS FOR NOT BREASTFEEDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 BREASTFEEDING PATTERNS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 SUPPLEMENTATION PRACTICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 DURATION OF BREASTFEEDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 WEANING OF CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 iv CHAPTER 9 9.1 9.2 9.3 CHAPTER 10 10.1 10.2 10.3 10.4 CHAPTER 11 11.1 11.2 11.3 11.4 CHAPTER 12 12.1 12.2 12.3 12.4 12.5 12.6 12.7 CHAPTER 13 Page PREVENTIVE MEASURES: CHILD VACCINAT IONS . . . . . . . . . . . . . . . . . 105 VACCINATION CARD COVERAGE AND AVAILABIL ITY . . . . . . . . . . . . . . . 106 VACCINATION COVERAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 REASONS FOR NOT HAVING CHILDREN VACCINATED . . . . . . . . . . . . . . . 111 MORBID ITY AND ACCIDENTS IN CHILDREN . . . . . . . . . . . . . . . . . . . . . . . 113 DIARRHEAL DISEASES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 ACUTE RESPIRATORY INFECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 ILLNESSES OTHER THAN DIARRHEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 SERIOUS ACCIDENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 CURATIVE MEASURES: TREATMENT OF CHILD I LLNESS . . . . . . . . . . 125 DIARRHEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 TREATMENT OF THE LAST EPISODE OF DIARRHEA . . . . . . . . . . . . . . . . . . 130 ACUTE RESPIRATORY INFECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 FEVER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 INFANT AND CHILD MORTAL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 ASSESSMENT OF DATA QUAL ITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 PREVALENCE OF CHILD LOSS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 LEVELS OF INFANT AND CHILD MORTAL ITY . . . . . . . . . . . . . . . . . . . . . . . . 148 DIFFERENTIALS IN INFANT AND CHILD MORTAL ITY . . . . . . . . . . . . . . . . . 154 REPRODUCTIVE FACTORS AND CHILD SURVIVAL . . . . . . . . . . . . . . . . . . . 156 ENVIRONMENTAL FACTORS AND CHILD MORTAL ITY . . . . . . . . . . . . . . . 159 CAUSES OF DEATH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 SUMMARY AND POL ICY IMPL ICAT IONS . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 APPENDIX A APPENDIX B APPENDIX C APPENDIX D APPENDIX E SURVEY STAFF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 SAMPLE DES IGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 EST IMATES OF SAMPL ING ERRORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 DATA QUALITY TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 QUEST IONNAIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 TABLES Table 1.I Table 2.1 Table 2.2 Table 2.3 Table 2A Table 2.5 Table 2.6.1 Table 2.6.2 Table 2.7 Table 2.8 Table 2.9 Table 2.10 Table 2.11 Table 3.1 Table 3.2 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 3. l 1 Table 3.12 Table 4.1 Table 4.2 Table 4.3 Table 4A Table 4.5 Page Results of the household, women's and children's interviews . . . . . . . . . . . . . . . 11 Housing characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Household durable goods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Household population by age, residence and ~ex . . . . . . . . . . . . . . . . . . . . . . . . 16 Population by age from selected sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Household composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Educational level of the male household population . . . . . . . . . . . . . . . . . . . . . 19 Educational level of the female hot sehcld population . . . . . . . . . . . . . . . . . . . . 19 School enrollment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Background characteristics of respondents . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Level of education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Exposure to mass media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Background characteristics of children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Current fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Fertility by background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Age-specif ic fertility lates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Fertility by marital duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Marital fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Currently pregnant women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Children ever born and living . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Birth intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Age ~t first birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Median age ~t first birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Teenage pregnancy and motherhood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Children born to teenagers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Knowledge of contraceptive methods and source for methods . . . . . . . . . . . . . . . 38 Knowledge of modem contraceptive methods and source for methods, by selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Ever use of contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Number of children at first use of contraception . . . . . . . . . . . . . . . . . . . . . . . . 41 Reproductive intentions at first use of contraception . . . . . . . . . . . . . . . . . . . . . 42 vii 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 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 5.10 Table 5.11 Table 6.1 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 7.1 Table 7.2 Table 7.3 Page Current use of contraception by age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Current use of contraception by background characteristics . . . . . . . . . . . . . . . . 44 Reasons for current use of contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Problems with current method of contraception . . . . . . . . . . . . . . . . . . . . . . . . . 45 Source of supply for modem contraceptive methods . . . . . . . . . . . . . . . . . . . . . 46 Time to source of supply for modem contraceptive methods . . . . . . . . . . . . . . . 47 Future use of contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Preferred method of contraception for future use . . . . . . . . . . . . . . . . . . . . . . . . 50 Reasons for not intending to use contraception . . . . . . . . . . . . . . . . . . . . . . . . . 51 Communication with husband and husband's opinion . . . . . . . . . . . . . . . . . . . . 52 Current marital status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Remarriage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Age at first marriage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Median age at first marriage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Age difference between spouses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Ideal age at marriage for women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Consanguinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Polygyny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Number of co-wives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Postpartum amenorrhca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Termination of exposure to the risk of pregnancy . . . . . . . . . . . . . . . . . . . . . . . 63 Fertility preferences by selected background characteristics . . . . . . . . . . . . . . . . 66 Desire to limit childbearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Preferred sex of next child . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Ideal number of children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Mean ideal number of children by background characteristics . . . . . . . . . . . . . . . 72 Ideal family size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Wife's ideal family size compared to her husband's . . . . . . . . . . . . . . . . . . . . . 73 Mean difference in wife's and husband's ideal family size . . . . . . . . . . . . . . . . . 74 Wanted fertility rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Antenatal care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Medical conditions associated with pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . 81 Tetanus toxoid vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 viii Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 7.9 Table 8.1 Table 8.2 Table 8.3 Table 8A Table 8.5 Table 8.6 Table 8.7 Table 8.8 Table 8.9 Table 8.10 Table 8.11 Table 9.1 Table 92 Table 9.3 Table 9.4 Table 10.1 Table 10.2 Table 10.3 Table 10.4 Table 10.5 Table 10.6 Table 10.7 Table 11.1 Table 11.2 Table 11.3 Table 11.4 Table 11.5 Page Place of delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Assistance during delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Type of delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Re~son for not delivering st health facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Treatment of cord stump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Postnatal c~re . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Prevalence of breastfeeding by background characteristics . . . . . . . . . . . . . . . . . 92 Prevalence of breastfeeding by birth order . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Re~sons for not breastfeeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Breastfeeding patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Starting age for supplementary food . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Food supplementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Bottle feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Duration of breastfeeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Postpartum amenorrhea and abstinence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Age at weaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Reason for weaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Health card . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Immunization of children under five . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Immunization of children 12-23 montt's . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Re~sons for children not being v~ccinatcd . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Prevalence of diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Signs and symptoms of diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Severity of diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Prevalence of ~cute respiratory infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Incidence of measles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Prevalence of fever and other infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Prevalence cf accidents among children under five . . . . . . . . . . . . . . . . . . . . . 124 Knowledge and u~,e of ORS packets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 Knowledge of sources for ORS pickets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Quantity of water used to prepare ORS packets . . . . . . . . . . . . . . . . . . . . . . . 129 Type of water used to prepare ORS p~ckets . . . . . . . . . . . . . . . . . . . . . . . . . . 130 U~,e of oral rehydration therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 ix Table 11.6 Table 11.7 Table 11.8 Table 11.9 Table 11.10 Table 11.11 Table 11.12 Table 11.13 Table 11.14 Table 12.1 Table 12.2 Table 12.3 Table 12.4 Table 12.5 Table 12.6 Table 12.7 Table 12.8 Table 12.9 Table B. 1 Table C. 1 Table C.2 Table C.3 Table C.4 Table C.5 Table C.6 Table D. 1 Table D.2 Table D.3 Table D.4 Table D.5 Table D.6 Page Treatment of diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Sources of treatment for diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Treatment for diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Feeding modes during diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Breastfeeding practices during diarrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Reasons for not taking child with diarrhea lot treatment . . . . . . . . . . . . . . . . . 139 Sources of treatment for acute respiratory int~ction (ARI) . . . . . . . . . . . . . . . . 140 Treatment of acute respiratory infection (ARI) . . . . . . . . . . . . . . . . . . . . . . . . 141 Treatment of fever . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Mean number of children ever born and proportion dead . . . . . . . . . . . . . . . . . 146 Number of children ever born by number of l iving children . . . . . . . . . . . . . . . 148 Infant and child mortality by gender of child . . . . . . . . . . . . . . . . . . . . . . . . . 149 Period-direct measures of infant and child mortality by residence and gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 Infant and child mortality estimates, age model and duration model . . . . . . . . . 153 Infant and child mortality by background characteristics . . . . . . . . . . . . . . . . . 154 Infant and child mortality by demographic characteristics . . . . . . . . . . . . . . . . 157 Infant and child mortality by environmental characteristics . . . . . . . . . . . . . . . . 160 Causes of death in early childhood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Sample implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 List of selected variables for sampling errors . . . . . . . . . . . . . . . . . . . . . . . . . 189 Sampling errors - Entire sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 Sampling errors - Urban area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Sampling errors - Rural area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Sampling errors - Northern and western govcmorates region . . . . . . . . . . . . . . 193 Sampling errors - Southern and eastern govemorates region . . . . . . . . . . . . . . . 194 Household age distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Age distribution of eligible and intervicwcd women . . . . . . . . . . . . . . . . . . . . 198 Completeness of reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Births by calendar year since birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20ll Reporting of agc at death in days . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Reporting of age at death in months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 FIGURES Figure 2.1 Figure 3.1 Figure 3.2 Figure 4.1 Figure 4.2 Figure 4.3 Figure 5.1 Figure 6.1 Figure 6.2 Figure 7.1 Figure 7.2 Figure 8.1 Figure 8.2 Figure 8.3 Figure 9.1 Figure 9.2 Figure 10.1 Figure 10.2 Figure 10.3 Figure 10.4 Figure 11.1 Figure 11.2 Figure 11.3 Figure 11.4 Page Population pyramid of Yemen, 1991/92 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Age-specific fertility rates by urban-rural residence . . . . . . . . . . . . . . . . . . . . . . 27 Total fertility rate (TFR) and mean number of children ever born (CEB) to women 40-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Contraceptive prevalence by selected background characteristics . . . . . . . . . . . . . 43 Source of family planning supply, current users of modem methods . . . . . . . . . . 47 Time to source of modem methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Median duration of postpartum amenorrhea . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Fertility preferences among currently married women 15-49 . . . . . . . . . . . . . . . . 67 Total wanted [~rtility rates and total fertility rates . . . . . . . . . . . . . . . . . . . . . . . 75 Reasons lor attending and not attending antenatal care . . . . . . . . . . . . . . . . . . . 80 Accessibility of antenatal care among currently pregnant women . . . . . . . . . . . . 80 Mean duration of breastfceding by residence and region . . . . . . . . . . . . . . . . . . 99 Pattern of weaning by selected background characteristics . . . . . . . . . . . . . . . . 100 Weaning of last birth under 3 months by selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Children under 5 with health cards and card availability by selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Immunization coverage for children 12-23 months (BCG, DPT3 and fully vaccinated) by residence and region . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Children under 5 who had diarrhea by selected demographic characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Children under 5 who had cough and difficult breathing by selected demographic characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Children under 5 who had measles by selected demographic characteristics . . . . 121 Children under 5 who had fever by selected demographic characteristics . . . . . . 123 Percentage of respondents who have heard of ORS by selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Children under 5 with diarrhea who rcccivcd no treatment by selected demographic characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Treatment of mild and severe diarrhea among children under 5 . . . . . . . . . . . . 134 Treatment of cough and difficult breathing among children under 5 by selected background characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 xi Figure 12.1 Figure 12.2 Figure 12.3 Figure 12.4 Proportion of children dead by age of mother and residcncc . . . . . . . . . . . . . . . 147 Trends in infant and child mortality for 5-year periods preceding the survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Infant and child mortality by selected background characteristics . . . . . . . . . . . 155 Infant and child mortality by selected demographic characteristics . . . . . . . . . . 158 xii FOREWORD The publication of this final report on the Demographic, Maternal and Child Health Survey in the Republic of Yemen is considered a great achievement and is part of the Central Statistical Organization's plan to establish a comprehensive statistical and informational database on population, health, and the socioeconomic status of the Yemen'i people. Such a demographic database will help planners, decisionmakers, and researchers in putting forward realistic regional, sectoral, and comprehensive development plans for execution throughout the country. The success of these development plans is dependent on the accuracy of the socioeconomic and health indicators obtained from the survey data. This report focuses mainly on issues related to maternal and child health and the circumstances which directly and indirectly affect the lives of mothers and children, a wide sector of Yemeni society. The Central Statistical Organization carried out the Demographic, Maternal and Child Health Survey and prepared this final report in collaboration with the Ministry of Public Health, PAPCHILD, the Arabian Gulf Program, UNFPA, UNICEF, the UN Statistical Division, Macro International Inc., and USAID in Yemen. The execution of the survey and the preparation of this report have gonc through several phases, starting November 16, 1991 and ending with the publication of this report in February 1994. These phases included the preparation for the survey, field operations, data processing and tabulation, publication of the preliminary report in August 1992, analysis of the survey data, preparation of the chapters for this report, and publication and distribution of the report. The importance of the achievement of this survey lies in it being the first statistical, scientific research project executed after the unification of the country on May 22, 1990. The survey data represent all the govemorates, both urban and rural. The results will be a major factor complementing and supporting the efforts of the government (all sector and agencies) to execute a population policy, especially since the National Population Council came into being in July 1992 and the General Secretariat was established in January 1993. The General Secretariat will be responsible for implementation of the plan of action emerging from the National Population Strategy and the proceedings of the First National Population Conference held in Sana'a during October 26-29, 1991. It gives us great pleasure to present this report to all scholars, researchers and concerned users. This report, which is a detailed and accurate representation of the final survey results, deals with and touches upon important aspects of maternal and child health in unified Yemen. It also emphasizes the important role played by statistical work in planning, development and scientific research in our country. A word of thanks and gratitude are in order here for all the persons who participated in the execution of this survey and the preparation of this report, especially the project's administrative staff, the General Department of Population Studies and Research Center in the Central Statistical Organization, the Ministry of Public Health, PAPCHILD, and the Demographic and Health Surveys (DHS) program at Macro International Inc. (USA) for their help and important role in producing this report. Finally, we hope that this research study meets the stated objectives of the survey and will serve all researchers, planners and decisionmakcrs. Abdoraboh Ahmed Gradah Chairman Central Statistical Organization and Survey Committee xiii PREFACE The Yemen Demographic and Maternal Child Health Survey (YDMCHS) was carried out by the Central Statistical Organization with the aim of providing detailed information on the factors affecting maternal and child health and survival: demographic, social, economic and environmental variables. The YDMCHS was designed to respond to the needs of the Republic of Yemen and to provide internationally comparable data. The survey was carried out using the most scientific and accurate survey methods available. The successful implementation of the YDMCHS was made possible by the active support and dedicated efforts of a large number of individuals from the Central Statistical Organization and the Ministy of Public Health in Sana'a, Yemen, Macro International Inc. in Calverton, Maryland USA, and the Pan Arab Project for Child Development (PAPCHILD) in Cairo, Egypt. Funds from the U.S. Agency for International Development (USAID) in Washington and Sana'a and the support of the Mission Director and Dr. Raga Uqba of USAID in Yemen are gratefully acknowledged. I wish to thank Dr. Mohamed Ayad, Regional Coordinator, Macro International, for his support and encouragement throughout all phases of the survey. I would also like to express my thanks to other staff at Macro International who contributed to the survey: Dr. J. Ties Boerma for the questionnaire and survey design, Dr. Alfredo Aliaga for the sample design, Ms. Jeanne Cushing for her assistance in data processing, Mr. Noureddine Abderrahim for data tabulations, Dr. Ann Way and Dr. Jeremiah Sullivan for reviewing two chapters of the report, and Mr. Jonathan Dammons and Mr. Robert Wolf for preparing graphs for the report. Special thanks go to Ms. Kaye Mitchell for word processing support and to Dr. Sidney Moore for editing the report. 1 'also wish to record my special thanks to Mr. Sushil Kumar, the DHS Country Monitor for the YDMCHS for his dedicated efforts throughout all phases of the survey, especially his valuable contribution to the final report. 1 wish to express my thanks to the League of Arab States (LAS) and the staff of PAPCHILD for the valuable and significant support they provided throughout the various stages of this survey. In particular, I would like to thank Mr. Mahdi El Hadi, LAS Under-secretary for Social Affairs and PAPCHILD Programme Director, Dr. Atef Khalifa (former), PAPCHILD Manager, Dr. Hoda Rashad, PAPCHILD Manager, Dr. Ahmed Abdul Moniem, Country Monitor for the YDMCHS, and other PAPCHILD experts: Mrs. Samia Charchour, Mr. Hafedh Chakir, Mr. Ahmed E1 Baz, and Dr. Laila E1 Zeiny. Special thanks are also due to Dr. Nabil Khorazaty for his contribution to the scientific editing. International support from AGFUND, UNFPA, UNICEF, WHO and UNSTAT and other funders of PAPCH1LD, is also highly appreciated. We are deeply indebted and grateful to Dr. Farida Allaghi (AGFUND), Mr. Abdul Monem Abu Nawar, Mr. Hrdi Jemai (UNFPA), Dr. Fatih E1Samani (UNICEF), and Mr. M. Ouakrim (WHO), for the valuable support they have provided throughout the survey. March 1994 Amin Marouf al Janad Project Director XV _<, YEMEN AL-MAHWEET ~ HAJJAH HODEIDAH RED SEA 0 HUNAISH ISL4NDS 5A'ADAH SANA'A >, i 5AN4 ~ CI~ DHAMAR IBB TAIZ KINGDOM OF SAUDI ARABIA / / -~ / / AL-JAWF I / I / ( MA'ARE~f" $HABWAH AL-BEIDA AIBYEN ~ GULF OF ADEN ADEN / / I HADRAMOUT ~UK4JJ~ ARAB~AN SEA I T " I AL-MAHRAH SULTANATE OF OMAN CHAPTER 1 THE SETTING This chapter presents a short review of the geography, history, natural resources, and population of the Republic of Yemen, and the socioeconomic and political conditions prevailing in the Republic. The pur- pose is to provide the reader with a comprehensive background on the country and the condition of the Yemeni community. The chapter also includes a description of the organization of the Yemen Demographic and Matemal and Child Health Survey (YDMCHS) and the implementation of the different phases of the survey. 1.1 GEOGRAPHY The Republic of Yemen is located in the southern part of the Arabian Peninsula between 12 ° and 20 ° north latitude and 41 ° and 54 ° east longitude. The total area, excluding A1-Rub AI-Khali desert, is 555,000 square kilometers. The boundaries of the Republic of Yemen are the Kingdom of Saudi Arabia in the north, the Arabian Sea and Gulf of Aden in the south, the Sultanate of Oman in the east, and the Red Sea in the west. The Bab Al-Mandab strait lies offthe southwestern tip of Yemen. Mayoun, a Yemeni island in themiddle of the strait, controls passage into and out of the Red Sea. There are about 112 Yemeni islands in the Red Sea and Arabian Sea. The largest is Sucotra, which has an area of 3650 square kilometers and lies 510 kilometers southeast of the Yemeni seaport of Mukalla. The other major islands, in order of size, are Kamaran, Hunaish Al-Kubra, Hunaish Al-Sughra, Zaqar A1- Zubair, and AI-Tair (Central Statistical Organization, 1991 a). 1.2 H ISTORY In ancient times, geographical location and favorable natural conditions played an important role in population settlement and the development of civilizations in the territory now occupied by the Republic of Yemen. This was particularly true regarding the valleys toward the A1-Rub aI-Khali desert in the east, the Arabian Sea in the south, and the Red Sea in the west. The eras of Mae'en Hadramout and Saba'a (Sheba) are considered to be the first organized political entities in Yemen before the birth of Christ. The Hemyar era flourished later and ended with the Ethiopian invasion in 525 A.D. The most important activities of these regimes were agriculture and trade. They invented agricultural terracing on the mountains and established dams, the most famous of which was Ma'areb dam. They also controlled caravans, which transported commodities from India and East Africa across the Arabian Peninsula to areas around the Mediterranean Sea (Al-Zoabi, 1990). Yemeni Civilization flourished in those years, and Yemen was called Arabia Felix (Arabia the "happy") by the Greeks. By the end of the Hamiarite era, however, Yemen was dominated first by the Ethiopians and then by the Persians. This lasted until the emcrgence of Islam, when the Yemenis embraced the Islamic religion and Yemen became part of the central Islamic state. This period lasted from 628 A.D. until 824 A.D. Thereafter, small independent states emerged, leaving Yemen weak and divided. The Turks occupied Yemen from the sixteenth century until the beginning of the seventeenth century, while the British occupied Aden in 1839. The Turks invaded the northern part of Yemen again in 1872, and continued their occupation until the outbreak of World War II, when Yemen came under the rule of the Hameed AI-Deen family in the north. This lasted until the revolution, 26 September 1962. The British remained in the south until the outbreak of the 14 October Revolution, which resulted in independence 30 November 1967 (Mustafa, 1984). With the success of the two revolutions, Yemen entered a new era of economic, social, cultural, and political change. The latest of these changes was the unification of the two parts of Yemen (north and south) on 22 May 1990. This led to democratic changes and the establishment of political parties. Parliamentary elections were held 27 April 1993, resulting in the first Parliamentary Council, which in tum elected the Presidential Council to abide by the unified country's Legislation. 1.3 NATURAL RESOURCES Yemen is divided into five regions: Mountain Area This area consists of two mountain ranges: the first is the North-South mountain range, parallel to the Red Sea; the second is West-East mountain range, parallel to the Gulf of Aden. The height of these mountains varies between 1,000 and 3,600 meters. The highest peak on Nabi Shuaib Mountain is 3,666 meters above sea level. It is the highest peak in the Arabian Peninsula and in the Arab region. The water from these mountains drains in all directions, forming valleys with basins and plains suitable for agriculture. Hill Area The hill area is to the east and north of the mountainous area parallel to it. The hill area becomes wider toward the AI-Rub Al-Khali region on the north. The highest peak is 1,000 meters. The boundaries of this region are contiguous with the A1-Rub Al-Khali region, which extends deep into the Arabian Peninsula and comprises one-quarter of its area. The Coastal Area The coastal area includes all areas adjacent to the Red Sea, Gulf of Aden, and the Arabian Sea. It is continuous, forming a coastal strip stretching from the Omani border in the east to Bab AI-Mandab Strait in the west, and north to the border of Saudi Arabia, a distance of approximately 2,000 kilometers. The coastal area varies in width from 30 to 60 kilometers. AI-Rub AI-Khali Area This area is part of the Yemeni desert and contains some desert plants, particularly in the outermost areas; it is contiguous with the hill area. As one goes deeper into the AI-Rub AI-Khali desert, there are fewer plants and the sand dunes increase. The Yemeni Islands These islands are scattered in the Yemeni territorial waters of the Red Sea and the Arabian Sea. Most of the islands are in the Red Sea, parallel to the Yemeni coast. The biggest and the most important island in the Red Sea is Kamaran, which is densely populated. In the Arabian Sea, the Yemeni islands are close together. The most famous island is Sucotra, on which are found ormosia, dracena draco and pterocarpus draco trees, from which gum, various medicine, incense, and pigments are obtained. 1.4 POPULAT ION Population Size and Distribution The latest two census of the population in the Republic of Yemen were conducted in 1986 and 1988. The 1986 Census was conducted in the northern govemorates (then called the Yemen Arab Republic, Y.A.R), while the 1988 Census was conducted in the southern govemorates (then called the People's Democratic Republic of Yemen, P.D.R.Y.). According to these censuses, the resident population size in the northem governorates was about 7.8 million and the resident population size in the southern governorates was about 1.8 million. The resident population size in the Republic of Yemen was estimated at approximately 11.3 million in 1990. (Central Statistical Organization, 1991b). At the national level, population density is about 21.4 persons per square kilometcr (1990), distributed among 17 govemorates and the Capital Mayorship (Sana'a City). The governorates include 238 directorates or districts (Public Survey Authority, 1990). The population is distributed between rural and urban areas. In 1990, 21.4 percent of the population lived in urban areas. The most densely populated cities arc Sana'a, Aden, Taiz, Hudaidah, and Mukalla. Population Growth Improvements in the living and health conditions in Yemeni society following the revolution had a significant impact on population growth. The growth rate increased slightly as a result of stable birth rates coupled with decreasing death rates. In 1988, the crude birth rate in Yemen was 52.6 per thousand, while the crude death rate was 21.8 per thousand. Accordingly, the annual rate of population growth is 3.1 percent. The total fertility rate, which is the total number of live births per woman by the end of the reproductive years, is 8.2 births per woman. Therefore, among developing countries, the Republic of Yemen is considered to have one of the highest rates of population growth and highest fertility rates. It is worth mentioning, that mortality in Yemen is still high, particularly among children. In 1988 the infant mortality rate was 130 per thousand (Central Statistical Organization, 1992a). Life expectancy at birth is moderate compared to many countries. In 1988 it was estimated at 46.3 years for both sexes. In the last decade there was a clear improvement in the standard of living and health conditions, which was reflected in the continuing increase in life expectancy. 1.5 SOCIOECONOMIC CONDIT IONS Education Education is one of the most important measures of social and economic development. Hence, the State has paid great attention to education since the revolution in the early sixties. Illiteracy among Yemenis was the highest among Arab and other developing countries. At the time of the revolution, there was not one secondary school in the country. Only Aden, being under the British occupation, was an exception. Statistics show that considerable progress has been achieved in this area. Almost 2 million students were enrolled in over 12,000 schools in 1990/91, while Sana'a and Aden universities had an enrollment of about 35,000 students in 1989/90 (Ganin and Mutahar, 1992). These figures confirm that there has been a great improvement in the education sector in Yemen, especially when compared to the previous lmami and Colonial regimes. Despite these efforts, figures show that more than half of the population is still illiterate. The 1986 and 1988 censuses in both the northern and southern govemoratcs show that illiteracy is about 67 percent and is higher among females than males. Figures also show that enrollment in basic education is about 57.4 percent for the population aged 6-15 years (Ganin and Mutahar, 1992). These figures indicate that Yemen needs to expend more resources to increase basic education, especially considering the high rate of population growth. Health Health conditions in any community arc the result of various social, economic, cultural, and environmental factors. Yemen witnessed many changes in the period lollowing the revolution. This is reflected in the improved level of health of the population, and in the increased coverage of health services. As a result, infant mortality has declined, while life expectancy has increased. For example, the infant mortality rate decreased from 171 per thousand live births in 1975 in the northern and western govemorates to 130 per thousand in 1990 (Central Statistical Organization, 1992a). The efficiency of health services has increased. In 1990, the number of hospitals reached 74, in addition to 94 health centers. These facilities have a capacity of 9,891 beds, with a staff of 2,854 doctors, 2,004 of whom are nationals (Central Statistical Organization, 1991a; 1992a). Despite progress in the health sector, health services cover only 40 percent of the population and are concentrated in urban areas. Rural areas are to some extent deprived of such scrviccs, especially in remote areas, which are difficult to reach. Population Activities According to 1990 population estimates, the labor force--all persons age 15-64 years--represents 44.3 percent of the population. Additionally, children under 15 years make up 53.5 percent, and 3.3 percent are elderly persons (65 years or more). Thus, the dependency ratio in Yemen is high: 126 per~ns in the non- productive age groups for every 100 persons aged 15-64 years. This is a result of the large number of children under 15 years, which is common in developing countries. Most of the Yemeni labor force (62.2 percent) is involved in agriculture and fishing, while the rest are distributed among other sectors (Sail et al., 1992). There are approximately 5.1 million acres of amble land in Yemen, 9.5 percent of the total land area. Only 1.1 million acres are under cultivation (Hasbem et al., 1992). Agriculture depends mainly on rain water. Eighty percent of the arable land receives rain water. Thus, production is subject to climatic conditions. Locally-grown fruits and vegetables are sufficient to meet the needs of the population; however, some essential food stuffs are imported, such as wheat, two-thirds of which is imported (Hashem et al., 1992). There are indications that Yemen has considerable unused mineral wealth, such as oil. The Yemeni government pays great attention to oil exploration operations, which have increased greatly since unification. It is likely that in the coming years Yemen will be an oil exporting country, exporting oil in large quantities. 1.6 POPULAT ION POL ICY AND STRATEGY In 1984, the National Committee for Population and Family Planning (NCPFP) was established to strengthen the government capacity to implement population policy in North Yemen. After the achievement of Yemeni unity in May 1990, the government drafted a national population strategy which, after revision, was adopted as national policy at the National Population Conference in October 1991. The National Population Council was established to oversee implementation of the policy. The overall objectives of the National Population Strategy for the year 2000 are (Central Statistical Organization, 1992a): 1. Reduce the infant mortality rate from 130 to 60 deaths per thousand live births; 2. Reduce the mortality rate for children age 1-4 years by 50 percent; . Increase immunization coverage for children under one year of age to 85 percent or more, and to expand tetanus immunization among women of reproductive age; . Reduce by 50 percent, compared with 1990 levels, the number of deaths among children due to diarrhea. The objectives of the family planning strategy are: . Increase the use of contraception to 35 percent among women of reproductive age, and expand family planning services to men; . Make family planning a free choice for couples, a basic human right, as well as a factor for social change. Family planning must also include the right to treatment of in fertility (Central Statistical Organization, 1992a). 1.7 OBJECT IVES OF THE SURVEY The General Department for the Population Studies and Research Center, with the cooperation of the Ministry of Public Health, the Pan Arab Project for Child Development (PAPCHILD), the Demographic Health Surveys (DHS), and the U.S. Agency for Intemational Development (USAID), has implemented the various phases of the 1991-92 Yemen Demographic and Maternal and Child Health Survey (YDMCHS). The survey was carried out as a part of the DHS program and also the PAPCHILD program. The DHS program is assisting governments and private agencies in the implementation of household surveys in developing countries; PAPCHILD has similar goals for developing countries in the Arab League. The main objectives of the DHS project are to: (a) pn~vide decisionmakers with a data base and analyses useful for informed policy choices, (b) expand the international population and health data base, (c) advance survey methodology, and (d) develop skills and resources necessary to conduct high quality demographic and health surveys in the participating countries. The YDMCHS was specifically aimed at furnishing information on basic population and household characteristics, matemal and child health, fertility, family planning, and infant and child mortality in Yemen. The survey also presents information on breastfeeding practices and the nutritional status of children under age five. The survey will provide policymakers and planners with important information for use in formulating programs and policies regarding maternal and child health, child mortality, and reproductive behavior. 1.8 ORGANIZAT ION OF THE SURVEY The Yemen Demographic and Maternal and Child Health Survey (YDMCHS) is the first national survey conducted in Yemen since unification of the country. It was designed to collect data on households, ever-m arried women of reproductive age, and children under age five. The subjects covered in the household survey were: characteristics of households, housing and living conditions, school enrollment, labor force participation, general mortality, disability, fertility, and child survival. The areas covered in the survey of women of reproductive age were: demographic and socioeconomic characteristics, marriage and reproductive history, fertility regulation and preferences, antenatal care, breastfeeding, and child care. For children under five in the survey, the topics included diarrheal and other morbidity, nutritional supplementation, accidents, vaccination, and nutritional status. Sample Design and Implementation The YDMCHS sample was designed to enable data analysis for Yemen as a whole, and separately for urban and rural areas, and for two regions: (1) the nonhero and western govemorates, and (2) the southern and eastern govemorates. The target sample was set at completed interviews for about 12,000 households with about 6,000 eligible women. No target number was fixed for children under five, for whom information was to be collected for all children in each household that was selected for the women's interview. In half of the selected households, only the Household Questionnaire was administered; in the other half, in addition to administering the Household Questionnaire, all eligible women were interviewed and information on eligible children was collected. The YDMCHS covered the entire country, except for nomadic peoples and those living on hard-to- reach Yemeni islands. The survey adopted a stratified, multi-stage sampling design. The sample was stratified by urban and rural areas in the two regions. In this report, the northem and western govemorates region includes: Sana'a City and the governorates ofSana'a, Taiz, Hodeidah, lbb, Dhamar, Hajjah, A1-Beida, Sa'adah, AI-Mahweet, Ma'areb, and AI-Jawf. The southern and eastern governorates region consists of Aden, Laheg, Abyen, Shabwah, Hadramout, and AI-Mahrah govemoratcs. In the first stage, sampling units or clusters were selected; the second stage involved selection of households. The initial objective of having a sel f-weighted sample was compromised in order to have reliable estimates for urban and rural areas within each region. Sana'a City, the urban (not rural) areas of Aden, and the rural areas of Laheg were oversampled. For the survey, 258 sampling units were selected, which contained 13,712 households. In half of the selected households, only the Household and Housing Characteristics Questionnaires were administered. In the other half, the Women's and Child's Questionnaires were also administered to all eligible women and children. The sample design is described in more detail in Appendix B. A discussion of sampling errors is presented in Appendix C, along with the sampling errors calculated for the entire sample, urban and rural samples, and regional samples. 1.9 IMPLEMENTATION OF THE SURVEY Preparation Stage Preparatory steps for the DHS survey project began in 1989 (before unification) at the Population Research Center of the Central Statistical Organization, in Sana'a. At the time, the Pan Arab Project for Child Development (PAPCHILD) was also involved in planning a survey in South Yemen. These efforts continued after unification and included both the DHS project and the PAPCHILD project in planning for the first national survey. The Prime Minister's Resolution No. 9, issued in 1991, initiated the survey and led to the formation of the Supreme Committee for the Yemen Demographic and Maternal and Child Health Survey. A technical committee was also formed and charged with preparation, organization, planning and implementation of the YDMCHS project, including all the technical and operational requirements (see Appendix A for the names of the members of the Supreme Committee and the Technical Committee, and other persons involved in the YDMCHS). Questionnaires Design, preparation andrevision of questionnaires. The YDMCHS survey includes the following questionnaires: Household Questionnaire Housing Characteristics Questionnaire Reproductive Health Questionnaire (also called the Women's Questionnaire Child Health Questionnaire (also called the Children's Questionnaire) Community Questionnaire The items included in these questionnaires were selected after reviewing similar surveys such as those carried out by the Pan Arab Project for Child Development (PAPCHILD), which was sponsored by the Arab League Organization, and the model questionnaires of the Demographic and Health Surveys (DHS) in Calverton, Maryland, USA. The final YDMCHS questionnaires were mainly based on PAPCHILD's model questionnaires. The questionnaires were modified to suit the conditions of Yemenl society and to meet the information requirements of the country. A large number of questions were included in the YDMCHS questionnaires in order to obtain as much information as possible on demographic and population dynamics, health and environmental issues, other indicators of standards of living, housing conditions, matemal and child health, and characteristics of local communities regarding provision of health services. English versions of the questionnaires (except the Community Questionnaire) are reproduced in Appendix E. The Household Questionnaire consists of a household roster, including questions on orphanhood, education level and economic activity of household members. It also collects information on general mortality, disability and, for ever-married women under age 55, information on fertility and child survival. The Housing Characteristics Questionnaire, was administered as pan of the household survey. It includes eight sections: housing, cooking, water, lighting, sanitation, waste disposal, ownership of objects and assets, and drainage. The YDMCHS Women's Questionnaire or Reproductive Health Questionnaire consists of nine sections: Respondent's background Marriage and co-residence 7 Reproduction and child survival Antenatal care: current pregnancy Maternal care: the last five years Child feeding Cause of death for children who died Family planning and childbearing attitudes Husband's background The Child Health Questionnaire, which is also referred to as Children's Questionnaire, consists of six sections: General child care Morbidity: diarrhea Morbidity: other illnesses Immunization Weight and height ~ Preparation of interviewer's instructions and training booklets. After completion of the design stage, interviewer's manuals containing instructions on interviewing techniques and on procedures for completing questionnaires were prepared for each questionnaire. The manuals provided simple, clear descriptions of each question as well as of basic terminology used in the survey (e.g., household, live birth). Survey Pretest and Printing of Documents Training of supervisors and female field editors. Twenty men were trained as supervisors for the survey fieldwork and 6 women were trained as interviewers for the pretest and as field editors for the main survey. All were trained in the theoretical aspects of survey research and the practical aspects of completing questionnaires, i.e., understanding the purpose of each question, understanding the instructions for completing the questionnaires, and leaming techniques to obtain specific and accurate answers from the respondents. Practicaltraining. The pretest field practice followed the theoretical training. Trainees interviewed selected households (that were not inchidcd in the main survey) as well as eligible women and eligible children in those households in both Sana'a City and rural areas around the capital. The pretest field practice lasted for 15 days from 25 August until 10 September 1991. Revision of basic survey documents. Based on the experience oftbe pretest interviewing and pretest field operations, the survey documents, questionnaires and instruction manuals were evaluated for appropriateness and adequacy for implementing the YDMCHS. As a number of problems were identified in the documents, some questions and instructions were revised accordingly. Printing of documents and questionnaires. After finalizing the questionnaires and fieldwork manuals, the basic survey documents and other forms to be used in the field for monitoring fieldwork operations and checking quality control were printed. The number of questionnaires printed was as follows: Household Questionnaire (16,500 copies), Housing Characteristics Questionnaire (15,000 copies), Reproductive Health Questionnaire (8,500 copies), IThe analysis of nutritional status is not included in this report. It will be published in a separate report covering a thorough assessment and detailed analysis of height/weight data. Child Health Questionnaire (6,000 copies), Community Questionnaire (350 copies). In addition to the questionnaires, 250 copies of the interviewer's manual were printed to guide interviewers, field editors and supervisors during the training and fieldwork. Training of Female Interviewers From past experience, it was known that survey fieldwork would be difficult to implement. The Technical Committee felt that, in the context of Yemeni culture, female interviewers would not be able to work away from home for long periods. Therefore, it was decided that fieldwork in all parts o f Yemen should be finished within two months. This meant cutting the duration of fieldwork drastically and doubling the number of field teams originally planned. In order for interviewers to be representative of the whole country, every attempt was made to recruit interviewers for the YDMCHS from all the govemorates. Because the number of interviewers to be trained was doubled, the training was conducted simultaneously at two sites: Sana'a in the north and Aden in the south. The Sana'a group included 75 persons (interviewers, editors and supervisors) who later formed nine field teams, which worked in the northern and westem goveroorates. The other group, trained in Aden, included 45 persons (interviewers, editors and supervisors) who later formed seven teams covering the southern and western goveroorates. The experience of pretest training and interviewing was useful in preparation for the training of interviewers for the main fieldwork. Intense theoretical and practical training for both groups was conducted for three weeks by professional trainers who were specialists in survey methodology, statistics and public health. Practical training for anthropometric measurements, that is, use of scales for weighing children and boards for measuring children's height (or recumbent length), followed completion of classroom training for the questionnaires. Female trainees spent one week completing practice interviews on selected households that were not included in the main survey. The fieldwork practice provided an opportunity for interviewers to become familiar with the process and problems of interviewing in the field, and for trainers to evaluate interviewer-trainees on their competence to carry out fieldwork. After the selection o f interviewers, 16 teams were formed to implcment the main survey fieldwork. Main Survey Fieldwork A plan was drawn up to use 16 teams to implement the main fieldwork. Each team included one field supervisor, one male or female field editor, and four or five interviewers. The data collection started on 16 November 1991 when the teams were dispatched to different govemorates to begin fieldwork. The teams retumed from the field on different dates because of differences in workloads and distances from Sana'a to the assigned areas. The last team returned from the field on 15 January 1992. All teams completed their work successfully. During the fieldwork period, the teams were encouraged to call the project's technical staffin Sana'a to keep the survey operations desk informed about the progress of the work and any problems rcquiring assistance. The senior project staff from Sana'a and Aden also visited teams to monitor the quality of fieldwork and to solve any technical or field problems the teams encountered. Whenever possible, the monitoring staff returned to Sana'a with completed questionnaires so that data preparation and data processing could be carded out simultaneously with fieldwork. On receipt of the questionnaires at the central office, the process of revising, editing, coding and processing the data was carried out. On average, an interviewer completed 4 to 6 household interviews daily and 2 or 3 interviews with eligible women. The overall duration of fieldwork was affected by travel time from one area to another. Travel time was often substantial because many of the household clusters were far apart or not linked by roads. The number of interviews completed by individual teams varied due to a number of factors including: the distance between households in rural areas, the number of persons in the households, and the need for repeat visits to households in urban areas to complete interviews with eligible respondents. Data collection was completed on schedule despite difficulties such as rough terrain and poor roads, inadequate accommodations for the field teams, and widely scattered households in some areas. The difficulties were overcome by the efforts and dedication of the supervisors and interviewers. Their enthusiasm and the survey officials' understanding of the nature of this undertaking bad a great impact on solving many of the difficulties the field teams faced. The survey operations desk in Sana'a maintained regular contact by phone with field supervisors. The supervisor called to report their progress, to ask technical questions they were not sure of or could not solve, and to discuss problems encountered in the field for which they needed assistance. Preparation of Data Editing and coding. Data preparation began one week after the start of fieldwork and continued simultaneously with the fieldwork activities. Field editors checked the questionnaires for completeness and consistency. Field supervisors also checked completed questionnaires on a sample basis. Completed questionnaires were then sent to the central office in Sana'a or brought by staff when they returned after visiting the teams. In the central office in Sana'a the questionnaires were edited again, and open-ended and other questions requiring coding were coded. This stage started on 22 November 1991 and was completed by the end of January 1992. Data Entry. Data entry was carried out using microcomputers and continued from January until the end of February 1992. The process of data entry, editing and cleaning was done with ISSA (Integrated System for Survey Analysis) programs specially designed for the DHS surveys. Data consistency checks and production of tables. Consistency checks and data cleaning started on 1 March 1992 and were completed by the end of June 1992. During July 1992 preliminary reports on the Yemen Demographic and Maternal and Child Health Survey were prepared, one in English by Macro International and one in Arabic by PAPCHILD. Both reports were published in August 1992. After the publication of the preliminary report, tabulations for the final report were carried out by both Macro International and PAPCHILD. 1.10 RESULTS OF THE HOUSEHOLD AND WOMEN'S INTERVIEWS Table 1.1 is a summary of the results from the household and women's interviews by urban-rural residence. Of the 13,712 households selected for inclusion in the survey, 13,206 were found and 12,836, or 97 percent, were successfully interviewed. In all, 6,150 ever-married women agc 15-49 years were identi fled in the households selected for individual interviews. Of these, 5,687 women were successfully interviewed and information was collected for 6,715 of 7,022 eligible children under five. The response rates for eligible women and children are 93 and 96 percent, respectively. The response rates for urban and rural areas are almost the same. The main reason for not completing some household interviews was that the dwellings were vacant at the time of fieldwork, although they were occupied when thc household listing was carried out. The principal reason for nonresponse in the case of eligible women was that respondcnts were not at home despite repeated visits by interviewers to the selected households. 10 Table t.1 Results from the household~ women's and children's questionnaires Number of households, eligible women, and eligible children, and response rates, Yemen 1991/92 Residence Result Urban Rural Total Household Interviews Households sampled 3333 10379 13712 Households found 3171 10035 13206 Households interviewed 3077 9759 12836 Household response rate 97.0 Individual interviews Number of eligible women 1568 Number of eligible women interviewed 1453 Eligible woman response rate 92.7 97.2 97.2 4582 6150 4234 5687 92.4 92.5 Children's questionnaires Number of eligible children 1561 5461 7022 Number of eligible children for whom quest, completed 1474 5241 6715 Eligible children response rate 94.4 96.0 95.6 11 CHAPTER 2 BACKGROUND CHARACTERISTICS: HOUSEHOLDS, RESPONDENTS, AND CHILDREN Throughout this report, nuptiality, fertility behavior and regulation, health of mothers and children, and infant and child mortality are examined in terms of the different subgroups of the population. One purpose of this chapter is to profile these subgroups and to describe the environment in which women and children live. First, the results of the hou~hold questionnaire are summarized. The general characteristics of the population are presented, including: housing characteristics (such as type of dwelling, water supply, sanitation facilities, and use of electricity), socioeconomic indicators measured by presence of durable goods in the household; age-sex structure, household arrangement (head of household, size of household) and literacy and educational level of household members. The data are presented by urban-rural residence to highlight where many of the indicators differ. The second purpose of the chapter is to provide a summary of the characteristics of women and children under five for whom data were collected in the YDMCHS. For the sample of women, percent distributions are shown for various demographic and socioeconomic characteristics, such as age, marital status, residence, and educational level. For children under five, age distribution, sex of child, place of residence and mother's educational level (for those whose mothers were interviewed) are presented. 2.1 HOUSING CHARACTERISTICS Table 2.1 summarizes the data obtained from questions in the YDMCHS on housing characteristics: type of dwelling, household congestion (number of persons per room) quality of housing, source of lighting and water, type of sanitation facilities, manner of garbage disposal, etc. Over 85 percent of the households reside in some type of house; 5 percent live in an apartment. Apartment living is more common in urban (18 percent) than in rural areas (2 percent). The quality of housing is better in urban areas; one-half of urban households live in dwellings with cement or tile floors. More than two-thirds of rural households, on the other hand, live in structures with earth or stone/mud floors. Residential congestion is common throughout Yemen, due to the large size of households and limited space for living and sleeping. One-fi fth of households have a density of 2 persons per room and another fifth have a density of 3 persons per room. One-fifth of households have 4 persons per room, while one-third have a density of 5 or more persons per room. Residential congestion is also evident from the figures for sleeping density (i.e., person per sleeping room). On average, sleeping density is 3.9 persons per sleeping room. Congestion in terms of sleeping space is slightly greater in rural than in urban areas. The congested living condition also accommodates farm animals in many cases. More than 61 percent of households in rural areas and 15 percent in urban areas keep animals in the dwelling. There are large differences between urban and rural areas in other socioeconomic indicators (see Table 2.1). Use of electricity is nearly universal (over 90 percent) in urban areas, while only one-third of households in rural areas have electricity. Almost three of five rural households use kerosene lamps or candles for lighting. Nine of 10 households in urban areas have piped water, whereas in rural areas, half of the households obtain water from a well (half of these with a pump and half without a pump); 1 in 5 rural housholds obtains water from a stream, the second most important source of water in rural areas. Flush toilets are present in half of the households in urban areas, while more than half of the rural household have no sanitation facilities at all. In urban areas, the majority of households (69 percent) dispose of garbage by putting it in a special place (dump); one-fourth throw it in the street. In rural areas, the overwhelming majority (9 of 10 households) throw garbage directly into the street. 13 Table 2.1 Housing characteristics Percent distribution of households by housing characteristics, according to urban-rural residence, Yemen 1991/92 Characteristic Urban Rural Total Characteristic Urban Rural Total Type of dwelling Source of drinking water H~se or villa 74.1 89.3 86.6 Government project 75.3 7.3 19.3 Apartment 17.8 1.9 4.7 Cooperative project 5.9 11.6 10.6 Hut 3.8 7.1 6.5 Private project 6.0 4.8 5.0 Sandaka 3.8 0.9 1.4 Well with pump 2.1 23.7 19.9 Cave 0.0 0.1 0.1 Well 3.3 25.6 21.7 Tent or hut 0.0 0.2 0.1 Stream 1.8 18.6 15.6 Temporary shelter 0.0 0.1 0.1 Covered pool 0.9 2.9 2.5 Other 0.3 0.2 0.2 Uncovered pool 0.2 2.7 2.2 Missing 0.1 0.2 0.2 Other 4.5 2.9 3.2 Missing 0.1 0.0 0.0 Total 100.0 100.0 I00.0 Total 100.0 100.0 100.0 Flooring Earth 10.2 41.4 35.9 Source of lighting Tiles/Cement 49.6 27.6 31.5 Govemmem electric 85.0 9.4 22.7 Stone/Mud 12.6 28.5 25.7 Cooperative electric 2.6 6.5 5.8 Gypsum 3.3 0.6 1.0 private electric 2.8 14.9 12.8 Tile 22.2 0.7 4.5 Own generator 0.8 3.2 2.7 Wood 0.4 0.0 0.1 Gas 0.1 2.1 1.8 Marble 0.1 0.0 0.0 Kerosene/lamp/candle 7.0 58.6 49.5 Gther 0.9 0.2 0.3 Other 0.1 0.8 0.7 Missing 0.8 1.0 0.9 None 1.3 4.1 3.6 Missing 0.4 0.4 0.4 Total 100.0 100.0 100.0 Total 100.0 100.0 100.0 No. persons per room <2 7.1 7.0 7.0 Sanitation facilities 2 23.5 18.7 19.5 Flush toilet with sewer 38.1 0.5 7.1 3 24.5 22.0 22.4 Flush toilet without sewer 15.5 1.6 4.1 4 18.9 18.1 18.2 Bucket 19.8 15.5 16.3 5 10.0 12.7 12.3 Pit 17.0 18.8 18.5 6 6.1 7.4 7.2 Toilet connected to 7 3.2 5.3 4.9 open drainage 1.9 7.2 6.3 8 2.2 3.3 3.1 Latrine shared 0.5 0.9 0.8 9 1.3 2.3 2. I Street toilet 0.4 0.5 0.5 I0 2.3 0.5 2.5 Open air 5.9 52.4 44.2 Missing 0.9 0.6 0.7 Other 0.8 2.6 2.2 Missing 0.1 0.0 0.0 Total 100.0 100.0 100.0 Total 100.0 100.0 100.0 Persons per sleeping room <3 40.7 33.4 34.7 Garbage disposal 3-4 36.9 36.5 36.6 Garbage collector 4.0 0.2 0.8 5-6 12.9 16.5 15.9 Dumping (special place) 68.5 5.0 16.2 7+ 8.6 13.0 12.2 Burning 0.8 1.6 1.4 Missing 0.9 0.6 0.7 Thrown m street 23.5 89.9 78.2 Other 2.5 3.0 2.9 Total 100.0 100.0 100.0 Missing 0.8 0.4 0.5 Mean persons/ Total 100.0 100.0 100.0 deeplng room 3.6 4.0 3.9 Number of households 2265 10571 12836 Farm animal in dwelling Yes 14.9 60.9 52.8 No 82.2 36.5 44.5 Missing 2.9 2.6 2.7 To~al 100.0 100.0 1(30.0 14 2.2 PRESENCE OF DURABLE GOODS IN THE HOUSEHOLD Table 2.2 shows the percentage of households with specific durable consumer goods by urban-rural residence. Overall, two-thirds of households have a radio/cassette recorder, one-half have a television, and very few (6 percent) own a video player/recorder. One-third of households have a gas or electric stove, less than one-fifth have a refrigerator, and even smaller proportions of surveyed households have other appliances such as a sewing machine, washing machine, vacuum cleaner, electric fan, or blender. Regarding ownership of a means of transportation, 12 percent of households have a car, 4 percent have a taxi, 4 percent have a bicycle, and 2 percent have a motorcycle. Table 2.2 Household durable goods Percentage of households possessing various durable consumer goods, by urban-rural residence, Yemen 1991/92 Characteristic Urban Rural Total Radio/Cassette recorder 81.8 62.6 66.0 B&W television 33.9 31.0 31.5 Color television 62.9 13.3 22.1 Any television 86.5 41.0 49.0 Video player/recorder 26.4 1.7 6.1 Refrigerator 62.8 7.2 17.0 Gas/electric cooking stove 75.6 23.9 33.0 Water heater 23.2 1.3 5.1 Sewing machine 33.5 9.3 13.6 Electric fan 40.6 4.6 11.0 Washing machine 58.9 4.1 13.7 Telephone 24.8 0.3 4.7 Air conditioner 10.2 0.4 2.l Vacuum cleaner 22.7 1.0 4.8 Blender 50.8 4.6 12.8 Bicycle 10.6 2.2 3.7 Motorcycle 3.2 1.8 2.1 Private car 20.0 10.6 12.3 Taxi 4.8 3.6 3.8 Number of households 2265 10571 12836 Ownership of durable consumer goods varies greatly by residence. As expected, the proportion of households where the specific items are present is much lower in rural areas than in urban areas. This urban- rural differential is particularly strong for video player/recorders, color televisions, refrigerators and other electrical appliances, reflecting the fact that urban households usually have more purchasing power and are three times as likely to have electricity as households in the mral areas. 2.3 HOUSEHOLD POPULATION BY AGE In many developing countries, data on age are affected by errors such as misstatements and preference for or avoidance of certain numberical digits. In order to improve age reporting in cases where age was not given, interviewers were instructed to estimate age using a specially designed historical calendar, with reference to other members of the household whose ages might be reasonably guessed, or based on physiological or sociological l~tctors. 15 Table 2.3 shows the percent distribution of the household population and the sex ratio for five-year age groups, according to urban-rural residence and sex.~ The table and the population pyramid in Figure 2.1 show a young population, a pattern typical of countries with high fertility and high mortality. The median age for the de jure population (i.e., usual residents) is 14.1 years, which means that half of the total population is under this age. The median age is lower (more than one year) for males than for females. Also half of the rural population is under 13.6 years, while half of the urban populaion is under 16.1 years. The proportion of the population under age five is lower, compared to the next older cohort (5-9 years) for all subgroups in the table. With no reason to believe that fertility has declined recently, this "inversion" indicates a possible shifting of children by interviewers from the lowest to the next higher age category to avoid asking a long series of questions about children under five. Another indication of intentional age shifting by interviewers (to lessen workload), is the higher proportion of females in age group 55-59, the age group just outside the age boundary for eligibility for the individual interviewing, than in age group 50-54 years in both urban and rural areas. 2 The sex ratios (number of males per female) below age 25 are uniform, then, through age 54, they are irregular, but slightly higher for urban populations. The latter is probably due to the migration of men from rural to urban areas or emigration of men to neighboring countries. However, the higher sex ratios Table 2.3 Household population by age, residence and sex Percent distribution of the de jure houseehold pepulation by five-year age groups, according to urban-rural residence, sex, and sex ratio. Yemen 1991/92 Urban Rural Total Age Sex Sex Sex ga)up Male Female Total ratio Male Female Total ratio Male Female Total ratio 0-4 15.2 14.9 15.0 1.11 18.2 17.4 17.8 1.0 17.6 16.9 17.2 1.0 5-9 16.9 17.3 17.1 1.0 ~.7 19.7 20.2 1.0 20.0 19.2 19.6 1.0 10 14 15.3 14.8 15.0 1.1 16.5 14.3 15.4 1.1 16.2 14.4 15.3 1.1 15-19 11.2 11.0 11.1 1.1 8.4 8.5 8.5 1.0 9.0 9.0 9.0 1.0 20-24 7.9 8.1 8.0 1.0 4.5 5.4 5.0 0.8 5.2 5.9 5.6 0.9 25-29 6.5 7.6 7.1 0.9 4.7 6.6 5.7 0.7 5.1 6.8 5.9 0.7 30-34 5.3 5.4 5.4 1.0 4.0 5.2 4.6 0.7 4.2 5.2 4.7 0.8 3539 4.6 4.8 4.7 1.0 4.(1 5.0 4.5 0.8 4.1 5.0 4.6 0.8 4044 3.9 3.2 3.6 1.3 3.6 3.5 3.5 1.0 3.6 3.4 3.5 1.1 45-49 3.2 2.5 2.8 1.3 2.8 2.9 2.8 0.9 2.8 2.8 2.8 1.0 50-54 2.2 2.2 2.2 1.0 2.8 2.1 2.4 1.3 2.7 2.1 2.4 1.3 55-59 1.8 3.0 2.4 0.6 1.8 2.9 2.4 0.6 1.8 2.9 2.4 0.6 60-64 2.2 2.0 2.1 1.1 2.9 2.5 2.7 1.1 2.7 2.4 2.6 1.1 65-69 1.0 0.9 0.9 1.2 1.3 1.0 1.1 1.2 1.2 1.0 1.1 1.2 70-74 1.3 1.1 1.2 1.2 1.8 1.4 1.6 1.3 1.7 1.3 1.5 1.3 75-79 0.5 0.4 0.4 1.2 0.6 0.5 0.6 1.2 0.6 0.5 0.5 1.2 80+ 0.8 0.7 0.8 1.3 1.3 1.0 1.1 1.3 1.2 0.9 1.1 1.3 Missing/Don't~ow 0.1 0.3 0.2 0.4 I).2 0.2 0.2 1.0 0.2 0.2 0.2 0.9 Total 100.0 100.0 1~.0 1.03 100.0 100.0 1~.0 0.98 100.0 1~.0 100.0 0.99 Number 8369 8147 16516 NA 34350 34970 69321 NA 42719 43117 85836 NA M~ian 16.0 16.2 16.1 NA 13.0 14.4 13.6 NA 13.5 14.8 14.1 NA Note: Table is based on de jure members; i.e., usual residents. Sex ratio is the number of males per female. NA = Not applicable ~The total number of households in urban and rural areas in Table 1.1 and Table 2.3 are different because Chapter 1 presents unwcighted data and Chapter 2 presents weighted data. Sana'a City, the urban areas of Aden govemorate and rural areas of Laheg governorate were ovcrsampled, 21n the YDMCHS all ever-married women age 10-54 were considered eligible and were interviewed. However, in this report analyses are based on data for ever-married women 15-49 only. 16 Age 80+ 75-79 70-74 65-69 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5-9 Figure 2.1 Population Pyramid of Yemen, 1991/92 0-4 10 5 0 5 10 Percent YDMCHS 1991/92 for ages 45-49 and 50-54 and the very low sex ratio (0.6) for age group 55-59 indicates artificial "aging" of women who were eligible for the individual interview (i.e., under 55 years of age). It is likely that interviewers were reporting some women as older than 54 so they could reduce their workload. Table 2.4 shows the age distribution of the Yemeni population by larger age groups. More than half of the population is under age 15, and around 7 percent are 60 and over. Forty-one percent are in the economi- cally active age category 15-59. Thus, the dependency ratio for Yemen is 144. 3 This extremely high depend- ency ratio is due to the large proportion of the population under age 15. As no national census has been done in Yemen and no other national survey has been imple- mented since unification, the only age distribution data with which the YDMCHS can be compared are the 1990 UN estimates (see Table 2.4) (United Nations, 1993). According to the 1990 UN estimates, almost half of the population is under 15 and 4 percent is over 59. Table 2.4 Population by age from selected sources Percent distribution of the population by age group, UN estimates and YDMCHS, Yemen UN 1 YDMCHS Age group 1990 1991/92 <15 49.6 52.1 15-59 46.4 40.8 60+ 4.0 6.9 Missing/don't know 0.2 Total 100.0 100.0 Median age U 14.1 Note: Toals may not add to 100 due to rotmding. U = Unknown ]United Nations, 1993 3The dependency ratio is 100 times the population under age 15 plus the population 60 and over (persons economically "dependent") divided by the population in the economically active ages, persons 15 to 59. Dependency ratios for selected Near Eastern countries are: Tunisia 1988: 86; Morocco 1992: 89, Jordan 1990: 94; and Sudan 1989/90: 95. 17 2.4 HOUSEHOLD COMPOSIT ION Table 2.5 presents information on household composition according to urban-rural residence for the de jure population (i.e., usual residents). Eighty-eight percent of the households covered in Yemen are headed by men, while about 12 percent have female heads. There is a difference between urban and rural areas in the proportion of households headed by men and wom- en. Less than 10 percent of households are headed by women in urban areas, whereas in rural areas 13 per- cent of households have a woman as the head. In terms of household size, large households are common in Yemen. The average number of members in a house- hold is 6.7 for the country as a whole, while urban households are larger than those in rural areas (7.3 and 6.6, respectively). One-third of urban households com- pared with one-fourth of rural households are com- posed of nine or more members. Joint and extended family living arrangements are the norm in Yemen. Single adult households are rare. About half of the households have three or more related adults, while almost 38 percent have two related adults of the opposite sex. Joint and extended families are more common in urban than in rural areas. 2.5 EDUCATIONAL STATUS OF HOUSEHOLD POPULAT ION In the YDMCHS, for all household membe~ 10 years and older, questions were asked to determine literacy and educational status: illiterate, can read only, literate (can beth read and write), completed primary, completed preparatory, secondary, post-secondary school or university. For each household member who Table 2.5 Household composition Percent distribution of households by sex of head of household, household size, and relationship sffucture, according to urban rural residence, Yemen 1991/92 Characteristic Urban Rural Total Household headship Male 90.5 87.2 87.8 Female 9.5 12.8 12.2 Total 100.0 100.0 100.0 Number uf usual members 1 3.4 3.1 3.2 2 5.9 8.1 7.7 3 6.2 8.2 7.8 4 9.0 9.0 9.0 5 9.5 11.8 11.4 6 11.5 11.8 11.7 7 11.6 12.4 12.2 8 10.7 10.5 10.6 9+ 32.2 25.1 26.3 Total 100.0 100.0 100.0 Mean size 7.3 6.6 6.7 Relationship structure One adult 4.9 8.2 7.6 Two related adults: Of opposite sex 30.3 39.5 37.9 Of same sex 1.9 2.2 2.2 Three or more related adults 58.9 49.0 50.7 Other 4.0 1.1 1.6 Total 100.0 100.0 100.0 Number of households 2265 10571 12836 Note: Table is based on de jure members; i.e., usual residents. was reported to have had no schooling or who had attended school but had not completed primary school, interviewers probed to determine the member's ability to read and write or his/her literacy status and then marked the appropriate level of literacy. Tables 2.6.1 and 2.6.2 show the percent distribution of the male population and the female population age 10 and over by level of education or literacy according to age, residence, and region. Almost one-third of men (see Table 2.6.1) in the YDMCHS household sample are illiterate, 15 percent can read (11 percent also can write) but have no schooling or have not completed the primary level, 38 percent have completed the primary level of school, and 15 percent have completed more than primary education. A substantial improvement in educational attainment over time can be seen in the fact that younger men have attained higher levels of education than older men. With increasing age, the proportion of men who cannot read increases and the proportion who have completed more than primary school decreases. As expected, urban men tend to be more educated than tbeir rural counterparts. For example, illiteracy in rural areas is more than double that in urban areas; and, urban men are three times as likely to have attained more than primary 18 Table 2.6.1 Educational level of the male household populat ion Percent distribution of the de jure male household populations age 10 and over by highest level of education completed, according to selected background characteristics, Yemen 1991/92 More Number Background Read than of characteristic Illiterate only Literate Primary primary Missing Total males Age 10-14 6.4 5.1 0.1 85.8 1.8 0.8 100.0 6930 15-19 7.8 1.2 0.3 56.9 33.3 0.5 100.0 3830 20-24 14.0 1.1 1.6 35.6 47.3 0.4 100.0 2215 25-29 29.6 2.5 7.5 32.4 27.6 0.3 100.0 2176 30-34 38.5 3.4 I7.2 19.2 20.9 0.7 100.0 1814 35-39 44.9 4.6 28.1 7.3 14.6 0.6 100.0 1760 40-44 55.1 3.5 29.2 3.7 8.0 0.5 I00.0 1558 45-49 59.7 4.5 26.7 3.0 5.7 0.6 I00.0 1216 50-54 64.1 4.4 25.6 1.5 3.1 1.3 100.0 1154 55-59 63.9 5,3 27.1 0.9 2.2 0.6 100.0 774 60-64 71.6 5.1 21.6 0.6 0.4 0.6 100.0 1173 65+ 73.9 4.4 19.2 0.2 0.7 1.6 100.0 2005 Missing/Don't know 87.2 7.8 0.0 0.0 0.0 5.0 100.0 24 Residence Urban 16.3 3.3 10.5 38.4 30.8 0.6 100.0 5674 Rural 35.4 3.8 11.2 38.4 10.5 0.7 100.0 20955 Region North./West. 34.2 3.9 11.5 38.0 11.8 0.6 100.0 21661 South./East. 18.8 2.6 9.0 40.1 28.2 1.2 100.0 4968 Total 31.4 3.7 11.1 38.4 14.8 0.7 100.0 26629 Table 2.6.2 Educational level of the female household populat ion Percent distribution of the de facto female household populations age 10 and over by highest level of education completed, according to selected background characteristics, Yemen 1991/92 More Number Background Read than of characteristic Illiterate only Literate Priraary primary Missing Total females Age I0-14 54.3 4.0 0.5 39.8 0.3 I.I I00.0 6202 15-19 60.4 1.9 I.I 26.5 9.2 0.9 I00.0 3881 20-24 70.1 2.0 0.8 16.1 I0.5 0.3 I00.0 2556 25-29 81.6 2.2 0.7 9.4 5.8 0.4 I00.0 2914 30-34 88.5 1.6 2.5 3.2 3.7 0.4 I00.0 2262 35-39 91.3 1.4 3.2 I.I 2.2 0.8 I00.0 2149 40-44 95.9 0.9 1.6 0.2 0.6 0.8 100.0 1467 45-49 97.1 0.3 1.3 0.2 0.4 0.7 100.0 1210 50-54 96.4 1.1 1.3 0.0 0.1 1.0 100.0 903 55-59 97.6 0.3 0.7 0.1 0.2 1.2 100.0 1270 60-64 96.6 0.4 0.3 0.0 0.3 2.3 100.0 1032 65+ 97.2 0.3 0.1 0.0 0.0 2.5 100.0 1612 Missing/Don't know 96.1 0.0 0.0 0.0 0.0 3.9 100.0 32 Residence Urban 45.9 3.1 2.8 33.9 13.7 0.6 100.0 5509 Rural 84.6 1.7 0.7 11.0 1.0 1.0 100.0 21980 Region North./West. 81.0 1.9 0.7 13.7 1.8 0.9 100.0 22378 South./East. 58.6 2.4 2.8 23.9 11.0 1.3 100.0 5111 Total 76.9 2,0 1.1 15,6 3.5 0.9 1(20.0 27489 19 education as men residing in rural areas. Regionally, there are marked variations in the educational status of men. In the northern and western governorates 34 percent of men are illiterate, compared to 19 percent in the southem and eastern govemorates. Likewise, whereas 70 percent of men in the southern and eastern governorates have completed primary or post-primary eduation, only 50 percent of men in the northern and westem governorates have attained this level. Women in Yemen have much less education than men but there has been more dram atic improvement in women's than men's educational attainment in recent years. More than three-quarters of women age 10 and over in the YDMCHS household sample are illiterate, 3 percent can read (1 percent can also write), 16 percent have completed primary school, and 4 percent have completed preparatory school or higher (see Table 2.6.2). The recent improvement in female education is indicated by the fact that only 54 percent of women in age 10-14 are completely illiterate, compared to 89 to 98 percent of women over 30. The percentage of women with more than primary education increases steadily in each successively younger age group. Not surprisingly, women who reside in urban areas have considerably more education than those living in rural areas. Eighty-five percent of rural women are illiterate, compared to 46 percent of urban women, and the proportion who have primary education is three times as high in urban as in rural areas. One percent of rural women have attained post-primary education while 14 percent of the urban women have completed preparatory or higher education. As with men, educational attainment for women in the southem and eastem govemorates is higher than for women in the northern and western govemorates. One-third of women in the southern and eastern governorates, or twice the proportion in the northern and westem govemorates, have completed primary or post-primary education. 2.6 SCHOOL ENROLLMENT The Household Questionnaire for the YDMCHS asked about the current status of school attendance for all persons in the household between the ages of 6 and 30 years. Table 2.7 indicates that 57 percent of school age children (6-15 years) are enrolled in school. The education of females lags far behind the education of males in Yemen. 4 Eight in 10 of school age males attend school compared to only about 3 in Table 2.7 School enrollment Percentage of the de jure household population age 6-29 years enrolled in school, by age group, sex, and urban rural residence, Yemen 1991/92 Male Female Total Age group Urban Rural Total Urban Rural Total Urb~m Rural Total 6-10 79.7 70.0 71.6 75.0 26.3 34.5 77.4 48.8 53.6 11-15 94.8 89.1 90.2 81.1/ 21.2 33.0 88.1 56.7 62.8 6-15 86.6 77.9 79.4 77.7 24.2 33.9 82.3 52.0 57.4 16-20 66.5 58.7 60.7 38.4 6.3 14.0 52.6 31.5 36.8 21-24 36.1 25.0 29.0 10.1 1.8 4.2 23.3 12.0 15.7 25-29 11.6 7.8 8,8 3.6 0.9 1.5 7.3 3.8 4.6 4 For example, according to the DHS survey in Jordan (Zou'bi et al., 1992), attendance of children 6-15 years is almost universal (95 percent), and enrollment of boys over girls is higher by only 1 percent nationally and by less than 4 percent in rural areas. Compared with the Morocco DHS survey (Azelmat et al., 1993), overall school enrollment for children 6-15 years in Yemen is higher (57 percent versus 47 percent) and gender enrollment ratios that is, the proportion of girls to the proportion of boys enrolled in school~o not differ fur children in urban areas in the two countries. However, enrollment ratios in rural areas are 49 percent in Morocco and only 31 percent in Yemen. 20 10 females. The difference in enrollment of school age children by gender is mainly due to the strikingly lower school attendance of girls in rural areas (24 percent) compared to boys (78 percent). The proportion of boys and girls attending schools in urban areas varies by less than 10 percentage points in favor of boys. More than one-third of the population age 16 to 20 attends school, but by age 21 to 24, a much smaller proportion is still in school. Relatively higher proportions of males than females and urban children than rural children remain in school in these two age groups. Less than 5 percent in the population age 25 to 29 is in school. 2.7 BACKGROUND CHARACTER- ISTICS OF WOMEN Age The low level of individual awareness of exact age in Yemen increased the likelihood of the occurrence of age misreponing. Age misreporting was also caused by estimating women's ages when they could not give their exact age. In some cases, age was not known at all and the interviewers had to estimate it by various means (scc section 2.3). Table 2.8 Background characteristics of respondents Percent distribution of ever-married women by selected background characteristics, Yemen 1991/92 Number of women Background Weighted Un- characteristic percent Weighted weighed Age 15-19 7.5 427 430 20-24 14.3 815 832 25-29 22.8 1295 1297 30-34 17.5 995 986 35-39 17.1 972 970 40-44 11.5 653 648 45-49 9.3 529 524 Marital status Married 94.2 5355 5336 Widowed 2.9 164 167 Divorced 2.6 150 168 Separated 0.3 18 16 Residence Urban 18.5 1054 1453 Rural 81.5 4633 4234 Region Nor th./West. 82.8 4708 4010 South.lEast. 17.2 979 1677 Education Illiterate 89.2 5075 4808 Literate 1.1 61 87 Primary 5.6 316 406 More than primary 4.1 235 386 Total 100.0 5687 5687 Table 2.8 shows the percent distribution of respondents by age group and various background characteristics. Nearly one-fi fth of ever-married women interviewed in the survey are under age 25, slightly more than one-fifth am in their late twenties, and one-third are in their thirties. More than one-fifth are in their forties, concentrating in the early forties. Current Marital Status Except for 3 percent of women who were widowed and 3 percent who were divorced, all women 15- 49 who had ever married were currently married at the time of the survey. Place of Residence Less than one in five (18.5 percent) of the women in the sample reside in urban areas with the remainder (81.5 percent) living in rural areas. More than four of five women surveyed live in the northem and western govcmoratcs and less than one-fifth reside in the southem and eastern governorates. 21 Literacy and Level of Education As can be seen in Table 2.8, an overwhelming majority of women surveyed in Yemen have received little or no formal education and are illiterate. Among ever-married women, only 6 percent have completed primary school and 4 percent have completed preparatory or higher schooling; 89 percent of Yemeni women are illiterate, irrespective of whether they had attended primary school or not. Table 2.9 examines variations in the educational attainment of women by age, residence, and region. As mentioned earlier, education is negatively associated with age; that is, older women are generally less educated than younger women. For example, whereas three-fourths of women age 15-19 are illiterate, more than 90 percent of women age 30-34 and 99 percent of women 45-49 are illiterate. The increase in the proportion of ever-married women who have completed primary school is most noticeable since the 1966 revolution. 5 Table 2.9 Level of education Percent distribution of ever-married women by the highest level of education completed, according to selected background characteristics, Yemen 1991/92 Level of education Number Background More than of characteristic Illiterate Literate Prunary primary Total women Age 15-19 74.4 2.2 16.2 7.3 100.0 427 20-24 79.4 1.2 11.7 7.7 100.0 815 25-29 87.9 0.8 6.0 5.3 100.0 1295 30-34 91.2 1.2 3.9 3.7 100.0 995 35-39 93.2 1.5 2.6 2.7 100.0 972 40-44 97.3 0.6 1,l 1.0 100.0 653 45-49 98.6 0.3 0.5 0.6 100.0 529 Residence Urban 66.2 2.7 14.8 16.3 100.0 1054 Rural 94.5 0.7 3.5 1.4 100.0 4633 Region Nor th./West. 93.5 0.6 4.0 1.8 100.0 4708 South./East. 68.6 3.5 12.8 15.1 100.0 979 Total 89.2 1.1 5.6 4.1 100.0 5687 Women who reside in urban areas have a lower level of illiteracy and a higher level of education than those living in rural areas. The proportion of ever-married women who cannot read and write is substantially higher in rural (95 percent) than in urban areas (66 percent). Converscly, the percentage who have post-primary education is more than ten times greater in urban areas than in rural. Regarding differentials by region, educational attainment is much higher in the southem and eastern govemorates where more than one-fourth of ever-married women have completed at least primary school compared with only 6 percent of women in the northern and western governorates. 5 The progress in female education is more obvious in Table 2.6.2 which indicates that among women 15-19, 60 percent are illiterate compared to 74 percent of ever-married women in Table 2.9. The longer women slay in school, the later they will marry, thereby raising age at marriage and lowering fertility. 22 Exposure to Mass Media Table 2.10 shows the percentage of ever-married women who are exposed to mass media, i.e., newspapers/magazines, television, and radio, according to selected background characteristics. The respondents were asked if they watched television or listened to radio (without reference to how often they were exposed to these media). Reading newspapers/magazines at least once a week was considered the measure of exposure to this medium. Less than 10 percent read a newspaper or magazine at least once a week, one-third listen to radio, and a slightly higher proportion watch television. Exposure to all media decreases with age; this is especially true for the print medium because the percentage of women who are literate decreases in the older age cohorts. Table 2.10 Exposure to mass media Percentage of ever-married women who usually read a newspaper once a week, watch television, or listen to radio, by selected background characteristics, Yemen 1991/92 Read Number Background newspaper Watch Listen to of characteristic weekly television radio women Age 15-19 17.8 45.9 46.1 427 20-24 15.8 44.2 42.8 815 25-29 8.7 39.2 36.3 1295 30-34 7.0 33.9 32.9 995 35-39 5.6 36.1 30.0 972 40-44 2.7 31.1 25.9 653 45-49 1.1 29.7 24.1 529 Residence Urban 28.4 82.6 57.7 1054 Rural 3.6 26.8 28.6 4633 Region Nor th./West. 4.8 31.9 31.7 4708 South./East. 24.2 62.4 44.9 979 Education Illiterate 0.5 31.9 29.9 5075 Literate 55.7 77.6 70.7 61 Primary 67.6 76.4 65.8 316 More than primary 80.8 86.2 70.6 235 Total 8.2 37.1 34.0 5687 The data on exposure to mass media also indicate a positive relationship between education and reading newspapers/magazines. Eighty percent of women with post-primary schooling read newspapers, compared with 56 percent of women who are just literate (can read and write) . 6 Level of education, which may also signify socioeconomic level, is related to exposure to radio and television. Only 3 in 10 illiterate women watch television or listen to radio, whereas 66 to 86 percent of educated women are exposed to television and radio. Women who are literate are slightly more likely to listen to radio than those who have completed primary school; however, there is practically no diffcrence among these two subgroups in the proportion of women who watch television. 6It should be noted that women who claim to be able to read but cannot write are classified as illiterate in Table 2.10. 23 Because of the higherlevels of literacy in urban compared to rural areas, more than one-quarter of urban women read the news- paper compared to less than 4 percent of rural women. As expected, exposure to other media is also related to residence; 83 and 58 percent of urban women watch television and listen to radio, respectively, compared to 27 and 29 per- cent of rural women, respectively. There are also substantial regional differences: 5 percent of women in the northem and western gover- norates and 24 percent of women in the south- em and eastern govemorates read newspapers or magazines. It is twice as likely for women in the southern and eastern governorates to watch television as in northern and western governor- ares. The same proportion of women listen to radio and watch television in the northern and western govemorates (32 percent), but in the southem and eastern govemorates while two- thirds of women watch television (62 percent) less than one-half listen to radio. 2.8 CHARACTERIST ICS OF CHILDREN In the YDMCHS, information about births in the five-year period preceding the sur- vey was collected in the Women's Question- naire, and a separate questionnaire, the Child's Questionnaire, was u~d to collect information about children under five living in the house- hold. The characteristics of children are pre- Table 2.11 Background characteristics of children Percent distribution of children by selected background characteristics, Yemen 1991/92 Number of children Background Weighted Un- characteristic percent Weighted weighted Child's age (months) 0-5 10.7 718 723 6-11 12.0 802 796 12-17 9.3 627 634 18-23 9.4 628 621 24-29 l l .0 735 752 30-35 10.3 694 683 36-41 11).8 724 722 42-47 8.9 597 588 48 53 10.0 673 684 54-59 7.7 515 512 Sex of child MNe 51.0 3427 3431 Female 49.0 3288 3284 Residence Urban 16.6 1113 1474 Rural 83.4 5602 5241 Region Nor th./West. 86.3 5793 5094 South./East. 13.7 922 1621 Mother 's education Illiterate 88.4 5939 5692 Literate 0.9 58 80 Primary 4.6 306 403 More than primary 3.1 211 343 Information not collected 3.0 202 197 Total 100.0 6715 6715 sented in Table 2. I 1. It is apparent from the table that some misdating of births has taken place. For example, the small proportion o f children 54-59 months compared to other age categories indicates that some children age 54-59 months were moved into the next older age group by interviewers. It is likely that this was done intentionally in some instances to reduce the workload. Questions about children under live in the Women's Questionnaire would not have to bc asked and the Child's Questionnaire would not have to be completed. Fifty-one percent of living children are males and 49 percent are females. The proportion of children residing in urban areas is 17 percent; 14 percent of children for whom the Child's Questionnaire was completed reside in southern and eastern govemorates and 86 percent live in northern and western governorates. The educational level of mothers of children under five is similar to that of ever-married women. Information on mother's educational level is not available because they were not included in the survey. Later in the report, it is shown as, "Information not collected." 24 CHAPTER 3 FERTILITY A major objective of the 1991/92 YDMCHS is to examine a number of important issues related to fertility and childbearing. This chapter briefly covers a number of substantive issues such as fertility levels, trends and differentials, indicators of future fertility, age at which women initiate childbearing, birth intervals, and adolescent fertility. The fertility estimates reported below are based on reported information collected from birth histories of ever-married women age 15-49. Although the YDMCHS collected birth histories for ever-married women only, it is possible to calculate fertility measures for all women. To do this, the assumption is made that single women and never-married women have had no children. While some births undoubtedly occur outside of marriage, most observers agree that the level of non-marital fertility in Yemen is negligible. Current, past and future fertility estimates are based on carefully collected data. First, a series of questions about live births were asked to know the number of children each ever-married woman had in her lifetime. To encourage complete reporting, the respondents were asked about the number of sons and the number of daughters who (a) were living at home, (b) were living elsewhere, and (c) died after birth. Each respondent was also probed by the interviewer to verify if the sum of live births recorded in these categories agreed with the total number of births the woman had. Second, a full birth history was collected from each woman, including the name, sex and date (year and month or season) of each birth, and age at death for children who died. It was also noted in the birth history if year of birth was checked with, or derived from, a document such as birth certificate. Third, the number of stillbirths and miscarriages or abortions reported by respondents were noted. Fourth, as an indicator of future fertility, all currently women were also asked if they were pregnant at the time of the interview. Despite efforts to minimize errors in collecting data on births, the YDMCHS is subject to the same type of errors that are typical of retrospective surveys. The birth histories are affected by underreporting of the number of children ever born and mistiming of births. The main reason for not getting accurate information on number of births (which includes underrcporting of children) is the omission of children who died in infancy or who married and left the parental or maternal home. Also, some women with no surviving children may report themselves as childless. Problems of this type more often occur in surveys in countries where the level of female literacy is low. A check on the data quality of the YDMCHS confirms that some omission of births in the five-year period preceding the survey, and the displacement of births out of this period has taken place. The apparent reason for these errors is that interviewers were trying to avoid completing the Child's Questionnaire in some households with children under five. Typically, births arc displaced more often than they are omitted. 3.1 LEVELS AND DIFFERENTIAL IN FERT IL ITY Table 3.1 presents the age-specific fertility rates and various summary measures such as the crudc birth rate (CBR), the general fertility rate (GFR) and the total fertility rate (TFR), calculated from survey data, according to urban-rural residence. The crude birth rate is the number of births per 1,000 population; it is the least refined measure of fertility, but it is the most commonly used and easily understood. The general fertility rate is the annual number of live births per 1,000 women; it is calculated by dividing the number of births occurring during a specified period of time by the total number of women of reproductive age (15-49 years). The age-specific fertility rate is defined as the number of births to women of a given age group per 1,0(X) women in that age group. The total fertility rate (TFR) is a summary measure that indicates the number of children a women would bear during hcr reproductive years if she were to experience the age-specific 25 fertility rates prevailing for the specified period. Mathe- matically, the TFR for women 15-49 is five times the sum of the age-specific fertility rates for the seven age groups. The crude birth rate (CBR) in Yemen, which is cen- tered on the year 1990, is 40 births per thousand population. It is substantially higher in rural areas (42) than in urban areas (34). The general fertility rate (GFR), estimated from the survey data for the country as whole, is 238 births per thousand women aged 15-49; it is also much higher in mral areas (254) than in urban areas (176). The TFR for Yemen for the three years preceding the survey is 7.7 births per wo- man, which is one of the highest total fertility rates in the world, t With respect to urban-rural differentials, the total fertility rate for urban areas (5.6) 2 is, as expected, much lower than the TFR for rural areas (8.2). The age-specific fertility rates for the three years preceding the survey indicate that the prime childbearing years in Yemen are ages 20-39 (see also Figure 3.1). How- ever, a substantial amount of childbearing is evident even among older women. Age-specific fertility rates increase with age from 102 births per 1,000 women in age group 15-19 to 315 births per 1,000 in age group 25-29, and then decline to 120 births per 1,000 in age group 45-49. Age-spe- cific fertility rates for rural areas are substantially higher than those for urban areas, and follow the same pattern (ex- Table 3.1 Current fertility Age-specific and cumulative fertility rates and the crude birth rate for the three years preceding the survey, by urban-rural residence, Yemen 1991/92 Age group Urban Rural Total 15+19 64 114 102 20-24 212 307 283 25-29 262 328 315 30-34 253 290 284 35-39 197 272 258 40-44 88 191 172 45-49 I351 [140[ I1201 TFR 15-49 5.6 8.2 7.7 TFR 15-44 5.4 7.5 7.1 GFR 176 254 238 CBR 34 42 40 Note: Rates are for the period 1-36 months preceding the survey. Rates for age group 45-49 may be slightly biased due to truncation TFR: Total fertility rate, expressed per woman GFR: General fertility rate (births divided by number of women 15-44) expressed per 1,000 women. CBRi Crude birth rate, expressed per 1,000 population. cept that women age 45-49 in urban areas report lower fertility than women age 15-19). Another way of looking at age-specific rates is to say that if current rates remained unchanged, the average woman in Yemen would have almost two children (1.9 children) by the time she reached age 25; she would have three more children during the next 10 years (between age 25 and 34), by her fortieth birthday she would have given birth to more than six children; and at the end of her reproductive years she would have had an average of more than 7.5 children. L The total fertility rates for five Arab countries in which DHS surveys have been implemented are: Egypt 1992:3.9 (EI-Zanaty et al., 1993), Morocco 1992:4.0 (Azelmat et al., 1993), Tunisia 1988:4.4 (Aloui et al., 1989), Sudan 1989/90:5.0 (DOS and IRD, 1991), and Jordan 1990:5.9 (Zou'bi et al., 1992). 2The TFR for small urban areas in Jordan is the same as the TFR for urban areas in Yemen. The 1PRS for urban women in other Arab countries range from 2.5 in the Morocco 1992 survey (Azehnat et al., 1993) and 2.9 in the Egypt 1992 survey (EI-Zanaty et al., 1993) to 3.4 in the Tunisia 1988 survey (Aloui et al., 1989) and 4.1 in the Sudan 1989/90 survey (DOS and IRD, 1991). 26 Figure 3.1 Age-specific Fertility Rates by Urban-rural Residence Births per 1,000 women 350 300 250 200 150 100 ) 50 ~ 0 15-19 20-24 25-29 30-34 35-39 40-44 Age 45-49 YDM CHS 1991/92 Figure 3.2 Total Fertility Rate (TFR) and Mean Number of Children Ever Born (CEB) to Women 40-49 TOTAL RESIDENCE Urban Rural REGION North,/West South./East EDUCATION No I 0 2 4 6 8 Number of children 10 YDMCHS 1991/92 27 Differentials in fertility by urban-rural residence and level of education are shown in Table 3.2 and Fig- ure 3.2. It has been pointed out that rural areas have higher TFRs than urban areas. Regionally, the difference in the TFRs is even greater. The TFR for the northern and western governorates (8.2 children per woman) is almost 50 percent higher than the TFR for the southern and eastem govemorates (5.5). Fertility rates are related not only to the geographical area in which a woman re- sides but also to her level of education. Women who have received no formal education have a TFR of 8.1,2.4 children higher than for women who have received pri- mary education (5.7), and 4.6 children higher than for women who have received more than primary education (3.5). Table 3.2 also shows the mean number of chil- dren ever born (CEB) to women age 40-49. This is a measure of completed fertility; i.e., the cumulative fer- tility of women of reproductive age, who are approaching the end of their childbearing years. A comparison of the cumulative measure of childbearing, CEB, with the TFR gives a rough indication of the trend in fertility over the recent decades. For all women, the mean number of chil- dren ever born is 7.8, which is very close to the TFR tbr the country as whole. Together, these figures indicate Table 3.2 Fertility by background characteristics Total fertility rate for the three years preceding the survey and mean number of children ever born to women age 40-49, by selected background characteristics, Yemen 1991/92 Mean number of children Total ever born Background fertility to women characteristic rate I age 411-49 Residence Urban 5.6 7.9 Rural 8.2 7.8 Region North./West. 8.2 8.0 South/East 5.5 6.9 Education No schooling 8.1 8.0 Primary 5.7 5.7 More than primary 3.5 3.9 Total 7.7 7.8 tRate for women age 15-49 years that overall fertility has remained virtually unchanged in Yemen. Fertility has increased slightly in some subgroups and decreased in others. A slight rise in fertility occurred among women in rural areas, those residing in the northern and western governorates and those who had no formal education; this is inl~rred from the higher TFRs compared with the CEBs. On the other hand, there is some indication that lertility may be declining in urban areas, in the southern and eastern governorates, and among women who have more than a primary level of education. 3.2 TRENDS IN FERT IL ITY The fertility indicators presented in Table 3.2 suggest that, overall, there has been almost no decline in fertility in recent years. Although there are no estimates from other sources with which to study trends in fertility, data from the birth histories collected in the YDMCHS make it possible to analyze fertility trends directly. Table 3.3 shows age-specific fertility rates for successive four-year periods during the 20 years preceding the survey. Four-year periods were used instead of the customary five-year periods 3 in order to avoid the effects of displacement of births from five years preceding the survey to six years before the survey. It should be noted that some of the fertility rates shown in the table are truncated; this is due to the fact that 3An examination of the same table for fivc-year periods shows that fertility levels peaked at 5 to 9 years bcibre the survey, indicating age displacement of births has taken place. The apparent reason for displacement is that interviewers may have tried to avoid (1) a set of questions about children included in the Woman's Questionnaire, (2) the Child Health questionnaire itself, and (3) the additional burden of measuring height and weight of children born during the live years preceding the survey. 28 the data on which these rates arc based pertain only to women who were under 50 years of age at the time of the sur- vey. Information presented in the table should be treated with caution due to the possible omission or incorrect dat- ing of events, especially by older wom- en, for the more distant time periods. Table 3.3 indicates that fertility peaked in Yemen between 8 and 15 years preceding the survey (approxi- mately 1976-83). An extremely sharp rise in the period 12-15 years before the survey (approximately 1976-79), does not seem plausible. The large increase in fertility may be partially explained by the difficulty older women have in recalling birth dates and/or births that occurred a long time ago. Table 3.3 Age-specific fertility rates Age-specific fertility rates for four-year periods preceding the survey, by mother's age at the time of birth, Yemen 1991/92 Number of years preceding the survey Mother's age 0-3 4-7 8-11 12-15 16-19 15-19 104 184 198 201 158 20-24 286 347 380 356 271 25-29 313 388 414 400 [3001 30-34 284 376 379 [3741 [3111 35-39 251 317 [3501 [3581 40-44 174 201 I243] 45-49 [1131 1309] Note: Age-specific fertility rates are per 1,000 women. Estimates enclosed in brackets are truncated. Table 3.4 shows the pattern of fertility by marital duration, i.e., the number of years since first marriage; the pattern is similar to that observed in Table 3.3. Fertility peaks among women who were first married 8 to 11 years before the survey, and is lowest among women who were first married 0 to 3 years before the survey. Women who had been married for more than 15 years reported lower fertility than women with shorter marriage durations. Again, this is probably due to recall problems of births which occurred a long time ago. Table 3.4 Fertility by marital duration Fertility rates for ever-married women by duration since first marriage in years, for four-year periods preceding the survey, Yemen 1991/92 Marriage duration at birth Number of years preceding the survey 0-3 4-7 8-11 12-15 16-19 0-4 328 356 330 307 222 5-9 350 413 409 398 316 10-14 328 404 436 414 317 15-19 289 354 386 372 [316] 20-24 233 296 [335] [3861 25-29 156 12161 [283] Note: Duration-specific fertility rates are per 1,000 women. Estimates enclosed in brackets are truncated. Another indicator that may be used to measure fertility within marriage is total marital ferti l ity rate (TMFR). This measure is calculated in the same way as the total fertility rate except that the denominator for calculating age-specific marital fertility rate (ASMFR) is married women instead of all women in the specific-age range. Table 3.5 shows the age-specific and total marital fertility rates for 0-2 years, 0-4 years, 29 Table 3.5 Marital fertility Age-specific marital fertility rates (per 1,000 women) mad total marital fertility rates for different periods preceding the survey, Yemen 1991/92 Maternal age at birth Number of years preceding the survey 0-2 0-4 5-9 10-14 15-19 315 295 359 322 20-24 365 354 435 404 25-29 338 327 429 429 30-34 289 291 407 372 35-39 260 254 344 [351] 40-44 172 171 [237] 45-49 [1201 [12ill TMFR 15-44 8.7 8.5 TMFR 15-49 9.3 9.1 TMFR = Total marital fertility rate, expressed per woman. Note: Estimates enclosed in brackets are truncated. 5-9 years, and 10-14 years before the survey. The pattern of age-specific marital fertility rates for the three years preceding the survey is similar to that observed for age-specific fertility rates. The total marital fertility rate for the three years preceding the survey is 9.3. The ASMFR rates for the three years before the survey compared with rates for the period 10-144 years preceding the survey, have declined for women in their twenties and thirties, but have remained unchanged for women under 20. Thus, no reduction has taken place in teenage fertility, and there has been no decrease in these elevated-risk births. 3.3 CURRENTPREGNANCY Another indicator of current fertility is the proportion of women who are currently pregnant. This measure represents, in a sense, the most current level of fertility since it anticipates fertility over the next few months. However, as a measure of current fertility, it should be treated with caution because it is an underestimate. Some women in the early stages of pregnancy may be unaware or uncertain that they are pregnant; others may deliberately avoid mentioning current conception due to local customs or traditions. Overall, 18 percent of currently married women 5 (see Table 3.6) and 13 percent of all women reported being pregnant (not shown). By age, the highest pregnancy rate is reported for currently married women age 20-24 (23 percent), although one-fifth of women in the next older and next younger age groups reported pregnancies at the time of the survey. Urban-rural differentials by age show an interesting pattern; urban women under 25 reported much higher pregnancy rates than rural women, but a higher proportion of 4The ASMFRs for the three years preceding the survey are considered for the reason mentioned earlier for ASFRs. Comparison with rates 10-14 years before the survey is more appropriate since the rates for 5-9 years before the survey are probably "inflated" by age displacement of children under five. 5The 1991 Egypt Maternal and Child Health Survey (Abdel-Azeem et al., 1993) reported that only 11 percent of currently married women were pregnant at the time of the survey. 30 Table 3.6 Currently pregnant women Percentage of currently married women who reported a current pregnancy, by age and selected background characteristics, Yemen 1991/92 Age of woman Background characteristic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Residence Urban 27.6 26.2 16.0 17.2 12.8 6.9 6.4 16.5 Rural 19.7 21.5 21.2 20.4 19.5 15.6 8.3 18.8 Region Nor th./West. 22.7 22.4 21.2 20.3 19.4 15.1 8.6 19.2 South/East 13.1 22.7 15.2 17.5 13.2 9.1 5.2 14.7 Children ever born 0 24.2 1 17.9 2 (15.5) 3 * 4 16.6 5+ NA 30.8 15.3 (14.6) * * * 23.7 19.3 23.0 (20.6) * * * 18.8 23.6 27.3 12.8 (14.7) * * 21.6 23.4 18.9 19.9 20.0 * * 19.4 15.8 15.2 19.7 17.5 (16.3) (5.3) 16.0 (9.9) 20.9 20.5 18.5 14.9 8.8 17.1 Total 21.0 22.5 20.2 19.8 18.4 14.1 8.0 18.4 Note: Figures in parentheses are based on 25-49 cases; an asterisk indicates that figure is based on fewer than 25 cases and has been suppressed. NA = Not applicable rural women were currently pregnant in the age group 25 and over. Regional differentials were more marked; higher proportion of women were pregnant in the northern and western governorates than in the southern and eastern governorates (19 percent and 15 percent, respectively). For age group 20-24, there is practically no difference between the regions, but for all other age groups the proportions of pregnant women was higher in the northern and western governorates. There was no observable pattern in the pregnancy rate by number of children ever bom; high rates were observed even among women with 5 or more children (17 percent). 3.4 CHILDREN EVER BORN Information on the number of children ever born (CEB) is presented in Table 3.7 for all women and for currently married women. (As noted earlier, it is assumed that never-married women had no births.) These data, which reflect cumulative births over time, indicate that all women have had an average of 3.7 children, while currently married women had 4.9 children. About one-fourth of currently married women have had eight or more births, and one in 10 has had at least 10 births. The difference in CEB between all women and currently married women is mainly due to the marital status of women under 25; a large proportion of teenagers and a substantial proportion of women 20-24 have never-married. The mean number of children ever bem increases with age, reflecting the natural family building process. For example, among all women, the average number of live births for age group 25-29 is 3.3, for age group 30-34 it is 5.3, and for women 35-39 the average is 6.7 children. At the end of the reproductive years, at age 45-49, the mean number of children ever bern for all women is 8. The high fertility of Yemeni women is evident from the substantial proportion of currently married women in their forties who have large families; one-fourth of women 40-44 and one-third of those age 45-49 have given birth to 10 or more children. 31 The results in Table 3.7 show that early childbearing is uncommon but not rare; almost 11 percent of teenagers (15-19) have had a child, compared to 56 percent of women 20-24. Teenage childbearing is examined in section 3.7. The proportion of women over 40 who have never given birth can be taken as a measure of primary sterility. Less than 2 percent of currently married women 40-49 are childless and have never given birth. Compared to other countries in the region and the developing world as a whole, primary infertility is very low in Yemen. The last column in Table 3.7 shows the mean number of children still living. Differences in the mean number of children ever born and children surviving are small for currently married women under age 30 but are notable after age 30. Table 3.7 Children ever born and living Percent distribution of all women and of currently married women by number of children ever born (CEB) and mean number ever born and living, according to five-year age groups, Yemen 1991/92 Number of children ever born Number Mean no. Mean no. Age of of of living group 0 1 2 3 4 5 6 7 8 9 10+ "l'otal women CF .B children ALL WOMEN 15-19 89.1 6.7 3,1 0.6 0.4 . . . . . . . . . . 100.0 1729 0.19 0.15 20-24 44.0 19.1 13.7 12.5 6.9 2.6 0,8 0.2 0.2 . . . . 100.0 1135 1.33 1.16 25-29 16.1 8.3 13.1 15.7 15.5 12.8 8.6 5.8 2.6 1.2 0.3 100.0 1425 3.34 2.90 311-34 5.6 4,3 4.9 10.9 12.1 13.8 15 .5 14.3 7.6 6.1 5.0 100.0 1021 5.27 4.41 35-39 3.5 2.1 3.1 5.0 6.4 10.4 14.9 15.2 14.5 8.6 16.2 100.0 982 6.71 5.53 40-44 2.2 2.1 3.0 3.7 4.9 6.2 9.0 12 .5 15.7 14.6 26,1 100.0 654 7.66 6.17 45-49 0.7 2.1 2.7 2.5 6.5 6.5 10.4 11.6 12.2 12.3 32.5 100.0 529 8.05 6.11 Total 31.8 7.2 6.8 7.7 7.5 7.1 7.4 7.0 5.7 4.3 7.4 CURRENTLY MARRIED WOMEN 100.0 7475 3.72 3.07 15-19 55.9 28.1 11.7 2.5 1.8 . . . . . . . . 100.0 406 0.67 0.61 20 24 21.4 26.5 19.4 17.4 9.9 3.5 1.2 0.3 0.3 -- -- 100.0 788 1.87 1,63 25-29 7.6 8.2 13.6 17 .5 17.6 14.4 9.8 6.6 2.9 1.4 0.4 100.0 1229 3.75 3.26 30-34 2.5 3.8 4.8 11.0 12.2 14.7 16.0 15.2 7.9 6.6 5.4 100.0 933 5.53 4.63 35-39 2.4 1.4 2.9 4.6 6.2 10.7 15.0 15.6 15.0 9.3 16.9 1130.0 909 6.90 5.71 40 44 1.8 1.7 2.6 3.4 5.1 6,4 9.1 12.9 15.9 14.7 26.6 100.0 615 7.75 6.27 45-49 0.6 1.5 2.5 1.9 6.9 5.3 9.1 12.5 12.8 12,8 34.2 100.0 476 8.25 6.28 To~al 10.2 9.1 8.7 10.0 10.0 9.4 9.6 9.4 7.6 5.9 10.0 100.0 5355 4.94 4,09 -- Less than 0.05 percent 3.5 B IRTH INTERVALS A birth interval is the period of time between two successive live births. A short interval between births affects the health of mothers, and the survival chances of children. The risk of a child dying is high if the interval between the child's birth and a previous or next birth is less than 24 months. Table 3.8 examines the birth intervals of children born in the five years before the survey by number of month since previous birth. In Yemen, more than 4 in 10 births have birth intervals of less than 24 months and one-fourth of births occurred after an interval of less than 18 months. About one-third of birth intervals are between 24 32 Table 3.8 Birth intervals Percent distribution of births in the five years preceding the survey by number of months since previous birth, according to demographic and socioeconomic characteristics, Yemen 1991/92 Number of months since previous birth Characteristic 7-17 18-23 24-35 36-47 48+ Median number of months Number since of previous Total births birth Age of mother 15-19 44.1 14.8 34.1 5.7 1.3 100.0 96 20.3 20-29 25.3 21.5 30.8 13.2 9.2 100.0 2592 24.8 30-39 21.5 17.3 31.4 13.9 15.9 100.0 2781 28.2 40+ 20.6 14.4 28.4 16.4 20.1 100.0 1110 31.3 Birth order 2-3 25.3 20.2 30.4 12.0 12.1 100.0 1790 25.2 4-6 21.0 17.8 31.0 15.1 15.1 100.0 2471 28.0 7 + 23.9 17.8 30.6 14.2 13.6 100.0 2317 26.7 Sex of prior birth Male 22.8 18.6 30.8 13.9 13.9 100.0 3387 26.9 Female 23.6 18.3 30.5 14.0 13.6 100.0 3191 26.5 Survival of prior birth Living 21.8 18.3 31.5 14.5 13.9 100.0 5732 27.2 Dead 32.8 19.4 24.9 10.3 12.4 100.0 847 23.1 Residence Urban 21.8 19.7 30.7 12.5 15.3 100.0 1046 26.4 Rural 23.4 18.2 30.7 14.2 13.4 100.0 5533 26.7 Region Nor th./West. 23.6 18.5 31.1 14.0 12.9 100.0 5738 26.5 South/East 20.6 18.2 28.0 13.6 19.6 100.0 841 28.4 Education lUiterate 23.4 18.3 30.8 14.1 13.5 100.0 6115 26.6 l.iterate 12.7 14.4 38.3 11.2 23.4 100.0 53 32.7 Primary 21.0 19.5 29.6 14.5 15.5 100.0 256 27.4 More than primaa 3, 22.6 23.2 26.6 9.8 17.7 100.0 155 25.7 Total 23.2 18.4 30.7 14.0 13.7 100.0 6579 26.7 Note: First-order births are excluded. The interval for multiple births is the number of months since the preceding pregnancy that ended in a live birth. and 35 months, and about one-fourth are three years or more apart. Fifty percent of births occur within 27 months of a previous birth (the median birth interval). The shortest median birth intervals are reported for women 15-19 years, and for births that occurred following the death of a child (20 months and 23 months, respectively). The short birth interval following a child who died reflects shortening of the period of postpartum amenorrhea due to cessation of breastfeeding at the death of the child. With increasing age of the mother, birth intervals also increase; birth intervals are longest for women in their forties (31 months). The median birth interval is even higher for women who are literate but have not completed primary education (33 months). For other background characteristics presented in Table 3.8, the medians vary little and range from a low of 24 to a high of 28 months. 33 3.6 AGE AT FIRST BIRTH The age at which women start childbearing has important demographic and health consequences. For example, the magnitude of teenage pregnancy is a major concem of health professionals. Women are at greater risk of dying from the complications of pregnancy and delivery under the age of 20 and after age 34. Also, early childbearing usually results in women having larger families, which can have a negative effect on their socioeconomic status and participation in the labor force. Table 3.9 presents the percentage distribution of women by age at first birth according to their current age. The quality of data on age at first birth is affected by reporting errors, such as misreporting the woman's age, underreporting of first births, and misreporting the first child's date of birth. Such errors are usually more pronounced in reporting by older women and illiterate women. A large proportion of Yemeni women age 20-496 had their first birth before age 20, ranging from one third in age group 40-44 to more than one-half in age groups 25-29 and 30-34. Among women 25-49, the proportion of first births after age 24 increases with age--from 5 percent to 38 percent. These results indicate that a trend toward early childbearing continues in Yemen. Table 3.9 Age at first birth Percent dista'ibution of women 15-49 by age at first birth, according to current age, Yemen 1991/92 Current age Women Median with Age at first birth Number age at no of first births <15 15-17 18-19 20-21 22-24 25+ Tolal women birth 15-19 89.1 2.2 6.4 2.3 NA NA NA 100.0 1729 a 20-24 44.0 5.9 21.1 14.0 10.9 4.1 NA 100.0 1135 a 25 29 16.1 7.0 25.4 18.5 15.2 12.6 5.1 100.0 1425 19.9 30-34 5.6 7.0 28.3 18.9 14.7 16.5 9.0 100.0 1021 19.5 35-39 3.5 4.6 20~3 17.7 18.5 19.0 16.3 100.0 982 20.8 40-44 2.2 6.3 16.4 12.4 14.3 19.2 29.3 100.0 654 22.1 45-49 0.7 5.4 17.6 15.1 9.2 14.0 37.9 100.0 529 22.7 NA = Not applicable aMedians were not calculated for these cohorts because less than 50 percent of the women in the age group x to x+4 have had a birth by age x. Table 3.9 shows the median age at first birth for the different age cohorts. The medians are not defined for cohorts in which at least 50 percent of the women did not have a birth. The trend across age cohorts suggests a steady decrease in the median age at first birth from the oldest to the youngest cohorts. Indeed, the median age at first birth is 22.7 years for women age 45-49, 20.8 for age 35-39, and continues to decline to a median age of 19.5 for those 30-34 before rising slightly to 19.9 years for younger women. Although it is possible that childbearing is starting at younger ages than in the past, it should be noted, as mentioned earlier, that the data may suffer because of incorrect birthdates both for the mother and for her first birth. The problem of omission of births or misdating particularly affects data on older women who may have memory lap~s or find it difficult to remember the dates of their first births. These women have a tendency to push birth dates closer to the date of survey, thereby making themselves older at first birth than they actually were. 6Only women 20-49 are considered because women under 20 can still have a birth before reaching age 20. 34 Table 3.10 presents the median age at first birth by current age and ~lected background characteristics. Overall, the median age at first birth is 20.4 years for women aged 25-49. For most variables, the median age at first birth differs by around one year: the median age is higher for women who live in rural areas than urban areas (misreporting of data may be a factor), higher in the southem and eastern govemorates than in northem and western govemorates, and higher for women with no schooling compared with those who have only primary education. The differences in the median age at first birth are particularly striking between women with more than primary education and those with no education or those with primary education; the median age at first birth for the most educated women is more than four to five years higher. Table 3.10 Median age at first birth Median age at first birth among women age 25-49 years, by current age and selected background characteristics, Yemen 1991/92 Current age Women Background age characteristic 25-29 30-34 35-39 40-44 45-49 25-49 Residence Urban 20.6 18.9 19.8 20,0 19.9 19.8 Rural 19.8 19.7 21.0 22.5 23.7 20.6 Region Nor th./West. 19.4 19.3 20.7 22.3 23.4 20.3 South/East 22.1 20.1 21.2 20.8 20.9 21.2 Education No schooling 19.5 19.4 20.8 22.2 22.7 20.3 Primary 19.8 18.7 18.5 17.8 25.6 19.1 More than primary a 24.6 23.1 21.9 23.2 24.7 Total 19.9 19.5 20.8 22.1 22.7 20.4 Note: The medians for cohort 15-19 and cohort 20-24 could not be determined because half the women in each cohort have not yet bad a birth. aThe median was not calculated for this cohort because less than 50 percent of women in the age group x to x+4 have had a birth by age x. 3.7 TEENAGE FERT IL ITY Concerns about the health of teenagers and the mortality risks of births to teenagers (women 15-19 years) were mentioned earlier in this report. It was also noted that 11 percent of women aged 15-19 have given birth (see Table 3.7). Table 3.11 examines in more detail the level of pregnancies and births among teenagers in Yemen. The sum of the percentage of ever-married women 15-19 years who are mothers and the percentage who are pregnant with their first child represents the proportion of teenage women who have begun childbearing. One in seven (14 percent) 7 are mothers or are currently pregnant with their first child. Only 1.2 percent of women age 15 have become mothers and the same proportion are pregnant for the first time. The percentage of women who have become mothers is 4.7 percent for women age 16; from then on the proportion almost doubles with each single-year increase in age to over 35 percent among women age 19. The proportion of teenagers who have started childbearing is substantially higher in rural areas than in urban areas (15 percent and 11 percent, respectively). Childbearing among teenagers is twice as common in the 7The proportion of teenagers who have had one or more births or are pregnant is much higher in Yemen than in Jordan. According to the Jordan DHS survey, 5.3 percent of teenagers had become mothers and another 2.1 percent were pregnant for the first time. 35 Table 3.11 Teenage pregnancy and motherhood Percentage of teenagers 15-19 who are mothers or pregnant with their first child, by selected background characteristics, Yemen 1991/92 Percentage who are: Percentage who have Pregn~mt begun Number Background with first child of chm'acteristic Mothers child bearing teenagers Age 15 1.2 1.3 2.6 506 16 4.7 1.2 5,9 348 17 7.9 5.6 13.5 276 18 16.9 5.3 22.2 370 19 35.5 4.4 39.9 228 Residence Urban 8.0 2.8 10.8 446 Rural 11.6 3.3 14.9 1316 Region North./West. 12.5 3.5 15.9 1353 South/East 5.0 2.3 7.3 384 Education No schooling 13.3 3.7 17.0 1077 Primary 7.2 2.3 9.5 420 More than primary 5.5 2.5 8.11 256 Total 111.9 3.2 14.1 1729 northem and western governorates (16 percent) as in the southern and eastern govemorates (7 percent). The proportion of teenage women without schooling who had begun childbearing is more than double that of teenage women who have more than primary education (17 percent and 8 percent, respectively). Table 3.12 shows that among teenage~ under 18 years of age who had given birth, only a few had more than one child. Although only a small proportion of women age 18 have had two children (7 percent), almost one-fifth of 19-year-olds have had two or more children. The likelihood that a teenage mother will have had more than one birth reaches a level of over 50 percent among nineteen-ycar-olds. Table 3.12 Children born to teenagers Percent distribution of teenagers 15-19 by number of children ever Ix~rn (CEB), Yemen 1991/92 Age (1 1 2+ Number of Mean children ever born number Number of of Total CEB teenagers 15 98.8 1.2 0.0 100.0 0.0 5116 16 95.3 4.3 0.4 100.0 0.1 348 17 92.1 7.4 0.5 100.0 O. 1 276 18 83.1 10.1 6.8 100,0 0.3 370 19 64.5 16.4 19.1 100.(1 O. 6 228 Total 89.1 6.7 4.1 10f).0 0.2 1729 36 CHAPTER 4 FAMILY PLANNING Information on contraceptive use is of particular interest to policymakers, program managers, and researchers in the areas of population and family planning. Family planning is a powerful tool for improving the health of mothers and children as well as for reducing the levels of infant and maternal mortality. Providing easy access to reliable and safe methods of family planning increases women's control over their own fertility. This chapter describes women' s knowledge of family planning methods and sources where they can be obtained, and use of contraception in Yemen. Differentials in knowledge and use are also discussed. In addition, problems with current methods, reasons for first use and current use, and reasons for not intending to use in future are included in this chapter. 4.1 KNOWLEDGE OF FAMILY PLANNING METHODS AND SOURCES Familiarity with contraceptive methods and sources for methods are among the prerequisites for the adoption of fertility regulation. Knowledge o f methods is a necessary but not always sufficient condition for use. The Yemen Demographic and Maternal Child Health Survey (YDMCHS) provides information on the level of knowledge of family planning methods and providers of family planning services. Data on knowledge of contraceptive methods were collected in the Woman's Questionnaire. Each respondent was asked to name the methods or ways a couple could use to avoid or postpone pregnancy, The interviewer then read the names of specific methods (without any description), omitting those that the respondent had already mentioned, and asked whether the respondent had ever heard of the method. The questionnaire included seven specific modem methods: the pill, IUD, injection, vaginal methods (diaphragm/jelly/foam), condom, female sterilization, and male sterilization. Three traditional methods, safe period (rhythm/periodic absti- nence), withdrawal, and prolonged breastfeeding, t were included. Any other methods mentioned by re- spondents, such as herbs were also recorded. To determine knowledge of sources of modem contraceptive methods except sterilizations, for each method the respondent recognized, she was asked if she knew any source for obtaining the method. Table 4.1 indicates that 60 percent of currently married women 2 have heard of at least one family planning method. The most widely known method is the pill, which is known to more than half of currently married women; almost all women who know of a modem method (53 percent) have heard of the pill) About one-third of the women know about the IUD and injection, and one-fourth have heard of female sterilization. Knowledge of other modem methods is much lower: male sterilization (13 percent), condom (10 percent), and vaginal methods (7 percent). Four in I0 women know of a traditional method, mainly breastfeeding (38 percent), while safe period and withdrawal were known by much smaller proportions of respondents. aYemen, Jordan, and Egypt are the only DHS countries that included prolonged breastfecding in the list of family planning methods on which respondents were prompted. ~l'he results for ever-married and currently married women are almost the same. 3Contraceptive knowledge in Mauritania is ahnost the same as in Yemen, according to a 1990/91 PAPCHILD survey. In Mauritania, 61 percent of currently married women know a family planning method, and 48 percent know a modem method. 37 Table 4,1 Knowledge of contraceptive methods and source for methods Percentage of ever-married women and currently married women who know a specific contraceptive method and who know a source for services, by specific methods, Yemen 1991/92 Know method Know a source Ever- Currently Ever- Currently Contraceptive married married married married method women women women women Any method 60.3 60.2 27.2 27.0 Modern method 53.5 53.2 27.2 27.0 Pill 51.7 51.3 24.9 24.5 IUD 34.2 33.5 15.7 15.3 Injection 32.1 31.9 12.8 12.8 Diaphragm/foam/jelly 7.2 7.0 3.5 3.3 Condom 10.4 10.3 5.2 5.0 Female sterilization 24.1 24.0 NA NA Male sterilization 13.6 13.4 NA NA Any traditional method 39.7 39.7 NA NA Safe period 13.2 13.1 NA NA Withdrawal 8.6 8.5 NA NA Prolonged breastfeeding 37.7 37.7 NA NA Other traditional methods 1.1 1.1 NA NA Number of women 5687 5355 5687 5355 NA = Not applicable The results presented in Table 4.1 show that only 27 percent of married women know where to go to get a modem method of contraception (the YDMCHS did not include a question on sources for female and male sterilization). Knowledge of sources for specific methods shows that one-fourth of married women know a source for the pill. Only 15 and 13 percent, respectively, know a source for the IUD and injection; sources for other methods are even less well known. The percentage of married women who know at least one modem contraceptive method and know a source for the method, by various background characteristics, is presented in Table 4.2. Only small differences by age exist in knowledge of modem methods among women 15-44:54-58 percent of women 20- 44, and almost half of the women 15-19 arc aware of a modem method of family planning. Older women, 45-49, are the least knowledgeable (42 percent). There is a greater variability in the level of contraceptive knowledge by residence and region. About 9 in 10 urban women are aware of at least one modem method, compared with less than half of rural women. Knowledge of modem methods is much higher in the southern and eastern govcmoratcs (over 70 percent), while only half of women in the nonhero and western govemorates know a modem method. Differentials in knowledge by education are even more striking. Among illiterate women, only half know a modem method, compared with 86 percent of women who have completed primary education, and an overwhelming majority of women with higher education (94 percent). The differences in knowledge of family planning by 38 education may account, in large part, for the greater knowledge of family planning in urban areas and in the southern and eastem govemorates, since a greater proportion of educated women live in those areas. Differentials in knowledge of service providers show relatively more variability than differentials in knowledge of modem methods (see Table 4.2). By age groups, between 24 and 31 percent of married women age 15-44 know a source for a modem contraceptive method. The differences are much greater by residence. Sixty-seven percent of urban women know a source compared to only 18 percent of rural women. Regionally, it is twice as likely for a woman in the southern and eastern governorates (46 percent) to know a contraceptive source as a woman who lives in the northern and westem govemorates. The most striking difference in knowledge of service providers is by educational level. Only 22 percent of illiterate women know a source. Among women who are literate or have completed primary school, knowledge of service providers is almost three times that of illiterate women, or approximately 40 percentage points higher, while knowledge of a source is 60 percentage points higher among women who have completed more than primary school. Table 4.2 Knowledge of modern contraceptive methods and source for methods r by selected background characteristics Percentage of currently married women who know any contraceptive method and a modem method and who know a source for services, by selected background characteristics, Yemen 1991/92 Know a Know Know source for Number Background any a modem modem of characteristic method method I method 2 women Age 15-19 55.4 50.7 23.7 406 20-24 63.7 57.7 27.9 788 25-29 62.1 54.3 27.6 1229 30-34 60.6 54.1 31.4 933 35-39 60.9 53.5 27.4 909 40-44 60.7 53.5 26.0 615 45-49 50.3 42.1 18.4 476 Residence Urban 88.8 86.2 67.3 951 Rural 54.0 46.0 18.3 4404 Region North./West. 57.3 49.3 23.2 4458 South./East. 74.3 72.4 45.8 897 Education Illiterate 56.6 48.9 22.0 4792 Literate 91.8 91.8 60.5 57 Primary 87.8 86.0 6¢/.8 291 More than primary 95.0 93.6 84.3 216 Total 60.2 53.2 27.(} 5355 tlncludes pill, IUD, injection, vaginal methods (diaphragm/foam/jelly), condom, female sterilization, and male sterilization. 2Source for female and male sterilization not asked. 39 4.2 EVER USE OF CONTRACEPT ION In the YDMCHS, all respondents were asked if they had ever used the methods they reported knowledge of. As shown in Table 4.3, almost one-fifth of ever-married women have used a method to regulate their fertility at some point in their lives. Thirteen percent have used a modem method and l0 percent a traditional method (including prolonged breastfeeding); thus, the majority of ever-users have had experience with a modem method, and some have used more than one method. Currently married women have about the same level of ever-use experience as ever-married women. The age differentials for ever-use among currently married women show that the rate for women 15- 19 is only 8 percent, it doubles for women 20-24, and increases to 20 percent among women 25-29; ever-use peaks at around 25 percent among women in their thirties before declining among older women (19 percent and 15 percent for women 40-44 and 45-49, respectively). Two methods, the pili (11 percent) and prolonged breastfeeding (8 percent), are used most commonly. At the most, 3 percent of currently married women have ever used any other method-- IUD (3 percent), injection and safe perio(~l (2 percent each), and female sterilization (1 percent). Table 4.3 Ever use o f contraception Among ever-marrled women and currently married women, the percentage who have ever used a contraceptive method, by specif ic method and age, Yemen 1991/92 Age Modern method Traditional method Any Dia- Female Male Pro- modem phragm/ steri stcri- Any With- longed Number Any meth- Injec- foam/ Con- liza- tlza- trad. Safe draw~ breast of method od Pill IUD tion jelly dora tion Lion method period ai feeding Other women EVER-MARRIED WOMEN 15-19 g.0 3,0 2.6 0.1 0.3 -- 0.1 -- 6.0 0.8 0.3 5.7 -- 427 20-24 16.1 9.1 7,2 1.8 I . I -- 0.6 -- -- 9.3 1.9 0.9 7.2 0.2 815 25-29 19.1 13.5 9.8 3.2 2.6 0.6 1.2 0.2 -- 9,2 2.1 2.0 7.1 0.2 1295 30-34 25.3 17.8 14.8 3.2 1.7 0.6 1.4 0.7 0.1 12,2 3.0 3.0 9.1 0.1 995 35 39 24.2 17.3 13.9 3.4 1.6 0.4 0.9 1.6 0.3 12.9 2.3 1.7 10.5 0.3 972 40-44 19.3 15.0 11,3 1,6 2.8 0.4 1.7 2.1 0.6 9.7 1,6 1.6 8.0 0.2 653 45-49 15.1 11.1 7.8 1.2 1.5 0,0 0,8 1.4 0.3 6.3 1,0 2.0 5.3 0.5 529 Total 19.5 13.4 10.4 2.4 1.8 0.4 1.0 0.8 0.2 9.9 2.0 1.8 7.9 0.2 5687 CURRENTLY MARRIED WOMEN 15-19 7.6 2.7 2.2 0.2 0. I 6.0 0.5 0.4 5.7 -- 406 20-24 16.2 8.9 7.0 1.9 1.1 - 0.6 9.6 2.0 1.0 7.5 0.2 788 25-29 19.6 13.9 10.1 3.4 2.7 0.6 1.3 0.2 - 9.4 2.1 2.0 7.3 0.2 1229 30-34 25.9 18.4 15.3 3.4 1.7 0.4 1.4 0.8 0.2 12.4 3.1 2.8 9.3 0.1 933 35-39 25.1 17,9 14.3 3.4 1.7 0.d 0.9 1.7 0,3 13.4 2.5 1.8 10.9 0.3 909 40-44 19.2 15.0 11.1 1.6 2.9 0.3 1.8 2.1 0.6 9.3 1.7 1.4 7.8 0.2 615 45-49 15.2 11.5 7.9 l.d 1.7 -- 0.9 1.6 0.3 6.1 1.0 2.0 5.2 0.3 476 Total 19.8 13.7 10.6 2.5 1,9 0.3 I . l 0.8 0,2 10.1 2,1 1.8 8.0 0.2 5355 -- Less than 0.05 percent 40 4.3 NUMBER OF CHILDREN AT FIRST USE Table 4.4 shows the percent distribution of ever-married women by the number of living children at the time they first used contraception. It is interesting to note that 5 percent of women, or one-fourth of ever- users, started using contraception before having their second child. This tendency to start contraception early, probably for the purpose of spacing births, is most noticeable among younger women, i.e., women under 35 years of age. For example, more than three-quarters of women 20-24 who had experience with contraception had initially started using a method before they had three children. Table 4,4 Number of children at first use of contraception Percent distribution of ever-man-ied women by number of living children at the time of first use of contraception, according to current age, Yemen 1991/92 Number of living children at time Never of first use of contraception Number Current used of age contraception 0 1 2 3 4+ Missing Total women 15-19 92.0 1.4 5.5 0,7 0,3 0.0 0.1 100,0 427 20-24 83.9 0.9 8.0 3.6 2.2 0.8 0.5 100,0 815 25-29 80.9 0.2 5.9 3.3 3.8 5,8 0.1 100.0 1295 30-34 74.7 0.7 4.6 4.1 3.7 11.6 0.6 100.0 995 35-39 75.8 0.3 2.9 3.3 2.9 14.2 0.7 100.0 972 40-44 80.7 0.3 2.5 1.1 2.1 13.3 0.0 100.0 653 45 49 84.9 0.2 2.0 0.9 1.2 10.7 0.1 100.0 529 Total 80.5 0.5 4.7 2.8 2.7 8.4 0.3 100.0 5687 4.4 REASON FOR FIRST USE Women were asked about their reproductive intentions when they first began using family planning. Table 4.5 shows among ever-married women who ever used contraception (20 percent), more than half (11 percent) had first used family planning for spacing births because they wanted more children later. Seven percent of ever-users wanted to stop childbearing at the time they began using a method. The majority of women 40 and over started using contraception in order to limit family size. Proportionally, ever-users who adopted a family planning method for the first time for purposes of spacing births rather than for limiting family size were higher among urban women than rural women, and higher among more educated women than among those with primary education or those who are illiterate. 41 Table 4.5 Reproductive intentions at first use of contraception Percent distribution of ever-marrled women by reproductive intentions at the time of first use of contraception by selected background characteristics, Yemen 1991/92 Never used Wanted Wanted Number Background comra- child no more Other/ of characteristic ception later children Missing Total women Age 15-19 92.0 5.0 2.7 0.4 100.0 427 20-24 83.9 11.0 4.1 1.0 100.0 815 25-29 80.9 12.9 5.5 0.7 100.0 1295 30-34 74.7 15.0 8.9 1.4 100.0 995 35-39 75.8 12.1 10.6 1.5 100.0 972 40-44 80.7 8.9 10.4 0.1 100.0 653 45-49 84.9 6.6 8.2 0.3 100.0 529 Residence Urban 50.8 30.4 17.1 1.8 100.0 1054 Rural 87.3 6.9 5.2 0.7 100.0 4633 Region Nor th.AVest. 83.3 8.8 7.1 0.8 100.0 4708 South./l~ast. 67.2 22.7 8.9 1.2 100.0 979 Education Illiterate 84.2 8.4 6.6 0.8 100.0 5075 Literate 47.2 36.1 16.7 0.0 100.0 61 Primary 57.2 27.1 13.5 2.2 100.0 316 More than primary 41.3 44.5 13.1 1.0 100.0 235 Total 80.5 11.2 7.4 (1.9 100.0 5687 4.5 CURRENT USE OF CONTRACEPTION Only nonpregnant currently married women were asked if they were using any contraceptive method at the time of the survey. Including prolonged breastfeeding as a method, the contraceptive prevalence rate in Yemen for allcurrently married women (pregnant and nonpregnant) is 10 percent. It is 7 percent if breast- feeding is not included (see Table 4.6). Although the contraceptive prevalence in Yemen is low compared with many other Arab countries,4 it is much higher than the one percent prevalence rate reported in 1979 by the Yemen Arab Republic, which, in this report is referred to as the northern and western govemorates and has a prevalence rate of 8 percent. The pill, the most popular method, is used by 3 percent of women and accounts for more than half of the modem method use in Yemen. The other modem methods-- IUD and female sterilization--are each used by one percent of women. Prolonged breastfeeding is the most frequently used traditional method and the second most popular method overall (2 percent). Around one-half of one per- cent of married women rely on the safe period or withdrawal. Table 4.6 also shows the percent distribution of currently married women by contraceptive method currently used, according to age. The pattem of differ- entials in current use by age is similar to the pattern for ever-use of contraception, i.e., women in their thirties have the highest levels of use. An extremely large difference in prevalence is observed between urban and rural areas. 5 At the time of survey, 28 percent of urban women were using a contraceptive method compared 4For example, in countries in North Africa and the Middle East where a PAPCH1LD survey and the DHS surveys have been conducted, the contraceptive prevalence rates range from a low of 4 percent in a PAPCH1LD survey in Mauritania (1990/91 ) and 9 percent in Sudan (1989/90), to a high of 50 percent in Tunisia (1988). The prevalence rates are 35 percent in Jordan (1990), 42 percent in Morocco (1992), and 47 percent in Egypt (1992). 5Contraceptive rates are higher in urban areas than in rural areas in 'all of the North African and Middle Eastern countries in which DHS surveys have been conducted. In terms of absolute percent differences between urban and rural women, the difference is greatest in Tunisia (1988) where the rates are 60 and 35 percent in urban and rural areas, respectively. However, relatively speaking, the ratio (over 4 times) of urban to rural use of contraception is highest in Yemen and Sudan (1989/90), and lowest (1.5 times) in Egypt (1992). 42 Tab le 4 .6 Cur rent use o f cont racept ion by age Percent d i s t r ibut ion o f cur rent ly marr ied women by cont racept ive method cur rent ly used, accord ing to age, Yemen 1991/92 Modern method Traditional method Background characteristic pi-o~ All Female Male All longed Not modem steri- steri- tred. With breast- cur- Number Any meth- lnjec- Con- liza- liza- meth- Safe draw- feed- rently of method otis I Pill IUD tion dorn tion tion ods 2 period al ing using Total women 15-19 5.1 1.3 1.1 0.2 3.8 -- 0.1 3.7 94.9 1(30.0 406 20-24 8.5 4.3 2.8 1.0 0.4 4.2 0.2 0.5 3.5 91.5 1(30.0 788 25-29 10.5 7.2 3.8 2.3 0.9 0.1 0.2 -- 3.3 0.6 0,6 2.0 89.5 100.0 1229 30-34 11.7 6.8 3.7 1.0 0.7 0.3 0.8 0.1 4.9 0.5 0.6 3.8 88.3 100.0 933 35-39 12.5 8.3 4.4 1.3 0.6 -- 1.7 0.3 4.2 0.6 0.6 2.7 87.5 100.0 909 40-44 8.8 6.4 2.4 0.9 0.7 -* 2.0 0.4 2.4 0.6 0.6 1.1 91.2 100.0 615 45-49 5.1 4.1 1.5 0.6 0.3 0.1 1.6 0.1 1.0 0.3 0.6 0.2 94.9 IGO.0 476 Total 9.7 6.1 3.2 1.2 0.6 0.1 0.8 0.1 3.6 0.5 0.6 2.5 90.3 100.0 5355 -- Less than 0.05 percent llncindes users of vaginal methods 2Includes users of c4her traditional methods with only 6 percent of rural women (see Figure 4.1 and Table 4.7). Rural women are also relatively more likely than urban women to rely on prolonged breastfeeding. Figure 4.1 highlights the differentials in contraceptive prevalence by region, educational level and current family size. The prevalence rate in the southern and eastern governorates is twice as high in the nonhem and western govemorates, but the ratios of preference, of modem to traditional methods, are almost Figure 4.1 Contraceptive Prevalence by Selected Background Characteristics RESIDENCE Urban Rural REGION North./West. South./East. EDUCATION Illiterate Literate m 6 / 8 / 7 17 28 29 NO. OF CHll 0 10 20 30 40 50 Percent * Includes prolonged breastfeeding, YDM CHS 1991/92 43 Table 4.7 Current use of contraception by background characteristics Percent distribution of currently married women by contraceptive method currently used, according to selected background characteristics, Yemen 1991/92 All modem Background Any meth- characteristic method ods I Pill Modern method "rr aditinnal method Pro- Female Male All longed Not steri- steri- trad. With breast- cur- Number injec- Con- l lza- liza- raeth- Safe draw- feed+ rently of IUD tlon dora tion tion ods 2 period al ing using Total women Residence Urban Rural Region North.AVest. South-East Education Illiterate Literate Primary Mote than primary No. of living children 0 1 2 3 4+ Total 28.2 18.8 9.8 4.6 0.5 0.4 2.7 0.6 9.5 1.7 2.5 5.2 71.8 100.0 951 5.7 3.3 1.8 0.5 0.6 -- 0.4 2.3 0.2 0.1 1.9 94.3 100.0 4404 8.3 5,1 2.2 1.3 0.7 -- 0.8 -- 3.2 0.3 0.3 2.4 91.7 100.0 4458 16.7 l l . l 8.2 1.1 0.1 0.2 0.8 0.7 5.6 1.1 1.6 2.9 83.3 100.0 897 7.2 4.3 2.0 0.8 0.7 -- 0.8 0.1 2.9 0.3 0.4 2.2 92.8 100.0 4792 29.1 16.7 11.6 3.3 0.7 1.1 12.4 3.6 1.1 7.7 70.9 100.0 57 24.3 16.8 11.2 4.2 -- 0.6 0.6 0.3 7.5 0.8 1.1 5.5 75.7 100.0 291 38.5 27.7 16.8 7.3 0.2 1.0 1.4 0.9 10.7 2.7 3.8 4.2 61.5 100.0 216 0.7 0.7 0.5 0.1 0.1 . . . . . . . . 99.3 100,0 611 7.7 4.2 3.6 0.5 -- 0,1 3.6 0.2 03 3,1 92.3 100.0 560 9.2 4.8 3.0 1.1 0.2 0. I 0.3 0.1 4.4 0.5 0.8 3.1 90.8 l~) .0 568 9.8 5.7 3.2 1.1 0.8 0.2 0.4 0.2 4.1 0.4 0.7 2.9 90.2 100.0 672 11.9 7.9 3.7 1.7 0.9 0.1 1.3 0.2 4.1 0.6 0.6 2.7 88.1 100,0 2945 9.7 6.1 3.2 1.2 0.6 0.1 0.8 0.1 3.6 0.5 0.6 2.5 90.3 100.0 5355 -- Less than 0.05 percent I Includes users of vaginal methods 2Includes users of othez traditional methods the same in both regions (see Table 4.7). Pronounced differences in current use exist by level of education. The proportion of married women using contraception increases dramatically from 7 percent among illiterate women to 24 percent among those with primary school completed, and then jumps to 39 percent for women with postprimary education. Contraceptive use varies slightly with the number of living children. Current use is negligible among women who have yet to start childbearing and is 8 percent among women with a single child. The prevalence rate increases to around 10 percent among those with 2-3 children, and is slightly higher among women who have four or more children. 4.6 REASON FOR CURRENT USE Table 4.8 indicates that the major reason for using of family planning in Yemen is to limit family size (51 percent) rather than to space births (32 percent). A small proportion (18 percent) of currently married women are using contraception for "other" reasons. Among rural women, less than half are using to limit the number of children, and only one-third practice family planning to space births. On the other hand, those with more than primary education are more likely to be using contraception for spacing than for limiting purposes. The table also shows that for various background characteristics, the proportion of women using contraceptive method for limiting purposes varies only slightly, from 51 to 54 percent. 44 Table 4.8 Reasons for current use of contraception Percent distribution of currently married women who are using contraception by reason for use, according to selected background characteristics, Yemen 1991/92 Want Want Number Background child no more Other of characteristic later children reason Total users Residence Urban Rural Region Nor th./West. South./East. Education 29.6 53.6 16.8 10O.0 268 34.0 47.3 18.6 100,0 249 31,8 50.1 18.1 100.0 368 31.6 51.8 16.6 100.0 150 Illiterate 27,7 53.3 19.0 100.0 347 Literate * * * 1130.0 17 Primary 37.1 51.0 I1.8 100.0 71 More than primary 44,3 38.3 17.5 100.0 83 Total 31.7 50.6 17.7 100.0 518 Note: An asterisk (*) indicates that the figure is based on fewer than 25 cases and has been suppressed. 4.7 PROBLEMS WITH THE CURRENT METHOD OF CONTRACEPT ION Women who were us ing any contracept ive method except ster i l izat ion were asked i f they had exper ienced any prob lems with their current method. Few contracept ive users reported any problems. None o f the women who rel ied on w i thdrawal and only a smal l proport ion o f those us ing the safe per iod (10 percent) or pro longed breast feeding (8 percent) reported any prob lems (see Tab le 4.9). Ha l f o f IUD and in ject ion users, and over 60 percent o f pi l l users said they exper ienced no prob lems with their method. When Table 4.9 Problems with current method of contraception Percent distribution of currently married women who are using contraception by the main problem with current method, Yemen 1991/92 lnjec- Safe With- Breast- Main problem Pill IUD tion I period I drawal I feeding No problem 62.5 51.1 51.4 89.7 100.0 92.5 Health concerns 26.4 32.3 19.1 4.5 0.0 5.6 Method failed 0.5 0.0 0.0 1.6 0.0 0.5 Access/availability 0.2 0.0 0.0 0.0 0.0 0.0 Cost 0.0 0.0 3.9 0.0 0.0 0.0 Inconvenient to use 0,5 3,3 8.0 4.1 0,0 0.0 Other 1.5 1.9 2.0 0.0 0.0 0.0 Missing 8.3 11.4 15.7 0.0 0.0 1.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Number 171 66 32 25 30 136 Note: Excludes users of vaginal methods, condoms and other traditional methods (fewer than 25 cases). No questions were asked about current use of female/male sterilization. Figures are based on fewer than 50 cases. 45 problems were mentioned, health concerns were cited most frequently. One-third of IUD users, one-fourth of pill users, one-fifth of injection users, and a small proportion of women using other methods reported experiencing health problem. It should be pointed out that information for 8-16 percent of users of modem methods on the question of problems with contraception was not available. 4.8 SOURCE OF CURRENT CONTRACEPT IVE METHOD The identification of sources of contraceptive methods lor current users is important for evaluating the role played by various providers of family planning services and supplies in the public and private sectors. Women using modem methods at the time of survey wcre asked where they obtained the method last time. Their responses are summarized in Table 4.10. Table 4.10 Source of supply for modern contraceptive methods Percent distribution of currently married women who are using modem contraceptive methods by most recent source of supply, according to specific methods, Yemen 1991/92 Female All Injec- sterili- mc, dern Source of supply Pill IUD Lion I zation I methods 2 Public hospital 19.4 27.1 54.5 77.1 33.2 Public family planning clinic 18.4 18.6 3.6 1.4 14.2 Private family planning clinic 0.5 4.4 3.9 4.3 2.2 MCI] center 12.4 11.2 6.0 0.0 9.4 Private medical doctor/clinic 2.2 31.8 1.5 8.0 8.8 Pharmacy 37.5 1.4 26.6 0.0 24.2 Other 3.5 3.7 3.9 0.9 3.1 Don't know/missing 6.2 1.7 0.0 8.3 4.9 Total 100.0 100.0 100.0 100.0 100.0 Number 171 66 32 44 325 IFigures are based on fewer than 50 cases. 2Includes vaginal methods, condom, and male stcrili/ation. The public sector, including nongovernmental organizations (NGOs), is a major provider of family planning in Yemen. It provides methods to 6 out of every 10 users of modem methods (57 percent in government sources, and 2 percent in private family planning clinics), while the private sector serves almost 3 in 10 users. In the private sector, pharmacies are the major source of contraceptive methods; they serve one-fourth of all modem method users (sec Figure 4.2). As for sources of speci tic methods, Table 4.10 shows that the main source for pill users is the public sector: government hospitals (19 percent), public family planning clinics (18 percent) and Matemal and Child Health centers (12 percent). Pharmacies supply a substantial proportion of pills users (38 percent). For IUD users, private doctors and clinics are the largest single source. One-third of IUDs were obtained from these private sources; however, public hospitals (27 percent) and family planning clinics (19 percent) together with matemal and child health centers (11 percent) are the largest source of this method. More than half of injection users go to public hospitals and one-fourth to private doctors. Among women who had chosen voluntary female sterilization as their family planning method, three-fourths had their operations done in government hospitals; only a small proportion went to a private doctors and family planning clinic. 46 Figure 4.2 Source of Family Planning Supply Current Users of Modern Methods Other/Missing 8% Pharmacy 24% ate doctor/Clinic 9% NGO FP clinic 2% MCH center 9% ub]ic FP clinic 14% Public hospital 33% YDMCHS 1991/92 4.9 TIME TO SOURCE OF CONTRACEPTIVE METHOD Women who were using a modem contraceptive method were asked about the time it takes to go to their method source. The answers, summarized in Table 4.11, show time to source of contraceptive method by urban- rural residence. One-fifth of current users of modem methods can reach their source in less than 15 minutes, one-fifth reach it in 15-29 minutes, and an additional one-fifth take between one-half and one hour to arrive at the source they last used. A substantial proportion of users, almost one-fifth have to spend more than two hours in travel to go to the place where they obtain the family planning method they use. As expected, urban women have easier access to family planning than those living in rural areas (see Figure 4.3). The median travel time for urban users is just 16 minutes, whereas the median for rural users is over one hour; 40 percent of rural users spend more than 2 hours traveling to their source of contraceptive methods. Table 4.11 Time to source of supply for modern contraceptive methods Percent distribution of currently married women who are using a modem contraceptive method by time to reach a source of supply, according to urban-rural residence, Yemen 1991/92 Minutes to source Urban Rural Total 0-14 26.8 7.2 18.0 15-29 26.2 10.8 19.2 30-59 26.1 18.2 22.6 60-119 10.5 12.6 11.5 120-239 1.4 33.1 15.7 4 hours or more 0.6 6.8 3.4 Not stated 8.3 11.1 9.5 Total 100.0 100.0 100.0 Median time to source 16.0 60.9 30.4 Number of women 178 147 325 47 Figure 4.3 Time to Source of Modern Methods Missing 8% < 15 minutes 27% 2 hours + 2% .---V--~.,~ 60-119 minutes 11% 30-59 minutes 26% 15-29 minutes 26% Missing 11% 2 hours + 40% <15 minutes 7% ~ 15-29 minutes 11% 30-59 ~ minutes 18% 60-119 minutes 13% Urban Rural YDMCHS 1991/92 4.10 INTENTION TO USE CONTRACEPT ION IN THE FUTURE Women who were not using any contraceptive method at the time of the survey were asked if they would use a family planning method in the future. Those who responded in the affirmative were also asked how long they would wait to use a method and what method they would prefer to use. Table 4.12 shows the distribution of currently married women who were not using contraception, by their intention to use in the future. The results in the table are presented according to the number of living children (including a current pregnancy) and past experience with contraception. Eighty-three percent of currently married women who were not using any method indicated that they did not intend to use a contraceptive method in the future. Only 16 percent expressed a desire to use some method--6 percent in the next 12 months and 2 percent after 12 months; 8 percent were unsure about when they would use a method. Surprisingly, of the women who had used a method in the past, more that half (6 percent) reported that they did not intend to use any method in the future. 48 Table 4.12 Future use of contraception Percent distribution of currently married women who are not using a contraceptive method by past experience with contraception and intention to use in the future, according to number of living children, Yemen 1991/92 Past experience with contraception and future intentions Number of living children I 0 1 2 3 4+ Total Never used contraception Intend to use in next 12 months 0,2 3.2 2.8 3.5 4.1 3.4 Intend to use later 1,2 3.1 1.6 0.7 1.5 1,5 Unsure as to timing 9.9 7.3 5.1 4.8 6.1 6.3 Unsure as to intention 0.0 0.0 0.0 0.0 0.0 0.0 Do not intend to use 86.3 82.8 80.0 80.5 72.5 76.8 Missing 0.9 0.2 1.1 1.2 0.6 0.7 Previously used contraception Intend to use in next 12 months 0.0 0.5 2.2 0.9 3. l 2.2 Intend to use later 0.4 0.5 1.3 1.0 0.9 0.9 Unsure as to timing 0,0 0.8 1.9 2.1 2.7 2.1 Unsure as to intention 0.0 0.0 0.0 0.0 0.0 0.0 Do not intend to use 1.2 1.7 3,9 5.3 8.4 6.1 Missing 0.0 0,0 0.0 0.0 0,1 0.0 Total I00.0 100.0 100.0 100.0 100.0 100.0 All currently married nonusers Intend to use in next 12 months 0.2 3.7 5.0 4.4 7.3 5.6 Intend to use later 1.6 3.6 2.9 1,7 2.4 2.4 Unsure as to timing 9.9 8.1 7.0 6.9 8.8 8.4 Unsure as to intention 0.0 0.0 0.0 0.0 0.0 0.0 Do not intend to use 87.5 84.5 83.9 85.8 80.9 82.9 Missing 0.9 0.2 1.1 1.2 0.7 0.8 Total 100.0 100.0 100.0 100.0 100,0 100.0 Number of women 459 558 495 624 2701 4838 llncludes current pregnancy Method preferences of potential future users are shown in Table 4.13. The pill is by far the most popular choice (44 percent). Injection is a distant second (13 percent), followed by the IUD (12 percent). Female sterilization is the preferred method for 5 percent of women, but none of the women selected male methods---condoms, withdrawal, and male sterilization. Women's preferences for methods differ by the time in which they intend to use contraception (in the next 12 months or later). However, the pill, the IUD, and injection are the most popular methods regardless of timing. The pill and female sterilization are preferred more by those who intend to use a family planning method after 12 months, while the IUD and prolonged breast feeding are more often chosen by those who want to use family planning sooner. Compared to women who intend to use in the next 12 months, those who plan to use later show a slightly stronger preference for the pill and slightly weaker preference for prolonged breastfeeding. 49 Table 4.13 Preferred method of contraception for future use Percent distribution of currently married women who are not using a contraceptive method but who intend to use in the future by preferred method, according to whether they intend to use in the next 12 months or later, Yemen 1991/92 Intend to use In next After Unsure Preferred method 12 12 as to of contraception months months timing Total Pill 41.1 56.7 42.3 43.8 IUD 15.5 10.3 9.7 11.7 Injection 12.3 11.5 14.2 13.1 Diaphragm/Foam/Jelly l.l 0.4 1.0 0.9 Condom 0.6 0.0 0.1 0.3 Female sterilization 4.9 7.4 4.6 5.1 Male sterilization 0.2 0.0 0.7 0.4 Safe period 3.4 7.7 4.1 4.4 Withdrawal 0.9 0.0 0.4 (I.5 Prolonged breastfeeding 10.0 4.7 7.5 7.9 Other 9.5 1.2 14.6 10.8 Don't know 0.5 0.0 0.5 0.4 Missing 0.0 0.0 0.3 0.5 Total 100.0 100.0 100.0 100.0 Number 271 1 l 7 405 796 4.11 REASON FOR NONUSE OF CONTRACEPT ION Currently married non-pregnant women who had indicated that they would not use any method in the future were asked the reason for not planning to use family planning. The percent distribution of these women by reason proffered lot not intending to use any method are presented in Table 4.14. The results in the table are shown by (a) broad age groups, and (b) fertility intentions, according to past experience or no experience with contraception. "Other" is a residu'al category and includes women who could not be assigned to the categories "want more" or "want to limit." The expressed reasons for not intending to use family planning can be broadly grouped into the following categories: reasons related to contraceptive methods; attitude toward family planning, fatalistic attitude and "other" reasons. The reasons that are related to contraceptive method are: lack of knowledge, side effects, inconvenience of use, and difficulty obtaining method. About one-fourth of nonusers said they would not use any method because they lacked knowledge of contraceptive methods. One in 10 women had no plan to use contraception because of fear of side effects, 3 percent considered the methods inconvenient to use, and 1 percent mentioned difficulty in obtaining methods. One-third of women gave attitudinal reasons: although only 2 percent were personally opposed to family planning, the remaining were split equally between religious prohibition (15 percent) and disapproval of husbands (16 percent). Ten percent of women were fatalistic as far as family size was concerned and implied there was little they can do to control their fertility. Only 3 percent of women were menopausal; 13 percent gave "other" reason for nonuse. 50 Table 4.14 Reasons for not intending to use contraception Percent distribution of women who are not using a contraceptive method and who do not intend to use in the future by main reason for not intending to use, according to age, fertility intentions and ever use o f contraception, Yemen 1991/92 Fertility intentions Want more Want to limit Reason for not intending Age Never Never to use Used used Used used contraception 15-29 30-49 method method method method Other I Total Religious prohibitions 13.9 16.4 9.8 14.8 4.6 17.6 16.3 15.3 Opposed to family planning 2.3 1.0 3.4 2.3 0.0 0.9 0.5 1.6 Husband disapproves 18.2 14.1 13.4 17.1 9.1 15.9 14.2 15.9 Relatives disapprove 0.4 0.0 0.7 0.3 0.0 0.1 0.2 0.2 Side effects 8.9 10.2 29.9 5.6 46.1 11.0 6.6 9.6 Lack of knowledge 22.1 24.3 3.6 23.9 0.2 23.4 31.8 23.3 Difficult to obtain 0.6 1.5 0.7 0.7 1.6 2.1 0.5 1.1 Costs too much 0.2 0.8 0.0 0.3 0.9 1.2 0.3 0.5 Inconvenient to use 2.5 3.3 3.4 2.6 8.6 3.0 2.1 2.9 Fatalistic 8.6 10.5 12.8 8.8 11.4 8.9 13.5 9.7 Menopausal/subfecund 0.5 5.1 1.3 0.7 7.0 7.9 0.5 3.0 Other 17.5 8.8 19.3 19.0 6.5 4.7 6.7 12.7 Unsure/Don't know 4.2 3.9 1.6 3.9 4.0 3.3 6.6 4.0 Missing 0.1 0.1 0.0 0.0 0.0 0.1 0.2 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 1786 2223 101 2027 166 1183 532 4009 IAnswers other than "want more" or "want no more" to the question on future reproductive intentions. The major differences in reasons by age were attitudinal reasons, religious prohibitions and disapproval of husbands. Religious prohibition was mentioned more often by women over 30 than by those who were younger, and disapproval of husbands was cited by a greater proportion of younger women. The reason mentioned most often by those who had never used a method was lack of contraceptive knowledge. Among those who had some experience with contraception, the major reason for not intending to use was the side effects associated with contraceptive use. Thirty percent of women who want more children and 46 percent of those who want to limit their family size gave this reason. Religious prohibition and husband's opposition to using family planning are mentioned by a substantial proportion of women whether or not they want more children in the future. It is surprising that, irrespective o f reproductive intentions, a slightly higher proportion of women who have used a family planning method than those who have not used a method are fatalistic. 4.12 SPOUSAL COMMUNICATION AND APPROVAL OF FAMILY PLANNING An indication of the acceptability of family planning is the extent to which couples discuss the subject with each other. TheYDMCHSincludedanumberofquest ionsonatt i tudesofspouses. Respondents who were not using any method at the time of survey were asked, "In your opinion, in general, does your husband approve or disapprove of couples using a family plarming method?" Respondents were also asked if they had ever talked with their spouses about family planning. Respondents who had previously given "husband's disapproval" as the major reason for not intending to use contraception in the future were not asked these 51 questions. It was assumed that they had already talked with their husbands about family planning. Table 4.15 presents a cross tabulation of the responses on spousal communication and approval of family planning given by currently married women who know of a contraceptive method (includes assumed answers). Of 2,705 women included in the table, 1,225, or 45 percent have talked about family planning with their husbands. Among women who talked with their husbands, 51 percent said that their husbands approved, 3 percent said their husbands approved conditionally, and 44 percent said that their husbands disapproved of family planning. In contrast, one-tenth of those who had not talked about family planning believed that their husbands approved of couples using a method of family planning. However, understandably, most of the women who had not talked with their spouses did not know about their husband's attitude toward family planning (55 percent). Table 4.15 Communication with husband and husband's opinion Percent distribution of currently married women who know a contraceptive method and are currently not using any method by their husband's attitude toward family planning, according to whether they discussed family planning with their spouse, Yemen 1991/92 Discussed with spouse Husband's Not opinion Yes I No stated Total Husband approves 50.6 10.6 0.0 28.5 Husband conditionally approves 3.0 0.6 0.0 1.7 Husband disapproves 44.3 33.6 2.3 37.9 Don't know 2.1 55.2 97.7 31.9 Total 100.0 100.0 100.0 100.0 Number 1225 1432 48 2705 lIncludeswomen who were not askedifthey ever talked withtheir husbands about family planning because they cited husband's disapproval &s reason ~r not planning to use a con~aceptive method in the future. 52 CHAPTER 5 NUPTIALITY AND EXPOSURE TO RISK OF PREGNANCY This chapter presents results on the principal factors other than contraception which affect a woman's risk of pregnancy, i.e., nuptiality, postpartum amenorrhea, and secondary infertility. The marital structure of a population directly affects population dynamics. Marriage, divorce, and widowhood are demographic events that influence exposure to pregnancy and thereby affect fertility. Marriage and fertility are closely linked in Yemen because, like most Arab countries, marriage is a primary indicator of women's exposure to risk of pregnancy and almost no childbearing takes place outside the context of marriage. Also, the traditional social structure of the country supports a natural fertility regime in which the majority of women do not use any means of fertility regulation. Since marriage is the primary indicator of exposure to the risk of pregnancy, and the length of time women are married directly affects overall fertility, the study of marriage patterns is essential to the understanding of fertility. Of equal significance for examining the levels and trends in fertility are duration of postpartum amenorrhea, which leads to temporary insusceptibility to the risk of pregnancy, and the onset of menopause and terminal infertility. 5.1 CURRENT MARITAL STATUS In the household questionnaire, information was collected on the marital status of females and males 10 years of age and older. Table 5.1 presents the distribution of women 15-49 by current marital status. By age 50, almost all Yemeni women have entered into a marital union. At the time of the survey, among all women 15-49 about three in four women bad married: 72 percent currently married, 2 percent widowed, and another 2 percent divorced or separated. In comparison with the five other Arab countries in North Africa and the Middle East where the DHS surveys have been implemented, the proportion of women 15-49 who have ever married is highest in Yemen) Table 5.1 Current marital status Percent distribution of all women by current marital status, according to age, Yemen 1991/92 M~ital status Number Never of Age married Married Widowed Divorced Separated Total women 15-19 75.3 23.5 0.2 0.8 0.2 100.0 1729 20-24 28.2 69.4 0.5 1.5 0.3 100.0 1135 25-29 9.1 86.2 1.2 3.1 0.3 100.0 1425 30 34 2.5 91.4 2.8 2.7 0.6 100.0 1021 35-39 1.0 92.5 4.0 2.5 0.0 100.0 982 40-44 0.2 94.1 4.4 1.3 0.0 100.0 654 45-49 0.0 90.0 7.4 2.6 0.0 100.0 529 Total 23.9 71.6 2.2 2,0 0.2 100.0 7475 1The proportion of women 15-49 who have ever married: Egypt 1992:70 percent (EI-Zanaty et al,, 1993); Morocco 1992:60 percent (Azelmet et al., 1993); Sudan 1989/90: 61) percent (DOS and IRD, 1991); Tunisia 1988:60 percent (Aloui et al., 1989); and Jordan 1990:56 percent (Zou'bi et al., 1992). 53 Although in age group 15-19 only one in four women has married, the overwhelming majority of women 20-24 have married (72 percent), and among those 30 years and older almost no women are single. The proportion of women widowed increases with age from less than 1 percent for the youngest women to 7 percent for oldest women. There was no clear pattern for marriages dissolved by divorce. Marriage is not as stable in Yemen as it might appear from the small proportion of women who are divorced or widowed; remarriage is relatively common (see Table 5.2). However, a large proportion of women have married only once (89 percent); 11 percent of women have married more than once. Table 5.2 Remarriage Percentage of ever-married women who had married only once by current age and selected background characteristics, Yemen 1991/92 Cuffentage Background characteristic 15-19 20-24 25-29 30-34 35 39 40-44 45-49 Total Residence Urban 99.1 94.0 90.9 89.3 81.7 78.2 87.0 88.3 Rural 98.8 95.1 91.1 89,0 85.2 82.4 80.1 88.8 Region North./West. 98.8 94.6 90.4 88.7 84.1 81.5 80.2 88.3 South./East. 99.1 96.1 94.7 91.4 86.8 81.7 86.0 91.1 Education Illiterate 98.6 94.1 90.6 88.7 84.5 81.8 81.4 88.1 Literate 100.0 100.0 100.0 96.6 91.4 89.5 100.0 96.6 Primary 99.1 97.2 90.6 90.9 81).7 60.5 54.9 92.7 More than primary 100.0 98.7 98.2 95.0 88.1 81.4 65.5 96.0 Total 98.8 94.9 91.1 89.1 84.6 81.6 81.2 88.7 The proportion of women who have married more than once increases with age. Five percent of women age 20-24 have remarried. The proportion of women who have married at least twice increases to I0 percent in age groups 25-29 and 30-34, and peaks at 20 percent among women in their forties. Dissolution of the first marriage is as likely to occur among women in urban areas as in rural areas. Regionally, women in each age group have a higher proportion of undissolved marriages in the southem and eastem govemorates compared to those in northem and western governorates region; although the differences are very small. Illiterate women are more likely (12 percent) to be married more than oncc than those who arc literate or who have completed at least primary school (3-7 percent). These differences may be due to lack of education among older women who have longer exposure to marriage and hence a greater probability of dissolution of their first marriage due to divorce or death of a spouse. 5.2 AGE AT FIRST MARRIAGE Table 5.3 shows the percentage of women who were first married by speci fled ages and the median age at first marriage by current age. For each age cohort the accumulated percentages stop at the lower age boundary of the cohort to avoid censoring problems. For instance, for the cohort currently age 20-24, accumulation stops with percentage married by exact age 20. 54 Table 5.3 Age at first marriage Percentage of women who were first married by exact age 15, 18, 20, 22, and 25, and median age at first marriage, according to current age, Yemen 1991/92 Current age 15 18 20 22 Percentage of women who were Percentage Median first married by exact age: who had Number age at never of first 25 married women marriage 15-19 8.7 NA NA NA NA 75.3 1729 a 20-24 20.4 49.2 62.6 NA NA 28.2 1135 18.1 25-29 28.7 65.4 76.6 82.7 89.0 9.1 1425 16.2 30-34 33.1 72.5 82.4 89.1 94.4 2.5 1021 15.7 35-39 31.4 69.2 79.8 90.5 94.9 1.0 982 15.7 4044 31.6 71.3 79.8 89.5 92.4 0.2 654 15.7 45-49 31.7 70.6 79.3 85.4 89.9 0.0 529 15.7 20-49 28.9 65.2 76.0 83.5 88.1 8.5 5746 16.0 25-49 31.0 69.2 79.3 87.1 92.1 3.6 4611 15.8 NA = Not applicable aThe median age was not calculated because less than 50 percent of the women in the age group x to x+4 were first married by age x. Early marriage (by age 15), once prevalent in Yemen, is now on a decline. Almost one-third of women 30 and over had married by age 15, while in younger age groups the proportion married by this age has steadily decreased to less than 10 percent among teenagers. The proportion of women married by exact age 20 is also lower among women 20-24, than among older women. A similar pattem is seen for median age at first marriage. Overall, the median age at marriage for women 25-49 is 16 years; it is slightly lower for women 30-49 (15.7 years) and slightly higher for women 25-29 (16.2 years). One-half of women in age group 20-24 marry after their eighteenth birthday or more than two years later than the national average. These results indicate that age at marriage may be rising as early marriages become less common. Tab le 5.4 2 presents differentials in the median age at marriage for various groups of women. Because more than 50 percent of women in age group 20-24 are single for certain categories of background characteristics, median age at marriage in the following discussion is confined to women age 25-49. The median age at marriage in urban areas is almost the same as in rural areas, but in the two regions median age at marriage differs by more than one year. Women in the southern and eastem govemorates marry later than women in the northem and western govemorates. While there are only minor differences in age at first marriage by residence and region, education piays an important role in determining women's entry into marriage. Improvements in educational opportunities for women have resulted in their staying in school longer, thus pushing the age at marriage upward. The difference in age at marriage is more than seven years between women with no formal education (mcdian 15.7 years) and those who have attained more than primary schooling (median 22.9 years). Overall differences in the median age at marriage for various background characteristics, mainly due to variation in the age at marriage among younger women (i.e., under 35 years); age at marriage differs little among older women. 2Because the number of Yemeni women who can read and write is small, and in order to make the results presented in this table comparable with those presented in tables in Chapter 3 (Fertility), the education categories in this table are the same as those in Chapter 3. 55 Table 5.4 Median age at first marriage Median age at first marriage among women age 20-49 years, by current age and selected background characteristics, Yemen 1991/92 Current age Women Women Background age age characteristic 20-24 25-29 30-34 35-39 40-44 45-49 20-49 25-49 Residence Urban a 17.6 16.1 15.6 15.3 15.6 16.9 16.0 Rural 17.6 15.9 15.7 15.8 15.7 15.7 15.9 15.8 Region Nor th.AVest. 17.3 15.8 15.6 15.6 15.7 15.7 15.8 15.7 South./East. a 19.2 17.0 16.7 15.8 15.8 18.0 17.0 Education No schooling 17.3 15,9 15.6 15,7 15.7 15.7 15.8 15.7 Primary 18.8 17.3 16.7 15.5 15.3 22.7 17.6 16.7 More than primary a 23.6 23.3 20.8 20.5 2(1.1 a 22.9 Total 18.1 16.2 15.7 15.7 15.7 15.7 16.0 15.8 Note: Medians are not shown for women 15-19 because less than 50 percent have married by age 15 in all subgroups shown in the table. aThe median age was not calculated because less than 50 percent of the women in the age group were first married by age 20. Differences in Ages of Husbands and Wives Using the ages of husbands and wives from the Women's Questionnaire, the difference between husband's age and wife's age was calculated by subtracting wife's age from that of the husband. Table 5.5 presents tbe difference in couple's ages by age of wi re. It is clear that Yemeni men rarely marry women older than themselves. In one-third of marriages, husbands were either the same age as their wives or at most four years older than their spouses; in another one-third, they were 5-9 years older. Six percent of women had married men who were 15-19 years older than they were. Furthermore, in 6 percent of couples this difference was 20 years or more. The average difference in age between wives and their spouses was 7 years--6 years for younger women (under age 30) and 8 years for older women. In considering the differences in ages of wives and husbands in the regions--7.3 in the northem and western governorates and 7.0 in the southern and eastern govemorates--it should be noted that in the southem and eastern governorates husband's age was not reported for a relatively high proportion of wives. Ideal Age at Marriage All women surveyed in the YDMCHS were asked: "In your opinion, what is the most suitable age for (your daugbter/a girl) to marry?" Before considering the results, it should be noted that almost 1 in 10 respondents gave non-numeric answers. However, two ages, 15 years and 20 years 3 were most often mentioned as the most suitable ages for girls to marry (see Table 5.6). More than half of respondents gave these responses. One in four ever-married womcn said the ideal age at marriage was 20-21 (actually, age 20); 3Although the age category 20-21 includes both ages 20 and 21, an examination of the data indicate that less than one percent of women reported 21 years as the ideal age at marriage; the rest reported 20 years. 56 Table 5.5 Age difference between spouses Percent distribution of currently married couples by age difference between spouses and mean age difference, according to wife's age and region, Yemen 1991/92 Number of years by which husband is older ltusband's Number Mean age Wife's agel Wife age of difference region older 0-4 5-9 10-14 15-19 20+ missing Total women (years) Wife's age 15-19 0.0 43.8 35.0 13.8 4.1 1.7 1.7 100.0 406 6.0 20-24 1.9 37.7 39.3 12.6 3.6 3.1 1.8 100.0 788 6.3 25-29 2.3 40.4 31.6 14.5 5.4 3.9 1.7 100.0 1229 6.4 30-34 4.3 26.2 33.2 21.0 6.3 6.6 2.4 100.0 933 7.9 35-39 4.0 31.1 30.3 18.2 7.6 7.5 1.4 100.0 909 7.7 40-44 3.9 26,4 27.6 23.4 6.5 9.6 2.6 100.0 615 8.4 45-49 5.4 25.0 31.3 15.8 11.1 9.0 2.3 100.0 476 8.4 Raglon North./West. 3.1 33.3 32.7 17.3 6.4 6.1 0.9 100.0 4458 7.3 South./East. 3.3 32.7 31.8 16.0 5.1 4.2 6.8 100.0 897 7.0 Total 3.2 33.2 32.6 17.1 6.2 5.8 1.9 100.0 5355 7.2 Note: MeeLn age difference is calculated by subtracting the mean age of wives from the mean age of husbands. Table 5.6 Ideal age at marriage for women Percent distribution of ever-married women by age at first marriage considered suitable most for daughter/girl, according to woman's current age and age at first marriage, Yemen 1991/92 Current age/ Age at first marriage considered Median Age at most suitable for daughter/girl Non- age at Number first numerical first of marriage <15 15 16-17 18-19 20-21 22-23 24+ response Missing Total marriage women Current age 15-19 4,0 30.8 12.2 11.4 26.1 1.5 6.5 6.0 1.4 100.0 17.8 427 20-24 6.8 26.9 11.6 8.6 27.4 1.4 8.6 7.4 1.3 100.0 18.1 815 25-29 6.1 31.0 12.0 7.8 23.7 1.7 7.9 8.8 1.0 100.0 17.2 1295 30-34 6.2 29.0 14.1 6.2 24.8 1.5 7.9 9.5 0.7 100.0 17.1 995 35-39 5.6 29.2 13.3 8.0 22.5 1.1 9.1 10.8 0.6 100.0 17.3 972 40-44 7.1 30.4 11.1 7.2 24.7 1.6 6.4 10.7 0.8 100.0 17.0 653 45-49 7.6 26.6 12.9 9.6 20.7 1.0 8.8 11.8 0.9 10(I.0 17.1 529 Age at first marriage < 15 9.2 32.6 10.0 7.0 21.7 1.8 6.8 10.0 0.9 100.0 16.4 1810 15 5.1 37.3 14.0 6.3 21.1 0.8 5.4 9.3 0.6 1011.0 16.4 1321 16-17 3.6 26.0 19.0 9,5 26.0 1.0 4.9 8.8 1.3 l(g).0 17.5 999 18-19 6.4 22.3 10.8 10.7 28.5 1.5 9.3 9.5 1.1 100.0 18.7 665 20-21 4.7 17.6 11.0 7.6 32.8 1.6 14.1 9.8 0.8 100.0 20.1 430 22-23 5.9 22.3 9.6 10.2 24.0 3.3 15.5 8.9 0.2 100.0 18.9 191 24+ 4.0 20.2 7.5 10.1 25.9 1.8 22.0 6.7 1.7 100.0 20.2 270 Total 6.2 29.3 12.5 8.0 24.2 1.4 8.0 9.3 0.9 100.0 17.4 5687 57 a slightly higher proportion mentioned age 15 as the ideal age. More than one-third of women were of the opinion that it was "best" if a girl/daughter married at age 15 or younger. Twenty-one percent mentioned ages between 16-19 years, and only 9 percent thought that ideal time for a girl to marry was after completing 22 years. There is a close association between the age at which a woman marries and the age she considers ideal for marriage. A majority of women who had first married before age 20-21 considered the ideal age for marriage either the same as or later than the age at which they themselves had married. But a majority of those who first married after age 21 consider their own age at marriage as the maximum age by which a girl should be married. The median ideal age at marriage in Yemen is 17.4 years. The median for women under 25, (18 years) is higher than for women 25 and older (17 years). Marriage Between Relatives In Yemen, as in many Arab countries, marriage between blood relatives (consanguineous marriages), usually occurs between first cousins. Table 5.7 presents data on first marriages between relatives for ever- married women. One in three women in Yemen has married her first cousin, mostly on the father's side, 5 percent have married other relatives, and 63 percent were not related to their husbands before marriage. Marriage to first cousins is also prevalent among women whose previous marriage was dissolved by widowhood or divorce. The last marriage of 1 in 5 women who had remarried was to her first cousin (data not shown). It appears from Table 5.7 that cousin marriages are increasing in Yemen as indicated by the higher proportion of women marrying first cousins among younger women (under age 25) and also among women who have first married within the last 10 years. Consanguineous marriages are more prevalent in rural (23 percent) than in urban areas (18 percent). This is probably because most women whose educational level is above primary, who are less likely to be married to relatives, live in urban areas. The difference by region is small: in the southem and eastem govemorates, women are slightly more likely to be married to a relative. 5.3 POLYGYNY As a Muslim country, Yemen considers polygyny legal. Islam permits a man to have up to four wives at a time, provided the husband treats all of them equally. In order to collect information on the practice of polygyny, all currently married women in the survey were asked whether their husbands had other wives. Table 5.8 shows the percentage of currently married women in polygynous unions by age and selected background characteristics. Only six percent 4 of currently married women live in a polygynous union, with no difference in urban and rural areas. Although the overall difference in prevalence of polygynous marriages in urban and rural areas is negligible, urban women of all ages except those in age groups 30-34 and 45-49 are more likely to be in polygynous marriages than women living in rural areas. The most noticeable difference in the prevalence of polygyny is between regions. Less than 3 percent of currently married women have one or more co-wives in the southem and eastern governorates, whereas, in the northem and westem governorates the rate of polygynous marriage is more than double. Women's education has an inverse impact on polygyny. Women who have completed primary or higher education are less likely to be in a polygynous marriage. '~The prevalence of polygyny is almost the same in Morocco (5 percent) (Azelmet et al., 1993) as in Yemen, but much higher in Sudan (20 percent) (DOS and IRD, 1991). Polygyny is illegal in Tunisia. The question on number of co-wives was not included in the Individual Questionnaire in the 1990 DHS survey in Jordan; but in the Household Questionnaire information was collected on the number of wives each married man had. Of all married men listed, 5 percent had more than one wife (special tabulation June 14, 1993). 58 Table 5.7 Consanguinity Percent distribution of ever-married women by relationship to their first husband, according to selected background characteristics, Yemen 1991/92 First cousin: Number Background Father's Mother's Other Not of characteristic side side relative related Missing Total women Age 15-19 29.1 12.8 5.9 51.7 0.5 100.0 427 20-24 25.0 10.2 5.2 59.3 0.3 100.0 815 25-29 23.3 10.2 5.5 60.7 0.3 100.0 1295 30-34 19.9 7.7 5.8 66.4 0.2 100.0 995 35-39 21.8 6.4 5.5 65.9 0.4 100.0 972 40-44 19.9 6.4 3.5 69.5 0.7 100.0 653 45-49 19.1 7.7 3.2 69.5 0.4 100.0 529 Age at first marr iage < 15 23.7 8.9 4.6 62.5 0.4 100.0 1810 15 22.6 7.5 5.4 63.9 0.5 100.0 1321 16-17 21.0 9.3 5.0 64.4 0.2 100.0 999 18-19 22.1 10.5 6.6 60.5 0.4 100.0 665 20-21 21.9 9.1 5.2 63.8 0.0 100.0 430 22-23 21.3 9.1 5.8 63.9 0.0 100.0 191 24+ 19.3 4.5 3.1 72.5 0.6 lO0.0 270 Years since first marriage < 5 24.6 9.9 4.8 60.5 0.2 100.0 852 5-9 23.8 10.7 5.8 59.2 0.5 100.0 929 10-14 21.8 9.9 5.6 62.5 0.2 100.0 1119 15-19 22.7 7.3 6.2 63.5 0.3 100.0 1046 20-24 22.6 6.6 4.4 66.0 0.4 100.0 825 25+ 18.9 7.2 3.5 69.9 0.6 100.0 916 Residence Urban 18.4 7.8 7.5 66.1 0.3 100.0 1054 Rural 23.3 8.8 4.6 62.9 0.4 100.0 4633 Region North./West. 22.2 8.4 5.0 64.1 0.4 100.0 4708 South.lEast. 23.2 9.8 5.8 60.8 0.3 100.0 979 Education Illiterate 22.6 8.3 4.8 63.8 0.4 100.0 5075 Literate 25.4 12.5 8.2 53.9 0.0 100.0 61 Primary 25.4 10.5 5.8 58.4 0.0 100.0 316 More than primary 12.5 12.1 9

View the publication