Ghana - Demographic and Health Survey -1989

Publication date: 1989

Ghana Demographic and Health Survey 1988 Ghana Statistical Service ®DHS Demographic and Health Surveys Institute for Resource Development/Macro Systems, Inc. Ghana Demographic and Health Survey 1988 Ghana Statistical Service Accra, Ghana Institute for Resource Development/Macro Systems, Inc. Columbia, Maryland USA September 1989 This report presents the findings of the Ghana Demographic and Health Survey (GDHS). The survey was a collaborative effort between the Ghana Statistical Service and the Institute for Resource Devalopment/Macro Systems, Inc. (IRD). The survey is part of the worldwide Demographic and Health Surveys Program, which is designed to collect data on fertility, family planning, and maternal and child health. Funding for the survey was provided by the U.S. Agency for International Development (Contract No. DPE-3023-C-00-4083-00), the Government of Ghana and the United Nations Population Fund. The United Nations Children's Fund loaned vehicle* for use during the survey fieldwork. Additional information on the GDHS can be obtained from the Ghana Statistical Service, P.O. Box 1098, Accra, Ghana. Additional information about the DHS Program can be obtained by writing to: DHS Program, IRD/Macro Systems, Inc., 8850 Stanford Blvd., Suite 4000, Columbia, ME) 21045, USA (Telephone: 301-290-2800; Telex: 87775; Fax: 301-290-2999). CONTENTS Page CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii L IST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi i L IST OF F IGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi MAP OF GHANA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiv CHAPTER 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 HISTORY, GEOGRAPHY AND ECONOMY . . . . . . . . . . . . . . . . . . . . . 1 EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 POPULATION AND FAMILY PLANNING PROGRAMME . . . . . . . . . . . . 2 HEALTH PRIORITIES AND PROGRAMMES . . . . . . . . . . . . . . . . . . . . 3 OBJECTIVES OF THE SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 ORGANISATION OF THE SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . 4 BACKGROUND CHARACTERISTICS OF SURVEY RESPONDENTS . . . . . 5 CHAPTER 2 2.1 2.2 2.3 2.4 2.5 NUPT IAL ITY AND EXPOSURE TO THE RISK OF PREGNANCY . . . . 9 CURRENT MARITAL STATUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 POLYGYNY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 AGE AT FIRST UNION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 BREAST~t~EDING AND POSTPARTUM INSUSCEPTIBILITY . . . . . . . . 12 MEAN DURATION OF BREASTI~EEDING AND POSTPARTUM INSUSCEPTIBILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 iii CHAPTER 3 3.1 3.2 3.3 3.4 3.5 3.6 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 4.13 4.14 4.15 4.16 Page FERTILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 CURRENT AND CUMULATIVE FERTILITY BY BACKGROUND CHARACTERISTICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 FERTILITY TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 CHILDREN EVER BORN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 CHILDREN EVER BORN AND AGE AT FIRST MARRIAGE . . . . . . . . 26 AGE AT FIRST BIRTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 AGE AT FIRST BIRTH BY BACKGROUND CHARACI 'ERISTICS . . . . . 27 FERTILITY REGULATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 KNOWLEDGE OF METHODS AND SOURCE . . . . . . . . . . . . . . . . . . . 30 KNOWLEDGE OF MODERN METHODS AND SOURCE BY BACKGROUND CHARACTERISTICS . . . . . . . . . . . . . . . . . . . . . . . . 31 ACCEPTABILITY OF METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 KNOWLEDGE OF SUPPLY SOURCES . . . . . . . . . . . . . . . . . . . . . . . 34 EVER USE OF CONTRACEPTION . . . . . . . . . . . . . . . . . . . . . . . . . . 34 CURRENT USE OF CONTRACEPTION . . . . . . . . . . . . . . . . . . . . . . . 37 CURRENT USE BY BACKGROUND CHARACTERISTICS . . . . . . . . . . 38 NUMBER OF CHILDREN AT FIRST USE . . . . . . . . . . . . . . . . . . . . . 40 KNOWLEDGE OF THE FERTILE PERIOD . . . . . . . . . . . . . . . . . . . . . 40 SOURCE OF SUPPLY OF CONTRACEPTION . . . . . . . . . . . . . . . . . . . 42 ATTITUDE TOWARD BECOMING PREGNANT . . . . . . . . . . . . . . . . . 42 REASONS FOR NONUSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 FUTURE USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 PREFERRED METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 ACCEPTABILITY OF MEDIA MESSAGES ON FAMILY PLANNING . . . 46 ATTITUDE TOWARD FAMILY PLANNING . . . . . . . . . . . . . . . . . . . . 47 iv 4.17 4.18 CHAPTER 5 5.1 5.2 5.3 5.4 CHAPTER 6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 CHAPTER 7 7.1 7.2 7.3 7.4 7.5 7.6 Page ATTITUDE TOWARD FAMILY PLANNING BY BACKGROUND CHARACTERISTICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 DISCUSSION OF FAMILY PLANNING . . . . . . . . . . . . . . . . . . . . . . . 48 FERTILITY PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 DESIRE FOR CHILDREN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 NEED FOR FAMILY PLANNING . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 IDEAL FAMILY SIZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 FERTILITY PLANNING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 MORTALITY AND HEALTH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 MORTALITY DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 MORTALITY TRENDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 MORTALITY DIFFERENTIALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 PRENATAL CARE AND DELIVERY ASSISTANCE . . . . . . . . . . . . . . . 67 CHILDHOOD IMMUNISATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 DIARRHOEA PREVALENCE AND TREATMENT . . . . . . . . . . . . . . . . 73 PREVALENCE AND TREATMENT OF FEVER AND RESPIRATORY ILLNESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 NUTRITIONAL STATUS OF CHILDREN . . . . . . . . . . . . . . . . . . . . . . 79 HUSBAND'S SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 CHARACTERISTICS OF THE SAMPLE . . . . . . . . . . . . . . . . . . . . . . . 89 MARRIAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 FERTILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 CONTRACEPTIVE KNOWLEDGE AND USE . . . . . . . . . . . . . . . . . . . 94 PROBLEMS WITH METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 SOURCE FOR METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 V Page 7.7 7.8 7.9 7.10 NUMBER OF CHILDREN AT F IRST USE . . . . . . . . . . . . . . . . . . . . . 99 INTENTION TO USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 ATT ITUDE TOWARD FAMILY PLANNING . . . . . . . . . . . . . . . . . . . 100 FERT IL ITY PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 APPENDIX A APPENDIX B B.1 B.2 B.3 APPENDIX C APPENDIX D SURVEY PERSONNEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 SAMPLE DESIGN AND IMPLEMENTATION . . . . . . . . . . . . . . . . 117 DESCRIPT ION OF SAMPLE DESIGN . . . . . . . . . . . . . . . . . . . . . . . 119 SAMPLING PROBABIL IT IES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 SAMPLE IMPLEMENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 SAMPL ING ERRORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 QUEST IONNAIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 vi LIST OF TABLES Page Table 1.1 Percentage Distribution of Women by Age, Urban-Rural Residence, Region and Level of Education, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . 7 Table 1.2 Percentage Distribution of Women of Reproductive Age, 1984 Census and 1988 GDHS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 1.3 Percentage Distribution of Women by Level of Education, According to Age, Urban-Rural Residence, and Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 2.1 Percentage Distribution of Women by Current Marital Status, According m Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 2.2 Percentage of Currently Married Women in a Polygynous Union, by Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . 11 Table 2.3 Percentage Distribution of Women by Age at First Union and Median Age at First Union, According to Current Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Table 2.4 Median Age at First Union Among Women Aged 20-49 Years, by Current Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Table 2.5 Percentage of Births Where Mothers are Still Breastfeeding, Postpartum Amenorrhoeic, Abstaining, and Insusceptible, by Number of Months Since Birth, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Table 2.6 Mean Number of Months of Breastfeeding, Postpartum Amenorrboea, Postpartum Abstinence, and Postpartum Insusceptibility, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Table 3.1 Total Fertility Rate (TFR) for Calendar Year Periods and for Five Years Preceding the Survey, and Mean Number of Children Ever Born to Women 40-49 Years of Age, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Table 3.2 Percentage of All Women Who are Currently Pregnant by Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Table 3.3 Age-Period Fertility Rate (Per 1000 Women) by Age of Woman at Birth of Child, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 vii Page Table 3.4 Percentage Distribution of Children Ever Bom (CEB) to All Women and to Currently Married Women, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Table 3.5 Mean Number of Children Ever Born to Ever-Married Women, by Age at First Marriage and Years Since First Marriage, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Table 3.6 Percentage Distribution of Women by Age at First Birth, and Median Age at First Birth, According to Current Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Table 3.7 Median Age at First Birth Among Women Aged 20-49 Years, by Current Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Table 4.1 Percentage of All Women and Currently Married Women Knowing Any Contraceptive Method and Knowing a Source (For Information or Services), by Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . 30 Table 4.2 Percentage of Currently Married Women Knowing at Least One Modem Method, and Knowing a Source for a Modem Method by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 32 Table 4.3 Percentage Distribution of Women Who Have Ever Heard of a Contraceptive Method by Main Problem Perceived in Using the Method, According to Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . 33 Table 4.4 Percentage Distribution of Women Knowing a Contraceptive Method by Supply Source Named, According to Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Table 4.5 Percentage of All Women and Currently Married Women Who Have Ever Used a Contraceptive Method, by Specific Method and Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Table 4.6 Percentage Distribution of All Women and Currently Married Women, by Contraceptive Method Currently Used, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Table 4.7 Percentage of Currently Married Women Currently Using Contraception by Method, 1979 and 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Table 4.8 Percentage Distribution of Currently Married Women by Contraceptive Method Currently Used, According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Table 4.9 Percentage Distribution of Ever-Married Women by Number of Living Children at Time of First Use of Contraception, According to Current Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 viii Page Table 4.10 Percentage Distribution of All Women and Women Who Have Ever Used Periodic Abstinence by Knowledge of the Fertile Period During the Ovulatory Cycle, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Table 4.11 Percentage Distribution of Current Users by Most Recent Source of Supply or Information, According to Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Table 4.12 Percentage Distribution of Non-Pregnant Women Who Are Sexually Active and Who Are Not Using Any Contraceptive Method by Attitude Toward Becoming Pregnant in the Next Few Weeks, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . 43 Table 4.13 Percentage Distribution of Non-Pregnant Women Who Are Sexually Active and Who Are Not Using Any Contraceptive Method and Who Would be Unhappy if They Became Pregnant by Main Reason for Nonuse, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . 44 Table 4.14 Percentage Distribution of Currently Married Women Who Are Not Currently Using Any Contraceptive Method by Intention to Use in the Future, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Table 4.15 Percentage 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, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 46 Table 4.16 Percentage of All Women Who Believe That it is Acceptable to Have Messages About Family Planning on the Radio, by Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 47 Table 4.17 Percentage Distribution of Currently Married Women Knowing a Contraceptive Method by the Husband and Wife's Attitude Toward the Use of Family Planning, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 48 Table 4.18 Percentage of Currently Married Women Knowing a Contraceptive Method Who Approve of Family Planning and Who Say their Husband Approves of Family Planning, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Table 4.19 Percentage Distribution of Currently Married Women Knowing a Contraceptive Method by Number of Times Discussed Family Planning with Husband, According to Current Age, GDHS, 1988 . . . . . . . . . . . . . 50 Table 5.1 Percentage Distribution of Currently Married Women by Desire for Children, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 ix Page Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Percentage Distribution of Currently Married Women by Desire for Children, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Percentage of Currently Married Women Who Want No More Children (including sterilised) by Number of Living Children and Selected Background Characteristics, ODHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 54 Percentage of Currently Married Women Who Are in Need of Family Planning and the Percentage Who Are in Need and Who Intend to Use Contraception in the Future by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Percentage Distribution of All Women by Ideal Number of Children and Mean Ideal Number of Children for All Women and Currently Married Women, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Mean Ideal Number of Children for All Women by Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 58 Percentage of Women Who Had a Birth in the Last 12 Months by Fertility Planning Status and Birth Order, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . 59 Indices for Detecting Underreporting of Infant Deaths, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Distribution of Child Deaths Occurring Between 6 and 24 Months of Age by Calendar Period and Age at Death When Reported, in Months and Years, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Infant and Childlioood Mortality for Calendar Periods, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Infant and Childhood Mortality by Socioeconomic Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Infant and Childhood Mortality by Demographic Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Mean Number of Children Ever Bom, Surviving, and Dead, and Proportion of Children Dead Among Ever Born by Age of Mother, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Percentage Distribution of Births in the Last 5 Years by Type of Prenatal Care for the Mother and Percentage of Births Whose Mother Received a Tetanus Toxoid Injection, According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 68 Page Table 6.8 Table 6.9 Table 6.10 Table 6.11 Table 6.12 Table 6.13 Table 6.14 Table 6.15 Percentage Distribution of Births in the Last 5 Years by Type of Assistance During Delivery, According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Among All Children Under 5 Years of Age, the Percentage with Health Cards, the Percentage Who Are Immunised as Recorded on a Health Card or as Reported by the Mother and, Among Children With Health Cards, the Percentage for Whom BCG, DPT, Polio and Measles Immunisations Are Recorded on the Health Card, by Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Among All Children 12-23 Months, the Percentage with Health Cards, the Percentage Who Are Immunised as Recorded on a Health Card or as Reported by the Mother and, Among Children With Health Cards, the Percentage for Whom BCG, DPT, Polio and Measles Immunisations Are Recorded on the Health Card, by Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Among Children Under 5 Years of Age, the Percentage Reported by the Mother to Have Had Diarrhoea in the Past 24 Hours and the Past Two Weeks, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Among Children Under 5 Years of Age Who Had Diarrhoea in the Past two weeks, the Percentage Consulting a Medical Facility and the Percentage Receiving Different Treatments as Reported by the Mother, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . 75 Among Mothers of Children Under 5 Years of Age, the Percentage Who Know About ORT by Education, and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Among Children Under 5 Years of Age, the Percentage Who Are Reported by the Mother as Having Had Fever in the Past Four Weeks, and, Among Children Who Had Fever in the Past Four Weeks, the Percentage Consulting a Medical Facility and the Percentage Receiving Various Treatments, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Among Children Under 5 Years of Age, the Percentage Who Are Reported by the Mother as Having Suffered from Severe Cough with Difficult or Rapid Breathing in the Past Four Weeks, and, Among Children Who Suffered from Severe Cough with Difficult Breathing, the Percentage Consulting a Medical Facility and the Percentage Receiving Various Treatments, by Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 xi Page Table 6.16 Table 6.17 Table 6.18 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 7.9 Percentage Distribution of Children Aged 3-36 Months by Percentage in Each Standard Deviation Category of Height-for-Age, Using the NCHS/CDC/WHO International Standard (Children With Exact Dates of Birth), According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 81 Percentage Distribution of Children Aged 3-36 Months by Standard Deviation Category of Weight-for-Height, Using the NCHS/CDC/WHO International Standard (Children With Exact Dates of Birth), According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Percentage Distribution of Children Aged 3-36 Months, by Standard Deviation Category of Weight-for-Age, Using the NCHS/CDC/WHO Intemational Standard (Children With Exact Dates of Birth), According to Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Percentage Distribution of Husbands by Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Percentage Distribution of Husbands by Level of Education, According to Age, Urban-Rural Residence and Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Percentage of Husbands in a Polygamous Union, by Age, and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 92 Percentage Distribution of Husbands by Number of Wives, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Percentage Distribution of the Age Difference Between Spouses and Mean Age Difference Between Spouses in the Sample of Married Couples, According to Wife's Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . 93 Percentage Distribution of Husbands by Number of Living Children, According to Age and Mean Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Percentage of Husbands Who Know a Contraceptive Method, Who Know a Source for a Method, Who Have Ever Used a Method, and Who Are Currently Using a Method, by Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Knowledge of Contraception Among Married Couples by Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Percentage of Husbands Who Are Currently Using Any Method and Any Modem Method of Contraception, by Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 xii Page Table 7.10 Percentage Distribution of Husbands Who Have Ever Heard of a Method by Main Problem Perceived in Using the Method, According to Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Table 7.11 Percentage Distribution of Husbands Knowing a Contraceptive Method by Supply Source Named, According to Specific Method, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Table 7.12 Percentage Distribution of Husbands by Number of Living Children at Time of First Use of Contraception, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Table 7.13 Percentage Distribution of Husbands Who Are Not Currently Using Any Contraceptive Method by Intention to Use in the Future, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . 100 Table 7.14 Percentage Distribution of Husbands Who Are Not Currently 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, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 100 Table 7.15 Percentage of Husbands Who Believe That It Is Acceptable to Have Messages About Family Planning on the Radio, by Age, and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 101 Table 7.16 Percentage of Husbands Knowing a Method Who Approve of Family Planning by Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Table 7.17 Percentage Distribution of Husbands Knowing a Contraceptive Method by Number of Times Discussed Family Planning With Wife, According to Age, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Table 7.18 Among Married Couples, Husband's Approval of Family Planning and Wife's Perception of Husband's Approval, GDHS, 1988 . . . . . . . . . . . . . . . . 104 Table 7.19 Percentage Distribution of Husbands by Desire For Children, According to Number of Living ChUdren, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 105 Table 7.20 Percentage of Husbands Who Want No More Children by Number of Living Children and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Table 7.21 Percentage Distribution of Couples by Desire For More Children, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 106 Table 7.22 Percentage Distribution of Husbands by Ideal Number of Children and Mean Ideal Number of Children, According to Number of Living Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 xiii Table 7.23 Table 7.24 APPENDIX B Table B.1 APPENDIX C Table C. 1 Table C.2 Table C.3.1 Table C.3.2 Table C.4.1 Table C.4.2 Table C.4.3 Table C.5.1 Table C.5.2 Table C.5.3 Table C.5.4 Table C.5.5 Table C.5.6 Table C.5.7 Table C.5.8 Table C.6 Page Percentage Distribution of Wives by Husband's Ideal Number of Children, According to Wife's Ideal Number of Children, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Mean Ideal Number of Children by Age and Selected Background Characteristics, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Percentage Distribution of Households and Eligible Women in the GDHS Sample by Results of the Interview ana Response Rates, According to Residence and Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 121 List of Selected Variables with Sampling Errors, GDHS, 1988 . . . . . . . . . . . . . . . 127 Sampling Errors for the Total Population, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 128 Sampling Errors for the Urban Population, GDHS, 1988 . . . . . . . . . . . . . . . . . . . 129 Sampling Errors for the Rural Population, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 130 Sampling Errors for Women Aged 15-24, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 131 Sampling Errors for Women Aged 25-34, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 132 Sampling Errors for Women Aged 35-49, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 133 Sampling Errors for Westem Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . 134 Sampling Errors for Central Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 135 Sampling Errors for Greater Accra, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . 136 Sampling Errors for Eastern Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 137 Sampling Errors for Volta Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Sampling Errors for Ashanti Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 139 Sampling Errors for Brong Ahafo Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 140 Sampling Errors for Upper West, East and Northern Regions, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Sampling Errors for Total Population of Husbands . . . . . . . . . . . . . . . . . . . . . . . . 142 xiv Table C.7.1 Table C.7.2 Table C.8.1 Table C.8.2 Table C.9.1 Table C.9.2 Table C.9.3 Table C.9.4 Table C.9.5 Table C.9.6 Table C.9.7 Table C.9.8 Page Sampling Errors for Husbands - Urban, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . 142 Sampling Errors for Husbands - Rural, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . 143 Sampling Errors for Husbands Aged 15-39, GDHS, 1988 . . . . . . . . . . . . . . . . . . . 143 Sampling Errors for Husbands Aged 40+, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . 144 Sampling Errors for Husbands - Western Region, GDHS, 1988 . . . . . . . . . . . . . . 144 Sampling Errors for Husbands - Central Region, GDHS, 1988 . . . . . . . . . . . . . . . 145 Sampling Errors for Husbands - Eastern Region, GDHS, 1988 . . . . . . . . . . . . . . . 145 Sampling Errors for Husbands - Greater Accra, GDHS, 1988 . . . . . . . . . . . . . . . . 146 Sampling Errors for Husbands - Volta Region, GDHS, 1988 . . . . . . . . . . . . . . . . 146 Sampling Errors for Husbands - Ashanti Region, GDHS, 1988 . . . . . . . . . . . . . . 147 Sampling Errors for Husbands - Brong Ahafo Region, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Sampling Errors for Husbands - Upper West, East and Nort_hem Regions, GDHS, 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 XV LIST OF FIGURES Page Figure 1.1 Figure 2.1 Figure 2.2 Figure 3.1 Figure 3.2 Figure 3.3 Figure 4.1 Figure 4.2 Figure 4.3 Figure 5.1 Figure 5.2 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Figure 6.7 Distribution of Women 15-49 by Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Union Status of Women 15-49 by Age Omup . . . . . . . . . . . . . . . . . . . . . . . . . 10 Duration of Breastfeeding, Amenorrhoea and Post- partum Abstinence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Total Fertility Rate 0-4 Years Before Survey and Children Ever Born to Women 40-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Cumulated Age-Specific Fertility Rates for Women Aged 15-34, GFS and GDHS 1965-1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Age-Specific Fertility Rates, GFS and GDHS . . . . . . . . . . . . . . . . . . . . . . . . . 24 Family Planning Knowledge and Use, Women in Union 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Current Use of Family Planning by Education and Number of Living Children, Women in Union 15-49 . . . . . . . . . . . . . . . . . . . 40 Source of Family Planning Supply, Current Users of Modem Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Fertility Preferences, Women in Union 15-49 . . . . . . . . . . . . . . . . . . . . . . . . . 52 Future Need for Family Planning, Women in Union Not Using Contraception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Trends in Infant and Child Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Immunisation Coverage Among Children 12-23 Months with Health Cards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Age Distribution of Measured Children and Living Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Nutritional Status of Children 3-36 Months Compared to International Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Height of Children 3-36 Months by Education of Mother, Compared to International Reference . . . . . . . . . . . . . . . . . . . . . . . . . 83 Nutritional Status of Children 3-36 Months Compared to International Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Cross-tabulation of Weight-for-Height and Height-for- Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 xvii Figure 7.1 Figure 7.2 Figure 7.3 Page Family Planning Knowledge and Use Among Husbands . . . . . . . . . . . . . . . . . 96 Mean Ideal Number of Children Among Husbands . . . . . . . . . . . . . . . . . . . . 103 Number of Times in the Past Year that Husband Discussed Family Planning With His Wife . . . . . . . . . . . . . . . . . . . . . . . . . . 109 xviii PREFACE The Ghana Demographic and Health Survey (GDHS) is a national sample survey designed to provide information on fertility, family planning and health in Ghana. The survey, which was conducted by the Statistical Service of Ghana, is part of a worldwide programme coordinated by the Institute for Resource Development/Macro Systems, Inc., in more than 40 countries in Africa, Asia and Latin America. The survey was conducted at a time when the government had launched an Economic Recovery Programme with a strong demographic and health component. The statistical information generated by the survey is expected to strengthen the planning, implementation and evaluation of programmes aimed at controlling fertility, promoting family planning, and improving the health status of the people. The results of the survey have also facilitated a further assessment of the demographic situation in the country. An innovative approach in the survey was interviewing a subsample of husbands of the women respondents. This was in recognition of the influence of husbands on the use or nonuse of family planning methods. The data from the survey on the attitudes of husbands about family planning and fertility preferences have provided insight into the factors influencing family planning practices in Ghana. The organisation of the survey benefited from the invaluable collaboration and support of several institutions and organisations both international and local. In particular, our profound gratitude goes to the Institute for Resouree Development/Macro Systems, Inc. for technical assistance, U.S.A.I.D. for funding the survey, UNFPA for additional funding for the training and fieldwork, and UNICEF for the use of vehicles for the fieldwork. We also wish to thank the Ministry of Health, the Department of Community Development, the Department of Social Welfare, the Department of Food and Nutrition, as well as all others who contributed to the success of the survey. Dr. Emmanuel Oti Boateng Government Statistician (Project Director) Statistical Service, Accra xix SUMMARY The Ghana Demographic and Health Survey (GHDS) is a nationally representative self-weighting sample survey of 4,488 female respondents aged 15-49 and a subsample of 943 co-resident husbands of the interviewed women. The survey was carried out by the Ghana Statistical Service between February and June 1988. The basic objective of the survey is to make available to planners and policymakers current information on fertility levels and trends, reproductive intentions of men and women, knowledge and use of contraception, and the current state of maternal and child health. Survey results indicate that fertility continues to be high in Ghana. At current rates, a woman will have an average of six children by the time she reaches her forty-fifth birthday. Urban women have 1.5 fewer births than their rural counterparts. There is a gap of about 3 children between uneducated women and women with more than middle school education. Early and nearly universal marriage among Ghanaian women appears to be one of the reasons for the high level of fertility. Survey data indicate that fewer than 1 percent of Ghanaian women aged 30 and over have never been married. The median age at first marriage has increased slightly over the past ten years, from less than 18.0 years to 18.5 years. In addition to its health benefits for children, breastfeeding is known to offer protection against pregnancy through its influence on the length of postpartum amenorrhoea. Mothers in Ghana breastfeed for an average of 20 months and are amenorrhoeic for an average of 14 months. Mothers abstain from sex for approximately 14 months after a birth. The duration of breastfeeding and postpartum abstinence among urban and more educated mothers is substantially shorter than among rural and less educated women. The low level of contraceptive use is another major factor contributing to high fertility. Even though three-quarters of married Ghanalan women know of some method of contraception, only 37 percent have ever used a method and only 13 percent are currently using a method. Twenty-one percent of married women have used a modem method sometime, with just 5 percent currently using a modem method. Periodic abstinence is the most commonly used method, followed by the pill. In spite of the overall high level of contraceptive knowledge, women who are not using any method attribute their nonuse to a lack of knowledge. Almost 80 percent of husbands interviewed know of some method of contraception. About 40 percent say that they have used a method sometime, while 20 percent are currently using a method. Almost half of the husbands who are currently using a method say they are using periodic abstinence; about 20 percent are using the pill. Despite the low level of contraceptive use, the data indicate that there is potential need for family planning. Twenty-three percent of married women want no more children, while 45 percent want to wait at least two years before having the next child. In other words, more than two-thirds of all married women are potentially in need of family planning either to limit or to space births. Both married women and their husbands continue to desire large families although husbands in the sample have considerable higher family size preferences than married women. The mean desired family size among married women is 5.5, whilst that among husbands is 7.6. xxi The GDHS data indicate that out of every 1,000 live births, 77 die before reaching their first birthday and 155 die before attaining age five. While these rates indicate high levels of mortality, the rates for earlier time periods are even higher, suggesting a decline in both infant and childhood mortality during the past fifteen years. Both infant and child mortality are higher in rural areas than in urban areas. Substantial regional differences exist in mortality, with Greater Accra having the lowest infant mortality rate (58) and the Central region the highest rate (138). Perhaps the most striking mortality differentials are those associated with the length of the preceding birth interval. Children born within two years of a preceding birth are more than twice as likely to die during the first year of life as those born four or more years after a preceding birth. The health of both mother and child is likely to be affected by the type of health care received during pregnancy. The GDHS data show that there is a high level of prenatal care by trained health personnel. For 82 percent of births in the five years before the survey, mothers received prenatal care from a trained doctor, nurse or trained midwife. One-third of the births were delivered by a trained nurse or midwife and 28 percent by a traditional birth attendant. Only 6 percent of births were delivered without assistance. Nearly three-quarters of children under 5 years of age have had at least one immunisation, but only 47 percent of children age 12-23 months with health cards are fully immunised. The GDHS data indicate fairly high levels of prevalence of certain childhood diseases. Among children under five, 26 percent had diarrhoea in the two weeks before the survey, and 35 percent had fever in the four weeks before the survey. Anthropometric measurements taken in the Ghana DHS permit an assessment of the nutritional status of children aged 3-36 months. Approximately one-third of children in this age group are classified as chronically malnourished; 8 percent are classified as acutely malnourished. xxii Ghana COTE d'IVOIRE BURKINA FASO UPPER WEST mWA ~ • mBOLGA ~ m TAMALE • NORTHERN SUNYANI 'm BRONG-AHAFO ASHANTI • • 7" = EASTERN KUMASI " • " KOFORID d 4L • WESTERN ,CENTRAL ~ ." " EAST TOGO i I • LARGE URBAN E,A. • URBAN E.A. ' • RURAL E.A. ACCRA SEKOND/ GULF OF GUINEA xxiv CHAPTER 1 BACKGROUND 1.1 HISTORY, GEOGRAPHY AND ECONOMY The Republic of Ghana, covering 238,537 square kilometres, lies along the west coast of Africa. Apart from the Atlantic Ocean that washes its 560 kilometres of coastline on the south, the country is surrounded by French-speaking countries--the Republic of Togo in the east, Burkina Fast in the north, and Cote d'Ivoire in the west. Ghana is divided into ten administrative regions made up of 110 districts which, under the present political structure, constitute the primary units of administration. On 6th March 1957, Ghana attained political independence from Britain and on 1st July 1960, adopted a republican constitution. Since independence, Ghana has made bold efforts at achieving rapid social and economic development. Successive governments which have administered the country have recognised that the country's population is an instrument for, as well as the beneficiary of, development and, consequently, population factors have been incorporated into all socioeconomic development plans. The Ghanalan population is composed of varied ethnic and linguistic groups. In 1960, ~ some 17 major groups were identified based on language (Census Office, 1964). The largest ethnic group, the Akans, constitutes 44 percent of the population and is concentrated in the Ashanti, Brong-Ahafo, Central, Western and Eastern regions. The Ewes, found mainly in the Volta region, are 13 percent, with the Ga- Adangbe concentrated in the Greater Accra and Eastern regions accounting for 8 percent. Finally, in the Northern and Upper regions are the Mole-Dagbani (16.0 percent), Grussi (2 percent) and Gurma (4 percent). Other, smaller ethnic groups make up the remainder. Christians make up over 50 percent of the population and are found mainly in the southern half of the country (Central Bureau of Statistics, 1983). Muslims and adherents of traditional religion .are concentrated in the northern half of the country. Ghana has a mixed economy consisting of a small, capital-intensive modem sector and a large, traditional agricultural sector. The modem sector focuses on mining and industrial activities. The traditional sector is composed of small farmers who make up 61 percent of the economically active adult population (Ghana Statistical Service, 1987). The economy experienced a steady decline throughout the 1970s and the early 1980s, with per capita real income falling substantially during that period. Economic activity stagnated due to a variety of factors including poor incentives for producers, lack of mw materials, and high operating costs. The production and export of commercial crops slumped drastically. Large budgetary deficits and poor fiscal management resulted in high inflation and reduced living standards for large segments of the population. Overvalued currency and a fixed exchange rate contributed to decreasing exports and periodic shortages of foreign exchange. Faced with deteriorating economic conditions, many professionals and skilled technical workers, as well as the semi-skilled and unskilled left the country seeking employment. Three years of severe drought and widespread bush_fires in the early 1980s resulted in acute food shortages in 1983 and the first half of 1984. The expulsion of an estimated one million Ghanaians from Information on ethnic groups has not been collected since the 1960 census. neighbouring Nigeria aggravated an already Ix)or employment and food situation. Extemal terms of trade worsened further with increases in crude oil prices and a decline in the price of the country's major foreign exchange earners; namely, cocoa, timber and gold. The gross national product dropped from 600 dollars per capita in 1974 to less than 200 dollars in 1981; real wages went down by 80 percent and the volume of imports fell by half. A third of export eamings were being spent on crude oil imports by 1981 with inflation at 117 percent. By 1983 the rate of inflation had reached an all-time high of 123 percent. In an effort to halt the economic recession, the government of the Provisional National Defence Council (PNDC) launched an Economic Recovery Programme (ERP) in April 1983. Phase one of the recovery programme (1983-86) was aimed at stabilising and consolidating the economy. It succeeded in restoring growth of per capita income over three consecutive years for the first time in over a decade. The first phase also achieved a substantial reduction in the macro-economic disequilibria. The second phase of the ERP, covering the period 1987-89, has a view to stimulating growth, encouraging savings and investment, and consolidating economic gains through a programme of structural adjustment and development. To complement the efforts under the Structural Adjustment Programme (SAP), the govemment has initiated a "Programme of Actions to Mitigate the Social Costs of Adjustment (PAMSCAD)." The primary objective under the PAMSCAD is to identify groups whose living conditions have been adversely affected by the structural adjustment programme and earlier periods of economic decline, and to address the needs of these groups. The implementation of the ERP has, so far, appreciably revived the economy with the Gross Domestic Product (GDP) growing at an annual average rate of 6 percent over the period 1984-88. Agricultural output between 1987 and 1988 increased from a growth rate of 0.04 percent in 1987 to more than 3 percent in 1988. Meanwhile, recovery and expansion in the industrial sector has continued with output growing at a rate of 10.3 percent in 1988 (Republic of Ghana, 1989). Domestic savings and investment have increased in both the public and private sectors, whilst the rate of inflation dropped from 40 percent hi 1987 to 31 percent in 1988 (Ghana Statistical Service, 1989). 1.2 EDUCATION The accelerated programme of education initiated during the immediate post-independence years has resulted in greater literacy. In particular, the proportion of females who have had some formal education has risen appreciably. For example, the proportion of women 15-24 who have never been to school declined from 79 percent in 1960 to 38 percent in 1984. While in 1960 only one out of twenty females 15-24 was in school, by 1984 one out of five in that age group was in school. 1.3 POPULATION AND FAMILY PLANNING PROGRAMME According to the results of the latest population census, Ghana's population as of March 1984 was 12.3 million, which represents an increase of 44 percent over the figure recorded in the 1970 census (Census Office, 1975; Ghana Statistical Service, 1989). Over two-thirds of the population live in rural areas, while nearly one-third live in urban areas (defined as localities with 5,000 or more persons). The vital registration system covers only about 40 percent of births and 25 percent of deaths, with the majority of the events being registered in urban areas where most registries arc located. In the absence of reliable vital statistics, censuses and demographic sample surveys have provided the baseline data for estimation of vital rates. The cmde birth rate is currently estimated at 44 to 48 per thousand, while the crude death rate is 16 to 17 per thousand. An intercensal growth rate of 2.6 percent per annum was recorded for the period 1970-1984. Declining mortality and comparatively high fertility have stood out as the principal factors influencing rapid population growth in the 1970s and 1980s. Traditionally, large families arc favoured. The ethnographic literature contains numerous accounts of young couples being urged to have as many 2 children as they can. The fate of a childless woman is reported to be a miserable one (Fortes, M. 1949; Gaisie, S.K. 1981; Goody J. 1956). In 1969, Ghana became the first sub-Saharan African country to adopt a population policy. A major objective of the policy is to reduce population growth to a rate of 2.0 percent per annum by the year 2000. Almost at the same time, the Ghana National Family Planning Programme (GNFPP) was instituted to offer family planning services to couples desiring to limit or space births. Outlets for the supply of family planning services were opened in hospitals and clinics, most of which were government operated. Currently, the Ministry of Health (MOH) distributes information and supplies through its family health clinics and primary health care centers. Since its inception in 1970, the GNFPP has focused on delivery of family planning services to such groups as girls under 18 years who are pregnant, women with children under two years, families with histories of poor child survival and development, and women aged 30-35 with four or more children. As recently as 1986, under a contrax;eptive social marketing programme begun that year, pharmacies and chemical sellers were permitted to sell condoms, vaginal foaming tablets and oral contraceptives through some 3600 retail outlets (Ampofo, 1988). Non-governmental agencies like the Planned Parenthood Association of Ghana (PPAG) and the Christian Council of Ghana (CCG) operate family planning clinics as a supplement to the efforts of the Ministry of Health. In addition, the Catholic Secretariat encourages use of the rhythm or ovulation method and private medical practitioners offer family planning services. 1.4 HEALTH PRIORITIES AND PROGRAMMES The Government is committed to the objective of attaining health for all by the year 2000. The strategy for achieving this objective is the Primary Health Care (PHC) programme which constitutes a vital component of the country's health delivery system. To ensure that the PHC programme operates efficiently, the health service is being decentralised to the districts with reorientation and retraining of personnel in PHC programme planning, implementation, and management. Furthermore, the efficiency of traditional birth attendants (I'BAs) is being improved through special training in modem midwifery practices as well as prenatal and postnatal care, health education, oral rehydration therapy, family planning, and aspects of child health. The PHC itself concentrates on six priority areas as follows: Maternal and Child Care Family Planning Nutrition Control of Diarrhoeal Diseases Expanded Programme on Immunisation and Malaria Control The implementation of all these aspects of the PHC programme requires multlsectoral action and close collaboration among the three main health systems, i.e., the Government Health Service, private practitioners, and traditional medicine. Maternal and child health (MCH), including the programme to combat the six major childhood diseases, has been incorporated in general medical care since the early sixties. The diseases included in a nationwide immunisation programme are diphtheria, tetanus, whooping cough, poliomyelitis, measles and tuberculosis. By adopting the WHO Expanded Programme on Immunisation (EPI) the government hopes to reduce the massive loss of children to these diseases. Under the EPI programme the country hopes to attain 80 percent immunisation coverage by the year 2000 for the target population--children under two years, pregnant mothers and puberty-aged girls (Adjei, et. al., 1988). 3 1.5 OBJECTIVES OF THE SURVEY The short-term objectives of the Ghana Demographic and Health Survey (GDHS) are to provide policymakers and those implementing policy with current data on fertility levels, knowledge and use of contraception, reproductive intentions of women 15-49, and health indicators. The information will also serve as the basis for monitoring and evaluating programmes initiated by the government such as the extended programme on immunisation, child nutrition, and the family planning programme. The long- term objectives are to enhance the country's ability to undertake surveys of excellent technical quality that seek to measure changes in fertility levels, health status (particularly of children), and the extent of contraceptive knowledge and use. Finally, the results of the survey will form part of an intemational data base for researchers investigating topics related to the above issues. 1.6 ORGANISATION OF THE SURVEY The Ghana Demographic and Health Survey is a stratified, self-weighting, nationally representative sample survey. The GDHS forms part of a worldwide programme to elicit data on fertility, contraceptive use, infant mortality and morbidity, and health-related issues for planning purposes. The survey which was funded mainly by the United States Agency for Intemational Development (USAID) and the Ghana Government, was carried out by the Ghana Statistical Service between February and June 1988. Contributions by the Ghana Government covered, among other things, the salaries of the survey personnel, the provision of offices and office equipment, as well as some of the vehicles used for the project. Funds from USAID were administered by the Institute for Resource Development/Macro Systems, Inc. (IRD), and were used for allowances of project personnel, data processing and anthropometric equipment, printing of questionnaires, publication of reports, and vehicle maintenance and fuel. Technical assistance to the survey was provided by IRD. The United Nations Fund for Population Activities (UNFPA) provided funds which were used for the training of interviewers and for fieldwork. Finally, UNICEF loaned 8 vehicles to the project for the duration of the fieldwork. The 150 clusters from which a representative sample of women aged 15-49 was selected form a subsample of the 200 clusters used for the Ghana Living Standards Survey (GLSS). All census Enumeration Areas (EAs) were first stratified by ecological zones into 3 strata, namely Coastal Savanna, Forest, and Northern Savanna. These were further stratified into urban, semi-urban, and rural EAs. The EAs (in some cases, segments of EAs) were then selected with probability proportional to the number of households. All households in the selected EAs were subsequently listed. (For details of the sample design, see Appendix B). Three different types of questionnaires were used for the GDHS. These were the household, individual and the husband questionnaires. The household and the individual questionnaires were adapted from the Model "B" Questionnaire for the DHS programme. The GDHS is one of the few surveys in which special effort was made to collect information from husbands of interviewed women on such topics as fertility preferences, knowledge and use of contraception, and environmental and health- related issues. All usual members and visitors in the selected households were listed on the household questionnaire. Recorded in the household questionnaire were data on the age and sex of all listed persons in addition to information on fostering for children aged 0-14. Eligible women and eligible husbands were also identified in the household questionnaire. The individual questionnaire was used to collect data on eligible women. Eligible women were def'med as those aged 15-49 years who spent the night prior to the household interview in the selected household, irrespective of whether they were usual members of the household or not. Items of information collected in this questionnaire are as follows: . 2. 3. 4. 5. 6. 7. 8. Respondent's Background Reproductive Behaviour Knowledge and Use of Contraception Health and Breastfeeding Marriage Fertility Preferences Husband's Background and Women's Work Weight and Height of Children Aged 3-36 Months. In half of the selected clusters a husband's questionnaire was used to collect data on eligible husbands. Eligible husbands were defined as those who were co-resident with their wives and whose wives had been successfully interviewed. Data on the husband's background, contraceptive knowledge and use, as well as fertility preferences were collected. All three questionnaires were translated into seven local languages, namely, Twi, Fante, Nzema, Ga, Ewe, Hausa and Dagbani. All the GDHS interviewers were able to conduct interviews in English and at least one local language. The questionnaires were pretested from mid-October to early November 1987. Five teams were used for the pretest fieldwork. These included 19 persons who were trained for 11 days. For the main survey, eight days' training was organised for the 16 supervisors and editors who had earlier taken part in the pretest. This was followed by a 3-week training for interviewers. Personnel involved in the fieldwork were 40 interviewers (26 males, 14 females), 8 supervisors (6 males, 2 females), 8 editors (7 males and 1 female) and 11 drivers. 2 Fifty-six of the field staff were recruited from the Ghana Statistical Service, whilst 11 persons were reeroited from the Department of Community Development and the Department of Social Welfare. Fieldwork began on 13th February 1988 and was completed on 5th June 1988. Completed questionnaires were collected weekly from the regions by the field coordinators. Coding, data entry and machine editing went on concurrently at the Ghana Statistical Service in Accra as the fieldwork progressed. Coding and data entry were started in March 1988 and were completed by the end of June 1988. Preliminary tabulations were produced by mid-July 1988, and by August 1988 preliminary results of the survey were published. 1.7 BACKGROUND CHARACTERISTICS OF SURVEY RESPONDENTS Of the 4966 households selected, 4406 were successfully interviewed. Excluding 9 percent of households that were vacant, absent, etc., the household response rate is 98 percent. Out of 4574 eligible women in the household schedule, 4488 were interviewed successfully. The response rate at the individual level is 98 percent. Of the 997 eligible husbands, 943 were successfully interviewed, representing a response rate of 95 percent. Age data in Ghana, as in many developing countries, are constrained by errors which result from omission of events, age misstatement, and preference for or avoidance of certain digits. As shown in Although the use of male interviewers is unusual in surveys of this type, a pilot survey for the Ghana Fertility Survey conducted in 1975 suggested that Ghanaian women have no preference for female over male interviewers (Central Bureau of Statistics, 1983). Figure 1.1, the age distribution of the successfully interviewed women in the sample by single years of age, reveals disproportionately large numbers of women reporting ages with terminal digits of 0 and 5 and somewhat less popular digits 2 and 8. Comparatively, ages ending in digits 1, 7 and 9 are avoided. Percent 6 5 4 3 2 1 0 15 Figure 1.1 Distribution of Women 15-49 by Age 20 25 30 35 40 45 50 Age Ghana DHS 1988 In the individual interview, each eligible woman was asked her date of birth and age. Where the respondent could not provide her date of birth nor age, various procedures were adopted to obtain an estimate of her age. The estimation techniques employed included the use of a historical calendar of national, regional and local events and the use of demographic information, such as the number of children ever born. To limit the effects of age heaping on the preferred digits, the conventional five-year age groupings are used in the analyses of the results of the GDHS. Distribution of the sample population by age, residence, region and level of education is shown in Table 1 A, and the distribution of women of reproductive age enumerated in the 1984 Census is presented in Table 1.2. The results in Table 1.1 indicate that the 15-19 age group comprises 19 percent of the sample, the same proportion as in each of the next two higher age groups. The proportion of the sample in the older age groups declines steadily from 14 percent in the 30-34 group to 8 percent for the 45-49 age group. One-third of the respondents live in urban areas with the remaining two-thirds residing in rural areas. A similar urban-rural distribution was found in the 1984 national population census. Some 18 percent of the sample population live in the Ashanti region with 16 percent and 13 percent residing in Eastern and Greater Accra regions, respectively. One out of every nine women in the sample lives in Bmng Ahafo, Volta or the three northemmost regions. In addition, while one out of every 10 women lives in the Central region, one out of every 12 resides in the Western region. Two-fifths of the women in the sample have never been to school while 16 percent have only primary education. More than a third (37 percent) have middle school education; only 8 percent have gone beyond middle school. An overview of the sample population according to level of education is provided in Table 1.3. Overall, three out of every five women in the sample have had at least primary school education, with the majority of these attaining middle school education. Educational differences between age groups reflect improvements in educational attainment in recent years. The proportion who have never been to school declines as younger age cohorts are considered. Nearly three-quarters (74 percen0 of women aged 45-49 have never been to school. In contrast, among women aged 15-19, less than a fifth have no education. In fact, more than half (53 percent) of them are either now at the middle-school level or have completed middle school. There is, therefore, an inverse relationship between the current age of women and educational attainment. As expected, urban respondents are better educated than their rural counterparts; 58 percent of the former have at least some middle school education compared to 37 percent for the latter. Regional differences in educational composition of women are also shown in Table 1.3. The more urban and modemised a region, the lower the proportion of women who have never been to school. For example, the proportion of women with no education is only 22 percent in Greater Accra compared to 82 percent in the combined Northern and Upper regions. Between those extremes lie the other regions; Central, Western and Brong Ahafo have proportions ranging from 43 to 48 percent; Eastern, Ashanti and Volta regions range from 28 to 38 percent. Table i.i Percentage Distr ibut ion of Women by Age, Urban- Rural Residence, Region and Level of Education, GDHS, 1988 Background Character ist ic Percent Number AGE 15-19 18.9 849 20-24 19.3 867 25-29 19.3 867 30-34 14.3 644 35-39 11.8 531 40-44 8.1 364 45-49 8.2 366 RESIDENCE Urban 33.9 1523 Rural 68.1 2965 REGION Western 8.7 392 Central 10.3 464 Greater Accra 13.3 598 Eastern 15.7 703 Volta II.i 5G0 Ashanti 18.3 823 Brong Ahafo ii.i 500 Upper West, East 11.3 508 and Northern LEVEL OF EDUCATION No educat ion 39.7 1783 Pr imary 16.3 731 Middle 36.5 1638 Higher 7.5 336 TOTAL 100.0 4488 Table 1.2 Percentage Distr ibut ion of Women of Reproduct ive Age, 1984 Census and 1988 GDHS Age 1984 Census 1988 GDHS 15-19 21.4 18.9 20-24 20.1 19.3 25-29 18.0 19.3 30-34 13.7 14.3 35-39 10.6 11.8 40-44 8.7 8.1 45-49 7.4 6.2 TOTAL i00.0 I00.0 Table 1.3 Percentage Distr ibut ion of Women by Level of Education, According to Age, Urban-Rural Residence, and Region, GDHS, 1988 Level of Educat ion No Background Educe- Character is t ic t ion Pr imary Middle Higher Total Number AGE 15-19 19.2 20.8 52.8 7.3 i00.0 849 20-24 30.9 15.0 44.5 9.6 100.0 867 25-29 36.1 14.8 38.3 10.8 i00.0 867 30-34 39.9 14.9 37.9 7.3 100.0 644 35-39 51.4 17.7 25.6 5.3 i00.0 531 40-44 66.2 13.7 15.4 4.7 I00.0 364 45-49 73.5 15.3 9.8 1.4 100.0 366 RESIDENCE Urban 26.7 15.0 43.7 14.6 I00.0 1523 Rural 46.4 17.0 32.8 3.8 100.0 2965 REGION Western 43.6 17.1 34.7 4.6 100.0 392 Central 47.6 15.5 31.7 5.2 100.0 464 Greater Accra 21.9 17.4 41.8 18.9 100.0 598 Eastern 27.6 21.1 44.7 6.7 100o0 703 Volta 38.0 19.8 35.2 7.0 I00.0 500 Ashant i 30.0 15.9 44.7 9.4 100.0 823 Brong Ahafo 42.8 13.2 41.2 2.8 100.0 500 Upper West, East 61.7 6.7 8.1" 1.6 100.0 508 and Northern TOTAL 59.7 16.3 36.5 7.5 I00.0 4488 CHAPTER 2 NUPTIALITY AND EXPOSURE TO THE RISK OF PREGNANCY Variations exist in the nuptiality pattem in Ghana among the ethnic groups. At one end of the spectrum, violation of the premarital sex injunction could result in expulsion from the village as was the practice among the traditional Adangbe. However, at the other end, strong emphasis on procreation as the goal of marriage demanded that some evidence of fecundity be exhibited before the marriage contract was concluded. Generally, early marriage was encouraged by all ethnic groups, but as the cost of marriage varied widely from one group to another, differences exist in the age at marriage depending, to some extent, on the ease with which the material prerequisites of marriage could be acquired (Aryee, 1985). In urban areas, however, new forms of sexual cohabitation, such as consensual unions, have emerged, and customary practices have been modified. For example, in mate selection, the role of kinsmen has been weakened by the modernising influences of the urban environment. These varied features of the institution of marriage in Ghana def'me the framework in which women can be exposed to the risk of pregnancy. The features of marriage also give indications of the regulatory mechanisms that operate to influence the level of fertility in Ghana. For the purpose of the Ghana DHS, marriage is def'med to include both formal unions and consensual (living together) arrangements. Table 2.1 Percentage Distribution of Women by Current Marital Status, According to Age, GDHS, 1988 Marital Status Not Never Living Living Age Married Married Together Widowed Divorced Together Missing Total Number AGE 15-19 75.6 17.0 3.5 0.i 1.3 2.4 0.1 i00.0 849 20-24 22.6 62.3 6.1 0.1 5.2 3.7 0.0 180.0 867 25-29 4.5 80.9 5.9 0.8 5.8 2.2 0.0 100.0 867 30-34 1.2 81.1 7.3 1.6 6.1 2.8 O.0 100.0 644 35-39 0.6 84.9 4.1 1.5 6,0 2.8 0.0 I00.0 531 40-44 0.3 78.3 6.9 4.4 7,1 3.0 0.0 i00.0 364 45-49 0.0 72.4 5.5 7.1 12,8 2.2 0.0 100.0 366 TOTAL 19.8 64.8 5.5 1.5 5,6 2.7 O.0 100.0 4488 2.1 CURRENT MARITAL STATUS Table 2.1 presents the distribution of women in the sample according to their current marital status. Overall, 20 percent of the respondents have never been married, while nearly two-thirds (65 percent) are lawfully married. Six percent of the respondents reported themselves as living in an informal union. Almost 3 percent had previously lived in an informal union but are now living separately, and the remaining 7 percent are either widowed or divorced. 9 Nearly all women in Ghana enter into some type of union during their reproductive years, since the percentage reporting themselves as "never married" falls off from 76 percent of women aged 15-19 years to less than one-half of one percent of women aged 40-44. In fact, by age 25-29, fewer than 5 percent of women have never been married (see Figure 2.1). The proportion currently in some type of union ranges from 21 percent among women 15-19 to 89 percent of women aged 35-39. Percent 100 75 50 25 0 Figure 2.1 Union Status of Women 15-49 by Age Group 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age Group Living together ~ Wldowed/dlv./sep. Ghana DHS 1988 An interesting observation is that while the women aged 15-19 years are least likely to be in an informal union, young adult women aged 20-34 arc more likely than any other age group to contract such a consensual relationship. In addition, the proportion divorced or widowed increases with age, reaching 20 percent of women aged 45-49. 2.2 POLYGYNY Among all currently married women in the sample, a third (33 percent) are in polygynous unions (Table 2.2). This proportion is slightly lower than the 35 percent found in the Ghana Fertility Survey of 1979-80. The differcnce is so marginal that one cannot tell whether the difference is a reflection of a real decline in polygyny or chance fluctuation. Younger women are less likely to be in polygynous unions than older women. As the data in Table 2.2 show, the proportion of currently married women in polygynous unions decreases from 43 percent for women aged 35-39 years to 16 percent for those aged 15-19 years, suggesting that the practice of polygyny is declining. Modemising influences in the urban setting seem to discourage the formation of polygynous unions. Some 34 percent of currently married women in the rural area arc in polygynous unions as compared to 28 percent in the urban area. 10 Table 2.2 Percentage of Current ly Marr ied Women in a Polygynous Union, by Age and Selected Background Characterist ics, GDBS, 1988 Age Background Character ist ic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total RESIDENCE Urban 18.8 20.5 20.5 26.6 41.1 36.8 39.3 28.3 Rural 15.1 27.4 31.1 37.8 43.8 44.6 59.9 34.5 REGION Western 29.5 16.9 31.6 29.8 I0.0 32.0 24.7 Central 15.2 26.1 23.0 42.6 40.0 51.9 28.6 Greater Accra 19.7 17.2 29.3 44.3 37.2 27.8 27.5 Eastern 10.0 20.7 22.0 28.0 35.2 35.1 34.5 26.6 Volta 37.9 39.6 44.3 59.2 51.4 47.1 43.8 Ashanti 5.6 18.0 29.6 32.1 25.9 46.3 40.0 28.1 Brong Ahafo 16.8 27.7 23.7 36.0 46.3 45.2 40.9 32.2 Upper West, East 36.0 38.3 46.3 48.1 62.3 60.5 43.9 48.3 and Northern LEVEL OF EDUCATION No education 25.9 36.5 33.6 40.0 45.0 46.4 40.4 39.5 Primary 13.5 16.3 26.6 33.7 40.7 35.0 36.2 28.3 Middle 10.8 19.5 25.4 31.3 42.7 33.3 40.7 27.1 Higher 21.9 17.9 17.5 23.8 19.1 TOTAL i6.1 25.5 28.1 34.3 42.7 41.9 39.3 32.6 - Fewer than 20 cases The practice of polygyny in Ghana shows considerable regional variation. The highest proportions of currently married women in such unions are in the three northernmost regions and the Volta region. The proportions range from 44 percent for Volta to 48 percent for the northern regions. The Western region has the lowest proportion (25 percent). The remaining regions have proportions ranging from 27-32 percent. The data show an inverse relationship between education and polygyny, such that the higher the level of education the lower the extent of polygyny. The practice of polygyny ranges from 19 percent for women with 11 or more years of education to 40 percent for women with no education. 2.3 AGE AT FIRST UNION In this section, we analyse information on the respondent's date of entry into first union. The survey collected information on the month and year women started living with their first husband or partner. Those who could not recall the year were asked how old they were at the time of the first marriage. The quality of data from these questions depends on how accurately the respondents place the event in tune. In addition to the difficulty in correct dating of events, the formalisation of marriage itself may span a number of years. Under these circumstances, caution must be exercised in interpreting the data. In the GDHS, 29 percent of ever-married women reported both a month and year of first marriage, 55 percent gave the year only and 14 percent reported their age at the time of first marriage. Less than 2 percent of respondents had the dates of their first marriage completely imputed. 11 Table 2.3 Percentage Distr ibut ion of Women by Age at First Union and Median Age at First Union, Accord ing to Current Age, GDHS, 1988 Age at First Union Current Never Median Age Marr ied <15 15-17 18-19 20-21 22-24 25+ Total Number Age 15-19 75.6 5.8 16.0 2.6 0.0 0.0 0.0 i00.0 849 20-24 22.6 8.7 32.3 22.4 ii.0 3.1 0.0 i00.0 867 18.7 25-29 4.5 12.3 30.8 23.2 14.8 11.6 2.8 100.0 867 18.5 30-34 1.2 14.0 34.3 23.3 12.6 9.9 4.7 100.0 644 18.1 35-39 0.6 10.5 38.0 23.7 13.0 7.9 6.2 100.0 531 18.1 40-44 0.3 14.8 40.4 21~4 11.5 6.0 5.5 100.0 364 17.6 45-49 0.0 13.4 39.1 20.5 9.8 7.9 9.3 i00.0 366 17.8 TOTAL 19.8 10.7 31.1 18.9 10.0 6.4 3.1 100.0 4488 - Omit ted due to censor ing Table 2.3 reveals that the median age at first union is about 18 years for women aged 20-49. In the GFS of 1979-80 the median age at first union for women aged 20-49 was also found to be 18 years (Central Bureau of Statistics, 1983). Nevertheless, the median age at marriage increases across age cohorts, from 17.7 for women currently aged 40-49 to 18.7 for women aged 20-24. Even though entry into union before age 15 is not rare, the practice is now decreasing. Fifteen percent of women in their forties entered a union before age 15, but only 6 percent of girls currently aged 15-19 did so. Furthermore, 31 percent of women aged 15-19 years in the GFS sample had been married sometime, compared to the 25 percent in the GDHS sample. Evidence from the latter survey also indicates that, while nearly two-thirds of the cohort of women 20-29 entered a union before age 20, for the two oldest cohorts, 72 to 77 percent were in a union before age 20. Table 2.4 compares the median age at first union for women 20-49 according to various background characteristics. Women aged 20-49 years living in urban areas are likely to enter a union about one-half year later than their counterparts in rural areas. The latter, on the average, contract a first union at age 18. Median age at first union varies slightly from one region to another. It ranges from a low of 17.8 years for Ashanti to a high of 19.0 years for Greater Accra. Women with higher education marry later, on the average, than less educated women. Median age at marriage rises rapidly from 17.8 years for women with no education to 22.3 years for women with eleven or more years of schooling. Thus, the expected inverse relationship between education of women and age at marriage is exhibited in the results of the survey. 2.4 BREASTFEEDING AND POSTPARTUM INSUSCEPTIBILITY In addition to age at first union, the GDHS collected information on breastfeeding and postpartum insusceptibility, which is presented in Tables 2.5 and 2.6. Susceptibility to pregnancy after a birth can be delayed by breastfeeding, which inhibits the resumption of ovulation and menstruation, and by practising postpartum sexual abstinence. 12 Table 2.4 Median Age at First Union Among Women Aged 20-49 Years, by Current Age and Selected Background Characterist ics, GDHS, 1988 Current Age Women Background Age Characteristic 20-24 25-29 30-34 35-39 40-44 45-49 20-49 RESIDENCE Urban 19.6 19.6 18.4 18.2 18.1 18.2 18.7 Rural 18.4 18.2 18.0 18.1 17.3 17.6 18.1 REGION Western 18.7 18.5 17.5 18.4 17.0 17.5 18.2 Central 18.6 18.8 17.9 19.0 17.7 18.2 18.4 Greater Accra 20.0 19.6 19.0 18.7 18.4 17.8 19.0 Eastern 19.7 18.9 18.5 18.8 17.9 17.2 18.7 Volta 19.0 18.8 19.2 18.0 18.7 19.2 18.8 Ashanti 18.5 18.5 17.6 17.3 16.4 17.2 17.8 Brong Ahafo 18.0 18.1 17.6 17.9 18.2 18.8 18.0 Upper West, East 18.3 17.7 18.0 18.2 17.7 17.4 17.9 and Northern LEVEL OF EDUCATION No education 18.1 17.8 17.7 17.7 17.4 17.7 17.8 Primary 18.1 17.9 17.2 18.1 16.7 17.4 17.7 Middle 19.0 18.8 18.2 18.6 18.0 18.7 18.7 Higher 21.3 22.4 20.7 22.3 TOTAL 18.7 18.5 18.1 18.1 17.6 17.8 18.3 * Less than 50 percent ever married - Fewer than 28 cases Respondents who gave birth in the five years preceding the survey were asked if they breastfed, the duration of breastfeeding, and the reason they stopped breastfeeding. Women were further asked how many months they were amenorrhoeic after each delivery and how long they abstained from sexual intercourse. In addition, the women were asked if they were currently breastfeeding, amenorrhoeic, and/or practicing abstinence. Since it may be difficult for respondents to recall the duration of these events, and since it may be difficult to precisely define when weaning takes place, data in Tables 2.5 and 2.6 are current status estimates, which refer to whether or not the woman was breastfeeding and/or amenorrhoeic at the time of the survey interview, rather than her reported durations following births in the last five years. All births occurring during the three years before the survey are considered in Table 2.5. It must be noted that Table 2.5 uses cross-sectional data, representing all women at a single point in time, rather than showing the experience of an actual cohort over time. For this reason, the proportions breastfeeding and amenorrhoeic at increasing durations do not decline in a steady fashion. For example, more mothers 10-11 months postpartum were breastfeeding at the time of the survey than were mothers of children 8-9 months old. To reduce such fluctuations, the births are grouped in 2-month intervals. The data in Table 2.5 reveal that the duration of breastfceding is fairly long. Nine out of 10 women with births 2-3 months before the survey were still breastfeeding and, for the period 20-21 months 13 after delivery, more than half (57 percent) of the women still breastfed. Indeed, 2 out of 5 women who gave birth 22-23 months before the interview were still breastfeeding. Table 2.5 Percentage of Births Where Mothers are Stil l Breast- feeding, Postpartum Amenorrhoeic, Abstaining, and Insusceptible, by Number of Months Since Birth, GDHS, 1988 Percentage of Births Where Mothers Are: Months Number Since B feast- /~menor- Abstain- Insuscep- of Birth feeding rhoeic ing t ible Births* Less than 2 92.7 95.1 96.7 96.7 123 2-3 89.9 88.5 92.1 94.2 139 4-5 92.4 77.9 73.3 88.5 131 6-7 91.9 71.6 56.4 80.5 149 8-9 91.6 69.9 48.3 79.0 143 18-11 94.4 67.1 43.4 78.3 143 12-13 87.2 59.6 41.0 70.2 188 14-15 78.2 47.4 36.1 60.2 133 16-17 64.3 33.6 31.5 46.9 143 18-19 68.1 32.6 26.1 40.6 138 20-21 56.6 20.8 24.5 35.8 106 22-23 41.2 13.1 23.5 29.4 153 24-25 19.2 8.8 16.9 17.6 182 26-27 18.0 7.2 15.1 18.0 139 28-29 11.7 5.0 9.2 13.3 120 30-31 9.2 3.8 11.5 12.3 130 32-33 7.1 3.6 10.1 11.2 169 34-35 5.9 2.9 6.6 7.4 136 Total 56.4 39.1 36.2 48.6 2565 * Includes births occurr ing in the per iod 0-35 months pr ior to the survey° Postpartum protection from conception can be prolonged by breastfeeding which can lengthen the duration of amenorrhoea. The protection from conception offered by breastfeeding is affected, however, by the frequency and intensity with which the child is breastfed. As many as 88 percent of women 2-3 months postpartum were amenorrhoeic. The proportion drops quite slowly to only 67 percent for women 10-11 months postpartum. Sexual abstinence after childbirth is practiced for a comparatively long time among women in Ghana. Less than 10 percent of women resumed intercourse 2-3 months after birth. Indeed, 8-9 months after birth slightly less than half of the women continued to abstain from sex. Only three-quarters of women resumed intercourse 22-23 months after delivery. Column four in Table 2.5 shows the proportion of women protected from pregnancy due to either amenorrhoea or abstinence. While 97 percent of women who delivered less than 2 months prior to the survey are insusceptible, by a year after delivery, this has dropped to 70 percent, and by two years after delivery less than 20 percent are still insusceptible. 14 2.5 MEAN DURATION OF BREASTFEEDING AND POSTPARTUM INSUSCEPTIBILITY Breastfeeding practices in a society are important to both health workers and demographers. To the former, breast milk is a superior source of good nutrition for a child. A decline in the lactation period could have serious implications for the nutritional status of infants. The demographer's interest stems from the suppressing effect which lactation exerts on the resumption of ovulation after childbirth. In analysing data on breastfeeding, one must be aware of two factors that can affect duration. The first relates to discontinuation of breastfeeding as a result of the child dying, while the second pertains to the intervention of the next pregnancy which, in turn, triggers hormonal action, leading to the cessation of the flow of breast milk. Table 2.6 presents the mean number of months of breastfeeding, postpartum amenorrhoea, postpartum abstinence, and postpartum insusceptibility by background characteristics of the mother. The mean durations were calculated by dividing the total number of women breastfeeding, amenorrhoeic or abstaining by the average number of births per month in the past 36 months. This technique is based on an epidemiological method of estimating the mean duration of a disease calculated by dividing its prevalence by its incidence. Table 2.6 Mean Number of Months of Breastfeeding, Postpartum Amenorrhoea, Postpartum Abstinence, and Postpartum Insusceptibility, by Selected Background Characteristics GDHS, 1988 Number Background Breast- Amenor- Absti- Insuscep- of Characteristic feeding rhoea hence t ibi l l ty Births AGE <30 20.6 13.8 13.5 17.9 1490 30+ 20.0 14.3 13.6 16.4 1098 RESIDENCE Urban 17.5 11.4 12.2 15.8 713 Rural 21.4 15.0 14.1 18.9 1875 REGION Western 20.0 12.6 10.2 15.6 229 Central 18.9 12.8 12.4 17.2 285 Greater Accra 14.7 9.3 8.8 12.7 249 Eastern 20.2 13.2 13.6 17.8 382 Volta 20.8 14.9 15.6 19.0 314 Ashanti 19.1 14.2 9.9 16.9 457 Brong Ahafo 21.4 13.7 10.7 16.2 351 Upper West, East 26.6 20.2 26.9 27.9 321 and Northern LEVEL OF EDUCATION No educat ion 22.8 16.4 17.1 21.3 1114 Primary 20.1 13.9 13.2 18.3 420 Middle 18.1 12.0 I0.0 14.9 922 Higher 16.1 9.0 9.3 12.0 132 TOTAL 20.4 14.0 13.5 18.1 2588 Note: Estimates are based on births 1-36 months before the survey. 15 Figure 2.2 Duration of Breastfeeding, Amenorrhoea and Postpartum Abstinence AGE *30 30* RESIDENCE Urban Rural EDUCATION 22 24 None "~''~""'~'~:;~;~:~'~-~-' '""~ . ""~"""1/1111"~ ' Primary ``~`~`~:~`~`~`~``~``~```~`~"~`~`~``~``~``~``~`~`~ ' . . . . . . . . . . . ' M idd le . """~"~:~::~:"~'~"~'~"'~"'~""~':"; Higher ~'~'~i~:~iiiii . illiii 0 2 4 6 8 10 12 14 18 18 20 Mean Durat ion in Months I ~Breastfeedlng ~Amenorrhoea ~ Abstinence Ghana DHS 1988 On the average, women in Ghana breastfeed for 20 months. More educated women and urban women breastfeed for shorter durations than less educated and rural women. While urban women breasffeed for an average of 17.5 months, rural women breastfeed almost 4 months longer. Uneducated mothers breastfeed 7 months longer than the most educated mothers (Figure 2.2). One plausible reason for the shorter duration of breastfeeding among urban women may be their greater participation in the modem sector of the economy. Evidence from the survey (not shown) indicates that, on the average, mothers fully breastfeed their children (i.e., give them breast milk and water only) for a period of five months. Mothers give supplementary food to their children during the remaining period of breastfeeding. In the Ghana Fertility Survey, the mean duration of full breastfeeding for mothers whose penultimate child survived for at least 12 months was found to be 5.2 months, while the mean length of the entire duration of breastfeeding was given as 18 months. Prolonged breastfeeding is, therefore, as common now as it was a decade ago. As part of an international child survival campaign, the health authorities in Ghana are educating and encouraging expectant and lactating mothers to feed their children at the breast instead of giving them formula food, since breast milk is nutritionally ideal, hygienic and, importantly, provides some immunity against disease during the first months of life. Studies in some developing countries suggest that mortality rates are higher for artificially fed infants than for breastfed infants (Knodel, J., 1982). It has been noted that, in the absence of breastfeeding, postpartum amenorrhoea generally lasts only for about two months, while it averages one to two years when breastfeeding is prolonged and intensive (Page et al, 1982). The mean duration of amenorrhoea in the GDHS is 14 months. The relationship between the duration of breastfeeding and menstruation is not consistent among different subgroups of the population. For example, apart from the three combined 3 regions of the north which 16 have the longest mean duration of breastfeeding (27 months) and the longest period of amenorrhoea (20 months) and the Greater Accra region which has the shortest mean duration of breasffceding (15 months) and the shortest period of amenorrhoea (9 months), the relationship between breasffeeding and menstruation is not consistent for the rest of the regions. For instance, the Ashanti region, which ranks sixth in duration of breastfeeding (19 months) ranks third (14 months) in the return of menstruation. The mean duration of breastfeeding, amenorrhoea and postpartum insusceptibility progressively declines as the level of education rises. Women with no education have the longest mean duration of breasffeeding (23 months), amenorrhoea (16 months), abstinence (17 months) and postpartum insusceptibility (21 months). This declines with increasing education, with the most educated women breasffceding an average of 16 months, amenorrhoeic 9 months, abstaining 9 months, and insusceptible 12 months. Women in the Northem, Upper East, and Upper West regions abstain the longest from sexual intercourse (27 months), while women in the predominantly urban region of Greater Accra abstain the shortest (9 months). The remainder of the regions range from 11 to 16 months with Volta, Eastern and Central regions at the upper end of the range and Brong Ahafo, Ashanti, and Western at the lower end. The mean duration of abstinence for the entire country is 14 months. As regards postpartum insusceptibility, again, the No,hem, Upper West, and Upper East regions record the longest period of insusceptibility (28 months), with the Greater Accra region recording the shortest period (13 months). Between these extremes lie the rest of the regions, with durations ranging 16 to 19 months. The Volta, Eastern, and Central regions occupy the upper portion of the range, with Ashanti and Brong Ahafo regions at the lower end. For the entire country, the mean duration of postpartum insusceptibility is 18 months. 17 CHAPTER 3 FERTILITY The first comprehensive national data obtained on fertility and mortality was collected in the 1979 Ghana Fertility Survey (GFS). The Ghana DHS is the second major national survey conducted in Ghana. One of the objectives oftbe GDHS is to update knowledge on fertility levels and trends. Two types of fertility data were collected in the survey. First, each woman was asked questions about the number of sons and daughters living with her, the number living elsewhere and the number that had died. Second, a complete live birth history was collected from each respondent including the sex and date of birth of each child, its survival status and, if dead, age at death and, if alive, whether the child was living with the mother. In dealing with birth history data, it is important to consider the accuracy of reporting. The omission of births is often evident, especially among older women, as is misstatement of dates of births and deliberate concealment of children not currently surviving. Age misstatement by mothers can also affect the accuracy of the fertility data. These factors are known to influence fertility data collected in any setting, but are thought to be particularly problematic in less developed countries. In this chapter, the discussion of fertility behaviour considers current and cumulative fertility by background characteristics, current pregnancies, fertility trends, children ever born, and age at first birth. 3.1 CURRENT AND CUMULATIVE FERTILITY BY BACKGROUND CHARACTERISTICS Table 3.1 shows the fertility behaviour of women for the period 0-4 years prior to the survey and for the calendar periods 1982-84 and 1985-88, as well as the number of children ever born to women who are in the latter part of the reproductive period (aged 40-49). For the period 0-4 years prior to the survey, the total fertility rate (TFR) for women aged 15-44 is 6.1. This is the number of children that a woman would bear during her reproductive years if she were to experience the age-specific fertility rates observed during the last five years. For the period 1982-84, the "FUR is 6.3, for 1985-88 it is 6.1. The "FUR (15-44) for the five years preceding the GFS (approximately 1975-79) was 6.3. Total fertility rates calculated for the age range 15-49 are also shown. It should be noted that the "FUR which includes women currently aged 45-49 uses data which are progressively truncated as one moves backward in time. The t lq, t for women 15-49 for the five years preceding the GDHS is 6.4; for the two calendar periods, it is 6.6 during the earlier period and 6.4 during the later period. The 'llq< (15- 49) recorded for the five years prior to the GFS was 6.5. A comparison of the total number of children ever bom among women aged 40-49 with total fertility in the last five years suggests that, at current rates, Ghanaian women will have slightly fewer children at the end of their reproductive lives than women currently in the 40-49 age group (see Figure 3.1). There is a large difference in fertility between urban and rural women. The total fertility rate for rural women is more than 1.5 child higher (6.6) than for urban women in the last five years (5.1). Approximately the same urban-rural difference is maintained in the two calendar periods shown. Regional differentials in fertility levels show that, in the last five years, women in Greater Accra experienced the lowest level of fertility at 4.6, followed by Eastern and Ashanti regions which have TFRs under 6.0. In the remaining regions, the TI~R ranges from 6.1 to slightly under 7.0. 19 Table 3.1 Background Character is t ic Total Fert i l i ty Rate (TFR) for Calendar Year Per iods and for F ive Years Preceding the Survey, and Mean Number of Chi ldren Ever Born to Women 40-49 Years of Age, by Selected Background Character ist ics, GDHS, 1988 Total Fert i l i ty Rate* Mean Number of Chi ldren Calendar Calendar 0-4 Years Ever Born Years Years Pr ior to (Women Age 85-89 82-84 the Survey 40-49) RESIDENCE Urban 8.13 5.20 5.05 6.15 Rural 6.63 6.90 6.64 7.31 REGION Western 5.91 7.15 6.10 6.67 Central 6.57 6.59 6.58 7.17 Greater Accra 4.54 5.08 4.64 6.20 Eastern 5.77 6.14 5.72 7.38 Volta 6.82 6.12 6.66 6.62 Ashanti 5.96 5.89 5.90 7.07 Brong Ahafo 7.11 7.02 6.86 7.11 Upper West, East 6.60 7.39 6.80 6.88 and Northern LEVEL OF EDUCATION No educat ion 6.74 7.10 6.77 7.06 Pr imary 6.10 6.27 6.09 7.17 Middle 5.93 5.94 5.87 6.54 Sigher 3.61 3.58 3,55 3.91 TOTAL 15-44 6.11 6.33 6.10 6.92 TOTAL 15-49 6.41 6.60 6.43 * Calcu lated for women 15-44 years of age Differences in fertility levels by education are striking. The total fertility rate in the five years prior to the survey declines from almost 6.8 among women with no education to 3.6 among women with more than middle school education. While women with primary and middle school education have lower fertility than women with no education, the most significant difference is between women with higher education and all other women. The rale for these women is about 40 percent (or about 2 children) less than that of other women. A similar pattern of differentials by education was found in the Ghana Fertility Survey. Results from that survey showed only a small difference in fertility between women with no education and primary education, a small difference for those with middle school education, and a substantial difference (more than 2 births) for women with higher education. It should be noted that the proportion of women with more than middle school education has nearly doubled in the period from the GFS to the GDHS, although they still comprise only a small fraction of all women of reproductive age (8 percent in 1988). Table 3.2 gives another indicator of the level of current fertility--the proportion of women pregnant at the time of the survey. Current pregnancy as a measure for estimating current fertility is highly subject to underreporting, especially when women are in the early months of pregnancy. 20 However, underreperting is likely to be similar among women at all ages; hence, it is not likely to distort variations between ages but can affect the overall estimate of current pregnancies. The table shows that among all women aged 15-49, almost 10 percent reported a current pregnancy. The proportion pregnant is approximately 4 percent among women 15-19, rises to more than 13 percent among women in their twenties, declines slightly to 12 percent among women in their thirties and then drops dramatically among women in their forties. Figure 3.1 Total Fertility Rate 0-4 Years Before Survey and Children Ever Born to Women 40-49 RESIDENCE Urban Rural EDUCATION None Primary Middle Higher O m TFR i k~ CEB 2 4 6 8 10 No. of Children Ghana DHS 1988 3.2 FERTIL ITY TRENDS With the collection of a full birth history in the GDHS, it is possible to examine fertility trends for various periods in the past. Birth history data from surveys often suffer from the omission of births, especially births among older women, births which took place in the distant past, and the births of children who have died. Further, errors in the dates of births of both women and children may affect the accuracy of fertility estimates and distort trends in fertility. ~ One common pattern of birth misplacement is the tendency for women to transfer births during the last five years to a period further in the past (usually the period 5-9 years prior to the survey). This pattern (the "Potter effect") is often accompanied by a tendency for women to place births which occurred ten or more years before the survey into the same ~Seventy-five percent of births in the GDHS had both a year and month of birth recorded, and more than 99 percent had at least a calendar year recorded. There is some evidence that interviewers calculated the date of birth from the age given by the mother in a substantial number of cases, but it is not possible to distinguish these cases precisely from those in which the birth date was provided entirely by the mother. The percentage of births with complete dates reported declines as one moves backward in time--from 86 percent of births in the last 5 years to 64 percent of births occurring 15 or more years before the survey. About 49 percent of respondents provided their own birth dates in the form of both month and ye~ and 97 percent provided at least a calendar year. Again, however, some calendar years recorded in the questionnaires may have been calculated by the interviewers from the ages provided by the respondents. 21 intermediate period. This type of error in the dating of births can give the misleading impression that a decline in fe,i l ity has occurred in the period immediately preceding the survey. Age-specific fertility rates for five-year periods preceding the GDHS are shown in Table 3.3. The rates are progressively truncated as periods further in the past are examined. Neven/aeless, a fairly complete picture of fertility up to age 34 can be gathered for the twenty years preceding the GDHS. Overall, the data suggest that a small decline in fertility has occurred during the course of the last twenty years. In the 15-19 age group, fertility appears to have declined fairly steadily at the rate of about 7-8 percent in each five-year period over the last 25 years. Among women currently in their twenties, smaller declines are recorded, amounting to between 1 and 5 percent in each five-year period, with the exception of the period between 20-24 and 15-19 years prior to the survey, in which fertility appears to have been stable or to have increased slightly. Among women currently in their thirties, slightly larger declines are recorded in the period between 5-9 and 0-4 years prior to the survey. There is some evidence among the group currently aged 30-34 that a transference of births into the period 5-9 years before Table 3.2 Percentage of Al l Women Who are Current ly Pregnant by Ago, GDHS, 1988 Women Who Are Total Age Pregnant Number 15-19 4.1 849 20-24 13.4 867 25-29 13.1 867 30-34 12.3 644 35-39 12.2 531 40-44 6.3 364 45-49 3.8 366 TOTAL 9.9 4488 the survey has occurred. The fact that relatively large decreases in fertility rates are apparent in the most recent 5-year period in comparison to earlier periods may be evidence of either a recent fertility decline or evidence that births have been transferred out of the most recent period into an earlier period, Table 3.3 Age-Per iod Fert i l i ty Rate (Per 1000 Women) by Age of Woman at Birth of Child, GDHS, 1988 Years Preceding Survey Mother 's Age at Birth 0-4 5-9 10-14 15-19 20-24 25-29 30-34 15-19 124 130 141 154 166 179 20-24 258 269 283 299 297 (305) 25-29 278 292 294 301 (323) 30-34 248 271 266 (265) 35-39 195 218 (241) 40-44 117 (145) 45~49 (60) (159) Note: Rates in parentheses are part ia l ly truncated. Further evidence on this point is provided by Figure 3.2, which compares data from the GDHS with data from the GFS. In this figure, age-specific fertility rates are cumulated from age 15 to age 34 for single calendar years. Three-year moving averages are presented. The GDHS data suggest that some births were shifted to the years 1979 and 1980, approximately 8-9 years before the survey. In addition to being consistent with the pattern described above, field experience from the GDHS suggests that these years were used as reference points for dating births, as Ghana experienced dramatic political and economic changes in those years. As shown in the figure, the GFS recorded a small fertility decline in the 22 five-year period preceding the survey. A detailed data-quality analysis of the GFS data suggested that no severe omission of births had occurred, but that mortality (and thus, the number of children who had died) was underestimated (Owusu, 1984). GDHS mortality estimates for the same period (the early to mid- seventies) confirm that mortality in the GFS was probably underestimated. This underestimation of mortality may be a factor in the higher fertility rates recorded in the GDHS for this period. Figure 3.2 Cumulated Age-Specific Fertility Rates for Women 15-34, GFS and GDHS 1965- 1987 No. of Children 6.0 5 .5 5 .0 4 .5 4 .0 3 .5 1965 I , I , I q , I , I , I , I , I I , 67 69 71 73 75 77 79 81 83 85 1987 Note: Rates ere for 3 year moving averages, except 1978 and 1987. GFS ~' GDHS i Ghana DHS 1988 Another analysis of GFS fertility data concluded that, in spite of reporting errors which might lead to a spurious decline in fertility, the beginning of a long-term decline in fertility was suggested consistently by the data (Shah and Singh, 1985). Nevertheless, in addition to the undereount of deceased children in the GFS, there are a number of additional factors which could contribute to the inconsistency between the two sets of data. First, a large-scale out-migration of young males accompanied by a severe economic crisis took place in Ghana during the mid to late seventies and early eighties. The exact magnitude of the migration is not known, but may have been large enough to depress fertility temporarily. In late 1983 and early 1984, many of these migrants returned to Ghana. It was also around this time that the economic recovery programme was launched by the government and the economic situation began to improve. It is difficult to assess the impact of these changes on fertility behaviour but, overall, the data suggest that, even if a short-term fertility decline occurred in Ghana during the mid to late seventies, the decline has not continued into the eighties (see Figure 3.3). In fact, the low level of use of modem contraceptives, together with the relatively improved economic situation, may have led to a short-term increase in fertility, since traditional restraints on high fertility, such as breasffeeding and postpartum abstinence, often decrease with modemisation. The fact that the duration of breasffeeding does not appear to have decreased in this period may have contributed to the stability of fertility in recent years. Furthermore, the very small increase in the use of contraception in the years since the GFS (see chapter 4) and the stability in age at first marriage are not consistent with a large decline in fertility. A more 23 Figure 3.3 Age-Specific Fertility Rates, GFS and GDHS Births per 1,000 women 300 250 200 150 100 50 15-19 20~-24 25'-29 30-'34 35-r39 40-'44 Age I GFS (1977-80) --~- GDHS (1985-88)i 45-49 complete explanation of fertility trends in Ghana must be suspended until a detailed analysis of available sources of fertility data is undertaken. 3.3 CHILDREN EVER BORN Table 3.4 presents the distribution of children ever born by age of the mother, for all women and for currently married women. The mean number of children ever bom among all women is 3.2. The number of children ever born increases with age, from 0.2 for women aged 15-19 to 1.3 and 2.7 for age groups 20-24 and 25-29, respectively. The mean number of children ever born rises sharply to 4.2 at age 30-34 and reaches the highest level of 7.3 at age 45-49. This picture indicates that completed family size is quite high in Ghana. A similar pattern was observed in the 1979 GFS, where fertility rose from a very low level of 0.2 for age group 15-19 to a completed family size of 6.7 for age group 45-49. Since voluntary childlessness is rare in Ghana, the data presented in Table 3.4 suggest that the level of primary sterility is quite low, with 77 percent of all women of reproductive age having at least one child. Childlessness declines rapidly with age. As many as 81 percent of women in age group 15-19 have never had a child, while this percentage is 28 for those aged 20-24 and declines sharply to 8 percent for those aged 25-29, and 2 percent for women 40-49. Among women aged 45-49, 22 percent have 10 or more children; this percentage is 10 for women 40-44. The mean number of children ever born among currently married women (4.0) is higher than that for all women (3.2). The proportion of currently married women that have never had a child is much lower (7 percent) than among all women (23 percent). The mean number of children ever born is higher for married women than for all women at every age, especially at the younger ages. This pattern implies that most childbearing in Ghana takes place within marriage, or at least after the first marriage. 24 Table 3.4 Percentage Distr ibution of Children Ever Born (CEB) to All Women and to Currently Married Women, According to Age, GDHS, 1988 Children Ever Born Mean Age 0 1 2 3 4 5 6 7 8 9 10+ Total Number CEB All Women 15-19 80.7 17.2 2.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100,0 849 0.2 20-24 27.9 34.8 24.7 9.6 2.7 0.3 0.0 0.0 0.0 0.0 0.0 I00,0 867 1.3 25-29 8.4 14.2 23.9 26.4 16.0 7.7 2.5 0.7 0.0 0.0 0.1 100,0 867 2.7 30-34 2.2 5.3 11.6 16.5 20.0 21.3 12.9 7.6 1.2 0.6 0.8 100,0 644 4.2 35-39 1.7 3.2 4.0 10.0 ii.i 18.8 20.0 14.3 9.0 5.1 2.8 100.0 531 5.5 40-44 1.6 2.2 4.7 3.3 8.5 9.3 15.7 15.1 16.2 13.2 10.2 100.0 364 6.6 45-49 1.6 2.2 5.2 3.0 6.6 5.7 i0.i 13.7 18.3 11.7 21.9 100.0 366 7.3 All Ages 23.1 14.2 12.7 11.0 9.0 8.1 6.8 5.3 4.1 2.7 3.1 100,0 4488 3.2 Currently Married Women 15-19 38.5 52.3 8.6 0.6 0.0 0.0 20-24 13.7 37.3 32.4 12.6 3.5 0.5 25-29 5.3 12.9 24.5 27.5 17.6 8.5 30-34 1.8 4.0 10.9 15.8 21.6 21.6 35-39 8.8 2.7 3.6 9.7 11.2 19.0 40-44 1.0 1.9 4.8 2.9 7.1 9.0 45-49 1.8 2.1 5.6 3.2 4.9 4.6 0.0 0.0 0.0 0.0 0.0 i00.0 174 0.7 0.0 0.0 0.0 0.0 0.0 i00.0 593 1.6 2.8 0.8 0.0 0.0 0.I 100.0 752 2.8 13.7 7.7 1.4 0.5 0.9 100.0 569 4.3 20.7 14.2 9.3 5.5 3.2 I00.0 473 5.6 16.5 14.2 17.1 14.2 11.3 i00.0 310 6.8 9.8 11.9 18.6 11.9 25.6 100.0 285 7.4 All Ages 6.7 14.5 15.9 13.8 11.6 10.2 8.7 6.2 5.0 3.4 4.1 I00.0 3156 4.0 3.4 CHILDREN EVER BORN AND AGE AT FIRST MARRIAGE Table 3.5 shows the mean number of children ever born among ever-married women, according to the age of the woman at first marriage and the duration since her first marriage. In Ghana, where the use of contraceptives is relatively low, the proportion of time a woman spends exposed to the risk of pregnancy determines in great part the number of children she will have. Thus, the earlier a woman marries, the more children she would be expected to have. Table 3.5 Mean Number of Chi ldren Ever Born to Ever -Marr ied Women, by Age at First Marr iage and Years Since First Marriage, GDHS, 1988 Age at First Marr iage Years Since First Marr iage <15 15-17 18-19 20-21 22-24 25+ Total 0-4 0.9 0.9 1.0 I.i i.I 1.5 1.0 5-9 2.2 2.4 2.4 2.4 2.5 2.5 2.4 10-14 3.4 3.9 3.8 4.0 3.9 3.6 3.8 18-19 5.1 5.3 5.2 4.8 5.5 4.6 5.2 20-24 6.2 6.7 5.9 6.7 6.5 5.8 6.4 25-29 7.2 7.4 6.7 7.4 7.2 7.2 30+ 8.0 8.3 6.3 8.0 TOTAL 4.8 4.3 3.6 3.3 3.0 3.1 3.9 The data in Table 3.5 do not show much evidence that age at marriage influences fertility levels. With the exception of women who marry at age 25 or older, the mean number of children born in the first 20 years of marriage does not vary significantly between women who married at different ages. The longer period of exposure of women who married early, however, results in a slightly greater number of children ever born by the end of the reproductive years. There is also some evidence in the table that women who marry later have higher fertility in the first few years of marriage than women who marry earlier. For example, the mean number of children ever born in the first five years of marriage is slightly less than one among women who marry before age 18; it is approximately one for those who marry between the ages of 18 and 19, somewhat more than one among those who marry at age 20-24 and 1.5 among those marrying at age 25 or older. 3.5 AGE AT FIRST BIRTH The age at which women have their first child is an important indicator because it is related to several aspects of fertility. First, delayed childbirth has been an important factor in fertility decline in some countries. Second, women who start childbearing early are often found to have higher lifetime fertility than those who start childbearing late and, finally, childbearing among teenagers is considered undesirable, as it is often associated with social and economic problems for both mother and child. It should be noted that, in Ghana, research suggests that, although a small percentage of females have their first birth while not yet married, these women often postpone furl_her childbearing until later years when they have completed their education (see, for example, Bleek, 1976). The data presented in Table 3.6 give the distribution of women by age at first birth according to current age. The prevalence of very early childbearing has declined over time. While 36 percent of women currently aged 45-49 had their first birth before the age of 18, only 23 percent of those currently 26 Table 3.6 Percentage Distr ibut ion of Women by Age at First Birth, and Median Age at First Birth, According to Current Age, GDHS, 1988 Age at First Birth Current No Age Birth <15 15-17 18-19 20-21 22-24 25+ Total Number Median 15-19 80.7 I.I 12.6 5.7 I00.0 849 20-24 27.9 2.8 20.6 27.2 17.1 4.4 i00.0 867 19.9 25-29 8.4 4.8 22.3 22.6 22.6 14.5 4.7 100.0 867 20.0 30-34 2.2 6.4 28.0 23.4 17.7 15.4 7.0 i00.0 644 19.2 35-39 1.7 4.7 28.4 22.6 17.5 15.6 9.4 100.0 531 19.5 40-44 1.6 7.1 33.8 21.7 17.0 II.0 7.7 100.0 364 18.8 45-49 1.6 6.7 30.6 21.0 15.3 12.3 13.4 I00.0 366 19.3 TOTAL 23.1 4.2 23.3 20.2 14.9 9.6 4.7 i00.0 4488 - Omitted due to censor ing aged 20-24 did so. The onset of childbearing is concentrated in the age group 15-19 with the overall median between 19 and 20 years. By age 21, 63 percent of the women had given birth to their first child. The trend across age cohorts is somewhat erratic, but suggests a small increase from the oldest to the youngest cohorts. 3.6 AGE AT FIRST BIRTH BY BACKGROUND CHARACTERISTICS Table 3.7 shows the median age at first birth among women aged 20-49 years by current age and background characteristics. Variations in the age at first birth by place of residence and level of educational attainment are evident. Women residing in rural areas begin childbearing about a year before their counterparts in urban areas. This difference increases from the oldest to the youngest age groups. The urban-rural difference is about 0.5 years among women aged 45-49, slightly less than 1 year among women aged 30-34 and 1.5 years among women aged 20-24. The urban-rural difference in age at first birth can be attributed to many factors, including early marriage, a high economic value placed on children (as part of rural labour), and inaccessibility of family planning facilities in rural communities. At the regional level, Greater Accra (the most urbanised region) has the latest age at first birth (20.6 years). Apart from Greater Accra, there is no significant variation in median age at first birth among the regions. There is an inverse relationship between age at first birth and level of education. The median age at first birth among women with higher education is difficult to assess, because there are few women in this category and because not enough women in the age group 20-24 have had a birth to enable calculation of the median. Nevertheless, women with higher education appear to have a median age at first birth of around 24 years, followed by women with middle school education at 19.9 years and the remaining women at approximately 19 years. 27 Table 3.7 Median Age at First Birth Among Women Aged 20-49 Years, by Current Age and Selected Background Character ist ics, GDHS, 1988 Current Age Background Total Character ist ic 20-24 25-29 30-34 35-39 40-44 45-49 (Ages 20-49) RESIDENCE Urban 21.1 21.2 19.9 20.0 19.3 19.6 20.3 Rural 19.6 19.6 19.0 19.3 18.5 19.1 19.3 REGION Western 20.2 20,i 19.3 19.0 19.0 17.8 19.5 Central 19.9 19.7 18.8 19.7 18.3 19.5 19.4 Greater Accra 21.2 21.6 20.8 20.4 19.4 19.6 20.6 Eastern 19.6 20.0 19.0 18.9 ]9.3 17.8 19.3 Volta 19.6 20.1 19.9 19.1 19.0 20.9 19.9 Ashant i 20.5 20.2 18.8 19.5 17.9 19.2 19.5 Brong Ahafo 19.5 19.9 19.4 18.9 19.0 20.5 19.5 Upper West, East 19.8 19.3 19.1 19.9 19.0 19.0 19.4 and Northern LEVEL OF EDUCATION No educat ion 19.3 19.4 19.0 19.1 18.9 19,2 19.2 Pr imary 18.9 19.6 18.6 19.3 17.9 18.7 19.0 Middle 20.1 20.3 19.1 20.8 19.0 19.6 19.9 Higher 23.1 23.5 24.0 * 23.7 TOTAL 19.9 20.0 19.2 19.5 18.8 I9.3 19.6 * Fewer than 20 cases - Less than 50 percent have had a b i r th 28 CHAPTER 4 FERTILITY REGULATION This chapter focuses on family planning in Ghana. It appraises the respondents' knowledge of methods, sources of supply and the perceived problems (if any) for different contraceptive methods. It also considers current and past practice, and knowledge of the ovulatory method. Special consideration is given to nonuse and intention to use family planning in the future. Also, information was collected on the availability of services and exposure to media coverage of family planning. The idea of the use of contraception as a measure for spacing and controlling the number of children a couple may have is not new to the Ghanaian community. Traditional methods have been used throughout the country from time immemorial. These methods include periodic abstinence, absence from the husband for a period of time, and withdrawal. That these methods proved ineffective is reflected in the rapid growth of the population (Republic of Ghana, 1969). The view that the rapid growth of population in Ghana is due to natural increase rather than migration is supported by research (e.g., Galsie, 1984) and is reflected in the 1969 Ghana population policy. The realisation of the need to curb rapid population growth, which is viewed as a threat to all developmental efforts by both individuals and government, made it necessary to adopt much more effective and reliable control measures. Approaches to the control of population growth, either through activities of voluntary organisations or central government policies, are directed towards the control of fertility rather than migration. The first attempt in this direction was in 1961, when the Christian Council of Ghana set up a Family Advice Centre with the objective of advising married couples on family planning and responsible parenthood. The second organised effort towards fertility control was made in 1967, when the Planned Parenthood Association of Ghana (PPAG) was set up with branches at various centres throughout the country. The PPAG is a voluntary organisation with the objectives of educating the public on the possibilities and the benefits of family planning and offering modem family planning services (Central Bureau of Statistics, 1983). The most comprehensive and positive move on the part of the Government of Ghana to control population growth was in 1969, when the first Ghana Population policy was launched. The policy was aimed at encouraging people to plan their families and to control family sizes through the reduction of births. In order to realize these objectives, a secretariat was established and given the responsibility to manage family planning logistics, clinics, public education, and field activities. The Ghana Govemment, private organisations, and individual efforts aimed at controlling fertility tend to emphasize the use of modem birth control techniques. These methods have been in limited use since their introduction, but there has been no comprehensive evaluation apart from the Ghana Fertility Survey. The 1988 GDHS, therefore, is intended, among other things, to serve as a source of information for ascertaining the impact of modem contraceptive methods. Findings from this chapter are of practical importance to planners and policymakers in connection with programmes related to contraception as a measure to control fertility in Ghana. Questions on contraception were intended to elicit information related to four main areas: knowledge, use, source of method, and problems associated with use of contraception. Spontaneous knowledge of methods was ascertained first by asking the question, "There are various ways or methods a couple can use to delay or avoid a pregnancy. Which of these ways or methods have you heard about?" Methods which were named spontaneously by respondents were recorded. Then, knowledge of methods which were not spontaneously given by the respondents was obtained by reading a sentence describing the method and subsequently asking if the respondent had heard of the method described. Then, for each method they reported they had heard of, women were asked whether they had ever used that method. In 29 addition, a question was asked about where the respondent would obtain the method if she wanted to use it, and the main problem she perceived with using the method. 4.1 KNOWLEDGE OF METHODS AND SOURCE Table 4.1 gives the percentage distribution of all women and currently married women by knowledge and source of contraceptive method. Knowledge about contraceptive methods is fairly high among women in Ghana. More than three-quarters of the female population interviewed claimed to know some method of contraception, 76 percent of all women and 79 percent of currently married women (see Figure 4.1). Knowledge of modem methods is also high, 74 percent for all women and 77 percent for currently married women. Furthermore, currently married women showed greater knowledge of sources of contraception than all women. Table 4.1 Percentage of Al l Women and Current ly Marr ied Women Knowing Any Contracept ive Method and Knowing a Source (For Information or Services), by Speci f ic Method, GDHS, 1988 Know Know Method Source Current ly Current ly All Marr ied All Married Method Women Women Women Women Any Method 76.2 79.4 69.8 73.4 Any Modern Method 73.8 76.5 66.4 69.6 Fil l 59.7 63.9 49.1 53.3 IUD 36.7 39.6 29.5 32.3 Inject ion 42.6 46.5 36.2 39.9 Vaginal Methods 36.6 38.0 31.9 33.1 Condom 48.5 49.5 38.3 39.3 Female Ster i l isat ion 54.1 57.3 48,8 52.0 Male Ster i l i sat ion 10.7 10.7 9.0 9,2 Any Tradit ional Method 49.2 51.9 35.1 37.1 Per iodic Abst inence 39.0 41.0 35.1 Withdrawal 31.0 32.6 Other 8.6 9.4 37.1 Number 4488 3156 4488 3156 Considering specific methods, the pill is the most well-known method among all women and married women, followed by female sterilisation. The pill is known by 60 percent of all women and 64 percent of married women, while female sterilisation is known by 54 percent of all women and 57 percent of married women. Knowledge of traditional methods is also high, 49 percent among all women and 52 percent among married women. More than a third of women know about each of the other methods, except male sterilisation, which was known by less than 11 percent. Comparing the evidence from the 1988 GDHS with the GFS data of 1979, it is observed that knowledge of contraceptive methods among currently married women is increasing. In 1979, 69 percent of currently married women knew of one or more methods; this rose to 79 percent in 1988. 30 100 80 60 40 20 Figure 4.1 Family Planning Knowledge and Use, Women in Union 15-49 Percent / o_/ Knows Method Ever Used Currently Using m Modern method m Any method Ghana DHS 1988 4.2 KNOWLEDGE OF MODERN METHODS AND SOURCE BY BACKGROUND CHARACTERISTICS Table 4.2 shows the percentage distribution of currently married women knowing at least one modem method and a source by selected background characteristics. More than 76 percem of all currently married women know at least one method of modem contraception. Apart from women aged 45-49, who show a relatively low level of knowledge (67 percent), the other age groups are all substantially higher without much variation between them; however, the level of knowledge about both methods and sources of modem contraceptives appears to be most common among currently married women aged 20-39. The variation in knowledge of method and source of modem contraception between the urban and rural sectors is significant. For urban women, knowledge about method and source are 88 percent and 82 percent, respectively, while, for rural women, these percentages are 71 and 64. With respect to regional variations, apart from the Upper East, Upper West, and Northern regions, where only 40 percent know a method and 32 percent know a source, more than 70 percent of married women know a method and more than two-thirds know a source in each of the regions. The very low level of education among the women in the three northernmost regions may be a factor in the low level of knowledge of methods there. Women in Greater Accra have the highest level of knowledge for both method (94 percent) and source (84 percent). This could be explained by the fact that Greater Accra is the most urbanised region and women there have better access to publicity and greater availability of family planning services. Significant differentials are observed amongst the women with respect to educational levels attained. Knowledge of both method and source are positively related to level of education. Women with no education have the least knowledge of both method (60 percent) and source (52 percent) and these 31 Table 4.2 Percentage of Current ly Marr ied Women Knowing at Least One Modern Method, and Knowing a Source for a Modern Method by Selected Background Characterist ics, GDHS, 1988 Number Background Know Know of Character is t ic Method Source Women AGE 15-19 71.3 62.1 174 20-24 77.6 71.7 593 25-29 80.7 75.1 752 30-34 77.0 71.0 569 35-39 76.5 68.7 473 40-44 74.8 66.5 310 45-49 66.7 57.5 285 RESIDENCE Urban 88.0 82.2 961 Rural 71.4 64.1 2195 REGION Western 84.2 82.1 279 Central 74.5 67.8 329 Greater Accra 93.6 83.9 360 Eastern 86.6 80.4 448 Volta 77.5 73.6 356 Ashanti 85.3 74.8 552 Brong Ahafo 71.6 67.1 401 Upper West, East 40.4 32.3 431 and Northern LEVEL OF EDUCATION No educat ion 60.3 52.0 1467 Pr imary 84.2 75.4 512 Middle 92.5 87.8 999 Higher 97.8 96.1 178 TOTAL 76.5 69.6 3156 percentages rise progressively to 98 percent and 96 percent, respectively, for women with higher education. Large urban-roral, education, and regional differentials in knowledge of methods among currently married women were also found in the GFS. The magnitude of these differentials does not seem to have lessened in the years since the GFS. 4.3 ACCEPTABILITY OF METHOD Table 4.3 gives the distribution of the responses given by women who know a particular method according to the main problem perceived in using the method. Evidence from this table may be useful in identifying obstacles affecting the use of specific methods with practical implications for future educational and publicity campaigns. Field experience from the GDHS suggests that the findings from this question must be handled with caution, due to the possibility that respondents often appeared confused by it but may have felt compelled to supply an answer. 32 Table 4.3 Percentage Distribution of Women Who Have Ever Heard of a Contraceptive Method by Main Problem Perceived in Uslng the Method, According to Specific Method, GDHS, 1988 Main Female Male Periodic Problem Vaginal Sterili- Sterill- Absti- Perceived Pill IUD Injection Methods Condom sation satlon hence Withdrawal No problem 22.5 17.2 25.1 35.7 27.3 27.3 23.7 63.3 38.3 Not effective 5.0 2.2 1.6 4.8 7.4 0.4 0.2 3.1 7.8 Partner disapproves 0.5 0.i 0.3 0.4 8.7 0.2 0.4 2.2 2°8 Health concerns 22.4 22.0 7.9 6.8 3.0 12.2 5.0 1.0 4.8 Difficult to get 0.i O,l O.l O.0 0.0 O.0 0.0 0.0 0.0 Costs too much 0.i 0.2 0.2 0.4 0.3 0.3 0.4 O.0 0.0 Inconvenient to use 1.5 2.0 l.O 3.5 5.4 1.3 0.6 2.2 5.4 Other 0.4 0.2 0.I 0.0 O.l 0.9 0.8 0.2 0.3 Don't know 47.4 55.6 63.4 48.3 55.5 57.0 68.0 27.4 39.7 Missing 0.2 0.2 0.4 0.2 0.3 0.3 0.8 0.5 0.9 TOTAL i00.0 I00,0 1O0.0 I00,0 i00.0 i00.0 I00.0 100.0 I00.0 Number 2681 1647 1910 1643 2175 2428 482 1750 1390 The table shows that the response "don't know" is the most common for all the methods, with the exception of periodic abstinence. After "don't know," "no problem" is the next most common response for all the methods, except the IUD. Apart from these two responses, what appears to be the most highlighted problem associated with use is health concerns, although a few women also mentioned ineffectiveness and inconvenience of use. Other reasons, such as cost, partner's approval and accessibility, do not appear to be important problems associated with the use of known methods. Periodic abstinence, which has been commonly practiced and needs no education or publicity, appears, in the opinion of the respondents, to have the fewest problems, with 63 percent claiming that there is no problem associated with its use. Withdrawal appears to be the most adjudged ineffective of all the methods, with almost 8 percent citing this as the main problem with using the method. Relative to the modem methods, parmer's disapproval with the traditional methods appears higher, with 2 percent for periodic abstinence and 3 percent for withdrawal. Among modem methods, the most recurring problem mentioned is health concerns, which was mentioned by 22 percent for the pill, 22 percent for the IUD and 12 percent for female sterilisation. Field experience suggests that, for the pill, complaints arc associated with abnormal fatness, dizziness and nausea. For the IUD, some respondents reported excessive bleeding, and complications associated with it. With the injection, respondents complain that it suppresses ovulation, and may result in infecundity. Female sterilisation is reported to be associated with weaker health, and the suppression of sexual interest. Respondents also expressed reluctance to use an irreversible method, especially given the possibility of dissolution of the present marriage or the death of children. In summary, most respondents who know a method are either not aware of any problems associated with its use or believe the method can be used without problems. Of those who named a problem, most mentioned health concerns. 33 4.4 KNOWLEDGE OF SUPPLY SOURCES The distribution of women knowing a contraceptive method by the supply source they would use if they wanted to use the method is given in Table 4.4. For the IUD, injection, and female and male sterilisation, the majority of women named government hospitals. Government hospitals and health centers were also named by about a third of the women as sources for the pill. Respondents indicate that pharmacy and chemical shops are also important supply sources for the pill, vaginal methods, and condom. PPAG clinics were identified by 10-15 percent of respondents as supply sources for the pill, IUD, injection, and vaginal methods. For periodic abstinence, friends and relatives are the most named source of information. It is worth noting that responses of "don't know" a supply source are quite high for all the methods. Table 4.4 Percentage Distribution Of Women Knowing a Contraceptive Method by Supply Source Named t According to Specific Method, GDHS, 1988 Female Male Periodic Vaglnal Sterili- Sterill- Absti- Source Pill IUD Injection Methods Condom sation satlon hence Government hospital 24.1 54.2 60.6 14.4 11.2 84.0 77.8 10.4 Government health center 7.0 6.4 9.0 3.5 2.5 1.5 1.0 3.2 PPAG clinic 15.1 14.8 11.2 11.7 8.O 2.4 3.1 9.4 Private maternity homo O.5 0.2 0.4 0.2 0.i 0.0 0.0 0.5 Field worker 0.8 0.2 0.4 0.8 0.2 0.I 0.0 0.7 Private doctor/cllnlc 0.4 0.7 1.3 0.4 0.3 0.4 0.6 0.6 Government ma~ernlty homo 0.3 0.4 0.2 0.i 0.0 0.0 0.0 0.5 Pharmacy/chemlcal soller 31.3 2.7 1.2 52.5 53.4 0.4 0.4 0.i Christian Council 0.0 0.i 0.i O.O 0.0 0.O 0.0 1.0 Frlends/relatives 1.7 0.5 0.4 2.3 0.8 0.5 0.4 40.6 Other 0.4 0.0 0.1 1.2 1.8 0.4 0.0 8.9 Nowhere O.4 0.4 0.5 0.8 0.5 0.3 0.4 14.0 Don't know 17.8 19.6 14.7 12.7 20.7 9.6 15.6 9.9 Missing O.I 0.I 0,2 0.2 0.2 0.2 0.6 O.l TOTAL iO0.0 i00.0 i00.0 I00.0 i00,0 i00.0 10O.0 i00.0 4.5 EVER USE OF CONTRACEPTION Table 4.5 presents summary data on all women and currently married women who have used any contraceptive method by age of the women. The table shows that only a third of all the women in the sample have ever used any contraceptive method. Also, only 21 percent indicated that they have ever used any modem method. Even the proportion that have ever used a traditional method is low; namely, 23 percent for any traditional method, 18 percent for periodic abstinence and 8 percent for withdrawal. The pill, which is the most popular modem method, has been used by 13 percent, followed by vaginal methods (8 percent), condom (5 percent), IUD (1 percent) and, lastly, injection and female sterilisation. Ever-use of contraceptives rises with age from 13 percent for ages 15-19 to more than 40 percent between ages 25 and 39, and declines thereafter. Female sterilisation, however, is most used by women aged 40 and older. 34 t~ Table 4.5 Age Percentage of All Women and Currently Married Women Who Have Ever Used a Contraceptive Method, by Specific Method and Age, GDHS, 1988 Any Any Female Tradi- Periodic Any Modern Injec- Vaginal Sterili- tional Absti- With- Method Method Pill IUD tlon Methods Condom sation Method hence drawal Other Number All Women 15-19 12.6 6.1 3.2 0.0 0.i 2.7 1.6 0.0 9.9 6.5 3.8 1.4 849 20-24 35.4 19.5 9.9 0.3 0.2 9.1 5.3 0.0 26.2 20.1 Ii.0 3.2 867 25-29 42.7 26.1 15.5 0.7 0.8 12.5 6.9 0.0 30.7 24.9 8.9 3.2 867 30-34 42.4 28.0 18.8 1.9 0.8 10.9 6.4 i.i 26.7 21.7 8.7 3.6 644 35-39 40.9 24.9 18.6 1.9 1.7 7.3 4.5 0.9 26.0 20.3 8.5 3.6 531 40-44 37.9 26.6 18.1 3.6 2.7 7.1 3.0 3.8 23.1 18.7 8.0 3.0 364 45-49 29.8 16.9 10.9 2.2 1.6 2.2 1.9 3.0 18.3 16.9 4.9 0.3 366 TOTAL 33.9 20.5 12.8 1.2 0.9 7.9 4.5 0.8 23.1 18.3 7.8 2.7 4488 Currently Married Women 15-19 21.8 12.1 6.3 0.0 0.0 5.2 3.4 0.0 17.2 9.2 8.6 2.9 174 20-24 32.5 18.2 10.3 0.3 0.2 8.9 4.4 0.0 23.6 17.0 9.3 3.5 593 25-29 41.0 24.2 15.4 0.7 0.8 10.2 5.9 0.0 29.9 24 .i 8.4 3.1 752 30-34 41.5 27.6 19.2 1.4 0.7 10.7 6.3 i.i 26.2 21 .3 8.3 3.2 569 35-39 39.7 23.7 18.0 1.9 1.9 7.2 3.6 i.i 25.4 19.9 8.2 3.8 473 40-44 39.4 27.1 19.0 3.5 2.9 7.1 2.3 4.2 24.2 19.7 8.4 2.9 310 45-49 29.1 16.5 10.5 2.5 1.4 2.1 1.8 2.5 18.6 16.8 6.3 0.4 285 TOTAL 37.0 22.5 14.9 1.3 1.0 8.3 4.5 1.0 25.1 19.7 8.3 3.0 3156 t~ C~ Table 4.6 Percentage Distribution of All Women and Currently Married Women, by Contraceptive Method Currently Used, According to Age, GDHS, 1988 Any Any F~a le Tradi- Periodic Any Modern Diaphragm/ Foaming Sterill- tlonal ~bst i- With- Not Age Method Method Pill IUD Injection Jelly Tablets Condom satlon Method hence drawal Other Using Total Number All Women l 15-19 5.8 1.3 0.6 0.0 0.0 0.i 0.5 0.i 8~5 ~ 4.5 3.1 0.8 0.6 94.2 I00.0 849 20-24 13.5 4.0 1.6 0.2 0.0 0.5 1.0 0.7 I|0 O 9.5 7.3 1.3 0.9 86.5 i00.0 887 25-29 14.1 4.6 1.8 0.2 0.3 0.2 1.3 0.7 i~.3 ~ 9.5 8.1 0.7 0.7 85.9 10O.O 867 30-34 15.1 7.0 2.8 0.8 0.0 0.6 1.7 0.0 1.7 #.~ 8.1 6.8 0.5 0.8 84.9 10O.O 644 35-39 14.5 5.8 1.9 1.3 0.4 0.2 0.9 8.2 0.9 ~ 8.7 7.0 0.9 0.8 85.5 10O.0 531 40-44 16.5 8.2 2.2 i.I 0.5 0.8 0.5 8.0 0.5 ~,~ 8.2 6.9 i.I 0.3 83.5 100.0 364 45-49 7.9 4.9 0.5 0.5 0.5 0.3 0.0 8.8 0.0 ~,~3.0 2.5 0.5 0.0 92.1 10O.O 366 TOTAl 12.3 4.7 1.6 0.5 0.2 0.3 0.9 0.3 0.9 7.6 6.1 0.8 0.6 87.7 10O.0 4488 Currently Married Women TOT~L 12.9 5.2 1.8 0.5 0.3 0.3 l.O 0.3 1.0 7.7 6.2 0.9 0.6 87.1 108.0 3156 15-19 4.6 2.3 2.3 0.0 8.0 0.0 8.0 O.O 0.8 2.3 8.6 0.6 i.i 95.4 i00.0 174 20-24 II.I 3.4 1.2 0.2 0.0 0.3 0.8 0.8 0.8 7.8 5.4 1.2 1.2 88.9 180.0 593 25-29 13.2 4.3 1.9 0.3 0.4 0.i 1.2 0.4 1.2 8.9 7.8 0.8 0.3 86.8 100.0 752 30-34 14.4 6.9 2.8 0.5 0.0 0.7 1.8 0.0 1.8 7.6 6.5 0.5 0.5 85.6 100.0 569 35-39 15.2 5.9 1.7 1.3 0.4 0.2 1.1 0.2 i.I 9.3 7.4 i.I 0.8 84.8 I00.0 473 40-44 18.4 9.0 2.3 1.3 0.6 O.0 0.6 0.0 0.6 9.4 7.7 1.3 0.3 81.6 100.0 310 45-49 7.7 4.2 0.7 0.4 0.4 0.4 0.0 0.0 0.0 3.5 2.8 0.7 O.0 92.3 I08.8 285 Among currently married women, a pattern similar to that observed among all women emerges. However, currently married women show a slightly higher level of ever-use. Ever-use is lowest among married women aged 15-19 years, increases with age to the highest level among women aged 25-44 years, then decreases for married women 45-49 years. A higher level of ever-use within the middle-age range may be an indication of married women's deliberate effort to space children. The slightly higher level of ever-use among currently married women, as compared to all women, is expected since married women are more likely to be exposed to the risk of pregnancy than their unmarried counterparts. 4.6 CURRENT USE OF CONTRACEPTION The level of current use of contraceptive methods is the most reliable measure to appraise the impact of a family planning programme. Further, it can be used to estimate the reduction in fertility attributable to contraception. Table 4.6 presents data on the proportion of all women and currently married women who are using contraception by age. Among currently married women, slightly less than 13 percent are currently using any method and 5 percent arc using any modem method. Traditional methods appear more popular among current users than modem methods. The proportion using any traditional method is close to 8 percent. Periodic abstinence is the most commonly used method overall. Among the modem methods, the piU is the most commonly used, followed by female sterilisation and foaming tablets. The proportion of currently married women using other modem methods is less than 1 percent. Current use of contraception among currently married women is lowest among women aged 15- 19 years (5 percent), increases to a peak of 18 percent for women aged 40-44, then declines sharply to 8 percent among women aged 45-49 years. The comparatively low level of use among younger and older women may indicate, for the younger women, the eagemess to have children as they are newly married and, for the older age group, the belief that they are no longer capable of bearing children. The relatively higher proportion of users between ages 35 and 44 suggests a conscious effort of some married women to space or avoid further births. Table 4.7 Percentage of Current ly Marr ied Women Current ly Using Contracept ion by Method, 1979 and 1988 Method GFS-1979' GDHS-1988 Pil l 2.4 1.8 IUD 0.3 0.5 Condom 0.6 0.3 Withdrawal 0.2 0.9 Abst inence 3.8 6.2 Female Ster i l isat ion 0.5 l.O Injection 0.I 0.3 Vaginal Methods 1.6 1.3 Other Methods 0.0 0.6 TOTAL 9.5 12.9 * Appiah, 1985. 37 Compared with the fmdings of the 1979 GFS, the proportion of currently married women using a contraceptive method has shown a slight increase, rising from 9.5 percent in 1979 to 12.9 percent in 1988 (Table 4.7). It should be noted that in the GFS, prolonged abstinence and rhythm (or periodic abstinence) were recorded separately, while, in the GDHS, prolonged abstinence was not asked about specifically. The results of the GDHS suggest, however, that women using prolonged abstinence to avoid pregnancy generally were recorded as using periodic abstinence. For purposes of comparison, the proportions using prolonged abstinence and rhythm in the GFS are added together. In both 1979 and 1988, abstinence was the contraceptive method used most often, with the proportion doubling since 1979. Among modem methods, the pill was the most frequently used method in both 1979 and 1988; however, the proportion of women using the pill declined from 2.4 percent in 1979 to 1.8 percent in 1988. At the same time, there was a slight increase in the use of injection, female sterilisation, withdrawal, and the IUD. 4.7 CURRENT USE BY BACKGROUND CHARACTERISTICS Table 4.8 examines current use of contraception among currently married women according to method and selected background characteristics. The relationship between place of residence and contraceptive use is strong. Comparing urban and rural dwellers, a greater proportion of married women in the urban areas are currently using a method of contraception than their rural counterparts. Specifically, the proportion using any method is almost 20 percent among urban women and about half that among rural women. The use of both traditional and modem methods is higher among urban women. Eleven percent of urban women are using a traditional method, while 6 percent of rural women are using these methods. The pill is the most popular modem method among both urban and rural women. Overall, periodic abstinence is the most commonly used method among urban and rural currently married women--9 percent of urban women and 5 percent of rural women. Regional variation in the current use of contraceptives among currently married women is considerable. Greater Accra shows the highest proportion of usage for any method (27 percent) and for any modem method (11 percent). The overall level of use is lowest in the Western region (8 percent). Current use in the rest of the regions varies between 10 and 15 percent. However, there is an interesting variation among the regions with respect to the use of modem versus traditional methods. While the Western region shows the lowest level of overall use, almost 40 percent of use is attributable to modem methods. In comparison, the overall level of use is almost 11 percent in the three no~lhemmost regions, but only 7 percent of users use modem methods. With respect to the specific modem methods, the pill appears to be the most commonly used modem method in all the regions with the exception of Greater Accra, where the IUD is more common, and the Central region, where the proportion using female sterillsation exceeds the proportion using the pill. The use of traditional methods also varies by region. Greater Accra shows the highest proportion of women using a traditional method (17 percent), followed by Volta (11 percent) and Northern, Upper East, and Upper West regions (10 percen0. The other regions show lower levels of use of traditional methods. There is a positive relationship between the level of educational attainment and current use of contraception among currently married women (see Figure 4.2). This relationship is maintained for both modem and traditional methods. It is interesting to note that higher proportions of women with primary and middle education use the pill than those with higher education. The higher education group is more likely than other groups to be using the IUD, condom and foaming tablets, as well as periodic abstinence. Women in the two lower education groups are more likely to use female sterilisation than other women. This difference is probably a reflection of the differential age distribution between the two groups (i.e., less educated women are likely to be older than more educated women). 38 %0 Table 4.8 Percentage Distribution of Currently M.~rrled Women by Contraceptive Method Currently Used, According to Selected Background Characteristics, GDHS, 1988 Any ~ny Dia- Female Tradi- Periodic Background Any Modern phragm/ Foa2alng Sterill- t lonal Absti- With- Not NLun- Characteristic Method Method Pill IUD Injection Jelly Tablets Condom sation Method hence drawal Other Using Total her RESIDENCE Urban 19.6 8.1 2.7 1.6 0.3 0.4 1.6 0.6 0.9 11.4 8.6 1,6 1.2 80.4 i00.0 961 Rural 8.9 3.8 1.5 0.I 0.2 0.2 0.7 0.i 1.0 6.1 5.1 0.6 0.3 90.1 100.0 2195 REGION Western 8.2 3.2 1.8 0.0 0.0 0.0 0.7 0.4 0.4 5.0 3.6 0.7 0.7 91.8 100.O 279 Central 9.7 4.9 1.2 0.3 0.3 0.6 0.9 0.0 1.5 4.9 4.0 0.8 0.0 90.3 i00.0 329 Greater Accza 27.2 10.8 2.2 3.1 0.0 0.0 2.8 0.8 1.7 16.7 10.8 3.6 2.2 72.8 100.0 360 Eastern 11.4 5.8 2.7 0.2 0.2 0.7 8.2 0.4 1.3 5.6 4.5 0.7 0.4 88.6 100.0 448 V~Ita 14.6 3.9 1.7 0.0 0.6 0.3 i.I 0.O 0.3 10.7 9.0 1.7 0.0 85.4 1O0.0 356 A~hanti 10.1 6.5 2.2 0.2 0.7 0.4 1.3 0.5 1.3 3.6 2.7 0.0 0.9 89.9 1O0.O 552 Brong Ahaf~ 12.0 8.2 2.5 0.5 0.0 0.2 1.8 0.0 1.0 6.7 6.2 0.2 0.2 88.0 i00.0 481 Upper West, East i0.7 0.7 0.2 0.2 0.0 0.0 0.0 0.0 0.2 I0.0 9.7 0.0 0.2 89.3 10O.0 431 and Northern LEVF~L OF EDUCATION No education 8.5 3.2 1.0 0.3 0.0 0.3 0.4 0.0 1.2 5.3 4.9 0.2 0.2 91.5 100.O 1467 Primary 12.1 6.1 2.5 0.6 1.2 0.2 0.2 0.2 1.2 6.i 3.9 1.4 0.8 87.9 100.0 512 Middle 16.8 6.7 2.8 0.6 0.i 0.4 1.7 0.5 0.6 10.1 7.7 1.3 i.i 83.2 100.0 999 Higher 28.7 10.1 1.7 1.7 0.6 0.0 3.9 1.7 0.6 18.5 15.2 2.8 0.6 71.3 i00.0 178 NO. OF LIVING CHILDREN None 3.8 1.9 0.8 0.0 0.0 0.0 0.0 0.8 0.4 1.9 0.8 0.4 0.8 96.2 i00.0 261 1 i0.i 3.5 1.5 0.2 0.O 0.2 0.7 0.7 0.2 6.6 4.8 0.9 0.9 89.9 100.O 546 2 11.7 3.0 1.2 0.0 0.0 0.0 0.9 0.3 0.5 8.7 7.7 0.3 0.7 88.3 10O.0 572 3 14.3 5.1 1.5 0.6 0.4 0.4 1.5 0.8 0.6 9.1 7.9 1.3 0.0 85.7 100.0 470 4+ 15.8 7.5 2.6 1.0 0.5 0.5 I.i 0.i 1.8 8.3 6.7 i.i 0.6 84.2 100.0 1307 TOTAl 12.9 5.2 1.8 0.5 0.3 0.3 1.0 0.3 1.0 7.7 6.2 0.9 0.6 87.1 100.0 3156 50 40 30 20 10 0 Figure 4.2 Current Use of Family Planning by Education and Number of Living Children, Women in Union 15-49 Percent Hone Primary MIddle Higher EDUCATION 1 2 3 4* NO. OF CHILDREN Ghana DHS 1988 There are indications that the number of surviving children influences current use of contraceptives among currently married women. It can be seen from Table 4.8 that there is a progressive increase in the proportion of women currently using contraceptives as parity increases. This reflects the desire of women with lower parity to have more children, while those with higher parity resort to measures to reduce the frequency of conception by spacing births or stopping childbearing altogether. 4.8 NUMBER OF CHILDREN AT FIRST USE Table 4.9 examines the percentage distribution of ever-married women by number of living children at the time of first use of contraception according to current age. The number of living children at the time of first use of contraception is useful as a measure of the willingness to postpone the first birth and of a deliberate effort at spacing further births. The table indicates a shift over time toward beginning the use of family planning earlier in the family building process. For example, while more than 17 percent of ever-married women currently aged 20-24 adopted family planning measures before the birth of their first child, only 2 percent of those aged 40-44 did so. 4.9 KNOWLEDGE OF THE FERTILE PERIOD All respondents in the GDHS were asked a question intended to ascertain their knowledge of the basic reproductive physiology of women. This background knowledge is necessary for the successful practice of periodic abstinence as a fertility control measure. The specific question asked of the respondents was, "When during her monthly cycle do you think a woman has the greatest chance of becoming pregnant?" Evidence from Table 4.10 indicates that, of all the women interviewed, about half claim to have no knowledge of the ovulatory cycle. Only 27 percent of women responded correctly overall. Among the women who have ever used periodic abstinence, 50 percent responded correctly. It must be noted, 40 however, that this was one of the questions that respondents found most difficult to answer and their responses were sometimes difficult to categorize. Thus, for example, some of the women who responded "after her period has ended" may, in fact, understand the ovulatory cycle but weren't able to communicate this to the interviewer. Table 4.9 Percentage Distr ibut ion of Ever-Marr ied Women by Number of L iv ing Chi ldren at Time of First Use of Contraception, According to Current Age, GDHS, 1988 Number of L iving Chi ldren at Time of F irst Use No Never Living Age Used Chi ldren 1 2 3 4+ Miss ing Total Number 15-19 79.1 16.5 4.4 0.0 0.0 0.0 0.0 i00.0 206 20-24 66.3 17.1 12.1 3.6 0.7 0.0 0.i 100.0 671 25-29 57.4 13.0 15.9 8.5 3.9 1.2 0.i 100.0 828 30-34 57.9 7.4 11.0 9.6 6.0 7.9 0.3 100.0 636 35-39 59.1 4.4 9.3 6.4 6.6 14.2 0.0 100.0 528 40-44 62.3 2.2 5.8 4.4 5.5 19.8 0.0 I00.0 363 45-49 70.2 I.i 6.6 2.5 3.8 15.6 0.3 100.0 366 TOTAL 62.4 9.4 10.7 5.9 4.0 7.3 0.I 100.0 3598 Table 4.10 Percentage Distr ibut ion of Al l Women and Women Who Have Ever Used Per iodic Abst inence by Knowledge of the Fert i le Period Dur ing the Ovulatory Cycle, GDHS, 1988 Ever Used Fert i le Al l Per iodic Per iod Women Abst inence Dur ing her per iod 9.5 0.5 After per iod has ended 21.3 33.7 Middle of the cycle 26.6 50.4 Before per iod begins 2.5 4.6 At any time 2.3 1.3 Other 0.i 0.0 Don't know 46.6 9.4 Miss ing 0.I 8.i TOTAL I00,0 i00.0 Number 4488 823 41 4.10 SOURCE OF SUPPLY OF CONTRACEPTION Table 4.11 and Figure 4.3 examine the supply source for contraceptive methods among current users. The methods are grouped into supply (pill, condom, vaginal methods, injection) and clinic methods (IUD, female sterilisation). Results are also shown separately for users of the pill, foaming tablets, female sterilisation and periodic abstinence. The private sector is the major source of supply methods, with the PPAG clinics supplying 20 percent of users, and pharmacy and chemical shops supplying about 33 percent. A surprising finding is that 20 percent of users of supply methods named friends or relatives as their most recent source of supply methods. Since rural users are somewhat more likely to name this source than urban users, and since supply outlets such as pharmacies are more accessible in urban areas, this result suggests that some respondents may have named as their source a person who purchased the method for them (e.g., her husband) rather than the place where the method was purchased. Relative to the private sector, the government sector, comprising hospitals and health centres is not a major source of the supply methods. Together, government hospitals and health centres were named as a source of supply methods by about 19 percent of users. Table 4.11 Percentage Distr ibut ion of Current Users by Most Recent Source of Supply or Information, According to Speci f ic Method, GDHS, 1988 Total Total Female Total Periodic Supply Foaming Cl inic Steri l l - Modern Abst i - Source Methods ~ Pil l Tablet Methods sation ~ Methods* hence Did not visit source 0.0 0,O 0.0 0.0 0.0 O.0 13.8 Government hospital ii.9 21,9 0.0 74.6 81.1 29.5 4.4 Government health center 6.6 5,5 7.1 1.7 0.0 5.2 1.8 PPAG cl inic 19.9 23,3 14.3 11.9 2.7 17.6 5.1 Field worker 1.3 2,7 0.0 0.0 0.0 l.O 0.4 Private doctor /c l in ic 1.3 2,7 0.0 3.4 2.7 1.9 2.6 Goverru~ent materni ty home 0.0 0,0 0.0 0.0 0.0 0.0 1.5 Pharmacy/chemical sel ler 32.5 30,1 38.1 0.O 0.0 23.3 0.0 Chr ist ian Counci l 0.7 1,4 0.0 0.0 0.0 0.5 1.5 Fr iends/re lat ives/school 19.9 II.0 35.7 0.0 0.0 14.3 62.8 Other 3.3 0.0 4.8 3.4 5.4 3.3 6.2 Inconsistent 1.3 0,0 0.0 3.4 5.4 1.9 0.0 Miss ing 1.3 1,4 0.0 1.7 2.7 1.4 0.0 TOTAL i00.0 I00.0 i00.0 100.0 i00.0 I00.0 100.O Number 151 73 42 59 37 210 274 Supply methods include pill, condom, vaginal methods, in ject ion • Cl in ic methods include female ster i l isat ion and IUD Modern methods include supply and cl inic methods Government hospitals appear to be the most important source for clinic methods followed by PPAG clinics. About three-quarters of users of clinic methods, primarily users of female sterilisation, named a government hospital as their source. Finally, respondents indicated that friends and relatives are their most important source of information or advice about periodic abstinence. 4.11 ATTITUDE TOWARD BECOMING PREGNANT Data presented in Table 4.12 show the attitude toward becoming pregnant in the next few weeks among married, sexually active, non-pregnant women who are not using contraception according to 42 Figure 4.3 Source of Family Planning Supply, Current Users of Modern Methods Government 35% Friends/relatives 14% PPAG 18% Other 8% Pr ivate Sources* z~ • Inc ludes pr ivate doctors , c l in ics and Chr i s t ian Counc i l Pharmacy 23°/, Ghana DHS 1988 number of living children. The responses are grouped into three categories: "happy," "unhappy," and "would not matter." On the whole, 50 percent of the respondents indicated they would be unhappy if they became pregnant, while 45 percent said they would be happy and 5 percent said it would not matter. Considering the responses by number of living children, there is an inverse relationship between the number of living children and being happy about becoming pregnant in a few weeks' time. The highest proportion of women who said they would be happy if they became pregnant is found among those with no children. This percentage declines as the number of living children increases. Table 4.12 Percentage Distr ibut ion of Non-Pregnant Women Who Are Sexual ly Active and Who Are Not Using Any Contracept ive Method by Att i tude Towards Becoming Pregnant in the Next Few Weeks, According to Number of Living Children, GDHS, 1988 Att i tude Towards Becoming Pregnant Number of Would Living Not Chi ldren Happy Unhappy Matter Miss ing Total Number None 61.7 35.7 2.2 0.4 100.0 457 1 56.7 40.2 2.5 0.6 100.0 356 2 51.0 45.3 3.4 0.3 I00.0 296 3 39.2 53.9 5°7 1.2 i00.8 245 4+ 27.7 64.4 6.8 1.2 100.0 763 TOTAL 44.5 50.2 4.5 0.8 100.0 2117 43 4.12 REASONS FOR NONUSE In an effort to ascertain why women who are nonusers of contraception and who would be unhappy if they became pregnant were not using, these women were asked their main reason for not using. The responses to this question are shown in Table 4.13, according to age. Table 4.13 Percentage Distr ibut ion of Non-Pregnant Women Who Are Sexual ly Act ive and Who Are Not Using Any Contracept ive Method and Who Would be Unhappy if They Became Pregnant by Main Reason for Nonuse, According to Age, GDHS, 1988 Reason for Nonuse Age <30 30+ Total Lack of knowledge 29.0 18.3 23.7 Opposed to family p lanning 3.9 3.2 3.6 Husband disapproves 2.8 4.7 3.8 Others d isapprove 0.7 0.6 0.7 Health concerns 7.1 11.9 9.5 Access/avai lab i l i ty 2.2 1.5 1.9 Costs too much 1.9 2.3 2,1 Inconvenient to use 2.2 0.9 1.6 Infrequent sex 12.7 7.2 10.0 Fata l is t ic 0.2 0.8 0.5 Rel ig ion 2.4 4.2 3.3 Postpartum/Breastfeeding 8.2 5.9 7.1 Menopausal/Subfecund 0.2 18.0 9.0 Other 12.2 13.8 13.0 Don't know 13.1 6.6 9.9 Miss ing 0.9 0.2 0.6 TOTAL i00.0 i00.0 100.0 Number 534 529 1063 Note: Women who have never had sexual intercourse and women who have not resumed sexual relat ions since the last b i r th are excluded. Generally speaking, the main reason given for not using contraception is lack of knowledge (24 percent), followed by other unspecified reasons (13 percent). Additional reasons worth noting include infrequent sex (10 percent) and health concerns (10 percen0. The "don't know" category is also significant (10 percent). Variations in reasons for nonuse are evident between women grouped into two age categories, below 30 years and 30 years and above. Women less than 30 years of age are more likely than older women to state infrequent sex, postpartum/breastfeeding, and "don't know," while older women are more likely to cite menopause, health concerns, and husband disapproval. 44 4.13 FUTURE USE In Ghana, where the level of use of contraception among married women is relatively low, an indication of intention to use contraception in the future provides a useful indicator for planners and policymakers in assessing future demand for services. It must, however, be noted that not only may declared intentions differ from actual behavior, but also that indecision may influence responses to questions probing into the future. Table 4.14 indicatos the intention for future use of contraception among currently married women who are not currently using any contraceptive method, with respect to number of living children. On the whole, the inclination towards use of contraception in the future is very low. More than half of all women do not intend to use in the future, irrespective of parity. A smaller proportion said they intended to use within the next 12 months (20 percent), while 14 percent said they would use later. Table 4.14 Percentage Distr ibut ion of Current ly Marr ied Women Who Are Not Current ly Using Any Contracept ive Method by Intent ion to Use in the Future, According to Number of L iving Children, GDHS, 1988 Number of l iv ing chi ldren Intention to Use In the Future None 1 2 3 4+ Total Intends to use: In next 12 months 1.7 12.3 18.6 19.8 26.7 20.0 Later 13.6 15.4 15.6 15.9 11.4 13.7 Unsure about t iming i.i 4.0 2.8 4.6 4.2 3.6 Unsure about use 9.1 6.7 6.9 5.5 ~.6 5.8 Does not intend to use 74.4 61.3 56.4 53.7 52.9 56.5 Miss ing 0.0 0.2 0.6 0.5 0.3 0.3 TOTAL 100.0 108.0 i00.8 100.0 i00.0 180.0 Number 176 494 495 415 1170 2750 Those unsure of future use and those with no intention to use comprise 62 percent of women. It is worth noting that the highest proportions who do not intend to use are found primarily among women with no children or with low parity. 4.14 PREFERRED METHOD Table 4,15 examines method preferences among currently married women who are not using a contraceptive method but who intend to use in the future, according to intention to use in the next 12 months or later. This information should be interpreted with caution, since there are two conditions implied: intention to use and method preferred if the intention is foUowed. Among those who indicated an intention to use in the next 12 months, the pill, injection and periodic abstinence are preferred, while about 24 percent said they did not know which method they 45 would choose. Among those who intend to use later, the pill still appears to be the most popular method and there is an indication that more women in this group intend to choose female sterilisation. Table 4.15 Percentage Distr ibut ion of Current ly Marr ied Women Who Are Not Using a Contracept ive 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, GDHS, 1988 Intend to Preferred Use in Next Intend to Method 12 Months Use Later Total Pi l l 24.0 24.5 24.2 IUD 2.2 2.9 2.5 Inject ion 23.8 14.6 20.i Diaphragm/ Je l ly 2.0 1.1 1.6 Foaming Tablets 2.0 2.4 2.2 Condom 0.7 0.8 0.8 Female Ster i l isat ion 6.7 15.7 10.4 Male Ster i l i sat ion 0.0 0.3 0.1 Per iodic Abst inence 9.6 6.9 8.5 Withdrawal 0.5 0.5 0.5 Other 4.7 7.2 5.7 Don't know 23.6 23.1 23.4 TOTAL I00.0 i00.0 i00.0 Number 550 376 926 4.15 ACCEPTABILITY OF MEDIA MESSAGES ON FAMILY PLANNING Both the National Family Planning Programme and non-govemmental organisations are engaged in the propagation and dissemination of family planning information. Publicity measures include programrnes on radio, television, newspapers, posters and other adult education programmes. On the whole, it appears that radio is the most wide-reaching media and, as such, a question aimed at ascertaining the acceptability of disseminating family planning messages on the radio was asked of all respondents. Table 4.16 presents the proportion of women who said it was acceptable to have family planning messages on the radio, by age and background characteristics. Generally, the majority of women claim that it is acceptable to use the radio for family planning messages. The youngest and oldest women are somewhat less likely than women in the middle age groups to find radio messages acceptable. Acceptability is higher for urban dwellers (83 percent) than their rural counterparts (70 percent). At the various age levels the same urban-rural disparity is maintained. Substantial variations are observed regionally with Greater Accra showing the highest level of approval (93 percent) and the Northern, Upper East and Upper West regions showing least approval (45 percent). The level of acceptability is between 66 and 86 percent for the rest of the regions. There is an increase in the level of acceptability as the respondent's level of education increases. 46 Table 4.16 Percentage of All Women Who Believe That it is Acceptable to Have Messages A~ut Family Planning on the Radio, by Age and Selected Background Characteristics, GDHS, 1988 Age Background Characteristic 15-19 20-24 25-29 30-34 35~39 40-44 45-49 Total RESIDENCE Urban 70.3 85.5 88.2 87.2 90.2 86.0 78.3 82.9 Rural 62.8 73.2 72.8 75.7 72.8 68.1 61.8 70.3 REGION Western 61.3 73.6 76.1 87.5 78.8 67.6 69.0 73.0 Central 64,5 89.1 92.4 93.2 93.7 85.7 84.6 86.2 Greater Accra 85.3 95.0 99.1 94.3 97.2 91.8 93.0 93.3 Eastern 80.3 83.7 86.9 90.5 95.2 88.6 75.3 85.1 Volta 69.9 78.9 73.8 78.6 77.2 65.0 64.9 73.6 Ashanti 52.2 69.3 67.2 73.0 74.7 71.8 60.3 66.0 Brong Ahafo 61.6 83.5 78.9 79.3 67.2 60.6 31.0 73.2 Upper West, East 34.8 42.4 45.5 48.2 47.2 45.8 50.9 44.9 and Northern LEVEL OF EDUCATION No educat ion 45.1 59,0 62.3 63.4 68.5 68.0 62.5 62.1 Primary 64.4 78.5 78.1 85.4 88.3 80.0 76.8 77.2 Middle 71.0 86.8 87.0 91.0 87.5 94.6 83,3 83.4 Higher 85.5 92.8 93.6 93.6 100.0 - 92.0 TOTAL 65.7 77.5 77.5 79.3 78.5 74.5 67.2 74.6 - Fewer than 25 women 4.16 ATTITUDE TOWARD FAMILY PLANNING Results presented earlier in this chapter show that knowledge about contraception in Ghana is quite high against a rather low level of use. In low-use countries widespread disapproval of contraception may act as a major barrier to the adoption of methods. Accordingly, respondents in the GDHS were asked a question aimed at ascertaining approval of a couple's using family planning methods, and a question on the wife's views of her husband's attitude toward use. Caution is needed in the interpretation of these findings, since the wife's perception of her husband's attitude may be incorrect. (See Chapter 7 for results on this issue from the husband's survey). This notwithstanding, the findings may give an indication of the climate of opinion and may be used as the basis for planning further educational and promotional activities on family planning. Table 4.17 examines currently married women knowing a contraceptive method, by the husband's and wife's attitudes toward the use of family planning. The information is based on responses of the wife only. Overall, a substantial proportion (74 percent) of the women approve of a couple's using contraception. Ten percent of women approve, but believe that their husband disapproves, 49 percent approve and indicate that their husband approves and 14 percent approve but indicate that they do not know whether their husband approves or disapproves. Sixteen percent of women disapprove of family planning and say that their husband also disapproves, 3 percent say that their husband approves and 7 percent of women disapprove and do not know their husband's opinion. 47 Table 4.17 Percentage Distr ibut ion of Current ly Marr ied Women Knowing a Contracept ive Method by the Husband's and Wife's Att i tudes Toward the Use of Fami ly Planning, GDHS, 1988 Husband's att i tude* Wife 's Dis- Don't Att i tude approves Approves Know Miss ing Total Disapproves 15.8 2.6 7.0 0.I 25.5 Approves 10.4 49.3 14.4 0.2 74.3 Miss ing 0.0 0.1 0.0 0.0 0.2 TOTAL 26.2 52.1 21.5 0.2 100.0 Number 658 1305 538 6 2507 * As perce ived by wife 4.17 ATTITUDE TOWARD FAMILY PLANNING BY BACKGROUND CHARACTERISTICS Table 4.18 examines the proportion of currently married women knowing a contraceptive method who approve of family planning and who say their husband approves of family planning, by background characteristics. Evidence of differentials will be a useful guide in the interpretation of GDHS data on adoption and use of family planning. For instance, differences between age groups may reflect generational change, with younger women being more responsive to new ideas. On the other hand, there may be a countervailing life-cycle effect. Older women have larger families and, thus, may feel a greater need for contraception than younger women. This need may bring a shift toward a more positive attitude. The table suggests that a higher proportion of women approve of couples using family planning than their husbands. However, about 22 percent of the women do not know whether their husband approves or disapproves, and evidence from the husband's survey indicates that about the same percentage of husbands as wives approve (see Table 7.16). The proportion of women who approve of family planning rises with age to a peak in the early to mid-thirties and then declines. More urban than rural women approve of family planning, and the level of approval rises with education. Women in Northern, Upper East and Upper West regions show the least approval of family planning while women in Greater Accra region rank highest. 4.18 DISCUSSION OF FAMILY PLANNING Table 4.19 summarises data on currently married women knowing a contraceptive method by the number of times they have discussed family planning with their husband by age (see Chapter 7 for a discussion of husbands' answers to the same questions). Fifty-eight percent of women indicated that they have never discussed family planning with their husband and 20 percent indicate that they have discussed it only once or twice. Thus, three-quarters of married women either never, or rarely, discuss family planning with their husband. Twenty-three percent of the women have discussed it more often with their husbands. It is significant to note that the highest proportions who have never discussed family planning with their husband belong to the youngest and oldest age groups. 48 Table 4.18 Percentage of Currently Married Women Knowing a Contraceptive Method Who Approve of F~i ly Planning and Who Say their Husband Approves of Family Planning, by Selected Background Characteristics, GDES, 1988 Woman Woman Doesn't Says Know Background Woman Husband Husband's Characteristic Approves Approves Opinion Number AGE 15-19 61.8 36.6 32.8 131 20-24 71.6 52.1 21.0 476 25-29 75.5 52.9 21.6 624 30-34 80.6 57.8 21.0 453 35-39 76.0 54.6 19.8 379 40-44 76.0 50.0 18.3 246 45-49 65.7 43.9 22.7 198 RESIDENCE Urban 82.9 57.7 20.8 859 Rural 69.8 49.1 21.8 1648 REGION Western 64.5 43.8 17.4 242 Central 75.5 46.7 21.8 257 Greater Accra 90.2 60.8 20.5 337 Eastern 78.7 56.8 22.4 389 Volta 76.8 57.0 22.9 284 Ashanti 72.5 55.1 17.6 472 Brong Ahafo 74.7 63.3 17.6 289 Upper Wsst, East 53.6 20.3 35.9 237 and Northern LEVEL OF EDUCATION NO sducation 65.2 41.2 26.5 957 Primary 73.3 48.7 21.9 439 Middle 82.6 61.9 17.5 936 Higher 82.3 67.4 13.7 175 TOTAL 74.3 52.1 21.5 2507 49 Table 4.19 Percentage Distr ibut ion of Current ly Marr ied Women Knowing a Contracept ive Method by Number of Times Discussed Family P lanning with Husband, According to Current Age, GDHS, 1988 Number of Times Discussed FP Once or More Age Never Twice Often Miss ing Total Number 15-19 71.8 11.5 16,0 0.8 100.0 131 20-24 59.9 22.1 17.9 0.2 100.0 476 25-29 58.5 21.2 20.2 0.2 100.0 624 30-34 49.2 23.4 27.4 0.0 i00.0 453 35-39 54.4 18.5 27.2 0.0 100.0 379 40-44 56.9 13.8 28.9 0.4 100.0 246 45-49 69.7 13.6 16.7 0.0 100.0 198 TOTAL 57.9 19.5 22.5 0.2 100.0 2507 50 CHAPTER 5 FERTILITY PREFERENCES A major reason for the establishment of the Ghana National Family Planning Programme was to enable couples to bear the number of children they desire, with the births spaced according to their preferences (Republic of Ghana, 1969). In this chapter, data on the desire for additional children, preferred and ideal birth intervals, ideal family size and the potential need for family planning will be examined. Data on the desire for additional children were based on responses of currently married women. Currently married women who were not pregnant were asked the question, "Would you like to have a (another) child or would you prefer not to have any (more) children?" If a currently married woman was pregnant, the question was rephrased to read, "After the child you are expecting, would you like to. " The rewording was to ensure that pregnant women did not think that they were being asked about the child they were then expecting. Women who wanted additional children were subsequently asked about their preferred interval before the next birth. All women, irrespective of marital status, were asked a question on their desired total family size. Those with no living child were asked, "If you could choose exactly the number of children to have in your whole life how many would that be?" The question was modified for those with a living child to read, " If you could go back to the time you did not have any children and could choose exactly." Data on fertility preferences are generally more difficult to interpret than objective phenomena, such as actual fertility or contraceptive use. For instance, a woman's fertility preferences might change over time or her ability to implement her preferences might be curtailed if her partner objects to her using contraception to achieve her fertility desires. Such phenomena may not be captured in a standardized questionnaire. There is, therefore, the need to be cautious in the interpretation of the data on fertility preferences. 5.1 DESIRE FOR CHILDREN Table 5.1 and Figure 5.1 show the percentage distribution of currently married women by desire for children in the future, according to the number of living children they already have. On the average, one out of every five currently married women in the sample wants to have a child within two years. Forty-five percent express the wish to postpone the next birth for two or more years. On the other hand, some 23 percent of the women do not want any more children, whilst 5 percent are undecided and 4 percent cannot tell when they want the next child. Among women with no living children, half of them want a child in less than two years, and 18 percent believe they are infecund. However, while 11 percent of the group would want to postpone birth for two or more years, some 16 percent are undecided about when they want to have their first child. The desire to have a child within two years declines as the number of surviving children increases. For example, while a little more than a quarter (27 percent) of women with one living child want to give birth to another child soon (that is, in under two years) only one-fifth of those with 3 living children and only 7 percent of women with 6 or more children express a similar desire. Furthermore, among childless women, as already noted, 11 percent want to postpone the first birth as compared to women with one child, some 63 percent of whom want to delay the second birth. 51 Table 5.1 Percentage Dist r ibut ion of Current ly Marr ied Women by Desire for Children, According to Number of L iving Children, GDHS, 1988 Number of L iving Children* Desire for Chi ldren 0 1 2 3 4 5 6+ Total Wan~ another: Soos = 50.0 26.6 21.7 20.I 16.0 12.8 7.3 19.5 Later* 11.3 63.0 61.9 57.7 48,4 34.0 15,5 44.9 Unsure when 16.1 6.0 4.4 2.7 1,6 1.8 2.5 4ol Undecided 3.8 1.6 3.0 4,6 7,8 6.9 8.I 5.1 Want no more 0.5 0.7 6.8 13.3 25,1 39.4 62.3 22.8 Declared infecund 17.7 1.8 2.1 1.7 i,I 5.1 4.2 3.5 Miss ing 0.5 0.2 0.0 0.0 0,0 0.0 0.0 0.i TOTAL I00.0 100.0 100.0 100.0 i00,0 100.0 100.0 i00.0 Number 186 549 562 482 450 335 592 3156 Including current pregnancy Wants next b i r th within 2 years Wants to delay next birth 2 or more years Figure 5.1 Fertility Preferences, Women in Union 15-49 Want to Space (2 or more yre) 45% Want No More or Sterilized 23% Want Sou. (within 2 yrs) 20% Undecided If Want More 5% ~/ant Another, Undecided When 4% Infecund 4% Ghana DHS 1988 52 It is interesting to note that, while women with one to four living children are most likely to want to wait two or more years before having the next child, those with five or more living children at the time of the survey are most likely to report that they want no more children. The proportion expressing a desire to have no more children begins to rise when the woman has two living children. The proportion nearly doubles for each succeeding birth up to 4 children; it then increases less sharply with the fifth, only to go up one and one-half limes for women with six surviving children. The results reveal that there is a potential demand for contraception for birth spacing, especially among women with one to three children. One-quarter to three-fifths of mothers with four or more living children report that they do not want any more children, and arc, therefore, potential users of contraception for limiting births. Table 5.2 shows that the pattern of fertility preferences by age of the woman closely follows that of the parity-specific pattem noted earlier in Table 5.1 This is consistent with expectation, since age and parity are closely related. The dcsire to postpone the next birth is highest among women aged 20-24 years--69 percent of them express the wish to delay the next birth by two or more years. They are closely followed by the 15-19 age group. Table 5.2 Percentage Distr ibut ion of Current ly Marr ied Women by Desire for Children, According to Age, GDHS, 1988 Age Desire for Chi ldren 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total Wants another: Soon ~ 16.1 19.4 23.7 20.2 20.I 16.1 12.3 19.5 Later I 67.2 69.0 57.8 45.9 27.9 14.8 6.0 44.9 Unsure when 8.0 4.0 2.1 4.0 4.9 5.2 4.6 4.1 Undecided 5.2 2.2 3.5 4.4 8.9 I0.0 5.3 5.1 Wants no more 0.6 3.9 11.2 23.7 35.3 49.4 55.4 22.8 Declared infecund 2.3 1.3 1.7 1.8 3.0 4.5 16.5 3.5 Miss ing 0.6 0.2 0.0 0.0 0.0 0.0 0.0 0.I TOTAL i00.0 100.0 100.0 100.0 i00.0 100.0 100.0 100.0 Number 174.0 593.0 752.0 569.0 473.0 310.0 285.0 3156.0 I Wants next birth within 2 years. 2 Wants to delay next birth 2 or more years. The proportion of currently married women wanting no more children increases from less than one percent for women aged 15-19 to 55 percent for women aged 45 and older. Also, the fact that not more than 24 percent of women in any age group want children soon gives an indication of the need for reliable methods of contraception. 53 In Table 5.3, the percentage of currently married women who want no more children is shown for each parity by selected background characteristics, Table 5.3 Percentage of Current ly Marr ied Women Who Want No More Chi ldren ( including steri l ised) by Number of Liv ing Chi ldren and Selected Back- ground Characterist ics, GDHS, 1988 Number of Living Chi ldren Background Character ist ic 0 1 2 3 4 5 6+ Total RESIDENCE Urban O.0 I.I 12.1 18.1 39.8 57.4 67.3 28.0 Rural 0.8 0.5 4.4 11.2 18.9 31.6 60.5 20.6 REGION Wsstern 0.8 7.8 7.1 6.8 31.3 56.0 16.8 Central 1.6 6.0 10.4 26.3 42.9 57.4 22.5 Greater Accra 0.0 1.5 13.2 30.0 60.4 77.4 79.0 36.1 Eastern 1.3 6.0 16.4 33.8 59.5 68.1 28.8 Volta 0.0 9.6 20.3 35.6 47.5 77.3 29.5 Ashant i 2.6 1.0 4.8 12.3 20.0 37.3 66.2 24.3 Brong Ahafo 0.0 0.0 6.7 7.1 18.2 31.8 53.5 18.0 Upper West, East 0.0 0.0 1.3 0.0 7.0 10.0 31.0 7.0 and Northern LEVEL OF EDUCATION No educat ion 1.2 1.0 2.4 7.3 15.9 27.1 59.0 22.4 Pr imary 0.0 0.0 6.3 17.6 31.9 55.3 63.2 26.6 Middle 0.0 0.5 8.4 15.9 32.0 52.0 73.2 21.7 Higher - 1.9 18.4 28.6 22.5 TOTAL 0.5 0.7 6.8 13.3 25.1 39.4 62.3 22.8 - Fewer than 25 cases Overall, urban women are more likely to want to stop childbearing than their rural counterparts. Large differences exist at each parity level, with the proportions being consistently higher in urban areas. A currently married woman in the Greater Accra region is more likely to want to cease childbearing than her counterpart in other regions. On the other hand, women in Upper West, Upper East and Northern regions are least likely to make a voluntary decision to stop childbearing. Among women with 3 children, 30 percent in Greater Accra express a desire to cease childbearing, but it would require 6 or more living children for the same proportion of women in Upper West, Upper East and Northern regions to express a similar fertility desire. An inverse relationship between education and wanting more children is indicated in this table. With two living children, only 2 percent of uneducated women want to stop having children, compared to 18 percent of women with 11 or more years of schooling. The proportions increase as the number of living children increases. 54 5.2 NEED FOR FAMILY PLANNING Table 5.4 permits the examination of womens' need for family planning in order to space or limit future births, according m their intention to use contraception. Table 5.4 Percentage of Current ly Marr ied Women Who Are in Need of Family P lanning and the Percentage Who Are in Need But Who Intend to Use Contracept ion in the Future, by Selected Background Character ist ics, GDRS, 1988 In Need In Need and Intend to of Fami ly P lanning I Use Contracept ion Want Want to Want Want to Background No Postpone/ No Postpone/ Character ist ic More Undecided* Total More Undecided ~ Total Number RESIDENCE Urban 20.1 41.0 61.1 9.8 16.5 26.3 961 Rural 17.4 50.6 68.0 9.0 16.6 25.6 2195 REGION Western 13.3 53.0 66.3 7.2 17.6 24.7 279 Central 18.2 52.9 71.1 10.6 13.1 23.7 329 Greater Accra 22.8 29.4 52.2 10.3 15.6 25.8 368 Eastern 24.3 44.2 68.5 13.8 25.0 38.8 448 Volta 23.6 46.1 69.7 12.4 11.2 23.6 356 Ashanti 21.4 50.0 71.4 10.9 16.1 27.0 552 Brong Ahafo 15.0 52.1 67.1 7.2 21.2 28.4 481 Upper West, East 5.8 53.4 59.2 0.9 11.4 12.3 431 and Northern LEVEL OF EDUCATION No educat ion 19.3 47.6 66.9 8.0 10.6 18.6 1467 Primary 20.5 45.5 66.0 10.4 18.0 28.3 512 Middle 16.2 50.8 67.0 10.4 24.2 34.6 999 Higher 14.0 37.6 51.7 9.0 19.1 28.1 178 TOTAL 18.2 47.7 65.9 9.2 16.6 25.8 3156 * Includes women who are not contracept ing and who want no more births or want to postpone the next b i r th for 2 or more years. * Includes women who are undecided about whether to have another b i r th or about the t iming for the next birth. Eighteen percent of women in union want no more children and are not using contraception. These women are defined as "in need" of family planning. Only about half of them (9 percent) intend to use a method in the future. Similarly, 48 percent of the currently married women want to postpone the next birth or are uncertain about having another child, and are not using contraception. These women are in need of family planning for spacing purposes. Of these women, only a third intend to use a method in the future. Overall, 66 percent of currently married women have an unmet need for contraception, but only one-third of them intend to use a method in the future. Among currently married women, a higher proportion in rural areas than in urban areas want no more children or want to postpone the next birth and are not using contraception. As regards intention to use in the future, about the same proportion (26 percent) of currently married women in both urban and 55 rural areas indicated their intention to use family planning to either postpone or regulate their future fextility (Figure 5.2). Women with higher education are less likely than other women to be in need of family planning. Of those in need, women with the highest education are the most likely to plan to use a method in the future. 60 50 40 30 20 10 0 Figure 5.2 Future Need for Family Planning, Women in Union Not Using Contraception Percent Urban Rural Urban Rural I m lntend to use - i Want to postpone ~ Want no more Ghana DHS 1988 The regional distribution shows that between 52 and 71 percent of married women are in need of family planning and are not using contraception. Upper East, Upper West and Northern regions have the lowest proportion in need who intend to use, while Eastern region has the highest. 5.3 IDEAL FAMILY SIZE Respondents were asked to consider a hypothetical situation independent of their current family size and to declare the number of children they would choose to have if they could start their reproductive years again. About 13 percent of all women in the sample either did not know or gave non-numeric answers to the question on desired family size. The most common non-numeric response was "Up to God." Nine percent of women in the sample gave this reply. Childlessness is deplored in Ghanaian society, which is confirmed by the fact that only one in 1000 women desired no children, and these were mainly women with either no children or one child. Table 5.5 shows that a third of the women in the sample would prefer to have 4 children and 21 percent preferred to have 6 children. The mean ideal number of children is 5.3, which is lower than actual current fertility. The implication here is that, on average, women are having more children than they want. Indeed, the evidence from the table suggests that there is some unwanted fertility. Among 56 women with 6 or more children, only one-quarter report an ideal family size of 6 children, while as many as a third report lower ideal numbers of children. Table 5.5 Percentage Distribution of All Women by Ideal Number of Ch i ld ren and Mean Ideal Number of Ch i ldren for Al l WOmen and Cur rent ly Mar r ied Women, Accord ing to Number of L iv ing Chi ldren, GDHS, 1988 Number of Living Children* Ideal Number of Ch i ld ren 0 1 2 3 4 5 6+ Tota l 0 0.2 0.1 0.0 0.0 g.o 0.0 0.0 0.I 1 0.2 0.5 0.3 0.2 0.2 0.0 0.5 0.3 2 6.8 3.5 2.7 1.3 1.8 3.1 1.1 3.3 3 11.2 12.5 6.0 4.4 3.6 2.1 3.5 7.1 4 34.9 40.0 39.5 35.6 23.6 24.4 22.1 32.5 5 11.3 8.8 12.1 11.8 7.9 7.1 6.1 9.5 6 15.5 16.9 17.8 25.6 32.3 23.4 25.0 21.2 7 I.i 1.1 0.5 1.3 1.4 2.4 4.1 1.6 8+ 6.0 7.7 11.4 10.0 14.5 I8.6 18.9 11.5 Have not thought of it 2.8 0.9 1.3 0.7 1.6 2.9 2.7 1.9 Up to God 7.3 6.8 6.3 6.5 10.5 13.4 13.6 8.8 AS many as can care for 1.0 0.7 1.0 I.I 0.8 0.8 i.I 0.9 Don't know 1.4 0.4 0.8 1.5 1.6 1.8 1.4 1.2 Missing 0.4 0.1 0.3 0.0 0.2 0.0 0.2 0.2 TOTAL I00.0 I00.0 I00.0 100.0 i00.0 I00.0 100.O I00.0 Number 1018.0 752.0 690.0 550.0 496.0 381.0 661.0 4486.0 Mean Ideal Number* All Women 4.6 4.8 5.2 5.4 5.7 6.0 6.1 5.3 Number 887 685 569 496 423 309 536 3905 Current ly Mar r ied Women 5.3 5.0 5.2 5.4 5.8 6.g 6.1 5.5 Number 162 498 511 433 380 271 481 2736 * Including current pregnancy. Exc ludes women who gave non-numer lc responses. Among all women in the survey, the mean ideal family size increases from 4.6 children for childless women to 6.1 children desired by women with 6 or more children. When only currently married women am considered, the mean ideal family size increases from 5.3 children for childless women to 6.1 children for those with 6 or more children surviving. The overall ideal family size for currently married women is 5.5 children. This is only 4 percent larger than for all women in the sample. In the Ghana Fertility Survey of 1979/80, the mean ideal family size of women currently in union was found to be 6.1 children. There are a variety of reasons why women with larger families declare a higher ideal family size. First, women with large families may genuinely desire more children than women with smaller families. Secondly, women with more children are likely to be older than women with fewer children. Their ideal family sizes may, therefore, reflect more traditional views. Finally, women may tend to rationalize the births they already have and, consequently, are unlikely to state a number that is less than their achieved family size. It is not easy distinguishing among the factors that influence the ideal family size of high- parity women. 57 Evidence in Table 5.6 clearly shows that older women do, indeed, prefer larger families than younger women. The ideal number of children increases from 4.7 for women aged 15-19 years to 6.5 for women aged 45-49. It can be inferred from this that if young women have only the number of children they want, fertility rates in the future can be expected to decline. Table 5.6 Mean Ideal Number of Chi ldren for All Women by Age and Selected Background Character ist ics, GDHS, 1988 Age Background Character is t ic 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Total RESIDENCE Urban 4.4 4.4 4.5 4.9 4.9 5.4 5.8 4.7 Rural 4.9 5.0 5.5 5.8 6.1 6.4 6.9 5.6 REGION Western 5.0 4.8 5.1 5.5 5.3 6.0 5.3 Central 4.6 4.6 4.9 5.2 5.5 5.5 5.0 Greater Accra 4.0 4.1 4.5 4.9 5.0 5.4 5.9 4.6 Eastern 4.6 4.4 4.8 5.1 5.1 5.3 5.9 4.9 Volta 4.1 4.5 4.5 5.0 5.4 6.5 6.1 4.8 Ashant i 4.8 4.8 4.7 5.0 5.2 5.1 6.4 5.0 Brong Ahafo 5.2 4.8 5.3 5.7 5.5 6.2 5.4 Upper West, East 6.2 6.8 8.7 8.5 9.5 8.8 10.3 8.2 and Northern LEVEL OF EDUCATION No educat ion 6.0 5.6 6.4 6.4 6.4 6.6 6.9 6.4 Pr imary 4.6 4.8 5.0 5.5 5.4 5.5 6.2 5.1 Middle 4.4 4.4 4.6 4.9 5.0 5.1 5.3 4.6 Higher 4.2 3.8 3.9 3.9 4.2 4.0 TOTAL 4.7 4.7 5.2 5.5 5.7 6.0 6.5 5.3 - Fewer than 25 cases As expected, not only is the mean ideal family size larger for rural women than for urban women, it is also larger for uneducated women compared to educated women. Differences in ideal family size by education level may be attributed partly to the fact that women with low levels of education tend to be older and of higher parity. But it is interesting to note that, within all levels of education, ideal family size increases with age. Differentials by region reveal that Greater Accra has the smallest ideal family size (4.6 children), with Upper West, Upper East and Northern regions having the largest ideal family size (8.2 children). 5.4 FERTILITY PLANNING Table 5.7 presents information on whether births in the last 12 months were planned, wanted later, or not wanted at all. Caution should be used regarding the data in this table, because the possibility exists that women with unplanned births will tend to rationalize such births. Overall, two-thirds of births in the last 12 months were wanted, while 30 percent were wanted later and 4 percent were unwanted. First and second order births were more likely to have been planned (71 percent) than third or higher 58 order births (62 percent). Conversely, less than 1 percent of lower order births were unwanted, compared with 6 percent of higher order births. DHS surveys in Burundi and Mall found that, of births in th 12 months preceding the surveys, unwanted births constituted 5.5 percent and 3.5 percent, respectively (Traore et al, 1989; Segamba L., et at, 1988). Table 5.7 Percentage of Women Who Had a Birth in the Last 12 Months by Ferti l ity Planning Status and Birth Order, GHDS, 1988 Birth Order * Planning Status 1-2 3+ Total Wanted child then 71.1 62.3 65.6 Wanted child later 28.3 31.1 30.0 Child not wanted 0.6 6.3 4.2 Not classif iable 0.0 0.4 0.2 TOTAL i00.0 i00.0 i00.0 Number 336 570 906 * Includes current pregnancy 59 CHAPTER 6 MORTALITY AND HEALTH This chapter presents estimates of infant and child mortality and selected indicators of maternal and child health. Current mortality estimates are important for the construction of population projections. Mortality and other health indicators are also useful for identifying sectors of the population at high risk, as well as for evaluating health programmes. Information on trends and differentials in mortality is presented first. This is followed by a presentation of selected maternal and child health indicators, such as prenatal care, assistance during delivery and childhood immunisations. Finally, the nutritional status of Ghanaian children aged 3-36 months is described through the use of anthropometric measurements. 6.1 MORTALITY DATA All female respondents in the GDHS were asked to provide a complete birth history, including the sex, date of birth, survival status, and current age or age at death of each live birth. The data obtained from the birth histories are used to calculate directly infant and childhood mortality rates. Mortality rates are presented for three age intervals and three time periods. The infant mortality rate (~ q 0) is the probability of dying between birth and exact age one. Childhood mortality (~ q ~) is the probability of dying between exact age one and exact age five, and under five mortality (5 q 0) is the probability of dying between birth and exact age five. Each of these rates is presented for three, five-year tune periods: 1973-77, 1978-82, and 1983-87. The 1983-87 rate includes information from the months in 1988 which preceded the interview (between 2 and 5 months for individual respondems). The reliability of mortality estimates calculated from retrospective birth histories depends upon the completeness with which deaths of children are reported and the extent to which birth dates and ages at death are accurately reported. While a complete evaluation of the quality of the mortality data from the GDHS has not been attempted here, some basic quality checks of the data are presented below. Underreporting of infant deaths is usually most severe for deaths which occur very early in infancy. The problem is rooted in cultural tradition. In Ghana, infants are traditionally considered "visitors" until they have survived some minimum period of time. Consequently, society imposes restrictions on discussion of children who die very early in infancy. A standard procedure for testing for underreporting of early infant deaths involves forming the ratio of deaths in the first week of life to deaths in the first month. These ratios are shown in Table 6.1. While the expected value of this ratio is not known, it is known that mortality declines throughout infancy, that the value of the ratio would be expected to increase as the overall level of mortality decreases and that a ratio of less than 0.25 would indicate severe underreporting of early infant deaths. As shown in Table 6.1, the ratio of deaths in the first week to all deaths in the first month is greater than 0.70 in all periods for both sexes combined, suggesting that early infant deaths are not severely underreported. In addition, the data do not indicate systematic underreporting of deaths of one sex compared to the other. Another indication of underreporting of early infant deaths would be a low rate of neonatal mortality relative to infant mortality. Table 6.1 shows neonatal mortality as a proportion of total infant mortality. These proportions range from 0.57 to 0.63 for the three periods under consideration and are 61 within the range of values calculated with WFS data for countries at similar levels of infant mortality (Rutstein, 1984). In the GDHS, if a child death was reported to have occurred within a month of birth, the age at death was to be elicited in days and recorded in days. If the child died within 24 months, the age at death would be recorded in units of months. If the child was 2 years old or older at death, age at death in years was recorded. In general, this protocol was followed well, with one notable exception. There were many child deaths apparently occurring during the 12-23 month age segment whose ages at death were recorded as "one year." Table 6.2 shows the distribution of Table 6.1 Indices for Detect ing Underreport ing of Infant Deaths, GDHS, 1988 Index 1983-87' 1978-82 1973-77 Deaths first week/ Deaths first month Total .77 .70 .75 Male .77 .65 .83 Female .77 .77 .65 Neonatal Mortal i ty/ Infant Morta l i ty .57 .60 .63 * Includes deaths and exposure during 1988 up to the month preceding the interview deaths between 6 and 24 months of age and the unit of age used to record the age at death. The distribution of deaths around 12 months suggests that the bulk of deaths reported at "one year" indeed occurred during the 12-23 age segment and probably should not be considered as infant deaths. There is, however, a slight deficit of deaths at 10 and 11 months, suggesting that a small number of deaths recorded at "1 year" or 12 months of age, in fact, occurred earlier. If so, the infant mortality rates presented below would be biased slightly downwards and childhood rates slightly upwards. There appears no striking trend in the extent of this problem across calendar time. It may be noted that, although the problem may not cause serious problems in the estimation of conventional demographic parameters, the age pattern of mortality in the second year of life cannot be recovered without enormous guesswork. Unreported age at death is also a potential problem. In the GDHS data, however, of 2436 deaths reported by the respondents, there are only 3 cases in which age at death was not given, of which only 1 occurred to a child bom in the 15 years prior to the survey. For purposes of this report, the missing data have been imputed using a hot-deck procedure) 6.2 MORTALITY TRENDS Infant and childhood mortaltiy rates for three five-year periods prior to the survey are shown in Table 6.3 and Figure 6.1. These rates demonstrate a clear and marked decline in infant and childhood mortality in Ghana since the mid-1970s. For example, the infant mortality rate, estimated at around 100 deaths per 1000 births in 1973-77, has declined by approximately 22 percent to 77 deaths per 1000 births in the most recent five-year period. Mortality during childhood has also declined during the period under consideration. The probability of dying between birth and age five (sq0) has dropped from 187 in 1973-77 to 155 in 1983-88. The apparent slight increase in ,q l and 5qo from 1978-82 to 1983-87 is probably due to the more severe heaping of age at death on 12 months in the recent period, which would cause some deaths to be attributed to childhood when they actually occurred during infancy. Thus, the decline in infant mortality may be somewhat overstated, and the decline in childhood mortality may be understated. t The procedure assigns the child an age at death which is the same as the last death in the data f'de of the same birth order. 62 Table 6.2 Distr ibut ion of Chi ld Deaths Occurr ing Between 6 and 24 Months of Age by Calendar Per iod and Age at Death When Reported in Months and Years, GDHS, 1988 1983-87' 1978-82 1973-77 Age at Death Mos. Yrs. Mos. Yrs. Mos. Yrs. 6 15 14 9 7 7 14 7 8 22 II 8 9 17 I0 6 I0 5 5 3 ii 4 2 3 12 (I yr.) 8 (79) 3 (57) 6 (56) 13 0 0 0 14 1 1 0 15 1 2 3 16 4 0 0 17 2 2 0 18 6 12 9 19 0 - 1 0 20 2 0 3 21 0 0 2 22 0 0 0 23 1 0 0 24 (2 yrs.) 1 103 1 (71) 0 (57) * Includes deaths dur ing 1988 up to the month preceding the interview Tab le 6.3 In fant and Ch i ldhoood Mor ta l i ty for Ca lendar Per iods , GDHS, 1988 In fant Ch i ldhood Both Per iod (~ q Q) (4 q*) (, q 0) 1983-1987" 77.2 84.0 154.7 1978-1982 86.4 72.3 152.4 1973-1977 99.6 97.3 187.2 * Inc ludes exposure dur ing 1988 up to the month preced ing the in terv iew. 63 250 200 150 100 50 Figure 6.1 Trends in Infant and Child Mortality Deaths per 1,000 100 97 187 0 Infant Mortal i ty Child Mortal i ty Under 5 Mortality m 1973-77 ~ 1978-82 ~ 1983-87 | Ghana DHS 1988 The Ghana Fertility Survey (GFS), conducted in 1979-80, reported an infant mortality rate of 71 and an under-five mortality rate (2 q o) of 121 for the five years preceding the survey (approximately 1975- 80). Based on a data-quality analysis of the GFS, however, these rates were judged to be an underestimate of true mortality levels during that period (Owusu, 1984). As shown in Table 6.3, GDHS estimates of mortality from approximately the same period (1978-82) are significantly higher than the GFS estimates. 6.3 MORTALITY DIFFERENTIALS In this section, we present infant and childhood mortality rates according to the socioeconomic background of the mother and various demographic characteristics of children and mothers. The rates are calculated for the ten-year period from 1978-19872 in order to ensure that there are enough cases to calculate rates for sub-groups. In Table 6.4, infant and childhood mortality rates are shown by urban-rural residence, the mother's level of education, and region. Both infant and child mortality are higher in rural than in urban areas. During the 1978-87 period, approximately 67 in 1000 children in urban areas and 87 in 1000 children in rural areas died before reaching their first birthday. The under-five mortality rate is about 24 percent higher in rural than in urban areas. Regional differences in mortality rates are also quite significant. The infant mortality rate ranges from 58 in Greater Accra to 138 in the Central region. Upper West, Upper East and Northern regions also have a high infant mortality rate at 103. In the remaining regions, infant mortality is in the 65-75 range. While the infant mortality rate is highest in the Central region, the childhood mortality rate in this region 2 These rates include deaths and exposure in 1988 up to the month preceding the interview. 64 Table 6.4 Infant and Chi ldhood Morta l i ty by Socioeconomic Characterist ics, GDHS, 1988 Infant Chi ldhood Both Background (x q ,) (, q i) (, q ,) Character ist ic 1978-87 1978-87 1978-87 RESIDENCE Urban 66.9 68.8 131.1 Rural 86.8 82.9 162.5 LEVEL OF EDUCATION No educat ion 87.7 95.2 174.6 Primary 84.8 68.5 147.6 Middle 69.7 64.0 129.2 Higher 79.3 22.2 99.5 REGION Western 76.9 80.4 151.2 Central 138.3 81.9 208.8 Greater Accra 57.7 48.9 103.8 Eastern 70.1 73.2 138.1 Volta 73.5 63.8 132.7 Ashanti 69.8 80.0 144.2 Brong Ahafo 65.0 61.6 122.6 Upper West, East 109.1 132.3 221.8 and Northern TOTAL 81.3 78.9 153.8 * Note: Rates include exposure dur ing 1988 up to the month preceding the interview. is similar to that found in most of the other regions, with the exception of Greater Accra, which has a much lower rate, and Upper West, Upper East and Northern regions, which have a very high rate. This pattem of regional differences in infant and child mortality is consistent with the pattern found in an analysis of GFS data in which the infant mortality rate for the Westem and Central regions combined was approximately twice as high as that found in all of the other regions, except Northern and Upper regions (Adansi-Pipim, 1985). The high level of infant mortality in the Westem/Central region was attributed to an unusually high neonatal mortality rate. The GDHS data suggest that infant mortality in the Central region alone is uniquely high, while mortality in the Western region is closer to that found in the Volta and Eastern regions. As will be shown later in this chapter, the incidence of diarrhoea and, especially, fever among children under age five is relatively high in the Central region, compared to other regions. The education of the mother is strongly associated with a child's chances of survival. Infant mortality generally declines with increasing education of the mother. Differences in infant mortality according to mother's education are not as large, however, as the differences in child mortality. The probability of dying between age one and age five is more than four tunes greater for children of mothers with no education than for children whose mothers have more than middle school education. The children of mothers with primary school education are 40 percent less likely to die between age one and five than children of uneducated mothers. 65 Table 6.5 presents mortality rates according to demographic characteristics of mothers and children. As expected, male infant mortality is higher than female infant mortality but mortality rates after the first year are approximately equal for both sexes. Table 6.5 Infant and Chi ldhood Morta l i ty by Demographic Characterist ics, GDHS, 1988 Infant Chi ldhood Both Background (* q 0) (, q ~) (5 q o) Character is t ic 1978-87 1978-87 1978-87 SEX OF CHILD Male 88.8 78.3 160.2 Female 73.5 79.4 147.1 MOTHER'S AGE AT BIRTH Less than 20 97.0 94.5 182.3 20-29 73.1 80.1 147.3 30-39 82.8 68.7 143.0 40-49 118.6 89.2 197.2 BIRTH ORDER First 86.3 81.8 161.1 2-3 67.9 84.7 146.8 4-6 82.6 79.8 155.9 7+ 101.8 57.9 153.8 PREVIOUS BIRTH INTERVAL <2 years 114.6 87.2 191.7 2-3 years 67.7 79.7 141.9 4 years or more 51.5 58.9 I07.4 * Note: Rates include exposure during 1988 up to the month prsceding the interview. The relationship between mother's age at the time of the birth and mortality is curvilinear. Children of teenage mothers have higher infant and child mortality than children of mothers in their twenties and thirties. The probability of dying increases for children of mothers more than 40 years old. Birth order differences in infant mortality are as expected. The infant mortality rate for first births is somewhat higher than that of second through sixth order births, and seventh and higher order births have dramatically higher rates of infant mortality than lower order births. Perhaps the most striking mortality differentials are those correlated with the length of the preceding birth interval. Children born within 2 years of a preceding birth are more than twice as likely to die during the first year of life as children born 4 or more years after a preceding birth. Higher mortality risks for children bom foUowing short intervals continue after the first year of life; the childhood mortality rate is 1.5 times higher for children born within 2 years of a preceding birth compared to those born 4 or more years after the preceding birth. 66 Another perspective on mortality during childhood in Ghana is offered by the data presented in Table 6.6. This table presents the mean number of children ever born and, of those ever born, the mean number surviving and dead at the time of the interview, by age of the mother. The proportion dead among those ever born increases with the age of the mother, as expected. Women aged 15-49 at the time of the survey had given birth to an average of 3.2 children. Of these, 2.6 children were surviving and approximately 0.5 (17 percent) were dead. Table 6.6 Mean Number of Chi ldren Evsr Born, Surviving, and Dead, and Proport ion of Chi ldren Dead Among Ever Born by Age of Mother, GDHS, 1988 Mean Number of Chi ldren Proport ion Dead Among Age of Ever Chi ldren Mother Born Surviv ing Dead Ever Born 15-19 0.22 0.20 0.01 0.07 20-24 1.25 1.08 0.17 0.14 25-29 2.65 2.26 0.39 0.15 30-34 4.18 3.50 0.68 0.16 35-39 5.47 4.57 0.90 0.16 40-44 6.58 5.39 1.19 0.18 45-49 ?.25 5.65 1.60 0.22 TOTAL 3.17 2.62 0.54 0.17 6.4 PRENATAL CARE AND DELIVERY ASSISTANCE Maternal and child health care is one of the priority areas addressed by the Primary Health Care (PHC) system in Ghana. As part of the PHC system the Ministry of Health has, in recent years, administered a programme in which traditional birth attendants are trained in pre- and postnatal care, as well as in child health care and family planning (Adjei, et. al., 1988). In the GDHS, women who had given birth in the five years prior to the survey were asked a series of questions concerning the type of health care they received prior to each birth during this period. Respondents were asked whether or not they had seen anyone for a check on the pregnancy. If they reported that they had seen someone, then they were asked who had provided the care. Women were also asked who assisted with the delivery of each birth. For both prenatal care and assistance at delivery, interviewers were instructed to record the most qualified person in cases in which more than one type of person provided care. Since neonatal tetanus is known to be an important cause of infant death in many developing countries, female respondents in the GDHS were also asked whether

View the publication