Sierra Leone - Multiple Indicator Cluster Survey - 2000

Publication date: 2000

THE STATUS OF WOMEN AND CHILDREN IN SIERRA LEONE A HOUSEHOLD SURVEY REPORT (MICS-2) undertaken by the GOVERNMENT OF SIERRA LEONE with technical and financial support from UNICEF, UNFPA, UNHCR, ACTION AID, CCF, CHASL. CENTRAL STATISTICS OFFICE MINISTRY OF DEVELOPMENT AND ECONOMIC PLANNING Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 1 Foreword It is with joy and satisfaction that my Ministry, in collaboration with our development partners, presents this national report for the 2000 Sierra Leone Multiple Indicator Cluster Survey (MICS2) – a nationally representative survey of households, women and children. The 1990 World Summit for Children Plan of Action set end-decade goals for 2000. This Plan required national efforts by government and its partners to ensure a high quality of life for our children in all spheres of their development. The Plan also sought to improve the quality of the health and education of the children of Sierra Leone and increase their opportunity to develop their unique abilities. Achievement of this laudable plan was to be monitored and assessed through socio-economic indicators in the area of health, education, and child protection. In preparing this report for Sierra Leone, the goals set for the World Summit for Children in 1990 were assessed and the present status of the socio-economic indicators that were recommended evaluated. The information generated in this report should therefore assist government and other development partners to plan for and develop meaningful development programmes that will ensure the survival, development, and protection of children in the coming decade. Sierra Leone occupies an unenviable position as reflected in the 2000 Global Human Development Report. That report was based on data that—accurate as it may have been—was badly out-of-date. Information in this report should be regarded as an accurate presentation of the current situation in Sierra Leone, considering the comprehensive nature in which the survey was planned and implemented. The Central Statistics Office (CSO) within the Ministry of Development and Economic Planning has been responsible for the collection, compilation and analysis of the data in this report. While CSO served as the technical arm of the ministry for the production of this report, administrative and technical support was provided by the MICS2 Task Force. The Task Force drew members from all ministries, UNICEF, UNFPA, UNHCR, FAO, CCF, Action Aid (S.L) and CHASL. UNICEF provided extensive technical support to the production of this report through its external consultant and national field staff. Useful comments and suggestions were also received from our development partners who read the preliminary report. Nevertheless, the Ministry of Development and Economic Planning assumes full responsibility for the contents of this report. Once again, I do believe that users of this document will find information that will help them plan and develop programmes and formulate policies for the advancement of women, children, and other vulnerable members of our society. Kadi Sesay, Ph.D. Minister of Development and Economic Planning November, 2000 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 2 Acknowledgements The people of Sierra Leone are acknowledged for the time they spent providing information about themselves and the warm hospitality that they showed to data collection personnel. The enumerators and supervisors are acknowledged for their dedication and hard work in the face of difficult and potentially dangerous conditions in the field. The 2000 Sierra Leone Multiple Indicator Cluster Survey—2 (MICS2) received the backing of many institutions and their dedicated personnel. The following organizations are acknowledged for the financial, logistical and/or technical support that they provided to date to the MICS2 survey effort. Government of Sierra Leone Central Statistics Office Ministry of Development and Economic Planning Ministry of Health and Sanitation Ministry of Information and Broadcasting Ministry of Local Government and Rural Development Ministry of Safety and Security Ministry of Social Welfare, Gender and Children’s Affairs Ministry of Youth, Education and Sports United Nations Agencies UNICEF Sierra Leone UNFPA UNHCR Non-governmental organizations Action Aid - Sierra Leone Christian Children’s Fund Christian Health Association of Sierra Leone Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 3 Table of Contents Foreword…………………………………………………………………………………………1 Acknowledgements……………………………………………………………………………….2 Table of Contents………………………………………………………………………………….3 List of Figures…………………………………………………………………………………….5 List of Tables………………………………………………………………………………………6 List of Abbreviations………………………………………………………………………………9 Executive Summary …………………………………………………………………………10 Summary Indicators ………………………………………………………………………….15 I. Introduction…………………………………………. ……………………………………….17 Background of the MICS2 Survey…………………………………………………………….17 Sierra Leone Background: Independence, conflict and consequences……………………….18 Survey Objectives.18 II. Survey Methodology……………………………………………………………………… …18 Background. 18 Sample Design.19 Questionnaires .19 Training and Fieldwork.21 Data collection and supervision.21 Survey personnel. 21 Data processing and management . 22 Limitations of the survey…………………………………………. 22 III. Sample Characteristics and Data Quality………………………………….………………….23 Response Rates.23 Age Distribution and Missing Data .23 Characteristics of the Household Population .25 IV. Results………………………………………………………………………………….……26 A. Infant and Under-Five Mortality.26 B. Education.28 Early childhood education. 29 Basic education .29 Literacy.30 C. Water and Sanitation.31 Use of drinking water……….31 Use of sanitation.33 D. Child Malnutrition .34 Nutritional status .34 Breastfeeding .36 Salt iodization.39 Vitamin A supplementation.40 Low birth weight .41 E. Child Health .43 Immunization Coverage .43 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 4 Acute respiratory infection .47 IMCI initiative . 48 Malaria .50 F. HIV/AIDS.51 AIDS knowledge.52 AIDS testing .54 G. Reproductive Health.55 Contraception.55 Prenatal care .56 Assistance at delivery .57 H. Child Rights.59 Birth registration .59 Orphanhood and living arrangements of children.59 Child labor .60 I. Maternal Mortality .61 J. Disability . 64 Appendices……………………………………………………………………………………….65 Appendix A: Survey costs……………………………………………….……………………….66 Appendix B: List of personnel involved in Sierra Leone MICS2……………………….……….67 Appendix C: Sample design………………………………………………………….………….70 Appendix D: Questionnaires………………………………………………………….………….73 Appendix E: Training schedule for enumerators…………………………………………………98 Appendix F: Tables of results…………………………………………………………………….99 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 5 List of Figures Figure 1: Single year age distribution of the household population by sex, Sierra Leone, 2000…………………………………………………………………………….24 Figure 2: Estimates of infant and under five mortality based on indirect estimation, Sierra Leone, 2000 ……………………………………………………………….….24 Figure 3: Infant mortality rate and under five mortality rate in Sierra Leone over time….26 Figure 4: Percentage of children of primary school age attending primary school, Sierra Leone, 2000…………………………………………………………………….27 Figure 5: Percent distribution of the population in Sierra Leone by source of drinking water, Sierra Leone, 2000……………………………………………………….30 Figure 6: Percentage of under-five children in Sierra Leone who are undernourished, Sierra Leone, 2000…………………………………………………………………….32 Figure 7: Percent distribution of living children under two years by breastfeeding status, Sierra Leone, 2000………………………………………………………………38 Figure 8: Percentage of children in Sierra Leone aged 12-23 months who received vaccinations by the time of the survey, Sierra Leone, 2000…………………….44 Figure 9: Percentage of women aged 15-49 who have sufficient knowledge of HIV/AIDS transmission by level of education, Sierra Leone, 2000…………….…………54 Figure 10: Percent distribution of women with a birth in the last year by type of personnel delivering antenatal care, Sierra Leone, 2000………………………………….57 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 6 List of Tables Table 1: Number of households and women, and response rates, Sierra Leone, 2000 Table 2: Age group distribution of household population by sex, Sierra Leone, 2000 Table 3a: Percentage of cases missing information for selected questions, Sierra Leone, 2000 Table 3b: Percentage of cases missing information for selected questions, Sierra Leone, 2000 Table 3c: Percentage of cases missing information for selected questions, Sierra Leone, 2000 Table 4: Percent distribution of households by background characteristics, Sierra Leone, 2000 Table 4a: Percent distribution of households by background characteristics, Sierra Leone, 2000 Table 5: Percent distribution of women 15-49 by background characteristics, Sierra Leone, 2000 Table 6: Percent distribution of children under 5 by background characteristics, Sierra Leone, 2000 Table 7: Mean number of children ever born (CEB) and proportion dead by mother's age, Sierra Leone, 2000 Table 9: Percentage of children aged 36-49 months who are attending some form of organized early childhood education program, Sierra Leone, 2000 Table 10: Percentage of children entering first grade of primary school who eventually reach grade 5, Sierra Leone, 2000 Table 11: Percentage of children of primary school age attending primary school, Sierra Leone, 2000 Table 12: Percentage of the population aged 15 years and older that is literate, Sierra Leone, 2000 Table 13: Percentage of the population with access to safe drinking water, Sierra Leone, 2000 Table 14: Percentage of the population with access to sanitary means of excreta disposal, Sierra Leone, 2000 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 7 Table 15: Percentage of under-five children with missing height or weight, Sierra Leone, 2000 Table 15a: Percentage of under-five children who are severely or moderately undernourished, Sierra Leone, 2000 Table 16: Percent of living children by breastfeeding status, Sierra Leone, 2000 Table 17: Percentage of households consuming adequately iodized salt, Sierra Leone, 2000 Table 18: Percent distribution of children aged 6-59 months by whether they received a high dose of Vitamin A supplement in the last 6 months, Sierra Leone, 2000 Table 19: Percentage of women with a birth in the last 12 months by whether they received a high dose of Vitamin A supplement before the infant was 8 weeks old, Sierra Leone, 2000 Table 20: Percentage of live births in the last 12 months that weighed below 2500 grams at birth, Sierra Leone, 2000 Table 20a: Percentage of live births in the last 12 months that weighed below 2500 grams at birth, Sierra Leone, Table 21: Percentage of children 12-23 months immunized against childhood diseases at any time before the survey, Sierra Leone, 2000 Table 22: Percentage of children age 12-23 months currently vaccinated against childhood diseases, Sierra Leone, 2000 Table 23: Percentage of under-five children with diarrhea in the last two weeks and treatment with ORS or ORT, Sierra Leone, 2000 Table 24: Percentage of under-five children with diarrhea in the last two weeks who took increased fluids and continued to feed during the episode, Sierra Leone, 2000 Table 25: Percentage of under-five children with acute respiratory infection in the last two weeks and treatment by health providers, Sierra Leone, 2000 Table 26: Percentage of children 0-59 months of age reported ill during the last two weeks who received increased fluids and continued feeding, Sierra Leone, 2000 Table 27: Percentage of caretakers of children 0-59 months who know at least 2 signs for seeking care immediately, Sierra Leone, 2000 Table 28: Percentage of children 0-59 months of age who slept under an insecticide- impregnated bednet during the previous night, Sierra Leone, 2000 Table 29: Percentage of children 0-59 months of age who were ill with fever in the last two weeks who received anti-malarial drugs, Sierra Leone, 2000 Table 30: Percentage of women aged 15-49 who know the main ways of preventing HIV transmission, Sierra Leone, 2000 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 8 Table 31: Percentage of women aged 15-49 who correctly identify misconceptions about HIV/AIDS, Sierra Leone, 2000 Table 32: Percentage of women aged 15-49 who correctly identify means of HIV transmission from mother to child, Sierra Leone, 2000 Table 33: Percentage of women aged 15-49 who express a discriminatory attitude towards people with HIV/AIDS, Sierra Leone, 2000 Table 34: Percentage of women aged 15-49 who have sufficient knowledge of HIV/AIDS transmission, Sierra Leone, 2000 Table 35: Percentage of women aged 15-49 who know where to get an AIDS test and who have been tested, Sierra Leone, 2000 Table 36a: Percentage of Sexually active women aged 15-49 who are using (or whose partner is using) a contraceptive method, Sierra Leone, 2000 Table 36: Percentage of married or in union women aged 15-49 who are using (or whose partner is using) a contraceptive method, Sierra Leone, 2000 Table 37: Percentage of mothers with a birth in the last 12 months protected against neonatal tetanus, Sierra Leone, 2000 Table 38: Percent distribution of women aged 15-49 with a birth in the last year by type of personnel delivering antenatal care, Sierra Leone, 2000 Table 39: Percent distribution of women aged 15-49 with a birth in the last year by type of personnel assisting at delivery Sierra Leone, 2000 Table 40: Percent distribution of children aged 0-59 months by whether birth is registered and reasons for non-registration, Sierra Leone, 2000 Table 41: Percentage of children 0-14 years of age in households not living with a biological parent, Sierra Leone, 2000 Table 42: Percentage of children 5-14 years of age who are currently working, Sierra Leone, 2000 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 9 List of Abbreviations AFRC Armed Forces Revolutionary Council ARI Acute Respiratory Infection CCP Commission for the Consolidation of Peace CSO Central Statistics Office DPT Diphtheria, Pertussis and Tetanus EA Enumeration Area HIV Human Immunodeficiency Virus IDP Internally Displace Person IMCI Integrated Management of Childhood Illnesses IMR Infant Mortality Rate MICS2 Multiple Indicator Cluster Survey–2 MMR Maternal Mortality Rate NGO Non-governmental Organization NID National Immunization Day OPV Oral Polio Vaccine ORS Oral Rehydration Solution ORT Oral Rehydration Therapy RHF Recommended Home Fluid RUF Revolutionary United Front RUFP Revolutionary United Front Party SOWCR State of the World’s Children Report (UNICEF publication) U5MR Under-5 Mortality Rate UNFPA United Nations Population Fund UNHCR United Nations High Commission For Refugees UNICEF United Nations Children’s Fund VAD Vitamin A Deficiency VAS Vitamin A Supplementation WHO World Health Organization WSC World Summit for Children Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 10 Executive Summary The 2000 Sierra Leone Multiple Indicator Cluster Survey—2 (MICS2) is a nationally representative survey of households, women, and children. The main objectives of the survey are to provide up-to-date information for assessing the situation of children and women in Sierra Leone at the end of the decade and to furnish data needed for monitoring progress toward goals established at the World Summit for Children and as a basis for future action. Overall findings · This report paints a portrait of a country in crisis. Following the publication of results from MICS2 surveys across the world, the rates of infant, under-five and maternal mortality in Sierra Leone will likely continue to be among the highest in the world. The values of many other indicators of health and educational status reported below clearly show the desperate social and living conditions faced by most Sierra Leoneans. · The problems in Sierra Leone cut across sectors. Education, health, and child rights are all in need of urgent action. · Problems in Sierra Leone can be acted upon and programs can achieve success. National immunization campaigns have achieved high coverage in the face of overwhelming logistical and political constraints. Communication programs have made their mark; caretakers have high knowledge levels regarding when to seek medical treatment for their children’s illness, and many make use of oral rehydration solution when their children fall ill with diarrhea. · The Northern region of Sierra Leone—decimated by the conflict of the past decade and off- limits to almost all aid initiatives—clearly stands out as the area of the republic in greatest need of support. The population of this region experiences the greatest morbidity, has the least access to resources, and has the lowest knowledge levels in the country. · Discrepancies in the opportunities available to boys and girls do not appear to be pervasive in Sierra Leone for most issues. Roughly equal proportions of girls and boys attend schools, and treatment of disease does not vary by gender. Rural-urban divisions are apparent, however, with urban populations enjoying greater services and displaying greater knowledge on most issues as compared to rural populations. · Many problems in Sierra Leone are regional in nature. Values for many indicators in Sierra Leone are in line with indicator levels in neighboring West African countries and in Sub- Saharan Africa as a whole. Sample Characteristics · 4021 out of 4207 households during data collection were found to be occupied and were successfully surveyed, yielding a household response rate of 97. In the interviewed households, 5750 eligible women (age 15-49) were identified. Of these, 4923 were successfully interviewed, yielding a response rate of 85.6 percent. In addition, 3060 children under age five were listed in the household questionnaires of surveyed households. Questionnaires were completed for 2764 of these children for a response rate of 90.3 percent. Infant and Under-five Mortality · The methodology used to calculate these indicators in this survey for Sierra Leone used an indirect technique to provide estimates for the infant mortality rate (IMR) and under-five mortality rate (U5MR) for 1998. The data suggest that the IMR was 170 per 1000 and the U5MR 286 per 1000 in 1998. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 11 Education · Forty-two percent of children of primary school age in Sierra Leone are attending primary school. School attendance in the North and East is lower than in the rest of the country at 28 and 35 percent, respectively. At the national level, there is little difference between male (44 percent) and female (40 percent) primary school attendance. · Eighty-five percent of children who enter the first grade of primary school eventually reach grade five. · Thirty percent of the population over age 15 years is literate. The percentage literate declines from 38 percent among those aged 15-24 years to 15 percent among the population aged 65 years and older. Water and Sanitation · Fifty-four percent of the population has access to safe drinking water—74 percent in urban areas and 46 percent in rural areas. The situation in the North with regard to this indicator is considerably worse than in other regions; only 30 percent of the population there gets its drinking water from a safe source. · Sixty-three percent of the population of Sierra Leone is living in households with sanitary means of excreta disposal. Child Malnutrition · Twenty-seven percent of children under age five in Sierra Leone are underweight or too thin for their age. Thirty-four percent of children are stunted or too short for their age and ten percent are wasted or too thin for their height. Breastfeeding · Only 2 percent of children aged under four months are exclusively breastfed, an extremely low level. At age 6-9 months, 52 percent of children are receiving breast milk and solid or semi-solid foods. At age 20-23 months, 51 percent of children continue to breastfeed. Salt Iodization · Twenty-three percent of households in Sierra Leone have adequately (15+ PPM) iodized salt. The percentage of households with adequately iodized salt ranges from 5 percent in the North (where local mining of salt is prevalent) to 47 percent in the Eastern province (where no mining of salt occurs). Vitamin A Supplementation · Within the six months prior to the MICS, 58 percent of children aged 6-59 months received a high dose Vitamin A supplement. Approximately 2 percent did not receive a supplement in the last 6 months but did receive one prior to that time. · The mother’s level of education appears to be unrelated to the likelihood of Vitamin A supplementation. · Somewhat surprisingly, Vitamin A supplementation coverage rates were lower in the South and Western regions—where access to health activities is thought to have been better in the last three years during the civil conflict—than in the East and Northern regions. This is most likely due to the Vitamin A supplementation activities that have been conducted during the last year as part of the National Immunization Day campaigns. Vigorous efforts by NGOs in the Eastern and Northern regions to provide preventive health care may also have played a part. · Approximately 33 percent of mothers with a birth in the year before the MICS2 received a Vitamin A supplement within eight weeks of the birth. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 12 Low Birth weight · The technique used to estimate the percentage of infants that weigh less than 2500 grams at birth suggests that 52 percent of infants in Sierra Leone have low birth weight. This percentage is much higher than the average for Sub-Saharan Africa (15 percent) and is higher than the reported percentage1 for every country in the world, including Bangladesh (50 percent). Immunization Coverage · Seventy-three percent of children aged 12-23 months received a BCG vaccination and the first dose of DPT was given to 68 percent. The percentage declines for subsequent doses of DPT to 59 percent for the second dose and 46 percent for the third dose. · Similarly, 82 percent of children aged 12-23 months had received Polio 1 and this percentage declines to 61 percent by the third dose. · Sixty-two percent of children aged 12-23 months had received measles vaccine. · Only thirty-nine percent of children aged 12-23 months had currently received all eight recommended vaccinations. · Vaccination coverage for male children is consistently about five percent higher than for female children for all antigens. Vaccination rates for children living in urban areas are 10- 20% higher than for children living in rural areas, depending on the antigen. · The neediest groups of children in Sierra Leone have the lowest coverage. Coverage rates measured in this survey were generally found to be moderately lower among traditionally underserved groups: girls, children living in rural areas, children living in areas of the country with limited access due to armed conflict (i.e., Northern and Eastern regions), and children of uneducated mothers. Diarrhea · Approximately eighty-six percent of children with diarrhea received one or more of the recommended home treatments (i.e., were treated with Oral Rehydration Solution or a Recommended Home Fluid). · Only 28 percent of children with diarrhea received increased fluids and continued eating as recommended. Acute Respiratory Infection · Nine percent of under-five children had an acute respiratory infection in the two weeks prior to the survey. Approximately 50 percent of these children were taken to an appropriate health provider. IMCI Initiative · Among under-five children who were reported to have had diarrhea or some other illness in the two weeks preceding the MICS, 30 percent received increased fluids and continued eating as recommended under the IMCI program. · Seventy-seven percent of mothers know at least two of the signs that suggest that a child should be taken immediately to a health facility. Malaria · Fifteen percent of under-five children in the Sierra Leone MICS2 slept under a bednet the night prior to the survey interview. However, only about ten percent of the bednets used were impregnated with insecticide. 1 The State of the World’s Children 2000, UNICEF, 2000 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 13 · Approximately 66 percent of children with a fever in the two weeks prior to the MICS2 interview were given Paracetamol to treat the fever and 60 percent were given Chloroquine while only 4 percent were given Fansidar. A relatively large percentage of children (43 percent) were given some other medicine. HIV/AIDS · Only 54 percent of women aged 15-49 at the national level (18 percent in the North) have ever heard of HIV/AIDS. · Twenty-one percent of women aged 15-49 know all three of the main ways to prevent HIV transmission – having only one uninfected sex partner, using a condom every time, and abstaining from sex. · Nineteen percent of women correctly identified three misconceptions about HIV transmission – that HIV can be transmitted through supernatural means, that it can be transmitted through mosquito bites, and that a healthy looking person cannot be infected. · Nine percent of women of reproductive age in Sierra Leone know a place to get tested for HIV and 2 percent have been tested. · The percentage of women who have sufficient knowledge of HIV transmission and the percentage who know where to get tested for HIV increases with the level of education. Contraception · Current use of contraception (both modern and traditional) was reported by 4 percent of married or in union women. The only two methods with a significant level of use are the pill and injections. Seven percent of women who said that they were sexually active reported contraceptive use. Prenatal Care · Fifty-eight percent of women with recent births in Sierra Leone are protected against neonatal tetanus. The vast majority of these women received two or more doses of tetanus toxoid within the last three years. · Eighty-five percent of women in Sierra Leone receive some type of prenatal care and 68 percent receive antenatal care from skilled personnel (doctor, nurse, midwife, auxiliary midwife). Assistance at Delivery · A doctor, nurse, or midwife delivered about 42 percent of births occurring in the year prior to the MICS2 survey. This percentage is highest in the Eastern region at 66 percent and lowest in the North at 22 percent. Birth Registration · The births of 47 percent of children aged less than five years in Sierra Leone have been registered. There are no significant variations in birth registration across gender or age categories. The birth registration breakdown across regions shows that 27, 50, 67 and 81 percent of children have been registered in the Northern, Eastern, Southern and Western regions, respectively. Orphanhood and Living Arrangements of Children · Overall, 61 percent of children aged 0-14 are living with both parents. Children who are not living with a biological parent comprise 16 percent and children who have one or both parents dead amount to 12 percent of all children aged 0-14. · The situation of children in the Western Area differs from that of other children in Sierra Leone. In the Western Area, less than half of children live with both parents. Nineteen Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 14 percent of children live with their mother only although their father is alive and a relatively large proportion (21 percent) live with neither parent. Child Labor · About two percent of children aged 5-14 years engage in paid work. A very large percentage—48 percent—participates in unpaid work for someone other than a household member. · Almost three in four children (74 percent) engage in domestic tasks, such as cooking, fetching water, and caring for other children, for less than four hours a day while 10 percent spend more than four hours a day on such tasks. Maternal Mortality · The maternal mortality ratio (MMR) was directly estimated in the MICS2. The MMR estimate was 1,800 (± 800) maternal deaths per 100,000 live births. This estimate is exactly equal to the working estimate that was used in Sierra Leone prior to the survey, suggesting that it is reasonably accurate. Summary Indicators World Summit for Children Indicators Under-five mortality rate Probability of dying before reaching age five 286 per 1000 Infant mortality rate Probability of dying before reaching age one 170 per 1000 Maternal mortality ratio Ratio of maternal deaths to l ive births: reported per 100,000 l ive births 1,800 ± 800 Underweight prevalence Percentage of under-f ives who are too thin for their age 27.2 percent Stunting prevalence Percentage of under-fives who are too short for their age 33.9 percent Wasting prevalence Percentage of under-f ives who are too thin for their height 9.8 percent Use of safe drinking water Percentage of populat ion who use a safe drinking water source 54.0 percent Use of sanitary means of excreta disposal Percentage of population who use a sanitary means of excreta disposal 63.1 percent Children reaching grade five Percentage of children entering first grade of primary school who eventually reach grade five 84.7 percent Net primary school attendance rate Percentage of children of primary school age attending primary school 41.9 percent Literacy rate Percentage of population aged 15+ years who are able to read a letter or newspaper 29.6 percent Antenatal care Percentage of women aged 15 -49 attended at least once during pregnancy by skilled personnel 68.0 percent Contraceptive prevalence Percentage of married women aged 15-49 who are using a modern contraceptive method 3.9 percent Childbirth care Percentage of births attended by ski l led health personnel 41.7 percent Birth weight below 2.5 kg. Percentage of l ive births that weigh below 2500 grams 52.5 percent Iodized salt consumption Percentage of households consuming adequately iodized salt 23.4 percent Children receiving Vitamin A supplementat ion Percentage of children aged 6 -59 months who have received a Vitamin A supplement in the last 6 months 58.2 percent Mothers receiving Vitamin A s upplementation Percentage of mothers who received a Vitamin A supplement before infant was 8 weeks o ld 32.6 percent Exclusive breastfeeding rate Percentage of infants aged less than 4 months who are exclusively breastfed 2.4 percent Timely complementary feeding rate Percentage of infants aged 6 -9 months who are receiving breast milk and complementary food 52.5 percent Continued breastfeeding rate Percentage of children aged 12 -15 months and 20-23 months who are breastfeeding 85.0 percent (12- 15) 51.3 percent (20- 23) DPT immunization coverage Percentage of children aged 12-23 months currently immunized against diphtheria, pertussis and tetanus 45.5 percent Measles immunization coverage Percentage of children aged 12-23 months currently immunized agains t measles 61.7 percent Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 16 World Summit for Children Indicators Polio immunization coverage Percentage of children aged 12-23 months currently immunized against polio 61.2 percent Tuberculosis immunization coverage Percentage of children aged 12 -23 months currently immunized against tuberculosis 72.8 percent Children protected against neonatal tetanus Percentage of one year old children protected against neonatal tetanus through immunization of their mother 57.7 percent ORT use Percentage of under-five children who had diarrhea in the last 2 w eeks who were treated with oral rehydration salts or an appropriate household solution 86.1 percent Home management of diarrhea Percentage of under-five children who had diarrhea in the last 2 weeks and received increased fluids and continued feeding duri ng the episode 28.4 percent Care seeking for acute respiratory infections Percentage of under-five children who had ARI in the last 2 weeks and were taken to an appropriate health provider 50.2 percent Preschool development Percentage of children aged 36 -59 months who are attending some form of organized early childhood education program 11.7 percent Indicators for Monitoring Children’s Rights Birth registration Percentage of under-f ive children whose births are reported registered 46.8 percent Chi ldren’s l iving arrangements Percentage of children aged 0 -14 years in households not living with a biological parent 16.1 percent Orphans in household Percentage of children aged 0 -14 years who are orphans living in households 2.4 percent Chi ld labor Percentage of children aged 5 -14 years who are currently working 71.6 percent Indicators for Monitoring IMCI and Malaria Home management of i l lness Percentage of under-five children reported ill during the last 2 weeks who received increased f luids and continued feeding 29.5 percent Care seeking knowledge Percentage of caretakers of under-five children who know at least 2 signs for seeking care immediately 77.2 percent Bednets Percentage of under -five children who sleep under an insecticide-impregnated bednet 1.5 percent Malaria treatment Percentage of under-five children who were ill with fever in the last 2 weeks who received anti -malarial drugs 60.9 percent Indicators for Monitoring HIV/AIDS Knowledge of preventing HIV/AIDS Percentage of women who correctly state the 3 main ways of avoiding HIV infection 21.1 percent Knowledge of misconceptions of HIV/AIDS Percentage of women who correctly identify 3 misconceptions about HIV/AIDS 19.0 percent Knowledge of mother to child transmission of HIV Percentage of women who correctly identify means of transmission of HIV from mother to chi ld 26.9 percent Attitude to people with HIV/AIDS Percentage of women expressing a discriminatory attitude towards people with HIV/AIDS 23.4 percent Women who know where to be tested for HIV Percentage of women who know where to get a HIV test 9.2 percent Women who have been tested for HIV Percentage of women who have been tested for HIV 2.5 percent Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 17 I. Introduction Background of the MICS2 Survey At the World Summit for Children held in New York in 1990, the government of Sierra Leone ("Government") pledged itself to a Declaration and Plan of Action for Children (“Plan”). The Government subsequently developed a National Program of Action for Children (“Program”) in 1992. The Program envisaged that programmatic efforts in line with the Convention on Rights of the Child2 (CRC) would be implemented, targeting the women and children of Sierra Leone. The Program, unfortunately, has only been partially implemented due to the internal conflict in Sierra Leone described below. The Government has, over the years, ensured the formulation and the implementation of policies and laws that are in harmony with provisions of the CRC. The Plan also called for the establishment of mechanisms for monitoring progress toward the Program's goals and objectives (taken from the World Summit for Children (WSC) Goals that are to be achieved by the year 2000). Toward this end, UNICEF—in collaboration with other international organizations—has developed a core set of 75 indicators that measure specific aspects of the situation of children. A MICS survey was conducted in Sierra Leone in 1995 to measure progress at mid-decade, although the survey only covered the country's Western Area. In 1999 the Government—in collaboration with UNICEF, UNFPA and NGOs both international and local—made the decision to conduct a survey in 2000 to determine the degree to which the goals and objectives laid out in the Plan have been achieved in Sierra Leone. The Multiple Indicator Cluster Survey–2 (MICS2) was designed to measure a wide variety of key indicators that are associated with the goals set out in the Plan. This report presents results on the principal topics covered in the survey and on the WSC indicators. The Central Statistics Office (CSO) of the Government conducted the MICS2 survey. Funding was provided by the UNICEF Sierra Leone Office, United Nations Population Fund (UNFPA), United Nations High Commission for Refugees (UNHCR), Action Aid - Sierra Leone, Christian Children’s Fund, and the Christian Health Association of Sierra Leone. The total cost of the survey was $93,539 (Appendix A). An Inter-Agency Working Group, consisting of many governmental, non-governmental and United Nations agencies (see Appendix B), was established to oversee the management of the MICS2 survey and monitor its progress. An international consultant was fielded to assist the CSO staff on the technical implementation of the survey. This MICS2 Survey Report details the results of a survey that is truly national in scope. Officials of the Government of Sierra Leone ("Government") worked closely with officials of the Revolutionary United Front Party (RUFP) and the Commission for the Consolidation of Peace (CCP) to reach an agreement whereby the survey team would include members of these two groups. This approach was pursued in order to further the process of national reconciliation and ensure access to the entire country for the data collection effort. Only the outbreak of hostilities in May 2000 prevented the survey teams from covering 100 percent of the country. Ninety-two percent of the country was successfully surveyed and statistical weighting techniques have been used to construct unbiased national estimates of all indicators. This report therefore documents the first methodologically rigorous survey in many years that provides accurate estimates of key health and education indices for areas in Sierra Leone controlled by rebel forces. 2 The CRC was adopted by the General Assembly of the United Nations in 1989 and was subsequently ratified by the Government on June 18, 1990. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 18 Sierra Leone Background: Independence, conflict and consequences Sierra Leone achieved independence from British rule in 1961. Living conditions in Sierra Leone at that time were difficult for many, yet the mood was upbeat. The country appeared to be well- positioned initially for a transition from a colonial state into an independent democratic republic. Sierra Leone had a multi-party system with a western-style parliament and well-organized political parties. In 1967, the first in a series of coups and counter-coups took place, with a subsequent negative effect on the political stability of the country. In 1978, following the model of many other African countries of that era, a one-party system was adopted with the goal of containing ethnic rivalry and unifying all sections of society. This goal was not achieved—the system was abused and the basic tenets of good governance in Sierra Leone were undermined. In 1991, the rebel war that persists to this day in Sierra Leone began with an attack on the country from rebels based in Liberia . The following nine years constitute a tragic chapter in the history of the Republic Armed forces of the Revolutionary United Front (RUF) controlled close to half of the country at the time of the data collection for this survey in April 2000. What have been the consequences of the unending political and military turmoil on those most vulnerable to poverty and disease in Sierra Leone—her women and children? Prior to this survey, the estimated under-five mortality in Sierra Leone was 316 per 1,000 births3, giving this country the dubious distinction of having the highest under-five mortality rate in the world. This report confirms and quantifies the catastrophic effect that the conflict in Sierra Leone has had on the health and education sectors in Sierra Leone. The levels of many health and education indicators at the national level are tragically low, and the residents of regions controlled by rebel forces are significantly more disadvantaged than those living in the rest of the country. This decade-long conflict is the defining context of the information presented in this report. Survey Objectives The 2000 Sierra Leone Multiple Indicator Cluster Survey—2 has as its primary objectives: · To provide up-to-date information for assessing the situation of children and women in Sierra Leone at the end of the decade and for looking forward to the next decade; · to furnish data needed for monitoring progress toward goals established at the World Summit for Children and a basis for future action; and, · to contribute to the improvement of data and monitoring systems in Sierra Leone and to strengthen technical expertise in the design, implementation, and analysis of such systems. II. Survey Methodology Background Reliable census data were not readily available at the time of planning the conduct of the MICS2 survey in Sierra Leone. These data were a prerequisite to the construction of a viable sampling frame. Armed members of the RUF effectively controlled over fifty percent of the country and only granted limited access to areas under their control. All parties involved in the conduct of the survey were committed to conducting a survey that would be truly national in scope, despite the 3 State of the World Children 2000, UNICEF Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 19 constraints imposed by the armed conflict. Within this context, the MICS2 survey team had to design the study in a creative manner in order to ensure a quality data collection effort. One of the most sensitive issues faced in the development of the sampling plan was how to carry out the survey in “inaccessible areas” that were controlled by rebel forces. The sampling plan that was developed assumed that permission would be secured from rebel leaders to conduct the survey in areas of Sierra Leone that were currently inaccessible. As a result of subsequent discussions with government officials and the rebel leaders, agreement was reached on both sides to conduct the survey throughout the country. The data collection teams included members of various rebel groups in Sierra Leone in order to ensure that enumerators achieved access to areas that were currently under the control of the rebel groups as well as to enhance a sense of ownership in the survey process among members of rebel groups. Sample Design The sampling methodology used in the MICS2 survey was designed to provide estimates of health and education indicators at the national level, for urban and rural areas, and for the four regions of Sierra Leone: North, South, East and West. The sample was selected in three stages using a stratified cluster sampling methodology. Sierra Leone has been demarcated into 2,546 enumeration areas (EAs) for the purpose of collecting census data. The current population of each EA in Sierra Leone was estimated through the use of key informants at the chiefdom level. Refugee camps containing Internally Displaced Persons (IDPs) were considered to be additional EAs and were included in the sample frame of EAs in order to ensure that IDPs were adequately represented in the survey. Based on the EA population estimates, randomized systematic sampling was used to select 225 EAs for the survey. The probability of an EA’s selection was proportional to the size of its population4, resulting in a self-weighting sample. Selected EAs were divided into segments of approximately 100 households each with the aid of a local informant. Maps of the selected EAs, maintained in the Cartographic Section of the CSO, were used to guide this “segmentation”. One segment in each EA was then randomly selected. The map of this segment was then updated. This segment was then divided into pieces of approximately 20 households, one of which was randomly selected for inclusion in the survey. All households in the selected piece were listed on the Household Listing Form for the EA and included in the survey. This approach to sampling yielded an overall sample size of 4,566 households. A household was defined as “a group of people who all eat from the same pot”, in line with the definition of household used in previous surveys in Sierra Leone. Five of the selected enumeration areas were not surveyed because they were not accessible during the enumeration period. Data from twelve additional EAs in the North that were surveyed were not processed because the fieldworkers had to flee Makeni when hostilities resumed between RUF and government forces in May 2000. Sample weights were used for reporting results to compensate for the seventeen clusters that were not included in the final database. Full technical details of the sampling plan are detailed in Appendix C. Questionnaires The MICS2 manual provides a carefully constructed core questionnaire that is extremely comprehensive in nature. The core questionnaire is made up of three questionnaires: Household questionnaire, Questionnaire for women, and Questionnaire for children. Explicit instructions are given regarding the addition and subtraction of questions to the core questionnaire. Very minimal 4 Note: large EAs may have had more than one cluster of 20 households selected for inclusion in the survey. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 20 modifications were made to the questionnaires for the purpose of the Sierra Leone MICS2. Optional modules such as the Disability Module and the Maternal Mortality Module were modified and added to the core questionnaire. Several additional questions were incorporated into existing modules to provide information pertinent to development and relief efforts in Sierra Leone. Any additions to or deletions from the core questionnaire were done within the guidelines detailed in the MICS2 manual. The household questionnaire was administered in each household. This questionnaire collects information on household members including sex, age, literacy, marital status, and orphanhood status. The household questionnaire also includes modules on education, child labor, water and sanitation, salt iodization, disability and maternal mortality. The Questionnaire for women and Questionnaire for children were administered in each surveyed household for women age 15-49 and children under age five. For children, the questionnaire was administered to the mother or caretaker of the child. The questionnaire for women contains the following modules: · Child mortality; · tetanus toxoid; · maternal and newborn health; · contraceptive use; and, · HIV/AIDS. The questionnaire for children under age five includes modules on: · Birth registration and early learning; · Vitamin A supplementation; · breastfeeding; · care of illness; · malaria; · immunization; and, · anthropometry. Translation of questionnaire The MICS2 manual recommends that enumerators never verbally translate the questionnaire “on- the-spot” to a respondent during the survey. It is instead recommended that enumerators are provided with a full complement of different language versions of the MICS2 questionnaire for all locally spoken languages. This allows interviewers to interview each respondent in a reliable manner with a fully pre-tested version of the questionnaire in a language in which the respondent is fluent. This protocol is grounded in rigorous research methodology and clearly should be followed to the extent possible. However, this approach was logistically impossible to adhere to in Sierra Leone. Five different languages are spoken in Sierra Leone —English, Krio, Temne, Mende and Limba—and the latter four are not written. In line with past surveys conducted by CSO, enumerators were provided with an English version of the questionnaire along with a list of key words (for each question on the questionnaire) in the language(s) spoken in the regions to be surveyed. Enumerators translated questions in an ad-lib fashion, while making sure to use the key words while translating. Pre-testing of questionnaire The finalized questionnaire was pre-tested in both rural and urban areas of the Western region. As noted above, there are five principal languages in Sierra Leone: English, Krio, Temne, Mende, and Limba. During the pre-test, the questionnaire was administered in each of these five Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 21 languages in at least six households per language. The finalized questionnaires are included in Appendix D. Training and Fieldwork Training Eight Technical Supervisors from the CSO were trained for two days in supervision protocols prior to the commencement of the enumerator training. Enumerators were trained for five days in Freetown based on a modification of the training plan outlined in the MICS2 manual. Although the enumerator training included lecture and demonstration components, a substantial practical component comprised the most important aspect of the training. Each trainee conducted at least three complete interviews during the training, and at least two of these interviews were observed in full by a (different) supervisor. Each supervisor made comprehensive written notes based on their observation. These notes were shared with both the pertinent enumerator as well as among supervisors at the daily supervisors’ meeting that took place each day at the close of training. Trainees had to be approved by each of the supervisors who observed their performance in order to be offered employment as an Enumerator in the survey. Following the training, thirty- one Enumerators and eight Technical Supervisors were hired to participate in the collection of data. Twelve Associate Supervisors from the two principle rebel groups (RUF and the Armed Forces Revolutionary Council (AFRC)) participated in the enumerator training and were hired to work alongside the CSO Technical Supervisors. An outline of the training schedule is included in Appendix E. Data collection and supervision Data collection commenced immediately following the training of enumerators in late March 2000. The first two days of data collection took place in the Western region under the strict supervision of the Survey Coordinator, the UNICEF M&E Officer and the MICS2 Survey Consultant. These three individuals served as supervisors for both the Enumerators and Technical Supervisors. Common mistakes in both enumeration and supervision were observed and immediate feedback was given to the enumerators and supervisors. Based on this experience a Supervision Plan and Supervision Checklist were developed. The Survey Coordinator and the UNICEF M&E Officer used these supervisory tools to guide their subsequent supervision in other regions. Enumerators worked together in EAs in teams of four, and were accompanied by Technical and Associate Supervisors throughout the survey. Enumerators worked individually within households. Data collection continued for approximately five weeks and finished in early May 2000. Public sensitization for the survey was done on radio in all regions prior to the survey. Survey personnel The MICS2 survey exercise was coordinated, managed and executed by Senior Statisticians at the CSO. The Director of Statistics at the CSO was administratively responsible for the conduct of the survey, while the Senior Statistician in charge of Household Survey Division coordinated the actual execution of the survey. Other CSO personnel who were involved in the exercise included Statisticians, Supervisors, Cartographers, Data Programmers, Data Entry Operators, Data Coders/Verifiers, Enumerators, Clerks, Typists, Drivers and Messengers. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 22 Although some enumerators were full-time CSO employees, the majority were private citizens hired for the duration of the survey. Most enumerators were residents of the regions that they were to survey. Approximately thirty percent of the enumerators were women. As noted above, the survey team included individuals who were members of rebel groups. These individuals were given the title of Associate Supervisors and worked under the terms of a written job description and service contract. The primary role of the Associate Supervisors was to ensure safe access of the survey team to rebel-held areas. Several Associate Supervisors became integral members of the teams they worked with and enumerated many households alongside the enumerators. Data processing and management Completed questionnaires were checked in the field by supervisors and then transported to Freetown, where Verifiers checked each questionnaire to make sure that it had been clearly and correctly completed. Questionnaires were then given to Data Entry Operators for entry into the database. Much of the software that was used to enter, process and analyze the data from the MICS2 survey had been previously developed by UNICEF Headquarters and merely needed to be modified for use in the Sierra Leone MICS2 survey. The software that was used to enter and process data was run in the EpiInfo6 software program. Data were entered on fourteen microcomputers that were rented for the purpose of the MICS2 survey. All questionnaires were double-entered and then verified using EpiInfo’s Validate routine. CSO staff oversaw all aspects of data processing and management. Procedures and standard programs developed for MICS2 by UNICEF and adapted to the Sierra Leone questionnaire were used throughout the data management process. Data entry/processing began in June 2000 and was completed in September 2000. Limitations of the survey Any data collection effort must confront methodological issues that will limit the validity of the results. Results can only be safely applied to the population from which the sample was drawn, and thus the manner in which the sample frame is constructed effects the generalizability of the results. In the Sierra Leone MICS2, two major potential limitations stand out. The first is related to the manner in which the questionnaire was translated into local languages, while the second concerns the technique used to construct the sample frame of EAs. Both are discussed below. As noted above, enumerators working in the Sierra Leone MICS2 worked with an English language version of the questionnaire and translated on-the-spot to respondents in the language preferred by the respondent. This method of translation during data collection exercises is standard practice in Sierra Leone and many other countries of the region. The result of this practice is that different enumerators will invariably ask a given question in (slightly) different ways. This generally will have the net effect of reducing the reliability (i.e., the precision) of the results, depending on the ambiguity of the question and the difficulty in its translation. If the surveyors make systematic mistakes in translating a question (e.g., if in the translation from English to Temne, most enumerators translate “health facility” to “place where a doctor or nurse or midwife or traditional healer works”—as occurred in the Northern region in this survey), then the validity of the results are compromised. No magic formula exists to counteract the limitations that result from this mode of questionnaire translation. Vigorous efforts must be made to supervise enumerators and standardize questioning and translation techniques (through monitoring and the use of standardized key words in different languages). Problems that arise Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 23 must be reported so that results can be most effectively interpreted. All of these efforts to maximize the quality of the data were carried out in the MICS2. The MICS2 sampling plan had clear limitations. The construction of the sample frame was based on estimates of EA populations by local informants. These estimates were certainly inaccurate to some extent. The implicit assumption behind this approach is that there will be no systematic differences between different types of EAs—for example, urban and rural EAs—with regards to the accuracy of the estimates. For example, it is assumed that if population estimates were generally overstated by 20%, this occurred to an equal extent in urban and rural areas of the country. Distortions would not be introduced into the results if this was the case. However, if rural estimates of EA population are generally 20 percent overstated while urban estimates are 20 percent understated, then rural EAs will be over-represented in the sample of EAs, and indicator estimates at the national level will be skewed towards the rural estimates. What this sampling plan sought to accomplish was to generate a sample that could be fairly regarded as nationally representative to estimate key indicators to an acceptable level of precision. The survey team feels that the plan that was implemented achieved this goal in spite of the resource constraints within which the survey was conducted. III. Sample Characteristics and Data Quality Response Rates As noted above, hostilities between the Revolutionary United Front (RUF) and forces aligned with the Government resumed during the final week of data collection for the MICS2 survey in May 2000. As a result, one cluster in the East and four clusters in the North could not be surveyed. Completed questionnaires from twelve clusters in the Northern region that had already been surveyed were stranded in Makeni when enumerators could not return to collect them. The completed questionnaires had been stored there for safekeeping while the survey team had traveled to remote regions to survey other clusters. Data from eight of these clusters have been retrieved recently and will shortly be added to the MICS2 database. In the original sample of 225 EAs there were 4566 households selected for the survey. There were 4207 households selected for the survey in the 208 clusters that were successfully surveyed and entered into the database. 4021 of these 4207 households were found to be occupied and were successfully surveyed, yielding a household response rate of 97 percent (Table 1; see Appendix F). In the interviewed households, 5750 eligible women (age 15-49) were identified. Of these, 4923 were successfully interviewed, yielding a response rate of 85.6 percent. In addition, 3060 children under age five were listed in the household questionnaires of surveyed households. Questionnaires were completed for 2764 of these children for a response rate of 90.3 percent. Age Distribution and Missing Data As shown in Figure 1, the single year age distribution of household members by sex exhibits notable distortions. A condensed version of the data used to construct this figure can be found in Table 2. There appears to be significant heaping of males on ages 12-17 and women on age 50. For both sexes, significant digit preference is evident for ages ending in 0 and 5, a pattern typical of populations in which ages are not always known. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 24 Figure 1: Single year age distribution of the household population by sex, Sierra Leone, 2000 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Age Male Female As a basic check on the quality of the survey data, the percentage of cases missing information on selected questions is shown in Table 3. Fewer than two percent of household members have missing information for both their level of education and years of education. This low level of missing data suggests that significant problems with the questions or the fieldwork did not exist, at least for this aspect of the survey. Among female respondents, 73 percent did not report a complete birth date (i.e., month and year). These data were missing for16 percent of female respondents, while 57 percent of female respondents did not know their complete birth date. Seventy-seven percent of women who had a birth in the 12 months prior to the survey did not report the date of their last tetanus toxoid injection. The degree of completeness of collection of data for these indicators is another sign of the quality and thoroughness of the fieldwork. In this case, data were collected for a very small proportion of respondents. Supervisors noted that most women in Sierra Leone are unable to state the month and year in which they were born, and that many even have great difficulty stating their age. Supervisors did feel like the high percentage of women unable to report a complete birth date may have been partially due to insufficient probing on the part of the enumerators. The data on a child’s weight and height are often the most likely information to be missing in a MICS database. In the Sierra Leone MICS2 survey, approximately ten percent of children are missing this information. The level of missing anthropometric data varied significantly among regions; the percentage of children missing data on weight and/or height is 16.1, 10.8, 0.8 and 1.5 percent in the North, East, South and West, respectively. Discussions were held with supervisors to try and explain the observed variation. The ability of survey teams to return to households where children had not been present varied among regions, due to security issues and varying distances between selected villages. In addition, surveyors in the North mistakenly followed a protocol of taking no anthropometric measurements on children aged less than six months, thus explaining a significant part of the high level of non-measurement for that region. Carelessness of enumerators in completing interviews and supervisors’ inability to check every questionnaire for completeness while in the field also accounted for some cases of non-measurement. By Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 25 international standards, the percentage of unmeasured children is moderate in comparison to other surveys in which anthropometric measurements are taken. Characteristics of the Household Population Information on the characteristics of the household population and the survey respondents is provided to assist in the interpretation of the survey findings and to serve as a basic check on the sample implementation. Table 1 presents the percent distribution of households in the sample by background characteristics. About 30 percent of the households (1187/3916) are urban and 70 percent (2729 households) are rural. Settlements with populations of 2,000 people or more were defined as urban, while those with populations below 2,000 were defined as rural. The Northern region comprises the largest of the four regions with 31 percent (1200/3916) of households interviewed while the Eastern region is the next largest with 25 percent (975/3916; Table 4). The remaining regions—the Southern region and the Western Area—contain 21 (817/3916) and 24 (924/3916) percent of the surveyed households, respectively. Most of the households in the sample (64%) have between four and seven members. Fifty-four percent of the households contain at least one child under age five and 87 percent contain at least one woman aged 15-49. Information was gathered on household possessions among surveyed households. The chart below shows the percentage of households in Sierra Leone that report owning selected items. Chart 1: Percentage of households owning selected items, Sierra Leone, 2000 Item # that own # questioned % that own Radio 1791 3905 45.9 Television 159 3910 4.1 Automobile 62 3910 1.6 Motorcycle 56 3910 1.4 Bicycle 248 3910 6.3 The house they live in 1875 3904 48.0 Table 3 shows the background characteristics of children under age five (and their mothers) who were surveyed. Approximately 78 percent of the mothers of the surveyed children are uneducated5. The proportion of children in each one-year age category between ages 0-11 months to children aged 36 – 47 months steadily increases from age zero years to age three years until falling for the group of children aged four years (i.e., 48-59 months). Fifty percent of the children are male. Among women of child-bearing age in the sample, 16 percent have completed some secondary education (i.e., those who have at least completed the first year of Form 1) while 78 percent are uneducated (Table 4a). Ten percent of women of child-bearing age are in the age group 15-19 years (Table 5). The lowest percentage (2.4) is in the age group 45-49 years. Eighty percent of women in the sample are married and 79 percent have given birth to at least one child (Table 5). 5 For children whose mothers did not live in the household, the education of the child’s caretaker is used. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 26 IV. Results A. Infant and Under-Five Mortality Child mortality indicators Indicator Rate Infant mortality rate (number of children dying before age one divided by the number of live births in the year) 170 per 1000 Under-five mortality rate (number of children dying before the age of five years divided by the number of live births) 286 per 1000 Note: These indicators were calculated using indirect techniques. Numerators and denominators for these indicators are therefore not reported. The infant mortality rate (IMR) is the probability of dying before the first birthday. The under- five mortality rate (U5MR) is the probability of dying before the fifth birthday. In the MICS2 survey, infant and under-five mortality rates are calculated based on an indirect estimation technique (the Brass method). The data used in the estimation are the mean number of children ever born for five year age groups of women from age 15 to 49 and the proportion of these children who are dead (also for the same five year age groups of women). The technique converts these data into probabilities of dying by taking into account both the mortality risks to which children are exposed and their length of exposure to the risk of dying. The Brass method therefore uses retrospective data to obtain estimates of IMR and U5MR. Several different estimates of these indicators are generated, all of which are pegged to different reference years in the past. Generally, following examination of the data, the most recent estimate that is felt to be robust is used. In the case of MICS2, although the survey has been conducted in the year 2000, the IMR and U5MR reported here represent estimates of the rates for1997 (precisely, 1997.3). Figure 2: Estimates of infant and under-five mortality based on indirect estimation, Sierra Leone, 2000 154 153 163 155 185 170 213 260 258 274 261 310 286 355 100 150 200 250 300 350 400 1985 1987 1989 1991 1993 1995 1997 1999 Reference year R at e Infant mortality Under five mortality Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 27 Mortality estimates were calculated using the United Nations QFIVE software program. When using the Brass method a model must be selected that best fits the mortality patterns in the country where the survey was conducted. There are four models: North, South, East and West. The North model life table was selected to estimate mortality rates in this survey. The North model is generally used in countries where both 1) the IMR is high and 2) the mortality rate of children aged one to four years is also high. The North model was used to estimate infant and child mortality for Sierra Leone in 1974 and 1985. Estimates of infant and under-five mortality calculated from the MICS2 data for several reference years are plotted in Figure 2. Visual inspection of these data show that the 1999 estimate is much higher than other estimates, and the 1995 estimates are also notably higher than the estimates from both 1993 and 1997. Estimates from the years 1986 to 1993 are very similar and the 1997 estimates are only slightly higher than estimates from these previous years. Which of these estimates should be chosen as the “correct” estimate—i.e., the most recent robust estimate—for the purposes of reporting? When using the Brass method of indirect estimation, the most recent estimates of IMR/U5MR—those based on the reports of women aged 15-19—are almost never used (and indeed, in this case, they are abnormally high). Some experts recommend that estimates based on the reports of women aged 20-24 should also not be used (i.e., the estimates for the year 1997 in Figure 2 above). However, there are two arguments for using the 1997 data in this case. First, based on visual inspection of Figure 2, the 1997 estimates appear to be reasonably robust, since they are generally in line with data from previous years. Secondly, previous estimates of IMR/U5MR in Sierra Leone have used data from women aged 20-24. The use of data from this age group therefore enhances the comparability of these estimates with previous estimates in Sierra Leone. Figure 3: Infant mortality rate and under five mortality rate in Sierra Leone over time 225 194 170 366 327 286 0 100 200 300 400 1974 1985 1997 D e a t h s p e r 1 0 0 0 l i v e b i r t h s IMR U5MR Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 28 In summary, the IMR/U5MR estimates for 1997 appear to be the most recent data that can be used with confidence. Figure 3 above shows the IMR and U5MR estimates from both MICS2 and previous surveys 6 in Sierra Leone. Discussion: Infant and child mortality · The infant and under-five mortality rates reported here are extremely high and highlight the need to intensify efforts to combat mortality among the children of Sierra Leone. · The data appear to be robust and suggest that the IMR and U5MR in Sierra Leone have slowly but consistently decreased over the last 25 years. One might ask why these rates have been decreasing during such a difficult period in the history of Sierra Leone. The data suggest that the victories of public health have made some impact on the mortality rates in Sierra Leone despite the trying conditions there. Cleaner sources of water prevent many water-borne diseases and the use of oral rehydration solution (ORS) to manage diarrhea saves the lives of countless children in Sierra Leone—children who, with the same disease but without management with ORS, would have died 30 years ago. Vaccinations protect many Sierra Leonean children from diphtheria, polio, measles and tuberculosis. Vitamin A supplementation programs have been proven to lower child mortality dramatically. · The World Summit for Children goal for these indicators is to reduce IMR and U5MR to 50 and 70 deaths per 1,000 live births, respectively. Sierra Leone is far from achieving this goal. B. Education Indicators for education Indicator % Preschool development Numerator: # of children aged 36-59 months who are attending some form of organized early childhood education program 12 Denominator: # of children aged 36-59 months in the sample Net primary school attendance rate Numerator: # of children aged 5-12 years attending primary school according to documentation or caretaker’s report 42 Denominator: # of children aged 5-12 years in the sample Children entering primary school who reach grade 5 Note: This indicator was calculated using indirect techniques. The numerator and denominator for this indicator are therefore not reported. 85 Literacy rate Numerator: # of people aged 15 years or more in the surveyed households who were reported to be able to read a newspaper or a letter “easily” or “with difficulty” 30 Denominator: # of people aged 15 years or more in the surveyed households 6 Source of 1997 estimates is the MICS2 survey. The 1974 and 1985 estimates of IMR and U5MR are documented in The Analytical Report: 1985 Population and Housing Census Sierra Leone (Central Statistics Office, Government of Sierra Leone) and derive from the Sierra Leone Population and Housing Censuses of 1974 and 1985, respectively. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 29 Universal access to basic education and the achievement of primary education by the world’s children is one of the most important goals of the World Summit for Children. Education is a vital prerequisite for combating poverty, empowering women, protecting children from hazardous and exploitative labor and sexual exploitation, promoting human rights and democracy, protecting the environment, and influencing population growth. Early childhood education Approximately one in nine children (11.7 percent) aged 36-59 months are attending an organized early childhood education program—such as kindergarten or community childcare—with organized learning activities (Table 9). Approximately equal percentages of girls and boys are attending these programs. There are large variations in this indicator according to region ranging from 4.3 percent of children in the North to 31.2 percent in the Western region. In addition, children in urban areas are over twice as likely to attend early learning activities as children in rural areas. Relatively few three-year-old children attend (8 percent of all three-year-olds) while a larger proportion of children of age four years attend (17 percent of all four-year-olds). Finally, the education of the mother is strongly related to the likelihood that a child will attend an early childhood education program. The percentage of children attending increases from 8 percent to 36.2 percent as the mother’s education increases from none to secondary or higher education. Basic education Overall, 42 percent of children of primary school age (6-12 years) in Sierra Leone are attending primary school (Table 11). In urban areas, 64 percent of children attend school while in rural areas 34 percent attend. School attendance in the North and East (28 and 35 percent) is significantly lower than in the South and West (48 and 75 percent). At the national level, differences in primary school attendance by gender are small; 44 percent of male children and 40 percent of female children attend primary schools. Eighty-five percent of children who enter the first grade of primary school eventually reach grade five (Table 10). However, there are regional and urban-rural disparities in the achievement of grade five. Approximately 93 percent of urban children who enter grade one reach grade five compared to 76 percent for children in rural areas. Only about 69 percent of the children in the South reach grade 5 as compared to 84 percent in the East and 97 percent in the Western region. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 30 Figure 4: Percentage of children of primary school age attending primary school, Sierra Leone, 2000 0 10 20 30 40 50 60 70 80 90 100 North East South West Urban Rural Percent Literacy Only 30 percent of the population over age 15 years in Sierra Leone is literate (Table 12). The literate population includes those who are reported to read “easily” or “with difficulty”. Overall, females are less likely than males to be literate (20 vs. 40 percent). The literacy rate is 69 percent in the Western region, while regional rates are generally lower outside the Western region – (14 percent in the North, 20 percent in the East and 28 percent in the South.). Literacy declines with increasing age. The percentage literate declines from 38 percent among those aged 15-24 to 15 percent among the population aged 65 and older. Discussion: Education · Early childhood education in Sierra Leone is a new concept. This is the best explanation for the low enrolment rate for the country. The higher rate in the Western Area is probably due to the higher proportion of educated and employed mothers in that region. Many of these mothers use kindergarten as a child care institution while they are at work. · Enrolment of children of primary school age shows a small improvement over that obtained in the 1985 Census (37 percent). · The adult literacy rate (30 percent) shows an improvement over the rate estimated in the 1985 Census (13 percent). However, the current estimate is still well below the goal set at the World Summit (goal: 60% adult literacy by the year 2000). The adult literacy rate has increased across gender between 1984 and 2000: from 17 percent (1985 Census) to 40 percent (MICS 2, 2000) for males and from 9 to 20 percent for females during the same period. This positive change is most likely due to changes in attitudes towards education, particularly for the girl child. · The finding that 85% of Sierra Leonean children who enter grade one complete grade five is—at first glance—surprisingly high. However, the comparison of this result with data from other countries of the region shows that it concurs with regional trends. The State of the Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 31 World’s Children 2000 (SOTWC; UNICEF, 2000) reports rates of primary school completion of 75 percent in the Ivory Coast, 80 percent in Guinea, 75 percent in Burkina Faso and 80 percent in Ghana. The Government of Sierra Leone has adopted a “6:3:3:4 approach” 7 to education in the early 1990s. Following the implementation of this plan, children in many schools have passed easily from one primary grade to another, almost irrespective of merit. Passing the national exams and gaining entry to secondary school is far more difficult under “6:3:3:4”. It is thus fairly easy—from a scholastic point of view—for a child to complete his or her primary education in Sierra Leone. · The two indicators of basic education taken together suggest that putting a child in school in Sierra Leone is a decision taken seriously by the child’s family. Once a child is put in primary school, the family does what is necessary to keep them in until they complete Class 5. This finding would seem to have clear programmatic implications for those working in the education sector in Sierra Leone. The push should be to encourage children to enter Class 1. Support mechanisms will need to be developed and implemented to help “new children”— that is, those who might not have attended school if not encouraged—to complete their primary education. C. Water and Sanitation Indicators for monitoring access to safe drinking water and sanitary means of excreta disposal Indicator % Use of safe drinking water Numerator: # of people in the surveyed households who use any of the following types of water supply for drinking: piped water; public tap; borehole/pump; protected well; protected spring; or rainwater. 54 Use of sanitary means of excreta disposal Numerator: # of people in the surveyed households who have any of the following within their dwelling or compound: toilet connected to sewage system; any other flush toilet (private or public); improved pit latrine; or traditional pit latrine. 63 Denominator: # of people in the surveyed households Use of drinking water Safe drinking water is a basic necessity for good health. Unsafe drinking water can be a significant carrier of diseases such as trachoma, cholera, typhoid, and schistosomiasis. Drinking water can also be tainted with chemical, physical and radiological contaminants with harmful 7 According to the 6:3:3:4 system of education, a student should complete their formal education in 16 years. The student enters primary school in Class 1 and completes their primary education in Class 6. After passing national exa ms, the student enters the first year of secondary education at the Junior Secondary School Level 1 (JSS1). Following three years of education at this level the student completes JSS3 and sits a second national exam. If the student is successful on this exam, he enters the Senior Secondary School Level 1 (SSS1). Following the student’s completion of SSS3 (the final year of secondary school) he takes a national exam that, if passed, allows him to attend the University for four years. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 32 effects on human health. In addition to its association with disease, access to drinking water may be particularly important for women and children, particularly in rural areas, who bear the primary responsibility for carrying water, often for long distances. The distribution of the overall population by source of drinking water is shown in Figure 5. Only 2 percent of the population uses drinking water from a source that is piped into their dwelling and 7 percent used water piped into their yard or plot. Rainwater collection is an insignificant source of drinking water. Potentially unsafe sources of drinking water such as unprotected dug wells, springs, ponds, rivers or streams are used by a significant percentage (46 percent) of the population. Figure 5: Percent distribution of the population in Sierra Leone by source of drinking water, Sierra Leone, 2000 Piped into yard 7% Public tap 15% TubeWell with pump 13% Protected dug Well 14%protected other 3% unprotected 19% River/stream 27% Piped into dwell. 2% The source of drinking water for the population varies strongly by region (Table 13). In the Western Area, 32 percent of the population uses drinking water that is piped into their dwelling or into their yard or plot. The corresponding percentages for the regions are 0.4, 3 and 0.2 for the North, East and South. Access to drinking water at public water taps was higher than access to private piped water in all regions, with percentage access at 3, 16, 12 and 41 respectively for the North, East, South and West regions. People who use safe drinking water sources are those who use any of the following types of supply: piped water, public tap, borehole/tubewell, protected well, protected spring or rainwater. Overall, 54 percent of the population has access to safe drinking water: 74 percent in urban areas and 46 percent in rural areas. The situation in the North with regard to this indicator is considerably worse than in other regions; only 30 percent of the population in the Northern region gets its drinking water from a safe source. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 33 Use of sanitation Inadequate disposal of human excreta and personal hygiene is associated with a range of diseases including diarrheal diseases and polio. Sanitary means of excreta disposal include: flush toilets connected to sewage systems or septic tanks, other flush toilets, improved pit latrines, and traditional pit latrines. Sixty-three percent of the population of Sierra Leone is living in households with sanitary means of excreta disposal (Table 14). This percentage is 88 in urban areas and 53 percent in rural areas. Access on regional basis is 53, 62, 51 and 95 percent in the North, East, South and Western regions, respectively. Most of the population without proper sanitary means of excreta disposal use open pits, rivers, the bush, fields, or other places for defecation. Discussion: Water and sanitation · Survey results suggest that improvement may have taken place in the population’s access to sources of safe drinking water in Sierra Leone between 1985 and 2000. The 1985 census reported that 16 percent of the population of Sierra Leone obtained their drinking water from (private or public) taps, while this survey found that 24 percent of the population drinks water from taps. The 1985 census did not dif ferentiate between wells that had safe and unsafe drinking water—a differentiation made in the MICS2 survey—making it impossible to assess overall change between 1985 and 2000 in the level of the population with access to safe drinking water. · Comparisons between the level of the indicator percentage of the population that has access to sanitary means of excreta disposal in 1985 and 2000 are impossible because of the way data was collected. The 1985 census did not differentiate between the different types of pit latrines (ventilated improved pit (VIP) latrine; traditional pit latrine; open pit) that, taken together, are used by more than half of the population of Sierra Leone. The MICS2 survey differentiated between these types of pit latrines, and considered the VIP and the traditional pit latrine to be sanitary means of excreta disposal, whereas an open pit was not. The primary difference between an open pit latrine (OPL) and a traditional pit latrine (TPL) is that the TPL is covered whereas the OPL is not covered. · Given the high percentage of the population that reports using a TPL (54 percent), it would be worthwhile to conduct a further investigation of the hygienic standards and quality of TPLs used in Sierra Leone. This could provide information that would guide the development of programs to improve the quality of sanitary facilities and ensure that TPLs in Sierra Leone are truly “sanitary means of excreta disposal”. · This survey found that 54 percent of Sierra Leoneans have access to safe drinking water. This result is well short of the World Summit goal of universal (100%) access to safe drinking water. Access to adequate sanitary means of excreta disposal as measured in MICS2 (63%) is also far below the goal of universal access that was set at the World Summit. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 34 D. Child Malnutrition Nutritional status Indicators of moderate and severe malnutrition among under-five children Indicator % Underweight prevalence Numerator: # of children aged 0-59 months who fall below minus 2 standard deviations from median weight- for-age of NCHS/WHO reference population 27 Stunting prevalence Numerator: # of children aged 0-59 months who fall below minus 2 standard deviations from median height- for-age of NCHS/WHO reference population 34 Wasting prevalence Numerator: # of children aged 0-59 months who fall below minus 2 standard deviations from median weight- for-height of NCHS/WHO reference population 10 Denominator: # of children aged 0-59 months who had complete data for weight, height, and age in the sample Children’s nutritional status is a reflection of their overall health. When children have access to an adequate food supply, are not exposed to repeated illness, and are well cared for, they reach their growth potential and are considered well nourished. In a well-nourished population, there is a standard distribution of height and weight for children under age five. Undernourishment in a population can be gauged by comparing children to this standard distribution. The standard or reference population used here is the NCHS standard, which is recommended for use by UNICEF and the World Health Organization. Each of the three nutritional status indicators are expressed in standard deviation units (z-scores) from the median of this reference popula tion. Weight-for-age is a measure of both acute and chronic malnutrition. Children whose weight-for- age is more than two standard deviations below the median of the reference population are considered moderately or severely underweight while those whose weight-for-age is more than three standard deviations below the median are classified as severely underweight. Height-for-age is a measure of linear growth. Children whose height-for-age is more than two standard deviations below the median of the reference population are considered short for their age and are classified as moderately or severely stunted. Those whose height-for-age is more than three standard deviations below the median are classified as severely stunted. Stunting is a reflection of chronic malnutrition as a result of failure to receive adequate nutrition over a long period and recurrent or chronic illness. Finally, children whose weight for height is more than two standard deviations below the median of the reference population are classified as moderately or severely wasted, while those who fall more than three standard deviations below the median are severely wasted. Wasting is usually the result of a recent nutritional deficiency. The indicator may exhibit significant seasonal shifts associated with changes in the availability of food or disease prevalence. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 35 In Table 15, the percentage of children who had missing weights and heights (approximately 10 percent of the interviewed children) is described. These children are excluded from anthropometric calculations, as are those children whose measurements fall outside a plausible range. In addition, the eight percent of children in the sample whose birth dates were unknown have also been excluded from these calculations. More than one child in four under age five in Sierra Leone is moderately (27 percent) underweight and 8 percent are severely underweight (Table 15a). Thirty-four percent of children are stunted (i.e., are too short for their age) and 9.8 percent are wasted (i.e., too thin given their height). Figure 6 displays percentages of children who are underweight, stunted, or wasted, across different age categories. Figure 6: Percentage of under-five children in Sierra Leone who are undernourished, Sierra Leone, 2000 0 5 10 15 20 25 30 35 40 45 50 0 6 12 18 24 30 36 42 48 54 60 Age (months) Stunted Underweight Wasted Higher percentages of children in the North and East are underweight (29 and 32 percent) than children in the South and West (24 and 20 percent). Levels of stunting are uniformly high in the North, East and South (range: 36-40 percent) and are lower (yet still high) in the West (22 percent). Levels of wasting are fairly constant across regions, ranging from 8 percent in the South to 11 percent in the North. Mothers who have secondary or higher education are less likely to have children who are underweight, stunted or wasted, as compared to children of mothers with lower levels of education. Boys appear to be somewhat more likely to be underweight, Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 36 stunted, and wasted than girls. The age pattern shows that malnutrition in under-fives starts manifesting itself as early as six months and reaches a peak before the child is two years old. In the case of underweight, the children are more likely to falter in growth as late as the fourth year of life (Table 15a). This pattern is typical in many developing countries, due to the fact that children at this age cease to be breastfed and are newly exposed to contamination in water, food, and environment. Discussion: Child malnutrition · A major contributing factor to under-five mortality in Sierra Leone has been childhood malnutrition. The present levels of moderate and severe malnutrition rates for underweight, wasting and stunting (27, 10 and 34 percent) in the MICS2 survey are comparable to the figures reported for Sierra Leone in the SOTWC (29, 9, and 35 percent). The unacceptably high figures are due to the overall economic depression the country has undergone for the past ten years. There has been inadequate food production in the country due to displacement of people because of war. This has subsequently led to the household food insecurity that triggers malnutrition among the under-five children. · The rates of malnutrition reported here for Sierra Leone are similar to those reported for other countries in West Africa and for the region as a whole (source: SOTWC), suggesting that the data are robust. Ivory Coast and The Gambia report underweight prevalence of 24 and 26 percent, while the regional underweight prevalence is 32 percent. Stunting prevalence is 24, 29, 30 and 41 percent in Ivory Coast, Guinea, The Gambia and the region, respectively. Similarly, wasting prevalence is 8, 12 and 9 percent in Ivory Coast, Guinea and the region. Two key conclusions that can be drawn here are that 1) the child malnutrition crisis is regional, not national, in nature, and 2) vigorous relief efforts in Sierra Leone by a variety of agencies may have helped to prevent the prevalence of child malnutrition in Sierra Leone from increasing beyond regional norms. Breastfeeding Indicators of breastfeeding status Indicator % Exclusive breastfeeding rate Numerator: # of children aged less than 4 months who were exclusively breastfed according to mother’s or caretaker’s report 2 Denominator # of children less than 4 months of age in the sample Timely complementary feeding rate Numerator: # of children aged 6-9 months who received breast milk AND complementary food according to mother’s or caretaker’s report 52 Denominator: # of children aged 6-9 months in the sample Continued breastfeeding rate Numerator: # of children aged: 12-15 months 20-23 months who were breastfeeding at time of the survey according to mother’s or caretaker’s report. 85 51 Denominator: # of children aged 12-15 months and 20-23 months in the sample Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 37 Breastfeeding for the first few years of life protects children from infection, provides an ideal source of nutrients, and is economical and safe. However, many mothers stop breastfeeding too soon, and there are often pressures to switch to infant formula milk, which can contribute to growth faltering and micro-nutrient malnutrition. Use of formula milk is unsafe if clean water is not readily available. The World Summit for Children goal states that children should be exclusively breastfed for four to six months, and that breastfeeding should continue with complementary food well into the second year of life. Many countries have adopted the recommendation of exclusive breastfeeding for about six months. In Table 16, breastfeeding status is based on women’s reports of children’s consumption in the 24 hours prior to the interview. Exclusive breastfeeding refers to children who receive only breast milk and vitamins, mineral supplements, or medicine. Complementary feeding refers to children who receive breast milk and solid or semi-solid food. The last two columns of the table include children who are continuing to be breastfed at 12-15 and 20-23 months of age. Exclusive breastfeeding status was investigated for children aged three months and below in this study. An extremely low percentage of Sierra Leonean children aged 0-3 months (2 percent) are exclusively breastfed. At age 6-9 months, 52 percent of children are receiving both solid or semi- solid foods and breastmilk. Encouragingly, 85 percent of children aged 12-15 months are still being breastfed, while 51 percent of children aged 20-23 months are breastfed. Breastfeeding rates were similar for boys and girls except at age 20-23 months, when girls were more likely to be breastfed than boys (58 vs. 45 percent). Figure 7 shows the percentage of children by breastfeeding status by age in months. Note that the percentage of children who are breastfeeding and drinking other liquids but not solid or semi- solid food is not shown on the graph. For example, among children three months of age, 2 percent are exclusively breastfed, 2 percent are not breastfed, and 9 percent receive both solid or semi-solid foods and breastmilk. From these data it can be determined that 87 percent (100 – [2 + 2 + 9]) of children aged 3 months receive breastmilk and other liquids but not solid or semi-solid food. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 38 Figure 7: Percent distribution of living children under two years by breastfeeding status, Sierra Leone, 2000 0 10 20 30 40 50 60 70 80 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Age in months Exclusively breastfed Breast milk and solid foods Not breastfed Discussion: Breastfeeding · The rate of exclusive breastfeeding (EBF) in Sierra Leone (2 percent) is essentially nil and well below the regional average (31 percent for Sub-Saharan Africa). Given the high rate of infant mortality, high levels of infant malnutrition, and the prevalence of diarrhea among children 0-3 months of age (9 percent) in Sierra Leone, there is clearly a need to promote EBF among Sierra Leonean mothers. · The rate of complementary feeding as measured in MICS2 (52 percent) is low with regards to previously reported rates in Sierra Leone (94 percent; source: SOTWC) but is only slightly lower than the regional average (66 percent, SOTWC). The percentage of children in the Sierra Leone MICS2 whose caretakers reported that they received breastmilk but no solid or semi-solid foods in the 24 hours prior to the survey interview is high across many age categories. For example, caretakers of 31 of 79 children (39 percent) aged 20-23 months who reported that the child received breastmilk during the 24 hours prior to the interview reported that the child did not receive solid or semi-solid food. The value of this indicator for children aged 12-15 months and 16-19 months is 40 and 32 percent, respectively. These data suggest that either a substantial percentage of very young breastfeeding children in Sierra Leone are not receiving complementary food—perhaps due to food shortage, current illness, or lack of caretaker knowledge—or that some respondents misunderstood the question and answered that their children had not received complementary food when in fact they had. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 39 Salt iodization Indicator of iodized salt consumption Indicator % Iodized salt consumption Numerator: # of households consuming adequately iodized salt 23 Denominator: # of households that provided salt to sample in the sample Deficiency of iodine in the diet is the world’s single greatest cause of preventable mental retardation and can lower the average intelligence quotient (IQ) of a population by as much as thirteen points. Salt iodization is an effective, low-cost way of preventing iodine deficiency disorders (IDD). Adequately iodized salt contains 15 ppm (parts per million) of iodine or more. In MICS, interviewers tested household salt for iodine levels by means of a testing kit. Approximately 93 percent of households had salt that was tested during the MICS2 (Table 17). Among these households, 23 percent had adequately iodized salt. The percentage of households with adequately iodized salt ranges from 5 percent in the North (where local mining of salt is prevalent) to 47 percent in the Eastern province (where no mining of salt occurs). Twenty-eight percent of urban households had adequately iodized salt as compared to 22 percent of rural households. Discussion: Salt iodization · Sierra Leone is a long way from achieving the goal of universal iodization of household salt for consumption. The government should consider formulating policy and introducing legislation that would help to achieve this goal. · The percentage of households consuming adequately iodized salt in Sierra Leone (23 percent) is similar to the percentages in other countries of Western Africa (9, 10, 23 and 37 percent in Senegal, Ghana, Burkina Faso and Guinea, respectively) but is well short of the average for Sub-Saharan Africa (62 percent; source: SOTWC). · Local salt mining is an important economic activity in the Northern region and the Western Area. Local salt mining (of uniodized salt) in these two regions may be the reason for the low prevalence of iodized salt there. Households in the Eastern and Southern regions generally depend on imported salt that is more likely to be adequately iodized. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 40 Vitamin A supplementation Indicators for Vitamin A supplementation coverage Indicator % Children receiving Vitamin A supplementation Numerator: # of children aged 6-59 months who received a high dose of Vitamin A supplement in the six months prior to the survey 58 Denominator: # of children aged 6-59 months in the sample Mothers receiving Vitamin A supplementation Numerator: # of mothers who received a high-dose Vitamin A supplement before infant was 8 weeks old 33 Denominator: # of mothers in the sample with live births in the 12 months prior to the survey Vitamin A deficiency (VAD) impairs children's immune systems, thereby increasing their chances of dying of common childhood diseases. VAD undermines the health of pregnant and lactating women and it can also cause eye damage and blindness in children. Yet it can be easily prevented by vitamin A supplementation or food fortification. UNICEF and WHO recommend that all countries with an under-five mortality rate exceeding 70 per 1000 live births, or where vitamin A deficiency is a public health problem, should put in place a program for control of vitamin A deficiency. Based on UNICEF/WHO guidelines, the Ministry of Health and Sanitation in Sierra Leone recommends that 1) children aged 6-12 months be given one dose Vitamin A capsule of 100,000 IU every six months, and that 2) children older than one year be given one high dose of 200,000 IU every six months. Sierra Leone does not have a formal program to control VAD. However, actions are taken in Sierra Leone to provide Vitamin A to malnourished children in therapeutic and supplementary feeding centers. Children who are given measles vaccine in Sierra Leone are routinely given a Vitamin A capsule. Vitamin A capsules have been given to children concurrent with polio immunization as part of the National Immunization Day (NID) campaigns. Within the six months prior to the MICS2 survey, 58 percent of surveyed children aged 6-59 months received a high-dose Vitamin A supplement (Table 18). Approximately 2 percent of the children had not received the supplement in the six months prior to the survey, but had received one earlier. A minor percentage (1.4 percent) of children received a Vitamin A supplement at some time in the past, although their caretaker was unable to specify when. The distribution of Vitamin A supplementation coverage in the regions is 60, 71, 40, and 50 percent for North, East, South and West, respectively. There was no significant association between the mother’s level of education and the likelihood of her child receiving Vitamin A supplementation. Vitamin A supplementation of mothers is much higher in the Eastern region (55 percent) than in the three remaining regions, where coverage levels range between 24 and 27 percent. Discussion: Vitamin A supplementation · Field trials of Vitamin A supplementation (VAS) have shown its outstanding potential to dramatically lower child mortality rates. Given the high IMR in Sierra Leone, the Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 41 implementation of a formal Vitamin A supplementation program should clearly be considered. Vitamin A supplementation in under-fives has recently received a boost during the National Immunization Days, as all children receiving polio vaccine during these campaigns have also received a Vitamin A tablet. In the short-term, the Ministry of Health might consider making VAS for under-fives at under-five clinics mandatory, so as to maintain and improve on current coverage levels. · Somewhat surprisingly, VAS coverage rates were lower in the South and Western regions— where access to health activities is thought to have been better in the last three years during the civil conflict—than in the East and Northern regions. This is somewhat puzzling, given that VAS is conducted during National Immunization Day (NID) campaigns, and NID coverage was thought to be highest in the Southern region and Western Area. Vitamin A tablets are also given to children receiving measles vaccine in Sierra Leone, but there have not been any major measles vaccination campaigns in the North or East recently that would explain such high VAS coverage rates. Supplementary and therapeutic feeding centers are (or have been) operated in the North and East by non-governmental organizations such as Medicines Sans Frontieres (MSF) in the 6 months prior to the survey. The existence of larger numbers of nutritional centers (and/or larger caseloads) in the North and East—as compared to the South and West—may explain the high VAS rates in the North and East. Low birth weight Indicator of low birth weight Indicator % Birth weight below 2.5 kg Numerator: # of live births in the 12 months prior to the survey among female respondents that are estimated to have weighed below 2500 grams at birth 52 Denominator: # of live births in the 12 months prior to the survey among female respondents in the sample Infants who weigh less than 2500 grams (2.5 kg.) at birth are categorized as low birth weight babies (LBWB). Since many infants are not weighed at birth and those who are weighed may be a biased sample of all births, reported birth weight cannot be used to estimate the prevalence of low birth weight among all children. Therefore, the percentage of births weighing below 2500 grams is estimated from two items in the questionnaire: the mother’s assessment of the child’s size at birth (i.e., very small, smaller than average, average, larger than average, very large) and the child’s weight as recorded on a health card if the child was weighed at birth. Birth weight documentation was shown to enumerators for seven percent (n = 67) of the total number of births (n = 910 unweighted, 964 weighted) in the MICS2. The following method is used to calculate the percentage of LBWB. First, for those 67 births where data are available for both size and weight, the two items (i.e., size and weight) are cross- tabulated for those children who were weighed at birth to obtain the proportion of births in each category of size who weighed less than 2500 grams (see Chart 2 below). For each category of size, the applicable proportion is then multiplied by the total number of children (out of the total pool of 910 births) to obtain the estimated number of children in the size category who were of low birth weight. The numbers for each size category are summed to obtain the total number of low birth weight children. This number is divided by the total number of live births (i.e., 910) to obtain the percentage of children with low birth weight. It should be noted that this method is Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 42 based on the estimations (of size) of a select group of women (i.e., women who had their children weighed at birth—presumably more educated, well-to-do women with access to higher-quality- than-average health care). In the case of this survey, the primary problem with this methodology is that the sample size of women who have recalled their child’s weight is very small, especially when it is stratified into different categories of size. The resulting estimate of percent LBWB is highly imprecise and should be interpreted with caution. Chart 2: Comparison of documented birth weight with caretakers’ estimate of size Size as reported by caretaker Very small Smaller than average Average Larger than average Very large Total Percentage of births less than 2500 grams 80 (4/5) 0 (0/7) 19 (5/27) 23 (3/13) 20 (3/15) 22 (15/6 7) Using the procedure described above, the data suggest that 52 percent of infants in Sierra Leone weigh less than 2500 grams at birth (Table 20). The prevalence of low birth weight babies ranges from 49 percent in the West to 56 percent in the East. High levels of low birth weight babies also prevail across urban and rural areas and among mothers of all education levels. A second estimate of low birth weight could be gained by calculating the percentage of low-birth- weight children among those who had documented birth weight, while bearing in mind that this is a non-representative group of children. As noted above, caretakers showed enumerators documentation of birth weight for 67 of the 910 children born in the last year. Twenty-two percent of these children (15/67) had documented birth weights below 2500 grams. Discussion: Low birth weight · The estimate of the percentage of Sierra Leonean children with a birth weight below 2,500 grams (53 percent) is much higher than previous estimates in Sierra Leone (11 percent), estimates in the region (15 percent for Sub-Saharan Africa) and neighboring countries (12, 13 and 21 percent in Ivory Coast, Guinea and Burkina Faso; source for all estimates: SOTWC)). · The high rate of low-birth-weight children suggests a high level of maternal malnutrition in Sierra Leone. · The World Summit for Children goal for this indicator is ten percent, well below its present value in Sierra Leone. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 43 E. Child Health Immunization Coverage Indicators for immunization coverage Indicator % Percentage of children aged 12- 23 months currently immunized against tuberculosis Numerator: # of children aged 12-23 months who received a BCG vaccination according to documentation or guardian’s report 73 Percentage of children aged 12- 23 months currently immunized against polio Numerator: # of children aged 12-23 months who received the OPV3 vaccination according to documentation or guardian’s report 61 Percentage of children aged 12- 23 months currently immunized against diphtheria, pertusis and tetanus Numerator: # of children aged 12-23 months who received the DPT3 vaccination according to documentation or guardian’s report 46 Percentage of children aged 12- 23 months currently immunized against measles Numerator: # of children aged 12-23 months who received a measles vaccination according to documentation or guardian’s report 62 Percentage of children aged 12- 23 months currently immunized against tuberculosis, polio, diphtheria, pertussis tetanus, and measles Numerator: # of children aged 12-23 months who received ALL of the following vaccinations: BCG, OPV3, DPT3, and measles, according to documentation or guardian’s report 39 Denominator (for all indicators): # of children aged 12-23 months in the sample According to UNICEF and WHO guidelines, a child should receive a BCG vaccination to protect against tuberculosis, three doses of DPT vaccine to protect against diphtheria, pertussis, and tetanus, three doses of polio vaccine (OPV), and a measles vaccination. All of these vaccinations should be administered by the age of 12 months. In the MICS2 survey, mothers were asked to provide vaccination cards (i.e., Health Cards) for children under the age of five. Interviewers copied vaccination information from the cards onto the survey questionnaire. If the child did not have a card, the mother was asked to recall whether or not the child had received each of the vaccinations and, for DPT and Polio, how many times. Mothers who recalled their children’s vaccination history were not asked to estimate the child’s age at vaccination. Surveyors did not exam children for the presence of a BCG scar. A perfect measure of vaccination coverage does not exist. The primary method used to measure vaccination coverage in the MICS2 survey has been to construct an indicator based on both documentation and a caretaker’s report. According to this method of calculating coverage, a child is considered to be vaccinated with a particular antigen if either the child’s Health Card shows that the pertinent vaccination was given or the child’s caretaker states that the vaccination was administered. This method of calculating coverage may yield an estimate of coverage that is somewhat higher than the true coverage level, if some caretakers exhibit faulty recall (i.e., if some caretakers recall that their children were vaccinated when they really weren’t). However, this combination indicator of recorded/recalled coverage is generally considered to be the most accurate indicator available for many countries. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 44 It should be noted that the indicator percent vaccine coverage for (antigen) by age one cannot be measured for all surveyed children from the data collected in the Sierra Leone MICS2, because vaccination data obtained through caretaker recall was not date-specific. Percent vaccine coverage for (antigen) by age one can be calculated for the sub-group of children with vaccination cards. However, this indicator will not be useful, as children with vaccination cards would be expected to have higher coverage rates than the general population of children. The denominator for all vaccination coverage indicators reported here thus consists of the number of children aged 12-23 months who were surveyed. This age group of children is used because they are the youngest age group of children who were old enough to be fully vaccinated at the time of the survey. Calculating coverage indicators for this group permits use of both recall and card-based data and yields the most useful and accurate indicator of coverage. Thirty-five percent of children in Sierra Leone aged 12-23 months had health cards. The percentage of caretakers who were able to show the interviewer a vaccination card for their child aged 12-23 months was 28 percent in the North, 35 percent in the East, 43 percent in the South, and 50 percent in West. Forty-three percent of caretakers were able to show vaccination cards to the surveyors in urban areas as compared to thirty-three percent in rural areas. Table 21 in the appendix, as well as the figure below, show the percentage of children aged 12 to 23 months who received each of the vaccinations. Approximately 73 percent of children aged 12- 23 months received a BCG vaccination while the first dose of DPT was given to 68 percent. The percentage coverage for subsequent doses of DPT was 58 for DPT2 and 46 percent for DPT3. Corresponding coverage figures for polio were 82 percent for OPV1, 76 percent for OPV2 and 61 percent for OPV3. The coverage rate for measles vaccine was 62 percent. Thirty-nine percent of surveyed children aged 12-23 months had received all of these vaccines. Figure 8: Percentage of children aged 12-23 months who received vaccinations by the time of the survey, Sierra Leone, 2000 73 68 82 6259 76 3946 61 0 20 40 60 80 100 BCG DPT Polio Measles All P e r c e n t Dose 1 Dose 2 Dose 3 Vaccination coverage for male children is consistently about five percentage points higher than for female children for all antigens (Table 22). Vaccination rates for children living in urban areas are 10-20% higher than for children living in rural areas, depending on the vaccine. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 45 Vaccination rates are generally higher in the Southern and Western regions than in the East and North, although the gap is lower for OPV coverage (5-15%) than for other antigens (15-30%). Vaccination coverage is consistently highest among children whose mothers have completed primary level education; the children of uneducated mothers and mothers who had completed secondary education had lower coverage rates. This latter finding is probably due to a statistical anomaly and is therefore best regarded as a spurious result. There were only 29 caretakers (a low sample size for drawing conclusions of this type) of children aged 12-23 months who had primary school education (as opposed to 443 uneducated caretakers and 77 caretakers with secondary or higher education). The 29 caretakers with primary education were disproportionably from the Western Area and from urban areas (where vaccination coverage is higher). The observed result is therefore probably more related to the place of residence of the 29 caretakers than to their educational background. Discussion: Immunization coverage · The neediest groups of children in Sierra Leone have the lowest coverage. Coverage rates measured in this survey were generally found to be moderately lower among traditionally underserved groups: girls, children living in rural areas, children living in areas of the country with limited access due to armed conflict (i.e., Northern and Eastern regions), and children of uneducated mothers. · Coverage rates among regions were most similar for polio vaccination, suggesting that the mass vaccination of children against polio through the conduct of National Immunization Days—in coordination with the RUFP—has achieved considerable success and should be continued. This conclusion is supported by the finding (reported above) that rates of Vitamin A supplementation coverage was higher in rebel-controlled areas than in the Southern region or Western Area. · Drop-out rates (i.e., the percentage of children who receive one vaccine in the OPV/DPT series but do not receive the subsequent vaccine) were much higher among children who did not have Health Cards as compared to those who did. This finding suggests that efforts to ensure that all children who are vaccinated be issued a Health Card and have their vaccinations recorded there should be intensified. · The World Summit for Children goal for vaccination coverage is 90 percent for all antigens for children one year of age. The coverage rates reported above, measured for children aged 12-23 months (i.e., a more liberal measure of coverage than that used for the WSC goal), fall far short of that goal. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 46 Diarrhea Indicators for home management of diarrheal disease Indicator % Prevalence of diarrhea Numerator: # of children aged 0-59 months whose caretaker reported that they had diarrhea in the previous 2 weeks 25 Denominato r: # of children aged 0 -59 months in the sample ORT use Numerator: # of under-five children whose caretaker reported that the child received oral rehydration salts, breast milk, gruel, other milk or infant formula, water with feeding, or a locally acceptable ORS substitute while they had diarrhea 86 Increased fluids during diarrhea episode Numerator: # of children aged 0-59 months whose caretaker reported that they drank more fluids than usual while they had diarrhea 56 Continued feeding during diarrhea episode Numerator: # of children aged 0-59 months whose caretaker reported that they ate somewhat less, the same, or more than usual while they had diarrhea 46 Home management of diarrhea Numerator: # of children aged 0-59 months whose caretaker reported that they d rank more fluids than usual AND ate somewhat less, the same, or more than usual while they had diarrhea 28 Denominato r: # of children aged 0-59 months whose caretaker reported that they had diarrhea in the previous 2 weeks Dehydration caused by diarrhea is a major cause of mortality among children in Sierra Leone. Appropriate home management of diarrhea—through the administration of either oral rehydration salts (ORS) or a recommended home fluid (RHF)—can prevent many of these deaths. Preventing dehydration and malnutrition by increasing fluid intake and continuing to feed the child who is ill with diarrhea are other important strategies for the home management of diarrhea. In the MICS2 questionnaire, mothers (or caretakers) were asked to report whether their child had had diarrhea in the two weeks prior to the survey. If the caretaker answered “yes”, s/he was asked a series of questions about what the child had to drink and eat during the episode and whether this was more or less than the child usually ate and drank. Overall, 25 percent of under- five children had diarrhea in the two weeks preceding the survey (Table 23). The prevalence of diarrhea was lower in the South (15%) and West (22%) than in the East and North (~28% in each region). Among children of different age groups in the survey, diarrhea prevalence peaked among children being weaned and/or being introduced to solid foods (i.e., children aged 6-23 months). Table 23 also shows the percentage of children receiving various types of recommended liquids during the episode of diarrhea. Since mothers were asked to name all types of liquid that the child received, and all liquids were recorded, the percentages sum to more than 100 across strata (e.g., for male children, or children living in the Eastern region). Seven in ten children aged less than twenty-four months received breast milk while they had diarrhea. About 14 percent of children with diarrhea received gruel while 42 percent received ORS. Children of mothers with secondary education and above appear more likely than other children to receive ORS. Eighty- six percent of children with diarrhea received one or more of the recommended home treatments (i.e., were treated with ORS or a RHF). Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 47 Slightly more than half (56%) of under-five children with diarrhea drank more than usual while 38 percent drank the same or less (Table 24). About 46 percent ate somewhat less, the same, or more than usual while 51 percent ate much less than usual or nothing. Overall, only 28 percent of children with diarrhea received increased fluids and continued eating as recommended. Discussion: Diarrhea · The prevalence of diarrhea reported by caretakers in this survey is quite high. The "diarrhea season" in Sierra Leone takes place at the beginning of the rainy season in May and June, at which time diarrhea prevalence rates in excess of thirty percent can be predicted. Data for this survey was collected throughout April and into the first week of May: that is, right before the commencement of the rainy season. · Although the percentage of children with diarrhea who receive ORT is high (86 percent), the percentage of children who receive increased fluids during their episode (56 percent) has substantial room for improvement. In Sierra Leone’s hot and humid climate, the high prevalence of diarrhea coupled with low levels of adequate fluid intake put the children in Sierra Leone at substantial risk of potentially fatal dehydration due to diarrhea. Acute respiratory infection (ARI) Indicators for home management of acute respiratory infection Indicator % Prevalence of ARI Numerator: # of children aged 0-59 months who had an illness in the previous two weeks with a cough accompanied by rapid or difficult breathing and whose symptoms were due to a problem in the chest, or both a problem in the chest and a blocked nose, or whose mother did not know the source of the problem. 9 Denominator: # of children aged 0-59 months in the sample Care seeking for ARI Numerator: # of children aged 0-59 months whose caretaker reported that they took their child to a hospital, health center, dispensary, village health worker, MCH or mobile/outreach clinic, or private physician while they had ARI 50 Denominator: # of children aged 0-59 months who had an illness in the previous two weeks with a cough accompanied by rapid or difficult breathing and whose symptoms were due to a problem in the chest, or both a problem in the chest and a blocked nose, or whose mother did not know the source of the problem. Acute lower respiratory infections, particularly pneumonia, are one of the leading causes of child mortality in the world. In the MICS2 questionnaire, children with acute respiratory infection are defined as those who had an illness with a cough accompanied by rapid or difficult breathing and whose symptoms were due to a problem in the chest, or both a problem in the chest and a blocked nose, or whose mother did not know the source of the problem. Only nine percent of under-five children had an acute respiratory infection in the two weeks prior to the survey according to these criteria (Table 25). Of these, 17 percent were taken to a hospital for treatment, and 22 percent were taken to a health center. Overall, 50 percent of children with ARI were taken to an Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 48 appropriate health provider (i.e., hospital, health center, dispensary, village health worker, MCH or mobile/outreach clinic, or private physician). Discussion: Acute respiratory infection · Data here again highlight the traditional mismatch of need and treatment. Although children in the North endure the highest prevalence of ARI in the country (11 percent in the North versus 4 to 9 percent in other regions), they are the least likely to be treated by an appropriate health provider (40 percent in the North versus 53 to 68 percent in other regions). · The prevalence of ARI as measured in this survey is modest. ARI season peaks in Sierra Leone in July and August. Data for this survey was collected in April and early May. Recognition of the symptoms for ARI is generally difficult for non-health workers The Integrated Management of Childhood Illnesses (IMCI) initiative Indicators for IMCI Indicator % Prevalence of illness Numerator: # of children aged 0-59 months whose caretaker reported that they were ill in the previous 2 weeks 62 Denominator: # of children aged 0-59 months in the sample Correct home management of illness Numerator: # of children aged 0-59 months whose caretaker reported that they drank more fluids than usual AND ate slightly less, the same, or more than usual while they were ill 30 Denominator: # of children aged 0-59 months whose caretaker reported that they were ill in the previous 2 weeks Correct care- seeking knowledge Numerator: # of caretakers of children aged 0-59 months who reported at least two of the following as symptoms that would cause them to take their child to a health facility right away: · develops a fever · blood in stool · child becomes sicker · inability of child to breastfeed · fast breathing · difficult breathing · drinking poorly 77 Denominator: # of caretakers of children aged 0-59 months The Integrated Management of Childhood Illnesses (IMCI) is a program developed by UNICEF and WHO that combines strategies for control and treatment of five major killers of children: acute lower respiratory tract infections, diarrheal dehydration, measles, malaria, and malnutrition. The program focuses on the improvement of case management skills by health workers, improvement of the health system, and improvement of family and community practices in the prevention and early management of childhood illnesses. Appropriate home management of illness is therefore one component of IMCI. The IMCI program advocates that mothers practice home management of diarrhea or any other illness by increasing the sick child's fluid intake and continuing to feed the sick child as s/he is normally fed. While Sierra Leone does not have a formally established IMCI program, many of the practices and principles of IMCI are advocated by the MOH and other non-governmental agencies and organizations in Sierra Leone. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 49 Table 26 presents information on the drinking and eating behavior of sick children. Almost two out of every three children were reported to have had diarrhea or some other illness in the two weeks preceding the survey. Of these, 55 percent drank more liquids during the illness and 48 percent continued eating (i.e., ate somewhat less, the same, or more). Overall, 30 percent of ill children received increased fluids and continued eating as recommended under the IMCI program. Promoting knowledge among caretakers about when it is appropriate to seek care for ill children is another important component of the IMCI program. In the Sierra Leone MICS2, caretakers of children were asked to name all of the symptoms that would cause them to take a child to a health facility right away. There was some variation in how this question was asked among regions. In the Northern region, there is often no health facility close enough so that a caretaker might feasibly take a sick child to be treated there. Enumerators in the North therefore (mistakenly) asked “During your child’s illness, what types of symptoms would cause you to take your child to see somebody who could treat the illness right away?” thus including traditional healers and witch doctors in their list of possible treatment alternatives. The question was asked in a uniform, correct manner in other regions, using the word “health facility” to define the care alternatives being referred to. Seventy-seven percent of caretakers were able to correctly state at least two symptoms that would cause them to take their child to a health facility immediately. The most common response, given by 82 percent of caretakers, was that they would take their child to a health facility right away if s/he developed a fever (Table 27). Forty-five percent said that blood in stool would cause them to take the child to a health facility and 43 percent mentioned a child becoming sicker. Between 7 and 28 percent of caretakers cited an inability to breastfeed, fast breathing, difficult breathing, and drinking poorly as reasons for taking a child to a health facility right away. Among the regions, the results surprisingly suggest that caretakers in the Western Area (by far the most developed area of Sierra Leone) are notably less likely than caretakers in other regions to know the signs for seeking care immediately. Overall, 47 percent of caretakers in the West know at least two signs for seeking care compared to 76 to 91 percent in the remaining regions. These regional differences are also reflected in the urban-rural and educational differentials. Rural caretakers and those with no education were more likely to mention at least two signs for seeking care than other caretakers. Discussion: Integrated Management of Childhood Illnesses · The level of reported child morbidity among under-fives during the two weeks prior to the survey was very high (62 percent). A large percentage of the children in Sierra Leone are malnourished and drinking water from unsafe sources, thus rendering them susceptible to different disease agents. This statistic is one of the most important indicators of the low state of preventive health care measures in Sierra Leone. · The low level of the indicator correct home management of illness (30 percent) mirrors the result for the same indicator for correct home management of diarrhea (28 percent). The majority of caretakers in Sierra Leone have either not received—or have not acted upon—the message that continued fluids and feeding are vital aspects of the home management of children’s disease. · Care-seeking knowledge as measured in the MICS2 is high in all regions except for the Western Area. Supervisors were closely questioned regarding the surprising result s among regions for the indicator correct care-seeking knowledge. Supervisors were adamant that the results showing lower levels of knowledge in the Western Area than in the remainder of the country reflected real differences and were not due to inter-regional differences in the way Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 50 that the question was asked (e.g., in the degree of probing by enumerators for additional answers). It was noted that urban mothers may be more competent in managing illness and thus may not feel the need to seek treatment at a health facility for as comprehensive a set of conditions as their rural counterparts. It was also felt that one should be careful not to assume that rural or uneducated caretakers would have lower levels of treatment-related knowledge than urban, educated caretakers. Several supervisors felt that it was plausible that uneducated mothers would take better care of their children than educated mothers, thus explaining these findings. Malaria Indicators for protection from and treatment of malaria Indicator % Use of bednets Numerator: # of children aged 0-59 months whose caretaker reported that they slept under an insecticide-impregnated bednet last night 2 Denominator: # of children aged 0-59 months in the sample Malaria treatment Numerator: # of children aged 0 -59 months whose caretaker reported that they received chloroquine or Fansidar while they were ill with fever 61 Denominator: # of children aged 0-59 months whose caretaker reported that they were ill with fever in the previous 2 weeks Malaria is the leading cause of death of children under age five in the world. It also contributes to anemia in children and is a common cause of school absenteeism. Preventive measures, especially the use of mosquito nets treated with insecticide, can dramatically reduce malaria mortality rates among children. In areas where malaria is common, international recommendations suggest treating any fever in children as if it were malaria and immediately giving the child a full course of recommended anti-malarial tablets. Children with severe malaria symptoms, such as fever or convulsions, should be taken to a health facility. Also, children recovering from malaria should be given extra liquids and food and should continue breastfeeding. The MICS2 questionnaire incorporates questions on the use of bednets among children. Fifteen percent of under-five children slept under a bednet the night prior to the survey interview (Table 28). This percentage is constant across age groups. Overall, only about ten percent of the few bednets that are used are impregnated with insecticide. Multiplying these two indicators together reveals that less than two percent (1.5 percent) of under-five children sleep under impregnated bednets in Sierra Leone. Questions on the prevalence and treatment of fever were asked for all children under age five. Almost one in two (46%) under-five children were ill with fever in the two weeks prior to the MICS2 (Table 29), a truly alarming level of morbidity. The prevalence of fever is approximately the same for children of all ages groups except for children aged less than six months, for whom it is somewhat lower (34 percent). Urban-rural differences in this indicator are not large, although modest differences are seen across regions, where the prevalence of fever ranges from 35 percent in the South to 50 percent in the North. Mothers were asked to report all of the medicines given to a child during their illness, both any medicine given at home and medicines given or prescribed at a health facility. Approximately 66 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 51 percent of children were given paracetamol and 60 percent were given chloroquine while only four percent were given Fansidar. A relatively large percentage of children (43 percent) were given some other medicine. Sixty-one percent of under-five children in Sierra Leone who were ill with fever received appropriate anti-malarial treatment. Children with fever in the North appear to be somewhat less likely to receive an appropriate anti-malarial drug (53 percent) than children in the East, South and West (72, 70 and 64 percent, respectively). Urban and rural children were equally likely to be treated appropriately and differences in this indicator across the education levels of mothers are modest. Discussion: Malaria · The portrait painted by these data is disturbing. Almost one in two under-five children in Sierra Leone has been ill with fever within the two weeks prior to the survey and six of ten ill children have taken anti-malarial medication. This suggests that thirty percent of under-five children are taking anti-malarial medication every two weeks. While it is encouraging that access to medication is reasonably high, it is far from ideal to constantly treat young children with strong medication. Among other potential harmful side effects, this practice eventually leads to the development of drug-resistant strains of malaria if treatment is not given correctly. · These data clearly show that the approach to malaria control in Sierra Leone is curative rather than preventive. Half of the children are sick and the majority of ill children are taking medication, while almost nobody is using bednets—much less impregnated bednets. Impregnated bednets have proven effect against malaria morbidity. Programs to promote and support their use in Sierra Leone should be developed and implemented. · The Northern region again stands out as the region with the greatest need (in terms of morbidity) and the least resources (in terms of prevention and treatment). The percentage of under-fives ill with fever in the North is higher than in any other region, while the percentage of ill children receiving appropriate anti-malarial drugs and the percentage of children sleeping under an impregnated mosquito net are lower in the North than in any other region. Within the constraints imposed by the ongoing conflict with rebel forces in Sierra Leone, the Northern region should clearly be the recipient of a disproportionately high percentage of resources targeting health and education problems. F. HIV/AIDS Indicators for monitoring knowledge of HIV/AIDS Indicator % Knowledge of preventing HIV/AIDS Numerator: # of women 15-49 years who correctly state the three main ways of avoiding HIV infection 21 Knowledge of misconceptions of HIV/AIDS Numerator: # of women aged 15-49 years who correctly identified three misconceptions about HIV/AIDS 19 Knowledge of mother-to- child transmission of HIV Numerator: # of women aged 15-49 years who correctly identify means of transmission of HIV from mother to child 34 Attitude to people with HIV/AIDS Numerator: # of women aged 15-49 years who express a discriminatory attitude towards people with HIV/AIDS 23 Women who know where to be tested for HIV Numerator: # of women aged 15-49 years who know where to get HIV test 9 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 52 to be tested for HIV where to get HIV test Women who have been tested for HIV Numerator: # of women aged 15-49 years who reported they had been tested for HIV 2 Denominator: # of women aged 15-49 years in the sample AIDS knowledge A major strategy for the reduction of the rate of HIV/AIDS infection is the promotion of accurate knowledge of how AIDS is transmitted and how to prevent transmission. Among women aged 15-49 in Sierra Leone, 54 percent have heard of AIDS (Table 30). This percentage is higher in urban areas (78%) than in rural areas (44%). Women interviewed in the MICS2 were given several statements about the means of HIV/AIDS transmission and asked to state whether they believed the statements were true. Among surveyed women, twenty-nine percent correctly believed that having only one uninfected sex partner can prevent HIV transmission. Twenty-seven percent correctly believed that using a condom every time one has sex can prevent HIV transmission and twenty-seven percent correctly believe that abstaining from sex prevents HIV transmission. Overall, 21 percent of surveyed women correctly knew all three ways of preventing HIV transmission and 33 percent were aware of at least one of the means of preventing transmission. Women who had never heard of HIV/AIDS were assumed to have inaccurate knowledge and were not given any statements about HIV/AIDS transmission. Accurate knowledge of the means of HIV/AIDS transmission varied dramatically across regions. The percentage of women who knew all three ways of preventing HIV transmission was 2 percent in the North, 43 percent in the East, 20 percent in South, and 33 percent in the West. Higher levels of a woman’s formal education were strongly associated with correct knowledge regarding HIV transmission. There is no clear association between a woman’s age and her knowledge regarding HIV transmission. Another category of inquiry that was assessed in the MICS2 survey concerned women’s misconceptions regarding HIV/AIDS. Thirty-three percent of women correctly stated that AIDS can’t be transmitted by supernatural means while 26 percent correctly stated that AIDS can’t be spread by mosquito bites (Table 31). Thirty-two percent of the women correctly believe that a healthy-looking person can be infected with HIV. Nineteen percent of respondents correctly identified all three misconceptions. Only 4 percent of the women in the North have knowledge of all three misconceptions about AIDS transmission, underscoring the low levels of knowledge regarding HIV-related issues in the North, as compared to other regions in Sierra Leone. Women in the Eastern and Western regions were more likely to recognize all three misconceptions than women in the North and South. Women in urban areas (32%) were more likely to recognize misconceptions about HIV prevention and transmission than women in rural areas (13%). Thirty-four percent of women in Sierra Leone correctly believe that AIDS can be transmitted from mother to child (Table 32). When asked specifically about the mechanisms through which mother to child transmission can take place, 32 percent said that transmission during pregnancy was possible, 30 percent said that transmission at delivery was possible, and only 30 percent agreed that AIDS can be transmitted through breast milk. Twenty-seven percent knew all three modes of transmission. Approximately, 67 percent did not know any specific way that HIV can be transmitted from mother to child. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 53 The MICS survey also attempted to measure discriminatory attitudes towards people who have HIV/AIDS. To this end, respondents were asked whether they agreed with two questions. The first asked whether a teacher who has the HIV virus but is not sick should be allowed to continue teaching in school. The second question asked whether the respondent would buy food from a shopkeeper or food seller who the respondent knew to be infected with HIV. The results are presented in Table 33. Twenty three percent of the women aged 15-49 years believe that a teacher with HIV/AIDS should not be allowed to work. This percentage is highest in the Eastern region at 44 percent and lowest in the Northern region at 8 percent. Taking into account that women who had not heard of HIV/AIDS were not asked this question, 42% (22.6/54.2) of women who have heard of HIV/AIDS expressed this discriminatory attitude. Urban women and those with secondary or higher education are more likely to express this discriminatory attitude than rural women and those who have completed primary education or who are uneducated. Five percent of women would not buy food from a person infected with AIDS. Interestingly, this measure shows a different regional pattern than the first question. Fourteen percent of women in the West stated that they would not buy food from a person with HIV/AIDS, while less than five percent of the women of the remaining regions express this sentiment Overall, 23 percent of women agree with at least one of the discriminatory statements. Table 34 summarizes information from two previous tables on AIDS knowledge (Tables 30 and 31). The second column shows the percentage of women who know all three means of preventing HIV transmission – having one faithful uninfected partner, using a condom during every sexual encounter, and abstaining from sex. Twenty one percent of women know all three ways. The third column of the table shows the percentage of women who correctly identified all three misconceptions about HIV transmission: that HIV can be transmitted through supernatural means, that it can be transmitted through mosquito bites, and that a healthy looking person cannot be infected. Nineteen percent of women correctly identified these misconceptions. Finally, the fourth column of the table shows the percentage of women who have “sufficient knowledge” of HIV/AIDS transmission. These are women who know all three ways of preventing HIV transmission and correctly identified all three misconceptions. Only 13 percent of women aged 15-49 fall into this category. Knowledge of HIV/AIDS transmission varies dramatically by level of education (Figure 9). Women with secondary or higher education are almost two times more likely to know all three ways to prevent transmission than women with no education. They are also three times more likely to correctly identify all three misconceptions about AIDS and twice more likely to have sufficient knowledge of HIV/AIDS transmission. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 54 Figure 9: Percentage of women aged 15-49 that has sufficient knowledge of HIV/AIDS transmission by level of education, Sierra Leone, 2000 0 10 20 30 40 50 60 70 80 90 100 Knows 3 ways to prevent transmission Correctly identified 3 misconceptions Has sufficient knowledge None Primary Secondary + AIDS testing Voluntary testing for AIDS, accompanied by counseling, allows those infected to seek health care and to prevent the infection of others. Testing is particularly important for pregnant women who can then take steps to prevent infecting their babies. The indicators shown in Table 35 are designed to monitor whether women are aware of places to get tested for HIV/AIDS, the extent to which they have been tested, and the extent to which those tested have been told the result of the test. In some places, a relatively large proportion of people who are tested do not return to get their results due to fear of having the disease, fear that their privacy will be violated, or other reasons. Nine percent of women of reproductive age in Sierra Leone know a place to get tested for AIDS. Women living in the Western Area are most likely to know a place, followed by those in the South, East and North, respectively. Only 4 percent of women with no education know of a place to get tested compared to 12 percent of women with primary school education and 29 percent of women with secondary or higher education. About 2 percent of women of reproductive age in Sierra Leone report that they have been tested for AIDS. Again, this percentage is highest in the Western Area at 5 percent and lowest in the Northern region (1 percent). The majority (59 percent) of women who have been tested were told the result. However, variation in this indicator exists across regions, age groups, and education levels. Among regions, women in the North are least likely to have been told their result. Adolescent women (age 15-19) are the least likely of any age group to have been tested and least likely to know the result. Women with no education are less likely than educated women to have been tested. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 55 Discussion: HIV/AIDS · The vast majority of women of child-bearing age in Sierra Leone have either never heard of HIV/AIDS or lack basic knowledge regarding this disease. This lack of knowledge about such a serious public health problem has clear potential consequences for the spread of HIV/AIDS in the country. Given the paucity of mass media channels in Sierra Leone, innovative and effective avenues for conducting information dissemination campaigns on this and other issues must be identified and exploited. The Northern and Southern regions should be targeted, given the exceptionally low knowledge levels there. · Women who have heard of HIV/AIDS expressed high levels of discriminatory attitudes towards individuals with this disease. Any communication campaign that is developed must clearly focus on minimizing discrimination through correct knowledge. The data suggest that the Eastern region, where 57 percent (44.2/77.4) of women who have heard of HIV/AIDS express discriminatory attitudes, should receive special attention in this effort. · The proportion of women who know where they can be tested for HIV is extremely low. Any effort to provide information about HIV/AIDS in Sierra Leone must also include information regarding where facilities for testing for HIV are located. Efforts in this regard may need to include efforts to increase the number of laboratory facilities that offer HIV testing, especially in the Northern, Southern and Eastern regions. G. Reproductive Health Contraception Indicators for contraceptive use Indicator % Contraceptive prevalence Numerator: # of married or in union women aged 15-49 years who report that they or their partner are using a contraceptive method (either modern or traditional) 4.3 Denominator: # of women aged 15-49 years in the sample. Indicators of contraceptive use, or prevalence, can be reported for three groups of women of reproductive age (i.e., aged 15-49): 1) for all women, 2) for all women who are married or in union, and 3) for all women who report that they are “sexually active”. In the MICS2 survey, 4.3% of women who reported that they were married or in union used contraceptives (3.9 percent modern, 0.4 percent traditional; see Table 36). Among women who reported that they were sexually active, 7.3 percent reported that they used contraceptives (Table: 36a). Six percent (297/4923) of all surveyed women of reproductive age reported that they use contraceptives. Among women of reproductive age who are married or in union in Sierra Leone, the pill and injections were the most commonly used contraceptive methods. One in forty respondents (2.5 percent) reported that she took the pill and 0.9 percent of respondents reported that they used injections. Contraceptive use among women who are married or in union is highest in the Western Area at 14 percent. This level of use is at least four times higher than in any other region. Women’s level of education is strongly associated with their use of contraceptives. The percentage of women using any method of contraception rises from 3 percent among those with Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 56 no education to 8 percent among women with primary education and 14 percent among women with secondary or higher education. Prenatal care Indicators for prenatal care coverage Indicator % Prenatal care Numerator: # of women aged 15-49 years who gave birth in the last year and were seen at least once for antenatal care during their pregnancy by a physician, nurse, midwife or auxiliary midwife 68 Child birth care Numerator: # of women aged 15-49 years who gave birth in the last year and whose delivery was attended by a physician, nurse, midwife or auxiliary midwife 42 Children protected against neonatal tetanus Numerator: # of women aged 15-49 who gave birth in the last year and reported that they had received either: 1) 2 doses of tetanus toxoid (TT), the last within 3 years; or, 2) 3 doses of TT, the last within 5 years; or, 3) 5 doses of TT during their lifetime. 58 Denominator: # of women in the sample who reported that they gave birth in the year prior to the survey Quality prenatal care can contribute to the prevention of maternal mortality by detecting and managing potential risk factors, including pre-eclampsia, anemia, and sexually transmitted diseases. Antenatal care also provides opportunities for women to learn the danger signs of pregnancy and delivery, to be immunized against tetanus, to learn about infant care, and be treated for existing conditions, such as malaria and anemia. Tetanus toxoid injections are given to women during pregnancy to protect infants from neonatal tetanus, a major cause of infant death that is due primarily to unsafe and unclean delivery practices. Two doses of tetanus toxoid during pregnancy offer protection from neonatal tetanus. However, if a woman was vaccinated during a previous pregnancy, she may only need a booster to give full protection. Five doses are thought to provide lifetime protection. Fifty-eight percent of mothers who gave birth in Sierra Leone in the 12 months prior to the MICS2 survey received sufficient doses of tetanus toxoid to protect their children against neonatal tetanus (Table 37). Almost all of these women (98 percent) achieved protection by receiving two or more doses of tetanus toxoid within the last three years. Among the regions, women living in the South are most likely to be protected (76 percent) while those living in the North are the least likely to be protected (45 percent). Long periods of deprivation of women in the Northern region from comprehensive reproductive health care is a major reason for this low coverage. A woman’s educational level is associated with the status of her protection against tetanus. The results of this survey show that a higher proportion of women with secondary or higher education (75 percent) are more likely to be protected against tetanus than those with either no education (55 percent) or primary education (64 percent). Female respondents who gave birth in the year prior to MICS2 were asked whether they had received antenatal care for the birth and, if so, what type of person provided the care. If the Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 57 woman saw more than one type of provider, all were recorded in the questionnaire. Table 38 presents the percent distribution of women with a birth in the year prior to the MICS2 by the type of personnel who delivered antenatal care. If more than one provider was mentioned by the respondent, she is categorized as having seen the most skilled person she mentioned. Figure 10: Percent distribution of women with a birth in the last year by type of personnel providing antenatal care, Sierra Leone, 2000 Physician 10% Nurse/ Midwife 56% Auxillary Midwife 2% Traditional Birth Attendant 14% None 15% Other 3% As the figure above shows, most women in Sierra Leone (85 percent) receive some type of antenatal care during their pregnancy. Approximately 68 percent of pregnant women receive antenatal care from skilled personnel (doctor, health assistant, nurse, midwife, auxiliary midwife). Women are more likely to receive antenatal care from a physician if she lives in the Western Area (38 percent) than if she lives in the other regions (where this indicator ranges from 6 percent in the South to 8 percent in the North). Most physicians in Sierra Leone, particularly obstetricians and gynecologists, reside in the Western Area. Assistance at delivery The provision of delivery assistance by skilled attendants can greatly improve outcomes for mothers and children by the use of technically appropriate procedures, accurate and speedy diagnosis, and treatment of complications. Skilled assistance at delivery is defined as assistance provided by a doctor, nurse, midwife, or auxiliary midwife. Approximately 42 percent of births occurring in Sierra Leone in the year prior to the MICS2 survey were attended by skilled personnel (Table 39). The percentage of births attended by skilled personnel is consistent across the Eastern, Southern and Western regions at 66, 57 and 64 percent, respectively. The exception to this trend was in the Northern region, where only 22 percent of births were attended by skilled personnel. The births of women with either primary or secondary levels of education were attended by skilled personnel more frequently than births by uneducated women. Thirty-four percent of births in Sierra Leone during the year prior to the MICS2 survey were attended by a nurse/midwife, while physicians attended 3 percent of births. Auxiliary midwifes attended 5 percent of births. The low percentage of deliveries attended by physicians suggests that very few women deliver at hospitals where comprehensive obstetric services are provided. Traditional Birth Attendants (TBAs) assisted at 38 percent of births nationally. Among regions, Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 58 TBAs were most widely used in the North, where they assisted at 44% of births. TBAs attended between 32 and 34% of births in the remaining three regions. Discussion: Reproductive health · The promotion of modern contraceptive methods among married or sexually active women and their partners in the Sierra Leonean population has been unsuccessful over the years. The prevalence of contraceptive use in Sierra Leone (4 percent) even lags behind the low standards of the region; the average value of this indicator is 17 percent in Sub-Saharan Africa, and is 12, 22 and 29 percent in The Gambia, Ghana and Guinea, respectively (source: SOTWC). Many would-be users of contraceptives lack adequate access to family planning commodities. Reasons for this lack of access include inadequate health facilities, a complete lack of family planning services within communities, and a lack of disposable income for the public to use to pay for these services. Another major constraint to the use of modern contraceptive methods is the unwillingness of partners to change from their old ways. For instance, despite the global promotion of condoms as a protective device against HIV/AIDS as well as an aid to family planning, this survey indicates that condoms are virtually unused as a method of contraception in Sierra Leone. · Even though many pregnant women in Sierra Leone attend prenatal clinics, a large proportion of them are only attended by TBAs during their deliveries. The data make this clear; a much larger percentage of women see skilled personnel for antenatal care (68 percent) than have their birth attended by skilled personnel (46 percent). The gap is made up by TBAs, who provide antenatal care for only 14 percent of mothers but attend 38 percent of births nationwide. Trained TBAs can adequately manage normal deliveries, however they are incapable of managing complicated deliveries. Any attempts by them to manage such cases may lead to delays in referral and proper management of complicated obstetric cases. Maternal deaths often result from delays or mistakes in the management of complicated deliveries. Given the heavy reliance on TBAs in Sierra Leone, TBAs need to be comprehensively trained to recognize at-risk pregnancies during ante-partum and complications during childbirth so that women at risk can be promptly referred to hospitals for specialized care. · The percentage of births attended by skilled personnel in Sierra Leone is as high or higher than most countries in the region. The value of this indicator is 36 percent in Sub-Saharan Africa, and is 27, 31 and 47 percent in Burkina Faso, Guinea, and Ivory Coast, respectively (source: SOWCR). Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 59 H. Child Rights Indicators for monitoring children’s rights Indicator % Birth registration Numerator: # of children aged 0-59 months whose births are reported registered. 47 Denominator: # of children aged 0-59 months in the sample Children not living with a biological parent Numerator: # of children aged 0-14 years who are not living with a biological parent 16 Orphans in household Numerator: # of children aged 0-14 years whose mother and father are dead 2.4 Denominator: # of children aged 0-14 years in the sample Child labor Numerator: # of children aged 5-14 years who do one or more of the following: 4) perform paid or unpaid work for someone who is not a member of the household 5) perform more than four hours of housekeeping chores per day in their own household 6) perform other family work (e.g., on a farm or in a business) 72 Denominator: # of children aged 5-14 years in the sample Birth registration The International Convention on the Rights of the Child states that every child has the right to a name and a nationality and the right to protection from being deprived of his or her identity. Birth registration is a fundamental means of securing these rights for children. The births of 47 percent of children aged less than five years in Sierra Leone have been registered (Table 40). There are no significant variations in birth registration across gender or age categories. The birth registration breakdown across regions shows that 27, 50, 67 and 81 percent of children have been registered in the Northern, Eastern, Southern and Western regions, respectively. During the MICS2 survey, caretakers who had not registered the birth of their child were asked to explain why. In the North, caretakers cited “didn’t know it should be registered” and “don’t know where to register” as the principal reasons for not having registered the birth. The cost of registration, having to travel too far for registration, and lack of knowledge regarding the registration process were all cited by significant proportions of caretakers in Sierra Leone’s remaining three regions as reasons why they had not registered the birth of their children. Orphanhood and living arrangements of children Children who are orphaned or living away from their parents may be at increased risk of impoverishment, discrimination, denial of property rights and rights to inheritance, various forms of abuse, neglect, and exploitation of their labor or sexuality. Monitoring the level of orphanhood and the living arrangements of children assists in identifying those who may be at risk and in tracking changes over time. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 60 In Sierra Leone 61 percent of children aged 0-14 are living with both parents (Table 41). Approximately 9 percent are living with their mothers only although their fathers are alive. About 10 percent of children are living with neither parent although both parents are alive. Sixteen percent of children live with neither of their biological parents and twelve percent of children aged 0-14 years have one or both parents dead. Older children are more likely to live away from their biological parents than young children; while only 7 percent of children under age five are not living with a biological parent, 23 percent of children aged 10-14 do so. The situation of children in the Western Area differs from that of other children in Sierra Leone. In the Western Area, less than half of children live with both parents. Nineteen percent of children live with their mother only although their father is alive and a relatively large proportion (21 percent) live with neither parent. This pattern is most likely due to the war and labor migration of men and, to some extent, women from the Western region to other regions and neighboring countries. Child labor It is important to monitor the extent to which children work and the type of work in which they participate for several reasons. Children who are working are less likely to attend school and more likely to drop out. This pattern can trap children in a cycle of poverty and disadvantage. Working conditions for children are often unregulated with few safeguards against potential abuse. In addition, many types of work are intrinsically hazardous and others present less obvious hazards to children, such as exposure to pesticides in agricultural work, carrying heavy weights and scavenging in garbage dumps. In Sierra Leone, the MICS2 survey estimates that only about two percent of children aged 5-14 years engage in paid work (Table 42). About twenty-five times as many (48 percent) participate in unpaid work for someone other than a household member. Domestic work is defined as cooking, shopping, cleaning, washing clothes, fetching water, and caring for children. Almost three in four children aged 5-14 years do these tasks for less than four hours a day while 10 percent spend more than four hours a day on such tasks. This result does not differ significantly across gender. Variations across regions are greatest in the percentage of children who engage in more than four hours of domestic work a day. This percentage ranges from 3 percent in the West regions to 14 percent in the North. Children who either 1) perform paid or unpaid work for someone who is not a member of the household, or 2) perform more than four hours of housekeeping chores in the household, or 3) perform other family work are considered to be currently working. Overall, 72 percent of surveyed children were classified as currently working. There is virtually no difference between boys and girls (72 percent of boys and 71 percent of girls). At the regional level, the percentage of children working is lowest in the West regions at 64 percent and highest in the North at 78 percent. Rural children (74 percent) are somewhat more likely to be currently working than urban children (66 percent). Discussion: child rights · The results regarding birth registration indicate that sensitization on Child’s Rights in Sierra Leone needs to be increased, especially in the Northern region. The large percentages of the population who are either ignorant of the birth registration process or who—despite their awareness—choose not to register the birth of their child suggests that the importance and significance of birth registration is not well understood. It is clearly significant that the areas Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 61 of lowest registration are those that are least accessible to mainstream society. This suggests a lack of facilities and systems to register births within these areas that is perhaps exacerbated by constraints in verifying supporting information needed in the birth registration process. Within a tense security environment, the issue of greater distances to travel to register a birth takes on increased importance. · Sensitization on child’s rights in terms of the responsibilities of parents towards children needs to be emphasized at all levels throughout the country, especially in the Northern region. Children must be aware of their rights and must be encouraged to attend school or learn a skill. In situations where assessments show that poverty will hinder a child’s development and threaten their rights, caretakers must be linked to poverty-alleviation programs and other programs that support income-generating activities. · The finding that 10 percent of children do not stay with their parents even though they are alive requires further investigation. If children living away from their parents do so with their consent and communicate regularly with them, then the risk that their rights will be abused is lower. Children are sent to live with relations at times because the parents believe that the relation has more to offer the child in terms of his or her education and development. · The percentage of orphans seems low and may not be representative of the number of orphans in the country as a whole. The MICS2 was a household survey. Many orphans live either in care systems or on the street and do not have an adult caretaker. These children would not be counted in a household survey. · The percentage of surveyed children that is classified as currently working is high (72 percent). This is due to the definition of work that was used in MICS2 and cultural norms as they relate to “work” in Sierra Leone. The performance of domestic work by children is routine in Sierra Leonean households as is agricultural work for children in rural communities. While the high value of an indicator of child labor such as children currently working should be noted and assessed, we cannot make assumptions about child labor and child development from this statistic alone. The work that a child performs is not necessarily exploitative and can have positive impact on a child’s life by 1) enhancing the child's "fit" or role in their family, 2) contributing to the overall functioning of the family's social and economic status, and 3) serving as one of the child's few entry points to subsistence. · In order to understand the true impact of the work a child performs we need to understand not only how much of the child's day is devoted to work but also what the child does when s/he is not working. This is a key issue in child labor. What is sometimes more important than the work itself is what else there is in the child's life to stimulate their development: for example, educational or recreational activities. I. Maternal Mortality Maternal mortality indicator Indicator Rate Maternal mortality rate (the number of maternal deaths out of 100,000 live births) 1,800 ± 800 One of the goals of the MICS2 effort in Sierra Leone was to attempt to estimate the Maternal Mortality Rate(MMR) at the national level. This indicator has not been measured in Sierra Leone for many years. The current MMR in Sierra Leone is estimated to be the highest of any country in the world at 1,800 maternal deaths per 100,000 live births. Even though it generally takes a much larger sample size than that of the MICS2 to estimate the MMR to a reasonable level of Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 62 precision, members of the MICS2 Advisory Committee did not want to pass up the chance to collect data to assess this important health indicator. The maternal mortality rate is defined as the number of [maternal deaths : live births] and is generally reported as the number of maternal deaths per 100,000 live births. In this survey, data was gathered to estimate the MMR through the recall of the head of household and of women of reproductive age. The head of every surveyed household was asked if a maternal death had occurred in his or her household in the previous year. The enumerators defined maternal deaths for all respondents according to standard international definition. If the respondent answered affirmatively, s/he was asked for more information about the death, including the date, the time at which the death took place (i.e., during pregnancy, during childbirth, or within 6 weeks after childbirth), and the cause of death. The Household Questionnaire in Appendix D contains the full sequence of instructions that were given and questions that were asked. Data on the number of live births during the year prior to the survey was gathered from births reported by women of reproductive age in the surveyed households. Respondents in 19 households reported that a maternal death had occurred in the 12 months prior to the survey. Deaths were only considered to be valid when the respondent could state the date of the death and when that date was less than 365 days prior to the date of the interview. Deaths were also considered valid if the respondent only recalled the month and year of the death (but not the day) and the recalled month was less than 12 months prior to the month of the interview. Women of reproductive age interviewed in this survey reported that they had given 910 live births in the year prior to the interview date. This value needs to be adjusted because a certain percentage of women (14.4%) of reproductive age who lived in the surveyed households were not interviewed. If these women had been interviewed, the number of live births reported to enumerators would presumably have been higher. Assuming that the proportion of women who gave birth in the year before the survey was equal among women who were both interviewed and not interviewed, we can estimate that there were 1,063 live births (910/0.856) among all women of reproductive age living in surveyed households in the year before the survey. The estimate of the MMR resulting from these data is 1,800 (19/1,063 x 100,000). This value is the same as current working estimates of the MMR, supporting the accuracy of the estimate. Given the small sample size of the survey, this estimate is highly imprecise. The level of precision can be calculated using the following standard formula: n pqnxp 96.1/ ±= The terms used in the equation are defined as follows: p the proportion of maternal deaths among all live births = 0.0179 x the number of maternal deaths in the observed sample = 19 n the estimated number of live births in the observed sample = 1063 q 1 – p = 0.9821 Plugging the values in the box above into the equation and then multiplying by 100,000 reveals that the 95% confidence limits for the MMR estimate are ± 800. The 95% confidence interval for the estimated MMR is thus equal to [1000, 2600]. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 63 Seven of the reported deaths occurred in the Northern region. Six of the deaths occurred in the Southern region, while three deaths took place in both the Eastern region and the Western Area. Respondents reported that six of the women died while pregnant, while another six died in labor. Four respondents died within 42 days following the birth of their children. Data was not available for the remaining three women. Of the six women who died while they were pregnant, four were reported to have died from an induced abortion. Four of the women died from obstructed labor, while three died from a ruptured uterus. Two died from sepsis while one each died from bleeding and eclampsia. Three women died from other causes. Three respondents did not know the cause of death. Seven of the women died at home, while another seven died at the hospital. Two died at a Primary Health Unit. Data on place of death is not available for the three remaining women. Data is available for seventeen women regarding their parity at the time of their deaths. One woman had never given birth while 14 of the women had given birth from one to three times. One of the two remaining women had given birth five times while the other had given birth six times. Discussion: Maternal mortality · The estimate of the MMR generated in this survey is extremely high. There is a 95 percent chance that the MMR is between 1,000 and 2,600 deaths per 100,000 live births. Even the lower end of this confidence interval represents a tragically high MMR. These data suggest that approximately 4,000 maternal deaths occur per year in Sierra Leone8. This can only be termed an epidemic and calls for drastic measures to reduce the MMR. · The number of maternal deaths about which information was gathered is too small to draw conclusions regarding where maternal deaths are the biggest problem in Sierra Leone, or which factors are associated with maternal deaths. It is clear that women from all regions of the country are at risk of maternal death and that observed maternal deaths were due to a variety of causes. Given that four of the maternal deaths were reported to be due to complications arising from induced abortions, it would appear that women who undergo abortions in Sierra Leone are at extreme risk of an adverse outcome. 8 A very rough approximation of the number of maternal deaths in Sierra Leone per year can be estimated by dividing the number of deaths measured in this survey by the proportion of the population in Sierra Leone that was surveyed. The number of people living the surveyed households was 24,375. In this case, 19 maternal deaths ¸ (24,375 / 5,000,000) = 3,897 (i.e., approximately 4,000) maternal deaths per year in Sierra Leone, assuming that the population of Sierra Leone is approximately 5 million. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 64 J. Disability Members of the MICS2 National Task Force felt that basic information about disability in Sierra Leone should be collected during the survey. During the survey, each head of household (or his or her surrogate) was asked if any member of the household was disabled. The cause of disability was investigated for each reported disability. The chart below describes the frequency of the disabilities that were reported as well as their causes. The data are reported for the 24,375 people who lived in the surveyed households. Disabilities and their causes were reported for 157 individuals living in surveyed households. Single disabilities were reported for 145 individuals, two disabilities for eleven people, and three disabilities for one individual, thus yielding a total of 170 disabilities among 157 individuals. The % disabled column on the far right shows the estimate generated by the MICS2 survey of the percentage of the Sierra Leonean population who are challenged by a particular disability. For example, 38 people in surveyed households were reported to be blind out of a total of 24,375 respondents (38/24,375 ´100 percent = 0.16 percent) Chart 3: Frequency of disability in Sierra Leone by type and cause, Sierra Leone, 2000 Cause of disability Type of disability From birth Accident War Illness Assault Other Missing Total % disabled Blindness 5 14 1 18 0 0 0 38 0.16 Crippled 1 10 3 16 1 0 3 34 0.14 Lost limbs 2 2 4 4 0 0 0 12 0.05 Deaf 12 4 2 17 0 0 1 36 0.15 Dumb 6 5 0 1 0 0 0 12 0.05 Mental disability 8 4 0 4 0 0 1 17 0.07 Other 3 2 2 12 0 2 0 21 0.09 Total (%) 37 (22) 41 (24) 12 (7) 72 (42) 1 (1) 2 (1) 5 (3) 170 (100) Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 65 Appendices Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 66 Appendix A: Survey costs Category Leones US $ Estimation of EA Population - 2,987,800 1,440 Demarcation of Segments - 12,400,000 5,976 Equipment and Stationery Supplies - 23,712,250 11,428 Pre – Test - 1,242,500 599 Printing and Photocopying - 26,200,000 12,627 Training of Supervisors & Enumerators - 6,407,000 3,088 Data Collection - 40,018,000 19,286 Identity Cards - 325,000 157 Data Processing - 11,270,000 5,431 Analysis and Report Writing - 900,000 434 Administration and Coordination - 4,430,000 2,135 Publicity and Dissemination - 6,000,000 2,892 Technical Assistance - 44,612,500 21,500 Contingency - 13,589,255 6,549 Total - 194,093,425 93,539 Monetary Percent Contribution Contribution UNICEF - 157,391,435 81.1 UNFPA - 23,250,500 12.0 Action Aid Sierra Leone - 6,693,215 3.4 Christian Children’s Fund - 6,432,500 3.3 Christian Health Association S/Leone - 325,775 0.2 Total - 194,093,425 100.0 Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 67 Appendix B: List of personnel involved in Sierra Leone MICS2 Members of the National Task Force on MICS2 Mrs. S A Bockarie-Torto - Central Statistics Office (CSO) Mr. B H Bola-William - Christian Health Association of Sierra Leone (CHASL) Mr. S A T P Horton - Ministry of Health and Sanitation (MOHS) Mr. S I Jalloh - Ministry of Youths, Education and Sports (MYES) Ms Nafisatu Jalloh - UNHCR Mr. B M Javombo - Ministry of Development and Economic Planning (MODEP) Mrs. Ethel Johnson - Ministry of Information and Broadcasting (MIB) Mr. Davidson Jonah - Christian Children’s Fund (CCF) Mrs. Iris Juxon-Smith - Ministry of Social Welfare Gender and Children’s Affair (MSWGCA) Mr. Daniel Kaindeneh - CCF Mr. L S Kain - MODEP Mr. I A K Kanu - Ministry of Local Government and Rural Development (MLGRD) Mr. A C Lahai - FAO Mr. M B Lappia - Ministry of Safety and Security (MSS) Dr. Bailah Leigh - MSWGCA Mrs. Hawa Musa - MODEP Mr. Sahr Ngayenga - Plan International Ms Miatta Njambawai - Action Aid Sierra Leone Mr. John S N Pessima - CSO Mr. Morten Peterson - UNHCR Mr. Paul A Sengeh - UNICEF Ms Theresa Vamboi - MSWGCA Mr. Moses L J Williams - CSO Mrs. Frederica Wyse - UNFPA Survey personnel Senior staff Dr. H.B.S Kandeh - Survey Director Mrs. S A Bockarie-Torto - Survey Director Mr. John S N Pessima - Survey Coordinator Mr. Andrew B Johnny - Supervisor/Cartographer Mr. Abdulai A Jalloh - Supervisor/Programmer Mr. Alimamy Kamara - Supervisor/Verifier Mr. George D Katta - Supervisor/Verifier Mr. Mohamed King Koroma - Supervisor/Data Specialist Mrs. Muriel V MacCormack - Supervisor Mr. Lawrence Moiguah - Supervisor Mr. Moses L J Williams - Supervisor/Data Analyst Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 68 Associate supervisors Mr. Joseph S Jayah - Associate Supervisor (AFRC) Mr. Omoroh B L Marah - Associate Supervisor (AFRC) Mr. Alimamy F Turay - Associate Supervisor (AFRC) Mr. Edward Warner Jr. - Associate Supervisor (AFRC) Mr. Peter E T Z Adonis - Associate Supervisor (RUF) Mr. Romeo Ansumana - Associate Supervisor (RUF) Mr. Joe Fatorma - Associate Supervisor (RUF) Mr. Jackson Jakema - Associate Supervisor (RUF) Mr. Gassious Jimmie - Associate Supervisor (RUF) Mr. Ibrahim Koroma - Associate Supervisor (RUF) Mr. Brima Njawhen - Associate Supervisor (RUF) Listers Mr. Sheku Golfa - Lister Mr. Gerald Woode - Lister Enumerators Mr. Augustine P Bangalie - Enumerator Ms Jean Bangura - Enumerator Ms Anita Cole - Enumerator Mr. A B S Conteh - Enumerator Mrs. Mariama Deensie - Enumerator Ms Mamusu Dumbuya - Enumerator Mr. N’pha Dumbuya - Enumerator Mr. Osman Kemoh Kabba - Enumerator Ms Agnes Fatu Kamara - Enumerator Ms Aminata B Kamara - Enumerator Mr. Edward S Kamara - Enumerator Mr. Jeromy Kamara - Enumerator Ms Karietta Kamara - Enumerator Ms Mamie Jue Kargbo - Enumerator Mr. Mustapha Kawa - Enumerator Mr. Abdul S Koroma - Enumerator Mr. Charles S Koroma - Enumerator Mr. James M Koroma - Enumerator Mr. M S Koroma - Enumerator Ms N’gadie Koroma - Enumerator Ms Siatta Lukuley - Enumerator Mr. Charles Macauley - Enumerator Mr. David M’bayo - Enumerator Mr. A B S Samura - Enumerator Mr. M M Sanusi - Enumerator Mr. Mohamed S Sesay - Enumerator Mr. M M Swaray - Enumerator Ms Jannie Taylor - Enumerator Mr. Samuel A Wilson - Enumerator Mr. Akintola Wyse - Enumerator Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 69 Data entry personnel Ms Marian M Conteh - Data Entry Clerk Ms Christiana Davies - Data Entry Clerk Mr. Billy K Hamilton - Data Entry Clerk Ms Isata Jigba - Data Entry Clerk Ms Maimunatu Karim - Data Entry Clerk Ms Mariatu S Mansaray - Data Entry Clerk Ms Matilda Ngondo - Data Entry Clerk Ms Gifty Palmer - Data Entry Clerk Ms Shuwai Pratt - Data Entry Clerk Ms Christiana Richards - Data Entry Clerk Ms Katumu Samba - Data Entry Clerk Mrs. Hawa Sesay - Data Entry Clerk Ms Cecilia Sillah - Data Entry Clerk Ms Cynthia Will - Data Entry Clerk Logistical support staff Mrs. Elizabeth Bangura - Secretary Mr. M D Kallon - Sub Accountant Mr. James Kamara - Driver Mr. Paul Koroma - Driver Mr. Mohamed Turay - Office Assistant Mr. Emmanuel Vincent - Finance Clerk Technical support staff Dr. Robert McPherson - Consultant Mr. Paul Sengeh - M&E Officer, UNICEF Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 70 Appendix C: Sample design The sampling design described below sought to guide a survey that would be truly national in scope. The sampling unit of the MICS2 survey is the household. However, depending on the indicator, the child, the caretaker, and the household all may serve as the unit of analysis. The MICS2 Manual9 was consulted extensively during the development of the sample design. Any significant deviations from the suggestions made in that manual are noted in the text below. Sampling strategy The sample was randomly selected using a three-stage stratified cluster sampling methodology. The strata that were used in the sampling process derive directly from the administrative structure of Sierra Leone that is based on regions, districts, chiefdoms, and enumeration areas (EAs). First stage of sampling The goal of the first stage of the sampling process was to prepare a list of administrative units that were widely recognized in Sierra Leone, along with current estimates of the population of each unit, and then randomly select a specific number of these units for further sampling. The Central Statistics Office of the Government of Sierra Leone has historically demarcated the country into Enumeration Areas (EAs) for the purpose of carrying out censuses and other data collection efforts. The decision was therefore taken to use EAs as the primary sampling unit (PSU), given their well-defined nature. The use of EAs as the PSU for the survey was complicated by the existence of camps for internally displaced persons (IDPs); in the latest count of IDPs in November 1999, it was estimated that 319,546 people currently resided in camps. While each camp was technically situated within one or more Enumeration Areas EAs, they were more easily viewed as separate EAs for the purposes of sampling. The overall approach in this stage of sampling used EAs as the primary sampling unit with the addition of camps to the list of EAs in order to assure that IDPs were adequately represented in the survey. In the first stage of the sampling process, a list of the 2,546 enumeration areas (EAs) of Sierra Leone was developed, along with an estimate of the current household population (i.e., number of households) of each EA. The Central Statistics Office (CSO) has a list of all EAs in Sierra Leone—the problem faced by the survey team was that the population of each EA had not been updated since the 1992 Demographic and Social Monitoring Survey. Updated estimates were established in the following manner. Senior staff from the CSO visited the country’s 149 chiefdoms. Within each rural chiefdom, the CSO staff person worked with key informants such as the Paramount Chief and the clerk of the Native Administration Office and other local informants to establish an estimate of the current population of each EA. In urban Chiefdoms the process was different. Urban chiefdoms are generally much larger than rural chiefdoms, and are divided into sections, which are in turn sub-divided into EAs. In urban chiefdoms, CSO personnel worked with the Section Chiefs or other local informants in order to establish population estimates for each EA. The population of each camp was ascertained from the UNHACU database of camp populations. The UNHACU database was updated monthly and it was felt by local experts to be reasonably accurate. 9 Monitoring progress toward the goals of the world summit for children: End-decade multiple-indicator survey manual. Division of Evaluation, Policy and Planning. Program Division. UNICEF. New York, November, 1999. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 71 The list of EAs was then ordered using implicit stratification according to five variables: 1) location [urban/rural]; 2) province/region; 3) chiefdom; 4) EA; 5) population size. The list of camps was appended to the lis t of 2,553 EAs. Systematic sampling was then used to select 225 EAs/camps for the survey. Second stage of sampling In the second stage of sampling, a Cartographer and a Mapper traveled to each of the 225 EAs that were selected in the first stage of sampling. Each EA was divided into pieces of approximately 100 households using the map of the EA held by the CSO. Local informants assisted in this division. One of the pieces was then randomly selected for the next stage of sampling. Rebels in 16 EAs that were selected for the survey refused access to the survey team for this stage of sampling because of inadequate information from their leaders. Plans were made to revisit these EAs during the data collection phase in order to complete the sampling process. The process in refugee camps that were selected was somewhat simpler; one household was randomly chosen from the list of households held by the camp administration using simple random sampling. The 19 households following the selected household were then selected in order to form a cluster of 20 households. Third stage of sampling In the third stage of sampling every household in the piece that had been selected was mapped. Using this map, the piece was then divided into segments . Each segment contained approximately 20 households. One of the segments in the selected piece was then randomly selected. A household list for use by surveyors was then constructed that included all houses in the selected segment. Sample size In the MICS2 survey, decisions regarding sample size were are made based on a number of factors that include 1) the probability of making a Type 1 error 10 that the survey team is willing to accept, 2) the design effect11 (or deff) of the survey design, 3) the estimated population prevalence of a key indicator, 4) the average number of people per household, 5) the estimated non-response rate, 6) the proportion of the target population represented by a key single-year age group (12-23 month-old children are often used), and 7) the desired level of precision for the indicator estimates. The MICS2 manual recommends that the probability of a Type 1 error be set at 0.05, that a design effect of 1.75 be assumed, that the average number of people per household be assumed to be 5 (unless available data suggest otherwise), that the survey team anticipate a non-response rate of 10%, and that the proportion of the target population in a key age group be considered to be 0.03. The manual also recommends that a specific indicator be designated as key by health and education officials in UNICEF and the national government. If the prevalence rate of this indicator is believed to be below 0.25 or above 0.75 (e.g., 19% vaccination with DPT3), a precision of +/- 0.03 is recommended. If prevalence rates are believed to be greater than 0.25, a precision of +/- 0.05 is recommended. 10 This probability, generally set at 0.05 (or 5%), is the probability that the true population value for a given indicator might fall outside of the confidence limits (in this case, the 95% confidence limits) that surround the estimate of the indicator. 11 The design effect is the factor by which the precision of the estimates of indicators in a survey decrease due to the sampling design, as compared to a survey that uses simple random sampling. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 72 UNICEF/Sierra Leone health and education personnel felt that almost all important indicators in those sectors were between 25 and 75 percent. In addition, the large sample size that would result from setting the desired precision at +/- 3 percent (8-10,000 households) was felt to be excessively large by UNICEF officials and members of the survey team. Available data suggest that the average household size in rural areas of Sierra Leone was approximately seven, while in urban areas it was approximately six. The MICS2 manual gives the formula that was used for calculating the sample size. This formula, along with a definition of the terms that it contains, is reproduced below. n = [4 (r) (1 – r) (deff) 1.1] / [(d2) (p) (nh)] Terms · n is the required sample size for the key indicator; · 4 is a factor to achieve a 5 percent chance of Type 1 error; · r is the predicted or anticipated prevalence for the key indicator; · 1.1 is the factor necessary to raise the sample size by 10 percent for non-response; · deff is the design effect; · d is the desired precision of the estimate of the key variable; · p is the proportion of the total population that the smallest population group comprises; and, · nh is the average household size. The recommended sample size for this survey was calculated using the sample size equation given above, based on the following values of the relevant variables: · r = 0.512 · deff = 1.75 · d = 0.05 · p = 0.03 · nh = 6.5 Plugging these values into the sample size equation above yielded a required sample size of 3949. This requirement was then reconciled with the recommended number of segments and the recommended number of households within each segment that was listed and surveyed. The MICS2 manual suggests that a minimum of twenty households be surveyed within each segment. The manual also recommends that from 250 to 300 segments be surveyed. Combining these two factors resulted in a minimum recommended sample size of 5-6,000. It was decided to split the difference between these two estimates. A sample size of 4,500 was therefore planned for the MICS2 survey, consisting of 225 segments of approximately 20 households each. 12 In the equation, the value of prevalence = 0.5 is used, since this gives the most conservative (i.e., maximum) sample size. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 73 Appendix D: Questionnaires HOUSEHOLD QUESTIONNAIRE WE ARE FROM (country-specific affiliation). WE ARE WORKING ON A PROJECT CONCERNED WITH FAMILY HEALTH AND EDUCATION. I WOULD LIKE TO TALK TO YOU ABOUT THIS. THE INTERVIEW WILL TAKE ABOUT (40) MINUTES. ALL THE INFORMATION WE OBTAIN WILL REMAIN STRICTLY CONFIDENTIAL AND YOUR ANSWERS WILL NEVER BE IDENTIFIED. DURING THIS TIME I WOULD LIKE TO SPEAK WITH ALL MOTHERS OR OTHERS WHO TAKE CARE OF CHILDREN IN THE HOUSEHOLD. MAY I START NOW? If permission is given, begin the interview. HOUSEHOLD INFORMATION PANEL ** 1. EA number: 2. Household number: ___ ___ ___ ___ ___ ___ ___ ___ ___ 3. Day/Month/Year of interview: 4. Interviewer number: ___ ___ / ___ ___ / ___ ___ ___ ___ ___ ___ 5. Name of head of household: Sex of household head: Male……….1 ______________________________________ Female…….2 6. Area: Urban. 1 Rural.2 7. Region:** North.1 East .2 South.3 Western Area.4 8. Material of dwelling floor:** Wood/Planks . 1 Tile/concrete. 2 Dirt/straw. 3 9. Number of rooms in dwelling:** ___ ___ Other(specify) _______________________ 4 9A. Which of the following do you have in this household? Number 1. Radios ……………………………. ____ 2. Television sets……………………. ____ 3. Private Vehicle …………………… ____ 4. Motorcycle ………………………… ____ 5. Pedal bicycle ……………………… ____ 6. Own house ………………………. ____ 10. Result of HH interview: Completed . 1 Refused. 2 Not at home. 3 HH not found/destroyed . 4 Other (specify)_______________________ 5 11. No. of women eligible for interview: 12. No. of women interviews completed: ___ ___ ___ ___ 13. No. of children under age 5: ___ ___ 14. No. of child interviews completed: ___ ___ 15. Data entry clerk: ___ __ Interviewer/supervisor notes : Use this space to record notes about the interview with this household, such as call- back times, incomplete individual interview forms, number of attempts to re-visit, etc. Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 74 EA no.__ __ __ __ __ __ Household no. __ __ __ HOUSEHOLD LISTING FORM FIRST, PLEASE TELL ME THE NAME OF EACH PERSON WHO USUALLY LIVES HERE, STARTING WITH THE HEAD OF THE HH. (Use survey definition of HH member). List the first name in line 01. List adult HH members first, then list children. Then ask: ARE THERE ANY OTHERS WHO LIVE HERE, EVEN IF THEY ARE NOT AT HOME NOW? (THESE MAY INCLUDE CHILDREN IN SCHOOL OR AT WORK). If yes, complete listing. Then, ask and record answers to questions as described in Instructions for Interviewers. Add a continuation sheet if there is not enough room on this page. Tick here if continuation sheet used ¨ Eligible for: WOMEN ’S MODULES CHILD LABOUR MODULE CHILD HEALTH MODULES For persons age 15 or over ask Qs. 8 and 9 For children under age 15 years ask Qs. 10-13 1. Line no. 2. Name 3. IS (name) MALE OR FEMALE ? 1 MALE 2 FEM. 4. HOW OLD IS (name)? HOW OLD WAS (name) ON HIS/HER LAST BIRTHDAY? Record in completed years 99=DK* 5. Circle Line no. if woman is age 15-49 6. For each child age 5-14: WHO IS THE MOTHER OR PRIMARY CARETAKER OF THIS CHILD? Record Line no. of mother/ caretaker 7. For each child under 5: WHO IS THE MOTHER OR PRIMARY CARETAKER OF THIS CHILD? Record Line no. of mother/ caretaker 8. CAN HE/SHE READ A LETTER OR NEWSPAPER EASILY , WITH DIFFICULTY OR NOT AT ALL? 1 EASILY 2 DIFFICULT 3 NOT AT ALL 9 DK 9. WHAT IS THE MARITAL STATUS OF (name)?** 1 CURRENTLY MARRIED / IN UNION 2 WIDOWED 3 DIVORCED 4 SEPARATED 5 NEVER MARRIED 10. IS (name’s) NATURAL MOTHER ALIVE? 1 YES 2 NO 9 DK 11. If alive: DOES (name’s) NATURAL MOTHER LIVE IN THIS HOUSE- HOLD? 1 YES 2 NO 12. IS (name’s) NATURAL FATHER ALIVE? 1 YES 2 NO 9 DK 13. If alive: DOES (name’s) NATURAL FATHER LIVE IN THIS HOUSE- HOLD? 1 YES 2 NO LINE NAME M F AGE 15-49 MOTHER MOTHER E D N DK M W D S N Y N DK Y N Y N DK Y N 01 1 2 ____ ____ 01 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 02 1 2 ____ ____ 02 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 03 1 2 ____ ____ 03 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 04 1 2 ____ ____ 04 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 05 1 2 ____ ____ 05 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 06 1 2 ____ ____ 06 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 07 1 2 ____ ____ 07 ___ ___ ___ ___ 1 2 3 9 1 2 3 4 5 1 2 9 1 2 1 2 9 1 2 ARE THERE ANY OTHER CHILDREN LIVING HERE – EVEN IF THEY ARE NOT MEMBERS OF YOUR FAMILY OR DO NOT HAVE PARENTS LIVING IN THIS HOUSEHOLD? INCLUDING CHILDREN AT WORK OR AT SCHOOL? If yes, insert child’s name and complete form. * See instructions: to be used only for elderly household members (code meaning “do not know/over age 50”). Multi-Indicator Cluster Survey-MICS2 Final Report November 2000 75 EA no.__ __ __ __ __ __ Household no. __ __ __ EDUCATION MODULE If interview takes place between two school years, use alternative wording found in Appendix 1. For persons age 5 or over ask Qs. 15 and 16 For children age 5 through 17 years, continue on, asking Qs. 17-22 14. Line no. 15. HAS (name) EVER ATTENDED SCHOOL? 1 YES ð Q.16 2 NO ø NEXT LINE 16. WHAT IS THE HIGHEST LEVEL OF SCHOOL (name) ATTENDED? WHAT IS THE HIGHEST GRADE (name) COMPLETED AT THIS LEVEL? LEVEL: 1 PRIMARY 2 SECONDARY 3 HIGHER 4 NON-STANDARD CURRICULUM 9 DK GRADE: 99 DK If less than 1 grade, enter 00. 17. IS (name) CURRENTLY ATTENDING SCHOOL? 1 YES ð Q.19 2 NO 18. DURING THE CURRENT SCHOOL YEAR , DID (name) ATTEND SCHOOL AT ANY TIME? 1 YES 2 NO ð Q.21 19. SINCE LAST (day of the week), HOW MANY DAYS DID (name) ATTEND SCHOOL? Insert number of days in space below. 20. WHICH LEVEL AND GRADE IS/WAS (name) ATTENDING? LEVEL: 1 PRESCHOOL 2 PRIMARY 3 SECONDARY 4 NON-STANDARD CURRICULUM 9 DK GRADE: 99 DK 21. DID (name) ATTEND SCHOOL LAST YEAR? 1 YES 2 NO ø NEXT LINE 9 DK ø NEXT LINE 22. WHICH LEVEL AND GRADE DID (name) ATTEND LAST YEAR? LEVEL: 1 PRESCHOOL 2 PRIMARY 3 SECONDARY 4 NON-STANDARD CURRICULUM 9 DK GRADE: 99 DK LINE Y NO LEVEL GRADE YES NO YES NO DAYS LEVEL GRADE Y N DK LEVEL GRADE 01 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 02 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 03 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 04 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 05 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 06 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ 07 1 2ðNEXT LINE 1 2 3 4 9 ___ ___ 1 2 1 2 ____ 1 2 3 4 9 ___ ___ 1 2 9 1 2 3 4 9 ___ ___ Now for each woman age 15-49 years, wr

View the publication