C o u n t r y r e p o r t s o n H n p a n d p o v e r t y Socio-Economic Differences in Health, Nutrition, and Population MADAGASCAR Davidson R. Gwatkin, Shea Rutstein, Kiersten Johnson, Eldaw Suliman, Adam Wagstaff, and Agbessi Amouzou April 2007 SOCIO-ECONOMIC DIFFERENCES IN HEALTH, NUTRITION, AND POPULATION MADAGASCAR 1997 Davidson R. Gwatkin, Shea Rutstein, Kiersten Johnson, Eldaw Suliman, Adam Wagstaff, and Agbessi Amouzou TABLE OF CONTENTS Foreword I Introduction III Part I. Basic Tables 1 A. Total Population 3 B. Female and Male Populations 9 C. Rural and Urban Populations 13 Part II. Technical Notes 19 A. Indicator Definitions 21 B. Data and Methods 35 C. Discussion 39 Part III. Supporting Tables 45 A. Sample Sizes 47 B. Standard Errors 53 C. Asset Distribution and Weights 59 Part IV. Annexes 61 A. Sources of Additional Information 63 B. Use of Information from this 65 Report to Monitor the Economic Status of People Served by HNP Programs C. Countries Covered by the 71 HNP-Poverty Report Project FOREWORD The World Bank shares the desire of its member states and client countries to ensure that the poor partake fully in the health gains that the countries achieve. To assist in this, the Bank, in cooperation with the Dutch and Swedish Governments, has sponsored the set of reports providing basic information about health inequalities within countries to which this document belongs. The information shows clearly that disparities in both health conditions and health service use are unacceptably large. As countries and the Bank work to reduce important inequalities among regions and countries, there is a clear need for equally vigorous efforts to lessen the inequity represented by intra-country differences among socio-economic groups. My colleagues and I hope that all concerned with equity in health will find this information useful in making the case for effective actions to improve the health of the poor, and in designing programs to achieve this crucial objective. Joy Phumaphi Vice President Human Development Network The World Bank - I - INTRODUCTION This report is one in a series that provides basic information about health, nutrition, and population (hnp) inequalities within fifty-six developing countries. The series to which the report belongs is an expanded and updated version of a set covering forty- five countries that was published in 2000. The fifty-six reports in the current series cover almost all DHS surveys undertaken during the period beginning in 1990 and ending with the date of the last survey for which data were publicly available as of June 2006.1 The report's contents are intended to facilitate preparation of country analyses and the development of activities to benefit poor people. To this end, the report presents data about hnp status, service use, and related matters among individuals belonging to different socio-economic classes. The principal focus is on differences among groups of individuals defined in terms of the wealth or assets of the households where they reside. The source of data is the Demographic and Health Survey (DHS) program, a large, multi-country household survey project. The figures in this and the other reports in the series draw on responses to questions about household wealth or assets included in the DHS questionnaire, which were similar for all the surveys covered. These responses served as the basis for the construction of a wealth index, which was used to rank individuals according to the index value for the household to which they belonged. The individuals were then divided into quintiles, and the mean value for each of up to approximately 120 indicators was calculated for each quintile. The report is organized in four principal parts: · Part one, which constitutes the report's core, consists of tables presenting quintile- specific data for each hnp indicator covered from the two most recent DHS surveys available at the time of the report's publication. In each part there are three sets of tables: the first provides quintile-specific information for the total population; the second presents data separately for quintiles of females and males; the third features quintile-specific information presented separately for rural and urban residents. Each of these sets is divided into four sections: one dealing with hnp status, the second with the use of hnp services, the third with hnp-related behavior, and the fourth with other hnp status determinants. · Part two provides supplementary technical information designed to help readers understand the data presented in part one. This information deals with such issues as how the covered hnp indicators were defined and how the quintile-specific estimates were derived. · Part three presents supporting tables that deal with three of the technical matters covered in part two: the size of the sample for each indicator covered; the standard error for each 1The average interval of approximately two years between data collection and availability means that the latest surveys covered were conducted in 2005. - III - quintile-specific estimate in the total population; and the items used in constructing the wealth index, along with the weight assigned to each. An additional, fourth part consists of three annexes, for readers interested in applying the approach used in the report or in learning more about the other reports in this series. The first annex is an annotated bibliography containing further information about the technical issues concerning the approach used in the report, and about employing that approach to examine additional issues using DHS or other data sets. The second shows how the report's approach can be applied to monitor the distribution of benefits from other hnp programs, and provides a tool for doing so. The third annex is a list of all fifty-six countries for which reports are available, along with an indication of how to obtain copies of the reports dealing with them. ********************************* The authors thank the Dutch and Swedish Governments for the generous support that made production of this report possible. - IV - PART I. BASIC TABLES A. TOTAL POPULATION B. FEMALE AND MALE POPULATIONS C. RURAL AND URBAN POPULATIONS Notes: ­ Each of the three sections referred to above consists of four divisions, presenting data for: I) hnp status; II) hnp service use; III) hnp-related individual and household behavior; and IV) other, underlying determinants of hnp status. ­ Full definitions of all indicators covered in the tables are provided in section A of the technical notes found in part II. ­ "na" appears in the table cells when data are not available, usually because the DHS survey concerned did not collect information about the indicator(s) in question. ­ Figures in the tables shown within parentheses indicate the absence of adequate observations to produce acceptably reliable values. Asterisks appear when the number of observations was too small to justify the presentation even of figures within parentheses. (For further information, see the section on "Sampling Errors" in the presentation of data and methods in part II.B.) Asterisks also will be found in columns showing statistical indices of inequality when the amount of quintile-specific information available is inadequate to permit computation of the value for the index concerned. ­ Female/male tables include only indicators relevant for both sexes; those pertaining to only one sex (e.g., fertility, women's nutritional status, antenatal care, attended deliveries) have been omitted. Madagascar 1997 - TOTAL POPULATION Part I: HNP STATUS Indicator Wealth Quintiles Low/High Low-High Concentration Index Ratio Diff. Value Standard Low 2nd 3rd 4th High Avg. (Abs. Val.) Errors A. Childhood illness and mortality Infant mortality rate 119.1 118.3 103.2 76.2 57.5 99.3 2.07 61.60 -0.1221 0.0011 Under-five mortality rate 195.0 186.1 175.0 129.6 101.4 163.9 1.92 93.60 -0.1095 0.0009 Prevalence of fever 33.1 32.0 34.8 29.7 30.0 32.2 1.10 3.10 -0.0325 0.0142 Prevalence of diarrhea 29.2 21.8 26.1 31.3 26.2 27.1 1.11 3.00 0.0090 0.0163 Prevalence of acute respiratory infection 27.3 25.6 25.0 21.4 16.0 23.9 1.71 11.30 -0.0971 0.0168 B. Fertility Total fertility rate 8.1 6.3 6.8 5.7 3.4 6.0 2.38 4.70 -0.1423 0.0009 Adolescent fertility rate 271.0 215.0 206.0 141.0 78.0 180.0 3.47 193.00 -0.2110 0.0023 C. Nutritional status (%) Children: Moderate stunting 24.7 26.4 31.0 26.5 25.2 26.7 0.98 0.50 0.0084 0.0169 Severe stunting 24.6 19.2 21.1 23.6 17.3 21.6 1.42 7.30 -0.0578 0.0190 Moderate underweight 29.1 27.2 26.5 27.5 23.6 27.2 1.23 5.50 -0.0364 0.0172 Severe underweight 16.0 12.9 12.6 11.6 8.2 12.8 1.95 7.80 -0.1179 0.0263 Mild anemia 18.5 19.4 22.4 23.4 23.0 21.0 0.80 4.50 0.0614 0.0239 Moderate anemia 46.6 51.1 48.0 39.9 36.7 45.2 1.27 9.90 -0.0486 0.0136 Severe anemia 10.7 8.4 6.9 6.5 3.4 7.7 3.15 7.30 -0.1814 0.0427 Women: Malnutrition 24.3 22.8 18.4 18.9 15.1 20.6 1.61 9.20 -0.0908 0.0214 Mild anemia 33.6 33.6 32.6 28.2 16.9 30.1 1.99 16.70 -0.0867 0.0149 Moderate anemia 11.4 13.3 11.6 11.3 6.3 11.1 1.81 5.10 -0.0942 0.0269 Severe anemia 1.3 0.6 0.3 1.1 0.8 0.9 1.63 0.50 -0.1463 0.1205 D. Female circumcision (%) Prevalence of circumcision: Girls na na na na na na na na na na Women na na na na na na na na na na Prevalence of occlusion: Girls na na na na na na na na na na Women na na na na na na na na na na E. Sexually transmitted disease Prevalence of genital discharge: Women 1.6 1.7 1.9 0.8 0.8 1.3 2.00 0.80 -0.1337 0.0584 Men na na na na na na na na na na Prevalence of genital ulcer: Women 0.8 0.9 0.8 1.0 1.1 0.9 0.73 0.30 0.0877 0.0730 Men na na na na na na na na na na - 3 - Madagascar 1997 - TOTAL POPULATION Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE Indicator Wealth Quintiles Low/High Low-High Concentration Index Ratio Diff. Value Standard Low 2nd 3rd 4th High Avg. (Abs. Val.) Errors A. Childhood immunization BCG coverage 47.5 60.0 62.0 82.3 92.5 66.0 0.51 45.00 0.1460 0.0104 Measles coverage 32.1 34.4 38.7 60.8 78.9 46.0 0.41 46.80 0.1987 0.0168 DPT coverage 32.3 37.7 41.4 65.3 81.1 48.4 0.40 48.80 0.2052 0.0156 Full basic coverage 22.0 25.2 27.1 54.7 66.0 36.2 0.33 44.00 0.2462 0.0205 No basic coverage 29.5 26.2 24.5 8.4 3.6 20.1 8.19 25.90 -0.2963 0.0310 Hepatitis B coverage na na na na na na na na na na Yellow fever coverage na na na na na na na na na na B. Treatment of childhood illnesses Treatment of fever: Medical treatment of fever 23.7 24.4 23.8 25.9 34.8 25.6 0.68 11.10 0.0420 0.0309 Treatment in a public facility 16.9 16.2 19.5 21.3 19.3 18.4 0.88 2.40 0.0003 0.0361 Treatment in a private facility 6.8 7.6 4.2 4.6 15.5 7.1 0.44 8.70 0.1393 0.0740 Treatment of acute respiratory infection: Medical treatment of ARI 35.1 33.5 32.1 36.9 59.3 36.6 0.59 24.20 0.0592 0.0281 Treatment in a public facility 26.4 25.4 25.7 32.3 36.8 28.0 0.72 10.40 0.0334 0.0324 Treatment in a private facility 8.7 8.1 6.4 4.5 22.5 8.6 0.39 13.80 0.1424 0.0838 Treatment of diarrhea: Use of oral rehydration therapy 60.2 61.0 59.5 66.7 92.7 65.7 0.65 32.50 0.0804 0.0124 Medical treatment of diarrhea 38.9 35.4 31.9 40.7 55.7 39.5 0.70 16.80 0.0726 0.0238 Treatment in a public facility 27.0 21.2 23.7 31.8 24.2 26.1 1.12 2.80 0.0169 0.0306 Treatment in a private facility 11.9 13.4 8.1 8.8 30.5 13.1 0.39 18.60 0.1821 0.0582 C. Antenatal and delivery care Antenatal care visits: To a medically trained person 67.1 71.7 75.3 86.1 95.6 77.0 0.70 28.50 0.0803 0.0054 To a doctor 14.8 15.5 21.5 24.8 20.2 18.8 0.73 5.40 0.1092 0.0207 To a nurse or trained midwife 52.3 56.1 53.9 61.4 75.4 58.2 0.69 23.10 0.0668 0.0083 Multiple visits to a medically trained person 55.2 58.1 61.5 72.3 82.6 63.9 0.67 27.40 0.0955 0.0071 Antenatal care content: Tetanus toxoid 45.1 48.5 45.6 53.7 58.3 49.3 0.77 13.20 0.0577 0.0102 Prophylactic antimalarial treatment na na na na na na na na na na Iron supplementation na na na na na na na na na na Delivery attendance: By a medically trained person 29.6 37.0 40.5 65.0 88.5 47.3 0.33 58.90 0.2382 0.0087 By a doctor 6.8 8.0 12.6 21.0 19.7 12.4 0.35 12.90 0.2488 0.0252 By a nurse or trained midwife 22.7 28.9 27.9 44.0 68.7 34.9 0.33 46.00 0.2346 0.0123 In a public facility 24.3 28.8 33.3 44.3 38.4 32.4 0.63 14.10 0.1177 0.0136 In a private facility 0.8 0.6 1.3 2.2 5.5 1.7 0.15 4.70 0.5818 0.1005 At home 73.7 69.7 64.3 52.7 54.9 64.8 1.34 18.80 -0.0754 0.0069 D. Contraceptive services Contraceptive prevalence: Women 2.3 3.0 7.3 12.4 23.8 9.7 0.10 21.50 0.4974 0.0275 Men na na na na na na na na na na - 4 - Madagascar 1997 - TOTAL POPULATION Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE (Cont.) Indicator Wealth Quintiles Low/High Low-High Concentration Index Ratio Diff. Value Standard Low 2nd 3rd 4th High Avg. (Abs. Val.) Errors D. Contraceptive services (cont.) Source of contraception - public sector: Women * (73.1) (71.1) 66.2 37.6 52.8 * * * * Men na na na na na na na na na na Source of contraception - private sector: Women * (26.9) (28.9) 32.3 61.2 46.2 * * * * Men na na na na na na na na na na E. Treatment of adult illnesses Treatment of genital discharge, ulcer sore: Women (36.7) (45.4) (40.7) (54.2) (53.1) 45.5 0.69 16.40 0.0746 0.0490 Men na na na na na na na na na na Treatment of genital discharge, ulcer, sore in public facilities: Women na na na na na na na na na na Men na na na na na na na na na na Voluntary counseling and testing for HIV/AIDS: Women na na na na na na na na na na Men na na na na na na na na na na - 5 - Madagascar 1997 - TOTAL POPULATION Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR Indicator Wealth Quintiles Low/High Low-High Concentration Index Ratio Diff. Value Standard Low 2nd 3rd 4th High Avg. (Abs. Val.) Errors A. Hygienic practices Disposal of children's stools: Sanitary disposal na na na na na na na na na na Handwashing: Wash hands prior to preparing food na na na na na na na na na na Handwashing facilities in household na na na na na na na na na na B. Bednet ownership and use Bednet ownership: Bednet ownership na na na na na na na na na na Treated bednet ownership na na na na na na na na na na Bednet use: By children na na na na na na na na na na By pregnant women na na na na na na na na na na C. Breastfeeding Exclusive breastfeeding 57.4 60.3 63.4 64.1 64.9 61.1 0.88 7.50 0.0199 0.0219 Timely complementary feeding 84.7 90.7 88.4 85.6 94.6 88.2 0.90 9.90 0.0208 0.0109 Bottle-feeding 1.5 0.6 1.1 3.9 14.2 3.1 0.11 12.70 0.7644 0.1281 D. Micronutrient consumption Iodized salt: Availability of iodized salt in household 75.1 67.9 72.2 88.0 91.5 78.9 0.82 16.40 0.0599 0.0033 Vitamin A: Children 2.4 2.6 1.5 2.0 3.2 2.3 0.75 0.80 0.1171 0.0826 Women na na na na na na na na na na E. Tobacco and alcohol use Tobacco: Women na na na na na na na na na na Men na na na na na na na na na na Alcohol: Women na na na na na na na na na na Men na na na na na na na na na na F. Sexual practices Non-regular sexual partnerships: Women na na na na na na na na na na Men na na na na na na na na na na Condom usage with non-regular partner: Women na na na na na na na na na na Men na na na na na na na na na na G. Domestic violence Ever experienced violence na na na na na na na na na na Experienced violence in past year na na na na na na na na na na - 6 - Madagascar 1997 - TOTAL POPULATION Part IV: UNDERLYING DETERMINANTS OF HNP STATUS Indicator Wealth Quintiles Low/High Low-High Concentration Index Ratio Diff. Value Standard Low 2nd 3rd 4th High Avg. (Abs. Val.) Errors A. Education School completion: Women 9.3 11.3 16.4 32.1 72.3 29.9 0.13 63.00 0.4646 0.0085 Men 10.9 14.6 19.0 31.5 71.1 32.3 0.15 60.20 0.4305 0.0079 School participation: Girls 40.5 45.3 58.0 72.6 92.1 59.8 0.44 51.60 0.1824 0.0078 Boys 36.0 42.4 56.9 68.5 93.4 57.6 0.39 57.40 0.2079 0.0078 B. Exposure to mass media Newspaper readership: Women 12.6 14.9 21.6 29.0 49.6 26.5 0.25 37.00 0.3405 0.0111 Men na na na na na na na na na na Radio listenership: Women 6.4 25.2 28.8 48.5 81.0 39.6 0.08 74.60 0.3966 0.0064 Men na na na na na na na na na na Television viewership: Women 0.9 1.4 1.4 5.1 51.5 13.8 0.02 50.60 0.9162 0.0212 Men na na na na na na na na na na C. Knowledge and attitudes about HIV/AIDS Knowledge about sexual transmission of HIV/AIDS: Women 18.6 24.4 33.1 49.3 78.6 42.3 0.24 59.92 0.3076 0.0065 Men na na na na na na na na na na Knowledge about mother-to-child transmission of HIV/AIDS: Women 48.3 44.4 52.6 67.6 89.4 58.3 0.54 41.10 0.2001 0.0047 Men na na na na na na na na na na Attitudes toward HIV/AIDS: Women na na na na na na na na na na Men na na na na na na na na na na D. Status of women Household decisionmaking: Can seek own health care na na na na na na na na na na Can seek children's health care na na na na na na na na na na Can make daily household purchases na na na na na na na na na na Can make large household purchases na na na na na na na na na na Can make meal-related decisions na na na na na na na na na na Freedom of movement: Can travel to visit family, relatives na na na na na na na na na na Other decisionmaking, attitudes: Can decide how to spend own money 16.9 20.5 23.2 22.4 45.1 25.3 0.37 28.20 0.2299 0.0136 Can decide whether to have sex na na na na na na na na na na Justifies domestic violence na na na na na na na na na na E. Orphanhood Paternal orphan prevalence 5.9 5.9 6.0 4.8 4.0 5.4 1.48 1.90 -0.0626 0.0194 Maternal orphan prevalence 2.3 4.4 3.7 3.9 2.7 3.4 0.85 0.40 0.0393 0.0233 Double orphan prevalence 0.3 0.7 0.5 0.3 0.4 0.4 0.75 0.10 0.0305 0.0701 - 7 - Madagascar 1997 - FEMALE / MALE POPULATIONS Part I: HNP STATUS Indicator Wealth Quintiles - Female Wealth Quintiles - Male Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Childhood mortality and morbidity Infant mortality rate 112.4 104.6 93.6 63.0 51.1 89.5 125.8 131.0 112.4 88.5 63.8 108.7 Under-five mortality rate 184.0 169.7 165.0 112.6 96.7 151.7 206.0 201.1 184.7 145.6 106.0 175.5 Prevalence of fever 31.8 29.4 34.7 31.8 29.8 31.6 34.5 34.9 34.9 27.8 30.1 32.8 Prevalence of diarrhea 29.7 20.5 24.4 29.6 25.8 26.2 28.5 23.3 27.7 32.7 26.5 27.9 Prevalence of acute 28.3 27.9 21.3 20.9 15.6 23.9 26.3 23.0 28.2 21.7 16.3 23.9 respiratory infection B. Nutritional status Children: Moderate stunting 21.4 29.4 30.6 26.8 20.9 25.8 28.1 23.1 31.4 26.2 29.3 27.6 Severe stunting 23.8 16.8 21.1 15.9 14.2 19.1 25.4 22.0 21.0 30.6 20.2 24.2 Moderate underweight 31.1 28.4 23.2 27.5 22.4 27.3 27.0 25.8 29.3 27.5 24.8 27.1 Severe underweight 13.6 10.5 14.3 4.9 5.8 10.4 18.6 15.5 11.1 17.7 10.4 15.2 Mild anemia 17.3 21.4 25.3 24.6 28.9 22.4 19.9 17.2 19.9 22.3 18.1 19.6 Moderate anemia 45.9 44.4 45.6 38.7 33.9 42.7 47.3 58.4 50.1 41.1 39.1 47.6 Severe anemia 9.7 7.9 5.0 4.8 1.2 6.5 11.7 8.9 8.4 8.0 5.3 8.8 C. Sexually transmitted disease Prevalence of genital discharge: Women 1.6 1.7 1.9 0.8 0.8 1.3 Men na na na na na na Prevalence of genital ulcer: Women 0.8 0.9 0.8 1.0 1.1 0.9 Men na na na na na na - 9 - Madagascar 1997 - FEMALE / MALE POPULATIONS Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE Indicator Wealth Quintiles - Female Wealth Quintiles - Male Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Childhood immunization BCG coverage 49.0 67.5 64.2 80.9 92.5 67.7 45.9 51.8 59.8 83.4 92.5 64.2 Measles coverage 31.5 36.2 37.9 56.5 76.3 44.5 32.8 32.5 39.5 64.4 81.6 47.4 DPT coverage 31.1 38.4 42.6 66.7 82.6 48.5 33.5 36.9 40.3 64.2 79.5 48.3 Full basic coverage 21.8 24.9 25.2 52.3 65.1 34.8 22.1 25.6 28.9 56.7 66.9 37.6 No basic coverage 29.9 19.5 22.6 8.9 2.4 18.6 29.0 33.6 26.4 7.9 4.9 21.5 Hepatitis B coverage na na na na na na na na na na na na Yellow fever coverage na na na na na na na na na na na na B. Treatment of childhood illnesses Treatment of fever: Medical treatment of fever 21.9 27.9 28.0 24.4 36.4 26.6 25.5 21.3 20.1 27.4 33.2 24.7 Treatment in a public facility 12.9 18.7 23.4 21.4 20.7 18.6 20.9 14.0 16.2 21.1 18.0 18.1 Treatment in a private facility 9.0 7.9 4.6 3.0 15.7 7.7 4.6 7.3 3.9 6.3 15.2 6.6 Treatment of acute respiratory infection: Medical treatment of ARI 33.1 35.5 38.2 42.3 (58.7) 38.2 37.5 30.9 28.1 32.2 (59.9) 35.0 Treatment in a public facility 23.3 29.0 33.0 40.2 (43.2) 30.6 30.1 20.7 20.9 25.6 (30.5) 25.3 C. Sexually transmitted disease 9.8 6.5 5.2 2.1 (15.5) 7.6 7.4 10.3 7.3 6.6 (29.4) 9.7 Treatment of diarrhea: Use of oral rehydration therapy 61.3 54.9 62.8 67.4 96.6 66.3 58.9 67.0 56.9 66.2 88.8 65.2 Medical treatment of diarrhea 39.6 30.9 37.0 44.0 57.7 41.0 38.1 39.7 27.9 38.0 53.6 38.1 Treatment in a public facility 26.1 17.4 26.9 35.5 28.6 27.0 28.0 24.8 21.3 28.9 19.9 25.3 Treatment in a private facility 13.5 11.9 10.0 8.4 29.1 13.7 10.0 14.9 6.6 9.2 32.0 12.6 C. Contraceptive services Contraceptive prevalence: Women 2.3 3.0 7.3 12.4 23.8 9.7 Men na na na na na na Source of contraception - public sector: Women * (73.1) (71.1) 66.2 37.6 52.8 Men na na na na na na Source of contraception - private sector: Women * (26.9) (28.9) 32.3 61.2 46.2 Men na na na na na na D. Treatment of adult illnesses Treatment of genital discharge, ulcer, sore: Women (36.7) (45.4) (40.7) (54.2) (53.1) 45.5 Men na na na na na 0.0 Treatment of genital discharge, ulcer, sore in public medical facilities: Women na na na na na na Men na na na na na na Voluntary counseling and testing for HIV/AIDS: Women na na na na na na Men na na na na na na - 10 - Madagascar 1997 - FEMALE / MALE POPULATIONS Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR Indicator Wealth Quintiles - Female Wealth Quintiles - Male Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Hygienic practices Disposal of children's stools: Sanitary disposal na na na na na na Handwashing: Wash hands prior to preparing food na na na na na na Handwashing facilities in household na na na na na na B. Bednet ownership and use Bednet use: By children na na na na na na C. Breastfeeding Exclusive breastfeeding 59.4 (52.1) (59.9) (66.0) (58.6) 58.8 55.5 (70.6) 65.6 (61.8) (72.4) 63.4 Timely complementary feeding 84.9 (88.7) (87.9) (78.8) (88.4) 85.6 84.5 (92.6) (88.6) (91.2) (100.0) 90.4 Bottle-feeding 1.9 1.1 0.5 6.5 16.2 4.1 1.0 0.0 1.6 1.4 11.7 2.2 D. Micronutrient consumption C. Sexually transmitted disease Children 1.6 2.8 2.1 2.6 2.5 2.2 3.3 2.3 0.9 1.5 3.9 2.4 E. Tobacco and alcohol use Tobacco: Women na na na na na na Men na na na na na na Alcohol: Women na na na na na na Men na na na na na na F. Sexual practices Non-regular sexual partnerships: Women na na na na na na Men na na na na na na Condom usage with non-regular partner: Women na na na na na na Men na na na na na na - 11 - Madagascar 1997 - FEMALE / MALE POPULATIONS Part IV: UNDERLYING DETERMINANTS OF HNP STATUS Indicator Wealth Quintiles - Female Wealth Quintiles - Male Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Education School completion: Women 9.3 11.3 16.4 32.1 72.3 29.9 Men 10.9 14.6 19.0 31.5 71.1 32.3 School participation: Girls 40.5 45.3 58.0 72.6 92.1 59.8 Boys 36.0 42.4 56.9 68.5 93.4 57.6 B. Exposure to mass media Newspaper readership: Women 12.6 14.9 21.6 29.0 49.6 26.5 Men na na na na na na Radio listenership: Women 6.4 25.2 28.8 48.5 81.0 39.6 Men na na na na na na Television viewership: Women 0.9 1.4 1.4 5.1 51.5 13.8 Men na na na na na na C. Knowledge and attitudes about HIV/AIDS Knowledge about sexual transmission of HIV/AIDS: Women 18.6 24.4 33.1 49.3 78.6 42.3 Men na na na na na na Knowledge about mother-to-child transmission of HIV/AIDS: Women 48.3 44.4 52.6 67.6 89.4 58.3 Men na na na na na na Attitudes toward HIV/AIDS: Women na na na na na na Men na na na na na na D. Orphanhood Paternal orphan prevalence 5.7 6.4 6.0 4.6 2.9 5.2 6.1 5.4 5.8 5.0 5.1 5.5 Maternal orphan prevalence 2.6 4.4 4.1 4.0 2.4 3.5 2.0 4.3 3.4 3.8 2.9 3.3 Double orphan prevalence 0.2 0.9 0.5 0.3 0.4 0.4 0.5 0.6 0.4 0.3 0.4 0.4 - 12 - Madagascar 1997 - RURAL / URBAN POPULATIONS Part I: HNP STATUS Indicator Wealth Quintiles - Rural Wealth Quintiles - Urban Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Childhood illness and mortality Infant mortality rate 121.2 116.4 105.0 77.9 (50.2) 105.0 * * * (69.2) 60.5 77.9 Under-five mortality rate 200.5 184.0 178.6 131.4 (95.7) 173.8 * * * (122.5) 103.8 127.1 Prevalence of fever 32.6 32.8 34.4 29.9 32.3 32.6 36.7 25.6 38.2 29.0 28.9 30.6 Prevalence of diarrhea 27.7 21.3 26.1 31.8 21.2 26.3 41.7 26.0 26.4 29.3 28.3 30.0 Prevalence of acute 26.3 26.6 24.5 21.4 16.5 24.6 36.1 16.8 29.6 21.0 15.7 21.1 respiratory infection B. Fertility Total fertility rate 8.2 6.2 6.8 5.8 * 6.7 * * * * 3.1 4.2 Adolescent fertility rate 280.0 216.0 211.0 126.0 * 204.2 * * * * 73.0 120.7 C. Nutritional status Children: Moderate stunting 25.0 26.3 32.2 25.5 26.4 27.0 21.4 27.8 19.5 30.9 24.7 25.3 Severe stunting 24.0 18.4 20.8 25.3 22.5 22.2 30.3 26.0 23.9 16.4 14.8 19.3 Moderate underweight 28.7 26.8 26.3 29.0 24.5 27.6 33.0 30.3 28.1 21.4 23.2 25.4 Severe underweight 15.9 12.6 13.0 12.4 8.3 13.4 17.0 15.2 8.7 8.0 8.1 10.1 Mild anemia 18.5 18.5 21.1 23.5 22.5 20.2 (19.1) (27.8) (33.1) 23.1 23.2 24.0 Moderate anemia 46.3 52.9 50.3 39.4 30.1 46.6 (48.4) (34.0) (29.1) 41.9 39.5 39.7 Severe anemia 11.4 9.0 6.7 5.1 4.0 8.3 (4.9) (2.6) (8.2) 11.6 3.2 5.5 Women: Malnutrition 24.4 23.6 18.8 17.4 12.2 21.1 22.7 17.1 (14.7) 24.4 16.4 18.6 Mild anemia 33.6 34.0 32.2 28.7 13.6 31.5 33.2 (30.0) (36.4) 26.6 18.2 24.7 Moderate anemia 11.9 13.8 12.3 8.5 5.3 11.5 7.8 (9.3) (4.8) 20.8 6.7 9.7 Severe anemia 1.5 0.4 0.4 1.2 0.9 0.9 0.0 (1.6) 0.0 1.0 0.8 0.7 D. Female circumcision Prevalence of circumcision: Girls na na na na na na na na na na na na Women na na na na na na na na na na na na Prevalence of occlusion: Girls * * * * * * * * * * * * Women * * * * * * * * * * * * E. Sexually transmitted disease Prevalence of genital discharge: Women 1.8 1.8 1.9 1.1 0.4 1.6 0.0 0.8 1.3 0.0 0.9 0.7 Men na na na na na na na na na na na na Prevalence of genital ulcer: Women 0.7 0.7 0.8 1.1 1.6 0.9 1.2 2.1 0.0 0.8 1.0 1.0 Men na na na na na na na na na na na na - 13 - Madagascar 1997 - RURAL / URBAN POPULATIONS Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE Indicator Wealth Quintiles - Rural Wealth Quintiles - Urban Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Childhood immunization BCG coverage 47.1 58.5 60.8 83.6 (89.9) 62.1 (50.7) * * (78.1) 93.5 80.3 Measles coverage 32.2 35.5 38.1 61.3 (72.6) 42.0 (31.8) * * (59.0) 81.2 60.8 DPT coverage 33.0 37.3 41.5 69.0 (84.3) 45.5 (26.9) * * (53.0) 79.9 59.2 Full basic coverage 22.8 26.0 26.4 57.3 (70.0) 33.5 (16.4) * * (45.9) 64.5 46.4 No basic coverage 28.6 27.1 24.9 7.6 (5.2) 22.1 (35.7) * * (10.8) 3.0 12.4 Hepatitis B coverage na na na na na na na na na na na na Yellow fever coverage na na na na na na na na na na na na B. Treatment of childhood diseases Treatment of fever: Medical treatment of fever 22.7 24.7 24.2 20.9 (37.8) 24.0 (30.9) * * (46.3) 33.4 32.5 Treatment in a public facility 16.4 16.3 20.7 16.8 (23.5) 17.8 (20.2) * * (39.4) 17.3 20.6 Treatment in a private facility 6.3 7.7 3.6 4.0 (14.3) 6.0 (10.7) * * (6.9) 16.1 11.9 Treatment of acute respiratory infection: Medical treatment of ARI 34.4 34.3 31.6 33.5 * 34.7 (39.4) * * * 56.7 45.7 Treatment in a public facility 26.1 25.5 25.5 29.0 * 27.0 (28.5) * * * 32.7 32.6 Treatment in a private facility 8.3 8.7 6.1 4.5 * 7.7 (10.9) * * * 24.0 13.1 Treatment of diarrhea: Use of oral rehydration therapy 59.3 61.7 60.4 65.9 * 62.9 (65.2) * * (70.1) 92.4 75.9 Medical treatment of diarrhea 38.2 38.4 33.4 38.2 * 37.5 (43.0) * * (51.1) 59.0 46.7 Treatment in a public facility 26.0 23.0 25.5 30.5 * 25.8 (32.6) * * (37.6) 27.7 27.1 Treatment in a private facility 12.1 14.5 7.9 7.7 * 11.5 (10.5) * * (13.5) 30.1 19.1 C. Antenatal and delivery care Antenatal care visits: To a medically trained person 67.8 71.7 75.3 87.4 94.8 75.3 60.7 72.0 (75.7) 81.2 95.9 83.7 To a doctor 15.6 15.9 21.8 29.9 26.6 20.2 7.8 13.2 (18.3) 5.6 17.6 13.5 To a nurse or trained midwife 52.2 55.8 53.5 57.6 68.3 55.1 52.8 58.8 (57.4) 75.6 78.3 70.2 Multiple visits to a medically trained person 55.2 59.4 61.1 73.4 81.3 62.1 55.1 48.4 (66.1) 68.1 83.2 70.9 Antenatal care content: Tetanus toxoid 44.7 49.1 44.7 54.2 57.6 48.1 48.5 44.0 (54.6) 51.9 58.7 53.9 Prophylactic antimalarial treatment na na na na na na na na na na na na Iron supplementation na na na na na na na na na na na na Delivery attendance: By a medically trained person 29.8 35.9 40.2 62.7 84.1 42.1 27.7 45.7 43.3 74.2 90.3 68.3 By a doctor 7.4 8.7 12.9 24.7 32.4 13.2 2.0 3.1 8.9 6.6 14.3 9.2 By a nurse or trained midwife 22.4 27.2 27.2 38.0 51.7 28.8 25.7 42.6 34.5 67.6 76.0 59.1 In a public facility 24.8 27.2 33.4 44.6 38.4 31.5 20.0 41.2 33.0 43.1 38.4 36.3 In a private facility 0.7 0.7 1.2 2.4 2.9 1.2 2.2 0.0 2.4 1.2 6.6 3.8 At home 73.6 71.1 64.3 52.1 58.7 66.4 74.2 58.8 64.6 54.9 53.3 58.5 D. Contraceptive services Contraceptive prevalence: Women 2.2 3.0 6.7 12.6 23.0 7.1 2.9 3.1 13.1 11.5 24.1 17.6 Men na na na na na na na na na na na na Source of contraception - public sector: Women * * (71.0) 66.6 (33.2) 61.1 * * * * 39.1 43.1 Men na na na na na na na na na na na na - 14 - Madagascar 1997 - RURAL / URBAN POPULATIONS Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE (Cont.) Indicator Wealth Quintiles - Rural Wealth Quintiles - Urban Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. D. Contraceptive services (cont.) Source of contraception - private sector: Women * * (29.0) 31.7 (64.2) 37.6 * * * * 60.1 56.3 Men na na na na na na na na na na na na E. Treatment of adult illnesses Treatment of genital discharge, ulcer, sore: Women (33.2) (45.9) (38.1) (59.0) * 43.0 * * * * (55.5) (54.6) Men na na na na na na na na na na na na Treatment of genital discharge, ulcer, sore in public medical facilities: Women na na na na na na na na na na na na Men na na na na na na na na na na na na Voluntary counseling and testing for HIV/AIDS: Women na na na na na na na na na na na na Men na na na na na na na na na na na na - 15 - Madagascar 1997 - RURAL / URBAN POPULATIONS Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR Indicator Wealth Quintiles - Rural Wealth Quintiles - Urban Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Hygienic practices Disposal of children's stools: na na na na na na na na na na na na Sanitary disposal Handwashing: Wash hands prior to preparing food na na na na na na na na na na na na Handwashing facilities in household na na na na na na na na na na na na B. Bednet ownership and use Bednet ownership: Bednet ownership na na na na na na na na na na na na Treated bednet ownership na na na na na na na na na na na na Bednet use: By children na na na na na na na na na na na na By pregnant women na na na na na na na na na na na na C. Breastfeeding Exclusive breastfeeding 60.5 61.6 68.6 63.0 * 63.6 * * * * (60.5) 50.6 Timely complementary feeding 83.7 89.8 87.8 83.9 * 86.0 * * * * (100.0) 98.5 Bottle-feeding 1.6 0.0 1.2 2.7 (6.6) 1.6 * * * (9.2) 17.3 10.4 D. Micronutrient consumption Iodized salt: Availability of iodized salt in household 74.5 68.0 73.2 88.8 92.3 77.0 79.5 67.2 63.2 84.5 91.2 84.6 Vitamin A: Children 2.2 2.4 1.7 1.9 2.6 2.1 4.1 (3.8) 0.0 2.4 3.5 3.1 Women na na na na na na na na na na na na E. Tobacco and alcohol use Tobacco: Women na na na na na na na na na na na na Men na na na na na na na na na na na na Alcohol: Women na na na na na na na na na na na na Men na na na na na na na na na na na na F. Sexual practices Non-regular sexual partnerships: Women na na na na na na na na na na na na Men na na na na na na na na na na na na Condom usage with non-regular partner: Women na na na na na na na na na na na na Men na na na na na na na na na na na na G. Domestic violence Ever experienced violence na na na na na na na na na na na na Experienced violence in past year na na na na na na na na na na na na - 16 - Madagascar 1997 - RURAL / URBAN POPULATIONS Part IV: UNDERLYING DETERMINANTS OF HNP STATUS Indicator Wealth Quintiles - Rural Wealth Quintiles - Urban Low 2nd 3rd 4th High Avg. Low 2nd 3rd 4th High Avg. A. Education School completion: 9.2 11.5 16.8 30.4 64.9 20.2 10.0 9.1 12.3 38.6 74.6 55.5 Women 10.8 15.0 18.9 30.5 62.0 22.9 11.8 11.7 20.5 36.0 74.5 57.1 Men School participation: Girls 38.9 45.3 57.5 71.9 88.4 54.7 (54.0) (45.0) (62.7) 76.0 93.6 78.2 Boys 33.0 42.1 56.7 72.7 89.3 52.1 (62.4) (44.6) (58.9) 54.0 95.3 75.9 B. Exposure to mass media Newspaper readership: Women 12.3 14.0 21.0 27.3 39.3 19.8 15.2 22.1 27.1 35.4 52.7 43.5 Men na na na na na na na na na na na na Radio listenership: Women 6.1 25.8 27.8 49.3 79.5 30.3 8.8 20.4 39.2 45.4 81.5 63.5 Men na na na na na na na na na na na na Television viewership: Women 0.7 1.0 1.0 3.3 29.6 3.6 2.2 4.2 5.6 12.0 58.0 39.9 Men na na na na na na na na na na na na C. Knowledge and attitudes about HIV/AIDS Knowledge about sexual transmission of HIV/AIDS: Women 91.2 92.2 92.4 95.6 99.3 93.3 86.6 91.1 91.0 98.8 98.7 96.8 Men na na na na na na na na na na na na Knowledge about mother-to-child transmission of HIV/AIDS: Women 31.6 45.0 52.5 64.7 88.1 50.7 32.2 39.9 53.6 78.4 89.8 77.8 Men na na na na na na na na na na na na Attitudes toward HIV/AIDS: Women 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Men na na na na na na na na na na na na D. Status of women Household decisionmaking: Can seek own health care na na na na na na na na na na na na Can seek children's health care na na na na na na na na na na na na Can make daily household purchases na na na na na na na na na na na na Can make large household purchases na na na na na na na na na na na na Can make meal-related decisions na na na na na na na na na na na na Freedom of movement: Can travel to visit family, relatives na na na na na na na na na na na na Other decisionmaking, attitudes: Can decide how to spend own money 16.7 20.2 21.7 19.5 31.3 20.2 18.1 22.5 38.8 35.3 50.0 41.2 Can decide whether to have sex na na na na na na na na na na na na Justifies domestic violence na na na na na na na na na na na na E. Orphanhood Paternal orphan prevalence 5.8 6.0 6.1 4.8 3.7 5.6 6.9 5.0 4.8 5.0 4.1 4.7 Maternal orphan prevalence 2.4 3.9 3.7 4.1 2.7 3.4 1.9 8.4 3.6 3.2 2.6 3.3 Double orphan prevalence 0.4 0.7 0.5 0.2 0.2 0.4 0.0 1.4 0.4 0.4 0.5 0.5 - 17 - PART II. TECHNICAL NOTES A. INDICATOR DEFINITIONS B. DATA AND METHODS C. DISCUSSION INDICATOR DEFINITIONS Part I: HNP STATUS A. CHILDHOOD MORTALITY AND ILLNESS 2 Infant mortality rate: number of deaths to children under 12 months of age per 1,000 live births, based on experience during the ten years preceding the survey. Under-five mortality rate: number of deaths to children under five years of age per 1,000 live births, based on experience during the ten years preceding the survey. Prevalence of fever: percent of children who had fever, whether or not accompanied by cough or rapid breathing, in the two weeks before the survey. Prevalence of diarrhea: percent of children who had diarrhea in the two weeks before the survey. Prevalence of acute respiratory infection: percent of children who had a cough accompanied by rapid or difficult breathing in the two weeks before the survey. B. FERTILITY Total fertility rate (TFR): average number of births a woman could expect to have during her lifetime if she followed the levels of fertility currently observed at every age. The TFR is calculated as the sum of average annual age-specific fertility rates for all reproductive age groups (usually 15-49 years) in the three years before the survey. Adolescent fertility rate: age-specific fertility rate for women 15-19 years of age. This is the average number of births among women aged 15-19 years per 1,000 women in that age group, based on births in the three years before the survey and expressed as annual averages. 2Figures for the prevalence of fever, diarrhea, and acute respiratory infection refer to percent of children under three, four, or five years of age, depending upon the country. (The specific ages covered for in particular country may be determined by consulting the full report on that country's DHS, which may be found at: www.measuredhs.com/countries.) - 21 - C. NUTRITIONAL STATUS Children 3, 4, 5 Moderate stunting (height-for-age): percent of children with a height-for-age Z-score of between ­2 and ­3 standard deviations of the median reference standard for their age (as defined in fn. 4). Severe stunting (height-for-age): percent of children with a height-for-age Z-score of below ­3 standard deviations of the median reference standard for their age (as defined in fn. 4). Moderate underweight (weight-for-age): percent of children with a weight-for-age Z-score of between ­2 and ­3 standard deviations of the median reference standard for their age (as defined in fn. 4). Severe underweight (weight-for-age): percent of children with a weight-for-age Z-score of below ­3 standard deviations of the median reference standard for their age (as defined in fn. 4). Mild anemia: percent of children with a hemoglobin level of between 10.0g/dl and 10.9 g/dl, the World Health Organization criterion for mild anemia. Moderate anemia: percent of children with a hemoglobin level of between 7.0g/dl and 9.9g/dl, the World Health Organization criterion for moderate anemia. Severe anemia: percent of children with a hemoglobin level of below 7.0g/dl, the World Health Organization criterion for severe anemia. Women 6, 7 Malnutrition: percent of women aged 15-49 years with a Body Mass Index (BMI) of less than 18.5, where BMI ­ commonly used to indicate adult nutritional status ­ is defined as weight in kilograms divided by the square of height in meters. Mild anemia: percent of women aged 15-49 years with a hemoglobin level of between 10.0g/dl and 10.9g/dl for pregnant women and between 10.0g/dl and 11.9g/dl for non-pregnant women, the World Health Organization criterion for mild anemia. Moderate anemia: percent of women aged 15-49 years with a hemoglobin level of between 7.0g/dl and 9.9g/dl, the World Health Organization criterion for moderate anemia. Severe anemia: percent of women aged 15-49 years with a hemoglobin level of less than 7.0g/dl, the World Health Organization criterion for severe anemia. 3 All figures related to children's nutrition status refer to children under three, four, or five years of age, depending upon the country. (The specific ages covered for in particular country may be determined by consulting the full report on the DHS of the country concerned, which is available at: www.measuredhs.com/countries.) 4 The reference standards used for stunting and underweight are those established in the 1970s by the World Health Organization, the U.S. Centers for Disease Control, and the U.S. National Center for Health Statistics. Updated stunting and underweight figures based on the recently-released, 2006 World Health Organization reference standards are currently under preparation. When complete, they will be available at: www.worldbank.org/povertyandhealth/ countrydata. 5The anemia figures for children living at an altitude above 1,000 meters have been adjusted to reflect the higher level of hemoglobin required. 6In some countries, the BMI is presented for all women; in others, the figure is available only for mothers of children under five years of age. The reference population for any given country can be determined by consulting the full report on the DHS for the country concerned. An electronic version of this report is located at: www.measuredhs.com/ countries. 7 Anemia cut-off points for respondents who live at an altitude above 1,000 meters and/or who smoke have been adjusted to account for their higher hemoglobin requirements. - 22 - D. FEMALE CIRCUMCISION Prevalence of Circumcision Girls: percent of women aged 15-49 years with one or more daughters, at least one of whom had been circumcised. Women: percent of women aged 15-49 years who had been circumcised. Prevalence of Occlusion Girls: percent of women aged 15-49 years with one or more daughters, at least one of whom had been circumcised with the vaginal area sewn closed. Women: percent of women aged 15-49 years whose vaginal area had been sewn closed. E. SEXUALLY TRANSMITTED DISEASES Prevalence of Genital Discharge Women: percent of women aged 15-49 years who had had abnormal genital discharge in the twelve months before the survey. Men: percent of men aged 15-54 years who had had abnormal genital discharge in the twelve months before the survey. Prevalence of Genital Ulcer or Sore Women: percent of women aged 15-49 years who had had a genital ulcer or sore in the twelve months before the survey. Men: percent of men aged 15-54 years who had had a genital ulcer or sore in the twelve months before the survey. - 23 - Part II: INTERMEDIATE DETERMINANTS OF HNP STATUS ­ HEALTH SERVICE USE A. CHILDHOOD IMMUNIZATION 8 BCG coverage: percent of children who had received a dose of BCG vaccine by the time of the survey. Measles coverage: percent of children who had received a dose of measles vaccine by the time of the survey. DPT coverage: percent of children who had received three doses of DPT vaccine by the time of the survey. Full basic coverage: percent of children who had received a dose of BCG vaccine, measles vaccine, and three doses of DPT and polio vaccines by the time of the survey, excluding polio vaccine given at birth. No basic coverage: percent of children who had received no vaccination against the six early- childhood diseases (TB, measles, polio, diphtheria, pertusis, and tetanus) by the time of the survey. Hepatitis B coverage: percent of children who had received three doses of hepatitis B vaccine by the time of the survey. Yellow fever coverage: percent of children who had received a dose of yellow fever vaccine by the time of the survey. B. TREATMENT OF CHILDHOOD ILLNESSES 9 Treatment of Fever Medical treatment of fever: percent of children with fever, with or without cough or rapid breathing, in the two weeks before the survey who had sought medical advice for fever from any health facility or health provider, whether public or private. Treatment in a public facility: percent of children with fever, with or without cough or rapid breathing, in the two weeks before the survey who had sought medical advice for fever from a public-sector health facility or provider (as defined in fn. 9). Treatment in a private facility: percent of children with fever, with or without cough or rapid breathing, in the two weeks before the survey who had sought medical advice for fever from a private-sector health facility or provider (as defined in fn. 9). 8Childhood immunization figures refer to rates among children 12-23 months of age in all countries except those in Latin America and the Caribbean. There, the figures refer to rates among children 18-29 months of age. All figures are based on information recorded on the child's vaccination card; or, in cases where a card was not seen by the interviewer, on the mother's report. 9Figures for illness treatment in a public facility refer to treatment in government hospitals, health centers, health posts, or dispensaries; or in facilities operated by government-affiliated social securing programs. Figures for treatment in private facilities cover treatment in private hospitals or clinics, in private doctors' offices, or in facilities operated by other private medical providers (such as non-governmental organizations) as defined in the country concerned; but exclude treatment obtained in private pharmacies or shops. - 24 - Treatment of Acute Respiratory Infection (ARI) Medical treatment of ARI: percent of children with a cough and rapid breathing in the two weeks before the survey who had been taken for treatment at any medical facility or provider, whether public or private. Treatment in a public facility: percent of children with a cough and rapid breathing in the two weeks before the survey who had been taken for treatment at a public-sector health facility or provider (as defined in fn. 9). Treatment in a private facility: percent of children with a cough and rapid breathing in the two weeks before the survey who had been taken for treatment at a private-sector health facility or provider (as defined in fn. 9). Treatment of Diarrhea Use of oral rehydration therapy: percent of children with diarrhea in the two weeks before the survey who had received oral rehydration therapy (ORT) (defined as including consumption of oral rehydration salts, other recommended home fluids, or other increased liquids). Medical treatment of diarrhea: percent of children with diarrhea in the two weeks before the survey who had been taken for treatment at any medical facility or provider, whether public or private. Treatment in a public facility: percent of children with diarrhea in the two weeks before the survey who had been taken for treatment at a public-sector health facility or provider (as defined in fn. 9). Treatment in a private facility: percent of children with diarrhea in the two weeks before the survey who had been taken for treatment at a private-sector health facility or provider (as defined in fn. 9). C. ANTENATAL AND DELIVERY CARE 10 Antenatal Care (ANC) Visits To a medically-trained person: percent of women with one or more births in the five years before the survey who had received at least one antenatal care consultation from a medically- trained person (as defined in fn. 10) before her most recent birth. To a doctor: percent of women with one or more births in the five years before the survey who had received at least one antenatal care consultation from a doctor before her most recent birth. To a nurse or trained midwife: percent of women with one or more births in the five years before the survey who had received at least one antenatal care consultation from a nurse or trained midwife (as defined in fn. 10) before her most recent birth. Multiple visits to a medically-trained person: percent of women with one or more births in the five years before the survey who had received at least three antenatal care consultations from any medically-trained provider (as defined in fn. 10) before her most recent birth. 10When speaking of antenatal and delivery care, medically-trained persons are defined as doctors, nurses, and trained midwives. The definition excludes traditional midwives or other traditional birth attendants, whether trained or untrained. - 25 - Antenatal Care (ANC) Content Tetanus toxoid: percent of women with one or more births in the five years before the survey who had received at least one tetanus toxoid injection during her most recent pregnancy. Prophylactic antimalarial treatment: percent of women with one or more births in the five years before the survey who had received prophylactic treatment with any anti-malarial drug during her most recent pregnancy. Iron supplementation: percent of women with one or more births in the five years before the survey who had taken iron tablets during her most recent pregnancy. Delivery Attendance By a medically-trained person: percent of births in the five years before the survey attended by a medically-trained person (as defined in fn. 10). By a doctor: percent of births in the five years before the survey attended by a doctor. By a nurse or trained midwife: percent of births in the five years before the survey attended to by a nurse or a trained midwife (as defined in fn. 10). In a public facility: percent of all deliveries in the five years before the survey occurring in a public-sector health facility (as defined in fn. 9). In a private facility: percent of all deliveries in the five years before the survey occurring in a private-sector health facility (as defined in fn. 9). At home: percent of all deliveries in the five years before the survey occurring at home (defined as the woman's own or any other home). D. CONTRACEPTIVE SERVICES Contraceptive Prevalence 11 Women: percent of married or in-union women aged 15-49 years who used any modern means of contraception (as defined in fn. 11). Men: percent of married or in-union men aged 15-54 years who used any modern means of contraception (as defined in fn. 11). Source of Contraception - Public Sector Women: percent of married women who obtained their current method of contraception from a public-sector health facility or provider (as defined in fn. 9). Men: percent of married men who obtained their current method of contraception from a public- sector health facility or provider (as defined in fn. 9). 11Figures refer to use of modern means of contraception, defined as male/female sterilization, oral contraceptive pill, contraceptive injection, intrauterine device, male/female condom, diaphragm, cervical cap, contraceptive jelly or foam, implant, or some country-specific modern method. - 26 - Source of Contraception - Private Sector Women: percent of married women who obtained their current method of contraception from a private-sector health facility or provider (as defined in fn. 9, except that private pharmacies and shops are included rather than excluded). Men: percent of married women who obtained their current method of contraception from a private-sector health facility or provider (as defined in fn. 9, except that private pharmacies and shops are included rather than excluded). E. TREATMENT OF ADULT ILLNESSES Treatment of Genital Discharge, Ulcer, or Sore Women: percent of women with genital discharge, ulcer, or sore who sought any medical treatment for resulting symptoms. Men: percent of men with genital discharge, ulcer, or sore who sought any medical treatment for resulting symptoms. Treatment of Genital Discharge, Ulcer, or Sore in a Public Facility Women: percent of women with genital discharge, ulcer, or sore who sought treatment from a public-sector health facility or provider (as defined in fn. 9). Men: percent of men with genital discharge, ulcer, or sore who sought treatment from a public- sector health facility or provider (as defined in fn. 9). Voluntary Counseling and Testing for HIV/AIDS Women: percent of women aged 15-49 years who had been tested for HIV at any time before the survey. Men: percent of men aged 15-54 years who had been tested for HIV at any time before the survey. - 27 - Part III: INTERMEDIATE DETERMINANTS OF HNP STATUS ­ INDIVIDUAL AND HOUSEHOLD BEHAVIOR A. HYGIENIC PRACTICES Disposal of Children's Stools Sanitary disposal: percent of mothers with at least one child under five years of age who disposed of the stools of their youngest child in a sanitary manner (defined as dropping stool into a latrine, burying it, or using disposable diapers). Handwashing Wash hands prior to preparing food: percent of women aged 15-49 years preparing meals who washed hands before handling food. Handwashing facilities in household: percent of households that had hand-washing materials or facilities, as determined by direct observation of interviewers. B. BEDNET OWNERSHIP AND USE Bednet Ownership Bednet ownership: percent of households owning one or more bednets. Treated bednet ownership: percent of households owning one or more bednets that had recently been treated with insecticides. Bednet Use By children: percent of households with at least one child under five years of age, some or all of whom had slept under a bednet the night before the survey. By pregnant women: percent of currently pregnant women who had slept under a bednet the night before the survey. C. BREASTFEEDING Exclusive breastfeeding: percent of children 0-3 months of age who had received only breast milk in the 24 hours before the survey. Timely complementary feeding: percent of children 6-9 months of age who had received breast milk and solid or semi-solid foods in the twenty-four hours before the survey. Bottle-feeding: percent of children under 12 months of age who had received any food or drink from a bottle with a nipple in the twenty-four hours before the survey. - 28 - D. MICRONUTRIENT CONSUMPTION Iodized Salt Availability of iodized salt in household: percent of households with cooking salt testing positive for iodine/iodate at the recommended level of 15 or 25 parts per million or more (depending on the country). 12 Vitamin A Children: percent of children13 who had received at least one dose of vitamin A in the six months before the survey, as reported by the mothers. Women: percent of women who had received a dose of vitamin A within two months of the last birth, in the five years before the survey. E. TOBACCO AND ALCOHOL USE Tobacco 14 Women: percent of women aged 15-49 years who currently were smoking or chewing tobacco products. Men: percent of men aged 15-54 years who currently were smoking or chewing tobacco products. Alcohol Women: percent of women aged 15-49 years who had gotten intoxicated due to excessive consumption of alcohol in the three months before the survey. Men: percent of men aged 15-54 years who had gotten intoxicated due to excessive consumption of alcohol in the three months before the survey. F. SEXUAL PRACTICES Non-Regular Sexual Partnerships Women: percent of women aged 15-49 years who had had sex with a non-regular partner at least once in the twelve months before the survey. Men: percent of men aged 15-54 years who had had sex with a non-regular partner at least once in the twelve months before the survey. 12Which of these two levels is recommended in any particular country may be determined by consulting the full report on that country's DHS, which may be found at: www.measuredhs.com/countries. 13Figures refer to children over six months of age and under three, four, or five years of age, depending upon the country. (The specific ages covered for in a particular country may be determined by consulting the full report on that country's DHS, which is available at: www.measuredhs.com/countries.) 14Tobacco products include cigarettes, pipes, cigars, leaves, etc. - 29 - Condom Usage with Non-Regular Partner Women: percent of women aged 15-49 years with non-regular sexual partner who had used a condom in the last sexual intercourse with such a partner in the twelve months before the survey. Men: percent of men aged 15-54 years with non-regular sexual partner who had used a condom in the last sexual intercourse with such a partner in the twelve months before the survey. G. DOMESTIC VIOLENCE Ever experienced violence: percent of women aged 15-49 years who had ever been hit or beaten by current or former husband/partner. Experienced violence in past year: percent of women aged 15-49 years who had been hit or beaten by current or former husband/partner in the twelve months before the survey. - 30 - Part IV: UNDERLYING DETERMINANTS OF HNP STATUS A. EDUCATION School Completion Women: percent of women aged 15-49 years who had completed the fifth grade. Men: percent of men aged 15-54 years who had completed the fifth grade. School Participation Girls: percent of girls aged 6-10 years who were attending school at the time of the survey. Boys: percent of boys aged 6-10 years who were attending school at the time of the survey. B. EXPOSURE TO MASS MEDIA Newspaper Readership Women: percent of women aged 15-49 years who read a newspaper at least once a week. Men: percent of men aged 15-54 years who read a newspaper at least once a week. Radio Listenership Women: percent of women aged 15-49 years who listened to radio at least once a week. Men: percent of men aged 15-54 years who listened to radio at least once a week. Television Viewership Women: percent of women aged 15-49 years who watched television at least once a week. Men: percent of men aged 15-54 years who watched television at least once a week. C. KNOWLEDGE AND ATTITUDES ABOUT HIV/AIDS Knowledge about Sexual Transmission of HIV/AIDS 15 Women: percent of women aged 15-49 years who knew of HIV/AIDS and of at least one of the following ways to avoid it through interruption of its sexual transmission route: abstinence; using a condom; avoiding multiple sex partners, sex with prostitutes, and unprotected homosexual sex. Men: percent of men aged 15-54 years who knew of HIV/AIDS and of at least one of the ways to avoid HIV/AIDS referred to in the preceding definition. 15In most countries, the survey sample included both married and unmarried individuals. Where this was the case, all respondents, regardless of marital status, were asked the question covered in this section. Where the survey covered only individuals who were or had been married, the data pertain only to individuals who had ever been married. (The marital status of people covered for in particular country is indicated in the full report on that country's DHS, which is located at: www.measuredhs.com/countries.) - 31 - Knowledge about Mother-to-Child Transmission of HIV/AIDS Women: percent of women aged 15-49 years who knew of at least one way HIV/AIDS can be transmitted from mother to child during pregnancy, delivery, or breastfeeding. Men: percent of men aged 15-54 years who knew of at least one way HIV/AIDS can be transmitted from mother to child during pregnancy, delivery, or breastfeeding. Attitudes toward HIV/AIDS Women: percent of women aged 15-49 years who believed that people with HIV/AIDS should be allowed to continue working or that HIV test results should remain confidential. Men: percent of men aged 15-54 years who believed that people with HIV/AIDS should be allowed to continue working or that HIV test results should remain confidential. D. STATUS OF WOMEN Household Decisionmaking Can seek own health care: percent of women age 15-49 years who could decide by themselves to seek their own health care. Can seek children's health care: percent of women aged 15-49 years, whose children live with them, who could decide by themselves to seek health care for their children. Can make daily household purchases: percent of women aged 15-49 years who could decide by themselves or jointly with others to make daily household purchases. Can make large household purchases: percent of women aged 15-49 years who could decide by themselves or jointly with others to make large household purchases. Can make meal-related decisions: percent of women aged 15-49 years who could decide by themselves what food to cook daily. Freedom of Movement Can travel to visit family, relatives: percent of women aged 15-49 years who could decide by themselves to visit family and relatives. Other Decisionmaking, Attitudes Can decide how to spend own money: percent of women aged 15-49 years who work for cash who could decide by themselves on how to use the money they earn. Can decide whether to have sex: percent of women aged 15-49 years agreeing that they can refuse to have sex with their husband for at least one of the following reasons: he has a sexually- transmitted disease; he has had sexual relations with another woman; or the woman is tired, not in mood, or recently has given birth. Justify domestic violence: percent of women aged 15-49 years believing that a husband/male partner would be justified in beating his wife/female partner for at least one of the following reasons: he suspects her of being unfaithful; she argues with him; she goes out without telling him; she neglects the children; she burns the food; or other, country-specific reasons (for example, she shows disrespect for her in-laws or her family does not give the expected dowry). - 32 - E. ORPHANHOOD Maternal orphan prevalence: percent of children under 15 years of age whose natural mother had died before the survey. Paternal orphan prevalence: percent of children under 15 years of age whose natural father had died before the survey. Double orphan prevalence: percent of children under 15 years of age both of whose natural parents had died before the survey. - 33 - DATA AND METHODS Any assessment of the figures featured in this report requires an appreciation of how they were prepared. The first need is to understand the basic features of the data and methods employed. A. SOURCE OF FIGURES The figures appearing in this report are all derived from data collected under the Demographic and Health Surveys (DHS) program conducted by ORC Macro, with support from the U.S. Agency for International Development and other external assistance organizations. Large DHS household surveys have been carried out, usually at periodic intervals, in approximately seventy- five countries across Africa, Asia, Latin America, the Middle East, and the former Soviet Union.16 This series of reports covers the fifty-six of those countries that had one or more DHS surveys undertaken since 1990, for which data were publicly available as of June 2006. (Annex C is a list of the countries for which reports have been prepared.) In each country, the DHS program has gathered information on a large number of indicators about health, nutrition, and population (hnp) status and service use; about relevant behaviors of household members; and about household characteristics like those described below. It has done this through a set of questionnaires, similar in all countries, to collect data at the individual, household, and community levels. The data presented here draw on responses to the individual and household questionnaires. In most cases, they are based on responses from women or other family members interviewed. The principal exceptions concern nutritional status, based on anthropometric measurement; immunization, which typically relies to the extent possible on record cards maintained at the household level; and those other items where a source other than interviewer response is specifically identified. B. MEASUREMENT OF ECONOMIC STATUS Wealth or Asset Approach Economic status has been expressed in terms of wealth or assets: specifically, on the basis of information about household characteristics gathered through the DHS household questionnaire. (Such information was normally provided for at least 25-30, and often many more, characteristics like the presence, availability, or use of a fan, radio receiver, or automobile; housing materials like wood or concrete flooring, tile or tin roofing, or cement block walls; superior sources of water like piped or a protected well; and other attributes related to economic status.) 16Further information about the DHS program is available at the program's website: www.measuredhs.com. - 35 - Index Construction A single, consolidated index of living standards17 was constructed by using principal components analysis (PCA) to generate a weight for each household item with available information. A wealth index score was calculated for each household by weighting the response with respect to each item pertaining to that household by the coefficient of the first principal component as determined by application of principal components analysis, and summing the results. The resulting household scores were standardized in relation to a standard normal distribution with a mean of zero and a standard deviation of one. All individuals usually present in each household were assigned the household's standardized wealth index score, and all individuals in the sample population were ranked according to that score. The sample population was then divided into quintiles of individuals, with all individuals in a single household being assigned to the same quintile. The same standardized household wealth index scores originally derived for the total population sample, as just described, were also used in preparing the disaggregated estimates for female and male members of the sample population, and for rural and urban residents. In preparing those disaggregated estimates, the entire population sample was divided into quintiles of individuals; the females and males, and the rural and urban residents in each quintile of the entire sample were then separated from one another; and the mean for each of the ten resulting subgroups (five female, five male; or five rural, five urban) was calculated. C. CALCULATION AND PRESENTATION OF RATES Use of Sampling Weights Rates for all health, nutrition, and population indicators are calculated after applying the DHS sampling weights. (DHS surveys often over-sample certain small subgroups of interest ­ residents of a particular geographic area, for example ­ in order to get sample sizes large enough to produce statistically-significant results. The DHS sampling weights are used to compensate for such over-sampling in order to ensure that the results are representative of the population as a whole and not just of the DHS sample.) Calculation of Total Population Averages The average for the total population presented alongside the quintile-specific rates for each indicator is calculated without reference to quintile divisions. It thus equals the weighted mean of the quintile rates, with the weight assigned to each quintile rate being the proportion of the number of individuals at risk (as defined on p. 41) for the indicator concerned. Sampling Errors Information needed to assess the statistical significance of differences among the quintile-specific rates is presented in three ways: · First, in all the basic tables presented in part I, rates are shown in parentheses or replaced by asterisks in cases where the standard error is likely to be unacceptably high because of small sample size. The number of observations used to determine how to present the data for the different indicators covered were as follows: 17 Such an index is usually referred to as either an "asset index" or a "wealth index." The two expressions are used interchangeably in this report; for ease of communication, "wealth index" appears more frequently despite the inexact correspondence between the items included in the index's construction and those appearing in more conventional, financially-based definitions of wealth. - 36 - Indicator Unit of measure Number of observations used to determine how quintile-specific rate was presented Without With Replaced by parentheses parentheses asterisk Infant and child Number of deaths >500 250-499 <250 mortality Total and adolescent Number of births >250 125-249 <125 fertility All other indicators Number of >50 25-49 <25 individuals · Second, the standard error for each quintile-specific rate (except for any rate replaced by an asterisk) appearing in the total population table is provided in part III.B. (Standard errors for the quintile-specific rates presented in the female-male and rural-urban tables are not available.) · Third, the right-hand column of the total population table provides the standard error for the concentration index, one of the measures of inequality shown, as indicated below. D. MEASUREMENT OF INEQUALITY Accompanying each of the indicators presented in the total population table are the values for three statistical measures of inequality: · Low/High Quintile Ratio: the ratio between the rate prevailing in the lowest (poorest) population quintile and that found in the highest (least poor) quintile. · Low-High Quintile Difference: the value of the lowest quintile minus the value of the highest, expressed as an absolute value. · Concentration Index: twice the area in a Lorenz-type diagram between the line of equality and the concentration curve for the indicator in question, the curve being the graph of the cumulative share of the indicator against the cumulative share in the asset distribution. (The value, which can range from -1 to +1, is negative when the hnp indicator is higher among the poor (e.g., fertility), positive when it is higher among the better-off (e.g., modern contraceptive use), and zero when on balance the indicator shows no systematic relationship with wealth.)18 18Adam Wagstaff, Pierella Paci, and Eddy van Doorslaer, "On the Measurement of Inequalities of Health," Social Science and Medicine 33 (1991): 545-57. (See also chapter eight in the volume by O'Donnell, van Doorslaer, Wagstaff, and Lindelow described in the annotated bibliography that constitutes annex A.) - 37 - DISCUSSION While a basic understanding of the data and methods employed is necessary to adequately appreciate the figures appearing in this report, it is not sufficient. For the application of the approach taken involves many subtleties that also need to be understood. Among the more important are: A. DESCRIPTIVE NATURE OF THE RELATIONSHIPS The hnp-poverty relationships shown in this report are no more than descriptive. They should not be taken to imply any direct causal relationships, for several reasons. One reason is the possibility that it is not wealth or asset possession per se that determine a person's health condition. Rather, the determining factors could be other characteristics (such as education or ethnic background) that are simultaneously associated with both asset ownership and health status. It is also possible that the health-poverty relationships shown are driven primarily by particular items included in the index (e.g., water and sanitation). Should this be the case, improvements in health conditions among the poor might be more effectively brought about by focusing on changing those particular components of the wealth index rather than by a general effort to increase economic status as measured by the index as a whole. B. IMPLICATIONS OF A WEALTH/ASSET APPROACH Wealth or Assets as a Measure of Economic Status Reliance on a wealth index to measure economic status is a rather recent development in research on economic disparities, where such status traditionally has been defined in terms of consumption or income. The principal reason for the choice of the wealth index is pragmatic rather than conceptual: the DHS surveys, which are of interest because of the plethora of hnp information that they contain, do not collect consumption or income data; but they do have detailed information on households' physical characteristics, and on the household-level presence of and access to a wide range of goods and services. While there is some disagreement about the relative merits of using such wealth/asset information or consumption data to measure economic status, several recent studies suggest that the asset-consumption relationship is quite close.19 To the extent this is the case, an index of wealth or asset possession/availability can be taken as a 19See, for example, Deon Filmer and Lant H. Pritchett, "Estimating Wealth Effects without Expenditure ­ or Tears: An Application to Educational Enrollments in States of India," Demography 38, no. 1 (February 2001): 115-32; Shea Rutstein and Kiersten Johnson, The DHS Wealth Index, DHS Comparative Reports No. 6 (Calverton, Maryland, USA: ORC Macro, August 2004) esp. 10-14; Adam Wagstaff and Naoko Watanabe, "What Difference Does the Choice of SES Make in Health Inequality Measurement?" Health Economics 12 (2003): 885-90. - 39 - reasonably satisfactory proxy for consumption, rather than or in addition to serving as an indicator of economic status in its own right. C. ECONOMIC STATUS AS A MEASURE OF SOCIO-ECONOMIC WELL-BEING Like consumption or income, a wealth index defines disparities that are primarily economic. This is by no means the only way to define inter-group inequalities that are of potential concern. Other possibilities include gender, place of residence, education, ethnic background, or other factors associated with social exclusion. Thus the economic perspective provides only a partial view of the multidimensional concepts of poverty, inequality, and inequity. By including tables on female-male and rural-urban inequalities, this report pays adequate attention to two of inequality's other important dimensions to justify the use of the term "socio- economic" rather than simply "economic" in its title. However, the centrality of tabulations based on the wealth index means that the primary focus is on the economic dimension. The justification for this lies not in the greater importance of economic considerations, but rather in the recently-improved ability to analyze and thus begin dealing with them. Until the development of the wealth/asset approach, the assessment of economic status had been based on consumption, expenditures, or income, all of which are far more difficult to measure than such other, non- economic dimensions of inequality as gender, ethnic identity, educational level, and place of residence. As a result, assessments of health inequalities by economic status had lagged well behind measurements in terms of these other dimensions, especially gender and education. The focus on inequality's economic aspect applied here represents an effort to redress this imbalance. D. INDEX CONSTRUCTION Choice of Items Use of a wealth index requires decisions about which items to include in it. In the case of secondary analyses like the one featured in this report, the choice is limited to those items included in the data sets being used. But even with this constraint, there nonetheless remains considerable room for choice, given the large number of items for which information is collected by the DHS. The decision made in preparing this report was to include all items in each DHS household questionnaire that relate to ownership of household goods; to dwelling unit construction and characteristics; and to access to services and resources like electricity, water, and sanitation facilities. Also included were other potential indicators of wealth, such as live-in domestic servants. This decision, admittedly somewhat arbitrary, has both advantages and disadvantages. The principal advantage is practical: use of a large number of assets increases the degree of variation across household asset scores and facilitates a more regular distribution of individuals across quintiles. It also reduces the possibility of subjectivity in selecting only some of the variables for inclusion on some a priori basis; and it may increase a wealth index's accuracy as a proxy for consumption. However, including all variables is far from satisfying conceptually. For example, it means failing to discriminate with respect to the items' differing natures. It is not clear, for instance, whether access to water, sanitation, electricity, or other publicly-provided resources should be included in an index that purports to measure private household wealth. - 40 - Further, many items that are candidates for inclusion in a DHS-based wealth index might be seen as directly influencing health status: water and sanitation for infant and child mortality, for example. It would be desirable to include quintile-specific estimates for such items; but to the extent that such items have large index coefficients, any estimates for those items would be suspect. Such items appear to be relatively few and of limited statistical significance in the index used here. However, for the sake of caution, quintile-specific estimates for items appearing in the index have nonetheless been excluded from the basic tables and appear only in supporting table III.C Weighting of Items A further decision required in construction of an index concerns the weight to attach to each of the respective items. As noted earlier, the method used in this report is principal components analysis (PCA). Adoption of this method was based on the findings, referred to earlier, that its use resulted in outcomes that approximated reasonably well those produced by taking a consumption or expenditure approach. Further, it often provides greater discrimination in economic status than does the use of consumption/expenditures. It has also emerged as the standard approach for use in analyses of the sort presented here, so that its adoption is largely non-controversial. Yet this choice, too, is not without an arbitrary aspect; for alternative plausible methods exist. Examples include the "inverse possession" approach, which gives more weight to items possessed by only a few and less to those possessed by many;20 or, perhaps, the common practice of simply assigning the same weight to each index item. Use of Principal Components Analysis with Dichotomous Variables An additional issue concerns the use of a technique like PCA, developed for use with continuous variables, in the construction of an index based primarily on dichotomous variables. While legitimate in principle, any reservations in this regard are of limited practical consequence, since the considerable experimentation undertaken in preparation for the tabulations presented here indicated that any inaccuracy introduced by applying PCA to the analysis of the dichotomous values used is minimal. Economies of Scale Calculating the values for a household wealth index also requires a decision concerning economies of scale that exist in the households covered. The calculations presented here assume complete economies of scale. The implicit assumption is that additional members do not add to household expenses on items included in the index. E. DEFINITION OF QUINTILES Quintiles of Individuals As noted earlier, the quintile-specific figures presented in this report refer to quintiles of individuals in the household population. Such quintiles need to be distinguished from quintiles of households or quintiles of only those people in the population who are "at risk": that is, subject to the particular condition, eligible for the particular service, or capable of behaving in a particular way (children born alive for infant and under-five mortality, for example; or adult men and women for condom use or non-regular sexual partnerships). 20Saul S. Morris et al., "Validity of Rapid Estimates of Household Wealth and Income for Health Surveys in Rural Africa," Journal of Epidemiology and Community Health 54 (2000): 381-87. - 41 - The expression of findings in terms of quintiles of individuals has several implications: · Because fertility is often higher in lower economic households than among better-off ones, the number of individuals per household will frequently be larger among the poor than among higher-income groups. In such cases, the number of households will vary systematically across quintiles of individuals, and the results expressed in terms of quintiles of households can differ significantly from those presented here. · The proportion of individuals "at risk" with regard to a particular indicator is also likely to vary across quintiles in many cases. (For example, in cases where fertility is higher among poor people, a higher-than-average proportion of poor populations will consist of newborns at risk from infant mortality, young children subject to malnutrition, and pregnant women for whom antenatal care is relevant.) To facilitate the work of any investigators wishing to undertake calculations based on people at risk, the number of such people in each quintile of individuals is shown in part III.A. · As previously indicated, the population average figure provided for each indicator is equivalent to the weighted sum of the quintile rates for that indicator, where the weight assigned to each quintile rate is the number of people at risk in each quintile as presented in part III.A. As a result of this weighting, the population average will usually differ from a simple mean of the population quintile estimates. Quintiles of Males and Females, of Rural and Urban Residents As also reported in the data and methods section, the tables on rural and urban residents and on men and women were prepared using the same asset scores as for the total population; and rural- urban residents and females-males were separated from one another only after the entire sample had been disaggregated into quintiles of individuals. This means that the figures given in the rural-urban and female-male tables refer to females-males and rural-urban residents belonging to each quintile of individuals in the total population, as distinct from quintiles of females, of males, of rural residents, or of urban residents alone. The consequence of this distinction is particularly evident with regard to rural and urban residents. Since rural residents tend to be poorer than urban dwellers, they normally form a considerably higher proportion of individuals in the lower economic quintiles of the total population than in the higher ones. Conversely, urban residents tend to be concentrated in the higher economic groups. As a result, the number of individuals in each of the urban and rural quintiles usually varies greatly and systematically; and when this is the case, the figures presented in the rural-urban tables can differ significantly from those produced by a computation procedure that places the same number of rural and urban residents in each rural quintile or each urban quintile. (The results may also differ significantly from application of an approach featuring the separate calculation of index values for urban and for rural groups. While such separate index values may well be preferable conceptually, their calculation involves complexities that prevented their preparation for this report.) - 42 - F. COMPARISON OF QUINTILES ACROSS COUNTRIES Reliance on population quintiles as basic presentational format for the data appearing in this report implicitly incorporates a relative concept of poverty. This differs from an absolute concept of poverty under which the population would be divided into groups of different sizes according to some absolute standard of living (such as people earning less than one dollar a day, between one and two dollars a day, and more than two dollars a day). This means that, when comparing values of an indicator among people in a given quintile across countries, the comparison is between groups of people whose economic status can be quite different. The lowest quintile of a Latin American population, for example, will usually be considerably better-off than the lowest quintile in an African country. G. STATISTICAL INDICATORS OF INEQUALITY The available statistical indicators of inequality are far too numerous to permit use of more than a small proportion of them in presenting the findings featured in this report. The three indicators employed have been selected to provide a wide range of perspectives. Two are designed for ease of understanding, the third for greater technical accuracy. The low/high quintile ratio and low-high quintile difference are the two presented for ease of understanding. The former is a relative measure, the latter an absolute measure that can produce a significantly different impression from that provided by the former. The concentration index is provided for the benefit of technical specialists wishing greater accuracy. It measures the degree of inequality in an hnp indicator across the full wealth index distribution, rather than differences between only two of the five quintiles, and also reflects the relative size of the different asset-based divisions of the study population. H. COMPARABILITY WITH OTHER REPORTS Tabulations similar to those presented here can also be found in the initial series of hnp/poverty country reports, issued in 2000, and in the recent country reports issued by the DHS program. The figures presented in those reports normally resemble quite closely those appearing here for any given indicator; but there are often slight differences for one or more of several reasons. The most common is a difference in the definition of the indicator in question. (These differences are usually small and subtle. But there is one important exception: the definition of moderate malnutrition among children. In the 2000 reports, this was defined as second and third degree malnutrition taken together. Here, it is defined as second degree malnutrition alone.) Another reason, with respect to infant and child mortality, is that the figures in the DHS documents are typically based on experience during the five years before the survey in question, rather than during the ten previous years as in this report. A further frequent reason is the use of an improved computational technique. In addition, asset-based, quintile-specific tabulations of hnp indicators have begun to appear in an increasing number of other documents. Given the basic similarity of approach, such tabulations usually produce results that are generally congruent with those shown in part I of this report. However, significant divergences have occasionally been reported; and the absence of adequately detailed information about data and methods often prevents any fully-satisfying understanding of the approaches used. As a result, occasional doubts and frequent uncertainties about approach comparability remain. - 43 - PART III. SUPPORTING TABLES A. SAMPLE SIZES B. STANDARD ERRORS C. ASSET DISTRIBUTION AND WEIGHTS Madagascar 1997 - SAMPLE SIZES TOTAL SAMPLE Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total Number of household members All 7,050 7,169 6,883 7,074 7,050 35,226 Urban 758 755 675 1,318 5,117 8,623 Rural 6,292 6,414 6,208 5,756 1,933 26,603 Female 3,577 3,583 3,433 3,461 3,438 17,491 Male 3,473 3,586 3,448 3,613 3,613 17,733 Part I: HNP STATUS Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total Mortality rates All 3,407 2,472 2,259 2,165 1,721 12,024 Urban 374 268 206 448 1,208 2,504 Rural 3,033 2,204 2,053 1,717 513 9,519 Female 1,713 1,184 1,102 1,050 859 5,908 Male 1,694 1,288 1,156 1,115 862 6,115 Prevalence of fever, diarrhea, acute respiratory infection All 1,003 716 697 653 468 3,537 Urban 104 76 66 132 326 704 Rural 899 641 631 522 141 2,834 Female 527 376 326 307 237 1,773 Male 477 340 371 346 230 1,764 Total fertility rate All 4,078 3,852 3,418 3,855 4,597 19,801 Urban 442 425 316 827 3,545 5,555 Rural 3,636 3,428 3,102 3,028 1,052 14,246 Age-specific fertility rate 15-19 All 951 968 807 894 1,043 4,664 Urban 112 130 78 192 831 1,343 Rural 839 839 729 702 212 3,321 Children's nutritional status All 836 625 605 605 409 3,080 Urban 83 64 55 117 280 599 Rural 753 561 550 488 129 2,481 Female 427 330 281 289 200 1,527 Male 409 295 324 316 209 1,553 Children's anemia status All 654 466 447 436 319 2,321 Urban 70 43 47 89 226 475 Rural 584 423 400 346 93 1,846 Female 313 225 242 228 172 1,180 Male 341 241 204 208 147 1,141 - 47 - Madagascar 1997 - SAMPLE SIZES Part I: HNP STATUS (Cont.) Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total Women's nutritional status All 725 553 485 479 363 2,604 Urban 59 63 47 102 250 521 Rural 666 489 438 377 112 2,082 Women's anemia status All 922 681 635 605 457 3,299 Urban 96 68 62 134 328 688 Rural 825 612 572 471 128 2,608 Girls' circumcision All na na na na na na Urban na na na na na na Rural na na na na na na Women's circumcision All na na na na na na Urban na na na na na na Rural na na na na na na Prevalence of genital discharge, ulcer, sore Female 1,346 1,284 1,082 1,174 1,320 6,206 Urban Female 149 145 98 251 1,015 1,658 Rural Female 1,197 1,139 984 922 305 4,547 Male na na na na na na Urban Male na na na na na na Rural Male na na na na na na - 48 - Madagascar 1997 - SAMPLE SIZES Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total Immunization coverage All 324 241 233 251 160 1,209 Urban 41 20 20 58 117 256 Rural 283 221 213 193 43 953 Female 167 125 114 115 82 603 Male 157 115 119 137 78 606 Treatment of fever All 332 229 242 194 140 1,138 Urban 38 19 25 38 94 214 Rural 293 210 217 156 46 922 Female 164 119 129 96 69 577 Male 167 111 113 98 71 560 Treatment of acute respiratory infection All 274 183 174 140 75 845 Urban 38 13 20 28 51 150 Rural 236 170 155 112 23 696 Female 149 105 70 64 37 425 Male 125 78 105 75 38 421 Treatment of diarrhea All 292 156 182 204 122 957 Urban 44 20 17 39 92 212 Rural 249 136 165 166 30 746 Female 156 77 79 91 61 464 Male 136 79 103 113 61 492 Antenatal and delivery care All 921 689 641 608 447 3,305 Urban 93 79 58 128 317 675 Rural 827 610 583 479 130 2,629 Contraceptive prevalence Female 1,005 888 765 840 936 4,435 Urban Female 114 106 64 165 684 1,133 Rural Female 891 782 701 676 252 3,302 Male na na na na na na Urban Male na na na na na na Rural Male na na na na na na Contraceptive source Female 23 27 56 104 223 432 Urban Female 3 3 8 19 165 199 Rural Female 20 24 47 85 58 233 Male na na na na na na Urban Male na na na na na na Rural Male na na na na na na Treatment of genital discharge, ulcer, sore Female 33 35 31 24 31 154 Urban Female 2 4 1 2 24 33 Rural Female 31 31 30 22 7 121 Male na na na na na na Urban Male na na na na na na Rural Male na na na na na na - 49 - Madagascar 1997 - SAMPLE SIZES Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total Sanitary disposal of stools All na na na na na na Urban na na na na na na Rural na na na na na na Wash hands prior to preparing food All na na na na na na Urban na na na na na na Rural na na na na na na Handwashing facilities in household? All na na na na na na Urban na na na na na na Rural na na na na na na Bednet ownership All na na na na na na Urban na na na na na na Rural na na na na na na Bednet use by children All na na na na na na Urban na na na na na na Rural na na na na na na Bednet use by pregnant women All na na na na na na Urban na na na na na na Rural na na na na na na Exclusive breastfeeding All 135 93 92 74 44 438 Urban 17 9 9 15 32 82 Rural 118 84 83 58 11 354 Female 70 41 57 34 20 222 Male 65 52 35 40 24 216 Timely complementary breastfeeding All 121 88 76 72 55 412 Urban 8 8 4 15 38 73 Rural 114 80 73 58 17 342 Female 56 46 48 40 29 219 Male 65 42 28 32 26 193 Bottle-feeding All 383 263 267 217 151 1,281 Urban 32 27 22 40 106 227 Rural 351 235 244 177 45 1,052 Female 185 122 154 111 68 640 Male 198 141 113 106 82 640 - 50 - Madagascar 1997 - SAMPLE SIZES Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR (Cont.) Wealth Quintiles Pop. Indicator Low 2nd 3rd 4th High Total Iodized salt in household All 1,102 1,404 1,414 1,354 1,321 6,595 Urban 121 145 143 239 951 1,599 Rural 981 1,259 1,271 1,114 370 4,995 Vitamin A supplementation All 801 582 550 537 398 2,868 Urban 84 62 51 112 275 584 Rural 717 520 499 424 124 2,284 Female 380 284 281 290 203 1,438 Male 421 298 269 247 196 1,431 Tobacco and alcohol use, casual sexual partners, condom use for casual sex Female na na na na na na Urban Female na na na na na na Rural Female na na na na na na Male na na na na na na Urban Male na na na na na na Rural Male na na na na na na Domestic violence All na na na na na na Urban na na na na na na Rural na na na na na na - 51 - Madagascar 1997 - SAMPLE SIZES Part IV: UNDERLYING DETERMINANTS OF HNP STATUS Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Total School completion (Grade 5) Female 1,492 1,541 1,383 1,507 1,773 7,695 Urban female 164 161 135 307 1,356 2,124 Rural female 1,327 1,380 1,248 1,200 416 5,571 Male 1,348 1,494 1,450 1,540 1,951 7,782 Urban male 148 160 135 270 1,423 2,137 Rural male 1,200 1,333 1,315 1,270 528 5,646 School participation Female 606 558 527 544 427 2,661 Urban female 64 65 48 97 301 575 Rural female 542 492 480 447 125 2,086 Male 646 566 509 524 483 2,727 Urban male 66 70 49 117 333 634 Rural male 580 496 460 407 149 2,093 Mass media exposure Female 1,436 1,377 1,214 1,374 1,659 7,060 Urban female 158 152 109 288 1,281 1,987 Rural female 1,279 1,225 1,106 1,086 378 5,073 Male na na na na na na Urban male na na na na na na Rural male na na na na na na Knowledge of HIV/AIDS prevention Female 2,591 1,377 1,214 1,374 1,659 8,215 Urban female 158 152 109 288 1,281 1,988 Rural female 1,279 1,225 1,106 1,086 378 5,074 Male na na na na na na Urban male na na na na na na Rural male na na na na na na Household decisionmaking and justification of violence All 1,436 1,377 1,214 1,374 1,659 7,060 Urban 158 152 109 288 1,281 1,987 Rural 1,279 1,225 1,106 1,086 378 5,073 Orphanhood All 3,773 3,212 3,023 3,063 2,593 15,665 Urban 395 350 294 622 1,838 3,499 Rural 3,378 2,862 2,730 2,441 755 12,166 Female 1,905 1,552 1,477 1,474 1,288 7,697 Male 1,868 1,661 1,543 1,589 1,305 7,965 - 52 - Madagascar 1997 - STANDARD ERRORS OF QUINTILE ESTIMATES FOR TOTAL POPULATION Part I: HNP STATUS Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Avg. A. Childhood illness and mortality Infant mortality rate 8.87 9.38 7.90 7.81 6.81 4.78 Under-five mortality rate 12.50 10.41 11.31 9.73 9.27 6.64 Prevalence of fever 1.64 2.03 1.86 2.24 2.28 1.08 Prevalence of diarrhea 2.47 1.87 2.20 2.32 2.84 1.32 Prevalence of acute respiratory infection 1.85 2.00 1.96 1.47 1.65 0.94 B. Fertility Total fertility rate 0.24 0.20 0.23 0.28 0.18 0.14 Adolescent fertility rate 15.00 14.00 16.00 14.00 11.00 8.00 C. Nutritional status Children: Moderate stunting 1.67 2.01 2.18 2.15 2.31 0.98 Severe stunting 1.78 1.66 2.10 2.45 1.95 0.95 Moderate underweight 1.75 1.76 1.87 2.65 1.87 0.92 Severe underweight 1.36 1.54 1.76 1.70 1.35 0.74 Mild anemia 1.69 2.17 2.19 2.14 2.71 1.01 Moderate anemia 2.19 2.47 2.90 2.56 3.18 1.33 Severe anemia 1.43 1.40 1.28 1.25 1.04 0.72 Women: Malnutrition 2.16 1.89 1.77 2.05 1.84 0.94 Mild anemia 2.05 2.03 2.05 2.17 1.78 1.02 Moderate anemia 1.46 1.73 1.42 1.55 1.38 0.80 Severe anemia 0.41 0.28 0.24 0.44 0.46 0.18 D. Female circumcision Prevalence of circumcision: Girls na na na na na na Women na na na na na na Prevalence of occlusion: Girls na na na na na na Women na na na na na na E. Sexually transmitted disease Prevalence of genital discharge: Women 0.49 0.44 0.57 0.30 0.22 0.23 Men na na na na na na Prevalence of genital ulcer: Women 0.30 0.27 0.28 0.34 0.27 0.15 Men na na na na na na - 53 - Madagascar 1997 - STANDARD ERRORS OF QUINTILE ESTIMATES FOR TOTAL POPULATION Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Avg. A. Childhood immunization BCG coverage 3.68 4.20 3.55 2.87 2.09 1.99 Measles coverage 3.01 3.41 3.54 3.74 2.95 1.93 DPT coverage 3.77 4.13 3.71 3.63 2.90 2.16 Full basic coverage 2.97 3.06 3.45 3.90 3.75 1.88 No basic coverage 3.19 3.68 3.45 1.78 1.54 1.67 Hepatitis B coverage na na na na na na Yellow fever coverage na na na na na na B. Treatment of childhood illnesses Treatment of fever: Medical treatment of fever 2.96 3.23 3.05 3.54 4.90 1.60 Treatment in a public facility 2.42 2.98 2.89 3.51 3.79 1.47 Treatment in a private facility 1.39 2.35 1.42 1.62 3.01 0.87 Treatment of acute respiratory infection (ARI): Medical treatment of ARI 3.29 4.38 4.49 4.57 6.10 2.20 Treatment in a public facility 3.07 3.90 4.35 4.67 6.51 2.20 Treatment in a private facility 1.72 3.06 2.12 1.76 4.84 1.10 Treatment of diarrhea: Use of oral rehydration therapy 3.96 4.18 4.56 3.63 2.31 2.17 Medical treatment of diarrhea 3.72 3.70 4.72 4.23 4.25 2.20 Treatment in a public facility 3.02 3.42 4.28 4.24 3.63 1.76 Treatment in a private facility 2.11 3.08 2.57 2.05 4.70 1.36 C. Antenatal and delivery care Antenatal care (ANC) visits: To a medically trained person 2.67 2.64 2.44 2.18 1.46 1.60 To a doctor 2.11 2.21 2.73 3.51 2.13 1.83 To a nurse or trained midwife 3.43 3.27 3.37 3.81 2.35 2.30 Multiple visits to a medically trained person 2.48 2.26 2.38 2.39 2.25 1.46 Antenatal care content: Tetanus toxoid 2.60 2.74 2.58 2.88 2.66 1.61 Prophylactic antimalarial treatment na na na na na na Iron supplementation na na na na na na Delivery attendance: By a medically trained person 3.08 2.83 3.02 2.95 2.01 2.23 By a doctor 1.68 1.61 2.22 3.63 2.41 1.70 By a nurse or trained midwife 2.87 2.48 3.04 3.32 2.67 2.08 In a public facility 2.99 2.71 2.79 3.28 3.03 1.93 In a private facility 0.49 0.27 0.55 1.14 0.96 0.46 At home 2.98 2.66 2.76 3.03 3.27 1.88 D. Contraceptive services Contraceptive prevalence: Women 0.72 0.99 1.48 1.75 1.49 0.82 Men na na na na na na Source of contraception - public sector: Women * 7.87 7.87 6.81 3.51 3.84 Men na na na na na na - 54 - Madagascar 1997 - STANDARD ERRORS OF QUINTILE ESTIMATES FOR TOTAL POPULATION Part II: Intermediate Determinants of HNP Status - HNP SERVICE USE (Cont.) Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Avg. D. Contraceptive services (cont.) Source of contraception - private sector: Women * 7.53 7.12 5.56 3.32 3.11 Men na na na na na na E. Treatment of adult illnesses Treatment of genital discharge, ulcer, sore: Women 6.34 10.42 9.06 10.57 7.56 4.23 Men na na na na na na Treatment of genital discharge, ulcer, sore in public facilities: Women 0.49 0.44 0.57 0.30 0.22 0.23 Men na na na na na na Voluntary counseling and testing for HIV/AIDS: Women na na na na na na Men na na na na na na - 55 - Madagascar 1997 - STANDARD ERRORS OF QUINTILE ESTIMATES FOR TOTAL POPULATION Part III: Intermediate Determinants of HNP Status - INDIVIDUAL AND HOUSEHOLD BEHAVIOR Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Avg. A. Hygienic practices Disposal of children's stools: Sanitary disposal na na na na na na Handwashing: Wash hands prior to preparing food na na na na na na Handwashing facilities in household na na na na na na B. Bednet ownership and use Bednet ownership: Bednet ownership na na na na na na Treated bednet ownership na na na na na na Bednet use: By children na na na na na na By pregnant women na na na na na na C. Breastfeeding Exclusive breastfeeding 4.11 5.46 4.78 7.00 6.83 2.46 Timely complementary feeding 3.66 4.14 3.31 4.02 3.17 1.63 Bottle-feeding 0.68 0.61 0.67 1.38 2.96 0.52 D. Micronutrient consumption Iodized salt: Availability of iodized salt in household 2.49 2.73 2.54 1.46 1.31 1.43 Vitamin A: Children 0.69 0.77 0.54 0.74 0.88 0.36 Women na na na na na na E. Tobacco and alcohol use Tobacco: Women na na na na na na Men na na na na na na Alcohol: Women na na na na na na Men na na na na na na F. Sexual practices Non-regular sexual partnerships: Women na na na na na na Men na na na na na na Condom usage with non-regular partner: Women na na na na na na Men na na na na na na G. Domestic violence Ever experienced violence na na na na na na Experienced violence in past year na na na na na na - 56 - Madagascar 1997 - STANDARD ERRORS OF QUINTILE ESTIMATES FOR TOTAL POPULATION Part IV: UNDERLYING DETERMINANTS OF HNP STATUS Indicator Wealth Quintiles Pop. Low 2nd 3rd 4th High Avg. A. Education School completion: Women 1.19 1.08 1.28 1.92 1.71 1.24 Men 1.23 1.30 1.29 2.19 2.58 1.34 School participation: Girls 3.59 3.40 3.35 2.86 1.36 2.03 Boys 3.21 3.92 3.07 2.77 1.39 2.17 B. Exposure to mass media Newspaper readership: Women 1.39 1.23 1.53 1.58 1.68 1.08 Men 0.00 0.00 0.00 0.00 0.00 0.00 Radio listenership: Women 0.75 1.49 2.00 2.58 1.28 1.36 Men 0.00 0.00 0.00 0.00 0.00 0.00 Television viewership: Women 0.36 0.36 0.36 0.94 2.30 0.97 Men 0.00 0.00 0.00 0.00 0.00 0.00 C. Knowledge and attitudes about HIV/AIDS Knowledge about sexual transmission of HIV/AIDS: Women 2.01 1.80 2.00 2.44 1.47 1.82 Men na na na na na na Knowledge about mother-to-child transmission of HIV/AIDS: Women 2.54 2.32 2.34 2.41 0.90 1.57 Men na na na na na na Attitudes toward HIV/AIDS: Women na na na na na na Men na na na na na na D. Status of women Household decisionmaking: Can seek own health care na na na na na na Can seek children's health care na na na na na na Can make daily household purchases na na na na na na Can make large household purchases na na na na na na Can make meal-related decisions na na na na na na Freedom of movement: Can travel to visit family, relatives na na na na na na Other decisionmaking, attitudes: Can decide how to spend own money 1.55 1.67 1.82 1.99 1.93 1.06 Can decide whether to have sex na na na na na na Justifies domestic violence na na na na na na E. Orphanhood Paternal orphan prevalence 0.68 0.58 0.78 0.60 0.68 0.29 Maternal orphan prevalence 0.41 0.67 0.52 0.66 0.35 0.24 Double orphan prevalence 0.10 0.22 0.13 0.13 0.13 0.06 - 57 - Madagascar 1997 - ASSET DISTRIBUTION AND WEIGHTS (FACTOR SCORES) Asset Variable Unweighted Wealth Quintiles Factor Low 2nd 3rd 4th High Avg. Score Mean Std. Percentage of Population Devia- tion Has electricity 0.154 0.361 0.0% 0.0% 0.1% 0.4% 55.1% 11.1% 0.15080 Has radio 0.417 0.493 0.9% 33.3% 31.4% 56.5% 86.4% 41.7% 0.10360 Has television 0.094 0.291 0.0% 0.0% 0.0% 0.2% 33.8% 6.8% 0.13631 Has refrigerator 0.020 0.138 0.0% 0.0% 0.0% 0.0% 6.3% 1.3% 0.08574 Has bicycle 0.052 0.223 0.2% 1.7% 3.6% 8.0% 17.4% 6.2% 0.04035 Has motorcycle 0.006 0.075 0.0% 0.0% 0.0% 0.0% 2.9% 0.6% 0.04297 Has car 0.016 0.125 0.0% 0.0% 0.0% 0.0% 5.2% 1.0% 0.07643 Has telephone 0.009 0.096 0.0% 0.0% 0.0% 0.0% 2.9% 0.6% 0.06839 Works own or family's agric. land 0.500 0.500 98.6% 70.6% 64.5% 56.9% 12.0% 60.5% -0.09713 Uses piped drinking water in residence 0.059 0.236 0.0% 0.0% 0.0% 0.7% 19.4% 4.0% 0.11394 Uses a well in residence 0.028 0.166 0.0% 0.7% 2.3% 6.4% 8.9% 3.7% 0.01202 Uses water from an open well, hole, cesspool in residence 0.035 0.183 0.0% 1.8% 4.6% 7.9% 7.5% 4.3% 0.00524 Uses river, canal or surface water for drinking 0.542 0.498 92.2% 73.9% 64.2% 59.3% 6.7% 59.3% -0.10437 Uses rain for drinking water 0.000 0.017 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% -0.00163 Uses a public faucet (piped) 0.149 0.356 0.0% 0.4% 4.3% 10.6% 40.7% 11.2% 0.08025 Uses water from open well, hole, cesspool outside residence 0.095 0.293 6.4% 17.6% 15.4% 5.3% 2.4% 9.4% -0.02915 Uses piped water outside residence 0.018 0.134 0.0% 0.0% 1.4% 2.3% 5.7% 1.9% 0.01861 Uses borehole with pump in residence 0.011 0.103 0.0% 0.0% 0.4% 1.1% 4.3% 1.2% 0.01642 Uses borehole with pump outside of residence 0.018 0.133 1.3% 3.0% 2.3% 0.8% 1.0% 1.7% -0.01110 Uses well outside of residence 0.038 0.191 0.0% 1.7% 3.7% 5.4% 3.1% 2.8% -0.00205 Uses water from a tanker truck 0.005 0.071 0.2% 0.9% 1.3% 0.2% 0.2% 0.6% -0.00265 Uses other source of drinking water 0.001 0.026 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% -0.00177 Uses own flush toilet 0.037 0.189 0.0% 0.0% 0.0% 0.3% 11.0% 2.3% 0.09723 Uses bush, field as latrine 0.587 0.492 100.0% 99.2% 67.2% 30.5% 5.4% 60.5% -0.12692 Uses a traditional pit toilet 0.334 0.472 0.0% 0.0% 26.8% 61.7% 75.7% 32.8% 0.08954 Uses a VIP latrine 0.042 0.200 0.0% 0.8% 5.9% 7.5% 7.9% 4.4% 0.00948 Uses other type of latrine 0.000 0.017 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.00208 Has dirt, sand, dung as principal floor in dwelling 0.071 0.258 0.0% 6.3% 10.8% 13.7% 5.8% 7.3% -0.00042 Has wood plank, principal floor in dwelling 0.140 0.347 0.0% 0.3% 12.5% 25.6% 19.4% 11.5% 0.03547 Has cement principal floor 0.142 0.349 0.0% 0.0% 1.4% 11.4% 50.7% 12.7% 0.08999 Has other type of flooring 0.000 0.017 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% -0.00141 Has parquet or polished wood floors 0.048 0.214 0.0% 0.0% 0.2% 2.4% 16.7% 3.9% 0.07280 Has tiles for main flooring material 0.003 0.055 0.0% 0.0% 0.0% 0.0% 1.0% 0.2% 0.03367 Has palm or bamboo for floor materials 0.591 0.492 100.0% 93.4% 75.1% 46.4% 4.8% 64.0% -0.12683 Has carpeted flooring 0.003 0.058 0.0% 0.0% 0.0% 0.3% 1.4% 0.3% 0.01799 Number of members per sleeping room 3.725 2.123 5.8 4.8 4.8 4.0 3.4 4.6 -0.05370 - 59 - PART IV. ANNEXES A. SOURCES OF ADDITIONAL INFORMATION B. USE OF INFORMATION FROM THIS REPORT TO MONITOR THE ECONOMIC STATUS OF PEOPLE SERVED BY HNP PROGRAMS C. COUNTRIES COVERED BY THE HNP- POVERTY REPORT PROJECT ANNEX A. SOURCES OF ADDITIONAL INFORMATION ADDITIONAL INEQUALITY DATA World Bank HNP/Poverty Country Report Website: http://www.worldbank.org/hnp/povertyandhealth/countrydata. This World Bank website provides the full texts and tables for all fifty-six countries covered by the HNP/Poverty Country Report Project. (A list of the countries covered appears in annex C, at the end of this report.) Also available at the site are summary tables, organized by indicator, designed to facilitate cross-country comparisons in inequality with respect to particular indicators. DHS Country Reports: http://www.measuredhs.com/countries. All DHS final country reports produced since 2003 include quintile-specific tabulations in approximately 50-100 of the reports' HNP indicator tables. The tables deal with some of the indicators covered in this volume, and with many that are not. UNICEF Multi-Indicator Cluster Survey Website: http://www.childinfo.org/MICS2/natlMICSrepz/MICSnatrep.htm. The UNICEF Multi-Indicator Cluster Survey (MICS) project is generally similar to the DHS program, but covers a somewhat different set of countries and indicators. The "standard tables" section for each country listed at the MICS website provides wealth- based, quintile-specific information in around 40-45 of the tables related to hnp, education, and child labor. In deriving these quintile-specific estimates, the MICS investigators have employed a wealth index similar to the one used here. World Health Organization World Health Survey Website: http://www.who.int/healthinfo/survey/en/index.html. The World Health Organization's World Health Survey (WHS) includes such issues as self-assessed adult health status; coverage of interventions against adult chronic diseases and against maternal and child health problems; household health expenditures; insurance coverage; and health system responsiveness. Approximately seventy countries ­ developed as well as developing ­ have been covered thus far. Household wealth information has been collected and used to prepare quintile-specific estimates for many of the indicators appearing in the reports on these countries. - 63 - METHODS AND RESOURCES FOR FURTHER INEQUALITY ANALYSIS Shea Oscar Rutstein and Kiersten Johnson, The DHS Wealth Index, DHS Comparative Reports No. 6 (Calverton, Maryland, USA: ORC Macro, August 2004) (Available at: http://www.measuredhs.com/pubs/pub_details.cfm?ID =470&srchTp=type). This DHS publication, by two of the co-authors of the current report, describes in detail the construction of the wealth index that underlies the data presented in the basic tables. Deon Filmer and Lant H. Pritchett, "Estimating Wealth Effects without Expenditure Data ­ or Tears: An Application to Education Enrollments in States in India," Demography 38, no.1 (February 2001): 115-132. This seminal piece gave birth to the wealth index procedure used in the current volume. It also includes three of the previously-cited country case studies demonstrating the close relationship between results produced using wealth and those based on consumption as an indicator of household economic status. Owen O'Donnell, Eddy van Doorslaer, Adam Wagstaff, and Magnus Lindelow. Quantitative Techniques for Health Equity Analysis. Washington D.C.: The World Bank, forthcoming. Among the topics covered in this comprehensive overview of available quantitative techniques are the measurement of living standards using a wealth index and other approaches (chapter 6) and the concentration index as a measure of inequality (chapter 8). DHS Country Data Sets: http://www.measuredhs.com/accesssurveys/search. The data sets for all DHS surveys undertaken since 2003 include two pieces of information for each household that are designed to help investigators prepare quintile- specific tabulations for any indicator. These are: 1) the household wealth score; and 2) the economic quintile to which individuals in the household belong. Any tabulation using these pieces of information will be comparable to the figures appearing here. - 64 - ANNEX B. USE OF INFORMATION FROM THIS REPORT TO MONITOR THE ECONOMIC STATUS OF PEOPLE SERVED BY HNP PROGRAMS The wealth or asset approach employed in this report can be used to monitor the economic status of people served by health, nutrition, and population (hnp) programs in two ways. The first, simpler way is suitable for monitoring nationwide, facility-based programs. A second, fuller version can also be employed for other types of programs, such as initiatives undertaken only in some parts of a country, or activities like mass education or outreach programs that do not operate through facilities. BASIC MONITORING OF NATIONWIDE FACILITY-BASED PROGRAMS The economic distribution of patients in a nationwide, facility-based program (say, a network of rural health posts, antenatal care clinics, emergency obstetrical facilities, or hospitals) can be determined through an exit survey of facility patients, using the wealth questionnaire and the set of quintile cut-off points that immediately follow this text, and which have been created using the information presented in part III.C. The questionnaire can be employed to measure the economic status of any individual responding to the questions on it. The set of cut-off points can serve to compare the distribution of the respondents' economic status with that of the nationally- representative sample of people interviewed by the DHS survey on which the present report is based. The first step is to use the questionnaire in interviewing an adequately-large sample of patients attending the facility-based services of interest. The wealth score for each patient can then be calculated by multiplying the response to each question by the item scores also provided on the questionnaire, and summing the results. After this has been done, the quintile cut-off points can be used to place each individual in the economic quintile to which (s)he belongs. The number of patients and percentage of total patients in each quintile can then be calculated. Since each quintile defined by the cut-off points contains 20 percent of the individuals in the nationally-representative DHS sample, the patients belonging to any such quintile containing significantly more (or less) than 20 percent of the total are over- (under-) represented relative to the national population. When the percentage of patients in each of the five quintiles is viewed as a whole, the result is a frequency distribution that indicates the spread of service beneficiaries across economic classes of individuals.21 For example: 21That is, across economic classes of all individuals in the sample population. Estimates pertaining to quintiles of only those individuals needing services require adjusting the results of the procedure described here through application of the relevant quintile-specific, sample-size figures presented in part III.A. - 65 - · A service that favors the poorest people would have substantially more than 20 percent of its patients in each of the lowest one or two economic quintiles; considerably less than 20 percent of its patients in each of the highest quintiles. · A service that reaches all economic classes equally would draw roughly the same proportion of total patients from the lower and upper quintiles. · A service that favors the least poor population groups would have well over 20 percent of its patients in each of the highest one or two quintiles, considerably less than 20 percent of its patients in each of the lowest quintiles. FULLER MONITORING OF FACILITY-BASED AND OF OTHER PROGRAMS While capable of providing far more information than currently exists about the distribution of a program's beneficiaries, the approach just described has important limitations. For example, it cannot deal with the many important types of health programs that do not deliver services primarily through facilities ­ mass media health education, household visits by health workers, and many social marketing initiatives, for example. It is also limited in its ability to assess programs working only with certain areas within a country: it can compare the economic status of the programs' beneficiaries with that of the national population, but not with that of the specific sub-national areas where the programs are active. Further, it focuses primarily on only one of the two important dimensions of monitoring the distribution of program beneficiaries: that is, incidence or focus ­ the percentage of program benefits that flow to the poor. It cannot deal nearly so well with the second dimension, which concerns coverage, or the percentage of the poor that the program reaches. These limitations can be overcome by a modified version of the approach described above that relies on a household- rather than facility-based survey. A household survey can generate a set of data containing the full range of information needed to produce an equity assessment by collecting two types of information: first, about the household's wealth or assets, using the questions in the left-hand column of the attached questionnaire;22 and second, about the household members' use of or exposure to the services provided by the program(s) of interest. The collected data can be analyzed in either (or both) of two ways, depending on the type of information desired: · One way would be to use only data from the household survey. The procedure would be analogous to that for a DHS survey employed in this report: ­ Asset information from the survey-generated data set would be used as the basis for the construction of a wealth index, weighting the individual items using some method like principal components analysis. ­ The individuals in the sample would be ranked in order of the index values for their households, then divided into groups like quintiles. ­ The coverage rates in each quintile for the service of interest would be calculated. 22Or, if one is willing to forego the benefits of the second analytical approach described below, using any of several other asset questionnaires that exist. Examples include the INDEPTH health equity survey tool (available at: www.indepth-network.org/core_documents/indepthtools.htm) or the model questionnaire developed by M. Mahood Khan and David Hotchkiss of the PHR Plus project (which can be found at: www.phrplus.org). - 66 - · A second approach would be to use the weights for each item appearing on the attached questionnaire in determining the wealth of each individual, instead of calculating the weights from the new household data set. Once the individuals' wealth is determined, the individuals would be ranked, divided into quintiles, and the coverage rate in each quintile would be calculated. In the case of programs undertaken in only one region of the country, it would provide a comparison of the economic status of the people served with that of the entire country rather than of only the region where the programs are active;23 in the case of nationwide as well as regional programs, it would permit a comparison with the other service programs covered in this report. 23This additional perspective could be particularly helpful in an assessment of a program seeking to reach the poor by focusing on especially backward districts. A report presenting only a finding that the program was reaching the better- off people in those districts could produce an impression that it had failed to reach its intended beneficiaries. But a comparison between the economic status of the program's beneficiaries with that of the national population might well reveal that most of the beneficiaries were poor by national standards and that the program was thus considerably more successful than otherwise thought. - 67 - Madagascar 1997 - ASSET QUESTIONNAIRE Question Score If Score If Item "Yes" "No" Score 1. In your household, is/are there? Electricity 0.35405 -0.06422 One or more radios 0.12243 -0.08765 One or more televisions 0.42387 -0.04383 One or more refrigerators 0.60761 -0.01210 One or more bicycles 0.17152 -0.00949 One or more motorcycles 0.56660 -0.00326 One or more cars 0.60131 -0.00971 One or more telephones 0.70954 -0.00659 2. Do the members of your household work their own or family's agricultural land? -0.09717 0.09708 3. What is the principal source of drinking water for your household? Piped water in residence 0.45393 -0.02860 Piped water in public faucet 0.02767 -0.00099 Piped water in yard, plot 0.13587 -0.00255 Well in residence 0.07041 -0.00205 Well in yard, plot -0.01033 0.00041 Open well, hole, cesspool in residence 0.02767 -0.00099 Open well, hole, cesspool in yard, plot -0.08998 0.00944 Borehole with pump in residence 0.15757 -0.00171 Borehole with pump in yard, plot -0.08166 0.00151 River, canal, surface water -0.09593 0.11354 Rainwater -0.09762 0.00003 Tanker truck -0.03729 0.00019 Other -0.06691 0.00005 4. What is the principal type of toilet facility used by your household? Private flush toilet 0.49630 -0.01904 VIP latrine 0.04538 -0.00198 Traditional pit toilet 0.12636 -0.06344 Bush, field as latrine -0.10656 0.15116 Other 0.12450 -0.00003 5. What is the principal material used for the floors in your household? Dirt, sand, dung -0.00150 0.00012 Cement 0.22147 -0.03656 Wood plank 0.08807 -0.01429 Parquet, polished wood 0.32332 -0.01639 Carpet 0.31049 -0.00104 Tiles 0.60695 -0.00187 Palm, bamboo -0.10548 0.15250 Other -0.08460 0.00002 6. How many people are there for each sleeping room in your household? # people- 3.72× -0.054 2.12 Total Household Asset Score (sum of individual item scores) - 69 - Madagascar 1997 - QUINTILE CUT-OFF POINTS Asset Index Value Wealth Quintile Bottom Cut-Off Top Cut-Off Low Low -0.81093 Second -0.81093 -0.60068 Third -0.60068 -0.26565 Fourth -0.26565 0.37056 High 0.37056 High - 70 - ANNEX C. COUNTRIES COVERED BY THE HNP - POVERTY REPORT PROJECT* East Asia and Pacific Sub-Saharan Africa Cambodia 2000 Benin 1996, 2001 Indonesia 1997, 2002-03 Burkina Faso 1992-3, 1998-9, 2003 Philippines 1998, 2003 Cameroon 1991, 1998, 2004 Vietnam 1997, 2002 Central African Rep. 1994-95 Chad 1996-97, 2004 Europe and Central Asia Comoros 1996 Armenia 2000 Cote d'Ivoire 1994 Kazakhstan 1995, 1999 Eritrea 1995 Kyrgyz Rep. 1997 Ethiopia 2000 Turkey 1993, 1998 Gabon 2000 Turkmenistan 2000 Ghana 1993, 1998, 2003 Uzbekistan 1996 Guinea 1999 Kenya 1993, 1998, 2003 Latin America and the Caribbean Madagascar 1997 Bolivia 1998, 2003 Malawi 1992, 2000 Brazil 1996 Mali 1995-96, 2001 Colombia 1995, 2000, 2005 Mauritania 2000-01 Dominican Rep. 1996, 2002 Mozambique 1997, 2003 Guatemala 1995, 1998-99 Namibia 1992, 2000 Haiti 1994-95, 2000 Niger 1998 Nicaragua 1997- 98, 2001 Nigeria 1990, 2003 Paraguay 1990 Rwanda 2000 Peru 1996, 2000 Senegal 1997 South Africa 1998 Middle East and North Africa Tanzania 1996, 1999, 2004 Egypt 1995, 2000 Togo 1998 Jordan 1997 Uganda 1995, 2000-01 Morocco 1992, 2003-04 Zambia 1996, 2001-02 Yemen 1997 Zimbabwe 1994, 1999 South Asia * Note: electronic versions of reports for all countries Bangladesh 1996-97, 1999-2000, 2004 are currently available at: www.worldbank.org/ povertyandhealth/countrydata. While supplies last, India 1992-93, 1998-99 paper copies may be obtained at no charge by sending Nepal 1996, 2001 a request to the World Bank's health advisory service: healthpop@worldbank.org. Pakistan 1990-91 - 71 - about this series... This series is produced by the Health, Nutrition, and Population (HNP) Family of the World Bank's Human Development Network. The find- ings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations or to the members of its Board of Executive Directors or the countries they represent. For free copies of the papers in this series please contact the HNP Advisory Service (healthpop@worldbank.org, tel 1-202-473-2256, fax 1-202-522-3234). tHe world bank 1818 H Street, NW Washington, DC USA 20433 Telephone: 202 473 1000 Facsimile: 202 477 6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org