Issuu

Inequalities in women’s and girls’ health opportunities and outcomes: A report from sub-Saharan Africa

©Pau Fabregat

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes: A report from sub-Saharan Africa

December 2016

This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

Contents Authorship and acknowledgements ................................................................................ 8 Abbreviations....................................................................................................................... 9

Executive summary..................................................................................................10

Chapter 1. Introduction...............................................................................................................11 1.1 Context setting....................................................................................................................................12 1.2 Inequality of opportunity ..............................................................................................................15 References.............................................................................................................................................18

Chapter 2. Methodology..........................................................................................................19 2.1 2.2 2.3 2.4 2.5 2.6 2.7

The Human Opportunity Index................................................................................................... 20 Shapley decomposition.................................................................................................................21 Study population................................................................................................................................21 Data sources and country inclusion criteria............................................................................22 Selecting and defining opportunities.....................................................................................24 Defining a set of circumstances ................................................................................................. 26 Data management............................................................................................................................ 29 References........................................................................................................................................... 30

Chapter 3. What is the state of health inequalities of women of reproductive age?........31 3.1

HOIs by country and multi-country pooled averages............................................... 32 3.1.1 Women of reproductive age (15-49 years old).............................................................32 3.1.2 Pregnant women...................................................................................................................... 38 3.1.3 Older adolescent girls (15-19 years old).......................................................................... 50

3.2 Comparing HOIs among groups of countries.............................................................. 54 3.2.1 Comparing African regions.................................................................................................. 54 3.2.2 Comparing SSA countries with different HIV prevalence......................................57 Key messages....................................................................................................................................... 59 References..............................................................................................................................................60

A report from sub-Saharan Africa

Chapter 4. What explains inequalities in health for women of reproductive age?...........63 4.1 Explaining inequality at country level............................................................................ 64 4.1.1 Women of reproductive age (15-49 years old) and pregnant women.............. 66 4.1.2 Older adolescent girls (15-19 years old)............................................................................71 4.1.3 HIV-related indicators.............................................................................................................73 4.2 Explaining inequalities across countries: a multi-country pooled analysis............75 4.2.1 Women of reproductive age (15-49 years old) and pregnant women..............77 4.2.2 Older adolescent girls (15-19 years old)..........................................................................79 4.3 Adolescent girls and marital status: the major source of inequalities.................... 80 4.3.1 Country level analysis results............................................................................................... 80 4.3.2 Multi-country pooled analysis results.............................................................................82 Key messages....................................................................................................................................... 85 References..............................................................................................................................................86

Chapter 5. Conclusions............................................................................................................. 87

5.1 Conclusions.......................................................................................................................... 88 5.2 Policy options for adolescent girls................................................................................ 93 5.3 Final considerations........................................................................................................... 94 Key messages....................................................................................................................................... 97 References..............................................................................................................................................98

Inequalities in women’s and girls’ health opportunities and outcomes

Appendix A. Additional tables with all results..........................................................................99

Boxes 1 Progress from MDGs to SDGs.........................................................................................................13 2 Opportunities.......................................................................................................................................16 3 How to interpret the HOI................................................................................................................33 4 Country cases: Ethiopia and Sierra Leone............................................................................... 44 5 D-index: country cases................................................................................................................... 66 6 Women of reproductive age and pregnant women’s opportunities: country cases...................................................................................................................................... 69 7 Older adolescent girls’ opportunities: country cases.......................................................72 8 The role of religion in women’s health indicators’ inequalities........................................77 9 Strengths and limitations of the study....................................................................... 92 10 Data gaps identified...................................................................................................96

Figures 2.1 Map of the countries included in the analysis........................................................................23 2.2 Summary of opportunities, circumstances and groups of women included in the analysis...................................................................................................................................... 29 3.1 HOI for access to reproductive health .................................................................................... 34 3.2 Comparison between “met need for family planning” for older adolescent girls (15-19) and women of reproductive age (20-49)..........................................................35 3.3 HOI for general women’s health .................................................................................................37 3.4 HOI for maternity care .....................................................................................................................41 3.5 HOI for “delivery attended by skilled personnel” by place of delivery .................. 42 3.6 Composite HOI for maternal care............................................................................................... 43 3.7 Correlations between the D-index of the maternity care package and IMR or MMR.................................................................................................................................................. 45 3.8 HOI for malaria and HIV indicators.............................................................................................. 46 3.9 HOI for infant care.............................................................................................................................. 48 3.10 Correlation between HIV prevalence and the HOI for “six months of exclusive breastfeeding”.................................................................................................................................. 49 3.11 HOI for access to reproductive health and education for older adolescent girls...........................................................................................................................................................52 3.12 Average HOIs for women of reproductive age and pregnant women’s opportunities by African region.....................................................................................................55 3.13 Average HOIs for adolescent girls’ indicators by African region..................................57 3.14 Average HOIs by HIV prevalence regions............................................................................. 58 4.1 Average D-index by opportunity (unweighted)................................................................. 65 4.2 Women of reproductive age and pregnant women: circumstances’ contributions to the D-index..........................................................................................................67

A report from sub-Saharan Africa

4.3 Older adolescent girls’ opportunities: circumstances’ contributions to the D-index....................................................................................................................................................71 4.4 HIV prevalence country groups: circumstances’ contributions to the D-index.......74 4.5 Average D-index by opportunity (multi-country pooled analysis)..............................76 4.6 Multi-country pooled analysis for women of reproductive age and pregnant women: circumstances’ contributions to the D-index........................................................78 4.7 Multi-country pooled analysis for older adolescent girls: circumstances’ contributions to the D-index..........................................................................................................79 4.8 Country level analysis – Older adolescent girls’ opportunities by marital status: circumstances’ contributions to the D-index...........................................................................81 4.9 Multi-country pooled analysis – Older adolescent girls’ opportunities by marital status: HOI and D-index by opportunity................................................................... 83 4.10 Multi-country pooled analysis – Older adolescents girls’ opportunities by marital status: circumstances’ contributions to the D-index............................................. 84

Tables 2.1 List of countries and DHS surveys................................................................................................22 2.2 List of opportunities and baseline population for whom they have been analysed.................................................................................................................................................24 2.3 List of circumstances.........................................................................................................................27 5.1 Country level average HOIs and multi-country pooled HOIs........................................ 88 5.2 Opportunities ranked by inequality level (multi-country pooled analysis).............. 89 5.3 Three main contributors to inequality for each opportunity and subgroup of women............................................................................................................................................. 90 A1 Levels: Currently attending school (older adolescent girls) .........................................100 A2 Levels: Having never been pregnant (older adolescent girls)...................................... 101 A3 Levels: Met need for family planning (older adolescent girls)..................................... 102 A4 Levels: Not having anaemia ......................................................................................................... 103 A5 Levels: Having the recommended BMI ..................................................................................104 A6 Levels: Met need for family planning ...................................................................................... 105 A7 Levels: Knowledge of a place where to get an HIV test ................................................ 106 A8 Levels: Four antenatal care visits ................................................................................................ 107 A9 Levels: Delivery attended by skilled personnel................................................................. 108 A10 Levels: Delivery attended by skilled personnel, by place of delivery .................... 109 A11 Levels: Postnatal checkup ............................................................................................................. 110 A12 Levels: Maternity care package....................................................................................................111 A13 Levels: Malaria prophylaxis during pregnancy ....................................................................112 A14 Levels: HIV test offered during pregnancy.............................................................................113 A15 Levels: Infant checkup within two months after delivery ................................................ 114 A16 Levels: Six months of exclusive breastfeeding.................................................................... 115 A17 Shapley decomposition: Currently attending school (older adolescent girls) ..... 116 A18 Shapley decomposition: Having never been pregnant (older adolescent girls)...................................................................................................................117

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

A19 Shapley decomposition: Met need for family planning (older adolescent girls).................................................................................................................. 118 A20 Shapley decomposition: Not having anaemia .................................................................... 119 A21 Shapley decomposition: Having the recommended BMI ............................................ 120 A22 Shapley decomposition: Met need for family planning ...................................................121 A23 Shapley decomposition: Knowledge of a place where to get an HIV test .......... 122 A24 Shapley decomposition: Four antenatal care visits ........................................................... 123 A25 Shapley decomposition: Delivery attended by skilled personnel............................ 124 A26 Shapley decomposition: Postnatal checkup....................................................................... 125 A27 Shapley decomposition: Maternity care package ........................................................... 126 A28 Shapley decomposition: Malaria prophylaxis during pregnancy.............................. 127 A29 Shapley decomposition: HIV test offered during pregnancy .................................... 128 A30 Shapley decomposition: Infant checkup within two months after delivery .......... 129 A31 Shapley decomposition: Six months of exclusive breastfeeding ............................. 130 A32 HOI comparisons among African regions...............................................................................131 A33 HOI comparisons between HIV prevalence regions...................................................... 132 A34 Levels and Shapley decompositions for the multi-country pooled analysis: women of reproductive age and pregnant women........................................................ 133 A35 Levels and Shapley decompositions for the multi-country pooled analysis: older adolescent girls................................................................................................................... 134 A36 Levels and Shapley decompositions for the multi-country pooled analysis: older adolescent girls by marital status.................................................................................. 135 A37 Country data....................................................................................................................................... 136 A38 Circumstancesâ€™ variables codification..................................................................................... 138

A report from sub-Saharan Africa

Authorship and acknowledgements This report was conducted as a joint project between the Barcelona Institute for Global Health (ISGlobal) and the Poverty and Equity Global Practice of the World Bank Group (WBG). The report is authored by Clara Pons Duran (Junior Researcher, ISGlobal), Andrew Dabalen and Ambar Narayan (Lead Economists, Poverty and Equity Global Practice, WBG), Anna Lucas and Clara Menéndez (Coordinator and Director of the Maternal, Child and Reproductive Health Initiative, ISGlobal, respectively). Clara Pons Duran developed the concept and led the quantitative analysis. Clara Pons Duran and Anna Lucas led the writing of the first draft. All authors drafted sections of the paper and provided input to the overall direction and content. The authors would like to acknowledge Carlos Felipe Balcázar (WBG), Llorenç Quintó (ISGlobal) and Elisa Sicuri (ISGlobal and Imperial College of London) for their instrumental input and suggestions throughout the drafting of the report. We are grateful for the support of Gonzalo Fanjul and Rafa Vilasanjuan, Director, Policy and Global Development, ISGlobal, and Ana Revenga, Director, Poverty and Equity Global Practice, WBG.

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

Abbreviations

BMI Body Mass Index Congo DR Democratic Republic of Congo Congo/Congo Rep. Congo Republic DHS Demographic Health Surveys D-index Dissimilarity Index HEP Health Extension Program HEWs Health Extension Workers HIV Human Immunodeficiency Virus HOI Human Opportunity Index IMR Infant Mortality Rate IPTp Intermittent Preventive Treatment of malaria in Pregnancy ISGlobal Barcelona Institute for Global Health LMICs Low and Middle Income Countries MDGs Millennium Development Goals MMR Maternal Mortality Ratio PCA Principal Component Analysis 2 R Correlation Coefficient RMNCAH Reproductive, Maternal, Newborn, Child and Adolescent Health SDGs Sustainable Development Goals SP Sulfadoxine-Pyrimethamine SSA Sub-Saharan Africa / Sub-Saharan African UHC Universal Health Coverage UN United Nations USAID United States Agency for International Development WBG World Bank Group WHO World Health Organization

A report from sub-Saharan Africa

Executive summary The Millennium Development Goal (MDG) 5 (to improve maternal health) was not achieved by the majority of the sub-Saharan African (SSA) countries. Women in SSA account for two thirds (201,000 deaths in 2015) of total maternal deaths globally. Despite progress for all essential maternal, newborn, and child health interventions between 1990-2015 in SSA, substantial disparity remains in coverage levels of interventions among and within countries. As a result, the most vulnerable women are not accessing essential health care services and undergo their pregnancies and childbirths outside the health system. The Sustainable Development Goals (SDGs) offer a new opportunity to address these inequalities. New tools and knowledge are needed to put equity at the heart of all strategies, a pre-requisite to achieve the common set of goals and targets set out by the SDGs, such as the reduction of the maternal mortality ratio (MMR) to less than 70 deaths per 100,000 live births by 2030. The report uses the most recent data available to analyse 15 opportunities for women of reproductive age (15-49), including two subgroups: pregnant women and older adolescent girls (15-19), within and across 29 SSA countries. The introduction of new metrics, such as the Human Opportunity Index (HOI), a composite indicator that determines how many opportunities are available (the coverage rate), and how equitably those opportunities are distributed across circumstance groups (sets of individuals with the same characteristics), allows new understanding of the constraints and opportunities to achieving equity in perinatal and reproductive health. The HOI allows simultaneous consideration of different health determinants to assess the magnitude and sources of inequality for different indicators and, thus identify which circumstances are generating the highest inequalities both at a country level and across the SSA region. Results reveal that overall reproductive and maternal health opportunities for women and girls are scarce – half of women and girls are not receiving the most essential interventions, and these are unequally distributed both at country level and across countries. Importantly, the most unevenly distributed opportunities were “maternity care package”, “delivery attended by skilled personnel” and “school attendance” while “not having anaemia” and “exclusive breastfeeding” are more equally available. Generally, wealth and related circumstances such as education and area of residence are the main sources of inequality for women of reproductive age. For the adolescent subgroup early marriage appears to be the main contributor to poor maternal and reproductive health opportunities. In the SSA context of low coverage and high inequalities, universal health coverage (UHC) strategies are the core mechanism to ensure effective and equitable provision of essential health care and reach “every woman, everywhere”. As governments and other members of the Reproductive, Maternal, Newborn, Child and Adolescent Health (RMNCAH) community mobilise efforts for the SDG period, the descriptions of inequality of opportunity in this report may be relevant for setting broad strategic priorities in public health policy, including those outside the health sector, and for identifying opportunities, the largest inequality gaps and the most underserved groups.

Â©Andalu Vila San Juan

Chapter 1. Introduction

A report from sub-Saharan Africa

1.1 Context setting Sub-Saharan Africa (SSA) is home to more than 500 million women who account for about half of the continent’s population and 14 percent of the female population worldwide1. About 47 percent of them are of reproductive age, defined as between 15 and 49 years. Despite the significant advancements that have been made on many of the Millennium Development Goals (MDGs) targets during the 19902015 period, a high proportion of SSA women face a wide range of problems and constraints in their daily lives, originating from their lower status than men in all spheres of life – i.e. family, community, labour market, religion or politics. This pervasive gender inequality in the region results in women being more likely to live in poverty and suffer ill health throughout their life cycles. As a consequence, African women carry an excessive share of the global burden of disease and death, particularly as it relates to maternal and reproductive health2. Despite progress during the MDGs period, in 2015, the maternal mortality ratio (MMR) in SSA was estimated at 546 maternal deaths per 100,000 live births, accounting for two-thirds (201,000) of the total maternal deaths worldwide (303,000)3. The fifth MDG set by the global development community in 2000 for improvement of maternal health, with the specific target of reducing MMR by 75 percent in each country between 1990 and 2015, has not been achieved by the majority of low and middle income countries (LMICs). In SSA, only four countries, Eritrea, Equatorial Guinea, Cabo Verde and Rwanda, reached the 75 percent MMR reduction, while others reduced the ratio by over 60 percent (e.g. Mozambique, Angola and Ethiopia)4. Despite an overall improvement in maternal survival and a 45 percent decline in MMR worldwide since 1990, SSA women continue to bear an unacceptable health burden4. Among the reasons for the reduction of maternal mortality in SSA are the investments made by some countries in quality maternity services accessible to the population2. However, as in other regions, in SSA, universal access of essential services and interventions is not a reality, and maternal health related services are not an exception3. As a result, millions of women are not accessing services, and undergo their pregnancies and childbirths outside the health system. Moreover, the second target of MDG5 – universal access to contraceptive methods – remains an important challenge for women of reproductive age in SSA. Despite the fact that the proportion of women of reproductive age using contraceptives more than doubled during the MDGs period, contraceptive use is still low and insufficient4. In SSA, one in four married or in-union women of reproductive age who wanted to delay or avoid pregnancy were not using any contraceptive method in 20152. Given current trends, the prevalence of unwanted pregnancies in SSA is predicted to further increase over the next few decades as a result of a combination of early sexual activity and low use of contraceptive methods2.

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

The recently agreed development agenda, the Sustainable Development Goals (SDGs), includes new and ambitious targets for maternal and reproductive health including ending preventable maternal mortality by reducing the global MMR to less than 70 per 100,000 live births by 2030 (target 3.1 of SDG3)4. Achieving universal coverage of essential maternal and reproductive health interventions should be the ultimate goal for all countries in the SDG era (SDG target 3.8). However, this is challenging in the short term given the low coverage rates in most SSA countries and the inequality gaps. Notably, one of the criticisms of the MDGs has been that the targets set in terms of average outcomes might have encouraged efforts in some countries to improve indicators by focusing on easier to reach segments of the population rather than those most in need6. As a result, large and avoidable disparities remain in coverage of health interventions for mothers, children and adolescents both across and within countries7,8. Inequity, unjust and avoidable inequalities, persists in maternal and reproductive health indicators and outcomes, posing a serious threat to the achievement of the agreed SDG targets.

Box 1. Progress from MDGs to SDGs MDGs

Baseline after MDGs

SDGs goals Targets By 2030

MDG4: Reduce child mortality MDG5: Improve maternal health

Global MMR was 216 deaths per 100,000 live births in 2015.

SDG3: Good health and wellbeing

Global under-five mortality rate was 43 deaths per 1,000 live births. The neonatal mortality rate was 19 deaths per 1,000 live births in 2015. Approximately three in four women of reproductive age who were married or in union satisfied their need for family planning by using modern contraceptive methods in 2015.

3.1 Reduce the global MMR to less than 70 per 100,000 live births. 3.2 End preventable deaths of newborns and children under five years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1,000 live births and under-five mortality to at least as low as 25 per 1,000 live births. 3.7 Ensure universal access to sexual and reproductive health-care services, including family planning, information and education, and the integration of reproductive health into national strategies and programmes. 3.8 Achieve universal health coverage (UHC), including financial risk protection, access to quality essential health-care services and access to safe, effective, quality and affordable essential medicines and vaccines for all.

box continues next page

A report from sub-Saharan Africa

Box 1. Progress from MDGs to SDGs (continued) MDG2: Achieve Universal Primary Education

Globally, two thirds of the adults (aged 15 and over) who were illiterate were women in 2013. One in ten girls was out of school, compared to one in 12 boys. Children from the poorest 20 percent of households are nearly four times more likely to be out of school than their richest peers. Out-of-school rates are also higher in rural areas.

SDG4: Quality education

4.6 Ensure that all youth and a substantial proportion of adults, both men and women, achieve literacy and numeracy. 4.7 Ensure that all learners acquire the knowledge and skills needed to promote sustainable development, including, among others, through education for sustainable development and sustainable lifestyles, human rights, gender equality, promotion of a culture of peace and non-violence, global citizenship and appreciation of cultural diversity and of cultureâ€™s contribution to sustainable development.

Completion rates for primary education in both developed and developing regions exceeded 90 percent in 2013. At the lower secondary level, the gap was at nearly 20 percentage points in 2013 (91 percent for developed regions and 72 percent for developing regions).

MDG3: Promote gender equality and empower women

In 63 countries, the legal age of marriage is lower for women than for men. Globally, the proportion of women aged between 20 and 24 who reported that they were married before their eighteenth birthday was 26 percent in 2015.

SDG5: Gender equality

Between 2007 and 2012, 56 of 94 countries with data available increased the income of the poorest 40 percent of the population more rapidly than its national average

5.1 End all forms of discrimination against all women and girls everywhere. 5.3 Eliminate all harmful practices, such as child, early and forced marriage and female genital mutilation. 5.6 Ensure universal access to sexual and reproductive health and reproductive rights as agreed in accordance with the Programme of Action of the International Conference on Population and Development and the Beijing Platform for Action and the outcome documents of their review conferences.

Twenty-one percent of girls and women aged between 15 and 49 experienced physical and/or sexual violence at the hands of an intimate partner in the previous 12 months. MDG1: Eradicate extreme poverty and hunger

4.5 Eliminate gender disparities in education and ensure equal access to all levels of education and vocational training for the vulnerable, including persons with disabilities, indigenous peoples and children in vulnerable situations.

SDG10: Reduced inequalities

10.2 Empower and promote the social, economic and political inclusion of all, irrespective of age, sex, disability, race, ethnicity, origin, religion, economic or other status. 10.3 Ensure equal opportunity and reduce inequalities of outcome, including by eliminating discriminatory laws, policies and practices and promoting appropriate legislation, policies and action in this regard.

Source: United Nations. SDGs. Sustainable Development Knowledge Platform [Internet]. 2016. Available from: https://sustainabledevelopment.un.org/sdgs

Inequalities in women’s and girls’ health opportunities and outcomes

Given this context, this report focuses on the analysis of maternal and reproductive health inequalities among women of reproductive age in SSA. It pays special attention to older adolescent girls – those between 15 and 19 years old –, a neglected population subgroup in terms of visibility and resources channelled to address their specific needs, calling for an in-depth examination of their health and reproductive issues (see Chapters 3 and 4).

1.2 Inequality of opportunity Access to maternal and reproductive health services is unequally distributed among women in SSA countries, as is typically the case when coverage of a service falls far short of universal access. Scarcity by its very nature produces inequality between those who have access (and better outcomes as a result) and those who do not, which is often manifested as systematic and persistent gaps between individuals belonging to different socio-economic groups. Large gaps exist in coverage and access to quality maternal health services between the poorest and richest households, and between rural and urban areas. In SSA only 56 percent of births are attended by skilled health personnel in rural areas, compared with 87 percent in urban areas4. When services are scarce, typically, an individual’s chances of accessing them are influenced by their circumstances, namely the economic and social attributes of the individual and the family. This in turn produces inequalities in access to services (and to outcomes linked to those services) between groups differentiated by characteristics such as geographic location, wealth status, education levels, family structure, depending on the country and the type of health service or outcome. These characteristics can be seen as the social determinants of health status, which act by influencing the physical environment (including the availability of services) and behavioural factors that matter for use of services or adoption of practices. In most societies there is broad consensus around the notion that granting access to a basic set of goods and services to every individual, regardless of the circumstances s/he was born into, is fundamental to building a just society and fostering economic and social development. However, in most LMICs, including those in SSA, the goal of universal and equal access to basic goods and services remains distant—a person’s circumstances still matter a great deal in determining his/her opportunities. Finally, a distinction between children and adults’ opportunities can be made since the opportunities of an adult could be “affected” by his/her own decisions (Box 2).

A report from sub-Saharan Africa

Box 2. Opportunities The World Bank Group (WBG) has published several human opportunity reports since 2009 to document unequal access to basic goods and services such as education, health services, safe water, sanitation and nutrition in different countries and regions around the world9. Opportunities in this context are understood as the minimum set of essential goods and services that enable individuals to realise their human potential. The concept of equality of opportunity, first formalised by the economist John Roemer in 1993 and 199810,11, requires that individuals’ opportunities are independent of their life circumstances. These circumstances are the characteristics that an individual is born into and has no influence over such as race, religion, gender, place of birth, or the wealth and education of one’s parents. Most of the previous WBG reports were focused on children’s opportunities to access basic goods and services in education, health and infrastructure12 – where individual effort and choice do not matter as these are considered irrelevant for children. Whilst most societies can agree on a set of basic goods and services that constitute a minimum level of opportunities for children, consensus around what could be considered opportunities for adults is less clear, because choices made by adults play some role in accessing basic services. Access to basic services, such as higher education or having a delivery attended by skilled personnel, is no doubt influenced by an individual’s own decisions, which is an argument against considering these as “opportunities” in the strict sense. However, there is a strong argument for going beyond this strict view and considering certain types of essential services or indicators of well-being as opportunities even for adults, and particularly for women. This is because the choices made by most women in LMICs – e.g. whether they should go to a hospital to deliver a baby, access pre-natal care or use family planning methods – are affected by external factors on which they have almost no influence. These include family, economic and social status, or location – circumstances that can effectively constrain the choices available to women in making these decisions. This argument is even more salient when it comes to health indicators such as anaemia and malnutrition, which are even more likely to be influenced by constraints imposed by life circumstances. As mentioned earlier, because women are a particularly vulnerable group in many situations, it is even harder for them to exercise free choice to access opportunities that are essential for their well-being.

A major focus of this report is the extent of inequalities associated with life circumstances for SSA women in reproductive and maternal health that they have no control over. Following the rationale described above (see Box 2), opportunities here will be interpreted as a “desirable situation” for a woman in terms of her reproductive and maternal health status. Thus, opportunities will refer to both health outcomes (such as being well-nourished), and the use and knowledge of essential maternal and reproductive health services (such as antenatal care, deliveries attended by skilled personnel, and family planning). This is clearly an expansive view of opportunities as it ignores the role of personal effort or decision-making by a woman in accessing these services or adopting healthy practices

Inequalities in women’s and girls’ health opportunities and outcomes

(in diet, for instance) and instead considers a lack of any of these “desirable situations” to be an absence of opportunity. The expansive view of what qualifies as opportunities has the disadvantage of ignoring the role of individual responsibility. However, this criticism is less relevant for the purposes of this report, which focuses on quantifying how opportunities are distributed by circumstances, as opposed to finding causal explanations for these inequalities. Accordingly, the findings of this report should be interpreted as a description of the extent to which women’s opportunities, in maternal and reproductive health are differentiated by life circumstances, and not as causal relationships pointing to the underlying reasons for these inequities, some of which could very well relate to individual behavioural patterns driven by intrinsic preferences and cultural norms. Whilst other studies have analysed maternal and reproductive health inequalities in the past, showing that almost all indicators are unequally distributed among population groups – with different wealth characteristics, areas of residence or educational levels13 –, this report aims to go one step further by considering all such health determinants simultaneously, to assess the magnitude and sources of inequality for different indicators of access to health care and health outcomes. Following the SDGs trend, and aligned with the SDG framework that advocates for strengthened stakeholder engagement and keeping pace with policy developments from an inter-sectorial perspective, we include many different factors in the same analysis to account for all possible inequalities. This is done using the Human Opportunity Index (HOI), a methodology developed by the WBG. The HOI is a measure of the coverage rate of an opportunity, discounted by inequality in its distribution across circumstance groups – sets of individuals with the same circumstances. It summarises two elements in a composite indicator: how many opportunities are available (the coverage rate), and how equitably those opportunities are distributed. If the coverage rate is close to the HOI, the distribution of the opportunities is equitable; when the HOI is lower than the coverage rate, the gap between them suggests inequality9. Interestingly, this methodology allows us to disaggregate the HOI into the marginal contribution (or weight) of each circumstance to the inequality of opportunity, meaning that data become available about which circumstances generate the highest inequalities between groups of individuals. The HOI is comparable across countries and indicators, and allows for the contributions or weights of different characteristics to be quantified. This report uses recent Demographic Health Surveys (DHS) data (year 2010 or later) to cover around 79 percent of the SSA population, allowing for comparisons across countries and analyses for the region as a whole. A more detailed discussion of the concepts underlying the HOI can be found in Chapter 2, methodological section.

A report from sub-Saharan Africa

References 1.

The World Bank Group. World Development Indicators. (2015). at <http:// data.world bank.org/data-catalog/world-development-indicators>

2. WHO African Region. The African Regional Health Report 2014. (2014). 3. WHO, UNICEF, UNFPA, World Bank Group & UN Population Division. Trends in Maternal Mortality : 1990 to 2015. 32, (2015). 4. United Nations. The Millenium Development Goals Report 2015. (2015). 5. Maternal Health Task Force. Post-2015: What’s next for maternal health? (2016). at <https://www.mhtf.org/topics/post-2015-whats-next-for- maternal-health/> 6. Melamed, C. After 2015. Contexts, politics and processes for a post-2015 global agreement on development. (2012). 7.

Alkenbrack, S., Chaitkin, M., Zeng, W., Couture, T. & Sharma, S. Did Equity of Reproductive and Maternal Health Service Coverage Increase during the MDG Era? An Analysis of Trends and Determinants across 74 Low- and Middle- Income Countries. PLoS One 10, e0134905 (2015).

8. Victora, C. G. et al. How changes in coverage affect equity in maternal and child health interventions in 35 Countdown to 2015 countries: An analysis of national surveys. Lancet 380, 1149–1156 (2012). 9. de Barros, R. P., Ferreira, F. H. G., Vega, J. R. M. & Chanduvi, J. S. Measuring Inequality of Opportunities in Latin America and the Carribean. Latin American Development Forum Series 46827, (2009). 10. Roemer, J. A Pragmatic Theory of Responsibility for the Egalitarian Planner. Philos. Public Aff. 22, 146–66 (1993). 11. Roemer, J. Equality of Opportunity. Cambridge, MA: Harvard University Press (1998). 12. Dabalen, A., Narayan, A., Saavedra-Chanduvi, J. & Suarez, A. H. Do African Children Have an Equal Chance? Human Opportunity report, Sub-Saharan Africa. (2015). 13. WHO: Department of Health Statistics and Information Systems. State of Inequality Reproductive, maternal, newborn and child health. (2015).

Inequalities A report from in women’s sub-Saharan andAfrica girls’ health opportunities and outcomes

©Andalu Vila San Juan

Chapter 2. Methodology

A report from sub-Saharan Africa

2.1 The Human Opportunity Index The Human Opportunity Index (HOI) is an aggregate measure that summarises the equitable availability of services. The endowment is the percentage of a population with access to a health service or with a health outcome that is necessary to progress in life. Unlike standard indices, such as coverage rate, to capture access to a particular service, the HOI also takes into account the (in) equitable access to the service among different groups of the population. The HOI is best understood as a coverage rate discounted for inequality of access. The HOI was developed by the World Bank Group (WBG) with external researchers and first presented in 2009 (Barros et al. 2009)1. The simplest way to express the HOI (H) for a particular opportunity is to take the average coverage rate for this opportunity (C) and apply a discount (P) due to inequality in coverage between population groups with different circumstances: H = C - P (1) Alternatively, the HOI can be expressed as: (2) P = C ( 1-D C

(

H = C 1-

Notice that from equation (2), the HOI is equal to the average coverage rate (H equals C) if access to the opportunity is independent of the circumstances (that is D=0). D is usually referred to as the dissimilarity index (D-index), and can be interpreted as the share of the total number of opportunities (that is, places available in a service) that needs to be reallocated between circumstance groups to ensure equality of opportunities. P is the penalty that the coverage (C) suffers due to inequality and it depends on the D-index (D) and on the coverage (C). For each circumstance group k, D can be computed as follows: 1 D= 2C

ak C -C k

(3)

k=1

where k is a group with a specific set of circumstances, Ck the coverage rate of group k, ak the share of group k in total population; and m the number of groups defined by circumstances. When analysing household survey data, the procedure consists of running a logistic regression model to estimate, at an individual level, the relationship between the access to a particular opportunity (binary dependent variable) and the circumstances of an individual (independent variables), on the full sample for which the HOI measure will be constructed. The estimated coefficients of the regression are used to obtain for each individual his/her predicted probability of access to the opportunity, which is then used to estimate the D-index, the coverage rate

Inequalities in women’s and girls’ health opportunities and outcomes

and the HOI1. Detailed information regarding construction, properties and limitations of the HOI has been described elsewhere1.

2.2 Shapley decomposition The Shapley decomposition is the decomposition of the D-index according to the Shapley value concept, first described by Shorrocks in 20122. The Shapley decomposition estimates the relative contribution of each circumstance to the inequality index so that the contributions add up to the value of the D-index, when it is computed with all the available circumstances in the data3. The D-index can change according to the set of circumstances used to define groups. In particular, it can only increase or remain constant when more circumstances are added to any existing set of circumstances. This in turn implies that the measured D-index is always a lower limit of the actual inequality that would be estimated if one were to use the set of all relevant circumstance variables. This property also allows defining the contribution of each circumstance to inequality as the increase in D-index due to the addition of a circumstance, or the marginal value added by a “new” circumstance to the D-index. Circumstances that add higher marginal value (in terms of the Shapley values) to the D-index are interpreted as “contributing” a larger share of the inequality between groups3. Detailed information regarding construction, properties and limitations of the Shapley decomposition and Shapley value has been described elsewhere1,4.

2.3 Study population The study population is comprised of women of reproductive age, between 15 and 49 years old. Three subgroups of this population are used to analyse certain indicators (Table 2.2) that are only relevant for a specific subgroup, taking into account the data available from the data sources (i.e. the Demographic Health Surveys (DHS)): Older adolescent girls between 15 and 19 years old, for whom the indicators of interest are those related to reproductive health and educational attainment. Women who had children in the last few years (five or two years, depending on the indicator) before the interview, for whom indicators related to pregnancy and infants’ health are analysed. Women who had a child within six months of the survey, used for the analysis of exclusive breastfeeding.

A report from sub-Saharan Africa

The analysis of “met need for family planning” is done for two non-overlapping subgroups: older adolescents (15 to 19 years) and women of reproductive age (20 to 49 years)I. Importantly, while it would have been relevant to analyse some of the indicators among younger adolescent girls between 10 and 14 years old, there is no data source. In fact, almost all studies on reproductive health among younger adolescent girls are conducted using data from retrospective questions addressed to adult women and older adolescent girls. The lack of information among this particular age group is one of the existing knowledge gaps for which it is necessary to generate reliable and timely data. Breaking the data gap to break the gender gap is highlighted in the new agenda of the Sustainable Development Goals (SDGs) era.

2.4 Data sources and country inclusion criteria Table 2.1 List of countries and DHS surveys Country

Survey year

African UN region

Country

2011-2012

Western

Malawi

Survey year

African UN region

2010

Eastern

2012-2013

Western

Benin

Burkina Faso

2010

Western

Mali

Burundi

2010

Eastern

Mozambique

2011

Eastern

Cameroon

2011

Central

Namibia

2013

Southern

Comoros

2012

Eastern

Niger

2012

Western

Congo Rep.

2011-2012

Central

Nigeria

2013

Western

Congo DR

2013-2014

Central

Rwanda

2014-2015

Eastern

Côte d’Ivoire

2011-2012

Western

Senegal

2014

Western

Ethiopia

2011

Eastern

Sierra Leone

2013

Western

Gabon

2012

Central

Tanzania

2010

Eastern

The Gambia

2013

Western

Togo

2013-2014

Western

Ghana

2014

Western

Uganda

2011

Eastern

Guinea

2012

Western

Zambia

2013-2014

Eastern

Kenya

2014

Eastern

Zimbabwe

2010-2011

Eastern

Liberia

2013

Western

Note: Congo DR = Congo Democratic Republic, Congo Rep. = Congo Republic, UN = United Nations. Research undertaken in March 2016.

For more information, see baseline populations in Table 2.2.

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

The data source for this study is the DHS financed by United States Agency for International Development (USAID)5. The DHS are community level, household surveys carried out in developing countries, including 33 sub-Saharan African (SSA) countries. They contain hundreds of questions related to household characteristics and household members, women of reproductive age and their children and men of reproductive age. The questionnaires administered to women of reproductive age provide useful information about maternal and reproductive health that makes the DHS the ideal data source for this report. The high degree of consistency in DHS questionnaires and sampling methodology across countries also make it particularly suitable for a multi-country study, as it allows for cross-country comparisons and/or aggregations of results. The countries included in the analysis were those having at least one available standard and complete DHS conducted between 2010 and 2015. The most recent dataset of each country was selected for the study (see Table 2.1 and Figure 2.1).

Figure 2.1 Map of the countries included in the analysis

Africa Regions Central East South West

A report from sub-Saharan Africa

2.5 Selecting and defining opportunities This report focuses on the study of health opportunities for women of reproductive age. The final selected indicators include health outcomes and the use or knowledge of certain health services (Table 2.2). Although most of these indicators can be influenced by individuals’ decisions, following the reasoning previously explained, in this report they will be treated as health opportunities for women. Health is only one dimension of women’s needs, but it becomes particularly salient during their reproductive age due to the elevated risk of death that they face6 (see also Chapter 1). The opportunities selected for this study can be interpreted as a necessary and minimum set of conditions to be met for a woman during her reproductive years from the perspective of her own and her children’s well-being. These include a number of variables measuring access to specific health services, as well as two “outcome” variables related to anaemia and adequate body mass index (BMI) that represent key aspects of maternal health associated with lower risks of mortality. For the specific study of older adolescents, education has also been selected as an opportunity because it is linked to adolescents’ reproductive health, early marriages and high-risk pregnancies7,8. Table 2.2 provides the list of the opportunities and the baseline population analysed in each case.

Table 2.2 List of opportunities and baseline population for whom they have been analysed Opportunity

Description

Not having anaemia

Women without anaemia Baseline population: all women of reproductive age (15-49)

Having the recommended BMI

Women with a BMI between 18.5 and 24.99 Baseline population: all women of reproductive age (15-49)

Met need for family planning

Women currently using contraceptive methods Baseline population: women of reproductive age (20-49) or older adolescent girls (15-19) with a need for family planning

Knowledge of a place where to get an HIV test

Women who know where to get an HIV test Baseline population: all women of reproductive age (15-49)

Four antenatal care visits attended by skilled personnel

Women who received at least four antenatal care visits attended by skilled personnel Baseline population: all women with newborns in the five years preceding the interview date

Delivery attended Women who had a delivery attended by a doctor, nurse, midwife or auxiliary midwife by a skilled attendant Baseline population: all women with newborns in the five years preceding the interview date Mother’s checkup after delivery

Women who had a checkup after delivery Baseline population: all women with newborns in the two/five years preceding the interview date table continues next page

Inequalities in women’s and girls’ health opportunities and outcomes

Table 2.2 List of opportunities and baseline population for whom they have been analysed (continued) Maternity care package

Women who attended at least four antenatal care visits, had a delivery attended by skilled personnel AND had a checkup after delivery Baseline population: all women with newborns in the five years preceding the interview date

Malaria prophylaxis during pregnancy

Women who took at least one dose of IPTp (SP) Baseline population: all women with newborns in the five years prior to interview and received at least one antenatal care visit

Being offered an HIV test during antenatal care

Women who were offered an HIV test during antenatal care visits Base population: all women with newborns in the two years prior to interview and received at least one antenatal care visit

Infant checkup within two months after delivery

Women whose last child had a checkup within two months after delivery Base population: all women with newborns in the two/five years prior to the interview date and the child survived

Six months of exclusive breastfeeding

Women who are breastfeeding and are not giving the children any other type of food or beverage Base population: all women with newborns in the six months prior to the interview date and the child survived

Having never been pregnant

Women who have never had a child, a stillbirth or an abortion, and are not currently pregnant Base population: Older adolescent girls (15-19)

Currently attending school

Women who are currently attending school (or university) Base population: Older adolescent girls (15-19)

Note: HIV = Human Immunodeficiency Virus, IPTp = Intermittent Preventive Treatment of malaria in Pregnancy, SP = Sulfadoxine-Pyrimethamine.

Notably, seven of the 13 opportunities are related to the health indicators listed in the recommendations of the World Health Organization (WHO) Commission on Information and Accountability for Women’s and Children’s Health. “Met need for family planning”, “antenatal care coverage with at least four visits during pregnancy”, “deliveries attended by skilled personnel”, “postnatal care visits” for mothers and newborns after delivery, and “six months of exclusive breastfeeding” are some of the reproductive and maternal health indicators used by the WHO to monitor progress on maternal and child health, and are used by other organisations and ongoing initiatives as well9,10. While “postnatal checkups” for mothers and newborns indicators are recommended within hours after delivery, this report will use a different time period for each indicator – two months for infant checkups and undetermined for mothers – because of the number of missing values for the recommended indicators in some of the datasets. The definition of all opportunities is the same across countries to allow comparisons. In general, DHS interviewers ask questions to all women who meet the criteria for a given question. In a few cases only, the baseline population is not the same because of the country-specific characteristics of the surveysII.

For example, all women who had a childbirth in the last five years answered questions regarding pregnancy, but for some countries and indicators, DHS program selected women who had their child during the last two years before the survey instead of five years, or randomly asked certain questions to a half or a third of the sample.

A report from sub-Saharan Africa

The composite HOI – an essential maternity care package A “composite HOI” that reflects access to multiple services for pregnant women has been defined, recognizing that none of these services are substitutes for each other, and underscoring that having access to all of them is critical for maternal health. Since the three key stages of pregnancy are the gestation months, the childbirth and the postpartum period, the three opportunities related to these stages constitute an essential maternity care package. For this analysis, “opportunity” refers to a woman attending at least four antenatal care visits, having a delivery attended by skilled personnel, and having a checkup after delivery. The calculation of the HOI then follows the methodology described earlier. The interpretation of this composite HOI is intuitive: it reflects the extent to which women who had a newborn were covered by an essential maternity care package. The package that has been considered is not the ideal, because a woman’s checkup should be within hours after delivery, but it can be interpreted as meeting a minimum standard.

2.6 Defining a set of circumstances Circumstances can be defined as the exogenous characteristics of women that, in absence of inequalities, should not be associated with having access to a service or having a particular health outcome (opportunities); contrarily, circumstances and opportunities are associated in the presence of inequalities. Some of the characteristics selected for this analysis such as education level, occupation and marital status can present the problem that they may be influenced by individual behaviour rather than being circumstances that are pre-determined at birth. For the purpose of this analysis, we favour this inclusive definition over a strict interpretation of circumstances for two main reasons. First, we are interested in seeing how access to opportunities varies by characteristics that differentiate groups of women – which is more important for policymaking purposes than finding differences in access by birth circumstances only. Second, characteristics like occupation and education are key contributors to the socio-economic situation of a woman of reproductive age, are extremely difficult to change and therefore exogenous for most practical purposes to a woman (or adolescent girl) at that point of time. Therefore, in assessing inequalities across groups, it seems important to take these characteristics into account, even though they do not conform to the standard definition of circumstances. In the rest of this report, these characteristics will be often referred to as circumstances to be consistent with how inequality of opportunities is typically presented. However, they must be understood as characteristics that are essentially beyond the control of a woman of reproductive age (or an older adolescent girl), rather than as circumstances determined purely at birth.

Inequalities in women’s and girls’ health opportunities and outcomes

Table 2.3 List of circumstances Women of reproductive age

Pregnant women

Older adolescent girls

Age

Age at delivery

Marital status

Number of children

Sex of the household head

Religion

Educational level

Occupational status

Location

Area (urban/rural)

Household status

Wealth index (quintiles)

Women’s characteristics

Household head characteristics Socio-cultural background

Note: The set of circumstances for Niger and Tanzania does not include religion, and the one for Mali and Senegal does not include occupational status, because these data were not available.

The circumstances that matter for women’s health opportunities could be slightly different across countries, but a common set is selected to allow cross-country comparisons. The list of selected circumstances can be categorised into five groups: women’s characteristics, socio-cultural background, household head characteristics, location and household status. Table 2.3 shows the complete list of circumstances. The codification of each variable (circumstance) is detailed in the Appendix A. The majority of the circumstances are used in the analysis of all women of reproductive age, but age is substituted by age at delivery for the analysis of maternity related opportunities. The reason for this change is the fact that age at delivery could condition certain aspects of maternity while age at the moment of the interview does not have any relationship with the time of pregnancy. For the analysis of adolescents’ opportunities, the list varies because education is considered an opportunity, age is taken into account in the selection of the group (age 15-19) and the inclusion of the number of children does not make sense when analysing whether adolescents have ever been pregnant. In the analysis of women of reproductive age, occupation of the woman appears to be highly correlated with wealth index and thus does not contribute significantly to the D-index and the HOI. Hence, occupation has been excluded from the analysis for women of reproductive age. But in the case of adolescents, occupation has been included as a circumstance because it is less correlated with wealth and could matter for the likelihood of older adolescent girls going to school, having more children or having their family planning needs met. Additionally, other identified circumstances relevant for women and girls’ health opportunities in the SSA region and thus potentially included in the analysis are

A report from sub-Saharan Africa

domestic violence and migration status. With a 36.6 percent prevalence, Africa ranks among the worst affected regions for intimate partner violence, the type of violence against women for which more data are availableIII. For combined intimate partner and non-partner sexual violence among all women of 15 years or older, estimated prevalence rate is 45.6 percent11. Migration has a complex effect on health, and women migrants may face adverse health conditions, such as poorer pregnancy outcomes and perinatal health indicators, or higher risk of sexually transmitted diseases, including HIV, resulting from voluntary and forced migration12. However, these factors were not included because of lack of data related to domestic violence in a third of the countries in this study, and the inclusion of migration questions in only three of these countries5. As mentioned earlier, the list of circumstances selected for constructing the HOI for an opportunity matters a great deal for the measure. Given this, all results that follow in the next chapters are subject to the limitation that the HOI is estimated for a specified list of circumstances and therefore subject to change if this list changes. However, while the HOI for an opportunity is not unique and depends on the number of circumstances considered, it cannot be higher if more circumstances are added to the existing set. In other words, the measure of the HOI used in this report will represent an upper boundary to the “true” HOI that would consider all circumstance groups (and a lower limit of the true D-index). Notably, the estimates always carry an error that could cause misleading comparisons between country HOIs, being a minor limitation to the analysis13. Having a common set of circumstances for a given opportunity across all countries also implies that certain circumstances important for inequality in a particular country are absent from the list. This could lead to the HOI (D-index) estimated for that country to be over-(under) estimated and not reflect the “true” inequality of opportunity in a specific country. Given this potential issue, the results throughout this report should be interpreted as the upper and lower boundaries of the HOI and D-index, respectively, for an opportunity in any particular country, computed for a set of circumstances common to all countries. Country-specificity is sacrificed to enable comparability of results across countries. Finally, it is important to take into account that all potential interactions between circumstances have been excluded from the analysis. The simplified specification is essential for the analysis to be tractable, and implies that the HOI (D-index) should be interpreted as the upper (lower) boundary of what the estimates would be if interactions were included.

III

Based on aggregated data from: Botswana, Cameroon, Democratic Republic of Congo, Ethiopia, Kenya, Lesotho, Liberia, Malawi, Mozambique, Namibia, Rwanda, South Africa, Swaziland, Uganda, United Republic of Tanzania, Zambia and Zimbabwe.

Inequalities in women’s and girls’ health opportunities and outcomes

2.7 Data management The country-level analyses were weighted using the sample weights available in the DHS programme datasets. These sample weights are expansion factors applied to adjust for differences in probability of selection across observations in the sample14. However, these country specific sample weights cannot be applied to the cross-country analysis when all study countries are pooled. An adjustment to the country specific sample weights needs to be performed to make country samples representative of women’s population in each country. Thus, more weight is given to those women belonging to a more populous country than those residing in smaller ones. The recalibration of the sample weights was performed by dividing each weight in a particular survey by the sum of the original sample weights and multiplying the result by the total number of women of reproductive age in the country15. No other data treatment has been applied to the data analysis.

Figure 2.2 Summary of opportunities, circumstances and groups of women included in the analysis Women of reproductive age

Not having anaemia

Four antenatal care visits

BMI between 18.5 and 24.99

Delivery attended by skilled personnel

Met need for familiy planning Knowledge of where to get an HIV test

Older adolescent girls

Postnatal checkup Maternity care package

Met need for family planning Having never been pregnant Currently attending school

Malaria prophylaxis during pregnancy HIV test offered during pregnancy

Opportunities

Pregnant women

Infant checkup after delivery

Age

Age at delivery

Marital status

Sex of the household head

Number of children

Religion

Sex of the household head

Area

Religion

Wealth index

Education level

Educational level

Occupational status

Area

Wealth index

Note: Women of reproductive age = 15-49 years old, except for met need for family planning (20-49 years old). Pregnant women = newborns born two/five years or six months prior to the interview date. Older adolescents = 15-19 years old.

Circumstances

Six months of exclusive breastfeeding

A report from sub-Saharan Africa

References 1. de Barros, R. P., Ferreira, F. H. G., Vega, J. R. M. & Chanduvi, J. S. Measuring Inequality of Opportunities in Latin America and the Carribean. Latin American Development Forum Series 46827, (2009). 2. Shorrocks, A. “Decomposition Procedures for Distributional Analysis: A Unified Framework Based on the Shapley Value. J. Econ. Inequal.11, 99–126 (2012). 3. Dabalen, A., Narayan, A., Saavedra-Chanduvi, J. & Suarez, A. H. Do African Children Have an Equal Chance? Human Opportunity report, Sub-Saharan Africa. (2015). 4. Hoyos, A. & Narayan, A. Inequality of Opportunities among Children: How Much Does Gender Matter? Background paper for the World Development Report 2012: Gender Equality and Development, World Bank, Washington, DC. itle. (2011). 5. USAID. The DHS Program. at <http://www.dhsprogram.com/> 6. WHO, UNICEF, UNFPA, World Bank Group & UN Population Division. Trends in Maternal Mortality : 1990 to 2015. 32, (2015). 7.

UNFPA. ADOLESCENT PREGNANCY : A Review of the Evidence. (2013).

8. Woog, V., Singh, S., Browne, A. & Philbin, J. Adolescent Women’s Need for and Use of Sexual and Reproductive Health Services in Developing Countries. (2015). 9. Victora, C. G. et al.Countdown to 2015: a decade of tracking progress for maternal, newborn, and child survival.Lancet (London, England) (2015). doi:10.1016/S0140-6736(15)00519-X 10. WHO Commission on Information and Accountability for Women’s and Children’s Health. Recommendation 2: Health indicators. at <http://www.who.int/woman_child_accountability/progress_information/ recommendation2/en/> 11. World Health Organization, London School of Hygiene & Tropical Medicine & South African Medical Research Council. Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. (2013). doi:10.1007/s13398-0140173-7.2 12. Carballo, M., Grocutt, M. & Hadzihasanovic, A. Women and migration: a public health issue. World Health Stat. Q. 49, 158–64 (1996). 13. Balcázar, C. F. Lower bounds on inequality of opportunity and measurement error. Econ. Lett. 137, 102–105 (2015). 14. Rutstein, S. O. & Rojas, G. Guide to DHS statistics. Demographic and health surveys methodology. 1–161 (2006). 15. The World Bank Group. World Development Indicators. (2015). at <http://data.worldbank.org/data-catalog/world-development-indicators>

Inequalities in women’s and girls’ health opportunities and outcomes

©Andalu Vila San Juan

Chapter 3. What is the state of health inequalities of women of reproductive age?

A report from sub-Saharan Africa

3.1 HOIs by country and multi-country pooled averages This section presents findings from the analysis of women’s health opportunities in 29 sub-Saharan Africa (SSA) countries using the most recently available Demographic Health Survey (DHS) data. The results are disaggregated by three subgroups of women: women of reproductive age (15-49), pregnant women and older adolescent girls (15-19).

3.1.1 Women of reproductive age (15-49 years old) Opportunities Not having anaemia BMI between 18.5 and 24.99 Met need for familiy planning Knowledge of where to get an HIV test

Context

Estimates suggest that at present about 26 percent of women of reproductive age – 225 million women – have an unmet need for family planning worldwide1,2. In low and middle income countries (LMICs), the number increases to 56 percent of the female population. However, in SSA, the percentage is 40 percent because of the desired big family sizes in the region3. Universal access to family planning would improve maternal health and survival by decreasing maternal deaths, including those attributable to unsafe abortions (eight percent) associated with unwanted pregnancies and the reduction of human immunodeficiency virus (HIV) transmission2. Behavioural, socio-economic and structural factors – those related to society structure and gender roles – make women in general more vulnerable to HIV infection4. In SSA, women account for 58 percent of the total population living with HIV. The disease disproportionally affects young women and adolescents. Every year, there are 380,000 new HIV infections among young women (10-24 years old). Fifteen percent of women aged 15 years and older living with HIV belong to the youth group (15-24 years old), and 80 percent of them live in SSA, where women become HIV infected, on average, five to seven years earlier than men. Regarding HIV knowledge in SSA, only 26 percent of adolescent girls have comprehensive knowledge about the disease, while among boys the percentage is 36 percent5. Additionally, two indicators relating to general women’s health outcomes were analysed for all women of reproductive age. Nutritional status is among the prin-

Inequalities in women’s and girls’ health opportunities and outcomes

cipal causes of morbidity and mortality in SSA6. Thus, not having any type or level of anaemia was considered as an opportunity for African women. It is estimated that about 468 million women aged 15-49 years worldwide are anaemic and between 48 percent and 57 percent of them live in Africa. Anaemia is an important health indicator for women because it can be produced by multiple causes, from poor nutrition, hormonal disorders or cancer, to malaria. Anaemia is associated with fatigue, increased susceptibility to infections, anaemia in pregnancy and postpartum haemorrhage, the latter being one of the principal causes of maternal mortality7,8. Therefore, not being anaemic constitutes a necessary condition for the well-being of women. The second indicator related to women’s general health and nutrition analysed is body mass index (BMI), which is calculated using the height and weight of the individual (weight/(height x height)). The recommended BMI values for an adult are between 18.5 and 24.99, where women with values lower than 18.5 are considered underweight and those above 24.99 are considered overweight9. Under nutrition is a persistent problem in LMICs, where nearly two percent of women have been recently assessed to have a BMI lower than 1610. Malnutrition can lead to fatigue and susceptibility to infections, and malnourished women are more likely to give birth to a newborn who has low birth weight, is more susceptible to diseases, and thus, has a higher probability of dying prematurely11.

Box 3. How to interpret the HOI %

The Human Opportunity Index (HOI) is the difference between the coverage rate and a penalty due to inequality:

30 25 20

HOI = Coverage - Penalty

The penalty comes from the dissimilarity index (D-index, the measure of inequality, see Chapter 2), but it also depends on the coverage rate of the opportunity:

10 5 0

HOI Coverage rate Penalty

Penalty = D-index · Coverage The gap between the grey bar and the blue bar reflects the reduction that the coverage rate suffers due to inequality (D-index), but taking into account that the penalty is also correlated with the coverage rate. Thus, when the coverage rate increases, the penalty due to inequality increases too; therefore in those opportunities where the coverage rate is high, for the same D-indices the penalties will be higher than for those opportunities with poor coverage rates.

A report from sub-Saharan Africa

Findings for reproductive health opportunities

Note: In order to avoid overlap with older adolescent girls’ analysis results (15 to 19 year old), the analysis of “met need for family planning” is restricted to women between 20 and 49 years old. Results from the two opportunities for reproductive health analysed vary significantly across countries. This is especially remarkable with regard to the level of “knowledge of where to get tested for HIV”, which has an almost 80 percent difference in the HOI between the best (Rwanda) and the worst performing country (Mali).

Figure 3.1 HOI for access to reproductive health a. Met need for family planning 100 80 60 40

kenya

Zimbabwe

Namibia Rwanda

Malawi

kenya

Congo

Zambia

Namibia

Zambia

Rwanda

Uganda

Gabon

Malawi Tanzania

Tanzania Gabon

Zimbabwe

Mozambique

Nigeria

Liberia

Cameroon

Ghana

Ethiopia

Senegal

Ghana

Uganda

Congo

Burundi

Senegal

Sierra Leone

Burkina Faso

Niger

Côte d’Ivoire

Togo

Liberia

Congo DR

Togo

Burkina Faso

Burundi

Comoros

Mozanbique

Mali

Benin

Guinea

The Gambia

Coverage (%) HOI (%) Pooled average HOI (%)

b. Knowledge of where to get an HIV test 100 80 60 40

Coverage (%) HOI (%) Pooled average HOI (%)

Sierra Leone

Ethiopia

The Gambia

Nigeria

Côte d’Ivoire

Benin

Niger

Congo DR

Cameroon

Guinea

Comoros

Mali

Inequalities in women’s and girls’ health opportunities and outcomes

The HOI for “met need for family planning” ranges between 20 (Guinea) and 78 (Namibia) with a multi-country pooled HOI of 46 (Figure 3.1). Notably, half of the countries rank below the average. The HOIs of family planning for adult women between 20 and 49 years old are on average higher than those for older adolescent girls (Figure 3.2). To compute the HOI for older adolescent girls, six circumstances were used, while the model for adult women used eight circumstancesIV. However, this is not necessarily the source of difference. Since the D-index (and therefore, the HOI) is sensitive to the number of circumstances selected for its calculation, adding a new circumstance always increases the D-index and lowers the HOI. Therefore, should the model for adolescent girls include two additional circumstances, similar to the model for adult women, their HOI would be even lower than it currently is.

Figure 3.2 Comparison between “met need for family planning” for older adolescent girls (15-19 years) and women of reproductive age (20-49 years)

Coverage (%) Older adolescent girls

100

80 Congo

Namibia

Gabon Zimbabwe Cameroon Sierra Leone kenya Tanzania Benin Côte d’Ivoire Togo UgandaNigeria Malawi Ethiopia Zambia Congo RD Ghana Senegal Rwanda Burkina Faso Liberia Comoros Niger Mozanbique Burundi Mali Guinea

The Gambia

40 60 Coverage (%) Women 20-49 years old

100

45o line

Both the coverage rates and HOIs for “knowledge of where to get an HIV test” seem to have a gradual gradient. The HOIs for this opportunity seem to be highly correlated with HIV prevalence rates in the countries. Thus, countries with the highest HIV prevalence rates in the SSA region (i.e. Zimbabwe, Tanzania, Uganda, Zambia, Malawi, Kenya and Namibia, with the exception of Mozambique), also have the highest HOIs for the knowledge of where to get tested. Encouragingly, the inequality across women for this opportunity is low, suggesting that the policy to test for HIV status appears to be more equitable.

Older adolescent girls circumstances: wealth index, region of residence, marital status, occupational status, religion and sex of the household head. Women between 20 and 49 years old circumstances: wealth index, region of residence, marital status, educational attainment, religion, sex of the household head, number of children and age.

A report from sub-Saharan Africa

Findings for women’s general health

In terms of coverage rates, overall women’s health outcomes examined – measured by the HOI of “not having anaemia” and “having the recommended BMI”V – are poor in SSA, for both indicators, although penalties for inequalities seem to be lower than for reproductive health opportunities (Figure 3.3). Differences across countries are less marked than for other indicators. The multi-country pooled HOI for “not having anaemia” and “having the recommended BMI” is in both cases slightly more than 60. For anaemia, the values range between 37 (Gabon) and almost 81 (Ethiopia), and for BMI, between 41 (Gabon) and 74 (Burundi).

Many women are anaemic (multi-country pooled prevalence of anaemia: 35 percent) and the rates are similar across most countries for which the data were available. Similarly, in almost every country, two out of three women have BMI within the recommended range, and there is little variation across countries. Notably, six of the 29 SSA countries selected do not have available data about anaemia levels. There is no correlation between levels of anaemia and BMI within countries except for Gabon, which has the lowest HOI for both, and Burundi and Rwanda, with some of the highest HOIs for both opportunities. For example, Namibia has an HOI of almost 75 for anaemia, but for the recommended BMI the HOI is less than 50, one of the lowest. There does not appear to be any geographical or income pattern in the distribution of anaemia and BMI. However, five out of the six countries with the highest HOI for not having anaemia are also the ones with the lowest incidences of malaria in SSA. Zimbabwe, Namibia, Burundi, Rwanda and Ethiopia have less than 17,000 new cases of malaria reported each year per 100,000 people12.

For most countries, data on these two opportunities were collected for a randomly selected half or a third of the sample of all women interviewed in the country. However, the results generalise to the whole population of adult women.

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.3 HOI for general women’s health a. Not having anaemia 100 80 60 40

Uganda

Namibia

Rwanda

Burundi

Ethiopia

Uganda

Congo DR

Mali

Rwanda

Malawi

Zimbabwe Guinea

Congo DR Tanzania

Niger

Tanzania

Cameroon

Benin Benin

Zambia

Ghana Liberia

Côte d’Ivoire

Niger

Sierra Leone

Togo

Nigeria

Togo

Burkina Faso

Mali

Guinea

Côte d’Ivoire

Congo

Mozanbique

The Gambia

Gabon

Coverage (%) HOI (%) Pooled average HOI (%)

b. Having the recommended BMI 100 80 60 40

Burundi

Mozambique

Malawi

Burkina Faso

Sierra Leone

Zimbabwe

The Gambia

Cameroon

Kenya

Congo

Comoros

Ghana

Namibia

Gabon

Coverage (%) HOI (%) Pooled average HOI (%)

Conclusions

Differences across countries also exist for reproductive health opportunities and outcomes. “Met need for family planning” shows slightly higher HOIs among women aged 20 years or older (multi-country HOI: 46 percent) than among adolescent girls (multi-country HOI: 40 percent). “Knowledge of a place where to be tested for HIV” follows a pattern related to HIV prevalence rate. Penalties for inequalities decrease as coverage of the opportunity increases.

A report from sub-Saharan Africa

Opportunities for “not having anaemia” and “having the recommended BMI” exhibit significantly lower inequality among SSA women. In those countries where malaria is less prevalent, anaemia prevalence is low, which leads to a high HOI of “not having anaemia”. The results could be due to the fact that malaria is one of the most important causes of anaemia in endemic areas. One in three women in most SSA countries does not have the recommended BMI meaning that they could be either undernourished or obese. Therefore, the countries that under perform in terms of BMI are not only those where a sizeable share of the population may be stunted, they might also have poor nutritional habits that lead to obesity.

3.1.2 Pregnant women Opportunities Four antenatal care visits Delivery attended by skilled personnel Postnatal checkup Maternity care package Malaria prophylaxis during pregnancy HIV test offered during pregnancy Infant checkup after delivery Six months of exclusive breastfeeding

Context

Good quality maternity care is crucial for the survival and health of both the mother and the newborn child. However, in regions such as SSA the proportion of births attended by skilled health personnel – i.e. doctors, nurses and midwives – is still very low, 52 percent13. Of note, in November 2016 the World Health Organization (WHO) launched the new antenatal care guidelines where the recommended minimum number of antenatal care visits increased from four to eight14. However, in this report we used the indicator of four antenatal visits given the low antenatal care coverage in SSA. Since 1990, the utilization of the recommended four antenatal care visits has remained low at 47 percent to 49 percent in 2015, in SSA13. Thus, highlighting the need to speed the efforts to achieve the full life-saving potential of antenatal care for women and newborns.

Inequalities in women’s and girls’ health opportunities and outcomes

In SSA, infectious diseases contribute to the burden of maternal and perinatal deaths15. HIV and malaria are known to be two relevant infectious diseases particularly during reproductive age and pregnancy. Globally, 85 percent of pregnant women living with HIV live in SSA5. HIV prevention and treatment during pregnancy protects against transmission of the disease to the newborn during gestation or delivery. In the 21 countries with the highest HIV rates, all in the SSA region, the number of women in need of mother-to-child transmission prevention procedures for HIV is 1.3 million. In 2013, half of women in LMICs were not tested for HIV during pregnancy, a health procedure that is essential in these settings, and therefore, they were not able to access HIV treatment and care in case of need5. Regarding malaria infection, in high transmissions areas, the risk of low birth weight doubles when there is placental malaria and up to five percent of the neonates can be born with congenital disease. Further, pregnant women infected with malaria more frequently show higher parasitaemia, severe anaemia, hypoglycaemia and acute pulmonary oedema16. In SSA, 10,000 pregnant women and 200,000 of their infants die every year due to malaria infection during pregnancy17. Intermittent Preventive Treatment of malaria in Pregnancy (IPTp) with sulfadoxine-pyrimethamine (SP) is considered one of the most cost-effective interventions to prevent these deaths18. Since 2014, the WHO has recommended the intake of three doses of SP as IPTp for all women living in moderate to high transmission areas, at each scheduled antenatal care visit, starting in the second trimester of gestation19. It is estimated that in the African countries that adopted this policy, 52 percent of pregnant women received at least one dose of SP in 2014, 40 percent received two or more doses and only 17 percent received three or more doses20. Neonatal care, strictly speaking, is not an opportunity specific to women. However, neonatal health is inextricably linked to maternal health, and therefore it is worth including this component in the analysis. Newborn health was not specifically addressed in the Millennium Development Goal (MDG) framework. Over that period, progress in the rate of child survival among children aged one to 59 months outpaced advances in reducing neonatal mortality; as a result, neonatal deaths now represent a larger share (45 percent) of all under-five deaths globally, resulting in 2.7 million deaths each year21. More than 80 percent of all newborn deaths result from three preventable and treatable conditions – complications due to prematurity, intrapartum-related deaths (including birth asphyxia) and neonatal infections. Cost-effective, proven interventions exist to prevent and treat each main cause. Improving effective coverage of care around the time of birth – the most risky period for mothers and their newborns – requires educated and equipped health workers, and availability of essential commodities22. The WHO recommends that women who have delivered in a health facility should receive postnatal care for at least 24 hours after birth. If a birth is at home, the first postnatal contact should be as early as possible within 24 hours of birth23. Postnatal care offers an opportunity to provide a number of interventions including counselling on exclusive breastfeeding, birth spacing and contraceptive methods, and

A report from sub-Saharan Africa

educating women on the benefits to their own and their newborn’s health of doing so. The WHO recommends three additional postnatal care contacts on day three, between days seven and 14 after birth and six weeks after birth. Newborn health has been given more prominence in the Sustainable Development Goals (SDGs) targets (target 3.2: by 2030, end preventable deaths of newborns and children under five years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1,000 live births and under-five mortality to at least as low as 25 per 1,000 live births)24. In 2015, the global neonatal mortality rate, that is, the likelihood of dying in the first 28 days of life, was 19 deaths per 1,000 live births13.

Findings for maternity care

In this section, the availability of each intervention was assessed individually, first “four antenatal care visits”, “delivery attended by skilled personnel” and “postnatal women’s checkup”, after which the joint availability of the minimum maternity care package, which is labelled as the “Composite HOI”, is examined. Results show that: Maternity care has generally low coverage and is unequally distributed in SSA. On average, for each one of the three individual opportunities, only 35 to 40 percent of women in the SSA region can claim to have access. As has been the pattern for most of the opportunities examined thus far, there is substantial country heterogeneity: coverage ranges between 80 to below 20 percent for the three opportunities. These opportunities also vary widely within countries, without any clear trend, reflecting that performing better in one of the opportunities does not mean that the performance is also good in the others. For example, “delivery attended by skilled personnel” and “four antenatal care visits” score very low in Burkina Faso (HOIs: 18 percent and nine percent, respectively), but the country performs very well on “postnatal checkups” (HOI: 81 percent). The HOI of “delivery attended by skilled personnel” for women that delivered at home is much lower than for those delivering in a health facility, when the sample is split by location of delivery (Figure 3.5). The highest HOI of “delivery attended by skilled personnel” for deliveries at home is 23 (Comoros), and in some countries it is virtually zero, which suggests that few women receive services from trained health personnel when delivery takes place at home. By contrast, for women who gave birth in a health facility, three quarters of the countries have a HOI of “delivery attended by skilled personnel” higher than 90.

41 40

20 Côte d’Ivoire

Malawi

Rwanda

Tanzania

Mozambique

Congo DR

Burundi

Niger

Mali

Kenya

Ethiopia

Liberia Congo Gabon Namibia Ghana Sierra Leone

Benin Namibia Rwanda Gabon Congo

The Gambia

Malawi Comoros

Cameroon Zimbabwe

Benin

Cameroon Ghana

Comoros

Zimbabwe Zambia

Guinea Zambia

Kenya

Senegal

The Gambia

Liberia

Nigeria

Burundi

80 Togo

100 Uganda

b. Delivery attended by skilled personnel

Sierra Leone

Coverage (%) HOI (%) Pooled average HOI (%)

Côte d’Ivoire

Togo

Senegal

Uganda

Tanzania

Congo DR

Guinea

Mali

Nigeria

Niger

Burkina Faso

Mozambique

Burkina Faso

Ethiopia

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.4 HOI for maternity care

a. Four antenatal care visits 100

Coverage (%) HOI (%) Pooled average HOI (%)

figure continues next page

A report from sub-Saharan Africa

Figure 3.4 HOI for maternity care (continued) c. Postnatal checkup 100 80 60 40

Ghana

Namibia

Senegal

Burkima Faso

Togo

Sierra Leone

Côte d’Ivoire

Congo

The Gambia

Liberia

Gabon

Zambia

Benin

Mozambique

Comoros

Kenya

Cango DR

Zimbabwe

Mali

Rwanda

Cameroon

Guinea

Nigeria

Niger

Burundi

Ethiopia

Uganda

Coverage (%) HOI (%) Pooled average HOI (%)

Figure 3.5 HOI for “delivery attended by skilled personnel” by place of delivery 100

HOI (%)

80 60 40

Rwanda

The Gambia

Namibia

Burundi

Comoros

Cameroon

Congo

Zimbabwe

Kenya

Côte d’Ivoire

Sierra Leone

Gabon

Uganda

Liberia

Malawi

Ghana

Ethiopia

Niger

Nigeria

Tanzania

Zambia

Benin

Guinea

Togo

Senegal

Mali

Congo DR

Mozambique

Burkima Faso

Delivery in a health facility Delivery at home

Findings for the minimum maternity care package (Composite HOI)

This section revisits the opportunities for perinatal care, but unlike the previous section, which looked at individual opportunities, it assesses how SSA countries perform when the opportunity of interest is a minimum package of essential maternity care (“four antenatal care visits”, “delivery attended by skilled personnel” and “postnatal checkup” regardless of time since delivery). Although the package does not include all services that women need, for example, checkup for women within hours after delivery is not included given data unavailability, it largely covers extent of the basic interventions necessary to avoid high risk situations for mothers.

Inequalities in women’s and girls’ health opportunities and outcomes

In general coverage is very low, as the multi-country pooled composite HOI (16) reveals. Despite this low average, ten countries have values below this value, meaning that the availability of the package of services is almost inexistent in one out of three countries. In all countries, there is substantial inequality of access to the package of services (Figure 3.6).

Figure 3.6 Composite HOI for maternal care 100 80 60 40

Namibia

Ghana

Congo

Gabon

Liberia

Sierra Leone

Gambia

Togo

Benin

Zambia

Comoros

Côte d’Ivoire

Senegal

Cameroon

Zimbabwe

Guinea

Nigeria

Mali

Rwanda

Uganda

Kenya

Congo DR

Niger

Burundi

Mozambique

Etiopia

Burkina Faso

Coverage (%) HOI (%) Pooled average HOI (%)

Differences across countries are large, ranging from moderate to no provision of the package (Ethiopia has a HOI near 0) to a HOI around 60 (Namibia). To further assess whether this essential maternal care package could have an impact on maternal and infant health, infant mortality rate (IMR)VI and maternal mortality ratio (MMR)VII have been correlated using the D-index of the composite HOI for each country and survey year (Figure 3.7). The correlations show how different infant or maternal mortality are between countries, depending on differences in inequality of opportunity. The higher the infant (or maternal) mortality is, the higher the inequality of access to the essential maternity care package (D-index) is in the country. The correlations appear to be significantly strong, especially in the case of IMR, suggesting that countries with higher inequality of the maternity care package also tend to have higher IMR and MMR. Ethiopia and Sierra Leone are outlier countries that make the correlation weaker. In both cases, the special situation of the countries regarding their health systems could explain these results (Box 4).

Infant mortality rate is the number of infants dying before reaching one year of age, per 1,000 live births in a given year.

VII

Maternal mortality ratio is the number of women who die from pregnancy-related causes while pregnant or within 42 days of pregnancy termination per 100,000 live births.

A report from sub-Saharan Africa

Box 4. Country cases: Ethiopia and Sierra Leone Ethiopia: The Ethiopian government launched the Health Extension Plan (HEP) in order to provide basic health prevention services and treatment to rural communities. The HEP was operative between 2004 and 2005 when the first Health Extension Workers (HEWs) graduated25,26. HEWs are trained during a year and then, they spread the adequate health messages to their communities to engage people in good health and hygiene practices. The indicators analysed regarding pregnancy are always constructed for antenatal care visits and deliveries attended by skilled personnel, thus all individuals attended by HEWs are counted as a “not skilled attended” case. Consequently, Ethiopia stands always at the bottom of the graphs with the lowest HOI for maternity care indicators. For this same reason, it can be observed that the inequalities in the indicators for Ethiopia are the highest in the group of countries analysed. Rural communities are the ones with the most important presence of HEWs, while in cities and among the wealthier groups of women, the conventional medical system is easier to access. Sierra Leone: Since the end of the Civil War in 2002, and despite political instability, Sierra Leone moved to process of peace and regeneration of the country. The war caused important damages in infrastructures, high death rates, migration of health personnel and the collapse of services, all of these issues that the government has worked to rebuild since then27,28. Despite being the country with the highest mortality rates in SSA, Sierra Leone is not among the regions with the biggest inequalities in maternal health. In fact, other studies found that, for example, concentration of nurses and midwives is similar between rural and urban areas, in contrast with other countries where the health workforce is concentrated in urban settings29. This fact could be caused by the general poverty situation in which the country was found after the end of the war, and the rebuilding process would have been quite similar in different regions, either rural or urban. Finally, although it is not observed in these results, the obstacles that pregnant women face in Sierra Leone under normal circumstances – owing to access barriers and the limitations of the weakest health systems in SSA – have significantly increased since the start of the Ebola outbreak (2014-2016). The breakdown of weak public health systems triggered by the epidemic has contributed to making medical resources scarcer and services (e.g. emergency maternity care, family planning, immunisation programmes or prevention of malaria) less available or halted, which could eliminate the gains achieved by the country, resulting in much higher maternal, newborn and child mortality rates30.

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.7 Correlations between the D-index of the maternity care package and IMR or MMR

D-index - maternity care package (%)

100

80 Ethiopia

Nigeria Mali Burkina Faso Mozambique Guinea Niger Cameroon Togo CongoDR Uganda Senegal Côte d’Ivoire Zimbabwe Ghana Zambia Benin Liberia Burundi Rwanda The Gambia Comoros Congo Namibia

Kenya

60 Infant Mortality Rate

Linear prediction (all countries) Linear prediction (whithout Ethiopia and Sierra Leone)

Sierra Leone

100

R2=0.05 R2 (without Ethiopia and Sierra Leone)=0.37

D-index - maternity care package (%)

100

80 Ethiopia

Mali Mozambique Niger Guinea Cameroon Togo Congo DR Senegal Uganda Côte d’Ivoire Ghana Benin Zimbabwe Liberia Burundi Zambia Comoros Rwanda Namibia Gabon Congo The Gambia Burkina Faso

Kenya

500

Linear prediction (all countries) Linear prediction (whithout Ethiopia and Sierra Leone) Note: R2= Correlation Coefficient

Nigeria

Maternal Mortality Ratio

Sierra Leone

1000

1500

R2<0.001 R2 (without Ethiopia and Sierra Leone)=0.17

A report from sub-Saharan Africa

Findings for malaria and HIV

Offering HIV testing during antenatal care has a multi-country pooled HOI of nearly 60, and follows a similar pattern to the previous opportunity examined, regarding women’s level of knowledge on a place to get tested for HIV. HIV testing coverage during pregnancy is substantially higher and relatively equal for women in countries with high HIV incidence. HOI scores for “malaria prophylaxis” vary greatly between countries, ranging from virtually zero in Burundi, to more than 90 in The Gambia and Zambia, with a multi-country pooled HOI of 43. The low scores for Burundi could be explained by the implementation of the WHO recommendations on malaria prophylaxis previously mentioned19. Burundi and Ethiopia are the only SSA countries included in this report that decided not to implement the recommendation, and Rwanda decided to stop recommending SP in 2008; in all cases the reason was the low incidence of malaria rates in the countries. Ethiopia and Rwanda do not have available information on IPTp in the DHS survey questionnaires and therefore were not included.

Figure 3.8 HOI for malaria and HIV indicators a. Malaria prophylaxis 100 80 60 40

Coverage (%) HOI (%) Pooled average HOI (%)

Zambia

The Gambia

Ghana

Malawi

Togo

Mail

Senegal

Niger

Liberia

Tanzania

Sierra Leone

Burkina Faso

Congo

Comoros

Cameroon

Uganda

Benin

Mozambique

Guinea

Nigeria

Congo DR

Kenya

Côte d’Ivoire

Zimbabwe

Gabon

Namibia

Burundi

Inequalities in women’s and girls’ health opportunities and outcomes

b. HIV test offered during pregnancy 100 80 60 40

Kenya

Rwanda

Namibia

Malawi

Zambia

Gabon

Tanzania

Zimbabwe

Togo

Uganda

Ghana

Liberia

Mozambique

Cameroon

The Gambia

Sierra Leone

Benin

Nigeria

Burundi

Ethiopia

Côte d’Ivoire

Burkina Faso

Niger

Congo

Mali

Comoros

Congo DR

Guinea

Coverage (%) HOI (%) Pooled average HOI (%)

Findings for infant care

The WHO recommends using the “newborn checkup within two days after birth” as a relevant indicator31. However, available data in SSA countries on this intervention is scarce due to a number of reasons including: weak information systems or poorly kept records, survey respondents’ inability to recall information, very low number of women attending the newborn checkup within two days after delivery or large number of women that deliver at home. As a result, this indicator had many missing values and could not be included. This is a limitation of the analysis, since this information would be relevant, especially in the current context of neonatal mortality accounting for 45 percent of the under-five child deaths globally21. Notably, after two months, an infant – or a child of less than one year of age – is no longer a newborn. This is relevant because newborn checkups immediately after delivery are intended to prevent or address the causes of neonatal mortality – that currently account for nearly half of child mortality32 – and to educate and counsel women on practices beneficial to both the mother and the newborn, such as breastfeeding, birth spacing or immunisation. Thus, the indicator “infant checkup within two months after delivery” has been used as an alternative indicator to the “newborn checkup within two days after birth” in the absence of information regarding neonates. Although “infant checkup after two months” is not a standard indicator, it conveys a measure of action taken to improve infant care by the mother. “Infant checkup after delivery” reveals high inequalities among countries ranging from Ethiopia (three) to Burkina Faso (81) in HOI. The multi-country pooled HOI is only 32. In general, West African countries appear to outperform the rest (East, Central and Southern African countries), because they are above the multi-country pooled HOI (green line) with only a few exceptions.

A report from sub-Saharan Africa

Looking at individual countries, Ethiopia is among the worst performing in both cases, while Ghana, Burkina Faso and Senegal have among the highest HOI scores for these two opportunities.

Figure 3.9 HOI for infant care a. Infant checkup within two months after delivery 100 80 60 40

Ghana

Senegal Burkina Faso

Togo The Gambia

Burkina Faso

The Gambia Niger

Liberia

Côte d’Ivoire Ethiopia

Burundi

Kenya

Sierra Leone

Togo

Guinea

Niger

Liberia

Zimbabwe Côte d’Ivoire

Senegal

Guinea Zambia

Kenya

Congo

Zambia Malawi

Gabon

Namibia

Namibia Ghana

Sierra Leone

Rwanda

Mali

Benin

Comoros

Malawi

Uganda

Nigeria

Burundi

Tanzania

Congo DR

Cameroon

Ethiopia

Coverage (%) HOI (%) Pooled average HOI (%)

b. Six months of exclusive breastfeeding 100 80 60 40

Rwanda

Cameroon

Mali

Congo DR

Uganda

Gabon

Nigeria

Tanzania

Comoros

Congo

Benin

Zimbabwe

Mozambique

Coverage (%) HOI (%) Pooled average HOI (%)

In stark contrast, SSA countries overall do very well with the opportunity “six months of exclusive breastfeeding”. The weighted HOI is around 80, which is the highest obtained among all opportunities examined in this study. Mozambique, the country that ranks lowest has an HOI above 50 percent, and the highest, Rwanda, has an HOI over 95.

Inequalities in women’s and girls’ health opportunities and outcomes

There is very little inequality in breastfeeding within countries. The most important inequality that can be observed is between regions. East African countries appear to be in general below the multi-country pooled HOI (80 percent). The negative correlation between HIV prevalence and the HOI for “exclusive breastfeeding” may be explained by the initial recommendation that HIV-infected mothers should not breastfeed to avoid the risk of transmitting the virus to their newborns. Currently, all lactating women should receive antiretroviral treatment, and thus breastfeeding is recommended at least up to six months of age, even if the mother is HIV positive33. However, in some settings exclusive breastfeeding might not be fully implemented, and the population remains cautious. This hypothesis would need to be tested with disaggregated data by HIV status of the women.

Figure 3.10 Correlation between HIV prevalence and the HOI for “six months of exclusive breastfeeding”

HOI “exclusive breastfeeding” (%)

100

Rwanda Burkina Faso Burundi Liberia Niger The Gambia Ethiopia Guinea Senegal Togo Kenya Côte d’Ivoire 80 Sierra Leone Mali Ghana NigeriaCameroon Congo DR Gabon

Benin

Malawi

Zambia

Namibia

Uganda

Tanzania

Congo

Zimbabwe

Mozambique

Linear prediction (all countries)

10 HIV prevalence (%)

R2=0.18

Conclusions

Maternity care is inadequate and is characterised by large inequalities within and across countries. These results also indicate that the high rate of home deliveries may be the reason why most births are not attended by skilled personnel. Although the package analysed is the minimum essential to avoid high risks during pregnancy, it is still unavailable for many women and is very unevenly distributed among the population of women in SSA countries.

These results suggest the possibility that improving access to this essential set of services for maternity care could lead to reductions in IMR and MMR.

A report from sub-Saharan Africa

The HOIs of “malaria prophylaxis during pregnancy” vary greatly between the countries. This could be caused by differences in antenatal care clinic attendance (because attending once might not be enough, especially when the visit occurs in the first trimester), uncertainty among health workers about SP administration and/or stock outs of SP at the health facility level, among other reasons34. Regarding HIV testing, high burden countries outperform the rest in terms of coverage and HOI. In countries where HIV prevalence is lower than five, results are significantly worse than in high prevalence countries. Therefore, actions leading to expanding this essential health service to offer HIV testing before and during pregnancy should be encouraged. Infant care indicators vary widely between countries and reveal disparate results. On the one hand, the “infant checkup after delivery” shows inequalities among and within countries, with low HOIs. Importantly, the high neonatal mortality rates could be reduced with newborn checkups within hours after delivery. However, there is no data available for this crucial indicator, highlighting the need for improving health information systems as a prerequisite to addressing the causes of newborn mortality and morbidity. On the other hand, “six months of exclusive breastfeeding” is a good example of an extended practice with low inequalities throughout SSA countries.

3.1.3 Older adolescent girls (15-19 years old) Older adolescent girls Met need for family planning Having never been pregnant Currently attending school

Context

In SSA, there are more than one billion people, and 23 percent of them are adolescents between ten and 19 years old21. Older adolescents – those between 15 and 19 years old – represent 11 percent of the SSA population. Half of these are girls, and 11 million are sexually active and want to delay childbirth for at least two years on average21,35. A third of these adolescent girls (3.6 million) are using a modern contraceptive method to avoid pregnancy, but the rest currently face an unmet need for contraception35. This unmet need is always higher among adolescents than among other women of reproductive age (60 percent and 26 percent, respectively)1. As a result, in SSA, almost half of pregnancies among older adolescents are unwanted, and half of them end in abortion in countries where abortion is illegal and usually performed under unsafe conditions. It is estimated that if

Inequalities in women’s and girls’ health opportunities and outcomes

all the adolescent girls in SSA who needed contraceptives used them, unintended pregnancies would drop by 2.7 million per year35. The fertility rate among older adolescent girls has not undergone important variations during the MDGs period. SSA is the region that has made the least progress – a four percent reduction between 2000 and 2015 –, and is currently at 102 births per 1,000 adolescent girls13,36. Pregnancies during adolescence are life-threatening events, due to a greater probability of high-risk pregnancy at this age37. Gestation and labour complications are among the leading causes of maternal death among older adolescent girls in LMICs. Further, newborns of adolescent mothers face major health risks compared to those born to older women1. Therefore improving adolescents’ access to sexual and reproductive health information and services is key. However, in 2015, less than half of women (15 to 49 years of age) in SSA who were married or in union satisfied their need for family planning by using modern contraceptive methods. Notably, child marriage is an important driver of adolescent pregnancy in SSA, where 90 percent of adolescents who give birth are married38. Adolescent pregnancy has strong negative effects on future educational and professional opportunities for girls, helping to perpetuate the cycle of poverty and ill health1. Education is tightly linked to adolescents’ current and future reproductive health outcomes, and this is the reason why education has been considered an opportunity39. Although the gender gap in educational attainment has narrowed since 1990, currently 68 percent of older adolescent girls from LMICs have completed seven or more years of education. This proportion is remarkably lower for Africa (51 percent) compared to other world regions (Latin America 82 percent, Asia 72 percent)1. Finally, preventing unintended pregnancy and reducing adolescent childbearing through universal access to sexual and reproductive health-care services are critical to advance the health of women, children and adolescents, a precondition to achieve the SDGs. Three of the goals— SDG3, SDG4 and SDG5, related to health, education and gender equality, respectively — depend largely on improvements in adolescents’ living conditions. However, adolescents have only been recently included in the global agenda as a separate group of individuals with specific needs40.

Findings

Across SSA, data show that from the entire older adolescent girls’ sample, only half of them attend school. Thirty-eight percent work and attend school at the same time. Twenty-three percent of older adolescent girls included in the study have been pregnant at some stage; of these, roughly 79 percent have been married or in a union, half are working and only eight percent are attending school. In general, the coverage of the three opportunities analysed for older adolescent girls is low, below 40 percent in two of the opportunities, while the penalties for inequalities are very high, meaning that there are important differences in cov-

A report from sub-Saharan Africa

erage rates between groups of adolescents with different characteristics, such as differences between urban-rural residents and between married and unmarried girls, among others.

Figure 3.11 HOI for access to reproductive health and education for older adolescent girls a. Met need for family planning 100 80 60 40

Kenya

Cameroon

Zimbabwe

Gabon

Congo

Namibia

Congo

Liberia

Kenya

Comoros

Namibia

Gabon

Benin

Sierra Leone Cameroon

Tanzania Sierra Leone

Congo DR

Malawi

Nigeria

Malawi

Côte d’Ivoire Ethiopia

Uganda

Togo

Uganda

Burundi

Zambia Togo

Zambia

Ethiopia

Congo DR

The Gambia

Benin

Ghana

Nigeria

Rwanda

Senegal

Liberia

Burkina Faso Zimbabwe

Comoros

Niger

Mozambique

Mali

Burundi

Guinea

Gambia

Coverage (%) HOI (%) Pooled average HOI (%)

b. Currently attending school 100 80 60 40

Ghana

Mozambique

Guinea

Tanzania

Côte d’Ivoire

Mali

Burkina Faso

Niger

Coverage (%) HOI (%) Pooled average HOI (%) figure continues next page

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.11 HOI for access to reproductive health and education for older adolescent girls (continued) c. Having never been pregnant 100 80 60 40

Rwanda

Burundi

Comoros

Ethiopia

Benin

Ghana

Togo

Senegal

The Gambia

Kenya

Namibia

Nigeria

Tanzania

Zimbabwe

Uganda

Burkina Faso

Cameroon

Malawi

Gabon

Congo DR

Zambia

Sierra Leone

Côte d’Ivoire

Liberia

Congo

Guinea

Mali

Mozambique

Niger

Coverage (%) HOI (%) Pooled average HOI (%)

The HOI for the three indicators varies greatly across countries. The HOI for “met need for family planning” ranges from 12 (The Gambia) to 68 (Namibia), with a multi-country pooled HOI of 40. The multi-country HOI for “currently attending school” is 40, but the differences across countries are wider, ranging from six (Niger) to 72 (Gabon). The HOI for “having never been pregnant” is generally higher than the HOI of other indicators for all countries, with a multi-country pooled HOI of almost 70 and ranging from 44 (Mali) to 90 (Rwanda). Cameroon, Kenya, Gabon, Rwanda, Congo and Namibia are among the top performing countries in terms of HOI for “met need for family planning” and “school attendance”. They also have some of the best scores with regard to adolescent girls avoiding pregnancies. In contrast, the HOIs for Guinea, Mozambique, Niger and Mali are among the lowest ranked for the three opportunities examined. Regionally, results also show that Sahel adolescents suffer larger disadvantages than adolescents from other regions. The countries included in this study belonging to the Sahel region are Senegal, Mali, Burkina Faso, and Niger. With the exception of Senegal in adolescent pregnancy, Sahel countries are always below the multi-country pooled HOI for the three opportunities analysed.

Conclusions

Important inequalities exist between adolescents with different life circumstances, since the coverage rates show important decreases due to inequalities when computing the HOIs. In some SSA countries, high proportions of adolescents avoid pregnancy, but regionally (in the multi-country pooled analysis of the 29 SSA analysed) about three in ten of older adolescent girls become pregnant at a very early age. 53

A report from sub-Saharan Africa

3.2 Comparing HOIs among groups of countries The previous section of this chapter has focused on analysing access to opportunities for SSA women at a country level and at a regional level – with the multi-country pooled HOIs –, while in this section we describe the comparisons studied between different sets of countries grouped by the United Nations (UN) region and HIV prevalence. For completeness of results, the comparisons will be made between both weighted and unweighted average country groups’ HOIs. The weighted averages (multi-country pooled HOIs) show the results of the African regions populations taking into account the women’s population of each country. For example, in West Africa, individuals from Nigeria have higher weight than the ones from other countries because Nigeria is the most populated country of the region. On the other hand, unweighted average HOIs are useful to see the simple mean HOI of a group of countries without losing the “effect” of small countries, which do not have an impact on weighted averages. As presented in the chapter, both types of comparison display quite similar trends, showing that the HOIs of the countries inside a group are very similar and the average does not change much when weighting.

3.2.1 Comparing African regions In the analysis of the HOIs by country, some geographical patterns have been detected and thus, have already been explained in the previous sections. In order to verify whether these differences between country regions are real, the average HOIs (weighted and unweighted) have been computed for Central, Eastern and Western Africa. The comparison with Southern Africa would have been meaningless because there is only one Southern African country included in the study (Namibia). The countries were classified as Western, Eastern or Central, following the UN classification (Table 2.1, Chapter 2). A non-parametric test – Wilcoxon rank sum testVIII – is needed to check for these differences but it can only be applied to the unweighted sample because it does not work with weights. In Figure 3.12, small violet squares mark the indicators where at least two of the regions show significant differences with a confidence level of 90 percent.

Women of reproductive age

East Africa is the region with the highest HOI for the “knowledge of a place where to get an HIV test”, which is consistent with the fact that the majority of the countries with high HIV prevalence in Africa are in the Southern and Eastern regions. For “met need for family planning”, West Africa has a lower HOI than the other regions. In the weighted analysis, HIV differences are still clear, while family planning seems to reduce the differences between West Africa and Eastern and Central regions.

VIII

Wilcoxon rank sum test is only applicable to pairs of samples, therefore in this analysis the test has been applied to each pair of regions to compare (West-East, East-Central and Central-West).

Average HOI (%)

55 Postnatal checkup

Maternity care package

Malaria prophylaxis HIV test offered during pregnancy Infant postnatal checkup Six months of exclusive breastfeeding

Central East West

Postnatal checkup

Maternity care package

Malaria prophylaxis

HIV test offered during pregnancy

Infant postnatal checkup

Six months of exclusive breastfeeding

Delivery attended by skilled personnel

Four antenatal care visits

Having the recommended BMI

Not having anaemia

Knowledge of a place where to get an HIV test

Delivery attended by skilled personnel

Met need for family planning

Central East West

Four antenatal care visits

Central East West

Having the recommended BMI

Central East West

Not having anaemia

100

Central East West

b. Weighted

Knowledge of a place where to get an HIV test

Note: * = significant differences with a confidence level of 90 percent.

Central East West

Note: * = significant differences with a confidence level of 90 percent.

Central East West

Met need for family planning

Average HOI (%)

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.12 Average HOIs for women of reproductive age and pregnant women’s opportunities by African region

a. Unweighted

100

A report from sub-Saharan Africa

“Not having anaemia” and “having the recommended BMI” do not appear to bear significant differences between regions in the unweighted analysis. However, it is remarkable that the opportunity “not having anaemia” shows large differences in terms of HOIs between West Africa in comparison with East and Central Africa in the weighted analysis.

Pregnant women

In the unweighted analysis, “postnatal checkup”, “malaria prophylaxis during pregnancy”, “HIV test offered during pregnancy”, “infant checkup”, “exclusive breastfeeding” and the “maternity care package”, all show significant differences between African regions. West Africa outperforms the other regions in “postnatal checkup”, “malaria prophylaxis” and infant opportunities, while East Africa fares better in “HIV test offered during pregnancy”. Central Africa has high HOIs for the rest of the indicators, but the power of the comparisons is low because there are only four Central African countries included in the analysis in comparison with the 11 and 13 countries of the two other regions. Although the results do not vary significantly in the weighted analysis, in general, it can be seen that average HOIs are lower than the ones obtained in the unweighted analysis.

Older adolescent girls

The differences in the unweighted analysis are not large, although they are significant for the “met need for family planning” and the “school attendance”. In general, Central Africa outperforms the other African regions.

The results from the weighted analysis are very similar to the previous ones. The only exception is the “met need for family planning” that seems to have fewer differences between regions, meaning that accounting for country populations, people from all over SSA regions show the same HOI for this indicator.

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 3.13 Average HOIs for older adolescent girls’ indicators by African region a. Unweighted

b. Weighted

Central East West

Central East West Having never been pregnant

Having never been pregnant

Central East West Met need for family planning

Central East West

Met need for family planning

Central East West

Currently attending school

Average HOI (%)

100

Currently attending school

Average HOI (%)

100

Note: * = significant differences with a confidence level of 90 percent.

3.2.2 Comparing SSA countries with different HIV prevalence Both indicators related to HIV analysed in this report – “knowledge of a place where to be tested” and “having been offered an HIV test during pregnancy” – showed a clear trend where high HIV prevalence countries outperform in general the rest of the SSA countries in access to these HIV services. To highlight the differences between HOIs, the weighted and unweighted analyses between the countries that have an HIV prevalence of more than five and the ones with prevalence equal or lower than five have been computed. Higher HIV prevalence countries included are Kenya, Malawi, Mozambique, Namibia, Tanzania, Uganda, Zambia and Zimbabwe. The HIV prevalence rates of all countries for the corresponding survey year are listed in Appendix A. Both indicators related to HIV show higher HOIs – both in weighted and unweighted analyses – for the group of countries with HIV prevalence higher than five. The HOIs tend to be lower in the weighted analyses than in the unweighted ones. Undoubtedly, it can be stated that in countries where HIV is a major public health problem with more than five percent of the population infected, knowledge and access to HIV services is considerably better than in other SSA countries. Despite the HIV prevalence being lower in the rest of the countries included in the

A report from sub-Saharan Africa

analyses, it is not insignificant. If these countries do not spread HIV awareness and prevention among the population, there is the possibility of an increase of HIV prevalence in the future.

Figure 3.14 Average HOIs by HIV prevalence regions Knowledge of where to get an HIV test a. Unweighted

b. Weighted

80 60 40 20 0

Prevalence >5 Prevalence 5

Average HOI (%)

100

Prevalence >5 Prevalence 5

Average HOI (%)

100

HIV test offered during pregnancy a. Unweighted

b. Weighted

80 60 40 20 0

Prevalence >5 Prevalence 5

Average HOI (%)

100

Prevalence >5 Prevalence 5

Average HOI (%)

100

Note: * = significant differences with a confidence level of 90 percent.

Inequalities in women’s and girls’ health opportunities and outcomes

Key messages

On average, there are fewer inequalities both at country level and across countries for the opportunities “not having anaemia” and “having the recommended BMI” than for reproductive and maternal opportunities analysed (e.g. “met need for family planning”, HIV-related opportunities and maternity care opportunities). Reproductive health needs are related more to service provision, which bear higher inequalities than health outcomes. Anaemia and BMI are indicators of general health and are more evenly distributed within the country populations.

The maternity care package (“four antenatal care visits”, “delivery attended by skilled personnel” and “postnatal checkup”) has very low coverage with large inequalities. Individually, each of these indicators also has high inequalities, although the most unequal is “delivery attended by skilled personnel”. The low HOIs obtained for “delivery attended by skilled personnel” are mainly due to home births. “Exclusive breastfeeding” has a good coverage in SSA with low inequalities within and across countries. Older adolescent girls have poorer reproductive health opportunities than older subgroups of women of reproductive age. On average, “met need for family planning” has lower coverage and higher inequalities among adolescents than women older than 20 years. No general geographical pattern has been detected for maternal and reproductive health opportunities distribution in SSA. Specific opportunities display particular patterns, but there is no general trend across indicators. High HIV prevalence countries show lower inequalities and higher coverage rates of HIV-related opportunities than low HIV prevalence countries, suggesting that progress is possible when interventions are prioritised and sufficiently funded.

A report from sub-Saharan Africa

References 1.

Darroch, J. E., Woog, V., Bankole, A. & Ashford, L. S. ADDING IT UP : Costs and Benefits of Meeting the Contraceptive Needs of Adolescents. (2016).

Victora, C. G. et al.Countdown to 2015: a decade of tracking progress for maternal, newborn, and child survival.Lancet (London, England) (2015). doi:10.1016/ S0140-6736(15)00519-X

3. Darroch, J. E., Sedgh, G. & Ball, H. Contraceptive Technologies : Responding to Women ’ s Needs. Guttmacher Inst. 1–51 (2011). 4. Ramjee, G. & Daniels, B. Women and HIV in Sub-Saharan Africa. AIDS Res. Ther. 10, 30 (2013). 5.

Joint United Nations Programme on HIV/AIDS (UNAIDS). The gap report. (2014). doi:ISBN 978-92-9253-062-4

6. Lozano, R. et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 2095–128 (2012). 7.

WHO African Region. The African Regional Health Report 2014. (2014).

8. Frass, K. A. Postpartum hemorrhage is related to the hemoglobin levels at labor: Observational study. Alexandria J. Med. 51, 333–337 (2015). 9. WHO. Global Database on Body Mass Index. at <http://apps.who.int/bmi/index.jsp?introPage=intro_3.html> 10. Razak, F. et al. Prevalence of Body Mass Index Lower Than 16 Among Women in Low- and Middle-Income Countries. Jama 314, 2164–71 (2015). 11. Blössner, M., Onis, M. De & Organization, W. H. Malnutrition: quantifying the health impact at national and local levels. Environ. Burd. Dis. Ser. 12, 43 (2005). 12. United Nations Statistics Division. Millennium Development Goals Indicators. at <http://mdgs.un.org/unsd/mdg/Data.aspx> 13. United Nations. The Millenium Development Goals Report 2015. (2015). 14. WHO. WHO recommendations on antenatal care for a positive pregnancy experience. (2016). 15. Desai, M. et al. Epidemiology and burden of malaria in pregnancy. Lancet Infectious Diseases (2007). doi:10.1016/S1473-3099(07)70021-X 16. White, N. J. et al. Malaria. Lancet 38322, 723–35 (2014). 17. Dellicour, S., Tatem, A. J., Guerra, C. A., Snow, R. W. & Ter Kuile, F. O. Quantifying the number of pregnancies at risk of malaria in 2007: A demographic study. PLoS Med. 7, 1–10 (2010). 18. Sicuri, E. et al. Cost-effectiveness of intermittent preventive treatment of malaria in pregnancy in Southern Mozambique. PLoS One 5, (2010). 19. WHO. WHO policy brief for the implementation of intermittent preventive treatment of malaria in pregnancy April 2013 (revised January 2014 ). WHO Dep. Matern. Newborn, Child Adolesc. Heal. (2014). 60

Inequalities in women’s and girls’ health opportunities and outcomes

20. WHO. World Malaria Report 2015. (2015). doi:ISBN 978 92 4 1564403 21. The World Bank Group. World Development Indicators. (2015). at <http://data. worldbank.org/data-catalog/world-development-indicators> 22. WHO & UNICEF. Every newborn. An action plan to end preventable deaths. (2014). at <www.who.int/about/licensing/copyright_form/en/index.html> 23. WHO. Recommendations on newborn health. 24. United Nations. Transforming our world: the 2030 agenda for sustainable development. (2015). 25. Banteyerga, H. Ethiopia’s Health Extension Program: Improving Health through Community Involvement. MEDICC Rev. 13, 46–49 (2011). 26. Kimura, K., Omole, D. W. & Williams, M. Yes Africa Can: Success Stories from a Dynamic Continent. (2011). doi:10.1596/978-0-8213-8745-0 27. Oyerinde, K. et al. The status of maternal and newborn care services in Sierra Leone 8 years after ceasefire. Int. J. Gynecol. Obstet. 114, 168–173 (2011). 28. Witter, S., Wurie, H. & Bertone, M. P. The free health care initiative: how has it affected health workers in Sierra Leone? Health Policy Plan. czv006- (2015). doi:10.1093/heapol/czv006 29. McPake, B. et al. Removing financial barriers to access reproductive, maternal and newborn health services: the challenges and policy implications for human resources for health. Hum. Resour. Health 11, 46 (2013). 30. Menéndez, C., Lucas, A., Munguambe, K. & Langer, A. Ebola crisis: the unequal impact on women and children’s health. Lancet. Glob. Heal. 3, e130 (2015). 31. WHO Commission on Information and Accountability for Women’s and Children’s Health. Recommendation 2: Health indicators. at <http://www.who.int/ woman_child_accountability/progress_information/recommendation2/en/> 32. Bhutta, Z. A. et al. Countdown to 2015 decade report (2000-10): taking stock of maternal, newborn, and child survival. Lancet 375, 2032–2044 (2010). 33. WHO. Breast is always best, even for HIV-positive mothers. Bulletin of the World Health Organization (2011). 34. WHO. Intermittent preventive treatment in pregnancy (IPTp). (2015). at <http:// www.who.int/malaria/areas/preventive_therapies/pregnancy/en/> 35. Make full SRHR for young women a priority. Sex Rights Africa Network (2016). at <http://www.sexrightsafrica.net/action/make-full-srhr-young-women-priority/> 36. Nations, U. Goal 3. Sustainable Development. Knowledge Platforn (2016). at <https://sustainabledevelopment.un.org/sdg3> 37. Mombo-Ngoma, G. et al. Young adolescent girls are at high risk for adverse pregnancy outcomes in sub-Saharan Africa: an observational multicountry study. BMJ Open 6, e011783 (2016)

A report from sub-Saharan Africa

38. UNFPA. Girlhood, not motherhood preventing adolescent pregnancy. (2015). 39. The International Community Must Prioritize Young Women’s Contraceptive Needs. Global Daily (2016). at <http://globaldaily.com/the-international-community-must-prioritize-young-womens-contraceptive-needs/> 40. United Nations. The Global Strategy for Women’s, Children’s and Adolescents’ Health (2016-2030). Survive thrive transform. (2015).

ÂŠPau Fabregat

Chapter 4. What explains inequalities in health for women of reproductive age?

A report from sub-Saharan Africa

This chapter describes the specific circumstances and to what extent they contribute to inequality of opportunities from different perspectives. Firstly, the contributions of different circumstances to inequality at country level are examined through the simple averages of the decomposition results (see specific country results in Appendix A). An additional analysis is performed dividing countries into two groups by human immunodeficiency virus (HIV) prevalence, to examine the differences in circumstances’ contributions for the HIV-related opportunities. Secondly, a multi-country approach is presented through a multi-country pooled data analysis of the Human Opportunity Index (HOI) decompositions. These results thus complement the country level analyses of circumstances by providing insights on how maternal and reproductive health services and outcomes among sub-Saharan African (SSA) women are associated with differences in their individual and household characteristics, when women from all countries are considered as a single group. Finally, a more in-depth analysis is presented to examine which circumstances drive inequality of opportunities among the older adolescent girls’ subgroup.

4.1 Explaining inequality at country level This section addresses how different circumstances contribute to inequality of opportunity in maternal and reproductive health indicators and outcomes at country level. The results of the analysis are summarised by showing the unweighted average contribution (simple mean) of each circumstance across all countries to the dissimilarity index (D-index) for each opportunity. In other words, the results show the contribution of each circumstance to the inequality of a certain health opportunity, averaged across all countries, where every country is treated equally, regardless of its share in the population of women. Presenting the averages rather than single country results enables us to highlight key patterns in how circumstances matter for inequality of different opportunities. This would be difficult if more disaggregated country-level information was shown, given the large number of decompositions included (29 countries, 15 opportunities, and multiple circumstances)IX. Another important consideration to take into account: Shapley decomposition results show the relative contribution of each circumstance to the D-indexX for an opportunity, but not the “absolute amount of inequality” that each circumstance is generating. For example, in the case of opportunities with a low D-index, a circumstance that may significantly contribute to the D-index could be responsible for a very small “amount of inequality” in terms of magnitude. Figure 4.1 shows the differences across opportunities in a graph that displays the unweighted average D-index for all countries. The average D-index shows that there are large differences in the D-index across opportunities, meaning that the magnitude of inequality is not the same for the different indicators analysed. This is relevant for interpreting the results on the

Decomposition results for all countries are accessible to interested readers in Appendix A.

Using the definition explained earlier, the D-index measures inequality between groups differentiated by circumstances.

Inequalities in women’s and girls’ health opportunities and outcomes

relative contributions of circumstances to inequality of opportunity, since the same percentage contribution of a circumstance to inequality would have different implications for different opportunities. Notably, the most unevenly distributed health opportunities are (Figure 4.1): “currently attending school” (D-index: 26 percent) – that applies exclusively to the older adolescents group –, “maternity care package” that includes “four antenatal care visits”, “delivery attended by skilled personnel” and “postnatal checkup” (D-index: 25 percent), “met need for family planning“ (older adolescents group) (D-index: 19 percent) and “delivery attended by skilled personnel” (D-index: 17 percent).

Figure 4.1 Average D-index by opportunity (unweighted) 30 20

Six months of exclusive breastfeeding

Infant checkup after delivery

HIV test offered during pregnancy

Malaria prophylaxis

Maternity care package

Postnatal checkup

Delivery attended by skilled personnel

Four antenatal care visits

Having the recommended BMI

Not having anaemia

Knowledge of a place where to get an HIV test

Met need for family planning

Currently attending school (Adolescents)

Having never been pregnant (Adolescents)

10 Met need for family planning (Adolescents)

D-index (%)

Note: The average D-indices (inequality) for the group of countries are calculated as the unweighted or simple average of the D-indexes for each country.

A report from sub-Saharan Africa

Box 5. D-index: country cases To demonstrate the average D-indices with some examples, a selection of countries and opportunities with their D-index is presented, which in some cases are far from the average D-index. The different results displayed highlight the importance of focusing also on the particular results of each country to take into account their specific features. Country

D-index (%) Average D-index

Example 1

Example 2

Currently attending school

Gabon

Niger

Not having anaemia

Zimbabwe

The Gambia

Delivery attended by skilled personnel

Rwanda

Nigeria

4.1.1

Women of reproductive age (15-49 years old) and pregnant women Country level inequalities are largely explained by differences in household wealth (i.e. which quintile of wealth in a country women belong to), educational level and area of residence (urban/rural) (Figure 4.2). In most countries, a pattern is observed where women who are more educated, living in urban areas and in relatively wealthier households have an advantage for almost all indicators. The only exception to this situation is for the body mass index (BMI); the wealthiest and most educated women are the ones with inadequate BMIs, mostly being overweight. A similar trend is observed for opportunities with the highest levels of inequality (i.e. “maternity care package” and “delivery attended by skilled personnel”), where household wealth, area of residence and the woman’s educational level are the most significant contributors to inequality, in respective order of importance (Figure 4.2). Marital status is one of the most significant contributors to inequality for some health indicators and outcomes: “malaria prophylaxis during pregnancy”, “met need for family planning”, “not having anaemia” and “exclusive breastfeeding”. Single women have a significant advantage over married women in some health indicators such as “not having anaemia” or “met need for family planning”, whereas married women have better malaria prophylaxis during pregnancy uptake. Overall inequality is high only for “met need for family planning” and quite low for the other mentioned indicators (Figure 4.1), suggesting that marital status contributes to a significant share of inequality only in the case of access to family planning. Certain circumstances stand out for specific indicators. In general, sex of the household head, number of children, age and religion are not significant contrib-

Inequalities in women’s and girls’ health opportunities and outcomes

utors to inequality. For example, age does not seem to be a major driver of inequalities in health for women of reproductive age except in the case of “having the recommended BMI”, where older women tend to have poorer BMI scores. The low level of overall inequality for this indicator (Figure 4.1) suggests that the differences in BMI attributable to age are not large. However, averages can conceal significant variations between different countries (Box 6).

Figure 4.2 Women of reproductive age and pregnant women: circumstances’ contributions to the D-index

30 20 10 0 Weath index Educational level Area Age Marital status Religion Number of children Sex of the household head

average % contribtion to the D-index

c. Not having anaemia 40 30 20 10 0 Weath index Educational level Area Age Marital status Religion Number of children Sex of the household head

average % contribtion to the D-index

b. Knowledge of where to get an HIV test 40 30 20 10 0 Weath index Educational level Area Age Marital status Religion Number of children Sex of the household head

d. Having the recommended BMI average % contribtion to the D-index

average % contribtion to the D-index

a. Met need for family planning

40 30 20 10 0 Weath index Educational level Area Age Marital status Religion Number of children Sex of the household head figure continues next page

A report from sub-Saharan Africa

Figure 4.2 Women of reproductive age and pregnant women: circumstancesâ€™ contributions to the D-index (continued)

20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

average % contribtion to the D-index

g. Postnatal checkup 40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

average % contribtion to the D-index

i. Malaria prophylaxis 40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

average % contribtion to the D-index

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

h. Maternity care package average % contribtion to the D-index

f. Delivery attended by skilled personnel

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

j. HIV test offered during pregnancy average % contribtion to the D-index

average % contribtion to the D-index

e. Four entenatal care visits

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head figure continues next page

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 4.2 Women of reproductive age and pregnant women: circumstances’ contributions to the D-index (continued)

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

l. Six months of exclusive breastfeeding

average % contribtion to the D-index

k. Infant checkup within two months after delivery

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

Note: The average contributions of circumstances to inequality of opportunity for the group of countries are calculated as the unweighted or simple averages (across all countries) of Shapley decompositions of the D-index for that opportunity.

Box 6. Women of reproductive age and pregnant women’s opportunities: country cases To illustrate with examples the average results previously presented, below is a selection of countries and opportunities that follow the average trend or, in contrast, show very distant results from the average. The variability of the results indicates the need to review the results of each country to take into account their specific features (see Appendix A). Not having anaemia: this opportunity shows significantly different results across countries. For example, Ethiopia and The Gambia do not share any similarity. Ethiopia The Gambia 0 Weath index Educational level

20 Area Marital status

40 60 % contribution to the D-index Religion Sex of the household head

100 Number of children Age

box continues next page

A report from sub-Saharan Africa

Box 6. Women of reproductive age and pregnant women’s opportunities: country cases (continued)

Met need for family planning: while Uganda seems to represent the average results obtained for SSA quite well, Ghana shows different contributors to inequality; marital status being the main one. Ghana Uganda 0 Weath index Educational level

40 60 % contribution to the D-index

Area Marital status

Religion Sex of the household head

100 Number of children Age

Maternity care package: this opportunity shows quite homogenous results across countries. Zambia’s results reflect the regional average of the 29 SSA countries analysed, whereas Rwanda is the most notable exception with the number of children being the main contributor to inequality. Rwanda Zambia 0

100

% contribution to the D-index Weath index Educational level

Area Marital status

Religion Sex of the household head

Number of children Age at delivery

Malaria prophylaxis during pregnancy: Nigeria is a country that in many cases is an outlier because of the important contribution of religion to inequality. Intermittent Preventive Treatment of malaria in Pregnancy (IPTp) is an example. While religion is not relevant for Mali’s inequality, for Nigeria it is the main contributor to the D-index. Mali Nigeria 0

100

% contribution to the D-index Weath index Educational level

Area Marital status

Religion Sex of the household head

Number of children Age at delivery

Inequalities in women’s and girls’ health opportunities and outcomes

4.1.2 Older adolescent girls (15-19 years old) The opportunities analysed for this age subset are among the most unevenly distributed health opportunities of the report (Figure 4.1): “currently attending school” (D-index: 26 percent), “met need for family planning” (D-index: 19 percent) and “having never been pregnant” (D-index: 15 percent). In general, the main drivers of inequality for the older adolescent group (Figure 4.3) are: marital status, which ranks first for all the opportunities examined (40 percent, 39 percent and 69 percent of the D-index), followed by wealth index, occupation and area of residence. The main circumstance contributing to inequality in the case of older adolescent girls’ pregnancies, i.e. the “having never been pregnant” opportunity, arises from marital status (i.e. being married) that accounts for as much as 70 percent of the D-index (Figure 4.3). Adolescent pregnancies are more prevalent among married adolescent girls than among those who are unmarried. Occupation is an especially relevant driver of inequality with regard to school attendance (28 percent of the overall D-index for “currently attending school”). Among older adolescent girls who are employed, school enrolment rate is lower than for girls who are not working. A more in-depth analysis of older adolescent girls with regard to marital status is presented at the end of this chapter.

Figure 4.3 Older adolescent girls’ opportunities: circumstances’ contributions to the D-index b. Met need for family planning

30 20 10 0 Marital status Occupation Religion Weath index Area Sex of the household head

average % contribution to the D-index

a. Currently attending school

30 20 10 0 Marital status Occupation Religion Weath index Area Sex of the household head figure continues next page

A report from sub-Saharan Africa

Figure 4.3 Older adolescent girls’ opportunities: circumstances’ contributions to the D-index (continued) c. Having never been pregnant 70 60

average % contribution to the D-index

50 40 30 20 10 0 Marital status Occupation Religion Weath index Area Sex of the household head

Box 7. Older adolescent girls’ opportunities: country cases School attendance: while for Zimbabwe marital status is the most important contributor to the inequality (D-index) followed by occupational status, Benin shows the inverse, with occupation as the main source of inequality. Benin Zimbabwe 0 Marital status Weath index

40 60 % contribution to the D-index Occupational status Area

100

Religion Sex of the household head box continues next page

Inequalities in women’s and girls’ health opportunities and outcomes

Box 7. Older adolescent girls’ opportunities: country cases (continued)

Having never been pregnant: in both examples marital status stands for the main contributor to the D-index. In the case of Togo occupational status also seems to contribute significantly to inequality, while in Malawi its contribution is minor. Malawi Togo 0 Marital status Weath index

40 60 % contribution to the D-index Occupational status Area

100

Religion Sex of the household head

The variability of the results indicates the need to review the results of each country to take into account their specific features (see Appendix A).

4.1.3 HIV-related indicators HIV-related indicators have been further analysed in order to ascertain possible differences for inequality in countries with different HIV prevalence rates. Thus, the study sample has been divided into countries with high HIV prevalence (those with an HIV prevalence rate of more than five percent of the population infected) and low prevalence (countries below or at the five percent threshold) (Figure 4.4), which might in turn have implications for the design of public health policies and programmes in countries with different epidemic settings.

Knowledge of where to get an HIV test

Education is a key contributor to inequality in high and low HIV prevalence countries, with D-index values of 25 percent. In high prevalence countries, the relative contribution of marital status is much higher (24 percent) than for low prevalence countries (14 percent). Further research would be needed to understand why marital status is more important than other circumstances (i.e. wealth, area of residence and other factors) in explaining differences in knowledge of where to get an HIV test (married women have better opportunities than those who are unmarried), and why this is so different as compared to low prevalence countries. Household wealth status and area of residence (urban/rural) contribute less to inequality in high prevalence countries – 14 percent and eight percent of the D-index compared to 24 percent and 18 percent of the D-index, respectively – than in low prevalence countries.

A report from sub-Saharan Africa

HIV test offered during pregnancy

The circumstances that most contribute to inequality are similar across high and low HIV-prevalence countries. Womenâ€™s wealth, area of residence and education, are the main drivers for both groups of countries, albeit with some differences in the order of importance.

Figure 4.4 HIV prevalence country groups: circumstancesâ€™ contributions to the D-index Knowledge of where to get an HIV test

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

b. HIV prevalence > 5 average % contribtion to the D-index

average % contribtion to the D-index

a. HIV prevalence < 5

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

HIV test offered during pregnancy

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

b. HIV prevalence > 5 average % contribtion to the D-index

average % contribtion to the D-index

a. HIV prevalence < 5

40 30 20 10 0 Weath index Educational level Area Age at delivery Marital status Religion Number of children Sex of the household head

Inequalities in women’s and girls’ health opportunities and outcomes

Conclusions

Overall, for SSA women of reproductive age (15-49 years old), wealth, area of residence (rural/urban) and the women’s educational level are the leading contributors to inequality in maternal and reproductive health indicators, including those with the highest levels of inequality such as access to a basic maternity care package and having the delivery attended by skilled personnel. Marital status is the main contributor to inequalities for older adolescent girls (15-19 years old), as well as for several opportunities for adult women, most notably, access to family planning services and malaria prophylaxis uptake. Married adolescents have fewer opportunities for reproductive health and education opportunities. However, for adult women marital status can contribute positively for some indicators (“malaria prophylaxis during pregnancy”) and negatively for others (“met need for family planning”). Other circumstances (i.e. number of children, age, sex of the household head and religion) are of marginal importance. However, there are significant differences across countries underlying the averages. For example, in Nigeria, religion stands out as the main contributor to inequality in “malaria prophylaxis during pregnancy”.

4.2 Explaining inequalities across countries: a multi-country pooled analysis This section describes the sources of inequalities among the population of women of reproductive age in SSA from a different angle. The analyses performed include all country samples in the same pool, all weighted by each country’s share of women’s population to the total. The results are subject to the caveat that the estimated contributions of circumstances could be affected by the presence of country-specific factors, correlated with the observed circumstances. Those could be unobservable but systematically present in some countries. For example, religion in a subset of countries could be geographically distributed in a way that results in a confounding factor. While the above limitation calls for caution in interpreting results, the findings are still instructive. The results are best seen as providing a picture of how maternal and reproductive health services and outcomes among SSA women are associated with differences in their characteristics, when women from all countries are considered as a single group. In performing the pooled analysis, the circumstances measured by the wealth index also have to be interpreted with caution. The wealth quintiles for the pooled analysis remain the same as those defined for each country analysis. This fact

A report from sub-Saharan Africa

implies that wealth in the multi-country pooled analyses indicates the relative wealth position (in terms of quintile) of an individual woman in her countryXI. Finally, as in the previous section, the D-indices of all opportunities have to be taken into account before interpreting the results because Shapley decomposition results are relative contributions to the inequality. For this analysis, the D-indices used are those computed from the multi-country weighted analysis pooling all samples included in the report (Figure 4.5). Although results do not differ significantly from the country level D-indices (Figure 4.1), some features merit attention. First, there is a marked reduction in the D-index (inequality) of “six months of exclusive breastfeeding” opportunity that results in it scoring the lowest inequality across SSA. Second, there is a significant increase in the D-index (inequality) of the “infant checkup after delivery” and “maternity care package” opportunities. This last one shows a D-index above 30 percent.

Figure 4.5 Average D-index by opportunity (multi-country pooled analysis)

30 20

Six months of exclusive breastfeeding

Infant checkup after delivery

HIV test offered during pregnancy

Malaria prophylaxis

Maternity care package

Postnatal checkup

Delivery attended by skilled personnel

Four antenatal care visits

Having the recommended BMI

Not having anaemia

Knowledge of a place where to get an HIV test

Met need for family planning

Currently attending school (Adolescents)

Having never been pregnant (Adolescents)

10 Met need for family planning (Adolescents)

D-index (%)

Note: The average D-indices (inequality) for the group of countries are calculated pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

The principal component analysis (PCA) used by the Demographic Health Surveys (DHS) programme to calculate the wealth index assigns a different number to each individual depending on the distribution of assets in the sample10. In other words, a person from Gabon could be as wealthy as another from Zimbabwe, but this methodology does not assign them the same wealth index value. By generating wealth quintiles, a country’s population is classified into a relative wealth scale. In the multi-country pooled analysis, the wealth quintiles have been left unchanged, which means that wealth as a circumstance has to be understood as the relative wealth position of the household (that the woman belongs to) in her country rather than the value of assets owned by the household.

Inequalities in women’s and girls’ health opportunities and outcomes

4.2.1 Women of reproductive age (15-49 years old) and pregnant women As previously observed for the country level analysis, wealth, education and area of residence (urban/rural) are the most important contributors to inequality for the majority of the opportunities for the subgroup of women of reproductive age and pregnant women examined (Figure 4.6). In general, the contributions of religion and number of children to inequality tend to be high in the pooled analysis (Figure 4.6). Possible explanations could be that these two circumstances are correlated with country-specific factors, since they appear more concentrated in some countries than in others or the contribution of these two factors to inequality actually reflect, at least in part, the effects of other circumstances that are country-specific but unobservable in the analysis. Religion accounts for a large share of inequality in the indicators of “not having anaemia” and “malaria prophylaxis during pregnancy” (Figure 4.6). However, since average D-index is very low in both cases (see Figure 4.5), the actual amount of inequality attributable to religion is quite small.

Box 8. The role of religion in women’s health indicators’ inequalities Religion is not a typical sociodemographic characteristic included in the studies on anaemia or prevention of malaria infection in pregnancy in the SSA region. Therefore, there is scant evidence available regarding the contribution of religion to health inequalities. However, many studies conducted in India and South-East Asia highlighted religion as a possible risk factor for anaemia and found significant differences in religion between groups of the population with and without anaemia1–4. Regarding IPTp uptake, few studies use religion as a covariate in the analysis of the uptake of malaria prophylaxis among pregnant women. In general, the scarce evidence available did not find a statistically significant relationship between religion and IPTp coverage5–7. In contrast, a systematic review of the literature showed that beliefs and religious practices are barriers to access, delivery and use of preventive interventions against malaria during pregnancy8. A possible explanation of the relevance of religion for anaemia and malaria indicators may be the geographical and country distribution of religious groups in malaria endemic countries overlapped with other unobservable factors. Some countries such as Guinea, Sierra Leone or The Gambia are mostly Muslim, while in others such as, Zambia, Cameroon, Congo or Malawi, Muslim religion is less prominent. In the majority of the SSA countries, there are also communities and regions where animism predominates. Another plausible explanation that could explain the high contribution of religion to inequality in these health indicators is the geographical distribution of box continues next page

A report from sub-Saharan Africa

Box 8. The role of religion in women’s health indicators’ inequalities (continued)

religious groups within countries where more than one religion is prevalent. In Nigeria, for example, Christian communities are placed mostly in the southern region, while most Muslims live in the north9. This regional distribution of religious groups combined with other factors – such as different climates, altitudes, types of crops and nutritional habits, or different malaria incidence/prevalence – could explain differences in anaemia prevalence and use of malaria preventive strategies across countries with different religions. In some cases, these relationships may be due to chance, while in others, particular religious practices and beliefs might explain the results. More in depth analysis at country level is needed to understand the role religion is playing, not only regarding anaemia prevalence and malaria prophylaxis coverage, but also with respect to other health opportunities (e.g. maternity care indicators, family planning, etc.).

Figure 4.6 Multi-country pooled analysis for women of reproductive age and pregnant women: circumstances’ contributions to the D-index Met need for family planning Knowledge of where to get an HIV test Not having anaemia Having the recommended BMI Four antenatal care visits Delivery attended by skilled personnel Postnatal checkup Maternity care package Malaria prophylaxis during pregnancy HIV test offered during pregnancy Infant checkup within two months after delivery Six months of exclusive breastfeeding 0

Weath index Educational level

Area Marital status

40 60 80 % contribution to the D-index

Religion Sex of the household head

100

Number of children Age/Age at delivery

Note: The average contributions of circumstances to inequality of opportunity for the group of countries are calculated pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

4.2.2 Older adolescent girls (15-19 years old) The results obtained from the pooled analysis for the older adolescent girls group are similar to those previously observed for the country level analysis, with the most significant variation from the earlier results being the greater role of religion as a contributor to inequality (Figure 4.7). This finding could be related to certain religions being concentrated in a few countries, where unobserved country-specific factors likely affect access to services and outcomes in maternal and reproductive health for this subset of girls.

Figure 4.7 Multi-country pooled analysis for older adolescent girls: circumstancesâ€™ contributions to the D-index Met need for family planning Having never been pregnant Currently attending school 0 Marital status Weath index

40 60 80 % contribution to the D-index

Occupational status Area

100

Religion Sex of the household head

Conclusions

In general, results are similar to those obtained in the country level analysis, with a few exceptions. Wealth, education and area of residence are the most important contributors to inequality for adult women, while for older adolescent girls marital status is the main contributor. Religion has a more important role in the pooled analysis and appears to be a relevant contributor to inequality across SSA.

A report from sub-Saharan Africa

4.3 Adolescent girls and marital status: the major source of inequalities Analyses of older adolescent girls’ opportunities reveal that a major share of inequalities in this particular age group is attributable to differences in marital status. The large contribution of marital status indicates that the drivers of inequality are likely to be very different for groups with different marital status, which in turn makes the interpretation of the contributions from other circumstances difficult. To account for these differences, the same analyses (country level and multi-country pooled analysis) have been performed in this section for older adolescent girls by dividing them into two groups: adolescent girls who are married or “in union”XII (i.e. living with the partner, widowed, divorced or separated), and adolescent girls who were never in union.

4.3.1 Country level analysis results Wealth appears to be the highest contributor to inequalities between both groups of older adolescent girls (i.e. “in union” and “never in union”) in most cases (Figure 4.8). The only exception to this pattern is observed for the group of “never in union” adolescents with regard to the “school attendance” opportunity, where differences in occupation contribute the most to the D-index. In all cases, there are differences in the order of importance of the contributors to inequality between the two groups of older adolescent girls, with the most significant differences seen for inequality in “school attendance”, where occupation accounts for as much as 46 percent of the D-index (inequality) for the “never in union” adolescents. For the other two opportunities (i.e. “having never been pregnant” and “met need for family planning”), an important difference between the two groups is that “religion” is an important contributor to inequalities among “in union” adolescents, while “sex of the household head” is relatively more important for “never in union” adolescents.

XII

Women currently or previously married or in union.

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 4.8 Country level analysis – Older adolescent girls’ opportunities by marital status: circumstances’ contributions to the D-index Currently attending school b. In union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion

average % contribtion to the D-index

a. Never in union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion

Met need for family planning b. In union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion

average % contribtion to the D-index

a. Never in union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion figure continues next page

A report from sub-Saharan Africa

Figure 4.8 Country level analysis – Older adolescent girls’ opportunities by marital status: circumstances’ contributions to the D-index (continued) Having never been pregnant b. In union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion

average % contribtion to the D-index

a. Never in union

30 20 10 0 Weath index Area Occupational status Sex of the household head Religion

Note: The average contributions circumstances to inequality of opportunity for the group of countries are calculated as the unweighted or simple averages (across all countries) of Shapley decompositions of the D-index for that opportunity.

4.3.2 Multi-country pooled analysis results In order to highlight the impact of marital status on the inequalities computed with the pooled data, weighted by the number of women in each country, the HOI and the D-index (inequality) are shown for each of the three opportunities applicable to adolescents for both subgroups (Figure 4.9). The HOIs for the “never in union” subgroup of adolescents are higher than those for the “in union” subgroup for all opportunities examined. Notably, there is a difference of almost 60 percent in adolescent pregnancies, over 50 percent in school attendance and about 25 percent in family planning needs between both groups of girls. Conversely, the D-index shows more inequalities among the subgroup of married or “in union” adolescents for all three opportunities – “school attendance”, “met need for family planning” and “having never been pregnant”; 38 percent, 18 percent and 9 percent of the D-index, respectively, compared to ten percent, eight percent and one percent of the “never in union” subgroup.

Inequalities in women’s and girls’ health opportunities and outcomes

Figure 4.9 Multi-country pooled analysis – Older adolescent girls’ opportunities by marital status: HOI and D-index by opportunity Currently attending school

Met need for family planning

a. Never in union

b. In union

a. Never in union

b. In union

100

HOI (%) D-index (%)

Having never been pregnant a. Never in union

b. In union

100

0 HOI (%) D-index (%)

HOI (%) D-index (%)

Note: The average HOIs and D-indices (inequality) for the group of countries are calculated pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

A report from sub-Saharan Africa

Figure 4.10 Multi-country pooled analysis - Older adolescent girls’ opportunities by marital status: circumstances’ contributions to the D-index Currently attending school

Met need for family planning

Having never been pregnant

Never in union In union Never in union In union Never in union In union 0

Weath index Occupational status

Area Religion

100

Sex of the household head

Shapley decompositions of the D-indices compared in Figure 4.10 show differences between the two subgroups that are qualitatively similar to what was observed in the averages obtained from country-specific analysis. For two of the three indicators (“met need for family planning” and “having never been pregnant”), religion is an important contributor to inequality among “in union” adolescents but not for the other group. For inequality in “school attendance”, occupation is an important factor among the “never in union” group but not for the other group. A possible explanation of this finding could be that those girls who do not have a partner work for themselves and their families, whereas “in union” girls do not work outside home in many cases. Given the caveats discussed earlier, the results cannot be interpreted as the direct effect of religion on inequality of opportunities among older adolescent girls. What they show quite clearly are significant differences in the opportunities of older adolescent girls by religion, with these differences being much higher among married adolescents when it comes to meeting needs for family planning and the likelihood of being pregnant. Whether this indicates differential access to family planning services among married adolescent girls of different religions, is a question that would merit more in-depth analysis.

Inequalities in women’s and girls’ health opportunities and outcomes

Conclusions

In general, wealth is the main contributor to inequality for “in union” and “never in union” adolescents at country level. Religion appears to be an important driver of inequalities among “in union” adolescents in the multi-country pooled analysis. For all three indicators analysed, the HOI is always higher for “never in union” adolescents, and the D-index (inequality) is always lower for the same group, meaning that older adolescent girls that have ever been in union (married, living with their partner, etc.) have large disadvantages in access to reproductive health and education opportunities.

Key messages

On average, wealth and related circumstances such as education and area of residence are the main sources of inequality for women of reproductive age health opportunities at country level in SSA.

In the multi-country pooled analysis of women from all countries, a more prominent role of religion and a reduced contribution of wealth are observed. For older adolescent girls’ education and reproductive health indicators and outcomes (i.e. access to contraception, pregnancy), the main source of inequalities is marital status. In general, once marital status is controlled for, wealth becomes the first contributor to inequalities al country level. For the “school attendance” opportunity, after adjusting for marital status, employment status of adolescents is the main source of inequality. For older adolescents, multi-country pooled analysis also shows a significant contribution of religion across countries. Religion is more associated with inequalities for older adolescent girls who are married or have ever been in union, particularly with regards to access to family planning and becoming pregnant.

A report from sub-Saharan Africa

References 1.

Bharati, P., Shome, S., Chakrabarty, S., Bharati, S. & Pal, M. Burden of Anemia and its socioeconomic determinants among adolescent girls in India. Food Nutr. Bull. 30, 217–226 (2009).

2. Dey, S., Goswami, S. & Dey, T. Identifying predictors of childhood anaemia in North-East India. J. Heal. Popul. Nutr. 31, 462–470 (2013). 3. Bentley, M. E. & Griffiths, P. L. The burden of anemia among women in India. Eur. J. Clin. Nutr. 57, 52–60 (2003). 4. Choudhary, A., Moses, P. D., Mony, P. & Mathai, M. Prevalence of anaemia among adolescent girls in the urban slums of Vellore, south India. Trop. Doct. 36, 167–9 (2006). 5.

Choonara, S., Odimegwu, C. O. & Elwange, B. C. Factors influencing the usage of different types of malaria prevention methods during pregnancy in Kenya. Afr. Health Sci. 15, 413–419 (2015).

6. Sangaré, L. R. et al. Determinants of use of intermittent preventive treatment of malaria in pregnancy: Jinja, Uganda. PLoS One 5, 1–7 (2010). 7.

Dako-Gyeke, M. & Kofie, H. M. Factors Influencing Prevention and Control of Malaria among Pregnant Women Resident in Urban Slums, Southern Ghana. Afr. J. Reprod. Health 19, 44–53 (2015).

8. Hill, J. et al. Factors Affecting the Delivery, Access, and Use of Interventions to Prevent Malaria in Pregnancy in Sub-Saharan Africa: A Systematic Review and Meta-Analysis. PLoS Med. 10, (2013). 9. Okehie-Offoha, M. U. & Sadiku, M. N. O. Ethnic and Cultural Diversity in Nigeria. (1996). 10. Rutstein, S. O. & Johnson, K. The DHS Wealth Index. DHS Comparative Reports No.6.

Â©Pau Fabregat

Chapter 5. Conclusions

A report from sub-Saharan Africa

5.1 Conclusions Improvements in maternal, neonatal and reproductive health can only be achieved if access and quality of care are ensured for all women and girls, including those who are currently underserved or excluded from health systems1. In the transition to the new Sustainable Development Goals (SDGs) era, there is a need to focus more deliberately on improving the health of women, children and adolescents from an equity standpoint. The aim of this report is to highlight the sources of unequal and preventableÂ health differences among women and adolescent girls in sub-Saharan Africa (SSA). It presents findings based on the most recent available information that can be used to inform policy at different levels. Additionally, the report introduces new metrics such as the Human Opportunity Index (HOI), which provides a novel approach to understanding the constraints and opportunities for achieving equity in maternal and reproductive health.

Table 5.1 Country level average HOIs and multi-country pooled HOIs HOIs (%) Opptunities

Women of reproductive age (15-49 years old)

Pregnant women

Older adolescent girls (15-19 years old)

Country level average

Multi-country pooled average

Not having anaemia

45.47

62.55

Having the recommended BMI

58.76

62.06

Met need for family planning

46.26

46.14

Knowledge of a place where to get an HIV test

66.48

60.90

Four antenatal care visits

46.20

34.21

Delivery attended by skilled personnel

53.10

36.96

Postnatal checkup

52.77

40.90

Maternity care package

26.08

15.87

Malaria prophylaxis during pregnancy

47.45

42.52

HIV test offered during pregnancy

61.26

57.51

Infant checkup within two months after delivery

45.51

31.65

Six months of exclusive breastfeeding

76.67

78.09

Met need for family planning

37.80

40.01

Having never been pregnant

63.72

66.24

Currently attending school

39.45

40.17

Note: BMI = Body Mass Index, HIV = Human Immunodeficiency Virus. The country level average is the average of individual country HOIs. The multi-country pooled average is the predicted HOI for the group of countries that has been calculated by pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

Inequalities in women’s and girls’ health opportunities and outcomes

In general, the findings of this analysis are in accordance with those of other articles and reports2,3: Overall, coverage of services is low; inequalities are driven by income, education and location (urban and rural) for most indicators across women in SSA. Baseline coverage is even lower, and inequalities higher, for those interventions that require higher provider-patient interaction (e.g. antenatal care or delivery attended by skilled personnel) than for interventions that could be delivered through strategies outside the health system (e.g. exclusive breastfeeding or HIV information and testing) (Table 5.2). This report provides a novel approach to understand inequalities of opportunities in Reproductive, Maternal, Newborn, Child and Adolescent Health (RMNCAH) by simultaneously analysing all the factors – wealth, education, place of residence and others such as, religion, marital status or age –, that may affect inequality and their relative contributions to it. The findings suggest that wealth, educational level and area of residence (urban/rural) are the three main variables associated with inequality of access to health care by women. Notably, these socio-economic barriers to demand are often interlinked (Figure 5.3). The HOI approach enables the establishment of associations between individual and household circumstances and inequality. It also allows quantification of the different levels of inequalitiesXIII that exist within and across countries and among different opportunities, as it is shown in Appendix A, where country-level specific data complement the regional findings. It also complements other existing data, such as the Countdown to 2015 reports4.

Table 5.2 Opportunities ranked by inequality level (multi-country pooled analysis) D-index (%)

D-index (%)

Maternity care package

32.67

Having never been pregnant

15.30

Currently attending school

23.73

Delivery attended by skilled personnel

11.46

23.32

Knowledge of a place where to get an HIV test

Four antenatal care visits

19.26

HIV test offered during pregnancy

11.36

Postnatal checkup

17.17

Malaria prophylaxis during pregnancy

7.34

Infant checkup after delivery

15.89

Met need for family planning (older adolescents)

Having the recommended BMI

6.51

15.39

Not having anaemia

4.46

Met need for family planning (20-49 years old)

15.37

Six months of exclusive breastfeeding

1.11

Note: D-index= dissimilarity index. The multi-country by pooling average is the predicted HOI for the group of countries that has been calculated by pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

XIII

Although the existence of inequalities in maternal and reproductive health opportunities is well known, this report highlights the different levels of inequalities that exist among countries and among different opportunities.

A report from sub-Saharan Africa

Table 5.3 Three main contributors to inequality for each opportunity and subgroup of women

Pregnant women

Women of reproductive age (15-49 years old)

Opportunities Not having anaemia

Having the recommended BMI

Met need for family planning

Knowledge of a place where to get an HIV test Four antenatal care visits

Delivery attended by skilled personnel

Postnatal checkup

Maternity care package

Malaria prophylaxis during pregnancy

HIV test offered during pregnancy

Infant checkup within two months after delivery Six months of exclusive breastfeeding

Country level

Multi-country

Wealth index

27.31

Religion

38.32

Marital status

16.79

Educational level

21.14

Educational level

15.18

Wealth index

13.85

Wealth index

26.38

Age

26.27

Age

23.11

Wealth index

25.65

Area (urban/rural)

17.60

Area (urban/rural)

22.39

Wealth index

22.15

Educational level

23.59

Marital status

19.47

Wealth index

20.29

Educational level

18.36

Religion

15.05

Educational level

25.20

Educational level

30.42

Wealth index

21.60

Wealth index

22.72

Marital status

17.11

Religion

18.79

Wealth index

31.83

Educational level

30.71

Educational level

20.25

Area (urban/rural)

24.91

Area (urban/rural)

17.24

Wealth index

23.20

Wealth index

32.45

Educational level

27.20

Area (urban/rural)

24.17

Wealth index

21.81

Educational level

18.36

Area (urban/rural)

20.40

Wealth index

32.62

Educational level

26.27

Area (urban/rural)

21.64

Area (urban/rural)

24.81

Educational level

19.61

Wealth index

22.13

Wealth index

32.26

Educational level

27.60

Area (urban/rural)

22.24

Area (urban/rural)

26.82

Educational level

20.58

Wealth index

24.23

Wealth index

26.09

Religion

65.77

Marital status

18.78

Educational level

15.88

Educational level

15.19

Marital status

6.39

Wealth index

30.67

Educational level

26.60

Area (urban/rural)

23.64

Wealth index

22.43

Educational level

21.09

Area (urban/rural)

18.81

Wealth index

31.40

Area (urban/rural)

23.94

Area (urban/rural)

17.50

Educational level

23.24

Educational level

16.65

Wealth index

19.06

Wealth index

25.09

Wealth index

28.56

Marital status

19.92

Educational level

25.90

Educational level

15.75

Number of children

15.59

table continues next page

Inequalities in women’s and girls’ health opportunities and outcomes

Table 5.3 Three main contributors to inequality for each opportunity and subgroup of women (continued)

Older adolescent girls (15-19 years old)

Opportunities Met need for family planning

Having never been pregnant

Currently attending school

Country level

Multi-country

Marital status

38.98

Marital status

33.31

Wealth index

24.15

Wealth index

22.61

Area (urban/rural)

16.03

Religion

20.99

Marital status

69.11

Marital status

75.24

Wealth index

10.54

Wealth index

8.66

Occupation

9.91

Occupation

5.69

Marital status

40.07

Marital status

47.63

Occupation

28.19

Occupation

19.69

Wealth index

15.43

Religion

12.26

Note: The country level average is the average of individual country HOIs. The multi-country pooled average is the predicted HOI for the group of countries that has been calculated by pooling all country samples and weighting them taking into account the number of women between 15 and 49 years old of each country.

Based on data analysed, the most pressing issues identified are: Baseline coverage of maternal and reproductive health services is very low: about half of women of reproductive age in SSA are not provided with routine maternity care components that have a potentially significant impact on maternal and infant health. Multi-country coverage rate of “four antenatal care visits” is 42 percent, “delivery attended by skilled personnel” is 48 percent and “mother checkup” is 49 percent. Adolescents are the most neglected group in terms of access to contraceptive information and services. While for all women of reproductive age, “met need for family planning” has a multi-country coverage rate of 55 percent and a country level average of 53 percent, for adolescent girls the multi-country coverage rate for “met need for family planning” is 47 percent, and the country level average is 46 percent. Importantly, the same proportion of women is not provided with preventive interventions for infectious diseases that contribute significantly to the burden of perinatal and neonatal deaths in the region. For example, multi-country coverage rate for “malaria prophylaxis during pregnancy” is lower than 50 percent (46) and coverage rates for HIV-related opportunities are around 65 percent. The combination of high inequality indices (D-index) with low HOIs and coverage rates for most RMNCAH indicators, suggests a situation of uneven distribution of available reproductive and maternal health opportunities. What this indicates is the dramatic situation for the poorest and most marginalised women, adolescent girls and newborns in SSA. With few exceptions for specific health opportunities such as “six months of exclusive breastfeeding”, these groups are essentially excluded from the health system.

A report from sub-Saharan Africa

For women of reproductive age as a whole (15-49 years old), wealth, education and area of residence (urban/rural) are the most prominent determinants of access to the health opportunities analysed. However, for older adolescent girls (15-19 years old), being married appears to be the main source of inequalities for all opportunities observed, with percentages ranging from 40 to 75 percent (see Table 5.3). The descriptions of inequality of opportunity in this report may be relevant for decision-makers and managers in low and middle income countries (LMICs), and other development stakeholders in guiding broad strategic priorities and designing equity-oriented policies. They will also help to identify services with the largest inequality, as well as the most underserved groups. In turn, this may inform decisions on better targeting RMNCAH resources – both domestic and foreign – to support the scale-up of health interventions. Actions directed to increase access to and use of quality maternal and reproductive health services and to reduce inequalities between women are urgently needed, especially in settings where baseline inequalities are high.

Box 9. Strengths and limitations of the study The large number of countries and observations included in the analysis provides strong statistical power to the study. The relationships presented here should not be interpreted as causal. Rather, the report provides information about associations between individual and household circumstances and inequality. In this respect, more context-specific research, including a number of causal relations to identify the determinants of inequalities, is needed in order to design the most appropriate interventions to address the observed inequities. Notably, each of the 29 SSA countries included in the study has distinctive features that should be taken into account by policy makers to generate appropriate and effective policies. Although individual country results cannot be extrapolated to the rest of the SSA region, the results of the multi-country pooled analyses allow for drawing policy implications that could be generalised throughout the region. The representativeness of the sample populations is an essential criterion in order to be generalisable. In this report the Demographic Health Surveys (DHS) samples used are representative of the populations they include5. The HOIs and the D-indices are always upper and lower bounds respectively. Thus, in all cases conclusions are drawn following the most conservative results. Finally, the opportunities analysed in this report are strongly associated with individual efforts and decisions, therefore they are not “opportunities” in the strict economic sense. As explained in Chapters 1 and 2, the opportunities selected for the study are not as exogenous as would be desirable, since they are to a certain extent related to personal choices in the case of adult women. This is a limitation of this study.

Inequalities in women’s and girls’ health opportunities and outcomes

5.2 Policy options for adolescent girls Overall reproductive health opportunities among older adolescent girls scored low HOIs and high inequalities in these analyses. They appear to be the most neglected group in terms of access to contraceptive information and services, putting them at risk of early pregnancy and contracting HIV and other sexually transmitted diseases. Only around half have their family planning needs satisfied and are enrolled at school, and, on average, more than 20 percent have had a pregnancy before the age of 19 (multi-country pooled analysis results). Addressing the needs of adolescent girls requires addressing potential factors that act as barriers to accelerating progress. Marital status, along with wealth and occupation, appear to be strongly correlated with inequality of opportunity for this group. Increasing access to schooling is a critical strategy in ending child marriage and ensuring that married girls have the opportunity to complete their education. Strategies aimed at retaining older adolescent girls in school, preferably through at least the end of secondary education, e.g. scholarships, conditional or unconditional cash transfers and economic assistance for material and transportation, are key. If expanded and promoted, they could help adolescent girls (and their families) to delay marriage and first pregnancy, while reducing the high fertility rates observed among adolescents6. Introducing economic incentives or schemes (cash transfers, scholarships microfinance, loans) to increase the economic security of girls and families can encourage families to avoid or postpone early entry of children into the workforce or other consider alternatives7. Beyond education, protection services need to be accessible via a number of channels (in the health facility, at school, in the community) in order to ensure that cases of child marriage in the community are responded to effectively. There is an urgent need for SSA countries to implement youth-friendly health services – for both unmarried and married girls –, as several African countries are already doing8, and to make them accessible, acceptable and appropriate for adolescents’ needs9. Advocating to strengthen, implement and fund laws and policies that prevent child marriage is crucial to upholding girls’ rights. For example, advocacy in favour of raising the legal marriage age for girls to 18 years old and enforcing compliance where this already exists are crucial9. In addition, policies impeding girls’ access to contraceptive methods by requiring parental or spousal consent need to be reviewed10. Further, policies should target adolescent girls as well as other influential family members, who are often the decision makers on their behalf, and communities, which can have a powerful influence over them11.

A report from sub-Saharan Africa

5.3 Final considerations Despite progress between 1990 and 2015 in some Millennium Development Goals (MDGs) indicators, the target of reducing maternal mortality by 75 percent was not achieved. Importantly, inequality within and among countries for this indicator is growing. The gap in levels of maternal mortality between the best and worst performing countries in the past 20 years has doubled12. Though more women and adolescent girls are receiving services (e.g. delivery with a skilled birth attendant, antenatal care visits), these are often of poor-quality. In addition, many still undergo pregnancy and childbirth outside the health system or do not access modern contraceptives, the most cost-effective intervention to curb the number of maternal deaths13,14. Ensuring access to and availability of these basic services for the most vulnerable women, adolescents and newborns is necessary, and remains one of the most pressing issues to address the high risk of death from causes related to pregnancy and childbirth that millions of women and girls in SSA face on a daily basis. Moreover, prioritising provision of equitable access to reproductive, maternal and perinatal healthcare is a prerequisite to achieve the SDG3 targets associated with maternal, reproductive health and other related issues such as SDG5, gender equality or SDG10, reduced inequalities (see Box 1, Chapter 1). The main circumstance that poses a major barrier to the health and wellbeing of women and girls in most opportunities analysed is the unequal distribution of wealth. Thus, efforts to increase the incomes of the poorest segments of the population may have a significant impact on maternal mortality reduction in all settings. Furthermore, the pathway towards universal health coverage (UHC) – defined by the World Health Organization (WHO) as the situation in which “all people receive the health services they need without suffering financial hardship when paying for them”15 –, appears as the goal to work for to alleviate the financial constraints that deter less affluent women and girls from seeking and accessing health services. An approach towards progressive universalisation can advance the RMNCAH agenda by ensuring focus on and acceleration of effective coverage of key lifesaving interventions (e.g. childbirth attended by skilled personnel, emergency obstetric care). Additionally, those efforts should be driven by locally designed and tailored policies oriented to favour first the poorest and most excluded subsets, especially for those services where baseline inequalities are very marked, to ensure that the most vulnerable women’s needs are addressed and prioritised16 (e.g. marginalised groups could be exempted from or receive subsidies for user fees, premiums or co-payments, transportation subsidized, etc.). The present situation in SSA countries is far from UHC, and thus governments and stakeholders should prioritise actions towards scaling-up coverage of quality maternal and reproductive health services with the above-mentioned pro-poor approach aimed to curb

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

inequalities. This requires political will and mobilisation, sustained financing for health systems strengthening as well as new tools and knowledge. The financial gap to scale-up coverage of an essential package of reproductive, maternal and newborn health services poses a major but attainable challenge that can secure large health, social, and economic returns17. The Global Financing Facility launched in 2015 offers a window of opportunity for increased coordination and dialogue between donors and LMICs to address the resource gap and guarantee sustainable financing mechanisms for woman, child and adolescent health over the next years18. As we transition to the SDGs era, a number of external factors can influence the progress of the new agenda16. Challenges range from future humanitarian crises, global health governance issues, political instability, ensuring sustained momentum for RMNCAH among many competing health targets, to LMICs ability to increase their domestic fiscal space for health. The ability of the RMNCAH community to navigate such complex issues will determine effective and equitable provision of maternal and reproductive health that leaves no one behind. Finally, research is a central component to advance the post-2015 agenda in a more equitable way. Commensurate with the magnitude of the problem, more efforts and resources should be devoted to the evaluation of inequalities in access to health services and health outcomes. It is imperative to measure the extent of gaps in access to maternal-perinatal and reproductive health services and outcomes between population groups and the determinants of (or contributors to) these gaps, as well as tracking how coverage and inequalities for interventions change over time. Moreover, there is a need to generate more data concerning highly vulnerable groups such as migrant women and adolescents including younger adolescents and women in humanitarian crisis settings.

A report from sub-Saharan Africa

Box 10. Data gaps identified Despite notable progress in recent years, important gaps remain in the availability of data collected through DHS: Data about younger adolescent girls (10-14 years old) are missing. Unlike data on older adolescent girls (15-19 years old), which are routinely recorded in the DHS as well as in other household surveys, information on younger adolescents is almost always obtained through retrospective questions asked to women 15 and older. DHS and other household surveys funders and stakeholders need to make an effort to include this subset of younger adolescent girls in their surveys to enable new data generation that identifies their needs. Indicators related to newborn health opportunities, such as newborn checkup within hours after delivery, are included in the DHS, but are of poor quality in many countries. Frequently, values are missing for questions regarding the first hours after birth that may be due to respondentsâ€™ inability to recall information or the low number of women attending the newborn checkup within hours after delivery. While these data gaps are not filled, possible alternatives to obtain quality information about newborn health opportunities could be obtained through health facility survey data and records. Six out of the 29 SSA countries included in this report do not have available data on anaemia for women of reproductive age. Additionally, domestic violence and migration status indicators â€“ two circumstances that can drive inequalities â€“ were not included in the analysis because of lack of data (see Chapter 2). This report covers around 79 percent of the SSA population. The DHS do not have available data on the other 21 percent for the period 2010 and later, although some surveys are currently being carried out. These and other existing gaps highlight the need for innovative measurement approaches and support to SSA countries to upgrade their capacity to develop and implement better and sound measurement approaches.

Inequalities in women’s and girls’ health opportunities and outcomes

Key messages

Despite notable progress achieved in the last decade, overall reproductive and maternal health opportunities for women and girls in SSA are scarce and unequally distributed. As a result, a lack of services and a major burden of disease are more concentrated among the worst-off, the less-educated and those living in rural settings. Further progress in improving women’s and adolescents’ health and well-being can only be achieved by expanding coverage and reducing inequalities. This requires scale-up of needed services currently unavailable to large proportions of women and girls, while targeting first underserved populations in order to curb the inequality gaps that can otherwise impede acceleration of overall progress and the achievement of the SDGs targets. SSA countries and the RMNCAH community at large need political mobilisation and sustained financing for health system strengthening. Ensuring access for all women everywhere to skilled attendance in childbirth is key to addressing the high risk of death from causes related to pregnancy and childbirth; this is especially critical for the poorest and most marginalised women, girls and newborns, and thus, they should be prioritised. However, this is one of the biggest obstacles that lies ahead because of the challenge to provide access to quality care 24/7 in the context of weak health systems poses. Adolescent girls are a highly vulnerable group among women of reproductive age with very poor results in terms of access to reproductive health services and educational achievements. Early marriage is the main contributor to poor maternal and reproductive health indicators and outcomes for this group.

Ensuring equitable expansion of health coverage should be the cornerstone of efforts to meet SDG3 – including the reproductive, maternal and child health targets – and the ultimate goal of achieving universal health coverage. Actions outside the health system focused on poverty reduction, raising educational achievements and improving communities’ physical access to healthcare or ending child marriage have the potential to reduce inequalities in maternal and reproductive health, and highlight the need to strengthen inter-sectorial co-operation and coordination mechanisms among health and other sectors concerned. Research has a key role to play to further ascertain the levels and causes of inequalities, bridge the existing data gaps for specific subgroups of vulnerable women and girls, as well as for monitoring and accountability purposes.

A report from sub-Saharan Africa

References 1.

Koblinsky, M. et al. Quality maternity care for every woman, everywhere: a call to action. Lancet (2016). doi:10.1016/S0140-6736(16)31333-2

2. WHO: Department of Health Statistics and Information Systems. State of Inequality Reproductive, maternal, newborn and child health. (2015). 3.

Kuruvilla, S. et al. Success factors for reducing maternal and child mortality. Bull World Heal. Organ 92, 533–544 (2014).

4. Victora, C. G. et al. Countdown to 2015: a decade of tracking progress for maternal, newborn, and child survival. Lancet (London, England) (2015). doi:10.1016/ S0140-6736(15)00519-X 5.

USAID. The DHS Program. at <http://www.dhsprogram.com/>

6. UNFPA. Girlhood, not motherhood preventing adolescent pregnancy. (2015). 7.

Girls Not Brides. at <http://www.girlsnotbrides.org/>

8. WHO African Region. The African Regional Health Report 2014. (2014). 9. Department of Maternal Newborn Child and Adolescent Health. Making Health Services Adolescent Friendly - Developing National Quality Standards for Adolescent Friendly Health Services. (2012). at <http://www.who.int/iris/bitstre am/10665/75217/1/9789241503594_eng.pdf?ua=1> 10. Tumlinson, K., Okigbo, C. C. & Speizer, I. S. Provider barriers to family planning access in urban Kenya. Contraception 92, 143–151 (2015). 11. Lowe, S. M. P. & Moore, S. Social networks and female reproductive choices in the developing world: a systematized review. Reprod. Health 11, 85 (2014). 12. Graham, W. et al. Diversity and divergence: the dynamic burden of poor maternal health. Lancet (2016). doi:10.1016/S0140-6736(16)31533-1 13. Singh, S. et al. Made possible by The Costs and Benefits of Investing in Sexual and Reproductive Health 2014. 14. Ahmed, S., Li, Q., Liu, L. & Tsui, A. O. Maternal deaths averted by contraceptive use: An analysis of 172 countries. Lancet (2012). doi:10.1016/S01406736(12)60478-4 15. WHO. Universal health coverage (UHC). Fact sheet N°395 (2016). 16. Kruk, M. E. et al. Next generation maternal health: external shocks and health-system innovations. Lancet (2016). doi:10.1016/S0140-6736(16)31395-2 17. Stenberg, K. et al. Advancing social and economic development by investing in women’s and children’s health: A new Global Investment Framework. Lancet 383, 1333–1354 (2014). 18. The World Bank. Global Financing Facility in Support of Every Woman Every Child. (2015).

Â©Pau Fabregat

Appendix A

A report from sub-Saharan Africa

Table A1 Levels: Currently attending school (older adolescent girls) Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

38.35

25.95

13.44

51.79

0.81

Burkina Faso

13.34

42.69

9.94

23.28

0.57

Burundi

37.93

29.92

16.19

54.13

1.03

Cameroon

45.43

23.09

13.64

59.07

0.75

Comoros

63.31

12.74

9.24

72.56

1.53

Congo

50.19

21.52

13.76

63.95

1.33

Congo DR

46.38

21.24

12.51

58.89

0.91

CĂ´te dâ€™Ivoire

17.79

40.97

12.35

30.14

0.90

Ethiopia

45.29

19.73

11.14

56.43

0.95

Gabon

72.33

9.86

7.91

80.25

1.61

The Gambia

37.40

25.80

13.00

50.40

1.01

Ghana

38.21

15.61

7.07

45.28

1.41

Guinea

23.30

37.32

13.88

37.18

0.89

Kenya

56.93

17.86

12.38

69.31

1.00

Liberia

57.32

12.98

8.55

65.87

1.32

Malawi

43.38

23.46

13.29

56.68

0.64

Mali

17.77

40.28

11.98

29.75

0.76

Mozambique

27.42

34.10

14.19

41.60

0.85

Namibia

65.71

10.34

7.58

73.29

1.12

Niger

6.01

61.81

9.73

15.74

0.41

Nigeria

34.04

31.25

15.47

49.51

0.43

Rwanda

39.78

22.78

11.73

51.51

0.99

Senegal

36.99

21.55

10.16

47.15

1.15

Sierra Leone

45.74

21.93

12.85

58.59

0.79

Tanzania

24.92

38.68

15.72

40.64

0.90

Togo

38.42

28.55

15.35

53.77

0.90

Uganda

43.81

24.21

13.99

57.80

0.85

Zambia

44.54

19.78

10.98

55.52

0.81

Zimbabwe

32.22

29.11

13.23

45.45

0.91

100

Inequalities in women’s and girls’ health opportunities and outcomes

Table A2 Levels: Having never been pregnant (older adolescent girls) Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

73.85

11.89

9.96

83.81

0.56

Burkina Faso

61.10

18.97

14.30

75.40

0.91

Burundi

83.19

7.98

7.22

90.40

0.56

Cameroon

62.17

16.92

12.67

74.83

0.78

Comoros

79.44

10.44

9.26

88.70

0.74

Congo

52.01

22.40

15.01

67.02

0.85

Congo DR

58.76

19.23

13.99

72.75

0.73

Côte d’Ivoire

59.89

15.07

10.63

70.52

1.35

Ethiopia

78.47

10.54

9.24

87.72

0.92

Gabon

63.63

12.05

8.72

72.35

1.52

The Gambia

71.31

13.72

11.33

82.64

1.00

Ghana

80.18

6.57

5.64

85.82

1.13

Guinea

50.95

22.50

14.80

65.75

1.14

Kenya

73.93

9.98

8.20

82.12

0.87

Liberia

59.87

12.76

8.76

68.63

1.37

Malawi

59.42

20.14

14.99

74.41

0.65

Mali

44.23

27.17

16.50

60.74

1.13

Mozambique

47.09

24.46

15.25

62.34

1.02

Namibia

76.35

6.26

5.09

81.45

1.08

Niger

45.12

24.24

14.44

59.56

1.44

Nigeria

63.93

17.60

13.66

77.58

0.59

Rwanda

89.51

3.45

3.20

92.71

0.50

Senegal

71.75

12.90

10.63

82.38

1.28

Sierra Leone

61.94

14.16

10.22

72.16

0.77

Tanzania

65.75

14.74

11.37

77.12

1.01

Togo

74.23

11.14

9.30

83.53

0.79

Uganda

62.48

17.96

13.68

76.16

0.93

Zambia

60.06

16.08

11.51

71.57

0.67

Zimbabwe

62.08

18.76

14.33

76.41

1.02

101

A report from sub-Saharan Africa

Table A3 Levels: Met need for family planning (older adolescent girls) Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

46.00

15.47

8.42

54.42

1.73

Burkina Faso

26.63

30.36

11.61

38.25

1.81

Burundi

23.38

24.11

7.43

30.82

4.27

Cameroon

49.72

16.06

9.51

59.24

1.73

Comoros

26.96

27.21

10.08

37.04

3.89

Congo

64.64

8.97

6.37

71.01

2.33

Congo DR

38.04

13.89

6.14

44.18

2.13

CĂ´te dâ€™Ivoire

38.86

19.75

9.56

48.42

2.38

Ethiopia

36.21

17.45

7.66

43.87

3.28

Gabon

55.84

12.11

7.69

63.53

2.46

The Gambia

12.40

26.87

4.56

16.96

3.14

Ghana

37.72

11.19

4.75

42.47

3.31

Guinea

19.63

28.82

7.95

27.58

1.93

Kenya

51.57

12.47

7.35

58.92

3.04

Liberia

32.53

14.34

5.45

37.98

2.00

Malawi

43.42

12.39

6.14

49.56

1.97

Mali

22.03

22.87

6.53

28.56

2.42

Mozambique

21.59

32.35

10.32

31.92

1.65

Namibia

68.45

7.67

5.69

74.13

2.18

Niger

29.06

14.28

4.84

33.90

3.50

Nigeria

36.43

26.94

13.44

49.87

1.37

Rwanda

25.97

34.41

13.63

39.60

3.52

Senegal

28.91

27.18

10.79

39.70

3.35

Sierra Leone

49.65

13.08

7.47

57.13

1.47

Tanzania

47.37

12.42

6.72

54.09

3.10

Togo

37.31

21.55

10.25

47.56

2.38

Uganda

37.74

21.52

10.35

48.09

3.34

Zambia

35.58

19.71

8.73

44.32

1.96

Zimbabwe

53.18

12.20

7.39

60.56

3.22

102

Inequalities in women’s and girls’ health opportunities and outcomes

Table A4 Levels: Not having anaemia Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

56.61

3.34

1.96

58.56

0.81

Burkina Faso

48.33

5.61

2.87

51.21

0.65

Burundi

79.12

2.94

2.39

81.52

0.76

Cameroon

58.37

3.34

2.02

60.38

0.68

Congo

43.95

4.10

1.88

45.83

1.21

Congo DR

59.99

2.67

1.65

61.63

0.74

Côte d’Ivoire

44.40

3.85

1.78

46.17

1.00

Ethiopia

80.81

3.06

2.55

83.36

0.51

Gabon

36.80

6.58

2.59

39.40

1.16

The Gambia

36.82

10.68

4.40

41.22

0.88

Ghana

54.76

4.99

2.88

57.64

0.91

Guinea

47.74

6.08

3.09

50.83

0.88

Malawi

68.78

2.77

1.96

70.75

0.74

Mali

45.75

5.84

2.84

48.59

0.85

Mozambique

43.90

4.60

2.12

46.01

0.58

Namibia

77.46

2.44

1.94

79.39

0.77

Niger

51.89

4.25

2.30

54.19

0.91

Rwanda

79.14

2.02

1.63

80.77

0.58

Sierra Leone

51.80

6.28

3.47

55.27

0.72

Tanzania

58.17

2.91

1.74

59.91

0.68

Togo

49.18

5.37

2.79

51.96

0.89

Uganda

74.63

3.01

2.32

76.95

1.11

Zimbabwe

70.35

2.00

1.44

71.79

0.61

103

A report from sub-Saharan Africa

Table A5 Levels: Having the recommended BMI Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

61.74

7.25

4.82

66.56

0.46

Burkina Faso

70.94

4.21

3.12

74.07

0.61

Burundi

74.14

3.67

2.82

76.96

0.74

Cameroon

54.15

10.76

6.53

60.68

0.65

Comoros

49.46

10.99

6.11

55.57

0.89

Congo

55.30

8.36

5.04

60.34

1.15

Congo DR

67.42

4.28

3.01

70.43

0.73

CĂ´te dâ€™Ivoire

61.44

8.08

5.40

66.84

0.96

Ethiopia

66.66

3.52

2.43

69.09

0.59

Gabon

41.49

14.18

6.85

48.35

1.12

The Gambia

57.37

5.93

3.62

60.99

1.07

Ghana

44.71

16.57

8.88

53.59

0.89

Guinea

64.50

6.47

4.47

68.96

0.87

Kenya

52.94

9.63

5.64

58.58

0.62

Liberia

60.31

9.01

5.97

66.28

1.08

Malawi

70.72

4.32

3.20

73.92

0.76

Mali

67.09

6.07

4.33

71.42

0.75

Mozambique

70.68

6.26

4.72

75.39

0.46

Namibia

49.11

10.59

5.82

54.93

0.90

Niger

65.15

5.62

3.88

69.04

0.77

Nigeria

59.27

7.72

4.96

64.23

0.32

Rwanda

67.77

5.50

3.95

71.71

0.63

Sierra Leone

69.07

4.85

3.52

72.59

0.70

Tanzania

63.44

6.75

4.59

68.04

0.65

Togo

54.99

11.46

7.12

62.11

0.83

Uganda

65.94

6.23

4.38

70.32

1.12

Zambia

62.03

7.52

5.04

67.07

0.52

Zimbabwe

56.06

8.87

5.46

61.52

0.62

104

Inequalities in women’s and girls’ health opportunities and outcomes

Table A6 Levels: Met need for family planning Country Benin

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

27.11

16.77

5.46

32.57

0.62

Burkina Faso

34.40

20.39

8.81

43.22

0.70

Burundi

35.73

11.34

4.57

40.30

1.05

Cameroon

47.51

17.62

10.16

57.67

0.71

Comoros

32.37

19.64

7.91

40.28

1.40

Congo

72.24

5.88

4.51

76.75

0.98

Congo DR

39.74

14.41

6.69

46.43

0.92

Côte d’Ivoire

39.03

16.46

7.69

46.72

1.04

Ethiopia

45.06

16.34

8.80

53.87

1.00

Gabon

55.39

10.34

6.39

61.77

1.26

The Gambia

22.67

24.55

7.38

30.04

1.08

Ghana

46.96

10.54

5.53

52.50

1.02

Guinea

20.38

32.51

9.82

30.19

0.88

Kenya

73.55

5.94

4.65

78.20

0.65

Liberia

40.45

10.96

4.98

45.43

1.06

Malawi

61.13

6.12

3.98

65.11

0.60

Mali

22.83

24.35

7.35

30.18

0.86

Mozambique

29.84

22.35

8.59

38.43

0.88

Namibia

78.43

4.53

3.72

82.15

0.68

Niger

42.07

11.70

5.58

47.64

1.20

Nigeria

44.67

20.66

11.63

56.31

0.53

Rwanda

68.42

4.92

3.54

71.97

0.70

Senegal

40.54

16.21

7.84

48.38

1.25

Sierra Leone

42.55

15.86

8.02

50.57

0.83

Tanzania

53.82

10.31

6.19

60.01

0.95

Togo

36.78

13.76

5.87

42.65

0.87

Uganda

43.65

15.49

8.00

51.65

1.01

Zambia

66.59

5.32

3.74

70.34

0.67

Zimbabwe

77.73

3.43

2.76

80.49

0.75

105

A report from sub-Saharan Africa

Table A7 Levels: Knowledge of a place where to get an HIV test Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

47.16

14.27

7.85

55.01

0.41

Burkina Faso

67.62

6.83

4.96

72.58

0.43

Burundi

81.38

4.82

4.12

85.50

0.52

Cameroon

35.78

9.96

3.96

39.74

0.45

Comoros

24.75

23.26

7.50

32.25

0.71

Congo

66.47

9.77

7.20

73.67

0.64

Congo DR

41.31

19.54

10.03

51.34

0.51

CĂ´te dâ€™Ivoire

53.14

14.45

8.98

62.12

0.65

Ethiopia

58.14

12.53

8.33

66.47

0.60

Gabon

85.15

4.34

3.86

89.02

0.65

The Gambia

62.91

8.25

5.65

68.56

0.65

Ghana

72.36

7.81

6.13

78.49

0.55

Guinea

32.83

16.93

6.69

39.52

0.59

Kenya

95.41

1.74

1.69

97.10

0.23

Liberia

70.07

7.97

6.07

76.14

0.61

Malawi

95.65

1.33

1.29

96.94

0.21

Mali

19.23

28.77

7.77

27.00

0.45

Mozambique

72.95

7.00

5.49

78.44

0.55

Namibia

96.14

1.23

1.20

97.34

0.26

Niger

37.15

16.22

7.19

44.34

0.59

Nigeria

49.05

18.72

11.30

60.35

0.30

Rwanda

98.77

0.39

99.16

0.12

Senegal

70.02

7.63

5.79

75.81

0.00

Sierra Leone

66.57

5.55

3.91

70.48

0.47

Tanzania

89.52

3.07

2.84

92.36

0.43

Togo

63.78

11.34

8.16

71.94

0.55

Uganda

92.40

2.42

2.29

94.68

0.38

Zambia

94.88

1.66

1.60

96.48

0.24

Zimbabwe

87.39

3.60

3.26

90.65

0.43

106

Inequalities in women’s and girls’ health opportunities and outcomes

Table A8 Levels: Four antenatal care visits Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

51.87

13.64

8.19

60.06

0.57

Burkina Faso

9.44

29.35

3.92

13.36

0.31

Burundi

31.83

4.85

1.62

33.46

0.79

Cameroon

53.58

14.65

9.19

62.78

0.68

Comoros

51.45

9.90

5.65

57.10

1.56

Congo

74.04

6.22

4.91

78.95

0.75

Congo DR

30.38

13.85

4.88

35.26

0.65

Côte d’Ivoire

35.89

18.82

8.32

44.21

0.85

Ethiopia

10.28

35.69

5.70

15.98

0.49

Gabon

73.86

6.98

5.54

79.41

0.94

The Gambia

75.00

3.40

2.64

77.64

0.89

Ghana

76.50

6.72

5.51

82.01

0.75

Guinea

42.92

16.87

8.71

51.64

0.84

Kenya

14.98

22.98

4.47

19.45

0.53

Liberia

74.00

5.72

4.49

78.49

0.76

Malawi

42.48

3.86

1.71

44.18

0.57

Mali

20.98

29.00

8.57

29.56

0.57

Mozambique

29.14

18.29

6.52

35.66

0.66

Namibia

77.96

4.19

3.41

81.37

0.92

Niger

28.65

10.01

3.19

31.84

0.66

Nigeria

36.17

26.63

13.13

49.30

0.37

Rwanda

40.38

7.99

3.51

43.89

0.71

Senegal

43.24

12.41

6.13

49.37

0.99

Sierra Leone

85.02

1.80

1.56

86.59

0.52

Tanzania

34.73

11.80

4.64

39.38

0.80

Togo

36.52

21.72

10.13

46.66

0.74

Uganda

44.51

7.29

3.50

48.01

0.90

Zambia

51.59

5.35

2.91

54.51

0.65

Zimbabwe

62.27

4.99

3.27

65.54

0.88

107

A report from sub-Saharan Africa

Table A9 Levels: Delivery attended by skilled personnel Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

75.21

8.41

6.91

82.12

0.50

Burkina Faso

18.01

30.45

7.88

25.89

0.41

Burundi

57.60

8.74

5.51

63.11

0.85

Cameroon

53.84

20.51

13.89

67.73

0.57

Comoros

80.18

6.11

5.21

85.40

1.20

Congo

90.23

3.56

3.34

93.57

0.38

Congo DR

45.91

10.10

5.16

51.06

0.69

CĂ´te dâ€™Ivoire

52.57

15.83

9.88

62.45

0.90

Ethiopia

4.91

58.41

6.90

11.81

0.30

Gabon

88.65

3.53

3.24

91.89

0.58

The Gambia

57.22

13.95

9.27

66.49

0.77

Ghana

63.45

13.03

9.51

72.96

0.85

Guinea

30.72

29.38

12.79

43.51

0.72

Kenya

56.85

14.83

9.90

66.75

0.72

Liberia

57.32

11.45

7.41

64.74

0.82

Malawi

69.88

5.67

4.20

74.08

0.53

Mali

28.46

29.62

11.97

40.43

0.62

Mozambique

15.70

27.95

6.09

21.79

0.50

Namibia

85.21

5.16

4.64

89.85

0.66

Niger

23.84

27.89

9.22

33.06

0.59

Nigeria

25.35

36.97

14.87

40.22

0.29

Rwanda

87.99

3.09

2.80

90.79

0.51

Senegal

50.69

16.37

9.92

60.61

0.83

Sierra Leone

56.45

9.88

6.19

62.64

0.69

Tanzania

42.89

19.02

10.07

52.96

0.82

Togo

48.43

21.49

13.26

61.69

0.74

Uganda

53.14

12.29

7.44

60.59

0.89

Zambia

59.85

12.95

8.90

68.75

0.62

Zimbabwe

59.20

12.11

8.16

67.36

0.86

108

Inequalities in women’s and girls’ health opportunities and outcomes

Table A10 Levels: Delivery attended by skilled personnel, by place of delivery Home deliveries Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

Deliveries in health facilities SD HOI (%)

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

4.29

38.37

2.67

6.97

0.53

89.92

2.74

2.53

92.45

0.40

Burkina Faso

0.17

50.73

0.17

0.34

0.06

27.49

23.14

8.28

35.76

0.59

Burundi

0.63

53.97

0.74

1.38

0.15

99.35

0.20

99.55

0.18

Cameroon

4.33

45.59

3.63

7.97

0.47

99.18

0.25

99.42

0.20

Comoros

22.78

27.01

8.43

31.20

0.00

99.54

0.21

99.75

0.27

Congo

18.24

23.66

5.65

23.89

1.76

98.97

0.34

99.31

0.16

Congo DR

15.72

11.01

1.95

17.67

1.06

54.81

7.17

4.23

59.04

0.79

Côte d’Ivoire

4.89

27.00

1.81

6.69

0.58

98.82

0.42

0.41

99.23

0.25

Ethiopia

0.19

64.50

0.34

0.53

0.05

94.49

2.07

2.00

96.49

1.20

Gabon

4.17

47.39

3.75

7.92

0.88

97.41

0.55

0.54

97.95

0.57

The Gambia

2.33

31.17

1.05

3.38

0.30

99.78

0.09

99.87

0.09

Ghana

1.52

51.22

1.59

3.11

0.34

94.57

1.43

1.37

95.94

0.50

Guinea

5.36

48.41

5.03

10.39

0.44

84.16

5.88

5.26

89.42

1.03

Kenya

2.31

25.78

0.80

3.11

0.34

98.92

0.25

99.17

0.24

Liberia

9.87

31.14

4.46

14.34

0.72

96.72

0.95

0.93

97.65

0.39

Malawi

0.39

51.20

0.41

0.79

0.11

96.32

0.51

0.49

96.81

0.27

Mali

1.70

35.92

0.96

2.66

0.26

59.01

13.12

8.91

67.92

0.99

0.00

30.83

13.45

4.79

35.62

0.87

Namibia

6.50

24.80

2.14

8.65

1.18

99.58

0.13

99.71

0.14

Niger

0.61

28.56

0.24

0.85

0.11

94.41

1.75

1.68

96.09

0.63

Nigeria

3.12

53.07

3.52

6.64

0.15

94.39

1.74

1.67

96.06

0.33

Rwanda

0.21

67.16

0.44

0.65

0.12

99.76

0.09

99.86

0.09

Senegal

1.83

64.80

3.37

5.21

0.31

70.59

6.49

4.90

75.48

0.98

Sierra Leone

9.37

35.50

5.16

14.52

0.53

98.58

0.35

98.93

0.22

Tanzania

1.47

33.99

0.76

2.23

0.26

94.37

1.69

1.62

95.99

0.62

Togo

0.61

62.27

1.00

1.61

0.20

73.99

9.49

7.76

81.75

0.83

Uganda

4.00

18.41

0.90

4.91

0.52

96.97

0.51

0.50

97.47

0.45

Zambia

0.42

38.71

0.27

0.69

0.13

93.21

2.16

2.06

95.27

0.40

Zimbabwe

2.96

27.59

1.13

4.09

0.41

98.94

0.36

99.31

0.26

Mozambique

109

A report from sub-Saharan Africa

Table A11 Levels: Postnatal checkup Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

62.38

8.04

5.45

67.83

0.56

Burkina Faso

81.21

3.95

3.34

84.56

0.47

Burundi

27.18

12.00

3.71

30.89

0.75

Cameroon

36.36

18.65

8.34

44.70

0.61

Comoros

60.49

7.85

5.15

65.65

1.45

Congo

73.74

5.87

4.60

78.34

0.73

Congo DR

44.44

10.60

5.27

49.71

0.70

CĂ´te dâ€™Ivoire

76.58

5.62

4.56

81.13

0.80

Ethiopia

5.22

44.10

4.12

9.34

0.34

Gabon

71.10

5.39

4.05

75.15

0.99

The Gambia

74.33

5.16

4.04

78.37

0.69

Ghana

81.72

5.40

4.66

86.39

0.71

Guinea

37.42

15.86

7.06

44.48

0.82

Kenya

52.53

12.57

7.55

60.09

0.75

Liberia

71.32

5.77

4.37

75.69

0.78

Mali

39.66

17.80

8.59

48.24

0.71

Mozambique

63.79

7.31

5.03

68.82

0.77

Namibia

82.33

4.58

3.95

86.29

0.73

Niger

36.61

14.87

6.39

43.00

0.69

Nigeria

31.62

28.15

12.39

44.02

0.35

Rwanda

42.18

6.38

2.87

45.06

0.72

Senegal

77.77

7.67

6.46

84.23

0.69

Sierra Leone

78.94

3.01

2.45

81.39

0.58

Togo

77.06

6.08

4.99

82.05

0.71

Uganda

32.04

15.29

5.79

37.83

0.81

Zambia

66.63

8.90

6.51

73.14

0.60

Zimbabwe

45.60

11.07

5.68

51.28

0.86

110

Inequalities in women’s and girls’ health opportunities and outcomes

Table A12 Levels: Maternity care package Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

35.63

18.21

7.93

43.56

0.53

Burkina Faso

6.25

37.04

3.68

9.93

0.25

Burundi

9.94

15.26

1.79

11.73

0.50

Cameroon

23.02

27.97

8.94

31.96

0.48

Comoros

31.85

14.64

5.46

37.31

1.37

Congo

57.46

10.45

6.70

64.16

0.84

Congo DR

10.89

25.99

3.82

14.71

0.46

Côte d’Ivoire

24.86

26.27

8.86

33.72

0.72

Ethiopia

1.17

70.41

2.79

3.96

0.12

Gabon

52.98

10.70

6.35

59.33

1.12

The Gambia

43.28

13.49

6.75

50.03

0.80

Ghana

53.82

15.57

9.93

63.75

0.88

Guinea

15.62

34.80

8.34

23.95

0.56

Kenya

8.42

34.58

4.45

12.86

0.39

Liberia

42.67

14.72

7.36

50.03

0.83

Mali

12.05

39.06

7.72

19.77

0.42

Mozambique

6.96

35.50

3.83

10.80

0.31

Namibia

57.77

11.17

7.26

65.03

0.98

Niger

8.21

32.42

3.94

12.14

0.36

Nigeria

16.23

42.22

11.86

28.10

0.24

Rwanda

17.50

13.78

2.80

20.30

0.53

Senegal

24.35

23.89

7.64

31.99

0.77

Sierra Leone

44.58

11.09

5.56

50.15

0.69

Togo

28.37

27.63

10.83

39.19

0.67

Uganda

14.44

23.74

4.49

18.93

0.59

Zambia

31.28

14.91

5.48

36.76

0.57

Zimbabwe

24.63

19.77

6.07

30.70

0.71

111

A report from sub-Saharan Africa

Table A13 Levels: Malaria prophylaxis during pregnancy Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

43.97

5.33

2.47

46.45

0.66

Burkina Faso

58.73

5.48

3.40

62.13

1.06

Burundi

0.22

46.98

0.20

0.42

0.08

Cameroon

48.21

6.20

3.18

51.39

0.75

Comoros

48.42

6.22

3.21

51.63

1.49

Congo

52.09

4.31

2.34

54.43

1.13

Congo DR

34.22

7.65

2.84

37.06

0.84

CĂ´te dâ€™Ivoire

27.79

8.01

2.42

30.21

0.86

Gabon

5.45

17.58

1.16

6.62

0.56

The Gambia

92.74

1.39

1.31

94.04

0.69

Ghana

84.94

2.18

1.89

86.83

0.84

Guinea

34.32

7.52

2.79

37.11

0.92

Kenya

23.82

13.70

3.78

27.60

1.30

Liberia

66.11

3.22

2.20

68.31

0.96

Malawi

88.31

1.50

1.35

89.66

0.38

Mali

68.52

4.95

3.57

72.09

1.00

Mozambique

44.28

7.93

3.81

48.09

0.89

7.13

17.42

1.50

8.63

0.48

Niger

70.31

1.65

1.18

71.49

0.73

Nigeria

33.02

14.95

5.81

38.83

0.54

Senegal

79.14

3.51

2.88

82.02

0.80

Sierra Leone

65.26

3.13

2.11

67.37

0.67

Tanzania

63.47

5.04

3.37

66.84

0.93

Togo

81.45

1.55

1.28

82.74

0.82

Uganda

47.21

5.99

3.01

50.22

0.93

Zambia

93.06

1.35

1.27

94.33

0.35

Zimbabwe

13.94

12.01

1.90

15.84

0.61

Namibia

112

Inequalities in women’s and girls’ health opportunities and outcomes

Table A14 Levels: HIV test offered during pregnancy Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

55.80

11.66

7.36

63.17

0.73

Burkina Faso

42.39

19.90

10.53

52.92

1.14

Burundi

56.29

5.26

3.12

59.41

0.95

Cameroon

69.15

10.32

7.96

77.11

1.07

Comoros

23.93

11.31

3.05

26.98

1.48

Congo

42.20

14.24

7.01

49.21

1.03

Congo DR

24.11

29.00

9.85

33.96

0.81

Côte d’Ivoire

49.08

15.29

8.86

57.94

1.08

Ethiopia

45.62

17.03

9.37

54.99

1.83

Gabon

87.32

3.11

2.80

90.13

1.14

The Gambia

65.71

3.55

2.42

68.13

0.86

Ghana

76.15

5.83

4.72

80.87

1.05

Guinea

10.10

36.71

5.86

15.96

0.63

Kenya

97.36

0.84

0.82

98.18

0.63

Liberia

74.10

5.37

4.20

78.30

0.99

Malawi

91.89

1.34

1.25

93.14

0.36

Mali

29.21

20.61

7.58

36.80

1.16

Mozambique

71.53

7.41

5.72

77.25

0.99

Namibia

97.36

0.80

0.79

98.15

0.43

Niger

29.25

22.74

8.61

37.86

0.85

Nigeria

50.81

16.13

9.77

60.58

0.53

Rwanda

99.11

0.23

99.34

0.17

Sierra Leone

71.44

4.89

3.67

75.11

0.67

Tanzania

80.56

4.82

4.08

84.64

0.88

Togo

75.14

7.83

6.39

81.52

0.99

Uganda

80.97

3.52

2.96

83.92

0.86

Zambia

94.13

1.34

1.28

95.40

0.38

Zimbabwe

85.94

2.97

2.63

88.57

0.79

113

A report from sub-Saharan Africa

Table A15 Levels: Infant checkup within two months after delivery Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

45.74

8.86

4.45

50.19

0.59

Burkina Faso

80.66

2.78

2.31

82.97

0.50

Burundi

16.90

13.55

2.65

19.55

0.81

Cameroon

11.36

24.94

3.77

15.13

0.68

Comoros

30.05

8.65

2.85

32.90

1.40

Congo

53.62

6.39

3.66

57.27

0.94

Congo DR

15.78

15.50

2.89

18.67

0.53

CĂ´te dâ€™Ivoire

68.92

2.82

2.00

70.92

0.93

Ethiopia

3.24

22.35

0.93

4.17

0.30

Gabon

53.88

5.71

3.26

57.14

1.22

The Gambia

68.92

4.95

3.59

72.50

1.50

Ghana

70.00

4.62

3.39

73.39

0.92

Guinea

55.65

7.35

4.41

60.07

0.87

Kenya

64.45

5.87

4.02

68.47

0.74

Liberia

61.45

4.69

3.02

64.47

0.97

Malawi

29.91

6.63

2.12

32.03

1.09

Mali

33.64

11.28

4.28

37.91

0.72

Namibia

51.29

7.07

3.90

55.19

0.97

Niger

58.23

5.83

3.60

61.83

0.74

Nigeria

20.42

29.95

8.73

29.15

0.32

Rwanda

50.39

4.67

2.47

52.86

1.04

Senegal

80.58

5.80

4.96

85.54

0.72

Sierra Leone

68.77

2.83

2.00

70.77

0.66

Tanzania

16.66

7.67

1.38

18.05

0.82

Togo

69.40

4.84

3.53

72.93

0.80

Uganda

28.56

11.14

3.58

32.14

0.83

Zambia

54.87

4.79

2.76

57.63

0.66

Zimbabwe

56.48

7.21

4.39

60.87

1.10

114

Inequalities in women’s and girls’ health opportunities and outcomes

Table A16 Levels: Six months of exclusive breastfeeding Country

HOI (%)

D-index (%)

Penalty (%)

Coverage (%)

SD HOI (%)

Benin

58.66

6.02

3.75

62.42

1.65

Burkina Faso

94.75

1.39

1.33

96.09

1.05

Burundi

92.86

1.79

1.70

94.56

1.62

Cameroon

72.93

5.59

4.32

77.25

2.46

Comoros

66.29

5.76

4.05

70.34

0.00

Congo

59.82

9.48

6.26

66.08

3.95

Congo DR

70.96

5.11

3.82

74.78

2.29

Côte d’Ivoire

79.05

5.87

4.93

83.98

2.72

Ethiopia

86.94

2.32

2.07

89.01

1.62

Gabon

67.64

7.65

5.60

73.24

3.96

The Gambia

88.58

3.08

2.82

91.39

2.38

Ghana

73.16

6.60

5.17

78.34

3.21

Guinea

85.95

3.23

2.87

88.81

2.14

Kenya

81.98

3.85

3.29

85.27

2.02

Liberia

90.81

3.68

3.47

94.28

1.90

Malawi

77.00

5.73

4.68

81.68

2.62

Mali

73.52

3.71

2.84

76.36

2.38

Mozambique

54.78

7.52

4.45

59.24

1.99

Namibia

76.95

4.79

3.87

80.83

3.28

Niger

89.26

1.56

1.42

90.68

1.45

Nigeria

71.14

3.38

2.49

73.63

1.09

Rwanda

95.17

1.69

1.63

96.80

1.35

Senegal

83.16

4.00

3.46

86.62

2.96

Sierra Leone

77.32

3.56

2.85

80.18

2.35

Tanzania

62.29

6.19

4.11

66.40

2.31

Togo

85.18

3.57

3.15

88.34

2.42

Uganda

69.80

6.63

4.96

74.75

3.38

Zambia

79.49

2.85

2.33

81.82

1.50

Zimbabwe

57.85

5.06

3.08

60.94

2.38

115

A report from sub-Saharan Africa

Table A17 Shapley decomposition: Currently attending school (older adolescent girls) Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Sex of the household head

Occupation

Religion

Marital status

Benin

13.97

5.39

2.55

45.02

8.01

25.06

Burkina Faso

18.19

15.08

2.08

27.77

10.66

26.23

Burundi

9.23

0.39

0.09

75.90

0.76

13.64

Cameroon

14.69

8.97

3.69

14.02

17.40

41.23

Comoros

24.78

7.86

0.75

18.47

0.65

47.49

Congo

15.59

7.47

0.62

35.12

3.70

37.50

Congo DR

11.24

11.22

0.27

24.82

1.12

51.34

CĂ´te dâ€™Ivoire

17.11

16.32

2.90

37.33

9.14

17.20

Ethiopia

15.19

9.17

2.22

5.00

5.77

62.65

Gabon

22.06

3.90

1.74

17.18

5.67

49.45

The Gambia

11.86

12.28

3.17

22.18

1.78

48.72

Ghana

9.13

1.32

5.69

51.94

3.92

28.00

Guinea

19.86

14.25

0.95

32.66

1.48

30.79

Kenya

8.76

9.59

0.67

37.59

3.82

39.58

Liberia

23.31

14.29

3.07

7.19

8.77

43.38

Malawi

12.29

2.95

3.36

10.08

1.63

69.69

Mali

31.22

16.64

1.25

1.22

49.67

Mozambique

21.75

14.46

1.50

8.30

3.30

50.69

Namibia

17.96

6.38

6.92

33.17

2.53

33.04

Niger

27.57

25.41

2.33

0.97

43.71

Nigeria

20.56

9.97

3.71

8.87

17.60

39.29

Rwanda

8.55

2.70

4.15

76.50

0.67

7.43

Senegal

18.00

22.73

6.77

4.60

47.90

Sierra Leone

13.92

13.55

4.21

31.53

3.32

33.47

Tanzania

8.40

5.13

0.43

63.12

22.91

Togo

2.52

1.92

1.06

61.33

7.80

25.38

Uganda

7.37

3.76

1.91

23.11

0.51

63.34

Zambia

14.12

8.61

1.60

20.20

0.32

55.14

Zimbabwe

8.22

1.03

2.29

28.27

2.18

58.02

116

Inequalities in women’s and girls’ health opportunities and outcomes

Table A18 Shapley decomposition: Having never been pregnant (older adolescent girls) Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Sex of the household head

Occupation

Religion

Marital status

Benin

9.11

5.02

2.76

6.46

3.51

73.16

Burkina Faso

8.06

7.15

1.26

2.22

4.00

77.30

Burundi

1.61

1.13

4.34

15.84

3.10

73.97

Cameroon

6.84

6.00

2.17

5.38

5.18

74.43

Comoros

7.28

2.20

4.69

4.11

0.19

81.53

Congo

10.10

6.39

0.54

16.40

1.54

65.04

Congo DR

6.95

5.07

0.09

15.02

0.48

72.38

Côte d’Ivoire

18.61

15.22

2.85

3.70

5.16

54.47

Ethiopia

7.81

6.30

2.06

1.70

1.52

80.62

Gabon

24.48

4.91

0.28

14.36

1.44

54.52

The Gambia

6.64

6.36

2.81

5.03

1.16

78.00

Ghana

18.44

5.00

4.72

14.52

8.44

48.88

Guinea

9.94

7.59

0.33

11.82

2.16

68.15

Kenya

7.46

1.27

2.32

19.17

1.05

68.72

Liberia

21.78

11.54

4.55

6.45

2.15

53.54

Malawi

5.73

1.20

3.21

4.50

0.81

84.54

Mali

9.23

8.34

1.50

1.51

79.42

Mozambique

6.14

2.99

2.12

5.13

1.90

81.72

Namibia

28.16

4.21

4.60

23.24

1.59

38.21

Niger

8.42

11.46

0.88

5.59

73.64

Nigeria

10.98

9.50

2.67

5.24

9.07

62.55

Rwanda

8.47

1.43

1.33

21.67

1.05

66.05

Senegal

11.42

8.79

1.18

0.65

77.96

Sierra Leone

9.24

8.92

1.94

17.61

1.26

61.03

Tanzania

6.48

5.48

0.47

23.19

64.37

Togo

7.01

4.60

1.98

17.67

5.08

63.66

Uganda

6.78

0.60

2.67

8.05

1.38

80.51

Zambia

14.07

9.05

0.58

9.94

0.22

66.13

Zimbabwe

8.41

3.68

2.72

3.40

1.97

79.81

117

A report from sub-Saharan Africa

Table A19 Shapley decomposition: Met need for family planning (older adolescent girls) Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Sex of the household head

Occupation

Religion

Marital status

Benin

11.59

4.06

5.88

4.15

10.11

64.20

Burkina Faso

27.99

29.08

1.56

3.69

4.96

32.71

Burundi

29.84

9.25

15.21

17.91

11.58

16.21

Cameroon

28.71

14.73

3.82

3.51

11.40

37.82

Comoros

13.91

28.41

2.28

8.95

0.46

46.00

Congo

13.76

8.13

0.69

6.77

8.00

62.66

Congo DR

13.77

15.48

2.42

4.10

1.74

62.50

CĂ´te dâ€™Ivoire

14.76

11.86

15.13

10.45

30.89

16.91

Ethiopia

21.57

11.15

8.97

9.95

26.85

21.51

Gabon

31.46

7.93

1.77

3.34

18.32

37.18

The Gambia

29.19

44.32

0.61

2.85

0.38

22.65

Ghana

20.04

1.80

15.13

8.48

14.82

39.73

Guinea

22.46

12.92

4.21

1.63

3.85

54.93

Kenya

28.41

22.78

2.80

29.57

2.87

13.58

Liberia

38.42

20.95

7.43

6.33

5.10

21.77

Malawi

20.17

8.61

6.54

6.40

14.31

43.96

Mali

33.77

36.99

2.31

1.99

24.94

Mozambique

44.15

25.76

0.96

4.69

4.13

20.31

Namibia

51.78

9.99

13.26

7.07

7.05

10.85

Niger

17.50

12.15

8.70

2.53

59.12

Nigeria

19.62

13.29

4.24

3.00

20.22

39.64

Rwanda

15.86

0.78

6.43

6.74

1.39

68.80

Senegal

30.19

33.70

15.23

2.12

18.76

Sierra Leone

10.81

8.73

2.96

14.47

8.94

54.09

Tanzania

43.45

20.62

5.89

14.51

15.53

Togo

30.48

16.37

4.17

9.51

7.30

32.17

Uganda

21.76

6.01

2.09

6.84

1.73

61.58

Zambia

7.71

12.39

10.65

1.55

0.99

66.70

Zimbabwe

7.29

16.75

4.45

3.29

4.54

63.68

118

Inequalities in women’s and girls’ health opportunities and outcomes

Table A20 Shapley decomposition: Not having anaemia Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Religion

Marital status

Sex of the household head

Education level

Age

Number of children

Benin

13.62

0.72

15.28

51.34

2.52

13.35

1.58

1.59

Burkina Faso

25.15

13.80

28.55

10.71

3.55

14.82

0.68

2.74

Burundi

32.40

2.03

1.78

34.52

8.85

14.28

2.62

3.52

Cameroon

21.51

11.69

23.54

14.98

1.72

20.65

1.70

4.22

Congo

22.13

30.55

15.14

20.42

0.93

3.65

4.16

3.02

Congo DR

14.78

13.55

5.52

11.23

2.16

32.40

14.52

5.83

Côte d’Ivoire

23.28

6.63

14.90

25.70

2.13

24.15

1.07

2.16

Ethiopia

11.45

10.68

30.16

13.40

0.35

18.50

8.02

7.43

Gabon

27.78

4.76

20.38

11.55

8.25

24.28

1.69

1.30

The Gambia

31.32

32.02

3.49

6.92

5.68

13.50

1.47

5.61

Ghana

43.00

4.77

4.23

8.19

6.63

12.34

17.21

3.63

Guinea

32.00

15.86

5.63

13.40

7.72

12.80

3.15

9.44

Malawi

21.42

8.82

28.35

20.30

4.39

7.13

5.44

4.15

Mali

29.49

16.49

6.38

6.02

6.90

25.45

2.27

6.99

Mozambique

45.94

8.64

10.40

8.68

0.83

23.66

0.90

0.94

Namibia

12.89

10.22

2.45

19.52

3.64

16.34

25.36

9.58

Niger

66.94

7.42

0.00

5.18

1.60

10.19

2.43

6.25

Rwanda

37.52

9.50

0.90

23.79

13.50

7.69

5.49

1.62

Sierra Leone

23.40

27.19

14.78

9.03

1.14

10.04

12.18

2.25

Tanzania

22.38

22.75

0.00

11.62

6.20

27.61

6.84

2.60

Togo

31.25

11.50

4.41

15.44

3.30

6.07

11.71

16.32

Uganda

28.73

5.53

4.13

20.80

1.41

6.04

13.33

20.03

Zimbabwe

9.72

22.16

3.86

23.50

18.82

4.21

13.66

4.08

119

A report from sub-Saharan Africa

Table A21 Shapley decomposition: Having the recommended BMI Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Religion

Marital status

Sex of the household head

Education level

Age

Number of children

Benin

29.19

15.93

8.93

5.47

1.82

7.23

25.88

5.56

Burkina Faso

25.36

15.65

5.17

8.94

0.94

9.47

29.35

5.13

Burundi

23.79

16.44

2.20

24.94

11.37

10.17

8.16

2.94

Cameroon

23.05

14.57

3.23

15.26

0.49

9.56

26.43

7.41

Comoros

9.29

6.36

0.43

35.06

0.59

6.54

24.13

17.60

Congo

25.85

19.92

3.18

8.69

2.19

8.98

25.74

5.44

Congo DR

40.67

21.24

2.37

3.39

2.93

8.45

16.65

4.29

CĂ´te dâ€™Ivoire

26.54

27.22

7.67

4.92

3.34

5.54

20.97

3.80

Ethiopia

13.23

9.59

2.69

30.61

7.19

20.21

10.21

6.28

Gabon

15.44

3.45

5.39

19.97

1.77

2.59

36.24

15.16

The Gambia

13.28

10.88

0.60

10.90

15.89

7.74

33.12

7.59

Ghana

30.34

12.57

2.18

16.48

1.82

10.04

21.40

5.17

Guinea

29.58

29.08

6.19

3.71

0.65

8.95

19.16

2.68

Kenya

30.35

12.88

2.22

13.22

0.24

5.08

29.81

6.19

Liberia

17.58

6.91

1.68

18.27

0.96

6.24

34.31

14.06

Malawi

35.67

19.41

0.41

5.62

0.94

7.54

22.68

7.73

Mali

33.20

29.28

2.76

2.21

0.56

7.21

18.85

5.92

Mozambique

38.13

17.72

4.52

7.30

1.17

9.10

19.33

2.72

Namibia

21.57

11.38

0.71

19.37

1.06

2.75

31.90

11.26

Niger

28.33

27.79

0.00

9.96

3.74

11.44

15.76

2.97

Nigeria

26.76

15.59

5.26

6.94

2.00

13.90

24.14

5.41

Rwanda

35.71

25.32

1.81

11.00

3.88

9.05

6.79

6.45

Sierra Leone

27.58

21.80

4.17

6.97

7.67

4.27

20.42

7.11

Tanzania

39.55

26.01

0.00

4.85

0.38

5.37

20.68

3.16

Togo

22.64

16.80

7.10

14.17

1.14

4.89

27.75

5.51

Uganda

24.67

22.28

2.11

7.16

5.14

11.89

22.25

4.50

Zambia

32.10

22.21

0.87

7.98

1.01

6.19

22.42

7.21

Zimbabwe

19.08

14.42

2.07

10.86

0.89

4.41

32.64

15.63

120

Inequalities in women’s and girls’ health opportunities and outcomes

Table A22 Shapley decomposition: Met need for family planning Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Religion

Marital status

Sex of the household head

Education level

Age

Number of children

Benin

12.07

6.21

7.10

25.02

7.18

19.31

12.88

10.23

Burkina Faso

28.55

20.52

7.48

8.23

1.59

21.94

2.15

9.54

Burundi

24.99

10.89

5.22

6.08

2.99

28.40

5.65

15.78

Cameroon

19.72

11.71

10.37

14.03

4.59

21.41

4.05

14.12

Comoros

11.71

34.81

0.12

21.09

0.71

17.78

6.13

7.67

Congo

15.89

7.12

6.98

27.30

3.83

12.24

9.79

16.85

Congo DR

21.22

20.38

0.91

10.49

0.54

30.58

5.55

10.35

Côte d’Ivoire

16.87

11.13

11.48

18.73

2.80

24.09

4.69

10.21

Ethiopia

26.27

16.17

12.78

6.50

0.74

12.28

5.15

20.11

Gabon

18.32

5.59

7.83

25.36

5.04

21.89

3.04

12.92

The Gambia

20.75

17.11

2.25

17.10

5.37

27.34

4.50

5.60

Ghana

8.56

2.30

8.58

40.98

2.56

13.92

5.03

18.06

Guinea

11.75

11.34

10.60

27.07

4.66

13.79

5.62

15.18

Kenya

33.59

8.58

10.49

7.60

2.86

14.58

2.79

19.51

Liberia

21.38

11.87

8.47

28.66

5.26

17.85

1.79

4.72

Malawi

19.83

6.61

14.76

28.49

6.08

12.07

9.35

2.80

Mali

41.09

26.28

1.47

4.33

1.14

17.49

1.36

6.83

Mozambique

30.65

19.87

3.95

9.91

1.14

20.20

3.28

11.00

Namibia

14.20

20.32

0.61

21.99

4.14

16.17

8.47

14.10

Niger

38.39

18.67

0.00

12.10

3.86

12.77

5.25

8.95

Nigeria

21.58

13.17

16.06

11.52

3.24

21.00

3.80

9.63

Rwanda

19.06

2.28

0.70

32.39

16.85

4.92

8.26

15.54

Senegal

16.59

20.98

2.07

24.05

1.00

27.75

2.73

4.82

Sierra Leone

16.50

16.24

5.29

31.20

3.29

15.14

3.59

8.74

Tanzania

38.23

14.17

0.00

10.54

1.55

14.08

4.36

17.07

Togo

9.87

7.34

11.21

29.30

3.98

26.49

4.28

7.53

Uganda

27.25

12.87

1.73

16.50

6.41

17.49

8.28

9.47

Zambia

29.59

22.34

0.84

12.01

2.17

17.35

3.58

12.12

Zimbabwe

27.84

8.28

4.06

35.97

4.53

12.10

0.98

6.23

121

A report from sub-Saharan Africa

Table A23 Shapley decomposition: Knowledge of a place where to get an HIV test Marginal contribution to the total inequality of opportunities (%) Country

Wealth index

Area

Religion

Marital status

Sex of the household head

Education level

Age

Number of children

Benin

34.97

17.59

9.54

6.27

1.76

27.05

0.78

2.02

Burkina Faso

26.01

22.61

12.30

10.64

1.24

23.42

2.26

1.53

Burundi

10.01

4.15

1.86

49.29

2.64

10.85

10.81

10.38

Cameroon

25.56

15.04

8.88

10.52

8.27

26.40

1.90

3.42

Comoros

24.59

15.13

0.12

15.77

0.41

35.12

4.80

4.06

Congo

27.81

16.66

4.27

8.71

0.86

29.87

9.11

2.72

Congo DR

35.19

30.11

0.57

4.26

0.63

25.18

2.61

1.46

CĂ´te dâ€™Ivoire

20.40

20.63

11.54

6.63

2.95

33.58

0.57

3.70

Ethiopia

27.40

18.72

6.38

5.84

2.85

30.91

2.21

5.69

Gabon

14.73

6.00

9.05

17.58

10.70

30.46

8.23

3.25

The Gambia

5.95

4.32

0.61

42.53

0.24

10.36

16.36

19.65

Ghana

29.57

16.07

4.88

9.89

2.20

29.24

4.66

3.49

Guinea

24.94

23.98

5.56

6.75

1.68

27.80

2.42

6.86

Kenya

14.82

8.29

9.95

23.17

0.56

23.99

12.91

6.30

Liberia

22.03

16.98

2.91

17.91

2.84

32.03

3.88

1.42

Malawi

9.01

2.12

1.78

38.66

1.38

25.67

10.44

10.94

Mali

34.02

32.22

1.38

3.96

0.88

24.21

0.96

2.37

Mozambique

37.36

20.11

7.65

10.83

1.12

19.28

1.12

2.52

Namibia

8.44

11.35

1.88

6.32

3.73

38.63

22.10

7.54

Niger

39.93

30.49

0.00

3.86

0.12

21.54

2.06

1.98

Nigeria

22.67

13.71

21.92

5.40

2.99

27.56

2.32

3.43

Rwanda

8.65

3.47

1.82

33.33

1.79

20.40

15.48

15.07

Senegal

25.91

18.95

1.04

9.55

3.39

27.04

10.50

3.63

Sierra Leone

22.29

24.63

3.25

21.61

2.84

20.69

2.52

2.17

Tanzania

17.90

13.61

0.00

20.54

1.70

25.53

12.66

8.06

Togo

28.07

22.79

10.19

11.16

2.62

19.92

1.60

3.66

Uganda

11.04

4.63

0.98

25.71

1.89

31.25

13.40

11.11

Zambia

7.05

4.08

1.43

34.26

1.34

15.20

20.85

15.79

Zimbabwe

10.21

3.64

1.96

35.36

2.20

17.70

19.95

8.99

122

Inequalities in women’s and girls’ health opportunities and outcomes

Table A24 Shapley decomposition: Four antenatal care visits Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Area

Sex of the household head

Religion

Marital status

Burkina Faso

2.82

9.82

16.90

27.22

28.12

1.22

8.80

5.11

Burkina Faso

2.82

9.82

16.90

27.22

28.12

1.22

8.80

5.11

Burundi

4.57

16.86

18.79

24.49

7.66

0.33

9.63

17.67

Cameroon

2.42

7.31

25.60

25.57

18.32

3.20

11.26

6.33

Comoros

5.38

13.78

34.64

32.46

3.17

4.64

1.39

4.54

Congo

1.92

10.07

25.24

31.34

24.46

0.16

4.33

2.48

Congo DR

2.15

7.87

22.92

30.60

29.71

0.38

2.10

4.27

Côte d’Ivoire

1.70

8.65

22.20

29.87

25.92

2.35

6.96

2.35

Ethiopia

2.79

8.83

23.08

35.20

22.26

1.64

3.33

2.88

Gabon

2.32

11.65

22.07

37.95

16.41

0.21

3.15

6.23

The Gambia

22.64

6.78

10.33

27.69

9.71

4.19

9.20

9.44

Ghana

2.05

9.29

22.07

32.78

17.13

1.45

8.83

6.41

Guinea

1.80

5.84

14.04

38.54

28.83

2.93

1.90

6.12

Kenya

2.60

14.72

15.99

36.63

21.24

0.93

3.52

4.36

Liberia

1.53

5.20

21.74

35.99

22.91

3.84

5.62

3.19

Malawi

6.54

18.95

22.29

32.90

7.05

4.85

0.92

6.50

Mali

0.71

4.56

14.82

47.24

29.28

0.42

1.60

1.38

Mozambique

2.55

6.16

15.34

38.31

19.23

1.85

2.90

13.66

Namibia

16.47

8.74

38.71

13.70

5.73

2.09

4.19

10.37

Niger

0.72

3.14

23.02

41.18

21.91

2.69

0.00

7.34

Nigeria

3.04

4.09

27.40

28.78

17.27

2.32

14.27

2.85

Rwanda

7.10

24.25

11.87

6.03

0.72

13.10

2.37

34.61

Senegal

2.88

14.37

14.14

38.98

13.07

4.17

0.55

11.86

Sierra Leone

7.68

11.90

22.50

22.90

20.17

0.81

3.34

10.70

Tanzania

3.24

11.41

12.55

32.77

24.43

3.37

0.00

12.23

Togo

1.29

7.75

11.34

35.43

27.74

1.57

11.92

2.96

Uganda

4.12

13.50

15.39

35.84

10.69

3.53

7.71

9.23

Zambia

20.23

4.71

13.67

43.65

10.20

0.73

1.79

5.02

11.61

15.42

28.42

20.56

5.77

0.69

3.33

14.19

Zimbabwe

123

A report from sub-Saharan Africa

Table A25 Shapley decomposition: Delivery attended by skilled personnel Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Area

Sex of the household head

Religion

Marital status

Benin

0.44

3.29

15.51

36.16

13.52

3.03

24.64

3.41

Burkina Faso

1.99

7.11

15.27

30.68

34.13

1.29

5.62

3.91

Burundi

15.73

25.35

22.39

11.35

2.04

1.57

5.83

Cameroon

1.65

6.19

24.46

29.66

17.32

3.10

8.94

8.68

Comoros

1.40

10.43

26.73

41.97

15.12

0.18

0.91

3.25

Congo

2.01

7.72

23.29

30.09

28.32

1.15

5.29

2.13

Congo DR

1.78

2.82

12.17

39.51

20.34

0.46

1.90

21.02

CĂ´te dâ€™Ivoire

1.46

6.91

12.83

32.01

37.19

1.63

5.82

2.15

Ethiopia

3.30

9.90

17.58

31.17

28.70

2.96

3.57

2.83

Gabon

1.51

6.47

19.63

34.21

28.78

0.19

6.37

2.84

The Gambia

1.35

6.48

12.98

26.67

44.96

4.38

0.73

2.46

Ghana

1.74

9.75

20.97

33.62

21.58

1.87

6.24

4.23

Guinea

1.96

6.03

14.70

34.90

31.94

1.73

2.56

6.18

Kenya

4.85

16.75

21.88

32.49

17.31

1.43

2.97

2.31

Liberia

2.30

5.62

20.76

33.26

24.57

4.94

3.91

4.64

Malawi

8.17

15.97

18.35

40.35

12.16

0.29

1.95

2.78

Mali

0.76

3.29

10.77

49.83

31.23

0.47

1.57

2.08

Mozambique

2.18

5.19

13.22

35.82

27.16

2.93

2.90

10.60

Namibia

4.95

15.82

32.37

18.15

16.29

4.13

2.54

5.76

Niger

1.89

7.02

18.72

36.43

33.17

0.58

0.00

2.19

Nigeria

2.65

5.65

26.24

25.46

16.85

2.14

18.11

2.90

Rwanda

14.25

27.47

12.05

22.16

7.99

6.02

1.38

8.66

Senegal

1.32

8.22

12.92

35.44

30.06

7.12

0.35

4.56

Sierra Leone

4.93

7.42

18.41

26.67

26.85

1.54

8.20

5.97

Tanzania

4.81

13.31

12.18

37.39

24.84

0.98

0.00

6.48

Togo

2.04

7.66

11.35

33.70

26.90

2.19

11.42

4.75

Uganda

4.77

10.29

22.65

31.10

19.07

2.69

6.22

3.21

Zambia

5.06

11.05

18.80

28.10

32.85

0.52

0.39

3.24

Zimbabwe

3.81

15.03

20.34

31.55

20.41

1.56

3.36

3.93

124

Inequalities in women’s and girls’ health opportunities and outcomes

Table A26 Shapley decomposition: Postnatal checkup Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Area

Sex of the household head

Religion

Marital status

Benin

1.18

4.69

18.37

35.42

16.96

1.28

17.38

4.72

Burkina Faso

3.85

12.74

8.92

37.73

15.34

0.53

18.63

2.25

Burundi

11.52

20.30

25.14

27.08

7.82

0.07

2.47

5.62

Cameroon

1.46

7.47

23.41

31.21

19.36

3.80

7.07

6.23

Comoros

1.93

10.51

20.48

37.37

20.78

2.82

0.75

5.36

Congo

2.84

6.09

21.62

33.33

27.83

0.15

3.27

4.88

Congo DR

2.73

3.53

27.34

31.43

31.31

0.66

1.29

1.72

Côte d’Ivoire

2.13

9.14

12.48

33.02

28.33

0.65

5.66

8.60

Ethiopia

2.26

7.87

18.17

32.56

26.51

4.01

4.76

3.86

Gabon

7.97

8.61

11.84

39.93

19.09

0.21

7.21

5.13

The Gambia

1.75

4.47

14.87

23.12

47.38

4.99

0.82

2.59

Ghana

2.14

9.92

20.47

33.51

17.21

2.13

9.96

4.65

Guinea

2.20

3.99

17.63

37.28

31.90

0.34

2.60

4.05

Kenya

3.84

15.64

23.07

28.47

17.42

1.83

3.69

6.03

Liberia

2.36

7.65

14.70

38.28

22.02

2.88

7.58

4.52

Mali

0.78

1.35

17.60

50.51

26.64

0.25

1.49

1.39

Mozambique

3.64

4.55

16.38

43.48

21.02

0.68

0.74

9.50

Namibia

3.50

12.13

34.42

20.63

10.77

1.58

5.27

11.71

Niger

1.12

6.06

21.48

38.24

29.76

1.00

0.00

2.33

Nigeria

1.80

6.31

27.63

27.81

14.45

2.04

16.85

3.12

Rwanda

7.08

23.39

17.48

21.80

4.06

4.33

1.65

20.21

Senegal

3.72

6.59

13.75

41.07

23.57

6.01

0.51

4.77

Sierra Leone

5.05

9.51

21.63

19.62

15.45

3.77

13.78

11.19

Togo

2.96

10.44

17.01

25.44

15.29

1.34

23.67

3.85

Uganda

4.02

11.93

25.04

30.75

17.81

1.00

2.95

6.50

Zambia

4.98

10.95

17.46

30.46

33.13

0.56

0.66

1.80

Zimbabwe

6.72

7.71

21.21

31.19

22.98

3.98

3.25

2.97

125

A report from sub-Saharan Africa

Table A27 Shapley decomposition: Maternity care package Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Benin

1.47

6.16

21.03

37.93

Burkina Faso

2.53

9.61

16.47

Burundi

6.86

21.36

Cameroon

2.31

Comoros

Sex of the household head

Religion

Marital status

14.26

1.67

16.11

1.37

29.00

29.05

1.48

7.39

4.46

24.67

28.11

8.59

1.58

3.27

5.56

7.97

25.10

29.03

18.10

3.59

8.39

5.51

6.60

13.98

32.06

33.76

8.10

0.17

0.50

4.83

Congo

1.75

8.04

21.86

33.48

26.68

0.33

3.85

4.02

Congo DR

1.64

6.44

23.11

31.64

32.51

0.89

1.71

2.06

CĂ´te dâ€™Ivoire

1.94

8.77

18.44

31.97

29.21

2.34

5.39

1.94

Ethiopia

2.24

9.13

17.84

32.59

28.84

2.73

3.77

2.86

Gabon

4.07

10.55

14.69

40.91

14.21

0.68

5.96

8.92

The Gambia

3.00

6.83

14.75

27.33

39.05

4.48

1.87

2.71

Ghana

1.63

9.89

20.84

34.97

20.79

1.57

5.87

4.43

Guinea

1.92

6.10

15.86

35.07

32.67

1.21

1.81

5.36

Kenya

2.95

15.39

19.20

34.79

21.99

0.52

1.47

3.69

Liberia

1.78

5.58

21.12

34.76

23.60

4.71

4.05

4.41

Mali

0.69

4.00

14.73

46.46

30.16

0.82

1.96

1.18

Mozambique

2.54

7.24

17.96

36.73

23.57

1.55

3.23

7.18

Namibia

4.42

13.24

38.10

19.08

13.57

0.84

3.32

7.43

Niger

1.04

6.40

21.85

37.06

29.18

0.96

0.00

3.50

Nigeria

3.31

6.01

26.28

28.42

17.22

1.75

14.64

2.37

Rwanda

7.51

34.46

16.50

11.81

3.52

1.69

1.43

23.08

Senegal

2.31

14.17

15.31

36.71

18.63

5.17

0.81

6.90

Sierra Leone

4.45

8.70

19.07

27.05

24.30

1.87

8.17

6.39

Togo

1.51

8.33

11.20

35.41

27.10

1.49

11.81

3.14

Uganda

2.60

9.83

23.92

35.89

18.57

3.40

3.97

1.80

Zambia

2.98

11.49

20.95

32.18

27.86

1.02

0.70

2.83

Zimbabwe

4.35

12.88

22.63

29.01

19.18

1.14

5.52

5.29

126

Area

Inequalities in women’s and girls’ health opportunities and outcomes

Table A28 Shapley decomposition: Malaria prophylaxis during pregnancy Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Area

Sex of the household head

Religion

Marital status

Benin

1.47

7.94

6.61

9.73

5.49

1.85

14.77

52.16

Burkina Faso

1.47

3.59

16.55

38.57

15.36

1.25

4.56

18.65

Burundi

3.46

10.73

8.83

10.16

0.28

10.78

5.90

49.87

Cameroon

3.51

6.64

22.63

14.12

1.82

1.18

8.27

41.83

Comoros

9.99

2.18

10.77

38.81

6.57

16.09

2.72

12.87

Congo

15.42

4.36

6.29

23.77

9.98

17.01

10.43

12.74

Congo DR

10.87

2.36

34.49

18.50

14.80

1.13

3.78

14.07

Côte d’Ivoire

12.76

4.23

22.03

21.65

9.65

2.87

16.59

10.21

Gabon

3.17

2.13

6.63

18.06

7.32

16.18

3.41

43.11

The Gambia

13.33

10.69

7.07

32.98

22.75

2.80

0.16

10.21

Ghana

0.92

1.51

8.62

10.40

16.11

1.54

20.60

40.30

Guinea

14.34

4.61

5.62

23.00

9.80

2.43

27.27

12.93

Kenya

8.06

12.88

8.28

21.20

5.18

2.82

27.86

13.72

Liberia

17.35

3.09

12.02

20.17

4.50

0.41

19.85

22.61

Malawi

7.37

5.69

27.34

30.57

12.55

2.12

5.80

8.58

Mali

4.88

0.89

9.98

62.19

10.50

0.31

6.99

4.26

Mozambique

4.65

6.05

15.99

18.98

28.29

1.11

10.49

14.42

Namibia

3.36

1.10

7.21

40.53

20.78

0.16

4.31

22.54

Niger

8.01

2.12

11.49

66.97

2.28

1.14

8.00

Nigeria

0.79

3.49

19.02

3.79

0.76

3.78

61.83

6.54

Senegal

8.89

1.93

26.42

27.72

17.69

7.50

1.17

8.68

Sierra Leone

2.99

8.91

28.20

26.59

20.42

0.21

2.16

10.52

Tanzania

6.06

5.57

22.70

38.51

14.74

1.03

11.39

Togo

10.71

12.12

16.70

11.30

28.37

1.82

13.27

5.70

Uganda

7.83

1.71

24.37

23.25

7.63

0.79

8.01

26.41

Zambia

14.39

6.18

17.17

28.42

19.18

1.42

6.25

6.98

Zimbabwe

10.61

6.03

7.10

24.49

30.53

1.34

2.12

17.78

127

A report from sub-Saharan Africa

Table A29 Shapley decomposition: HIV test offered during pregnancy Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Benin

0.57

6.69

18.49

35.63

Burkina Faso

2.51

7.88

15.30

Burundi

5.57

7.57

Cameroon

2.11

Comoros

Sex of the household head

Religion

Marital status

31.36

0.45

4.22

2.59

29.14

36.24

0.98

2.16

5.78

36.07

23.04

11.31

1.94

3.80

10.70

7.83

27.95

28.66

16.69

2.83

7.56

6.37

7.61

3.91

32.29

32.53

13.33

1.37

0.75

8.22

Congo

1.95

9.25

16.24

34.89

24.93

3.11

5.89

3.72

Congo DR

3.09

1.62

17.44

42.51

31.06

0.38

0.70

3.19

CĂ´te dâ€™Ivoire

0.84

4.58

14.13

25.94

40.63

2.80

4.70

6.38

Ethiopia

1.86

5.03

20.49

31.01

27.93

3.18

1.37

9.15

Gabon

13.65

2.21

14.80

37.23

7.34

6.38

12.51

5.88

The Gambia

7.59

2.89

4.67

18.54

41.73

1.74

0.69

22.15

Ghana

1.58

11.26

15.87

43.27

19.75

0.48

5.20

2.59

Guinea

1.57

8.36

19.67

31.11

28.59

1.04

2.98

6.67

Kenya

1.12

0.71

50.65

15.93

2.44

10.10

3.21

15.84

Liberia

2.79

7.96

27.12

26.14

24.01

3.26

2.07

6.66

Malawi

1.64

3.90

13.75

40.67

26.04

3.02

3.21

7.76

Mali

0.95

4.20

16.28

33.28

38.39

0.59

2.98

3.34

Mozambique

1.97

4.55

15.28

40.00

21.44

4.36

6.59

5.81

Namibia

2.04

3.65

41.61

5.97

13.34

4.54

9.08

19.78

Niger

1.48

3.74

14.32

37.29

39.19

1.30

0.00

2.68

Nigeria

3.38

5.94

28.27

25.50

17.04

1.98

15.03

2.85

Rwanda

8.52

14.25

8.17

40.72

5.56

0.64

4.13

18.01

Sierra Leone

3.16

9.01

21.05

26.09

24.84

3.45

4.97

7.43

Tanzania

2.89

8.67

19.78

30.99

26.22

1.67

0.00

9.78

Togo

1.10

6.60

12.80

35.53

27.00

1.60

12.70

2.66

Uganda

7.87

14.65

21.66

24.71

16.85

4.33

3.83

6.12

Zambia

4.28

8.58

23.76

28.96

29.48

0.28

1.54

3.12

Zimbabwe

3.19

5.53

22.68

33.56

19.35

4.96

3.84

6.90

128

Area

Inequalities in women’s and girls’ health opportunities and outcomes

Table A30 Shapley decomposition: Infant checkup within two months after delivery Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Area

Sex of the household head

Religion

Marital status

Benin

1.31

4.18

21.76

24.22

22.38

2.00

17.09

7.06

Burkina Faso

2.56

10.18

2.95

48.62

7.83

0.04

25.88

1.95

Burundi

8.43

18.50

8.52

32.45

6.93

4.77

5.13

15.26

Cameroon

3.43

3.05

16.17

36.17

17.36

4.60

11.81

7.41

Comoros

6.51

3.89

17.32

23.75

38.87

0.13

0.74

8.80

Congo

4.38

12.44

15.09

37.49

18.45

0.15

6.73

5.27

Congo DR

1.79

4.79

32.11

22.10

27.06

5.04

2.61

4.49

Côte d’Ivoire

1.91

1.03

12.90

45.89

2.79

1.43

8.65

25.39

Ethiopia

0.39

1.32

18.30

46.66

9.54

1.41

12.71

9.69

Gabon

7.30

5.20

19.82

30.25

11.39

1.74

3.91

20.39

The Gambia

2.72

7.71

12.93

20.49

44.51

1.89

0.25

9.49

Ghana

3.94

4.86

4.78

34.50

9.43

4.66

18.52

19.32

Guinea

2.75

6.69

15.42

31.21

26.46

0.49

12.99

3.99

Kenya

2.23

12.53

13.21

29.65

8.49

0.82

8.52

24.54

Liberia

0.62

1.57

21.95

18.99

6.17

4.22

28.72

17.77

Malawi

12.67

3.88

12.77

25.98

2.25

2.11

18.08

22.26

Mali

0.90

1.07

17.84

50.71

24.37

0.04

4.15

0.91

Namibia

16.67

8.44

15.68

18.77

4.09

0.87

4.63

30.85

Niger

3.49

7.00

29.67

30.75

19.27

2.18

0.00

7.63

Nigeria

1.91

5.66

28.08

28.38

16.59

1.99

14.35

3.04

Rwanda

2.08

7.54

11.07

38.59

27.64

1.12

5.47

6.48

Senegal

4.55

7.13

12.26

44.01

20.10

5.96

0.82

5.17

Sierra Leone

2.20

4.15

25.80

29.74

5.73

2.83

4.65

24.89

Tanzania

5.48

7.09

10.51

25.42

26.56

8.98

0.00

15.97

Togo

2.80

10.03

30.09

13.24

3.11

3.88

31.35

5.51

Uganda

3.42

10.59

14.67

39.70

12.78

2.04

2.70

14.10

Zambia

2.54

5.40

5.20

22.83

51.98

0.92

1.63

9.49

Zimbabwe

3.46

14.96

19.26

28.74

17.92

3.79

3.67

8.20

129

A report from sub-Saharan Africa

Table A31 Shapley decomposition: Six months of exclusive breastfeeding Marginal contribution to the total inequality of opportunities (%) Country

Age at birth

Number of children

Education level

Wealth index

Benin

8.40

5.62

14.19

15.70

Burkina Faso

12.17

21.99

13.40

Burundi

20.26

12.42

Cameroon

4.80

Comoros

Sex of the household head

Religion

Marital status

3.23

2.99

22.82

27.04

23.99

0.71

5.17

6.75

15.81

14.65

10.61

15.90

0.55

6.69

18.93

3.21

10.77

19.90

17.66

3.44

5.71

34.51

17.47

2.73

9.69

11.29

24.05

14.18

1.93

18.65

Congo

7.39

6.87

11.08

26.68

18.84

1.61

3.40

24.13

Congo DR

1.00

0.50

9.13

50.18

3.47

15.83

3.41

16.48

CĂ´te dâ€™Ivoire

2.08

1.49

13.66

41.50

3.67

0.70

18.79

18.10

Ethiopia

3.50

1.94

7.12

11.08

0.98

1.55

60.72

13.11

Gabon

5.30

2.69

25.21

24.24

2.43

7.99

14.57

17.58

The Gambia

1.70

4.95

7.06

45.32

22.89

8.16

1.97

7.94

Ghana

2.21

6.52

7.32

37.35

3.83

8.84

7.68

26.25

Guinea

6.53

2.15

19.75

22.42

4.76

14.16

3.86

26.36

Kenya

4.22

8.74

11.40

20.26

25.80

0.62

11.51

17.44

Liberia

2.92

13.22

17.63

13.24

2.42

17.65

8.08

24.84

Malawi

2.37

2.14

23.94

39.53

9.93

2.60

12.54

6.95

Mali

5.17

4.03

37.35

18.77

1.20

3.30

18.16

12.02

Mozambique

9.97

2.94

12.66

12.14

14.74

6.22

9.20

32.11

Namibia

0.69

7.29

26.94

42.39

8.39

0.50

4.25

9.56

Niger

3.79

7.64

25.26

18.16

3.00

7.95

0.00

34.20

Nigeria

0.75

6.79

22.35

36.23

17.63

0.49

4.66

11.10

Rwanda

5.46

17.02

22.34

17.40

6.57

5.09

1.59

24.52

Senegal

9.87

7.14

4.01

29.07

20.14

1.87

14.28

13.61

Sierra Leone

8.89

18.57

14.54

26.12

4.13

0.79

6.25

20.71

Tanzania

0.65

3.10

3.69

43.14

25.43

4.24

0.00

19.75

Togo

7.61

4.75

17.17

15.95

9.54

1.44

23.50

20.05

Uganda

2.53

6.33

18.43

27.15

5.49

5.74

12.56

21.77

Zambia

20.72

6.46

20.02

15.43

9.24

15.15

4.37

8.62

Zimbabwe

8.98

9.16

16.00

12.33

0.36

4.83

12.81

35.52

130

Area

Inequalities in women’s and girls’ health opportunities and outcomes

Table A32 HOI comparisons among African regions Opportunities

African regions – HOI (%) Unweighted analysis West

East

Central

Weighted analysis West

East

Central

Met need for family planning

32.09

36.63

52.06

34.62

39.56

42.09

Having never been pregnant

62.95

69.22

59.14

63.57

69.59

59.30

Currently attending school

31.13

41.78

53.58

30.82

41.27

46.78

Met need for family planning

35.42

53.45

53.72

40.86

53.36

43.14

Knowledge of a place where to get an HIV test

54.76

81.02

57.18

51.74

80.52

41.94

Not having anaemia

37.48

50.45

49.78

20.64

56.02

58.53

Having the recommended BMI

56.66

63.62

54.59

56.84

63.98

63.51

Four antenatal care visits

47.40

37.60

57.97

40.26

27.82

38.16

Delivery attended by skilled personnel

45.21

53.47

69.66

34.97

37.88

50.33

Postnatal checkup

63.59

35.97

56.41

48.15

26.29

44.39

Maternity care package

27.38

16.24

36.09

21.46

9.37

16.32

Malaria prophylaxis during pregnancy

62.02

38.43

34.99

47.28

31.20

37.60

HIV test offered during pregnancy

48.40

75.21

55.70

48.18

73.43

35.95

Infant checkup within two months after delivery

60.18

31.96

33.66

41.19

25.75

17.10

Six months of exclusive breastfeeding

80.81

74.95

67.84

75.48

76.24

70.85

131

A report from sub-Saharan Africa

Table A33 HOI comparisons between HIV prevalence regions Opportunities

HIV prevalence – HOI (%) Unweighted analysis >5

≤5

Weighted analysis >5

≤5

Knowledge of a place where to get an HIV test

57.28

90.54

52.51

90.08

HIV test offered during pregnancy

51.28

87.47

46.71

86.00

132

133

62.55

62.06

46.14

60.90

34.21

36.96

40.90

15.87

42.52

57.51

31.65

78.09

Having the recommended BMI

Met need for family planning

Knowledge of a place where to get al HIV test

Four antenatal care visits

Delivery attended by skilled personnel

Postnatal checkup

Maternity care package

Malaria prophylaxis during pregnancy

HIV test offered during pregnancy

Infant checkup two months after delivery

Six months of exclusive breastfeeding

HOI (%)

Not having anaemia

Opportunities

1.11

15.89

11.36

7.34

32.67

17.17

23.32

19.26

11.46

15.37

6.51

4.46

D-index (%)

0.87

5.98

7.37

3.37

7.70

8.48

11.24

8.16

7.89

8.38

4.32

2.92

Penalty (%)

78.96

37.62

64.88

45.88

23.58

49.38

48.20

42.37

68.78

54.52

66.38

65.47

Coverage (%)

0.52

0.18

0.37

0.27

0.12

0.19

0.17

0.18

0.14

0.25

0.17

0.22

SD HOI (%)

28.56

19.06

22.43

3.11

24.23

22.13

21.81

23.20

22.72

20.29

25.65

13.85

Wealth index

1.08

23.94

18.81

1.92

26.82

24.81

20.40

24.91

12.02

12.44

22.39

10.99

Area

3.18

4.24

4.12

1.41

1.66

2.68

2.64

1.94

4.97

2.54

1.83

1.55

Sex of the household head

25.90

23.24

26.60

15.88

27.60

26.27

27.20

30.71

30.42

23.59

12.32

21.14

Education level

13.41

5.90

12.51

65.77

4.33

5.70

10.23

4.83

18.79

15.05

2.25

38.32

Religion

9.27

8.12

5.59

6.39

3.45

6.94

7.48

4.71

6.19

9.34

4.85

10.76

Marital status

3.02

3.35

2.38

1.74

3.03

2.22

1.96

2.43

1.92

3.42

26.27

1.70

Age (at delivery)

Marginal contribution to the total inequality of opportunity (%)

15.59

12.15

7.57

3.78

8.88

9.25

8.29

7.26

2.97

13.34

4.45

1.68

Number of children

Table A34 Levels and Shapley decompositions for the multi-country pooled analysis: women of reproductive age and pregnant women

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

134

40.01

40.17

66.24

Currently attending school

Having never been pregnant

HOI (%)

Met need for family planning

Opportunities

15.30

23.73

15.39

D-index (%)

11.96

12.50

7.28

Penalty (%)

78.20

52.67

47.29

Coverage (%)

0.26

0.81

SD HOI (%)

8.66

12.18

22.61

Wealth index

4.79

5.65

17.77

Area

2.06

2.60

3.06

Sex of the household head

5.69

19.69

2.28

Occupation

3.56

12.26

20.99

Religion

75.24

47.63

33.31

Marital status

Marginal contribution to the total inequality of opportunity (%)

Table A35 Levels and Shapley decompositions for the multi-country pooled analysis: older adolescent girls

A report from sub-Saharan Africa

135

Having never been pregnant

Currently attending school

Met need for family planning

Opportunities

92.64

24.95

In union

3.70

In union

Never in union

60.31

28.90

In union

Never in union

53.64

HOI

Never in union

Marital status

8.81

1.29

37.64

9.87

18.27

8.44

Penalty (%)

2.41

1.21

2.23

6.60

6.56

4.95

Coverage (%)

27.36

93.85

5.93

66.91

35.36

58.59

SD HOI (%)

0.63

0.19

0.23

0.39

1.09

1.04

Wealth index

8.14

24.39

36.81

16.03

18.89

43.26

Area

5.07

5.12

18.61

5.15

12.54

42.90

Sex of the household head

6.73

26.10

8.77

0.69

0.80

1.28

Occupation

22.50

20.07

9.28

60.53

4.23

7.34

Religion

57.56

24.31

26.52

17.61

63.54

5.21

Marital status

Marginal contribution to the total inequality of opportunity (%)

Table A36 Levels and Shapley decompositions for the multi-country pooled analysis: older adolescent girls by marital status

Inequalities in womenâ€™s and girlsâ€™ health opportunities and outcomes

A report from sub-Saharan Africa

Table A37 Country data Country

UN Region1

Survey year2

Language2 (A/F)

Economy3

IMR3

MMR3

Population of women 15-493

HIV prevalence3

IPTp (2 or 3 doses) policy since4

Benin

Western

2011-2012

LIC

436

2311289

1.2

2005

Burkina Faso

Western

2010

LIC

417

3579459

1.1

2005

Burundi

Eastern

2010

LIC

808

2187036

1.6

Cameroon

Middle

2011

LMIC

652

4978935

2004

Comoros

Eastern

2012

LIC

365

178949

2003

Congo

Middle

2011-2012

LMIC

494

979629

3.1

2006

Congo DR

Middle

2013-2014

LIC

746

16167171

1.1

2004

Côte d’Ivoire

Western

2011-2012

LMIC

715

4701945

3.8

2005

Ethiopia

Eastern

2011

LIC

482

20811496

1.3

Gabon

Middle

2012

UMIC

314

387504

4.3

2003

The Gambia

Western

2013

LIC

730

439525

1.9

2003

Ghana

Western

2014

LMIC

322

6803551

1.5

2003

Guinea

Western

2012

LIC

695

2678217

1.6

2005

Kenya

Eastern

2014

LMIC

525

10853576

5.3

1999

Liberia

Western

2013

LIC

762

1002431

1.2

2004

Malawi

Eastern

2010

LIC

629

3297665

11.7

1993

Western

2012-2013

LIC

617

3517972

1.4

2003

Eastern

2011

LIC

596

5735866

2006

Namibia

Southern

2013

UMIC

283

624523

16.2

2005

Niger

Western

2012

LIC

619

3636832

0.6

2005

Nigeria

Western

2013

LMIC

821

39172542

3.3

2004

Rwanda

Eastern

2014-2015

LIC

304

2905877

2.8

2005 – until 2008

Senegal

Western

2014

LMIC

323

3546400

0.5

2004

Sierra Leone

Western

2013

LIC

1460

1492597

1.5

2004

Eastern

2010

LIC

514

10532046

6.1

2001

Western

2013-2014

LIC

386

1689457

2.5

2003

Uganda

Eastern

2011

LIC

408

7460696

7.1

2000

Zambia

Eastern

2013-2014

LMIC

237

3476200

12.6

2001

Zimbabwe

Eastern

2010-2011

LIC

446

3551962

2004

Mali Mozambique

Tanzania Togo

Note: all country data belong to the year of the particular survey. UN=United Nations. LIC=Low Income Countries. LMIC=Lower Middle Income Countries. UMIC=Upper Middle Income Countries. F=Francophone. A=Anglophone. IMR=Infant Mortality Rate. MMR=Maternal Mortality Ratio. IPTp=Intermittent Preventive Treatment of malaria in Pregnancy.

136

Inequalities in women’s and girls’ health opportunities and outcomes

References 1.

United Nations Statistics Division. Standard Country and Area Codes Classifications. at <http://unstats.un.org/unsd/methods/m49/m49regin.htm>

2. USAID. The DHS Program. at <http://www.dhsprogram.com/> 3. The World Bank Group. World Development Indicators. (2015). at <http://data.worldbank.org/data-catalog/world-development-indicators> 4. Van Eijk. A. M. et al. Coverage of malaria protection in pregnant women in subSaharan Africa: A synthesis and analysis of national survey data. Lancet Infect. Dis. 11. 190–207 (2011). 5.

137

WHO. Global Health Expenditure Database. at <http://apps.who.int/nha/ database>

A report from sub-Saharan Africa

Table A38 Circumstancesâ€™ variables codification Circumstance

Type of variable