Search

Background: Health-seeking behaviour is a key contributor to the widespread and unfair inequalities in health outcomes related to socio-economic position. This thesis compared the levels and determinants of maternal and child health-seeking between a national sample and the rural poor in Egypt, and examined whether existing inequalities could be explained by socio-cultural characteristics or ability to afford care. Methods: This quantitative analysis relied on two datasets: the Egypt Demographic and Health Survey from 2008 and a 2010/11 survey of households below the poverty line in rural Upper Egypt. Latent variables capturing several dimensions of socio-economic position were constructed and used in multivariable regression models to predict several dimensions of maternal (antenatal and delivery care) and child (diarrhoea and acute respiratory infection) health-seeking. Results: Latent constructs capturing socio-cultural and economic resources were identified in both datasets. Two further dimensions of socio-economic position in the Upper Egypt sample included dwelling quality and woman’s status. DHS analysis showed that sociocultural and economic capital were independently positively associated with seeking antenatal and delivery care among women, and with seeking timely and private child illness treatment. Free-of-charge public maternal care was not effectively targeted to poorest women. Poor households in Upper Egypt showed lower maternal healthseeking levels than nationally; both socio-cultural and economic resourcefulness positively predicted maternal health-seeking, dwelling quality was positively associated with private provider use, while women’s status was not associated with any dimension of maternal health-seeking behaviour. Conclusion: A better understanding of perceived and objective quality of care in both public and private sectors is required to reduce existing inequalities in the coverage of essential maternal and child health interventions. Improvement in free public care targeting is required to prevent catastrophically high expenditures for basic care among poor households.

Poverty and long travel time are barriers to using skilled care at birth, especially care provided at hospitals which can be located far and result in high direct and indirect costs. In parts of sub- Saharan Africa, about one third of births occur in hospitals. This thesis aimed to assess the relative contributions of poverty and long travel time to the probability of giving birth in a hospital in Kenya, Malawi, Nigeria and Tanzania. I first reviewed the literature related to measuring the distance/travel time between women and health facilities in sub-Saharan Africa. Although the measurements and standards adopted by included studies were diverse, the impeding effect of living far from health facilities on use of childbirth care was prominent. In the second study, we compared two approaches to create high-resolution poverty maps in Kenya, Malawi, Nigeria and Tanzania. We found that the spatial variation in poverty and its determinants differed across countries, which should be considered when choosing the most suitable mapping approach. For each country, we used the better-performing approach to construct a national poverty map. These maps showed the highest concentration of poverty in remote locations, where population density was low and the allocation of resources potentially expensive. Next, we assessed the wealth inequality in travel time to the nearest hospital and its trade-off against minimizing overall travel time in the four countries. Travel time was calculated by overlaying locations of the population, wealth subgroups and hospitals. We simulated alternative hospital locations to identify the shortest overall travel time and the narrowest equity gap possible. Results suggest that hospitals in the four countries are currently well placed to minimize overall travel time, but they create wide inequality gaps by wealth. Lastly, we assessed the relative contributions of poverty, travel time, and other factors on the probability of hospital birth in the four countries. Poverty and travel time were important, and they played different roles within and across countries, meaning different strategies are needed to increase hospital-based childbirth. Nonetheless, these strategies alone do not address all barriers, and further research of where they do not lead to the desire result is required to help devise tailor-made actions.

This thesis assesses the effects of having pluralistic systems of health financing and service provision on universal healthcare coverage with a case study on maternal health in Kenya. Through five research papers using a mix of systematic literature review, qualitative, and quasi-experimental quantitative methods, this thesis answers three primary research questions. First, how do researchers measure the contribution of the private sector to maternal health and family planning service provision and how much care does the private sector provide in sub-Saharan Africa (SSA)? Second, how did Kenya's pluralistic financing policies and public-private engagement strategies for health arise and evolve over time? Finally, what are the impacts of user fee removals and subsidized vouchers on use, sector, quality, continuity, and equity of maternal care in Kenya? The findings from the systematic review suggest that there is substantial heterogeneity in the way that the private health sector is defined in scientific literature, making it difficult to compare estimates of private sector health provision. The qualitative study reveals that Kenya's pluralistic health system results from the confluence of many historical, social, political, and economic factors and effective lobbying by the private for-profit sector. Finally, the three quasi-experimental studies highlight a complex set of outcomes resulting from user fee removal policies and the safe motherhood voucher program in Kenya. The 10/20 policy was associated with positive effects on the timing and number of ANC visits; however, these improvements were unrelated to use of the public primary care facilities that the policy targeted. The voucher program increased use of facility-based delivery care among poor women; however, it had no impact on use of four or more ANC visits or postnatal care. After the free maternity services policy was introduced, the voucher program no longer improved use of facility-based delivery among the poor; however, use of the private sector remained much higher in voucher counties. Both the voucher program and insurance coverage had positive impacts on continuity of maternal care for poor women, while introduction of the free maternity services policy did not. Many factors affect women's use of maternal health services beyond the cost of care. Making services free in the public sector is not sufficient to eliminate disparities in access to health services; policymakers must therefore simultaneously address both financial and nonfinancial barriers to service use. Health financing strategies involving private providers have the potential to equitably increase service use and continuity, provided that the cost of care is subsidized for users with the lowest ability to pay.