UCLA

Despite substantial global progress in reducing child mortality, the burden remains high in sub-Saharan Africa. Although affordable and effective treatments exist, many children die each year because of poor access to medical care. Expanding access to treatment and reducing barriers to care are key to improving child health outcomes. My three-paper dissertation examines how access to health services was associated with care utilization and child mortality in sub-Saharan Africa. In Chapter 2 (Study 1), I evaluated the association between two dimensions of access -- geographic distance and quality of care -- on care-seeking during childhood illness among rural households in Malawi. I used geospatial methods to link national household survey data with health facility data, to estimate households’ distance to health facilities and operationalize the quality of health services within households’ service environments. In accordance with previous literature, I found that longer distances to care and poor health facility quality were associated with reduced care utilization of sick child care: each additional kilometer in distance between households’ residence and health facilities was associated with a 5% reduction in the odds of care-seeking (aOR 0.95, 95% CI 0.91- 0.98; p<0.05); and those living in high quality health service environments were 36% more likely to have sought sick child care compared to mothers living in areas with low quality service environments (aOR 1.36, 95% CI 0.99 – 1.86, p=0.05). In an innovative new approach of evaluating the interaction between these two components (distance and quality), I found evidence of a trade-off between these two factors: the effect of higher health service quality on care-seeking decreased as the average distance to health facilities increased, indicating that geographic distance to facilities may be the most important influence on sick child care utilization. In Chapter 3 (Study 2) I examined how policies to make health care free for children affected child mortality in sub-Saharan Africa. I used a quasi-experimental difference-in-differences research design to compare countries that have, and have not, removed user fees (out-of-pocket payments) for children’s health services. I found that removing these fees was associated with a 19% reduction in the odds of infant mortality (aOR 0.81, 95% CI 0.72-0.91, p<0.001), and a 26% reduction in the odds of under-five mortality (aOR 0.74, 95% CI 0.70 – 0.87, p<0.001). The effects on reduced child mortality were strongest in the first year after the policy change, and attenuated over time. I expanded on these findings in Chapter 4 (Study 3) by examining differential effects of under-five fee removal policies across three key socioeconomic indicators: household wealth quintile, household residence (rural versus urban), and level of maternal education. I found that the impact of user fee removal on child mortality varied across all socioeconomic categories, and the largest reductions in child mortality were observed among children from the poorest households (1.7 percentage point reduction, 95% CI -2.42 - -0.009, p<0.001) and those residing in rural areas (1.4 percentage point reduction, 95% CI –0.019 - -0.008, p<0.001). Removing user fees also significantly narrowed socioeconomic disparities in child mortality across all indicators (household wealth, household residence, and level of maternal education). The greatest impacts of removing user fees (biggest reductions in child mortality) were observed among children from the poorest households: the gap in the predicted probability of child mortality between the wealthiest and poorest households prior to fee removal was 2.0 percentage points, and decreased to 0.6 percentage points after the policy change. Together, these papers provide new insights into factors influencing child outcomes in the highest-burden settings, which can be used to inform future research and policymaking about how to improve service coverage and access, and strengthen standards of health care delivery.