Essays on Patient and Provider Behaviors for Maternal and Child Health

Abstract

This thesis addresses research questions pertaining to maternal and child health using a multi-level perspective, building on existing evidence that emphasizes the essence of considering demand-side, supply-side and contextual factors in advancing health. The dissertation uses quantitative methods to particularly answer the following questions: 1. Assessing whether accuracy of perceived quality levels at facilities predicts pregnant women’s utilization of better care for delivery; 2. Investigating whether rewards and penalties predict superior provider effort in the context of primary care in Lebanon; and 3. Analyzing whether certain providers within the Palestinian health system experience higher demand for maternal care during conflict. The dissertation uses multiple datasets to answer these questions. The first and second papers are secondary analyses and use primary data collected for a randomized control trial in Kenya and an observational study in Lebanon, respectively. The third paper utilizes publicly available data from three sources: Nationally representative surveys of women within the reproductive age, and data on population size, and casualties. Altogether, the papers apply several types of models, including multi-level Poisson models and multivariate logistic and linear regressions. The papers also use various sensitivity analyses to assess the robustness of the findings.