The demand for medical care in Kenya

Abstract

We use a quantile regression method to analyse the demand effects of user fees over the entire distribution of visits in a sample of rural and urban populations controlling for other covariates of interest, notably income and demographics. We find that the negative effects of fees on attendance differ across the visit quantiles, with the smallest effects being felt at higher quantiles. The main contribution of the paper is to show the non-uniform effects of fees at different points of the visits distribution, in contrast to the average effect computed using the OLS method. The policy implications of the findings are briefly discussed.