Health effects of sanitation facilities: a Bayesian semiparametric analysis of compositional data
Diarrhoeal disease is one of the leading causes of illness and death in young children: a problem exacerbated by a lack of access to safe sanitation facilities. But the effects of different types of sanitation facilities, and the relationship between their levels of coverage in an area and disease risk, remain uncertain. We specify a hierarchical model that enables us to estimate the disease risk response surface across the multi‐dimensional space of the composition of sanitation facilities in an area. This permits a non‐linear relationship between coverage of improved sanitation, differential effects by sanitation type and unobserved differences between countries. The model is used to estimate the change in risk associated with different types of sanitation facility, from which it is estimated that only increases in the coverage of sewerage and private improved sanitation facilities will achieve clinically meaningful reductions in the risk of adverse outcomes.
No Supplementary Data
No Article Media