Geo-additive models of childhood undernutrition in three sub-Saharan African countries

We investigate the geographical and socioeconomic determinants of childhood undernutrition in Malawi, Tanzania and Zambia, three neighbouring countries in southern Africa, using the 1992 Demographic and Health Surveys. In particular, we estimate models of undernutrition jointly for the three countries to explore regional patterns of undernutrition that transcend boundaries, while allowing for country-specific interactions. We use geo-additive regression models to flexibly model the effects of selected socioeconomic covariates and spatial effects. Inference is fully Bayesian based on recent Markov chain Monte Carlo techniques. While the socioeconomic determinants generally confirm findings from the literature, we find distinct residual spatial patterns that are not explained by the socioeconomic determinants. In particular, there appears to be a belt transcending boundaries and running from southern Tanzania to northeastern Zambia which exhibits much worse undernutrition. These findings have important implications for planning, as well as in the search for left-out variables that might account for these residual spatial patterns.