A Tale of Two Continents: Explanations for Death Clustering in India and Africa

Nyovani Madise, University of Southampton
Zoe Matthews, University of Southampton
Alison Whitworth, Office for National Statistics, UK

This paper examines levels of infant death clustering using DHS data from twelve sub-Saharan African countries and twelve Indian states. Levels of death clustering are regressed with state and country-level economic indicators such as GDP and Gini coefficients as well as with development measures and health service uptake statistics. The results show that infant deaths are determined by known factors such as short birth intervals, the survival status of the previous sibling and socioeconomic factors in both regions. However, the magnitude of the estimates varies. Infant death clustering is more marked in India than in sub-Saharan Africa. HIV prevalence is related to infant death clustering within families only in Africa, where a higher prevalence is associated with less death clustering. Communal wealth, as measured by GNP or economic inequality indicators such as Gini coefficient are not very important in explaining infant mortality or death clustering.