Infant Mortality and Social Environment in Georgia: An application of hotspot detection and prioritization

Recent years have witnessed the growth of new information technologies and their applications to various disciplines. The goal of this paper is to demonstrate how the two innovative methods, upper level set scan (ULS) hotspot detection and the multicriterion prioritization scheme, facilitate population health and break new ground in public health surveillance. It is believed that the social environment (i.e. social conditions and social capital) is one of the determinants of human health. Using infant health data and 10 additional indicators of social environment in the 159 counties of Georgia, ULS identified 52 counties that are in double jeopardy (high infant mortality and a high rate of low infant birth weight). The multicriterion ranking scheme suggested that there was no conspicuous spatial cluster of ranking orders, which improved the traditional decision making by visual geographic cluster. Both hotspot detection and ranking methods provided an empirical basis for re-allocating limited resources and several policy implications could be drawn from these analytic results.