A new multidimensional population health indicator for policy makers: absolute level, inequality and spatial clustering - an empirical application using global sub-national infant mortality data

The need for a multidimensional measure of population health that accounts for its distribution remains a central problem to guide the allocation of limited resources. Absolute proxy measures, like the infant mortality rate (IMR), are limi- ted because they ignore inequality and spatial clustering. We propose a novel, three-part, multidimensional mortality indi- cator that can be used as the first step to differentiate interventions in a region or country. The three-part indicator (MortalityABC index) combines absolute mortality rate, the Theil Index to calculate mortality inequality and the Getis-Ord G statistic to determine the degree of spatial clustering. The analysis utilises global sub-national IMR data to empirically illu- strate the proposed indicator. The three-part indicator is mapped globally to display regional/country variation and further highlight its potential application. Developing countries (e.g. in sub-Saharan Africa) display high levels of absolute mortality as well as variable mortality inequality with evidence of spatial clustering within certain sub-national units (hotspots). Although greater inequality is observed outside developed regions, high mortality inequality and spatial clustering are com- mon in both developed and developing countries. Significant positive correlation was observed between the degree of spatial clustering and absolute mortality. The proposed multidimensional indicator should prove useful for spatial allocation of healthcare resources within a country, because it can prompt a wide range of policy options and prioritise high-risk areas. The new indicator demonstrates the inadequacy of IMR as a single measure of population health, and it can also be adapted to lower administrative levels within a country and other population health measures.