Comparing and decomposing the determinants of multiple health outcomes in southern and northeastern US states using county data

Abstract

Objectives: To investigate whether regional health disparities could explain health outcome differences between US regions. Methods: The 2010-2012 County Level Raking database, which contains measures on health outcomes and factors for the US states, are used in this study. First, a regional comparison of the determinants of various health outcome measures (premature death rates, low birth weight rate and mean unhealthy days) is undertaken. Next, using the Blinder-Oaxaca decomposition method, the differences in health outcomes are partitioned into the portion explained by regional health factor endowments and the portion unexplained. Key findings: Obesity rate and access to care determinants impact strongly each region's health outcomes. Half of the differences in clinically assessed (premature death rates and low birth weight rates) measures are unexplained by health endowments. The explained portion for the self-assessed (unhealthy days) outcomes is small. Conclusions: Study findings both justify the imperatives of accounting for regional variations in order to strengthen policy inferences of research findings and suggest the prevalence of regional disparities in health outcomes. It is important to focus on region-specific preventive care. More effective public education and information strategies raising population awareness to the health hazards of obesity are in order. Greater access to various health services personnel should be provided to the relevant population. Implications are discussed for pharmaceutical care. JPHSR