Geographic Variation in U.S. Mortality 2004-2008: a Spatial Durbin Approach

Tse-Chuan Yang, Pennsylvania State University
Aggie J. Noah, Pennsylvania State University
Carla Shoff, Pennsylvania State University

Identifying the determinants of mortality in the US counties is not an unexplored area; however, previous studies often ignored the well-documented spatial dependence of mortality and focused on the relationships between mortality and explanatory covariates within a county. We challenge the literature by arguing that the mortality of a certain county should be associated with the features of its neighboring counties, and examine our argument with spatial Durbin modeling. Our theoretical framework is drawn from spillover and relative deprivation perspectives, and substantively, we found that the mortality of a focal county is positively related to health insurance coverage rates and affluence in neighboring counties and negatively associated with neighbors’ social capital and income inequality. The former echoes the relative deprivation viewpoint, whereas the latter confirmed the spillover perspective. Methodologically, spatial Durbin modeling outperformed the traditional analytic approaches in ecological mortality research–ordinary least square, spatial error, and spatial lag regression.