Revisiting the Rural Paradox in US Counties with Spatial Durbin Modeling

The rural paradox refers to the phenomenon that the standardized mortality rates are lower in rural than in urban areas despite the relatively poor socioeconomic profiles among rural residents. Previous research on the geographic mortality differential has failed to recognize the complexity of the concept of rurality and the spatial structure underlying the ecological mortality data has not been fully utilized to advance our understanding of the rural paradox. Drawing from the drift and breeder hypotheses, this study first uses county-level data to measure “rural” with three distinct aspects, namely ecological dimension, economic integration, and natural resources dependency. Then, it employs the spatial Durbin approach to capture the exogenous relationships between the mortality of a county and the features of its neighbors. The key findings include that (1) the drift hypothesis (i.e., internal migration) did not appear to explain the rural paradox, but the breeder hypothesis (i.e., exposures to environments) partially accounts for the rural-urban mortality disparity, (2) the associations between the ecological dimension and economic integration with mortality were explained after accounting for the exogenous relationships, (3) the observed spatial feedback effects reflected the spatial dynamics across county boundaries, and (4) the spatial dynamic processes between mortality and its determinants were largely confined to the first- and second-order neighbors. The results of this study indicate that future ecological mortality research should further utilize the spatial structure to explain the variation of mortality across space.