Since 1980, average life expectancy in the United States has increased by roughly five years; however, in recent years it has been declining. At the same time, spatial variation in life expectancy has been growing. To explore reasons for this trend, some researchers have focused on morbidity factors, while others have focused on how mortality trends differ by personal characteristics. However, the effect community characteristics may play in expanding the spatial heterogeneity has not yet been fully explored. Using a spatial Durbin error model, we explore how community and demographic factors influence county-level life expectancy in 2014, controlling for life expectancy in 1980 and migration over time, and analyzing men and women separately. We find that community characteristics are important in determining life expectancy and that there may be a role for policy makers in addressing factors that are associated with lower life expectancy in some regions.

Introduction

In 2016, for the first time since 1962-63, average American life expectancy declined for two consecutive years (The Economist, Jan 6, 2018). Although American life expectancy is expected to increase by 2040, if current trends continue, the United States (US) is expected to fall 21 spots in world life expectancy rankings from 43rd to 64th (IHME, 2018) resulting in a life expectancy lower than China’s. Kontis et al. (2017) project that the US will fall farther behind in the future due to rising negative health outcomes and lack of universal health insurance. However, focusing on the national average masks the dramatic increase in the variation of life expectancies across the US. From 2010 to 2015, census tract-level average life expectancy ranged from 56 to 97 years; the lower number is on par with Somalia and the higher number exceeds the average for Japan, which has the world’s highest overall life expectancy, by 13 years (Arias et al., 2018; The Economist, Sep 25, 2018). Female life expectancy has been higher than that of males since the 19th century (Goldin and Llera-Muney, 2018). However, consistent with the theoretical model in Leung et al. (2004), as more women have entered the workforce, the gap between female and male life expectancy has shrunk. From 1980 to 2014, overall female life expectancy at birth in the US increased from 77.5 to 81.5 years, while overall male life expectancy increased from 70.0 to 76.7 years (IHME, 2016).