Determinants of geographic variation in life expectancy in urban and rural settings

Objective: To illustrate the variation in life expectancy that occurs within a community and to examine the social, economic, and demographic characteristics that contribute to the variation.

Methods: Using multiple years of birth and death data, we calculated life expectancy by the smallest geographic unit possible within six cities and two rural communities. We also calculated measurements such as the poverty rate, racial segregation, and educational attainment for all communities as well as additional measurements for some communities such as food access, environmental hazard exposure, social cohesiveness, crime, availability of exercise facilities, and others. Using GIS software and multivariate statistical methods such as linear, non-linear, logistic, and ordinal logistic regression analysis, we analyzed the relationship between life expectancy and its determinants.

Results: Life expectancy within a single community varied by as much as 33 years between smaller geographic units. Lower life expectancy areas were significantly more likely to have higher poverty rates, higher percentage of adults without a high school diploma, large minority communities, and fewer community assets (such as supermarkets or healthcare providers) as well as to exhibit a persistent pattern of these metrics over 4 – 5 decades compared to high life expectancy areas. GIS maps for specific communities will be presented.

Conclusions: Regional statistics oversimplify important geographic differences that exist between small areas within a community. Geographic disparities in health status reflect, in part, past and present neighborhood contexts and social characteristics. Areas of persistent distress consistently have lower life expectancy and higher morbidity and mortality rates.