Geographic variation in rural-urban obesity rates in older adults: Evidence from a national survey

Obesity affects over one-third of older adults in the US.  By 2050, the US population aged 65+ is expected to more than double to 88.5 million.  Both aging and obesity contribute to increased risk for chronic disease, risk of early mortality, and health care utilization.  Previous studies have found that obesity rates are higher in rural areas than in urban areas, although findings are mixed.  Therefore, the objective of this study is to determine if the association between rurality and obesity in older adults varies by state.  Using the 2012 Behavioral Risk Factor Surveillance System, we obtained national obesity data on adults aged 60+.  Logistic regression modeled the association between obesity and rurality using the Index of Relative Rurality.  Three crude and sociodemographic-adjusted models were used: (1.) for the entire US, (2.) stratified by state with national-level confounders, and (3.) stratified by state with state-specific confounders.  Overall, as urbanicity increased, the likelihood of obesity decreased (OR 0.961; 95%CI 0.951-0.972).  Each 10% increase in urbanicity resulted in a 3.9% decrease in the likelihood of obesity.  However, the stratified models revealed that although this significant association remained for nine of the fifty states, two states (Kentucky and Wyoming) showed a significant association in the opposite direction: obesity was more prevalent in urban areas.  These findings demonstrate that the overall rurality-obesity associations observed in older adults vary by geography.  Traditional “one-size-fits-all” approaches to reducing rural-urban health disparities in older adults may be more effective if tailored to the area-specific rural-urban gradients in health.