Relationship between measures of urban form and body mass index in non-urban diabetic adults: Results from Vermont

Objectives: We analyzed the relationship between measures of urban form and body mass index (BMI) across a settlement gradient ranging from rural areas to small regional cities. Methods: Using data from the Vermont Diabetes Information System on 610 adults with diabetes, we regressed BMI against several measures of urban form, controlling for a number of potential confounders as well as spatial autocorrelation. Results: We found a strong positive correlation between BMI and the density of commercial buildings within 250 m of the subject's home. There was about a 1 kg/m2 increase in BMI for each additional commercial facility per hectare. However, location within a designated downtown district was associated with a reduction in BMI of 1.2 kg/m2. Conclusions: The positive association of BMI and density is contrary to much of the previous literature and suggests a different mechanism in a predominantly rural and exurban region like Vermont than in the large metropolitan areas where most of the previous research was conducted. This association may relate to rural residents' greater access to outdoor recreation opportunities or to the physical requirements associated with rural property maintenance. The effect of downtown location suggests that additional design characteristics beyond density—possibly mix of uses, pedestrian infrastructure, or difficulty in parking— result in downtowns being associated with lower BMI.