Neighborhood Characteristics and Physical Activity

Our goal is to understand how selected neighborhood characteristics influence self-reported physical activity. Study data come from the CDC funded Nashville REACH 2010 project. Telephone interviews conducted with Nashville residents (n=11,167) asked about demographic characteristics, health conditions, and health behaviors. Sidewalk maps were obtained from the Metro Nashville Planning Department. Using GIS software, a half-mile buffer was created around each survey respondent and the total area of sidewalks within that half-mile was calculated. Population and housing characteristics were obtained from the U.S. Census for each block group in Nashville, TN (n=467) and the total area of sidewalks for each block group was calculated. Dependent variables included binary physical activity (yes/no) and an activity index measuring frequency, duration, and intensity of exercise behaviors. People reporting regular physical activity had significantly less sidewalk area within 0.5 miles of their home address (p<0.0001) than people reporting no physical activity. People who report no physical activity had significantly greater access to sidewalks than people in the upper two quartiles of those reporting some activity on the Activity Index (p<0.05). The median year in which housing was constructed and the total area of sidewalks were significantly correlated (p<0.05). Individual perception of neighborhood safety was associated with physical activity (p<0.0001). Multilevel models using census data were better predictors of physical activity than were individuals' access to sidewalks. Multilevel models also showed that sidewalks were more common in older, poorer neighborhoods. Policies for increasing physical activity may have to focus on issues of safety and socioeconomic inequality.