Aim: Reporting both relative and absolute measures of association allows for a comprehensive assessment of the public health impact of exposures. Our aim was to quantify the association between selected neighborhood characteristics (and demographics) and the outcome of leisure-time physical activity among school employees working on the Texas-Mexico border.

Methods: Subjects were school employees working in El Paso County who participated in a pilot study of a walking challenge intervention trial.  Cross-sectional analyses were performed in which the binary outcome was the frequency of leisure-time physical activity during the majority of the past year: Subjects who spent most of their leisure time without engaging in much physical activity (the sedentary group) were compared to those who were more active. Predictors included neighborhood environment variables. Adjusted prevalence ratios (PR), adjusted prevalence differences (PD) and 95% confidence limits (CL) were calculated.

Results: Mean age of the 80 employees was 44.4 y. Of the 80, 35% were classified as sedentary. No associations on the relative scale between the following neighborhood environment factors and the outcome of being sedentary were found: Crime rate; high traffic; and presence of sidewalks. Subjects >=50 y of age were 92% more likely than those <50 to be sedentary (Adjusted PR=1.92, 95% CL:  1.11, 3.31, p=0.02). The age-adjusted PD for subjects who somewhat agreed or strongly agreed that the crime rate in their neighborhood makes it unsafe to go on walks at night was -22.6% (95% CL: -42.3%, -2.9%, p=0.02).

Discussion: We found a positive correlation between age and sedentary behavior during leisure time on the relative scale and an inverse association between neighborhood crime and sedentary behavior on the absolute scale. It is possible that individuals who live in neighborhoods that are not conducive to walking outdoors may be engaging in physical activity at home, work, and/or in fitness centers.