Background: Previous studies have shown poor correlation between subjective assessments of neighborhood safety and physical activity. This study was performed to model relationships between adolescents' outdoor physical activity and objectively measured violent crime densities, along with other key environmental variables.
Methods: Participants' physical activity was measured using a validated questionnaire. Multiple regression analysis was performed using outdoor physical activity away from school (OPA) as the dependent variable. Environmental factors included the density of violent crime within 1/2 mi of each participant's home, distance to nearest open play space, per capita income, and participants' subjective assessment of neighborhood safety.
Results: Multiple regression analysis showed that distance to nearest open play space was inversely and significantly associated with boys' OPA, while density of violent crime within 1/2 mi of home was inversely and significantly associated with girls' OPA. Girls' perceptions that their neighborhood was safe for physical activity were also associated with higher levels of OPA, while boys' assessments of neighborhood safety were not significantly related to OPA.
Conclusions: Neighborhood violent crime may be a significant environmental barrier to outdoor physical activity for urban dwelling Mexican-American adolescent girls. Future studies of physical activity and the environment should use objective measures of environmental factors.