Interactive Effects of Built Environment and Psychosocial Attributes on Physical Activity: A Test of Ecological Models

BACKGROUND: The principle of cross-level interactions of influence on behavior in ecological models is seldom studied. PURPOSE: To examine built environment × psychosocial interactive effects on physical activity. METHODS: Multi-level mixed regression analyses used data from the Neighborhood Quality of Life Study conducted in neighborhoods in two US regions (n = 2,199 adults). Outcomes were (1) objectively measured moderate-to-vigorous physical activity, (2) reported transport walking, and (3) leisure-walking. Conceptually matched built environment variables were analyzed for domain-specific outcomes. RESULTS: With leisure walking as the outcome, built environment × psychosocial interactions were significant in 7 of 20 models tested. Directions of interactions were consistent, indicating a stronger built environment-leisure walking association in adults with less favorable psychosocial status. Little evidence supported such interactions with objectively measured moderate-to-vigorous physical activity or transport walking as outcomes. CONCLUSION: The results imply that the built environment may exert stronger influence on adults who are not psychologically predisposed to be active.