Perceived neighborhood safety as a moderator of activity participation frequency in suburban dwelling adults

Purpose: To understand the unique contribution of perceived neighborhood safety for physical activity (PA) participation among rural adults when other environmental correlates are considered.

Methods: A total of 441 residents (52% male, mean age 51.9 years) living outside the city limits of a mid-size southeastern US city returned self-administered questionnaires during fall and winter of 2007. One thousand single family households within 1.5 miles (Euclidean distance) of open space targeted for park development were invited to complete a questionnaire assessing their current physical activity levels ( I-PAQ short form), perceived neighborhood safety (Environmental Supports For Physical Activity Questionnaire), infrastructure for walking and cycling (Subsection of Saint Louis Environment and Physical Activity Instrument), and an investigator developed questionnaire which identified access to PA areas, preferences for the pending development and demographic information. The unique contribution of perceived neighborhood safety was determined using a blocked ordinary least squares regression.

Results: Individual age, race/ethnicity, level of income, and gender were entered in the first block. Socio-demographic factors explained for 3.8% (adj. R² = .038) of the variance in minutes of physical activity in the last 7 days. Next, known environmental correlates of activity participation were entered as a second block in the regression. Infrastructure for walking and cycling and reported access to PA facilities significantly improved the model (p < .01, R² change = 0.14). When perceived neighborhood safety items were added as a third block in the analysis, the model again improved significantly (p < .001, R² change = 0.15) with the inclusive model explaining 31.8% of respondents' variance in reported physical activity levels. Examination of standardized beta coefficients indicated the significant positive correlations between neighborhood safety (β = 0.143, p < .01) and infrastructure for walking and cycling (β = 0.115, p < .05) and physical activity. A significant negative association was observed between respondents' level of income and physical activity achievement (β = -0.055, p < .05).

Conclusions: The present results highlight the importance of perceived safety as a moderator of physical activity in suburban dwelling adults. Since modifications to the built environment (e.g., park development) can increase perceptions of safety, the results of this study support efficacy of such interventions to increase physical activity in adults.