LAUSR

Women use parks less for physical activity than men, and explanations include gendered concerns regarding personal safety and access to walking paths. This study conducted mediation analyses to examine the effects of gender and presence of park walking path on park use, participation in park programs, and park-based physical activity through the hypothesized mediator (perception of crime). The sample included 3213 park users from 48 parks in high poverty neighborhoods in Los Angeles surveyed between 2013 and 2015; park-level factors were assessed through systematic observations of study parks. Women reported fewer park visits than men in the last 7 days (β = -0.17, p = 0.02) and had significantly higher perceived crime (β = 0.12, p < 0.0001) and perceived crime partially mediated the gender association with park visits (βgender, direct = -0.09, p = 0.19; βgender, indirect = -0.07, p < 0.0001). Similarly, the existence of a walking path in the park was significantly related to increased park use (β = 0.27, p = 0.006) and a lower level of perceived crime (β = -0.25, p = 0.0034) and perceived crime partially mediated the association of walking path with park visits (βwalking path, direct = 0.18, p = 0.10; βwalking path, indirect = 0.15, p = 0.005). The associations between gender, walking path, and park-based exercise and program participation were not meaningfully mediated by perceived crime. Among park users in majority Latino, high poverty neighborhoods, addressing crime concerns are likely necessary to increase park use among women and adults whose parks do not have a walking path. For park-based exercise and participation in park programs, gendered preferences regarding park-based physical activity should be explored.