The effect of changing the built environment on physical activity: a quantitative review of the risk of bias in natural experiments
BACKGROUND Many assume that improving the quality and the perceived safety of facilities in parks and recreation centers is critical to attracting more users and increasing population physical activity. There are few studies in which these assumptions have been tested. PURPOSE This study aims to assess the impact of park improvements on park use and physical activity. METHODS Five intervention parks and five matched comparison parks were studied by objectively measuring park use and collecting self-reports of park use by residents before and after park improvements. After using the System for Observing Play and Recreation in Communities to count park users and measure their activity levels, and conducting household interviews and intercept surveys with park users, propensity score analyses were used to adjust for differences in respondents' characteristics between pre- and post-intervention and across conditions. RESULTS Overall park use and physical activity declined in both intervention and control parks, with 39% of the decline directly attributable to fewer scheduled organized activities. Perceptions of park safety increased more in the intervention parks than in the comparison parks. CONCLUSIONS Improvements to parks may not automatically result in increased use and physical activity, especially when programming decreases. Multiple factors contribute to park use and need to be accounted for in future community-level interventions. Further, improving perceptions of safety alone is unlikely to result in increased park use.