Crime, perceived safety, and physical activity: A meta-analysis

Perceived safety from crime and objectively-measured crime rates may be associated with physical inactivity. The purpose of this meta-analysis is to estimate the odds of accumulating high levels of physical activity (PA) when the perception of safety from crime is high and when objectively-measured crime is high. Peer-reviewed studies were identified through PubMed, Web of Science, ProQuest Criminal Justice, and ScienceDirect from earliest record through 2016. Included studies measured total PA, leisure-time PA, or walking in addition to perceived safety from crime or objective measures of crime. Mean odds ratios were aggregated with random effects models, and meta-regression was used to examine effects of potential moderators: country, age, and crime/PA measure. Sixteen cross-sectional studies yielded sixteen effects for perceived safety from crime and four effects for objective crime. Those reporting feeling safe from crime had a 27% greater odds of achieving higher levels of physical activity (OR = 1.27 [1.08, 1.49]), and those living in areas with higher objectively-measured crime had a 28% reduced odds of achieving higher levels of physical activity (OR = 0.72 [0.61, 0.83]). Effects of perceived safety were highly heterogeneous (I2 = 94.09%), but explored moderators were not statistically significant, likely because of the small sample size. Despite the limited number of effects suitable for aggregation, the mean association between perceived safety and PA was significant. As it seems likely that perceived lack of safety from crime constrains PA behaviors, future research exploring moderators of this association may help guide public health recommendations and interventions.