The Effects of Neighborhood Violence on Physical Activity Following the COVID-19 Stay-at-Home Order: A Multilevel...

COVID-19 contributed to decreases in physical activity and residents from communities with more crime and fewer resources were less physically active during the pandemic. However, the direct effects of COVID-related increases in crime on leisure-time physical activity has not been studied, extensively. Therefore, we examined how individual perceptions of neighborhood violence, community-level perceptions of safety, and the prevalence of violent crime are associated with leisure-time physical activity across neighborhoods in Chicago, Illinois, following the COVID-19 stay-at-home order.

We used cross-sectional data to build a multilevel logistic regression model in which the individual represented level 1 and the Chicago community area represented level 2. We used a stepwise approach, testing variables, one at a time, as random intercepts and then as random coefficients. We estimated adjusted odds ratios and 95% confidence intervals.

Respondents who reported violence occurring every day had significantly lower odds of leisure-time physical activity (OR = 0.65, 95% CI = 0.50-0.85). Compared to non-Hispanic Whites, Non-Hispanic Blacks (OR = 0.75, 95% CI = 0.57 - 0.99), non-Hispanic Asian American/Pacific Islanders (OR = 0.63, 95% CI = 0.45-0.89), and Hispanic/Latino respondents (OR = 0.69, 95% CI = 0.53-0.89) had lower odds of leisure-time physical activity. On the neighborhood level, the percent of adults who reported they feel safe in their neighborhood was significant (p < .001), but violence prevalence was not (p = 0.091).

These results can inform public health efforts to promote physical activity following the COVID-19 stay-at-home order.