Temporal and spatial shifts in gun violence, before and after a historic police killing in Minneapolis

Objective: To determine the impact of the police murder of George Floyd in Minneapolis, MN on firearm violence, and examine the spatial and social heterogeneity of the effect. Methods: We analyzed a uniquely constructed panel dataset of Minneapolis Zip Code Tabulation Areas from 2016–2020 (n = 5742), consisting of Minnesota Hospital Association, Minneapolis Police Department, Minneapolis Public Schools, Census Bureau, and Minnesota Department of Natural Resources data. Interrupted time-series and random effects panel models were used to model the spatiotemporal effects of police killing event on the rate of firearm assault injuries. Results: Findings reveal a rising and falling temporal pattern post-killing and a spatial pattern in which disadvantaged, historically Black communities near earlier sites of protest against police violence experienced the brunt of the post-killing increase in firearm assault injury. These effects remain after adjusting for changes in police activity and pandemic-related restrictions, indicating that rising violence was not a simple byproduct of changes in police behavior or COVID-19 response. Conclusions: The results suggest that the increases in firearm violence as a result of police violence are disproportionately borne by underserved communities.