Prior violent injury and subsequent homicide involvement: evidence from a linked trauma registry study.

BACKGROUND: Violence-related injury is a major source of preventable morbidity and mortality in urban communities. This study quantified the overlap between non-fatal assault-related hospitalisations and subsequent homicide involvement and examined the neighbourhood socioeconomic contexts in which both occurred. METHODS: We linked data from a level I trauma registry (2013-2022) and a citywide homicide surveillance system (2018-2022) to identify individuals involved in homicide as victims or perpetrators following prior hospitalisation for violent injury. Demographic and injury characteristics were compared by survivors of violence that did and did not have homicide involvement. We assessed neighbourhood socioeconomic conditions at residence and injury locations using the Index of Concentration at the Extremes for income and evaluated spatial clustering using Moran's I. RESULTS: Among 3409 patients admitted for assault-related injuries, 67 (2%) were subsequently involved in a homicide as a victim or perpetrator. Compared with trauma patients without homicide involvement, these individuals were younger, predominantly male and far more likely to have sustained firearm-related injuries. In adjusted analyses, prior trauma contact was strongly associated with homicide involvement (adjusted OR 16.5, p<0.001), with independent associations also observed for higher injury severity scores and firearm injury mechanism. Residential and injury locations across all cohorts concentrated in socioeconomically deprived neighbourhoods with strong spatial clustering. CONCLUSION: Hospital encounters for violent injury identify individuals at elevated risk for homicide involvement within structurally disadvantaged neighbourhoods. These findings highlight trauma care settings as potential points for secondary prevention to reduce future lethal violence.