Multisite Evaluation of Emergency Department Revisit Rates Among Participants in a Hospital-Based Violence Intervention Program in a Safety-Net Hospital System

Interpersonal violence is a leading cause of ED visits in the US, with high rates of recurrent injury and utilization disproportionately affecting the structurally marginalized. Hospital-based Violence Intervention Programs (HVIPs) engage patients during acute care use and aim to disrupt cycles of violent reinjury, yet evidence of their impact on ED revisits remains limited. This study examines ED revisit rates among patients presenting with interpersonal violence, compared with those enrolled in the Los Angeles (LA) County Hospital-based Violence Intervention Program (LAHVIP).

This retrospective cohort study used encounter level data for adult patients from two safety net hospital emergency and inpatient departments in LA from 2022-2024. Interpersonal violence was defined as any interpersonal violence documented via triage or diagnosis including stabbing, shooting, hit and run, physical or sexual assault. Groups were defined as patients presenting for interpersonal violence who were enrolled in the 6-month HVIP (treatment) and those not HVIP engaged (control). Analyses estimated 12-month rates of ED revisits for HVIP patients versus controls. We employ logistic regression controlling for age, sex, and LAHVIP site to compare the likelihood of a revisit between groups.

A total of 419 patients in the LAHVIP enrolled group were matched to their ED records vs 10,787 among the interpersonal violence group not engaged by LAHVIP. Rates of revisits for interpersonal violence for treatment and control patients at 0-12 months were 5.9% and 3.8%, respectively. Rates of revisits for interpersonal violence at 6-12 months, after completion of the 6-month program, were 0.72% for the treatment group and 0.92% for the control group. Rates of all cause visits among treatment and control groups at 0-12 and 6-12 months, were 18.4%, 5.7%, and 3.8%, and 1.2% respectively. After regression adjustment, odds of revisits for interpersonal violence at 6-12 months were 0.43 [95% CI: 0.12, 1.54] for treatment compared to control patients.

HVIP interventions at these two EDs in LA County demonstrated higher relative reductions in ED visits and interpersonal violence vs controls after 6-months. These data may demonstrate that HVIPs could be a useful tool in decreasing patient re-presentation to the emergency department with future episodes of interpersonal violence.