Youth violence is a leading cause of adolescent mortality, underscoring the need to integrate evidence-based violence prevention programs into routine emergency department (ED) care. To examine the translation of the SafERteens program into clinical care. Hospital staff provided input on implementation facilitators/barriers to inform toolkit development. Implementation was piloted in a four-arm effectiveness-implementation trial, with youth (ages 14-18 years) screening positive for past 3-month aggression randomized to either SafERteens (delivered remotely or in-person) or enhanced usual care (EUC; remote or in-person), with follow-up at post-test and 3 months. During maintenance, ED staff continued in-person SafERteens delivery and external facilitation was provided. Outcomes were measured using the RE-AIM implementation framework. SafERteens completion rates were 77.6% (52/67) for remote and 49.1% (27/55) for in-person delivery. In addition to high acceptability ratings (e.g., helpfulness), post-test data demonstrated increased self-efficacy to avoid fighting among patients receiving remote (incidence rate ratio [IRR] 1.22, 95% confidence interval [CI] 1.09-1.36) and in-person (IRR 1.23, 95% CI 1.12-1.36) SafERteens, as well as decreased pro-violence attitudes among patients receiving remote (IRR 0.83, 95% CI 0.75-0.91) and in-person (IRR 0.87, 95% CI 0.77-0.99) SafERteens when compared with their respective EUC groups. At 3 months, youth receiving remote SafERteens reported less non-partner aggression (IRR 0.52, 95% CI 0.31-0.87, Cohen's d -0.39) and violence consequences (IRR 0.47, 95% CI 0.22-1.00, Cohen's d -0.49) compared with remote EUC; no differences were noted for in-person SafERteens delivery. Barriers to implementation maintenance included limited staff availability and a lack of reimbursement codes. Implementing behavioral interventions such as SafERteens into routine ED care is feasible using remote delivery. Policymakers should consider reimbursement for violence prevention services to sustain long-term implementation.