PROJECT ABSTRACT Injury is the leading cause of loss of independence among those over the age 65, resulting in over 3 million Emergency Department (ED) visits, 800,000 hospitalizations, and greater than $50 billion in costs each year. Guideline-concordant triage of trauma patients – rapid identification of those with severe injuries and transfer to trauma centers – decreases mortality by 10 to 25%, reduces loss of independence, and diminishes pain at one year. Despite this evidence, however, under-triage persists (incidence: 50%), particularly among those over the age of 65 (incidence: 80%). To address this implementation gap, we propose to conduct a Type I hybrid effectiveness-implementation trial in which we test the effect of video game-based training on physician behavior. We designed this trial based on extensive preliminary studies, during which we isolated physician behavior as the single largest source of variation in triage practices, identified heuristics (pattern recognition) as a major cause of behavior, developed theory-based, customized video games to recalibrate physician heuristics (i.e., align them with clinical practice guidelines), and ensured their efficacy in the laboratory. For this trial, we will recruit a national sample of emergency medicine physicians (n=900), in collaboration with three large staffing organizations, and will randomize physicians to one of three interventions: video game-based training, text-based education (active control), or nothing (passive control). The specific aims of the trial are (1) to test the effect of the three interventions on behavioral (e.g., under-triage) and patient-centered outcomes (e.g., 30-day mortality) by linking trial data to Medicare claims, (2) to compare the mechanisms of action by which the active interventions affect physician behavior by using a suite of validated instruments, and (3) to identify contextual factors that influence implementation and maintenance of guidelines in practice by performing a comprehensive qualitative evaluation of trial participants (n=20) and a national sample of key decision makers (e.g., ED directors, patients [n=20/group]). Together these aims will provide critical insight into the effectiveness of our novel intervention and key data to inform future implementation efforts.