OBJECTIVES/GOALS: We aim to extend a novel statistical method called the Semi-Supervised Mixture Multisource Exchangeability Model (SS-MIX-MEM) and to implement the SS-MIX-MEM approach to supplement ALPS-COVID data with N3C data to achieve analyses with greater precision and actionable conclusions. METHODS/STUDY POPULATION: We will apply the SS-MIX-MEM to supplement the Angiotensin receptor blocker-based Lung Protective Strategy for COVID-19 (ALPS-COVID) RCTs with the National COVID Cohort Collaborative (N3C) database. ALPS-COVID includes both an inpatient and outpatient trial, which investigate losartan as a treatment for COVID-19. The outpatient trial sought to randomize 580 individuals but only enrolled 117, whereas the inpatient trial met its enrollment target and randomized 205 individuals. The N3C database has 3,237,344 COVID-19 cases alongside demographics, lab values, and more. RESULTS/ANTICIPATED RESULTS: In simulation studies, the proposed SS-MIX-MEM approach effectively leveraged a subgroup of supplemental real world data for RCT analyses, improving trial efficiency by increasing precision of treatment effect estimates, decreasing necessary sample size, and introducing minimal bias. In an influenza trial real world data application, the SS-MIX-MEM approach was able to effectively provide insight into treatment effect heterogeneity found in an RCT analogous to incorporating around 80 individuals into a subgroup analysis. We anticipate that leveraging external real world data in a re-analysis of the ALPS-COVID RCTs could provide new insights into losartan, a readily available, potentially beneficial therapeutic for COVID-19. DISCUSSION/SIGNIFICANCE: The high blood pressure drug, losartan, is readily available, has an established safety profile, and might be effective as a treatment for COVID-19. Given that we have very few effective treatment options and are still in the midst of a global pandemic, patients with COVID-19 would greatly benefit from a repurposed, readily available treatment.