INTRODUCTION: Variceal hemorrhage (VH) is associated with high mortality. Patients require urgent referral to GI specialists, acid suppression, somatostatin analogues, and antibiotic prophylaxis. Our institution noted a delay in initiating vasoactive agents like octreotide in residents' patients suspected to have VH. Therefore this project was initiated to create an electronic order set to help address the delay in initiating octreotide in patients suspected to have VH. Hepatology guidelines recommend the use of proton pump inhibitors, vasoactive agents, and antibiotic prophylaxis the moment a physician suspects VH [1]. Delays in initiating therapy not only can worsen patient outcomes but can increase liability. METHODS: An order set was created to include three medications (pantoprazole, octreotide, and ceftriaxone). A total of 200 patients were randomized & divided into two groups. The baseline group had a total of 50 patients in which no order set was used. The order set group had 45 patients. Each patient met the Baveno VI criteria. Three years of retrospective data was used to evaluate the primary outcome of minutes to medication onset. The secondary outcomes included start time variability between medications, mortality, days hospitalized, & number of blood products. RESULTS: Overall the minutes of onset for pantoprazole, octreotide, and ceftriaxone showed no statistical difference between the baseline & order set group (P = 0.935). All three medications did not have a reduced onset time when the order set was used. There was however a statistical difference in start time variability between the three medications (P = 0.022). The order set group had a median variation of 48 minutes. The baseline group had a median variation of 121.5 minutes. Other outcomes such as mortality, days of hospitalization, days in intensive care, and number of units of blood transfused were all statistically insignificant and showed no difference between the groups. CONCLUSION: Overall this project did not achieve its goal of reducing minutes of onset for initiating octreotide. Though the onset of initiating all medications could not be improved, there was a significant reduction in the start time variability. The order set group had decreased variability by 73.5 minutes when compared to the baseline group. This analysis did provide the insight that increased start time variability did not translate to clinically worse outcomes. Further studies will be needed to show the efficacy of order sets in improving VH outcomes.Table 1.: Demographic Characteristics of Patients in the Baseline and Order Set GroupsTable 2.: Minutes To Onset Of MedicationsTable 3.: Other Secondary Outcomes