Purpose Critically ill patients are at risk of stress ulcer related GI bleeding. To prevent gastrointestinal (GI) bleeding in critically ill patients, proton pump inhibitors (PPI) is the most commonly used for stress ulcer prophylaxis followed by Histamine 2 receptor antagonists (H2RA). The relative efficacy and harms of gastrointestinal bleeding prophylaxis with PPIs, H2Ras remains unclear. Methods This is a retrospective, observational, comparative cohort study using EHR data of Seoul National University Bundang Hospital, converted to the Observational Medical Outcomes Partnership Common Data Model. Among patients who admitted medical ICU and required invasive mechanical ventilation between May 1, 2003 and July 31, 2018, those prescribed either PPI or H2RA more than 2 days after ICU admission were enrolled. GI bleeding, pneumonia, Clostridioides difficile infection, and all-cause mortality within 90 days during index hospitalization were compared between PPI and H2RA group by Cox regression analysis after propensity score matching. Results Before matching, 1641 patients and 2128 patients were identified in PPI and H2RA group respectively. After propensity matching, 872 individuals from each group were analyzed. No significant difference was found in GI bleeding, pneumonia, Clostridioides difficile infection. In subgroup analysis of patients with an APACHE Ⅱ score of 25 or higher, there was also no significant difference in GI bleeding between PPI and H2RA. The 90-day mortality of the PPI group was 27.6%, which was significantly higher than the H2RA group in which 21.1% died. [Hazard ratio 1.33; 95% confidence interval 1.1-1.62; P<0.001]. Conclusions There was no difference in the efficacy of GI bleeding prophylaxis and the incidence of pneumonia and Clostridioides difficile infection between PPI and H2RA. However, the 90-day mortality was higher in PPI group compare to H2RA group. These findings do not support current recommendations to prefer PPI for GI bleeding prophylaxis in ICU patients.