Mortality associated with gastrointestinal bleeding events: Comparing short-term clinical outcomes of patients hospitalized for upper GI bleeding and acute myocardial infarction in a US managed care setting

OBJECTIVES: To compare the short-term mortality rates of gastrointestinal (GI) bleeding to those of acute myocardial infarction (AMI) by estimating the 30-, 60-, and 90-day mortality among hospitalized patients. METHODS: United States national health plan claims data (1999-2003) were used to identify patients hospitalized with a GI bleeding event. Patients were propensity-matched to AMI patients with no evidence of GI bleed from the same US health plan. RESULTS: 12,437 upper GI-bleed patients and 22,847 AMI patients were identified. Propensity score matching yielded 6,923 matched pairs. Matched cohorts were found to have a similar Charlson Comorbidity Index score and to be similar on nearly all utilization and cost measures (excepting emergency room costs). A comparison of outcomes among the matched cohorts found that AMI patients had higher rates of 30-day mortality (4.35% vs 2.54%; p < 0.0001) and rehospitalization (2.56% vs 1.79%; p = 0.002), while GI bleed patients were more likely to have a repeat procedure (72.38% vs 44.95%; p < 0.001) following their initial hospitalization. The majority of the difference in overall 30-day mortality between GI bleed and AMI patients was accounted for by mortality during the initial hospitalization (1.91% vs 3.58%). CONCLUSIONS: GI bleeding events result in significant mortality similar to that of an AMI after adjusting for the initial hospitalization.