BACKGROUND: accurate risk adjustment is needed to guide quality improvement initiatives and research to improve care of patients with acute myocardial infarction (MI). We developed and validated a model to predict the risk of in-hospital mortality for contemporary patients with acute MI treated in routine clinical practice. METHODS: the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry-Get With The Guidelines (GWTG) database of patients with acute MI was used to derive (n = 65,668 from 248 US sites) and validate (n = 16,336) a multivariable logistic regression model to predict the likelihood of in-hospital mortality (4.9% in each cohort). RESULTS: factors with the highest independent significance in terms of mortality prediction included age, baseline serum creatinine, systolic blood pressure, troponin elevation, heart failure and/or cardiogenic shock at presentation, ST-segment changes, heart rate, and prior peripheral arterial disease. The model showed very good discrimination, with c statistics of 0.85 and 0.84 in the derivation and validation cohorts, respectively. The model calibrated well overall and in key patient subgroups including males versus females, age <75 versus ≥ 75 years, diabetes versus no diabetes, and ST-elevation MI versus non-ST-elevation MI. The ACTION Registry-GWTG in-hospital mortality risk score was also developed from the model. Patients with a risk score of ≤ 40 had an observed mortality rate of <4% compared with those with a risk score of 41-50 (12%) and risk scores >50 (34%). CONCLUSION: the ACTION Registry-GWTG™ in-hospital mortality model and risk score represent simple, accurate risk adjustment tools for contemporary patients with acute MI.