BACKGROUND: Association between body mass index (BMI) and long-term mortality is poorly studied in older hospitalized populations. METHODS: The researchers prospectively studied the impact of the BMI, comorbidities, and malnutrition on long-term mortality in 444 patients (mean age 85.3+/-6.7 years; 74.0% women) receiving geriatric inpatient care. All-cause mortality was determined using simple and multiple Cox proportional hazard models. RESULTS: Higher BMI was associated with a higher prevalence of diabetes, hypertension, and heart failure, but with a lower prevalence of malignancies. Four-year all-cause mortality was inversely associated with a BMI greater than or equal to 30kg/m(2) (hazard ratio = 0.59, p = .037) and positively associated with age, male gender, several individual comorbidities, and the global disease load determined by the Cumulative Illness Rating scale. The inverse association between a BMI greater than or equal to 30 and mortality remained significant after adjustment for age, gender, smoking, individual comorbidities (including heart failure and malignancies), Cumulative Illness Rating scale scores, and malnutrition parameters (hazard ratio = 0.52, p = .015). One-year mortality was associated with the Cumulative Illness Rating scale score but not with BMI categories. There were no survival differences between patients in low (<20.0) and intermediate (20.0-24.9 and 25.0-29.9) BMI categories. CONCLUSIONS: A BMI greater than or equal to 30 is associated with better long-term survival in hospitalized older patients, even after extensive adjustment for comorbidities, malnutrition, and smoking. Conversely, a low BMI (<20-25) is not associated with excess mortality, likely due to the overriding impact of multiple comorbidities. The researchers' observations have important implications for the mortality risk stratification in older high-risk patients.