Objective: To evaluate the prognostic characteristics of body mass index (BMI) and standard exercise test variables in a consecutive series of patients with mild to moderate congestive heart failure (CHF) referred for standard exercise tests. Background: Controversy exists regarding the prognostic importance of BMI, etiology, and exercise test variables in patients with CHF. Methods: All patients referred for evaluation at two university-affiliated Veterans Affairs Medical Centers who underwent treadmill tests for clinical indications between 1987 and 2000 were determined to be dead or alive using the Social Security Death Index after a mean 6 years follow-up. Clinical and exercise test variables were collected prospectively according to standard definitions; testing and data management were performed in a standardized fashion using a computer-assisted protocol. Survival analysis was performed using all-cause mortality as the endpoint for follow-up. Results: A total of 522 patients with a history and clinical findings of CHF underwent exercise testing. Forty-two percent died during the follow-up period, for an average annual mortality of 6.7%. Cox proportional hazards model chose peak metabolic equivalents (METs), BMI, age, and ischemic etiology in rank order as independently and significantly associated with time to death. A score based on these variables classified patients into low (2% annual mortality), medium (5.2%), and high-risk groups (7% annual mortality). Conclusion: Standard exercise testing and BMI can be used to estimate prognosis in outpatients with heart failure. A score incorporating METs, BMI, age, and etiology efficiently stratified these patients. BMI was chosen by the survival analysis, confirming its surprising inverse relationship to prognosis in CHF patients (i.e., heavier patients do better).