An exercise tolerance test (ETT) is often performed to identify patients for early discharge after observation for acute chest pain, but the safety of this strategy is unproven. We prospectively studied 276 low-risk patients who underwent an ETT within 48 hours after presentation to the emergency department with acute chest pain. The ETT was considered negative if subjects achieved at least stage I of the Bruce protocol and the electrocardiogram showed no evidence of ischemia. There were no complications associated with ETT performance. The ETT was negative in 195 patients (71%); there was no identifiable subsets of patients at very low probability of an abnormal test. During the 6-month follow-up, patients with a negative ETT had fewer additional visits to the emergency department (17% vs 21%, respectively; p < 0.05) and fewer readmissions to the hospital (12% vs 17%; p < 0.01) than those with positive or inconclusive ETTs. No patient with a negative ETT died and only 4 patients with a negative ETT experienced a major cardiac event (myocardial infarction, coronary angioplasty, or bypass) within 6 months. Among these 4 patients, only 1 had an event within 4 months. In conclusion, our results suggest that ETT can be safely used to identify patients at low risk of subsequent events. Patients without a clearly negative test are at increased risk for readmission and cardiac events, and should be reevaluated either during the same admission or shortly after discharge.