Patients with non-ST-elevation chest pain constitute a heterogeneous population. Our aim is to compare the outcome of patients with chest pain, non-ST-segment deviation, and normal troponin, categorized using a risk score, with that of patients with ST depression or troponin increase. A total of 1,449 patients with non-ST-elevation chest pain were evaluated. A validated risk score (using pain characteristics and risk factors) was applied to patients without ST depression or troponin increase. Accordingly, 4 risk categories were defined: group 1, no troponin increase, no ST depression, and risk score <3 points (n = 633); group 2, no troponin increase, no ST depression, but risk score > or = 3 points (n = 158); group 3, no troponin increase, ST depression (n = 106); and group 4, troponin increase (n = 552). Median follow-up was 26 months, and the end point was death or myocardial infarction. Group 1 experienced fewer events at 30 days (1.7%, p = 0.0001) and long-term follow-up (9.4%, p = 0.0001) than groups 2 (10.8% and 26%), 3 (6.6% and 30%), and 4 (9.5% and 25%). Kaplan-Meier curves overlapped among groups 2, 3, and 4, whereas group 1 showed a flatter curve (p = 0.0001). Using multivariate analysis, risk group (group 1 vs remaining groups) predicted 30-day (p = 0.0003) and long-term (p = 0.0001) outcome. There were no differences among groups 2, 3, and 4. In conclusion, application of a risk score to patients without troponin increase or ST deviation identified a high-risk group with prognosis similar to that of patients with troponin increase or ST depression and affords a practical classification for the full spectrum of non-ST-elevation chest pain.