The diagnostic and prognostic value of symptom limited exercise tests (ET) performed before discharge and after one month were compared in men admitted to hospital after an episode of unstable angina or a non-Q-wave myocardial infarction (MI). A 'Positive ET' was defined as either a maximal work load below 100 W or ST-depression greater than or equal to 0.1 mV in 1-2 leads below 130 W or ST-depression greater than or equal to 0.1 mV in more than 2 leads at any load at the ET. During follow-up, severe angina was the only indication for coronary angiography and revascularization. There were no significant differences in diagnostic findings between the tests--Positive ET in 47% and Negative ET in 25% at both ETs. The occurrence of MI or death and the need of revascularization were related to signs of ischemia at both ETs. There were no differences in prognostic value between the early and late tests regarding MI or death or future severe angina during the 11 months' follow-up after the one month ET. However, half (10%) of the overall event rate (20%) during the one year follow-up occurred during the first months. The risk of these events could be identified by the predischarge but, for obvious reasons, not by the one month ET. Therefore, the present study suggests that a symptom limited ET should be performed before discharge in men stabilized after an episode of unstable angina or non-Q-wave MI.