BACKGROUND Abnormal heart rate recovery (HRR) has been shown to predict mortality. Although small studies have found that HRR can be improved with cardiac rehabilitation, it is unknown whether an improvement would affect mortality. The aim of this study was to determine whether HRR could be improved with cardiac rehabilitation and whether it would be predictive of mortality. METHODS AND RESULTS We evaluated 1070 consecutive patients who underwent exercise stress testing before and after completion of a phase 2 cardiac rehabilitation program. Heart rate recovery, defined as the difference between heart rate at peak exercise and exactly 1 minute into the recovery period, and mortality were followed up as the primary end points. Of 544 patients with abnormal baseline HRR, 225 (41%) had normal HRR after rehabilitation. Of the entire cohort, 197 patients (18%) died. Among patients with an abnormal HRR at baseline, failure to normalize after rehabilitation predicted a higher mortality (P<0.001). After multivariable adjustment, the presence of an abnormal HRR at exit was predictive of death in all patients (hazard ratio, 2.15; 95% confidence interval 1.43-3.25). Patients with abnormal HRR at baseline who normalized afterward had survival rates similar to those of the group with normal HRR at baseline and after cardiac rehabilitation (P=0.143). CONCLUSIONS Heart rate recovery improved after phase 2 cardiac rehabilitation in the overall cohort. There was a strong association of abnormal HRR at exit with all-cause mortality. Patients with abnormal HRR at baseline who normalized HRR with exercise had a mortality similar to that of individuals with baseline normal HRR.