Association of Home‐Based Cardiac Rehabilitation With Lower Mortality in Patients With Cardiovascular Disease: Results From the Veterans Health Administration Healthy Heart Program

Background Home‐based cardiac rehabilitation (HBCR) and traditional facility‐based cardiac rehabilitation (CR) programs have similar effects on mortality in clinical trials and meta‐analyses. However, the effect of HBCR on mortality in clinical practice settings is less clear. Therefore, we sought to compare mortality rates in HBCR participants versus nonparticipants. Methods and Results We evaluated all patients who were referred to and eligible for outpatient CR between 2013 and 2018 at the San Francisco Veterans Health Administration. Patients who chose to attend facility‐based CR and those who died within 30 days of hospitalization were excluded. Patients who chose to participate in HBCR received up to 9 telephonic coaching and motivational interviewing sessions over 12 weeks. All patients were followed through June 30, 2021. We used Cox proportional hazards regression models with inverse probability treatment weighting to compare mortality in HBCR participants versus nonparticipants. Of the 1120 patients (mean age 68, 98% male, 76% White) who were referred and eligible, 490 (44%) participated in HBCR. During a median follow‐up of 4.2 years, 185 patients (17%) died. Mortality was lower among the 490 HBCR participants versus the 630 nonparticipants (12% versus 20%; P<0.01). In an inverse probability weighted Cox regression analysis adjusted for patient demographics and comorbid conditions, the hazard of mortality remained 36% lower among HBCR participants versus nonparticipants (hazard ratio, 0.64 [95% CI, 0.45–0.90], P=0.01). Conclusions Among patients eligible for CR, participation in HBCR was associated with 36% lower hazard of mortality. Although unmeasured confounding can never be eliminated in an observational study, our findings suggest that HBCR may benefit patients who cannot attend traditional CR programs.