Association Between Initiation of Cardiac Rehabilitation after Hospitalization for Heart Failure and 1-Year Survival Among Medicare Beneficiaries.

Abstract

Heart failure (HF) is a leading cause of hospitalization and death for older Americans, with an approximately 50% mortality rate within 5 years. To examine the relationship between cardiac rehabilitation (CR) attendance and mortality after a hospitalization for HF. We identified Medicare beneficiaries hospitalized with a principal diagnosis of HF in 2017. We evaluated the association between initiation of CR within 180 days of hospital discharge and 1-year mortality with a model that used stabilized inverse probability of treatment weighting (SIPTW) and adjusted for covariates. We conducted a series of sensitivity analyses including landmark methods (restricted to those surviving 180 days), propensity score weighting, propensity matching, and a dose-response analysis. Of 116,302 patients (mean age 82 years, 54% women) admitted with HF, 1747 (1.5%) initiated CR within 180 days. Patients who did not attend CR were more likely to be older, female, frail, have diastolic HF, and live further from a CR facility. Overall, 35% of patients died within 1 year of discharge, including 10.7% among those who started CR and 35.2% among those who did not. Attributing time prior to CR initiation as untreated time and adjusting for unbalanced covariates and propensity to start CR, participation was associated with a 45% lower risk of death within 1 year of discharge (HR 0.55, 95% CI 0.46-0.66). Multiple sensitivity analyses yielded similar effect estimates, including a limited sub-group with a diagnosis of systolic HF. Although few patients with HF enroll in CR within 180 days of a HF hospitalization, participation is associated with lower risk of 1-year mortality. These results should inform our counseling of patients with HF about CR and should be considered in the development of clinical guidelines, research funding strategies, and payment policies. Not applicable.