Abstract 14404: Variation and Predictors of Referral for Cardiac Rehabilitation Among Patients With Heart Failure Before and After Expansion of Coverage and Association With 1-year Clinical Outcomes

Abstract

Introduction: The Centers for Medicare and Medicaid Services(CMS) expanded coverage for cardiac rehabilitation(CR) in 2014 for patients with clinically stable heart failure(HF) with reduced ejection fraction. Contemporary CR referral and participation rates among eligible patients with HFrEF are not known. Methods: Patients hospitalized for HF with ejection fraction ≤35% in the American Heart Association Get With The Guidelines®-HF(GWTG-HF) registry from 2010-2020 were included. Trends in rate of referrals and predictors of referral were determined. Among subset of participants with available Medicare-linked data, rate of CR participation was assessed, and 1-year outcomes were compared among patients referred vs. not referred to CR. Results: Of 69,441 HF patients eligible for CR, 17,076(24.6%) were referred to CR. There was substantial variability in referral across GWTG-HF participating centers (range 0-100%, IQR 4%-50%). Of patients with fee-for-service Medicare referred to CR, only 4.2% participated in CR in the year following HF hospitalization (median sessions: 3, range 1-76). Referral rate increased from 2010-2020, with significant increase since 2014 CMS coverage expansion (p trend <0.001). Patients not referred were more likely to be older, of minority race, and with greater burden of comorbidities. Patients admitted to rural hospitals and those in the Northeast were less likely to be referred to CR. Among patients free of HF events within 30 days post-discharge, CR referral was independently associated with lower risk of 1-year all-cause mortality vs. those not referred (HR 0.84, 95% CI (0.73 - 0.98, p = 0.0267), without significant difference in 1-year HF-related or all-cause readmission. Conclusions: Although CR referrals have increased among eligible HF patients since CMS expanded coverage in 2014, absolute referral and participation rates remain low. Age, race, and burden of comorbidities were independently associated with CR referral.