Association of Veterans Health Administration Home-Based Programs With Access to and Participation in Cardiac Rehabilitation

Performance measures recommend referral to cardiac rehabilitation (CR) after acute myocardial infarction (AMI), percutaneous coronary intervention (PCI), or coronary artery bypass grafting (CABG),¹ but CR is vastly underused.² Lack of transportation and limited access to CR programs have been identified as 2 of the largest barriers to participation.³ To address this issue, the Veterans Health Administration (VHA) has started implementing new home-based CR (HBCR) programs.^4,5 Therefore, we examined whether implementation of new HBCR programs is associated with improved CR participation in the VHA.

We conducted a prospective cohort study of 151 VHA facilities at which 99 097 patients were hospitalized for MI, PCI, or CABG (based on national electronic health records) from January 1, 2010, through December 31, 2015. Sixty-four VHA facilities were excluded from the analysis because they had fewer than 10 patients hospitalized for ischemic heart disease (IHD). During the 6-year study period, the number of VHA facilities that offered HBCR increased from 1 to 12. This study, including a waiver of the requirement to obtain informed consent, was approved by the institutional review board at the University of California, San Francisco.

We compared CR participation across 3 groups of facilities. The 3 groups of facilities offered referral to: (1) offsite facility-based CR programs reimbursed by the VHA, (2) either offsite CR or VHA onsite CR programs, or (3) offsite CR or VHA onsite CR or HBCR programs. Participation was defined as participating in at least 1 outpatient CR session within 12 months after the discharge date based on electronic health records from VHA facilities, non-VHA facilities that were reimbursed by the VHA, and Medicare claims data. The number of sessions was determined using Current Procedural Terminology codes (93797, 93798) and/or VHA Stop Code 231. We used mixed-effects logistic regression with clustering by facility to determine the association of offering HBCR and participation, adjusted for patient and hospital characteristics. We used 2-sided t tests for continuous variables and χ² tests for dichotomous variables. P < .05 was considered to be statistically significant.

From January 1, 2010, through December 31, 2015, a total of 99 097 patients were hospitalized for IHD (AMI, PCI, or CABG). Participation in at least 1 CR session increased from 8.1% to 13.2% overall (P < .001). At the 12 facilities that began implementing HBCR programs, participation increased from 6.0% (14 of 234) to 24.6% (724 of 2941) (P < .001) (Figure). At the 23 facilities that offered referral to offsite CR or VHA onsite CR, participation increased from 10.9% (815 of 7463) to 17.6% (928 of 5266) (P < .001). At the 52 facilities that offered referral to offsite CR only, there was no detectable change in CR participation (from 6.4% [752 of 11 771] to 6.6% [561 of 8504]; P = .63). In a sensitivity analysis that required 3 or more CR sessions, participation increased from 5.1% (12 of 234) to 16.6% (489 of 2941) at facilities that offered HBCR (P < .001), 8.3% (621 of 7463) to 9.6% (504 of 5266) at facilities that offered offsite or VHA onsite CR (P = .01), and 5.2% (615 of 11 771) to 6.0% (511 of 8504) at facilities that offered offsite CR only (P = .02).

Compared with patients hospitalized at a facility that offered referral to offsite CR only, those hospitalized at a facility that offered HBCR had 4-fold greater odds of participating, and those hospitalized at a facility that offered VHA onsite CR had 3-fold greater odds of participating in CR (Table). No significant difference was found in the number of weeks of CR completed among patients offered HBCR vs those not offered HBCR (median, 7.6 vs 8.7 weeks; P = .60). However, patients offered HBCR were less likely to drop out after the first session than were those for whom HBCR was not available (423 [16.8%] vs 1741 [20.2%]; P < .001).

In summary, veterans hospitalized with IHD were more likely to participate in CR when a home-based program was available. We recognize that results may be biased because facilities that developed HBCR programs were likely to be stronger proponents of CR, and overall CR participation remained low. Nonetheless, these findings demonstrate that HBCR may be an effective tool for increasing CR participation among patients who would otherwise decline to participate.

Accepted for Publication: November 22, 2017.

Published Online: January 22, 2018. doi:10.1001/jamainternmed.2017.8039

Corresponding Author: David W. Schopfer, MD, MAS, 4150 Clement St (111A1), San Francisco, CA 94121 (david.schopfer@ucsf.edu).

Author Contributions: Drs Schopfer and Whooley had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Schopfer, Krishnamurthi, Duvernoy, Forman, Whooley.

Acquisition, analysis, or interpretation of data: Schopfer, Krishnamurthi, Shen, Whooley.

Drafting of the manuscript: Schopfer.

Critical revision of the manuscript for important intellectual content: Krishnamurthi, Shen, Duvernoy, Forman, Whooley.

Statistical analysis: Schopfer, Krishnamurthi, Shen.

Obtained funding: Schopfer, Whooley.

Administrative, technical, or material support: Whooley.

Study supervision: Whooley.

Conflict of Interest Disclosures: None reported.

Funding/Support: Dr Schopfer is funded by grant KL2 TR000143 from the National Center for Advancing Translational Sciences, National Institutes of Health. This work was funded by grant QUE 15-283 from the Veterans Affairs Health Services Research and Development Service Quality Enhancement Research Initiative, contract IH-1304-6787 with the Patient-Centered Outcomes Research Institute, and the Veterans Health Administration Office of Rural Health.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.