A Qualitative Study of Home-Based and Hybrid Cardiac Rehabilitation During COVID-19

Background: Because of the need for distancing during the COVID-19 pandemic, many cardiac rehabilitation (CR) centers suspended in-person services and pivoted to home-based CR (HBCR). In this study, we qualitatively evaluated implementation of HBCR, which included weekly phone or video visits for individualized exercise plans, nutrition and health education counseling, wellness sessions, and optional mobile phone applications.


Introduction
Cardiac rehabilitation (CR) reduces mortality, morbidity, and hospitalizations. 1 CR has traditionally been offered in-person at rehab centers where patients receive exercise training, health education, psychosocial management, and tobacco cessation counseling from a multidisciplinary team. A recent scienti c statement endorsed home-based CR (HBCR) as an acceptable alternative for some patients. 2 Increasingly, digital tools (e.g., videoconferencing, mobile applications, and wearables) have augmented the ease of delivering programs outside of the traditional rehabilitation center setting. 3,4 In response to distancing policies during the COVID-19 pandemic, many CR centers suspended in-person services. 5 Some centers, including the University of California San Francisco (UCSF), pivoted to HBCR as an alternate mechanism of care delivery. Prior studies of HBCR have demonstrated feasibility and e cacy despite few of these including programs with modern videoconferencing. [6][7][8] Though many centers anecdotally transitioned to remote care during COVID-19, few have published their experiences, and even fewer have studied their efforts qualitatively. 5,9−12 In March of 2020, the University of California San Francisco (UCSF) began delivery of HBCR through telephone and video visits with optional use of mobile applications. We sought to qualitatively analyze the experience of patients and staff during the implementation of this program to understand the key barriers and facilitators to implementation.

DESIGN & POPULATION
We conducted a qualitative evaluation of the implementation of HBCR at UCSF. Patient participants included individuals enrolled in UCSF CR who had at least one telephone or in-person encounter with UCSF CR staff between March 16, 2020 and May 27, 2020. All participants were of adult age and English speaking.
Staff participants included UCSF staff members (e.g., physician, nurses, exercise physiologists, dietician, and administrative staff) providing patient care and operational support to the UCSF Health Cardiac Rehabilitation and Wellness Center as of March 16 th , 2020.
All participants provided verbal informed consent for this minimal risk study. The study was reviewed and approved by the UCSF Institutional Review and Ethics Board. The methods were performed in accordance with all relevant guidelines and regulations and approved by the ethics committee at UCSF.

INTERVENTION
To meet the need for CR during COVID-19, UCSF implemented HBCR, which included the core components of CR and was individually tailored to patients' needs and limitations at home (Figure I). 13 Patients interacted with CR staff through weekly telephone or video visits for individual exercise plans, health education, and motivational counseling. Patients engaged in 1-hour group wellness sessions focused on emotional wellbeing and health education over telephone or video with other patients and multidisciplinary staff. Patients were referred for telephone or video consultations with nutritionists, pharmacists, mental health professionals, and tobacco cessation counseling as indicated.
Patients were invited to use a free mobile phone application called Better Hearts (Chanl Health), which allowed for logging exercises and vitals, receiving medication reminders, viewing educational materials, and messaging with providers. Patient data was uploaded to a cloud-based server where providers could view the data through a patient management dashboard. Additionally, providers could create exercise prescriptions and exchange messages with patients. Patients were also invited to use a separate online platform called Tiatros designed to provide programming for psychosocial resilience and behavioral health support. Patients participated in an 8-week program with a peer group and trained mental health facilitators that included educational videos, learning materials, and psychosocial exercises.

INTERVIEWS
Patients and staff participated in semi-structured interviews about their experiences with health-related behaviors during COVID-19, HBCR, and technology tools used to facilitate HBCR. A single staff member interviewed participants following an interview guide [Supplement]. Interviews were audio-recorded and transcribed verbatim through an online transcription service (i.e., Rev.com).

QUESTIONNAIRES
All participants received a unique link to a Qualtrics questionnaire to evaluate Tiatros and Better Heart applications [Supplement]. Participants who were unable to complete the web-based questionnaire completed the questionnaire over the phone, administered by study staff. Responses were summarized using descriptive statistics.

ANALYSIS
Patient demographics were summarized using descriptive statistics. Interview transcripts were analyzed using a rapid qualitative template analysis which included themes from the Theory of Planned Behavior (e.g., attitudes, beliefs, subjective norms, perceived behavioral control, behavioral intention), Uni ed Theory of Acceptance and Use of Technology (e.g., performance expectancy, effort expectancy, facilitating conditions, social in uence, habit, price value, hedonic motivation, and technology use intention), and Consolidated Framework for Implementation Research [Supplement]. 14 Templates were iteratively revised to incorporate emergent themes. Two reviewers independently coded each interview and discussed discrepancies to achieve greater than 95% concordance. Key themes and quotes were extracted and collated for analysis.

SCREENING AND ENROLLMENT
From March 16, 2020 to May 27, 2020, 38 potential participants were identi ed from electronic health records. Of the 38 potential participants, 15 did not respond and 1 was not contacted due to privacy restrictions. Twenty-two participants were contacted, but 5 were ineligible due to a language barrier. Of the 17 eligible participants, 13 consented but 1 withdrew before the interview; ultimately, 12 participated.
Participants ranged in age from 48 to 83 (mean 64.8 years). There were 8 females (66.7%) and 4 males. There were 8 self-identi ed white individuals (66.7%), 2 were Asian (16.7%), 1 was Black (8.3%), and 1 was multi-racial (8.3%). Staff were all 18-64-years of age and include both men and women of different races and ethnicities (more speci c descriptions not reported due to small numbers and privacy protections).

QUESTIONNAIRES
Of the patient participants, eight (75%) reported using mobile apps or websites to help them participate in CR. Three participants used the Better Hearts app and responded to questions. Four participants used the Tiatros app and three responded to questions. The effects of the pandemic on health-related behaviors were variable. Some people found behaviors like physical activity easier to maintain, while others found them more di cult.
And ironically, I found more time and more ways to exercise, and I've been eating healthier and less.  Previously I got a great deal of exercise, making more frequent trips to the grocery stores and doing more shopping and doing more walking and riding the bus more. So, now that I'm holed up in the place, my exercise has been spotty. [ It has allowed them to certainly minimize their risk and their exposure, gives them peace of mind, allows them to have their family members participate, which has been really interesting and helpful.  Group telephone and video wellness sessions were praised by patients.
It was very helpful. I would say that it was, at the beginning of shelter in place, this regular routine. The check in was very helpful from not just the physical perspective, but the mental perspective. [ [About Tiatros] The whole thing was kinda rolled out, and not a lot of instructions were passed on. And then you're suddenly getting these emails and prompts that said, "Week one has started and you're paired with so-and-so." Well, paired to do what exactly?  If Fitbit and the Better Hearts could combine together it would be perfect.  [About Map My Walk] He gave me also an app that I use on my phone that now whenever I go out for a walk I hit the app and just measure the distance. This app is good 'cause it records everything. It's not as good as when you're able to be hands on and you could take an exercising blood pressure while they're in the middle, and you get to see their EKG the whole time. But other than really getting that diagnostic material, it's pretty close to being the same. [Staff-003] HBCR may be associated with different costs and reimbursement for centers but may reduce some patient out-of-pocket costs.
It was very cost effective for him because he was catching cabs.  We got a lot of support with getting laptops. [Staff-002] Even though we weren't going to be reimbursed we realized the importance of patient care.  We've created a billing process, so that's been a big hurdle and we'll see if that's successful.  Discussion UCSF rapidly implemented HBCR in response to COVID-19. Our qualitative interviews demonstrate key strengths of HBCR including the ability to deliver an individually tailored program (Table I). Strong relationships with staff who served as a source of accountability and technology support were a key facilitator of patient experience. Similarly, a strong team culture, institutional buy-in, and support for staff were important factors for implementation. Finally, heterogeneous attitudes toward remote care and technology highlight that exible hybrid delivery models may be needed to meet the varying preferences and needs of patients. The UCSF experience during the COVID-19 pandemic led to a exible hybrid delivery model in which patients attend an individualized program of in-person and/or remote telephone or video visit sessions ( Figure I).
As CR programs consider how to expand access, HBCR and hybrid programs, enhanced by new digital technology tools, may be an effective strategy. 3,15,16 Prior studies have shown that HBCR and hybrid CR are not signi cantly different from traditional CR in terms of outcomes and completion rates. 7,17−19 Qualitative studies of telerehabilitation outside of the US have reported the importance of individual tailoring and connections between staff and patients for promoting accountability. [20][21][22] Studies have also identi ed the potential barrier of limited peer engagement in remotely delivered CR. 20 Our study demonstrates that providing telephone or video group wellness sessions may address this barrier. The incorporation of technology tools into CR programs can pose challenges, but emerging evidence and our results suggest that it may be possible to include technology tools that meet expected performance, are easy to use, have appropriate training, and have their use supported by staff. 23 Organizational factors also contribute to the implementation of new delivery models for CR. Previous work has reported that factors such as leadership support, funding, and institutional contribute to CR delivery. 24,25 This study adds further qualitative data to support these facilitators, and adds information on organizational factors that contribute to technology adoption for CR programs, including equipment and work ow modi cations for CR staff.
Our study has limitations including the small number of participants. However, two reviewers found saturation of major themes on independent review. Further, some sampling bias may exist because patients who participated in interviews may have been more engaged. However, our sample did include one patient who chose not to continue participating in the program. Additionally, our study sample was limited to English-speaking participants. The CR program has delivered HBCR to patients with limited English pro ciency with the use of interpreter services, but this was not studied.
Despite the data supporting the bene ts of CR, the proportion of eligible patients participating in CR remains low. 26,27 Ongoing innovations in delivery of CR may increase participation. This study provides information that may be helpful for programs seeking to implement new models of CR delivery. Further research is needed to determine whether patient-centered, exible hybrid delivery models can improve access, patient experience, and clinical outcomes.  SupplementalMaterial.docx