Human‐Centered Design to Tailor Telehealth Cardiac Rehabilitation to Diverse Populations: The MCNAIR Study

Abstract

Background

Although telehealth cardiac rehabilitation (CR) may improve access, there are concerns about its long‐term effectiveness and impact on equity as compared with in‐person CR. Our objective was to tailor a patient‐centered telehealth CR program for diverse populations.

Methods

CR patients and caregivers were recruited between January and September 2023 from 4 US academic medical centers. Participants engaged in human‐centered design sessions to iteratively refine a telehealth CR program. Sessions had planned topics, but there was variation across sites to account for site‐specific needs and participant feedback. Sessions were qualitatively analyzed using rapid template analysis with preselected behavioral science constructs and other emergent codes.

Results

The study included 21 participants (71% aged ≥60 years, 48% women, 62% non‐Hispanic White individuals; 90% CR patients, 10% CR caregivers). Participants thought that telehealth CR could be helpful for personalized support at home and convenience but recognized that technology is not always easy to use. Some expressed concerns about the safety of telehealth CR, especially at the beginning, and desired monitoring through a mobile device or video observation of exercise. Safety protocols and technology training were developed, which addressed concerns about telehealth CR. Opportunities for social support with telehealth CR were also desired. From these findings, an implementation toolkit was developed, including a graphic program description, safety plan, home exercise plan for during and after CR, and scripts for technology training and individual and group telehealth visits.

Conclusions

A patient‐centered telehealth CR program and implementation toolkit were systematically tailored to address the needs of diverse populations.

Nonstandard Abbreviations and Acronyms

CR

cardiac rehabilitation

MCNAIR

Comparative Effectiveness of In‐Person and Telehealth Cardiac Rehabilitation

Clinical Perspective.

What Is New?

• We tailored a telehealth cardiac rehabilitation program to diverse participants at multiple health care sites and created a toolkit of materials that can be used to implement a telehealth cardiac rehabilitation program.

• Safety protocols and training materials were developed to address potential concerns about telehealth cardiac rehabilitation.

What Are The Clinical Implications?

• Patients and caregivers can use telehealth cardiac rehabilitation to receive personalized care support at home.

Cardiac rehabilitation (CR) is a multidisciplinary program of exercise training and health behavior counseling for chronic disease management of people with heart conditions. CR has traditionally been delivered in person at a CR center, with group sessions being held 3 times per week for 12 weeks. ¹ Despite its effectiveness with respect to cardiovascular morbidity and death, <25% of eligible patients attend CR, and <10% of eligible patients complete CR. ² , ³ , ⁴ , ⁵ , ⁶ Attendance is lowest among women, older adults, Hispanic and non‐Hispanic Black adults, and individuals residing in rural communities and of lower socioeconomic status. ² , ³ , ⁴ , ⁵ , ⁶ , ⁷

Common barriers to attendance and completion of CR include transportation, travel, costs, and conflicts with work or family duties. ⁸ Alternative delivery models for CR may overcome many of these common barriers to attending CR in person. ⁹ Terminology is evolving to describe delivery models for CR, including home‐based, virtual, telehealth, remote, and hybrid CR. With emerging delivery models, at least some of the CR encounters may occur outside of a CR center using telehealth, allowing patients to interact with cardiac rehab staff via video, phone, messaging, and apps for synchronous or asynchronous communication and exercise. ¹⁰ , ¹¹ Short‐term studies of select populations have shown that telehealth CR has similar safety and efficacy to in‐person CR. ⁹ , ¹² , ¹³ However, telehealth CR is not currently available in most health systems and does not have a standard delivery protocol.

There is concern that individuals who are less likely to attend CR may also face challenges accessing telehealth CR due to limited digital access and literacy. Smartphone ownership and home broadband access are high among US adults; however, digital access is lower among older adults and individuals with lower educational attainment and income. ¹⁴ Nevertheless, it is estimated that about half of US adults live in a CR desert (ie, in‐person CR meets <35% of demand) but up to 97% of these individuals may have broadband access. ¹⁵ Therefore, telehealth CR could help to expand CR availability to people without an accessible CR center. ¹⁵

The overall goal of the MCNAIR (Comparative Effectiveness of In‐Person and Telehealth Cardiac Rehabilitation) study is to determine whether telehealth CR, compared with in‐person CR, results in similar or better outcomes in diverse patients with cardiovascular disease. This investigation addresses only 1 aim of the overall MCNAIR study. The objective of this aim of the study was to tailor a patient‐centered telehealth CR program and implementation toolkit using a human‐centered design approach.

Methods

Design

Reverend Patricia McNair, a survivor of cardiovascular disease, cochairs the Patient Advisory Board that informs this study. The Patient Advisory Board chose to name the study after her to represent the many people who have been historically less likely to attend CR. This human‐centered design study collected qualitative and quantitative data from participants, with input into the design from study team members and Patient Advisory Board members at 4 large academic medical centers (University of California, San Francisco; University of Michigan; University of Pittsburgh; and Johns Hopkins University). The human‐centered design process included 4 phases: discover, define, develop, and deliver. ¹⁶ We started with materials previously used at the University of California, San Francisco for telehealth CR to deliver the core components of CR (except in‐person exercise) and supplemented through materials used at other participating CR centers, such as session scripts, telehealth training materials, and safety plans. We created and engaged with the Patient Advisory Board (composed of CR alumni, caregivers of CR patients, people who did not complete CR, and CR‐eligible patients who did not attend) to review and revise these materials. This study was approved by the WCG Institutional Review Board (single institutional review board), and procedures were followed in accordance with WCG Institutional Review Board and institutional guidelines. This study was reported according to the consolidated criteria for reporting qualitative studies (Table S1).

Population

For human‐centered design sessions, we recruited a convenience sample of 5 to 6 participants per site. Participants were eligible if they were aged at least 18 years, had an eligible CR diagnosis and were CR stakeholders (eg, caregiver of an eligible CR patient, prior CR participant, prior CR‐eligible patient who did not participate, current CR participant, or member of the study Patient Advisory Board), and were able to communicate in English. Participants were excluded if they had a diagnosis that would preclude telehealth CR (eg, unstable arrhythmias, enrolled in hospice care, or unable to consent for themselves). Invitations to participate in this study varied by recruitment site and were made via phone or email. Participants completed surveys collecting self‐reported demographic, health, and digital literacy information. ¹⁷ , ¹⁸ We chose the health literacy question since it is a validated, widely used single question for health literacy. ¹⁷ As there is no single accepted measure for digital literacy, we used a draft questionnaire developed from expert and formative research. ¹⁸ Quantitative data were collected and managed by individual sites using Research Electronic Data Capture tools.

Human‐Centered Design

Each site conducted 5 to 6 2‐hour human‐centered design sessions, between February 2023 and February 2024. The initial 3 to 4 sessions focused on discover, define, and develop within the human‐centered design process. ¹⁶ We discussed with participants their experiences with in‐person CR and presented our initial telehealth implementation toolkit and study materials (eg, recruitment materials and study questionnaires). Based on their suggestions, we refined these materials, and participants provided additional feedback on the iterative revisions to our materials. Feedback was gathered on the structure and details of the telehealth implementation toolkit and study materials, including agenda structure, additional materials to add, and interview and questionnaire wording, among others. The final 2 design sessions included a simulation of the proposed telehealth CR program (“deliver”) in which participants engaged in a mock study enrollment and telehealth sessions. At the final session, participants were asked to rate the program using the System Usability Scale (range, 0–100 with 100 representing greater usability). ¹⁸ A prespecified mean rating of 70% was considered acceptable usability. ¹⁹ If this was not achieved, additional iterative design and revision would continue until a site's mean score reached 70%. The coordinating site (University of California, San Francisco) provided the framework, scope, sample agendas and scripts, and initial telehealth implementation toolkit materials (Figures S1–S11). However, session formatting was flexible to meet individual site and participant feedback, and therefore was not uniform across sites.

Individual sites developed their own interview scripts based on the template provided by the coordinating center. Each site had a primary facilitator (M.R., E.M.S., S.R.K., J.G., S.M., N.A., D.F., T.S.) whose task was to lead the discussion during each session. Some investigators had prior relationships with participants either as prior research study participants or current patients of the investigators. Before the start of the first session, each participant provided written informed consent to participate in the study and provided verbal consent at the start of each session for audio recording. Each of the 6 sessions was conducted over a 2‐hour period either in‐person or via Zoom, at the participants' preference. Depending on participant availability, some sites hosted multiple dates for 1 session to ensure everyone had the opportunity to provide feedback and suggestions. In addition to the facilitator, additional research staff at each site were in attendance at each session and noted participant reactions and elicited additional comments. Sessions were audio‐recorded and transcribed. Upon review of session transcripts and participant feedback, study staff iteratively revised the starting materials and cataloged subsequent modifications. To minimize the possibility of unintentionally sharing information that could be used to reidentify private information, session transcripts will not be available; however, analyses are available from the corresponding author upon reasonable request.

Analysis

Audio recordings were transcribed for analysis but were not returned to participants for comment or correction. Recordings and transcripts were reviewed and analyzed by study team members at each site (Table S2) using rapid template analysis based on prespecified codes from the Theory of Planned Behavior, Unified Theory of Acceptance and Use of Technology, and Consolidated Framework for Implementation Research. ²⁰ , ²¹ , ²² , ²³ The coordinating center developed the analysis template that all sites used (Table S3). Emergent codes were generated at individual sites for concepts not captured by prespecified codes. Analysis was completed concurrently with data collection. Upon completion of each session, site‐specific coders individually reviewed and analyzed their own transcripts to extract representative quotations into the prespecified codes and generate emergent codes, if applicable. After individual analysis, site coders met to reach consensus on emergent themes and finalize representative quotations from transcripts. Once all sessions and subsequent analyses were completed, coders from all sites met to collaboratively discuss common themes and interpretations across sites.

Results

Among 21 participants, the majority were aged ≥60 years (N=15 [72%]) with 48% women (Table 1). Most of the participants were former patients who had attended at least 1 CR session (N=19 [90%]), and 2 identified as caregivers (N=2 [10%]); 3 (14%) University of California, San Francisco participants had participated in CR sessions over telehealth, with all other former patients having done only in‐person CR. Overall, participants had relatively high health and digital literacy, with all participants reporting that they were at least “somewhat” confident in filling out medical forms (health literacy) and using videoconference programs (digital literacy). After completing a walk‐through of the telehealth CR program, participants reported high usability, as measured by the System Usability Scale (Table 2), across all sites.

Table 1.

Participant Characteristics

Characteristic	N=21
Site	No.	%
University of California, San Francisco	5	23.8
University of Michigan	6	28.6
University of Pittsburgh	5	23.8
Johns Hopkins	5	23.8
Participant type*
Attended at least 1 CR session	19	90.5
Caregiver of a patient with a heart condition	2	10.5
Prefer not to answer	1	4.8
Age category, y
40–49	3	14.3
50–59	2	9.5
60–69	4	19.0
70–79	9	42.9
≥80	2	9.5
Prefer not to answer	1	4.8
Female sex	10	47.6
Race and ethnicity
Asian	1	4.8
Black	7	33.3
White	13	61.9
Subjective socioeconomic status, mean±SD (1=lowest, 10=highest), 6±2
Health literacy
How confident are you filling out medical forms?
Not at all	0	0
A little	0	0
Somewhat	1	4.8
Quite a bit	14	66.7
Extremely	6	28.6
Digital literacy
How confident are you using video conference programs (such as Zoom)?
Not at all	0	0
A little	0	0
Somewhat	8	38.1
Quite a bit	7	33.3
Extremely	6	28.6

CR indicates cardiac rehabilitation.

^*A participant can identify with >1 category.

Table 2.

System Usability Scale^* for Telehealth CR Program

Site	No.	Mean	SD
University of California, San Francisco	5	88.0	13
University of Michigan	2	78.8	8.8
University of Pittsburgh	5	81.5	8.4
Johns Hopkins	5	77.5	21.9

CR indicates cardiac rehabilitation.

^*System Usability Scale ranges from 0 to 100, with 100 representing greater usability.

Qualitative Results

Qualitative themes are summarized in Table 3 and described in additional detail below.

Table 3.

Summary of Qualitative Themes and Influence on Final Program Components

Theme	Description	Representative quotations	Final program component
Attitudes and beliefs about CR	Most people valued the experience of comprehensive CR	“It's not just exercise, it's not just medication. It's improving … mental health … lifestyle … all the kinds of things.”	Exercise training and multidisciplinary sessions are available
Social influence on CR	Social support from the clinical team, peers, and family is important	“I think being in a group is very good … just hearing what other people were doing and how they were handling [it].”	Group telehealth sessions are available
Self‐efficacy	Participants highlighted CR providing accountability and helping to build confidence	“I was held accountable, and I could compare myself to others to know where I'm at.” “It gave me the confidence that I could do more.”	Individual visits included motivational counseling and review of patient logs
Personalization	Participants desired personalized exercise plans as well as advice on other CR components	“As far as you know, tailoring the diet to each person is important, so it's important to be personal.”	Participants will have individual treatment plans
Desire for nonexercise CR components	Participants stressed the importance of nonexercise components of CR, especially nutrition and emotional well‐being	“The nutrition piece is very important, because you know oftentimes that's a great contributor to what might have led to the issues and the cardiac problems one is having. So, if you're going to go through this exercise program and you're, you know, gonna try to get yourself back in the shape but you…don't change your eating habits or your nutritional habits, it's kind of a moot point.”	Multidisciplinary sessions available. Education topics during individual telehealth visits
Post‐CR support	Participants wanted support for maintaining a healthy lifestyle after CR, with the suggestion to start supporting long‐term habits early	“Well, you know my thoughts on that would be if [this exercise tracking template] were to be there from day one, I think you'd be more likely to continue with [it] rather than just pick it up upon your graduation and start doing it. …”	Exercise plan developed as part of the toolkit
Attitudes and beliefs about technology	Though acknowledging the potential convenience of using technology, participants had mixed beliefs about using technology for CR	“I don't have anything against the Zoom thing. You know there is a certain level of convenience. But my personal preference, if I have time, you know I'd rather just do face to face.”	Video visits are the preferred mechanism for telehealth sessions, but telephone sessions are also allowed
Social influence on technology use	A potential advantage to using technology for CR is accountability to clinical staff	“You have a support team that will go with you. They will go home with you in this program.”	Individual telehealth sessions with a CR professional
Technology‐facilitating conditions	With training and support from family, people thought they could adopt technology for CR	“You might want to just have a quick, you know, crash course on ‘Okay, here's how you set it up. And this is what you have to do when you log in.’”	Training and telehealth tips developed as part of the toolkit
Cost	Participants noted time and money costs	“Cost is always going to be a factor, especially if it involves transportation, if it involves imposing on others, for their time, their efforts and so forth.”	No additional costs required related to purchase of equipment
Safety	Participants expressed concerns about the safety of exercise, and many desired initial monitoring and safety plans	“When you initially start there is a fear factor there in terms of ‘How much exercise can I do?’”	Tailoring of safety plan in toolkit
Privacy	Privacy concerns may be mitigated by presenting information about protections and reasons why the data are helpful.	“If you present it as a, you know, it's a lot more helpful to your medical team to have all this information entered.”	Group telehealth visit consent form
Tracking preferences	Tracking preferences were mixed, with some preferring not to track and some preferring paper tracking vs technology	“Yeah, where you could keep track yourself, that's a good idea [regarding an application/using existing resources such as exercise watches]. Show you how to do that and have you keep track yourself.”	Mobile application is available, but not required

CR indicates cardiac rehabilitation.

Attitudes and Beliefs About CR

CR was overall regarded as a valuable experience that improved participants' lifestyles. Participants who had access to additional components (eg, sessions with a dietitian or mental health specialist) regarded those components as equally valuable. Some, however, had neutral stances on the benefits of CR even after graduating.

“I really improved when I went to cardiac therapy [rehabilitation]. I could feel the difference by the end.” P12

“It's not just exercise, it's not just medication. It's improving … mental health … lifestyle … all the kinds of things.” P9

“I was more the one that went to cardiac rehab because I had to. I did what they wanted me to do while I was there. But then, when I left there, that was it until the next time. …” P13

Social Influence on CR

Participants reported many sources of support during the CR experience. There was a common desire for caregivers and other clinicians (eg, cardiologists, primary care physicians) to be involved in monitoring a participant's progress throughout CR. Encouragement from these individuals strengthened belief in the value of CR, and participants described trust and rapport with the clinical team as an important factor in CR success. In addition, participants found that support and connection with both other CR participants and family members were helpful in staying motivated to continue CR by increasing their confidence and accountability. The sense of community was noted as a vital component to the entire CR experience.

Clinical Team Influence

“It's like a team of doctors pushing you to your goal.” P1

“My cardiologist recommended it, and so it's been really good for me, especially having access to other services beyond cardiologist—pharmacists, nutritionists.” P5

Peer Influence

“Being around other people made me feel better, less alone.” P12

“I think being in a group is very good … just hearing what other people were doing and how they were handling [it], and you know any of the problems they were having.” P13

Family Influence

“If you have someone to report to or who's viewing your progress. A close family support, or a significant other, or a friend support of some sort, so you can say ‘Look what I did. I did this all week. Look at me getting healthier every day.’” P7

Self‐Efficacy

Participants described the importance of accountability towards making health behavior changes and building confidence during participation in CR.

“In cardiac rehab I was held accountable, and I could compare myself to others to know where I'm at.” P12

“It's somehow teaching us to be confident.” P2

“There's a certain level of accountability. And if you're not able to make it … the people there are like, ‘I didn't see you last week or yesterday.’” P10

Personalization

Participants often spoke about their desire for program personalization. They requested personalized exercise prescriptions based on their individual goals and capabilities, as well as frequent check‐ins to assess their progress. Participants noted that accounting for varying circumstances and capabilities would be helpful in reaching their individual goals, whether for exercise, nutrition, or other elements. It could also help build their confidence to restore and maintain their health.

“It would have been helpful to have some tailored advice to your specific circumstances.” P13

“I think it would have helped, you know, if you had somebody to talk to and talk about your progress.” P16

Desire for Nonexercise CR Components

Participants expressed the desire for nonexercise components in CR, especially in nutrition and emotional well‐being. This could be met by coordinating sessions with a registered dietitian, social worker, or mental health specialist. They also acknowledged the importance of involving family members and other support systems in these nonexercise components.

“The nutrition piece is very important, because you know oftentimes that's a great contributor to what might have led to the issues and the cardiac problems one is having. So, if you're going to go through this exercise program and you're gonna try to get yourself back in the shape but you … don't change your eating habits or your nutritional habits, it's kind of a moot point.” P10

“Besides the exercise, diet is the hardest thing, especially cutting out salt. That is the hardest part of all of this.” P8

“You can't really get to a successful, a rehab program, if you're not in a good mental state, or if you're constantly struggling with anxiety or depression or something of that nature.” P10

Post‐CR Support

Participants at all sites expressed a desire for additional support after CR graduation. Examples included tips for exercising independently, resources for inexpensive or free gym memberships and exercise equipment, a “lifestyle program prescription” that accounts for exercise and other health factors (eg, nutrition, stress, medication), and resources for maintaining accountability outside of the clinical space. Participants wanted tools that could help them establish routines early that they could continue to maintain after CR. For example, they desired exercise‐tracking tools and safety plans that could begin on Day 1 of CR and enable them to both monitor their progress during CR and establish sustainable habits.

“Being able to reach out at the end of the program. So, you're exercising at home and you're doing very well to be able to reach out with someone like, ‘Okay, I'm ready to increase my endurance. Can you tell me what I should do to increase my endurance. …’” P7

“Well, you know my thoughts on that would be if [this exercise‐tracking template] were to be there from day 1, I think you'd be more likely to continue with [it] rather than just pick it up upon your graduation and start doing it. … So now, after I graduate and I leave the program, I'm not looking at this form like, ‘Wait a minute. Where am I supposed to put this in?’” P10

Attitudes and Beliefs About Technology

Participants acknowledged greater societal use of technology, but expressed mixed attitudes toward use of technology in CR. Some preferred face‐to‐face interactions and paper tracking of program progress, or worried about loss of community and its concomitant sense of accountability in a telehealth setting. Despite these mixed beliefs, most reported that one can become more comfortable with technology with proper training, practice, and external support.

“I think it's better at home because you're apt to go at your own pace [based on] how you feel, and I feel that I know my body better than anybody. When I did the cardiac rehab, I always wanted to do more than what they wanted me to do, so I think the home‐based is pretty good. … But don't think intermittently it would hurt to come in on‐site. I think that might be helpful.” P16

“I don't have anything against the Zoom thing. You know there is a certain level of convenience. But my personal preference, if I have time, you know I'd rather just do face to face.” P10

“I think the more you [use technology], the more comfortable you'll be with it.” P16

Social Influence on Technology Use

Participants acknowledged the potential for technology to facilitate accountability with clinical staff. They noted that while real‐life interactions are important, continuous use and interaction with others who often use technology may reinforce technology use.

“You have a support team that will go with you. They will go home with you in this program.” P2

“I see my coach!” (About seeing picture of coach in the app) P1

Technology‐Facilitating Conditions

Many factors can facilitate or impede the use of technology. Some examples are compatibility, computer setup, convenience, training, external/family assistance, and access to reliable Internet and user‐friendly platforms. Once participants were trained in using the technology, they commented that they felt more confident in using it independently, although some used support from caregivers.

“You might want to just have a quick, you know, crash course on ‘Okay, here's how you set it up. And this is what you have to do when you log in.’” P10

“If it was interactive with the [patient portal], that would be super cool if everything could be in one place.” P2

“I definitely needed some help along the way, but I was able to figure some things out. So, I'm good.” P16

Cost

Participants recognized that attending to one's health (eg, healthy eating, purchasing equipment, and time) can come at both a financial and time cost but felt it was worthy of the time and effort in most cases.

“Cost is always going to be a factor, especially if it involves transportation, if it involves imposing on others, for their time, their efforts, and so forth.” P15

“Yeah, time costs. You know the amount of time that it seems like it takes a little bit longer to prepare healthier foods, and to do the shopping, to buy the healthier foods, and so on.” P7

Safety

In the beginning stages of CR, participants had safety concerns regarding independent exercise, especially within a telehealth context. For that reason, they often preferred in‐person exercise, at least during the initial CR sessions. To mitigate these concerns, they noted a desire for a variety of “checks” such as monitoring devices or an in‐person or remotely observed exercise. One participant suggested training caregivers, who may be present at home during telehealth sessions, to recognize and respond to symptoms such as increased heart rate or difficulty breathing.

“When you initially start there is a fear factor there in terms of ‘How much exercise can I do?’ It's comforting when you have a monitor on you, and there are people standing around looking at a computer watching your every move. There's a certain level of comfort there that allows you to be able to, you know, go about the business of your exercise, and knowing that if there's a sign of trouble they're gonna walk over to you and say, ‘Hey, wait a minute.’” P10

“A safety plan is an essential part of how it should be monitored.” P9

“I felt pretty safe and supported at home.” P5

Privacy

Some participants expressed concerns regarding maintaining privacy in the telehealth setting. After learning about the purposes of sharing private information and the protections involved in using a mobile application, these concerns were able to be mitigated.

“I think about confidentiality all the time. …” P16

“If you present it as ‘it's a lot more helpful to your medical team to have all this information entered.’” P4

Tracking Preferences

Overall, while participants expressed a desire for CR staff to maintain tracking and monitoring practices throughout their program, they held mixed preferences regarding self‐monitoring of their progress. Most appreciated sources of digital tracking, such as wearable devices and mobile applications. They viewed it as a valuable tool to maintain accountability and share their progress with friends, family, and clinicians. However, some participants preferred paper tracking and others preferred not to track anything themselves. Those interested in tracking their progress requested additional education on tracking methods as well as education on safe ranges for vital signs before and after exercise.

“Yeah, where you could keep track yourself, that's a good idea [regarding an application/using existing resources such as exercise watches]. Show you how to do that and have you keep track yourself.” P12

“Particularly when it comes to exercise as well as medications, all the things that are associated … make sure that [the patient is] doing it properly … checking those things are very essential.” P9

“They gave me log sheets on different things. I really didn't pay much attention because I had my own routine that I wanted to get into.” P16

Implementation Toolkit

The influence of the human‐centered design sessions and qualitative analysis on the final implementation toolkit materials is summarized in Table 3. The program materials are available as a CR implementation toolkit (https://beatty.ucsf.edu/cardiac‐rehab‐toolkit), including site‐level variations of materials. Examples of significant influences of the human‐centered design process on the tailoring of the materials includes the development of an additional Exercise Plan template and revision of the Safety Plan template to include more practical details relevant to participants. Final decisions about components to include as required core components (weekly individual telehealth sessions, individualized treatment plans, addressing the core components of CR) and optional available components (group telehealth sessions, multidisciplinary sessions, mobile application) were based on participant feedback about the value of the group experience, value of multidisciplinary sessions such as meeting with a dietitian, and perceived usefulness of the mobile application. For example, for items about which participants had mixed attitudes (eg, technology use), accommodations were allowed (eg, use of telephone instead of video, option to not use mobile application).

Discussion

We tailored a telehealth CR program to diverse participants at 4 medical centers (Figure) using human‐centered design. Although participants acknowledged the benefits of in‐person CR and potential challenges with telehealth CR (eg, social interaction, technology, concerns about safety), potential advantages of telehealth included personalized support in the home environment and convenience. Providing group telehealth sessions, technology training, and safety protocols addressed participant concerns.

Figure 1. Participant user journey for telehealth CR.

This represents a hypothetical participant journey for telehealth CR. This was presented to participants as part of the human‐centered design process and iteratively revised. Participants are referred to CR from the hospital or clinic, then attend an intake session at the CR center. Participants are offered an optional mobile application. Participants engage in a weekly individual telehealth session by video or phone with a CR professional. Participants are expected to exercise on their own. Group telehealth visits and sessions with other CR professions (eg, dietitian, pharmacist, behavioral health) are also available. If participants need to attend an in‐person CR session for safety, exercise prescription, or other reasons, they may attend in‐person CR sessions. At the completion of the 12‐week program, participants attend a graduation at the CR center.

Various models for CR delivery outside of a CR center suggest that telehealth CR can have similar safety and efficacy to center‐based CR in select populations. ⁹ When delivering telehealth CR, the core components of CR should still be addressed. ²⁴ One of the core elements of CR is the development of an individualized treatment plan. Participants valued personalization and saw telehealth CR as an opportunity to extend this personalization to the home environment.

Participants identified potential challenges with telehealth CR, such as reduced social interaction, difficulty using technology, and concerns about safety. A valued part of CR is interaction with peers, and participants were concerned about the loss of social connection using a telehealth delivery model. Evidence does suggest that psychosocial interventions can be delivered through telehealth. ²³ Some telehealth and hybrid CR models include group exercise sessions. ²⁴ Incorporating group exercise sessions into the program may address concerns about reduced social interaction. Additionally, participants highlighted the positive influence of the relationship with CR clinicians over telehealth.

Challenges using technology may be a barrier to telehealth CR for some people. Participants identified a number of ways to facilitate use of technology, such as training, enlisting the assistance of friends or family members for technology onboarding, and providing tips for setting up technology at home. Other studies have also demonstrated that it is feasible to train participants to use technology and supply loaner devices. ²⁵ The toolkit includes telehealth tips and training materials to help address participant concerns about technology use.

Many participants expressed fear about starting exercise as part of CR and feeling like the monitoring provided at the CR center helped to alleviate that fear. Though this fear must be acknowledged, evidence suggests that CR overall is a safe intervention and that delivery of CR outside of the center is no less safe. ⁹ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ Patient concerns about safety may be mitigated by support and training. We developed toolkit materials including a safety plan and individual and group telehealth scripts with an essential component of addressing safety concerns. For patients in whom there are initial safety concerns, a hybrid model may be appropriate, with initial sessions conducted in person before transitioning to a telehealth model once safety concerns have been addressed. The design of the program will also allow for patients to attend in‐person sessions as needed for concerns about safety.

In tailoring a telehealth CR program, we made iterative revisions based on the feedback of participants at 4 sites across the United States and included more women and fewer White people than historically attend CR. ² , ³ Since there are disparities in CR attendance, we hope that tailoring a telehealth CR program can provide access to more people. ⁵ Additionally, the methods used to tailor intervention materials to the needs of participants at a site could be used by sites planning to implement telehealth CR.

Despite many strengths, our study does have limitations. While we did not intend to exclude CR‐eligible patients who did not attend CR, our sample did not include any participants in this category, who may have had differing perspectives on telehealth CR. Though we do include more older adults, women, and non‐White people than many other studies, we did not have substantial representation of participants living in rural areas or those with low health literacy in our sample. However, our program is very similar to what has worked for the Veterans Affairs Office of Rural Health program at a time when their program mostly used telephone telehealth visits rather than video. ³¹ Additional personalization may be needed for participants with low health literacy. Additionally, our sample included only English‐speaking patients. At the University of California, San Francisco, the telehealth and hybrid CR program has successfully used interpreter services for patients who speak other languages. The upcoming trial will include Spanish‐speaking participants.

Substantial efforts are needed to reach the Million Hearts goal of 70% participation in CR. ³² The availability of safe and effective telehealth CR programs will be essential to expanding access and attempting to achieve equity in the delivery of CR. ¹⁰ This research has resulted in the development of a telehealth CR program and toolkit materials to help inform the delivery of telehealth CR. We are now testing the comparative effectiveness of in‐person and telehealth CR in a multicenter randomized trial (NCT05933083). The telehealth CR intervention uses the toolkit materials and results from this study.

Conclusions

We tailored a patient‐centered telehealth CR program and implementation toolkit to address the needs of diverse patients. Potential benefits of telehealth CR included personalized support in the home environment and convenience. Though people identified potential challenges to telehealth CR (eg, social interaction, technology, and safety concerns), these concerns could be addressed through program design changes.

Sources of Funding

This work was supported through a Patient‐Centered Outcomes Research Institute Project Program Award (IHS‐2021C3‐24 147). All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient‐Centered Outcomes Research Institute, its Board of Governors, or its Methodology Committee.

Disclosures

Dr Martin reports additional research support from the American Heart Association Health Technologies and Innovation Strategically Focused Research Network (20SFRN35380046 and 20SFRN35490003), a collaborative project of this network (878924), additional support from the American Heart Association (882 415 and 946 222), the National Institutes of Health (P01 HL108800 and R01AG071032), the David and June Trone Family Foundation, the Pollin Digital Innovation Fund, Sandra and Larry Small, Google, and Merck. Dr Martin also reports receiving equity from Corrie Health; material support from Apple; grants from Google; serving on the Care Access advisory board (unpaid); and personal fees for consulting from Amgen, Arrowhead, AstraZeneca, Bristol Myers Squibb, Chroma, Heartflow, Kaneka, NewAmsterdam Pharma, Novartis, Novo Nordisk, Premier, Sanofi, and 89bio outside the submitted work.

Dr Spaulding reports additional research support from the American Heart Association (20SFRN35380046 and 878924) and the National Institutes of Health (U01HL096812) and personal consulting fees from Corrie Health.

Dr. Golbus receives funding from the National Institutes of Health (L30HL143700 and 1K23HL168220).

Dr Nallamothu is a principal investigator or coinvestigator on research grants from the National Institutes of Health, Veterans Affairs Health Services Research and Development, and the American Heart Association. He also receives compensation as editor‐in‐chief of Circulation: Cardiovascular Quality & Outcomes, a journal of the American Heart Association. Finally, he is a coinventor on US Utility Patent Number US15/356012 (US20170148158A1) entitled “Automated Analysis of Vasculature in Coronary Angiograms” that uses software technology with signal processing and machine learning to automate the reading of coronary angiograms, held by the University of Michigan. The patent is licensed to AngioInsight, Inc., in which Dr Nallamothu holds ownership shares and receives consultancy fees.

All other authors have no disclosures to report.

Supporting information

Tables S1–S3

Figures S1–S11