Enhancing Participation in Cardiac Rehabilitation: Focus on Underserved Populations

Abstract

Participation in cardiac rehabilitation (CR) significantly decreases morbidity and mortality and improves quality of life following a wide variety of cardiac diagnoses and interventions. However, participation rates and adherence with CR are still suboptimal and certain populations, such as women, minorities, and those of lower socio-economic status, are particularly unlikely to engage in and complete CR. In this paper we review the current status of CR participation rates and interventions that have been used successfully to improve CR participation. In addition, we review populations known to be less likely to engage in CR, and interventions that have been used to improve participation specifically in these underrepresented populations. Finally, we will explore how CR programs may need to expand or change to serve a greater proportion of CR-eligible populations. The best studied interventions that have successfully increased CR participation include automated referral to CR and utilization of a CR liaison person to coordinate the sometimes awkward transition from inpatient status to outpatient CR participant. Furthermore, it appears likely that maximizing secondary prevention in these at-risk populations will require a combination of increasing attendance at traditional center-based CR programs among underrepresented populations, improving and expanding upon tele- or community-based programs, and alternative strategies for improving secondary prevention in those who do not participate in CR.

Benefits of and Attendance at Cardiac Rehabilitation

The benefits of Cardiac Rehabilitation (CR) are broad and compelling. Data from meta-analyses of randomized-controlled trials show that participation in CR results in decreases in overall and cardiac mortality as well as improvements in functional capacity, physical and psychologic function [1,2]. However, despite CR participation being a class 1A recommendation of the American Heart Association (AHA), the American College of Cardiology (ACC) and the European Society of Cardiology [3,4] participation rates are suboptimal.

CR participation rates have best been assessed by analysis of administrative databases such as that collected by the Centers for Medicare Services (CMS). In 2007 Suaya et al. analyzed CR participation rates for CMS patients, all aged 65 years or greater, who had experienced a myocardial infarction (MI) or underwent coronary bypass surgery (CABG) in 1997 [5]. They found that CR participation after MI was 13.9% and after CABG was 31.0% with an overall participation rate of 18.7% as defined by participation in at least one session of CR. This compares with a >90% use of aspirin [6] and a 93% use of statins early after MI [7] both of which are also class 1A recommendations. Surprisingly, CR use varied 9-fold among states ranging from 6.6% in Idaho to 53.5% in Nebraska.

A more contemporary analysis of CR participation among Medicare beneficiaries was performed by Ritchey et al in 2020, analyzing CMS data for a broad range of CR eligible patients in 2016 [8]. Overall, CR participation was 24.4%, which represents a 29% relative increase from years 1997 to 2016. Factors associated with non-participation included increasing age (within the Medicare range, 65+), geography (participation rates remained lowest in East South Central U.S. and highest in West North Central), sex (lower in women), race (lowest in Hispanics and Non-Hispanic Blacks), and diagnosis (lower after MI than CABG).

The Centers for Disease Control (CDC) has analyzed CR participation across the adult age range using data from the Behavioral Risk Factor Surveillance System [9]. An analysis of data from 20 states and Washington DC from 2013 and 2015 found that 34.7% of adults who reported a history of a MI also reported subsequent use of CR.

The process of increasing CR participation across the U.S. has been supported by the CR Collaborative of the CDC [10]. An ambitious goal of increasing CR participation from 20% to 70% has been set [11]. These processes include interventions to improve referral of appropriate patients to CR, enrollment of appropriate patients into CR and adherence to successful completion of the CR program. The purpose of this paper is to review what interventions have been used to successfully improve CR participation, while also commenting on how we may have to adjust how CR is offered to reach the ambitious goal of 70% participation.

Predictors of CR Participation

Researchers have assessed individual level, system level and geographic factors which can influence CR participation [12,13,14]. Some of these can be seen in Figure 1. Male sex, white ethnicity, age, higher socioeconomic status, presence of insurance coverage, strong physician recommendation, higher levels of education, spousal support, lower cost-share, shorter distance to program, and programs outside the southern region of the US have all been associated with increased rates of CR participation [8,13,15,16,17,18]. Using national level data from the Behavioral Risk Factor Surveillance System, Peters et al examined predictors of participation in CR following acute myocardial infarction from 2005-2015 [17]. Among the 32,792 survey respondents, CR non-participants were more likely to be female, black, with lower levels of educational attainment.

Figure 1.

Facilitators and Barriers to Attending CR

A recent study by Khadanga et al [19] examined the medical, psychosocial, and behavioral factors which influence CR participation in both sexes based upon in-hospital interviews. Of the 294 patients enrolled in the study, correlates for CR participation included: use of electronic referral during the hospital discharge process, surgical diagnosis, non/former smoker, ejection fraction >50%, higher educational attainment, stronger physician recommendation, higher physical function, and more social support (all p<0.04). Of these, four factors were identified as independent predictors: the presence of electronic referral, surgical diagnosis, non/former smoker and strength of physician recommendation (all p<0.02). Interestingly, there were no significant differences in participation due to sex, or questionnaire measures of anxiety, depression or executive function.

Participation and Adherence in Underserved Populations

Patient Characteristics

As mentioned above, several patient characteristics are associated with lower participation in CR. Some of these characteristics have been examined in depth. For example, while not always the case, women are generally found to have lower rates of CR participation and adherence (see [20], for more discussion of women in CR). In a meta-analysis of 297,719 individuals eligible for CR, women were 36% less likely to enroll in CR compared to men [21]. Lack of referral, increased family responsibilities, lower educational attainment, lack of social support and fear of exercise have been associated with lower CR participation in women [18,22,23]. While behavioral and psychosocial barriers are seen for both men and women, some tend to be more prevalent or pertinent among women. For a more in-depth discussion of psychological considerations see [24]. Physician recommendation is a strong predictor for CR participation, yet women are less likely to receive this endorsement compared to men [15,22]. Similarly, when asked about barriers to participation women are more likely to report lack of awareness of CR [18]. Regarding the reason for reduced adherence, it is likely that sex alone is not the reason for reduced adherence, instead it is certain characteristics commonly associated with females, such as single status, family responsibilities, and lower physical function, which contribute to dropping out of CR [25].

Minorities are under-represented in CR [8] and minority individuals residing in the most deprived communities are even less likely to initiate CR [26]. When examining the racial disparities in CR referral among Medicare patients, Li et al found that Black, Hispanics and Asian patients were, respectively, 20%, 36% and 50% less likely to receive CR referral [27]. A common barrier in these populations, seen in numerous studies, is lack of CR awareness; many report that they were not referred, did not understand the importance of CR, or that it was not endorsed by the physician [28]. Minority status often co-occurs with language barriers, which can hinder effective communication and create misunderstandings affecting perceptions of CR.

Minority women, in particular, are much less likely to engage in CR. Participation rates among Non-Hispanic black, Hispanic and Asian women average between 10-12% [8,22]. Mochari et al. [29] examined ethnic differences in CR referral among 304 women (15% Hispanic, 34% black, 48% white) and reported that compared to whites, minorities were less likely to receive endorsement of CR and more likely to cite financial issues as a barrier. Specific focus should be given to ethnic minorities, especially women, to increase participation and adherence.

Socio-economic status (SES) can be measured by educational attainment, income, insurance status, or occupation. Regardless of how it is measured, SES is a strong predictor of CR attendance and completion with lower-SES predicting fewer sessions completed [30,31]. This is concerning considering lower-SES patients are at a greater risk for cardiac recurrent events thus are more in need of CR than higher SES individuals [32]. Studies examining benefits of CR by SES generally show that lower-SES patients who complete CR derive similar benefits as compared to higher SES patients (e.g. [33]) although one recent study suggests that patients with lower educational attainment may not improve as much in fitness during CR as compared to patients with higher educational attainment [34].

Current smokers are less likely to attend CR and are substantially more likely to drop out. Current smoking may result in decreased participation for several reasons. First of all, patients may feel stigmatized for their inability or unwillingness to quit in the context of a secondary prevention program. Additionally, current smoking likely increases discomfort while engaging in exercise and can limit improvements in cardiovascular (CV) fitness [36]. Finally, continued smoking may reflect a more general inability or unwillingness to change health-related behaviors [32]. As continued smoking is the strongest predictor of recurrent CV events (e.g. [37]) the lack of attendance in this population is troubling.

Other patient characteristics can influence attendance. Age, for example, is a powerful predictor of attendance with both the youngest [31] and oldest [38] of those who are eligible being less likely to attend (see [39] for more discussion of older patients in CR). Cognitive characteristics are also important to consider, as cognitive deficits or impairments can interfere with CR attendance [40]. As these issues are more prevalent in older populations and as the CR population is becoming older on average [41], it will be important to further consider cognition. While these studies generally examined relatively minor deficits in cognition it would only be assumed that attendance would be impacted even more greatly in those with more severe intellectual disabilities.

Certain CR-eligible diagnoses are also associated with less participation. For example, patients with chronic stable systolic heart failure (HF; defined as ejection fraction ≤35%) have been eligible to attend CR since 2014 [42]. “Stability” of these patients is defined as no recent hospitalizations within the past 6 weeks and on optimal guideline-directed therapy [43]. While these insurance coverage criteria were based off of the HF-ACTION trial, this time delay has contributed to underutilization of CR. CR attendance among this population is extremely low ranging from 2-17% based on recent studies [8,44,45,46]. This low attendance is likely due to several factors. HF patients have high rates of comorbidities and very low levels of physical fitness [45]. Additionally, there is no smooth referral process for these patients. Patients who are hospitalized require a six-week period of stability (e.g. no hospitalizations) before they can begin CR, and while fairly stable HF patients are seen as outpatients, consistent systems for referring these patients to CR have not been established (For a more in-depth discussion of HF patients in CR see [47].

Program Characteristics

The geographic availability of CR programs can also have large effects on participation. CR programs are not distributed equally across geographic areas in the U.S.[48,49] and studies suggest that geographic differences in programs may have a large effect on CR participation [49,50]. The distance between a CR-eligible patient and a CR program can be quite variable and distance to the nearest facility is a significant predictor of participation [15,51]. Patients that reside in rural areas often receive care some distance away from where they reside and would likely be distant from a CR program, as well.as existing hospital-based CR programs tend to be concentrated in urban and suburban population centers. However, given the relatively low reimbursement rate for CR, it can also be cost-prohibitive to run programs in urban centers where property costs are high and thus CR “deserts” can occur in both rural and urban areas [10].

Interventions to Increase CR Participation

A number of interventions are associated with increased CR use (Figure 1). These interventions are divided into those that increase CR referral of appropriate patients, increase CR participation or uptake, and improve CR adherence to completion (Table 1).

Table 1:

Summary of Interventions to Increase CR Use

Interventions that Improve CR Referral

Automated Referral

CR Liaison

Interventions that Improve CR Uptake

Automated Referral

CR Liaison

Strong Physician Recommendation

Motivation Interviewing

Referral Close to Home

Reducing time to CR Entry

Hybrid Programs

Flexible Scheduling

Removing Co-Pays

Interventions that Improve CR Adherence

Flexible Scheduling

Motivation Interviewing

Hybrid/Home Programs

Patient Incentives

Interventions to Improve CR Referral

Computerized “Automated” Referral

The most thoroughly studied intervention to increase referral of appropriate patients to CR which, subsequently is associated with increased participation is that of a computerized “automated” referral as part of the hospital discharge process [52,53,54,55]. This process is facilitated by being an “opt out” decision pathway for individuals discharged with a CR-eligible diagnosis [54]. Compared with control status, automated referral can more than double CR referral (Table 2).

Table 2:

Automated Referral to Increase CR Use

Effects on Referral Rates

32.2% vs. 70.2% Referred [53]

31.6% vs. 85.7% Referred [54]

39.5% vs. 75.0% Referred [55]

Effects on Participation Rates

29.1% vs. 60.7% Enrolled [53]

26.1% vs. 33.3% Completed CR [55]

Implementation of automated referral in a non-research setting can be difficult and requires prioritization by hospital cardiologists and an interaction between a physician or mid-level practitioner who is very familiar with the discharge process and the information technology department of the hospital. The rationale for implementation of an automated process is that CR is a class 1A recommendation of the AHA and the ACC and will drive better clinical and functional outcomes for patients. Secondarily, that it will increase hospital billing for CR in a fee for service environment which can be an important factor for hospital administrators. Automated referral has such a strong evidence base that implementing automated referral in hospitals is the subject of nationwide efforts such as the American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR) change package [10].

CR Liaison

Also well-studied to increase CR referral and participation is the assignment of a CR liaison from the CR program during the hospital discharge process to help the patient understand and navigate the process from hospital discharge to CR participation. In one of the previously mentioned studies, a CR liaison increased CR referral from 32.2% to 59.0% with even better results from the combination of automated referral and CR liaison, resulting in referral in 85.8% of patients [53].

CR Referral as Performance Measure

Referral to CR was designated as a performance measure (or quality indicator) in 2007 by AHA/ACC/AACVPR [56]. This designation appears to have increased CR referral from 73-91% over the time period 2007-2012 in the U.S. [57]. The AHA’s “Get With the Guidelines Program” also made CR referral a quality indicator during the hospital discharge process with CR referral measured at 56% for data collected from 2000–2007 [58].

Interventions to Increase CR Uptake (Participation)

Computerized “Automated” Referral

Compared with controls, an automated referral to CR (described above) increased CR participation from 29.1% to 60.7% [53]. In another Canadian study, automated referral was also shown to increase CR completion rates by 28% [55]. However, it should be noted that results from analyses of CR participation rates over time have not necessarily seen increases in CR enrollment even though referral rates have increased [14,17]. Additional efforts towards enhancing access and optimizing insurance coverage may be required.

CR Liaison

Compared with controls, use of a CR Liaison, as described above, increased CR participation from 29.1% to 50.9% [53]. The combination of both computerized referral and CR liaison proved most effective, leading to a significantly higher participation rate of 74.0%.

Strong Physician Recommendation

A strong physician recommendation from the discharging hospital physician is an independent predictor of increasing CR participation rates [15,19]. Patients look to their physician to help make most health care decisions and CR participation is no different. In the first study of this issue, Ades et al showed that a strong MD recommendation (4-5 on a scale of 1-5) was associated with a CR participation rate of 66% whereas a weaker or no recommendation for CR (1-3 on a scale of 1-5), the participation rate was 2% [15]. These results were replicated in a more contemporary study [22].

Motivational Interviewing

Barriers to attendance can also be psychological as patients may not think that they need CR or may worry that increasing their physical activity will be risky. One randomized controlled trial showed that motivational interviewing can increase the likelihood of CR attendance by allowing patients to view CR as more necessary and assuaging concerns about the risk of exercise [59].

“Nearer to Home” Referral

Time to travel to a CR program is a powerful barrier to participation [51]. For patients who are referred to a secondary or tertiary care center for cardiology care, a “nearer to home” CR referral process is associated with a marked increase in CR participation [51]. Thus, the coordination of CR referrals from the discharging facility to the appropriately located outlying CR programs is a key obligation at the discharge process.

Reducing Time to Start CR

In many CR programs the mean time to CR participation form hospital discharge is over 30 days. As the time between discharge and enrollment increases the likelihood that a patient will enroll, and complete, CR decreases, with approximately a 1% loss of CR enrollment for every day of delay [60]. Furthermore, longer wait times to start CR have been associated with an attenuated training response of cardiopulmonary fitness in patients after coronary bypass surgery [25]. Accordingly, decreasing the time between hospital discharge and CR should improve enrollment [59,61].

One randomized-controlled trial demonstrated that being assigned to an early appointment to outpatient CR (10 days after discharge, compared with 35 days) significantly increased CR attendance [62]. Even if the patient is not ready for exercise at this early visit, they can be assessed for symptoms or medication side effects and scheduled for CR intake stress tests or move directly to participation depending on program policies.

Flexible Scheduling

As mentioned above, there is evidence that decreasing the time between hospital discharge and a scheduled appointment at CR will result in increased attendance at an orientation session [62]. One way to potential decrease wait times is by altering the way CR is delivered. For example, CR programs can implement group rather than individual orientation meetings. Additionally, programs can adopt an “open gym” format where patients are given blocks of times they can exercise in rather than strict appointments. In one program, implementation of a group enrollment and open gym format was associated with a significant decrease (3.74 fewer days) in wait times until first CR session [63].

Programs can also increase the number of patients served (and thus decrease time for any one patient to start) by adjusting the number of in-person session per week that a participant attends. Low-to-moderate risk patients could be seen in-person only once per week which would allow for the program to see more patients overall potentially decreasing the time until a new patient could start.

Elimination Patient Co-Pays

Costs of healthcare can be a significant barrier for many patients, and CR is no exception. In one study the presence of any co-pays or unmet deductibles in patients referred to CR was associated with 6 fewer sessions of CR completed, with a dose-response relationship between increasing levels of cost-share and fewer sessions completed [64]. While it is reasonable to assume that the elimination of co-pays and deductibles would result in a higher rate of participation in CR, to the best of our knowledge, such a study has not been performed. However, in a related area, a large randomized trial demonstrated that eliminating co-payments for preventive medications following myocardial infarction decreased incidence of recurrent major cardiovascular events and decreased overall patient costs [65]. It could be assumed that eliminating co-pays for CR would improve participation and have similar morbidity and cost benefits.

Transportation Assistance

Driving time to CR and/or a lack of efficient transportation to CR are also significant barriers to CR participation [15,66]. However, providing transportation to individual patients to attend CR would, in most situations, be unwieldy and expensive. One solution to inadequate transportation to CR, or a lack of geographically available CR programs, is the development of home and hybrid CR programs (see below).

Home/Hybrid CR Programs

It is likely that new delivery strategies will be required to expand CR participation to underserved populations. One prominent option is that of home-based or hybrid CR programs which combine on site-visits with home exercise and risk factor teaching, with a goal of expanding access to CR (for in-depth discussion of hybrid CR models see [67]). With a hybrid model, participants attend in-person less frequently than the traditional CR model with the expectation that they will complete more of their exercise outside of the program. The hybrid model of CR is appealing in that participants can maintain contact with medical professionals while establishing a routine of regular physical activity performed away from the formal CR setting. Developing a routine of regular physical activity at home while under the guidance of CR professionals may help with long-term adherence; the patient will not have to make an abrupt shift from performing most of their exercises at the CR program to having to do it all at home.

Guidelines for home-based CR have been published and approved by AHA, ACC and AACVPR [68]. Important lessons can be learned from the expansion of home-based CR in the Veterans Administration (VA) system of medical care in the US. Classically, the VA referred cardiac patients for participation in non-VA CR programs but referral processes were burdensome and participation was very low, in the range of 6% [69]. Between 2010 and 2015, at VA hospitals that began implementing home-based CR, participation increased from 6% to 24.6%. Patients who participated in home-based CR entered sooner and improved their functional capacity, measured by the 6-minute walk, to a greater degree than patients in facility-based programs [70]. Data from meta-analyses suggests that functional capacity gains from home-based CR are similar to those garnered from center-based CR, as are gains in health-related quality of life and effects on mortality [71]. It is quite clear that a judicious expansion of facility-based CR to include hybrid and home CR options can only increase access. However, it is notable that, in general, only lower risk CR patients are felt appropriate to participate in home CR-programs, again highlighting the importance of hybrid programs where patients can intermittently be assessed in person.

Given the enormous geographic inequality of available CR programs [48,50], hybrid CR programs will also play an important role in reaching rural patients or patients in CR deserts, where programs can be coordinated from the regional hospital or cardiology center. It is also of interest whether new models of care are benefiting the underserved, hybrid models seem to meet that consideration. One study implemented a hybrid model of CR among mostly lower-SES patients and found very few session absences, concluding that the flexibility of a hybrid program might particularly benefit these patients [33].

Interventions to Increase CR Adherence

Referring patients and having them enrolled in CR are important quality benchmarks. However, keeping patients engaged, having them complete the program, is also an important metric. For example, it is estimated that each 6 additional sessions of CR completed translates into 6% lower risk of MI [72]. Despite this important seemingly dose-dependent relationship, fewer interventions have been tested to improve CR adherence.

Motivational Interviewing

One study aimed to improve adherence among patients at risk of dropout by implementing a motivational nursing-based telephone intervention 1-3 days after CR orientation [73]. Those who were randomized to receive these calls were 62% more likely to attend their next CR session as scheduled.

Incentives

The use of financial incentives to improve health-related behaviors was first tested as a method to promote drug abstinence for cocaine addiction [74]. Additionally, the use of financial incentives have been expanded for use of other health-related behaviors such as promoting adherence to health-care visits for hepatitis vaccines and methadone treatment [75,76]. Incentives appear particularly effective at promoting attendance at health-care visits, and in CR, where health benefits appear to be dose dependent [5,72], the ability to sustain participation to completion would be of significant value.

One randomized clinical trial has been performed to test the use of financial incentives to increase CR participation in lower socioeconomic status patients [77,78]. This study demonstrated that the use of incentives almost doubled completion rates (55% vs 29%) and increased mean number of sessions completed from 14.7 to 22.4 in this at-risk population. A smaller quality improvement trial using incentives in a general CR population has also been conducted [79]. This study demonstrated that small rewards for consistent CR attendance such as parking passes, T-shirts, and water bottles increased median attendance from 12 to 20 sessions.

Home/Hybrid CR Programs

Given oft cited issues of transportation and inconvenient scheduling for early dropout (e.g. [80]), it seems likely that implementing a home or hybrid program would improve adherence to CR. There is some evidence to this effect, the home CR program offered by the VA, described above, has been shown to increase completion rates, particularly in patients after coronary bypass surgery [81]. Additionally, a controlled trial of smartphone-supported home CR vs traditional CR from Australia showed increased uptake, adherence and completion of CR [82]. Indeed, smartphone supported programs as alternatives or adjuncts to traditional CR has been tested multiple times. In a meta-analysis of 4 randomly controlled trials, the use of mobile applications led to a 40% increase in CR completion and adherence [83].

Suggested but Untested

The literature suggests many interventions that could improve attendance, but remain untested [84]. For example, it has been suggested that adherence would be higher if CR programs were available through local community facilities such as churches, or if interventions or educational resources were more individually tailored. However, the efficacy these ideas is unknown.

Interventions to Increase CR Participation in Underserved Populations

Overall, there is limited research examining ways to improve CR attendance specifically in underserved populations. However, some general interventions to improve CR participation have proven to be successful among underserved populations. Physician recommendation and use of automatic electronic referral have been found to increase participation in all patients including underserved groups [15,19,53]. Further, the use of an automated referral at time of hospital discharge can help overcome a potential bias from the provider level as to whether or not someone is suitable for CR; for example, women and minorities. This is particularly pertinent as these groups are often under-referred to CR [22]. As Grace et al found, a systemic referral process can improve CR referral by a factor of 2-7 times [53].

Some research has examined interventions for specific underserved populations. For example, to improve enrollment and attendance, different models of CR have been examined specifically in women with mixed results. One study compared adherence between women in standard CR, supervised women-only CR and home CR and found no difference in adherence [85]. This however, differs from a randomized controlled trial in which women were randomized to standard or a gender-tailored CR program using motivational interviewing [86]. Those in the gender tailored intervention attended 4 more sessions compared to those randomized to standard CR. Additional research, therefore, is needed to assess the role of gender tailored CR programs.

There are also a limited number of interventions that have been tested to improve CR participation in those with lower-SES. One approach is to adjust programs to better serve lower-SES patients. Two studies demonstrated that providing additional CR sessions or providing monthly phone calls focused on goal setting around secondary prevention to lower-SES patients improve health outcomes compared to control groups that were not offered additional sessions or contacts [87,88]. However, the benefits may not last long-term [89]. Another study implemented a hybrid model of CR among mostly lower-SES patients and found very few session absences, concluding that the flexibility of a hybrid program might particularly benefit these patients [33]. An additional study examined the effects of providing written and oral explanations of the benefits of CR to patients before hospital discharge and found that the intervention was particularly successful at increasing CR attendance among lower-income and lower-education patients [90]. An additional set of research studies has focused specifically on improving participation in traditional CR among lower-SES patients. These studies have demonstrated that providing financial incentives for completion of CR visits is successful at improving attendance, increasing number of sessions attended and doubling completion rates [77,78]. Additional on-going research is examining whether providing a case-manager to lower-SES patients in hospital can smooth entry into CR, or provide on-going encouragement and support, with the aim of improving attendance [91].

Other underserved populations appear not to have been the target of a controlled trial. For example, to our knowledge, those who smoke have not been the target of efforts to improve CR participation. There is some evidence that improving smoking cessation could translate into better CR attendance [92], however, an intensive trial of smoking cessation in those eligible for CR would be needed to test such a hypothesis. We also know of no interventions specifically targeting those with cognitive impairment, however, there is on-going work on improving CR among older populations where cognitive challenges would presumably be overrepresented [91]. Minority populations appear to also be understudied in regards to testing interventions to improve participation. Work has been done to document the barriers to participation that exist for these populations (e.g. [28,93]), and there have been efforts to adjust programming to better reflect the needs of minority populations (e.g. [94]), but whether these adjustments translate to better attendance has not been well studied.

Similarly we know of no controlled trials on improving CR participation among those who are distant from a CR program. Studies have demonstrated that distance to the CR program is a major determinant of participation [95]. Establishing community-based programs that are not associated with a hospital would seem to be key to expanding access to CR. Unfortunately, however, U.S. Centers for Medicare and Medicaid Services, the largest single provider of health insurance coverage for participants in CR, reimburses programs approximately 40% less for non-hospital based program. This discrepancy makes it difficult for physician or community-based programs to be fiscally solvent as even larger, hospital-based programs operate with very little margin [96]. Remedies to this reimbursement issue needs to be pursued to help expand the availability of non-hospital-based programs and increasing the number of programs available will be necessary to reach the 70% CR participation goal [97]. Additionally, expanding access in the community would hopefully have positive effects on attendance in underrepresented groups such as minorities and patients with lower-SES.

Overall, within the current literature, it appears that CR participation rates in many underserved populations have been documented, but controlled trials examining whether interventions can improve participation in these populations is sorely lacking. Accordingly, removing barriers to accessing CR that are overrepresented in these populations appears to be the best course of action.

Expanding the Model: Program Availability

Ideally, all eligible patients would participate in some version of in-person CR. There has been no randomized control trial that have demonstrated that alternatives to traditional, in-person CR are superior. Therefore, a primary goal should be optimize access to in-person CR. If efforts to increase referral and subsequent enrollment in CR are successful, programs will need to adjust to accommodate to the greater demand. One analysis suggest that the existing infrastructure could only expand enough to care for about 50% of eligible patients [97]. With a goal of increasing participation to 70% of eligible patients, there is a need to expand program availability [11].

In the short term, the most straight forward way to increase capacity is for existing programs to expand the hours and the number of days of operation. This would allow for patients and the individuals that could facilitate participation (i.e. family members and significant others) greater access to CR programs.

Programs can also increase the number of patients served by limiting the total number of in-person session per week that a participant can attend. Instead of attending 3 session per week, for example, low-to-moderate risk patients could be seen in-person once or twice per week in conjunction with providing advice and counsel to those individuals regarding a home exercise regimen and cardiovascular risk factor modification. Individuals that are deemed higher risk could be prioritized for more frequently in-person sessions. Seeing patients less frequently, in-person, would allow more patients, in total, to access CR. Moreover, spreading allotted sessions out over an extended period time has been demonstrated to be effective in improving outcomes [98,99]. Importantly, extending the program over a longer period of time can be done at no greater cost than the standard CR programming [98].

Program availability can also be increased by expanding beyond traditional CR settings. The use of technology has long held the potential to expand the reach of CR [82,100,101,102]. The technology that is available for monitoring measures of health is ever expanding. Wearable devices allow for the monitoring of a whole range of measures such as electrocardiogram, blood pressure, and physical activity. Accurate and responsive monitoring of home vital signs and ECG will help extend the reach of hybrid CR programs.

As an alternate to traditional CR, primarily home-based CR programs have proven to be effective in selected individuals [68,71,103]. Compared to participants in traditional CR programs, individuals enrolled in home-based programs have similar improvements in aerobic capacity and measures of quality of life. An important consideration is that all participants in these studies complete a thorough baseline medical evaluation prior to initiating a home-based program. A comprehensive baseline assessment most certainly enhances the safety of home-based programs as the program is crafted based on this medical evaluation. Therefore, home-based programs should be viewed as an extension of a center-based program

While reimbursement for alternatives to center based programs is suboptimal, cost from the would-be participant perspective is also barrier to participation. Lack of adequate insurance coverage restrict program access [95,104]. Other insurance related expenses such as co-pays and deductibles are also significant barriers [64]. Moreover, expenses related to transportation, parking fees and time away from work are other potential financial deterrents to CR participation. By expanding to hybrid and at-home options some of these barriers could be overcome.

Expanding the Model: Addressing CR Non-Participants

Enrolling in CR is the ideal process to continue secondary prevention strategies post hospital discharge. However, there will always be a significant portion of the CR-eligible population that cannot, or chooses not to, attend CR. Eligible patients who do not attend CR are a higher risk group than patients who choose to attend [38] with a greater burden of behavioral cardiac risk factors and, indeed, clinical benefits of secondary prevention may be greatest in this higher risk group, who are at increased risk for subsequent events and hospitalizations. Regardless of the reason for CR non-attendance, these patients merit aggressive secondary prevention efforts. Non-CR participants will be more likely to be women, minorities, lower SES, or those who smoke and will likely be the youngest and oldest of those eligible for CR [31,38].

A number of strategies have shown some promise in delivering secondary prevention to this group of patients. Jelinek et al [105] demonstrated that the COACH program (Coaching Patients on Achieving Cardiovascular Health) of telephone coaching, resulted in substantial improvement in multiple cardiovascular risk factors during a six-month period following hospital discharge [105]. Importantly, while there was some deterioration after 6 months, cardiovascular risk factor status was remained substantially improved compared to baseline at a two-year follow-up. Similar results were seen from a second study with the counseling provided via text message follow-up post-hospital discharge (the “TEXT ME” trial; [106]), where the program yielded a substantial increase in home physical activity vs. controls. Technically, these interventions were not home CR programs but rather focused on supporting control over a broad array of cardiac risk predictors. Similarly, case-managed cardiac prevention has a long history of success but has not been implemented broadly in the U.S. (e.g. [107]). Patients who cannot attend or sustain attendance at traditional CR should be enrolled into a hybrid or home CR program. Patients who choose not to attend any aspect of CR should be contacted by a case manager and considered for a telephone-based or text-message based program of cardiac prevention as described above [105,106].

Summary

Overall, with CR being a Class 1A recommendation, it is imperative that efforts be directed to increase participation [11]. While there is evidence that enrollment in CR has increased modestly over the past decade, much work needs to be done to realize the goal of increasing participation to 70% of eligible patients. The single biggest impact toward improving attendance at currently available CR programs would be for hospitals to universally adopt automated referral of patients to CR as part of the hospital discharge process. Other interventions reviewed could further improve attendance and adherence at these programs. However, there continues to be populations who are underrepresented at CR, where research on improving attendance is sorely lacking. It is likely, given the barriers that exist for some of those populations, programs will have to expand and adapt to best serve all of those who would benefit from CR. Given the capacity limitations of currently available programs, the existence of CR “deserts”, and the varied barriers to attending traditional CR programs, it is likely that the future of CR will need to encompass interventions to improve attendance as well as expansion of hybrid programs and, perhaps, the creation of secondary prevention programs for those who cannot, or will not, attend CR.

Alphabetical List of Abbreviations:

AACVPR

American Association of Cardiovascular and Pulmonary Rehabilitation

AHA

American Heart Association

ACC

American College of Cardiology

CABG

Coronary Artery Bypass Graft

CMS

Centers for Medicare Services

CR

Cardiac Rehabilitation

CV

Cardiovascular

HF

Heart Failure

MI

Myocardial Infarction

SES

Socioeconomic Status

VA

Veterans Administration