Maintenance Pulmonary Rehabilitation: An Update and Future Directions

Abstract

The Global Initiative for Chronic Obstructive Lung Disease guidelines recommend pulmonary rehabilitation (PR) for individuals with COPD to improve exercise capacity and health-related quality of life (HRQOL) and reduce symptoms of dyspnea. For cost-effectiveness in COPD care, PR is second only to smoking cessation. However, PR programs typically last 9–12 weeks. The benefits of PR in terms of exercise capacity and HRQOL often decrease toward pre-PR levels as early as 3–6 months after completing PR if patients do not continue to engage in exercise. This review will (1) briefly summarize the efficacy data that informed the 2023 American Thoracic Society (ATS) clinical practice guidelines for maintenance PR, (2) discuss exercise components of maintenance PR studied since 2020 when the last papers were included in the ATS guidelines, (3) explore future directions for delivery of maintenance PR using technology-mediated models, and (4) examine the need for behavior change techniques informed by theoretical models that underpin long-term behavior change. This review will focus on persons with COPD who have completed an out-patient core initial PR program as most of the data on maintenance PR have been published in this patient population. Core PR typically implies a facility-based initial intensive structured program. All patients who complete a core initial PR program should be counseled by PR staff at the discharge visit to engage in ongoing exercise. This usual care is equally as important as referral to a formal PR maintenance program. It is critical to emphasize that usual care after core initial PR means all patients should be supported to participate in regular ongoing exercise, regardless of whether supervised maintenance PR is available. Currently, the optimal frequency, exercise and/or physical activity content, and delivery mode for maintenance PR in persons with COPD and other chronic respiratory diseases remain unknown. Patient safety and degree of in-person supervision required due to the severity of the underlying lung disease need to be considered. Future research of maintenance PR should be underpinned by behavior change techniques. Finally, in the setting of finite resources, balancing the competing priorities of core initial programs with those of maintenance PR programs needs to be achieved.

Introduction

Pulmonary rehabilitation (PR) is standard of care for patients with COPD. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines recommend PR for individuals with COPD to improve exercise capacity and health-related quality of life (HRQOL) and reduce symptoms of dyspnea.¹ For cost-effectiveness in COPD care, PR is second only to smoking cessation.^2,3 However, PR programs typically last 9–12 weeks. The benefits of PR in terms of exercise capacity and HRQOL often decrease toward pre-PR levels as early as 3–6 months,⁴ and definitely by 12 months after completing PR if patients do not continue to engage in exercise.^4,5 The Pulmonary Rehabilitation for Adults with Chronic Respiratory Disease: An official ATS Clinical Practice Guideline, published in 2023, recommends either supervised maintenance PR or usual care after core initial PR.⁶ Core PR typically implies a facility-based initial intensive structured program. It is important to emphasize that usual care after core initial PR means all patients should be counseled to participate in regular ongoing exercise, regardless of whether supervised maintenance PR is available.

There is currently no consensus on the optimal approach to sustain the gains made in PR over the long term. Maintenance programs are defined as ongoing supervised exercise at a lower frequency than that delivered in the core initial PR program itself, with or without the components of education and/or disease self-management.⁷ It is unknown whether in-person versus remote supervision is an essential component of maintenance PR. New delivery models of core initial PR and thus maintenance PR have evolved that include home-based strategies and telerehabilitation, but their efficacy is still being studied.

Since level of supervision, exercise components, delivery modes, and frequency for core initial PR programs can vary between sites, maintenance PR programs will necessarily reflect those variations too. The efficacy of and adherence to core initial and maintenance programs depend on the frequency and intensity of supervised exercise, the strategies used to improve adherence to exercise, and facilitators and barriers to endurance exercise training.^8-10 The efficacy of maintenance programs also depends on the efficacy of the core initial PR program from where the patients graduated. Since maintenance is a long-term intervention, it needs to account for complex and interrelated individual-level factors that may be logistical and/or motivational, social and behavioral support needed to sustain behavior changes, and intercurrent illness that impact ability to exercise. Acute exacerbations (AEs) occur frequently following PR and predict worse PR outcomes at 12 months.¹¹ Wageck et al¹¹ reported 68% had a moderate exacerbation and 34% a severe exacerbation in a cohort of 166 subjects.¹¹ Attention to rebuilding the patient’s confidence and motivation to exercise after a medical setback is critical to ensuring ongoing engagement and maintenance of benefits after core initial PR. Provider-level factors such as personnel and space resources that are sustainable and health care system-wide factors such as reimbursement by payers for maintenance PR need to be optimized.

This review will (1) briefly summarize the efficacy data that informed the 2023 ATS clinical practice guidelines for PR,⁶ (2) discuss exercise components of maintenance PR studied since 2020 when the last papers were included in the ATS guidelines, (3) explore future directions for delivery of maintenance PR using technology-mediated models, and (4) examine the need for behavior change techniques informed by theoretical models that underpin long-term behavior change. This review will necessarily focus on persons with COPD who have completed an out-patient core initial PR program as most of the data on maintenance PR have been published in this patient population.

Efficacy of Maintenance PR

It is important for patients to continue to reap the benefits of the core initial PR program after they have spent time and effort on exercise training. The evidence supporting maintenance PR programs is growing, but heterogeneity of intervention design currently limits definitive conclusions.

The 2023 ATS clinical practice guidelines for PR were based on 2 existing systematic reviews^12,13 and examined the following treatment effects in stable patients with mild to very severe COPD.⁶ Exercise capacity: Compared to control, maintenance PR improved 6-min walk test (6MWT) distance, but the between-group difference did not reach statistical or clinical significance (mean difference [MD] 25.9 [95% CI −1.0 to 52.8] m, 10 randomized controlled trials [RCTs], 639 participants].⁶ Certainty of evidence was very low for this critical outcome.⁶ HRQOL: Compared to control, maintenance PR significantly improved Chronic Respiratory Questionnaire (CRQ) total score (MD 0.5 [95% CI 0–1.0] points, 4 RCTs, 258 participants), CRQ fatigue (MD 0.4 [95% CI 0–0.7] points, 4 RCTs, 210 participants), CRQ emotional function (MD 0.6 [95% CI 0.2–1.0] points, 4 RCTs, 210 participants), and CRQ mastery (MD 0.5 [95% CI 0.1–1.0] points, 4 RCTs, 210 participants).⁶ Certainty of evidence was low to moderate.⁶ Dyspnea: Compared to control, maintenance PR did not improve the CRQ dyspnea domain (MD 0.3 [95% CI −0.5 to 1.1] points, 4 RCTs, 210 participants).⁶ Certainty of evidence was very low.⁶ Risk of respiratory-cause hospital admissions: Compared to control, maintenance PR did not change risk for respiratory-cause hospital admissions (relative risk 0.74 [95% CI 0.41–1.37], 4 RCTs, 279 participants).⁶ Certainty of evidence was low.⁶ Outcomes were evaluated at 6–12 months. In-person supervision was provided in 11 studies, remote supervision in 6 studies, and a combination of modalities in 4 studies. Based on these data, the overall recommendation was “For adults with COPD, we suggest either supervised maintenance pulmonary rehabilitation or usual care after initial pulmonary rehabilitation (conditional recommendation, low quality evidence).”⁶

Usual Care

It is critical to emphasize that usual care after core initial PR means all patients should be counseled to participate in regular ongoing exercise, regardless of whether supervised maintenance PR is available. In fact, education on long-term exercise maintenance should begin at the core initial PR evaluation, with exploration of availability of home exercise equipment, use of physical activity (PA) monitors, social support, and safety of the built environment with respect to sidewalks and walking trails. One-on-one counseling on use of the Borg rating for dyspnea to guide exercise intensity while a patient exercises on a piece of gym equipment during the core initial PR program should extend to advice on use of the Borg rating when exercising unsupervised in the home environment. Education sessions during the core initial PR program should include multidisciplinary input from behavioral psychologists on promoting and maintaining motivation, confidence, self-efficacy, and behavior change. At the discharge appointment, PR staff should work with the patient to develop a feasible, concrete plan written on a discharge prescription for how the patient will continue regular exercise.

Indeed, usual care in research studies almost always have some component of recommendations focusing on continued exercise and adherence to medication and other disease management strategies learned in facility-based core PR. The clinical trials reviewed are thus not comparing maintenance PR versus no maintenance PR, but rather supervised maintenance PR versus recommended but unsupervised maintenance PR.

Studies on Maintenance PR Since 2020

Since 2020, when the last papers were included in the ATS clinical practice guidelines, studies of maintenance PR interventions have examined a variety of strategies with different outcome measures.^14-20 Frei et al¹⁴ examined 123 subjects with COPD who completed PR within a month prior to study entry.¹⁴ In an RCT, they were assigned to a home-based minimal-equipment strength training program or usual care. Training was delivered over 12 months using a combination of home visits and structured telephone calls. Results demonstrated improved 1-min sit-to-stand performance and subject-perceived fitness but no effect on dyspnea.¹⁴ Li et al¹⁵ used the social media platform WeChat with PR guideline videos uploaded weekly by a physiotherapist. Enrolled subjects had stable COPD and had completed an 8-week primary PR program. One hundred forty-eight subjects were randomized to one of 3 groups: maintenance PR via social media supervision at home, maintenance PR at hospital, and usual care. Subjects engaged in the intervention for 12 months. They found benefits were sustained in both PR groups with improvements of 6MWT distance, COPD Assessment Test score, and modified Medical Research Council dyspnea score.¹⁵

Galdiz et al¹⁶ studied a maintenance pulmonary telerehabilitation program, after core initial intensive PR, in a multi-center, open-label, pragmatic parallel-group RCT. Subjects first completed an 8-week out-patient hospital PR program. The intervention group was given training equipment, instructed to perform 3 weekly training sessions, and had to send performance data through an app. The usual care group was advised to exercise regularly. The investigators randomized 94 subjects with COPD. They found the program to be feasible and safe but not superior to control.¹⁶ Butler et al¹⁷ conducted a multi-center pragmatic RCT of community-based post-PR exercise compared to usual care. They enrolled 97 subjects with moderate to severe COPD who had completed PR within the previous 2 weeks. Subjects received a 1-y membership to a community center and were asked to exercise at least twice a week. There were group classes that ranged from unsupervised to supervised depending on the location of the community center. They found that the intervention maintained exercise capacity in those who attended at least 50% of the sessions over 1 y.¹⁷ These results highlight the need to focus on motivational factors that determine long-term adherence.

In a single-arm study, Candemir et al¹⁸ examined 33 subjects with COPD who received regular follow-up visits over 5 y after completion of an 8-week PR program. The follow-up visits occurred at 3-month intervals in year 1 and then every 6 months. Exercise training was re-prescribed at each follow-up. Incremental and endurance shuttle walk tests were preserved over 3 y.¹⁸ Blervaque et al¹⁹ studied a long-term PR maintenance program for subjects with COPD in a real-life setting over 5 y; the intervention was a supervised pragmatic multidisciplinary self-help PR maintenance program. In an RCT, they enrolled 144 subjects who had completed a short-term in-patient PR program and followed them yearly. Survival was higher at 5 y compared to a matched control group. At 4 y post-PR, benefits in 6MWT distance and HRQOL were maintained compared to pre-PR values within the intervention group.¹⁹ Finally, Souto-Miranda et al²⁰ performed a qualitative study with focus groups to assess perceptions by subjects with chronic respiratory disease, informal carers, and health care professionals of community-based strategies that included city resources and facilities close to home such as fire departments and primary care centers. Intrinsic motivation and professional and peer support were key elements identified that were important to maintain benefits of PR.²⁰ They identified barriers to maintenance PR including lack of awareness of disease progression and importance of PA, lack of motivation, fear/concern of doing exercises wrong when doing them alone, and need for regular monitoring and follow-up.

In summary, simple strategies such as regular telephone calls, community-based programs, as well as more complicated designs involving technology/social media may be promising maintenance PR interventions.^14-20 In a mixed-methods analysis of the intervention of Frei et al,¹⁴ participants shared that they appreciated the home-based exercises and the simplicity of the intervention equipment.^21,22 They liked that instructions were printed clearly on paper. However, the same participants endorsed that they missed the group dynamics of center-based PR. They reported feeling more motivated in a group environment and that having an individual coach did not lessen this need for peer support.^21,22 In general, it appears that less hands-on approaches have underperformed both on clinical and adherence outcomes.

Given the heterogeneity, absence of published standards, and lack of data to guide the optimal exercise content or the optimal delivery mode for maintenance PR, it is imperative that future models of maintenance PR are rigorously tested and based on what has already been proven to be efficacious. Core PR is a multicomponent, multidisciplinary intervention comprised of supervised group exercise training and education on disease self-management. Ideally, supervised maintenance PR would maintain fidelity to these key components. In addition, maintenance PR should be based on long-term behavior change techniques grounded in theoretical models and should be rigorously tested for efficacy. Such a model would be widely accessible to patients and scalable for implementation around the world. The overarching goal should be that everyone will know what maintenance PR looks like when they hear the words “maintenance PR.”

Exercise Content for Maintenance PR

The optimal choice of exercise content for maintenance PR is unknown. Theoretically, exercise content would depend on the choice for the main outcome of maintenance PR. Should maintenance PR aim to maintain exercise capacity (fitness) or PA (movement)? PA is defined as any bodily movement produced by skeletal muscles that requires energy expenditure. Since conventional PR has been proven to be efficacious when providing aerobic exercise and strength training to improve exercise capacity, it would seem obvious that if a patient has put in time and effort to achieve a certain level of physical fitness the goal should be to maintain that level. However, practically, after completion of a PR program, not all patients have access to exercise equipment. In addition, preserving a certain 6MWT distance may not be meaningful to patients who may be most interested in maintaining function in their daily activities. Maintaining or increasing PA may be a meaningful goal for maintenance PR. Ultimately, the exercise or PA that a patient can most feasibly maintain over time is the best maintenance regimen for that patient. The optimal exercise content needs to be personalized for each patient.

The pulmonary and systemic consequences of COPD limit aerobic capacity, but there is also a large component of deconditioning that limits functional ability. Improving exercise capacity and increasing PA are two related, but distinct, goals in maintenance PR.^1,23,24 Exercise capacity is the amount of exercise patients are physically and physiologically capable of doing, whereas PA reflects the amount of movement patients actually perform in their everyday lives.²³ The degree of exercise capacity does not equate to level of PA, from associations with genetic markers to effects on body composition.^25-27 Not surprisingly, increases in 6MWT distance from PR do not necessarily translate into increases in PA in the community.^4,28,29 Patients with physiologic abilities such that they “can do” may not have the behavioral skills to “actually do” in the home environment for a variety of reasons, including anxiety, kinesiophobia, lack of safe built environment, etc.³⁰ Traditionally, center-based supervised PR programs have focused on short-term improvements in exercise capacity, as measured by in-clinic tests like the 6MWT.³¹ Conventional PR, which improves exercise capacity, does not always increase PA in the home setting.^32-34 Promoting lifestyle and community-based PA has emerged as an equally important goal in the care of patients with COPD.^1,23,24 Higher levels of PA, self-reported or directly measured with pedometers and accelerometers, are associated with better outcomes in COPD: decreased risk of exacerbations, hospitalizations, and death, independent of lung function.^35-41

Regular walking after completing PR is associated with slower declines in overall HRQOL and walking self-efficacy as well as less worsening of dyspnea during activities of daily living.²⁹ The accumulation of evidence, over the past 10–15 y, that every step counts and the increased attention given to PA as a unique outcome in COPD support that PA is a valid goal to achieve. Persons with COPD with a higher daily step count have a significantly lower risk of dying, independent of FEV₁.^35,36,38,40 Persons who walked the least at study entry have risks that are 2 and 6 times higher for exacerbations and COPD-related hospitalizations, respectively, compared to those who walked the most over a median follow-up of 16 months.³⁷ Patient self-report of any moderate to vigorous PA predicted a lower risk of 30-d hospital readmission after an index COPD hospitalization compared to those who reported no moderate to vigorous PA.³⁹ These studies of daily step count, of any amount and intensity, support that every step walked can positively impact the disease course and should guide the exercise content of maintenance PR. Based on these compelling data, the GOLD guidelines recommend regular PA for all persons with stable COPD as standard of care.¹ In 2021, an international task force on PA proposed standardized methodology for assessment of PA, including standard operating procedures for future research and use as a clinical outcome.^42,43

In a review of the relationship between 6MWT distance and daily step counts as a measure of PA in persons with COPD, Troosters et al⁴⁴ made a novel observation that can be highly relevant for guiding exercise content for maintenance PR. They noted that PA levels were generally low when a 6MWT distance is < 300–350 m.⁴⁴ The relationship between change in PA and change in 6MWT is less clear, but it appears that a patient needs to have a minimum exercise tolerance to be able to perform PA. They concluded that once a certain threshold of exercise fitness is achieved then other factors such as motivation, social support, self-efficacy, built environment, season, climate, air pollution, etc, become more important in determining PA.^44,45 Applying this framework is helpful in personalizing maintenance PR. If a patient achieves a 6MWT distance > 350 m at the discharge visit after completion of core initial PR, then the goal should be to support behavior to maintain or increase PA during maintenance PR. If a patient achieves a 6MWT distance < 350 m after completing core initial PR, then the goal of maintenance PR should be to focus on physiologic limitations to improve exercise capacity and fitness. The supervision and modalities of maintenance PR needed to promote PA versus exercise capacity would be very different. Future studies will need to examine this hypothesis in a rigorous way. It is also possible that there is a subset of patients with COPD with such low fitness and are so deconditioned that they may need to increase their PA through activities of daily living to a certain threshold before they have the exercise self-efficacy and dyspnea management skills to comfortably engage in exercise training to improve fitness.

An interesting alternative to maintenance PR is repeating the core PR program at later intervals or when clinically indicated. Sandoz et al⁴⁶showed that in physician-referred subjects who underwent repeat PR there were clear benefits in exercise capacity following each course, albeit of smaller magnitude than after the core initial program, which was unaffected by the time interval between repeat PR programs. HRQOL and mood improved after the first 2 PR repeat programs but not after a third.^46,47 Issues of long-term access and availability of core PR programs remain unresolved. Details for billing and reimbursement in the United States and globally would need to be addressed in order to make repeat PR programs a sustainable option.

Mode of Delivery for Maintenance PR

Center-based PR programs have limited staff, open gym times, and resources, so their first priority should be to provide initial services to new patients. It may not be sustainable for patients to travel to sites for maintenance PR for indefinite periods of time. From the patients’ perspectives, they face the same barriers to access as for core programs. Core programs are typically based at medical centers or out-patient medical offices, require specialized resources, and ask patients to travel 2–3 times a week. Malla et al⁴⁸ showed that in a nationally representative sample of Medicare beneficiaries two-fifths of adults with COPD overall, and 8 in 9 of those in rural areas, have poor access to PR. Thus, center-based PR programs provide a limited solution for delivering maintenance PR. The optimal mode for delivery of maintenance PR is unknown.

Examples of various home-based exercise training programs⁴⁹ or telerehabilitation,^50-53 delivered as alternatives to core PR programs, abound in the literature, but their clinical efficacy and long-term feasibility are not well understood.^6,7 In a multi-site Canadian study, one home-based program was shown to be non-inferior to center-based programs at 3 months and 1 y on the primary outcome of CRQ dyspnea score.⁵⁴ Yet, home-based PR programs have not been widely adopted, nor have they become standard of care for core initial PR given lack of reimbursement and quality standards. Questions about fidelity to the conventional PR model remain as home-based programs are viewed as “better than nothing” alternative exercise programs rather than replacements for conventional center-based PR programs. In contrast to conventional PR, home-based models tend to provide individualized training, exercise of unknown or lesser intensity, limited multidisciplinary education, minimal social support, and variable patient accountability.

Telerehabilitation, defined as “the delivery of rehabilitation via a variety of technologies [which] encompasses a range of services that include evaluation, assessment, monitoring, prevention, intervention, supervision, education, consultation, and coaching,” holds promise to address the gap in PR access and adherence.⁵⁵ Cox et al⁵⁶ examined telerehabilitation in an equivalence trial and found that it is not equivalent to center-based PR for all outcomes but is safe and achieves clinically meaningful benefits. They concluded that when center-based PR is not available telerehabilitation may provide an alternative delivery model. The technological modalities employed to deliver telerehabilitation differ widely between studies and include telephone calls, web sites with or without telephone support, and videoconferences. A recent study by Zanaboni et al⁴⁹ randomized 120 subjects to telerehabilitation supervised by a physiotherapist, unsupervised training at home, or a control group. The intervention was delivered over 2 y, and subjects were followed during that time. Treadmills were delivered to and installed in participants’ homes, which may not be pragmatically feasible for patients outside of a clinical trial. For the telerehabilitation, there was a combination of videoconferencing and unsupervised sessions. The authors reported an incidence rate of hospitalizations and emergency department visits that was significantly lower compared to controls.⁴⁹

Two-way synchronous videoconferencing may be a platform that mirrors conventional PR the best in terms of direct supervision, particularly if delivered by core PR programs leveraging existing expertise and infrastructure, to increase access to patients who cannot physically attend a center-based PR program. This may also be an effective strategy for delivering maintenance PR. A 2-way synchronous, 12-week program in the United States using minimal equipment conferred benefits similar to those observed in center-based PR⁵⁷ and was efficacious in reducing the 30-d all-cause readmission rate in patients admitted with a COPD exacerbation compared to usual care.⁵⁸ A multi-site RCT based on this 2-way videoconferencing model with a clear path for exercise progression is currently underway (ClinicalTrials.gov NCT05119556). Supervised, accessible, feasible, and cost-effective, such a program can maintain fidelity to key elements of core PR and be well integrated into the existing infrastructure of center-based programs. Opportunities exist for patients to ask questions, providers to give immediate feedback about posture and form to minimize injury, and patients to derive motivation from real-time encouragement and progression of exercise.

At VA Boston Healthcare System, synchronous, real-time classes have been delivered and supervised by PR staff using 2-way audiovisual communication via VA Video Connect since 2020.⁵⁹ The center-based face-to-face PR program leveraged its resources to deliver virtual PR. The 2 modalities were integrated and delivered by the same PR staff in the dedicated PR gym that already had a large monitor for education presentations. Both in-person and virtual options are available twice a week on Tuesday and Thursday mornings. The virtual program occurs first with the PR staff demonstrating and performing the exercises in real-time while monitoring the patients on the screen (Fig. 1). After cool down, the virtual group is joined by the in-person group for education delivered in-person by PR staff. After the education session is completed, the virtual group logs off and the in-person group begins warm-up and exercises using the gym equipment.

Fig. 1.

Real-time pulmonary rehabilitation (PR) classes are delivered and supervised by PR staff using 2-way audiovisual communication. Note: For privacy reasons, patient images on the screen have been replaced with publicly available images.

Results were reported on 99 subjects (57 virtual and 42 in-person) enrolled from July 2020–April 2023.⁵⁹ The most common primary diagnoses included COPD (n = 57), interstitial lung disease (n = 5), pre-lung transplant (n = 4), and post COVID-19 (n = 3). Number of completed sessions did not differ between the 2 groups. At baseline, there were no differences in age, dyspnea score, or 6MWT distance. For the 42 subjects who completed the pre- and post-6MWTs, the virtual (n = 18) and in-person (n = 24) subjects had comparable improvements. Even for an urban site located in Boston, the virtual model increased geographic catchment. The distance from subjects’ homes to the PR center was over twice that for virtual versus in-person subjects (34 ± 23 miles vs 16 ± 15 miles, P < .001), and drive time to the PR center was also significantly longer (86 ± 35 min vs 57 ± 35 min, P < .001).⁵⁹ Two-way synchronous videoconferencing is a promising and safe model for core initial and maintenance PR that aims to maintain fidelity to the essential components of core PR, allowing direct supervision of exercise and real-time delivery of education, particularly as it leverages the expertise and resources of an existing in-person program. Access can be rapidly scaled up, and cost can be sustainable if existing center-based programs are leveraged to deliver virtual PR programs. Two-way synchronous videoconferencing programs may be safer for patients with more severe lung disease compared to home-based models where patients exercise asynchronously and independently.

In summary, telerehabilitation and home-based models may provide a PR maintenance option for patients who cannot otherwise attend a center-based program. It is important to acknowledge that the roles of mindfulness-based movements like tai chi and yoga in maintenance PR may be promising and remain to be elucidated.⁶⁰ A key challenge for all modalities is the ability to deliver exercise training that is individually prescribed and can be safely progressed.⁶ Patient safety and degree of in-person supervision required due to the severity of the underlying lung disease need to be considered.

The Need for Behavior Change Techniques

A critical point of consideration for any maintenance PR intervention is the need to incorporate behavior change techniques that promote exercise self-efficacy, motivation, confidence, and allow maintenance of a positive behavior change like exercise, which is a lifelong commitment. Counseling, motivational interviewing, individualized goal setting, frequent contact with health care professionals for coaching and reinforcement, rewards, and stress and time management have been proposed for behavior changes.^61-64 They are needed to address co-existing chronic pain, anxiety, and depression that are highly prevalent in the COPD population as well as for helping patients resume exercise after an exacerbation or worsening of a comorbidity.⁶⁵ A theoretically informed and systematic approach to behavior change is needed to address the complex issue of exercise adherence. Theoretical themes relevant to maintenance of behavior change include maintenance motives (ie, enjoyment, satisfaction with outcome), self-regulation, resources, habit, and environmental and social influences.^66-68 Health care professionals should collaborate with experts in the fields of health psychology and the behavioral sciences to develop theory-informed interventions that incorporate behavior change techniques to meet the individual needs of people with chronic lung disease.

In this last section, we present a web-based platform to promote PA in patients with COPD that is grounded in behavioral theory. The intervention is based on self-regulation theory that emphasizes an iterative process of behavior change^69-71 (Fig. 2). Efficacious when compared to usual care, it may represent one component of a sustainable strategy for maintenance PR. We developed Every Step Counts to increase lifestyle PA (decrease sedentary time), alleviate existing and prevent further deconditioning, and lessen the burden of COPD.^72-75 Every Step Counts is a dynamic, web-based program, easily accessed from home via a universal resource locator that interfaces with a pedometer. Persons working toward a behavioral goal learn from successes and failures and use this knowledge to develop effective behavioral strategies to achieve their goal. Accurate self-monitoring, feedback, goal setting, and social support are critical components of the cycle of self-regulation.^69-71 Every Step Counts supports the cycle of self-regulation with 4 unique components to promote walking: (1) objective walking assessment and iterative feedback, (2) individualized step-count goals, (3) motivational messages and educational content, and (4) online community.

Fig. 2.

Home page of web-based intervention with the 4 components of the Theory of Self-Regulation that emphasizes an iterative process of behavior change.

Individualized goal setting is calculated with an automated algorithm based on what the participant uploads each week from their pedometer. Each week’s goal is the lowest of 3 possible numbers: (1) the average of the most recent 7 d of step counts + 400 steps, (2) the previous goal + 400 steps, or (3) 10,000 steps/d. Each Sunday the participant receives an updated step-count goal for the following week, which is also displayed on their home page, allowing for progress over time. Feedback is provided on the face of the pedometer as well as on the home page where participants can view daily and weekly graphs of progress. Participants can be advised to engage in walking at a certain intensity level, using the Borg rating for dyspnea similar to that used in conventional center-based PR.

Educational tips and motivational messages are short paragraphs with links to publicly available web sites. The content is based on topics—such as medication knowledge; management of symptoms of dyspnea, anxiety, and depression; and early identification of exacerbations—commonly addressed in the education portion of conventional core PR programs and as part of COPD self-management programs. They provide general exercise and COPD-specific education, such as how to manage shortness of breath when walking, walking and wearing oxygen, and use of inhalers prior to walking. The educational and motivational content was developed through a multidisciplinary collaboration between pulmonologists and behavioral psychologists.⁷⁶ Participants can view this content on their personalized home page anytime in the comfort of their home. Like education in conventional center-based PR, the educational content on the web site is meant to enhance self-efficacy, motivation, and confidence.⁷⁷

The efficacy of the Every Step Counts intervention to increase PA has been proven in 3 RCTs.^78-81 The average increase of 800–1,300 steps/d observed is well within the published clinically important difference of 600–1,000 steps/d.^82,83 Although these increases may be perceived as modest, they are associated with significant benefits in this frail population. Participants who experienced an average increase of 804 ± 356 steps/d, compared to controls, had a significantly reduced risk of exacerbations (rate ratio = 0.51 [95% CI 0.31–0.85]) over 12–15 months.⁸⁴ Notably, the web-based intervention could reach older, rural veterans.^78,85-87 In one RCT that intentionally recruited rural participants, zip codes were matched with the Rural Urban Commuting Area Codes to determine whether one’s residence was urban or rural.^78,85

Rigorous follow-up examined the sustainability of benefits. Improvement in daily step counts started to diminish at 6 months, with convergence at month 8 between the intervention and control groups.⁷⁹ A deeper examination of the intervention group provided a glimpse into user behavior and how participants could be supported for ongoing engagement and adherence.⁸⁸ Waning pedometer adherence and web site engagement by the intervention group were observed over 12 months.⁷⁹ Such an intervention could be cost-effective for use in maintenance PR with the individual using monitoring, goal setting, and iterative feedback to maintain PA levels achieved after core PR completion.^89-91 Regular contact with peers and PR providers, virtually or in-person, can address social, behavioral, and environmental barriers. Finally, routine office visits with health care providers can provide counseling and education around exacerbations and flare-up of comorbidities to promote long-term exercise self-efficacy and adherence⁹² (Fig. 3).

Fig. 3.

Conceptual model of how long-term physical activity can be maintained. Note: Participants are shown wearing the StepWatch Activity Monitor (Modus Health).

Summary

All patients who complete a core initial PR program should be counseled and supported to engage in ongoing exercise. This usual care is equally as important as referral to a formal maintenance PR program. Currently, the optimal frequency, exercise content, intensity, and delivery mode for maintenance PR in persons with COPD and other chronic respiratory diseases remain unknown. Patient safety and degree of in-person supervision required due to the severity of the underlying lung disease need to be considered. Reimbursement and quality standards for maintenance PR are needed. Future research of maintenance PR should be underpinned by behavior change techniques. Finally, in the setting of finite resources, balancing the competing priorities of core initial programs with those of maintenance PR programs needs to be achieved.

Panel Discussion

MacIntyre: I don’t get it. Why does the ATS make a recommendation that usual care is okay?¹ I mean, it just baffles me. You can say that the evidence is low, you can say that the evidence is heterogeneous, all the disclaimers you want to put in there, but the notion that they’ve succeeded in core rehab and that it’s okay to then just toss them back into the real world doesn’t make sense. To me it makes the ATS look stupid. I don’t get it. Can you help me understand? That statement is going to hurt us, I think a lot, in trying to lobby for resources and the like for long-term support. To say usual care is enough and forget about them just seems like a silly statement to make.

Moy: Carly, do you want to take this one?

Rochester: Sure Marilyn, although I’d welcome your input too. This was a matter of vast discussion in the context of preparing the guidelines. And the final recommendation in the guideline regarding the PICO question on maintenance pulmonary rehabilitation simply comes down to having to follow very strict adherence to analysis and structured presentation of data according to ATS rules for clinical practice guidelines development. Based on the analysis of existing randomized controlled trials and the GRADE evidence analysis process, we were not able at this time to formally recommend routine participation in maintenance PR programs. This does not mean we do not inherently believe in the process of maintenance adherence to exercise. Indeed, our guideline committee felt strongly that individuals who have undertaken PR should continue to adhere to exercise and other health-enhancing behaviors after program completion. This is articulated in the guidelines document. We do not advocate that patients undertake PR and then stop exercising thereafter. Further research is needed regarding maintenance PR programs for individuals with COPD and other chronic respiratory diseases.

MacIntyre: But you did.

Rochester: We have no choice in a formal recommendation but to state what the data show as it currently stands. That’s the way it works in the clinical practice guidelines. But we do comment on that. The recommendation as shown by Marilyn in her talk, for the sake of brevity, does not note the whole section where we do talk about this. We qualify the recommendation by clearly saying we do recommend that every person go on to have regular physical activity and exercise behavior, but we can’t officially recommend something that the data don’t yet show. That’s really what it comes down to. Does anybody else who was involved with the guidelines development want to comment?

Nici: As a co-author of another CPG, I agree that you can’t recommend something without the evidence being there. It’s a real problem with CPGs, and in fact, the ATS is now going to allow clinical statements back into the fold which will review the evidence but not use GRADE. So, it will give expert opinion, because what you’re talking about is expert opinion, which is very different from a CPG. But I agree with you because with the COPD CPG² it was very challenging to realize you could not make a recommendation that you believed in because the data weren’t there.

MacIntyre: But I thought Marilyn made a nice case for physical activity.

Nici: Some of these studies were observational, right?

MacIntyre: But there were randomized trials in there, weren’t there, Marilyn?

Moy: Physical activity was not studied in the framework of maintenance pulmonary rehabilitation. It is frustrating, very frustrating. Other thoughts?

Bhatt: I have a follow-up question pertaining to the lack of data. If you were to design a clinical trial to show maintenance rehab works, what would be the ideal duration of such a study because you can keep on going. What would make sense to do?

Moy: The VA has recently put out an request for applications on rehab interventions allowing for 7 years of follow-up. I would say at least a year follow-up and then more if possible would be helpful to understand the behavior change and how we can maintain the benefits after completion of the core program.

MacIntyre: This discussion reminds me of the discussion about parachutes. I can see the ATS saying it’s okay to jump out of a plane without a parachute because we don’t have any data supporting the utility of parachutes.

Nici: Just to play devil’s advocate though. People have to do maintenance forever for this to work, right? As soon as you stop, the gains are gone unless you have behavior change. I think it’s not unreasonable to say that maintenance is probably not effective because we’re not promoting behavior change. I think our efforts should be focused on finding cost-effective ways to support increased physical activity. And that’s challenging, that’s really, really challenging because healthy adults often lapse with their exercise, life gets in the way. And these patients have symptoms that are progressive and they’re having exacerbations. I’m actually not sure that maintenance, just exercise or activity, is not where our efforts are going to be most fruitful. It’s going to be in the psychological part of this.

Rochester: I think there’s no question that patients benefit from continued exercise, and I agree with Marilyn that a goal should not just be exercise capacity but that physical activity is also important. If a maintenance program is to be undertaken, I think maintaining or increasing physical activity is an important goal. But we really simply don’t have a body of data right now that proves a definitive role for maintenance PR such that it can be formally recommended for all. We have some really great studies that have made great effort to show benefit, but the evidence is heterogeneous. I think we must not just to stop here – I think this is an area that that needs ongoing work. It’s a huge challenge, as Linda [Nici] said. But I think it needs ongoing work and that perhaps we have to reframe it and maybe not make it all about center-based maintenance programs. There’s also the suggestion that unsupervised programs aren’t as effective for maintenance, so we need to determine what is most effective. Programs such as the ones Marilyn was articulating this morning are very promising and hopefully will reach large numbers of people, and hopefully when the next iteration of guidelines comes out we’ll be able to say something different. We do actually, again, want to reiterate that in the guideline we discuss this and qualify the recommendation and note why it has to be the way it is. We share the disappointment over that, but we had to adhere strictly to the ATS guideline development process. On another note, Marilyn, I wonder if you could just tell us a little bit more about the virtual program that you do, what series of exercises are the patients doing?

Moy: The aerobic exercise is anchored in marching in place. We’re trying to increase the heart rate with an aerobic exercise, but it has to be feasible in the home environment without a piece of exercise equipment. For the strength training, they are using soup cans or good old-fashioned water bottles. We intermittently ask them to check their heart rate, check their oxygen saturation, and give us a Borg rating on their dyspnea and leg fatigue. We use those numbers to help us figure out where they are and then how to progress them next. They do increase their heart rate, so they do not have to be on a treadmill in order to accomplish that.

Rochester: Are they mirroring the physiotherapist doing it live at the same time?

Moy: Yes. That’s a really important point, the exercise is synchronous. Our physical therapist is doing it at the same time they are. In fact, we had an older person in the position who was doing the exercises twice a week, and she wasn’t able to do the teaching; she said this was not part of my job description. You know, my shoulder is hurting me, my hip is hurting me. I can imagine if we try to increase virtual classes to twice a day, which we have entertained, what that would mean in terms of physical burden on our staff needs to be recognized. But yes, patients are mimicking, following exactly what the physical therapist is doing. We definitely get them to have their full body visible on the screen so that we make sure they are doing the exercises properly and they’re not at higher risk for injuring themselves. We’re able to monitor them carefully.

Criner: Yeah, I think the standard for a rigorous clinical trial in trying to enhance physical activity and show an outcome is impractical mainly because you don’t really know if it’s cause or effect. And these patients because of their many lung and non-lung comorbidities, everything must in aggregate go together. I think physical activity is a requisite in a composite program to treat any patient with COPD where the objective is to improve the quality of life, not so much to show that you decrease hospitalizations. If you do that, that’s great. But there are many other factors that can go with that. And I would like to echo what Neil [MacIntyre] said. I think that statements like that from a major society that don’t definitively say that physical activity is a requisite of any program really don’t advance the field. It will not garner public support for that. It will not get regulatory attention. You hurt the development of the field rather than help it with lukewarm statements like that which are different from any other societies. It’s clearly in the cardiology recommendations from the American Heart Association³ that rehab is a requisite for patients with heart disease. That’s the kind of statement that we need. The difference from the GOLD report, it’s not a guideline, it’s a report.⁴ We don’t use methodologists for just that reason. There’s that clinical gestalt that must go into managing clinical care. Not a methodologist looking at the quality of the data that supports things. Like Neil said, parachutes work. There are some things that are biological consequences across human beings that are beneficial. As a society I think some of it’s on us. We should be strong to support and be clear-cut. Rehabilitation is an essential therapy.

MacIntyre: This is exciting. This is the best discussion we’ve had.

Moy: I want to interject that our discussion in the paper included all of these points. In addition to the lack of data, we were concerned about recommending maintenance pulmonary rehab because we were worried that with the limited resources for pulmonary rehab in general that such a recommendation would draw resources away from core programs.

MacIntyre: It’s one thing to say you don’t have a lot of evidence and you can’t give strong recommendations. I think Jerry [Criner] and I are both saying what really bothers us is the statement that usual care alone is okay.

Moy: But usual care in pulmonary rehab includes having a long-term home exercise plan for patients when they are discharged. Usual care means we are recommending to them to do X, Y, and Z exercise. It’s not go home and sit on your sofa.

MacIntyre: That’s not what the regulators are going to say.

Rochester: I just want to also qualify that every other aspect of the new ATS clinical practice guideline on PR strongly recommends pulmonary rehab for stable COPD, for COPD post hospitalization, and for those with interstitial lung disease. There’s a conditional recommendation based on evidence quality also in support for pulmonary hypertension, and there’s also a recommendation for telerehabilitation. The lukewarmly worded statement only pertains to maintenance pulmonary rehab. So don’t despair the guideline does come out and strongly recommend pulmonary rehabilitation.

MacIntyre: The problem is you’ve given the skeptics a sound bite they’re going to grab.

Moy: I also wanted to respond to Jerry’s point. In one of our clinical trials, the RCT⁵ looking at the web-based physical activity platform, we did show an improvement in HRQOL measured by the St George Respiratory Questionnaire. But we did not find that in the other 2 trials.^6,7 It really highlights that if we improve physical activity that does not necessarily translate into improvements in HRQOL. Studies need to consider and assess concepts such as fear of movement, fear of leaving one’s house, worry that you’re going to run out of oxygen, that you’re not going to have your inhaler with you. Just because you increase daily step counts in a trial or in the clinic doesn’t mean that all increases in physical activity translate into improvements in HRQOL.

Yohannes: I was also part of the guideline committee. Yes, we had very intensive discussion on this point as well. So, it wasn’t just lightly recommended. I know we didn’t have the evidence to support it, and like you, yes, we are concerned about it, but that’s what we had. But just to add to that. We conducted a comprehensive pulmonary rehabilitation in patients with COPD and observed a change in behavior in their exercise program habits whereby as part of the PR that individual has to exercise daily at least walk for 30 min. That had some significant benefits. Even if they didn’t have any maintenance, but we just followed up for 2 years, and we did show some benefit in quality of life and also a little bit of reducing anxiety symptoms. There is some benefit. That’s not part of the recommendation, but certainly there is some value as well.

Mosher: First I want to say, Marilyn, this was a really great presentation. Thank you, and I’ve really enjoyed the discussion by everyone. I can appreciate it seems as though you all were very handcuffed in what you were able to say in the recommendation. But I do want to echo it does feel to me like maybe we’re over complicating it? For example, we have good evidence outside of COPD that recommends older adults should get 150 min of moderate intensity exercise each week. I don’t know why we couldn’t recommend something to that extent, even though maybe that hasn’t been clearly shown in maintenance COPD. Just being practical about it.

Moy: Those minutes aren’t recommended in GOLD. There are no data from RCTs that show benefits of different thresholds of exercise time in COPD. It all makes sense, but they’re elderly patients. There’s also the recommendation for 10,000 steps/d, but there again we have to be very data driven for any guidelines in the COPD population. These values fall under the usual care recommendations. We send our patients home after they are discharged from pulmonary rehab with recommendations to walk every day. From the minute they start our program on day 1, and I’m sure this happens in all your programs, we’re trying to educate patients on how they are going to maintain exercise after they complete our program because they’re going to be with us for a very short amount of time. We ask if they have equipment at home, who else is in the home environment, what other exercises they enjoy doing that’s feasible such as yoga, tai chi, or swimming. This is all part of our initial and exit evaluations, as part of usual care in promoting exercise.

MacIntyre: What is your usual care is not the world’s usual care though. I’m sorry, let’s get the voice from this other side of the pond. What do you Europeans think of all this?

Evans: I suppose I have a different view, and please don’t shoot me down too much, but I think we do have lots of evidence, and the evidence is mostly negative for maintenance interventions with little around cost-effectiveness.^8-12 I thought your talk was really thoughtful, Marilyn, all the different techniques you’re utilizing; and I think, Linda, the evidence really speaks to what you’re saying. If somebody is engaged with rehab, changes their behavior, if you then randomize them to usual care, they just carry on doing what they’ve already learned (ie, they will be doing their own maintenance exercise regardless). If you have somebody that’s not really engaged, sort of went along with it, but didn’t really change their behavior, if you randomize them into a maintenance rehab arm, they end up not doing it. And I think that’s why we end up with these very negative results. The interventions don’t seem to alter the ongoing behavior change needed. I don’t think anyone here is saying we don’t want to promote that behavior change, but I do think at the moment we don’t know how to deliver it successfully. Marilyn, I think the key thing on your slide is the picture back in the office with the clinician. What we were trying to say yesterday about the trajectory, we don’t work enough with what’s actually happening to that person over time. I hear all the time in the advanced COPD clinic, “I loved rehabilitation, but oh my goodness, I couldn’t do that now. I’ve been in hospital twice” or “I’ve had several exacerbations, I’ve no idea what to do now with my exercises.” And we don’t reset. We don’t look at that exercise prescription alongside all the other things in the clinic. So, I think that’s a really positive thing in your next study, if that’s what you’re looking at, more integration across the trajectory of a person’s illness. I think we should absolutely investigate different techniques of how to achieve successful maintenance, but as a clinician in a socially funded health system, I don’t recommend the therapists deliver maintenance programs. We encourage ongoing exercise, patients to form their own groups, and embedding physical activity/exercise into voluntary groups such as “Breathe Easy.” We focus on delivering the main program to improve access for all those that are eligible, and we do repeat programs. There are real tensions between what we’d love to be able to do, what the evidence shows, and then what we need to do to sort it out.

Mosher: Your data show step counts are very important and when you have more steps there are more benefits. But I’m wondering are we selling our patients short? Are we conflating steps with exercise?

Moy: Oh no, not at all. I presented the graph initially showing that the association between daily step count and 6MWT distance, as a measure of exercise capacity, is quite weak. We’re not saying they’re the same thing, but I’m saying that based upon the data that we have about physical activity it’s worth aiming for that outcome in its own right because it’s an important outcome that gives us information that’s different from the 6MWT. In many ways, it’s more patient-centered. A patient knows what it means to walk extra steps to get to the grocery store. Getting an extra meter on our 6MWT means nothing to them. I would say that they are very different metrics, and I hope I was transparent about that and not conflating steps with exercise.

Mosher: No, I think you were. I guess the point I was trying to make is you know while it’s great for patients to walk 5,000 or 10,000 steps a day wouldn’t it be potentially more beneficial to say that’s great, you’re doing that in the background, but what I really want you to do is to exercise at a moderate intensity for 10 min. That’s very different than saying throughout the course of the day walk 7,000 steps.

Moy: In terms of thinking about what this platform is used for and how it has been studied to date, it’s really a better-than-nothing alternative to conventional pulmonary rehab. If it’s for maintenance pulmonary rehab, then it may look a little different. I want to end by saying that there is plenty of evidence for efficacy of this virtual platform. We’re adding tai chi as a virtual modality to those later time points at 3–6 months to try to increase engagement with our intervention based upon the evidence that we currently have for tai chi alone. It will take an act of Congress to implement this intervention in the VA. Literally an act of Congress. I need a senator to back this up. If you have any other thoughts about how to implement this in real life and bring it to the frontlines, to our patients, I would love to hear your thoughts.

MacIntyre: This has been a lot of fun and quite informative. Rachael [Evans], I’m always reminded when somebody makes comments like you did that absence of proof is not proof of absence. This is an area that clearly needs a lot of our attention.