Comments
Abstract
Rationale: Despite the benefits of pulmonary rehabilitation in chronic obstructive pulmonary disease (COPD), many patients do not access or complete pulmonary rehabilitation, and long-term maintenance of exercise is difficult.
Objectives: To compare long-term telerehabilitation or unsupervised treadmill training at home with standard care.
Methods: In an international randomized controlled trial, patients with COPD were assigned to three groups (telerehabilitation, unsupervised training, or control) and followed up for 2 years. Telerehabilitation consisted of individualized treadmill training at home supervised by a physiotherapist and self-management. The unsupervised training group performed unsupervised treadmill exercise at home. The control group received standard care. The primary outcome was the combined number of hospitalizations and emergency department presentations. Secondary outcomes included time free from the first event; exercise capacity; dyspnea; health status; quality of life; anxiety; depression; self-efficacy; and subjective impression of change.
Measurements and Main Results: A total of 120 participants were randomized. The incidence rate of hospitalizations and emergency department presentations was lower in telerehabilitation (1.18 events per person-year; 95% confidence interval [CI], 0.94–1.46) and unsupervised training group (1.14; 95% CI, 0.92–1.41) than in the control group (1.88; 95% CI, 1.58–2.21; P < 0.001 compared with intervention groups). Telerehabilitation and unsupervised training groups experienced better health status for 1 year. Intervention participants reached and maintained clinically significant improvements in exercise capacity.
Conclusions: Long-term telerehabilitation and unsupervised training at home in COPD are both successful in reducing hospital readmissions and can broaden the availability of pulmonary rehabilitation and maintenance strategies.
Despite the evidence of the benefits of pulmonary rehabilitation, many patients with chronic obstructive pulmonary disease (COPD) do not access or complete pulmonary rehabilitation, and long-term maintenance of exercise is difficult. Efforts to reduce hospital readmissions in COPD must be made to decrease the societal burden and improve patient outcomes. Long-term telerehabilitation and unsupervised training at home represent promising alternatives to traditional pulmonary rehabilitation and maintenance strategies.
Long-term unsupervised exercise training at home is an effective treatment strategy that can reduce hospital readmissions for patients with COPD, similar to the effect of a supervised telerehabilitation strategy. These interventions have the potential to improve uptake and access to pulmonary rehabilitation and support long-term exercise maintenance strategies. Unsupervised training at home could be offered to patients with COPD who do not access pulmonary rehabilitation or maintenance programs. Telerehabilitation may be useful for patients who are unsuitable for unsupervised training and need a closer follow-up.
Chronic obstructive pulmonary disease (COPD) imposes a substantial burden on patients, healthcare providers, and society (1, 2). Patients with COPD experience frequent exacerbations, which, in the most severe cases, may result in hospital admissions (3, 4). COPD exacerbations are commonly characterized by acute worsening of symptoms, including dyspnea, cough, sputum production, and sputum purulence (5). Patients with COPD also experience impaired exercise capacity, difficulty with activities of daily living, poor quality of life (6), anxiety, and depression (7). Chronic respiratory diseases contribute 7% to the global burden of disease, with COPD accounting for 56% of the costs of chronic respiratory diseases (8). Hospitalizations alone account for up to 70% of all COPD-related costs (9). Moreover, discharge from the hospital after a severe exacerbation is associated with an increased risk of readmission (10). Efforts to reduce recurrent exacerbations and hospitalizations must be made to improve patient outcomes and reduce societal burden (11).
Pulmonary rehabilitation (PR) is widely recognized as a core component of the management of COPD (12, 13). PR aims to improve the physiological and psychological condition of participants through exercise training accompanied by education and behavior change (14). PR leads to clinically important gains in exercise and functional capacity, dyspnea, health status, and health-related quality of life (12–14). PR has also been proven to be effective in reducing healthcare usage (15).
Despite the evidence of the benefits of PR for patients with COPD, there are several barriers to PR participation, including patient travel (16) and a severe shortage of programs because of a lack of knowledge, underfunding, and poor institutional support (17). The majority of PR programs are located in urban areas, thus limiting access for rural patients (18). Referral rates to PR after an exacerbation are low (17). Only 1.5% of patients are reported to initiate PR within 90 days of discharge (15), and fewer than 10% of patients complete PR programs (19). Sustaining long-term adherence to exercise training is difficult because of disease progression with intervening exacerbations, variation in day-to-day conditions, and transportation problems (13). In the absence of any maintenance program, the gains from PR typically wane over 6–12 months (14, 20). Maintenance rehabilitation consists of ongoing supervised exercise at a lower frequency than PR programs (21). However, the optimum maintenance intervention and supervision frequency are still unclear, and interventions have had varying impacts (20, 22).
Telerehabilitation, defined as the use of information and communication technologies to provide rehabilitation services remotely to people in their homes (23), has the potential to improve uptake and access to PR (24) and support long-term maintenance strategies (13, 25). A recent systematic review suggests that telerehabilitation achieves outcomes similar to those of traditional center-based PR (26). Patients with COPD have a lower likelihood of acute exacerbations and hospitalizations when undertaking maintenance telerehabilitation compared with no rehabilitation (27, 28). The duration of intervention for studies of maintenance telerehabilitation ranged from 4 months (28) to 12 months (27, 29). Few studies followed people up after the intervention was finished, and no intervention lasted longer than 1 year, making it difficult to draw conclusions about the long-term effectiveness. Unsupervised home-based structured exercise represents another promising strategy to deliver maintenance rehabilitation with minimal resources (24). Although unsupervised exercise interventions have been proven to be effective at improving health-related quality of life and exercise capacity in the medium term (24, 30), there is insufficient evidence for its provision to reduce hospital admissions and improve other outcomes, as well as long-term maintenance of benefits.
The aim of the present study was to compare the long-term telerehabilitation of patients with COPD or unsupervised exercise training at home with standard care with respect to the combined number of hospitalizations and emergency department (ED) presentations occurring during 2 years as well as other secondary outcomes (31).
The iTrain study was an international multicenter randomized controlled trial (RCT) conducted in three countries (Norway, Australia, and Denmark), in which 120 participants with COPD were randomly assigned to three groups (telerehabilitation, unsupervised training, or control) in a 1:1:1 ratio. Each participant was followed up for 2 years since the day of inclusion in the study, and the interventions were delivered for the entire period of follow-up. Web-based computerized block randomization was performed, with randomization stratified by center and disease severity (FEV1 under 50% vs. FEV1 of at least 50%). The RCT received approval from the Regional Committee for Medical and Health Research Ethics in Norway (2014/676/REK nord), the Alfred Hospital Human Research Ethics Committee (289/14), and the North Denmark Region Committee on Health Research Ethics (N-20140038). The complete study protocol, including full details of the interventions, has been previously published (31) and was prospectively registered (ClinicalTrials.gov: NCT02258646).
Eligible patients had: 1) a diagnosis of COPD on the basis of an FEV1/FVC ratio under 0.70; 2) moderate, severe, or very severe airflow limitation, with FEV1% predicted under 80%; 3) at least one COPD-related hospitalization or COPD-related ED presentation in the 12 months before enrolment; 4) age between 40 and 80 years; and 5) capacity to provide signed written informed consent.
Participants were excluded if they had at least one of the following criteria: 1) attendance at a rehabilitation program in the 6 months before enrolment; 2) participation in another clinical study that might have had an impact on the primary outcome; 3) physically incapable of performing the study procedures; 4) presence of comorbidities which might prevent participants from safely exercising at home; and 5) home environment not suitable for installation and use of rehabilitation and monitoring equipment (e.g., limited space for the treadmill and internet connection not good enough).
Participants were recruited by hospital facilities with a pulmonary medicine department treating patients with COPD. Supervision in the telerehabilitation intervention was provided by physiotherapists specialized in PR.
Participants in both intervention groups underwent a supervised in-person training session on the treadmill with an experienced physiotherapist to ensure safety.
Participants in the telerehabilitation group were offered an integrated intervention consisting of exercise training at home, telemonitoring, and self-management. Each participant received an individualized training program of regular exercise on a treadmill and strength training exercises according to guidelines (14). Depending on the participant’s exercise tolerance and the clinician’s preference, a program of continuous training (moderate intensity, Borg scale [32] ratings up to four) or interval training (1–4 minute intervals, high intensity, Borg scale ratings up to six) was assigned, with sessions lasting for at least 30 minutes (see the online supplement). The frequency prescribed was 3–5 times/wk for continuous training and 3 times/wk for interval training (31). Progression was made according to a standardized protocol (see the online supplement). The equipment included a treadmill, a pulse oximeter, a tablet computer, and a holder for the tablet computer (Figure E1 in the online supplement). The equipment was provided and delivered by the research team. A customized website was used by participants for self-management. They could access the individual training program (Figure E2), fill in a daily diary (Figure E3) and a training diary (Figure E4), review historical data, exchange electronic messages, schedule videoconferencing sessions, and facilitate individual goal setting and goal attainment. The information sent through the website was monitored and interpreted weekly by a physiotherapist. Participants had scheduled exercise sessions supervised by a physiotherapist via videoconferencing, which followed a standardized protocol (see the online supplement). After each supervised session, the physiotherapist could adjust the program if necessary and was also informed if a patient had been hospitalized. Telerehabilitation was delivered with two levels of supervision: 1) an intensive 8-week program (one videoconferencing session per week in the first 8 weeks, plus once-weekly for 1 month after any readmission, supplemented by unsupervised sessions); and 2) a lower-intensity maintenance program (one videoconferencing session per month commencing after the initial 8-week intensive program, supplemented by unsupervised sessions). Additional contacts with the physiotherapist could be arranged if necessary.
Participants in the unsupervised training group were provided with a treadmill only to perform unsupervised exercise at home. They also received an exercise booklet, a paper exercise diary to record their training sessions, and an individualized training program (see the online supplement) as prescribed to the participants in the telerehabilitation group, but without regular review or progression of the program. Participants were advised not to exercise if they felt unwell (more coughing, wheezing, breathless, or having more sputum than usual), had less energy, or had a loss of appetite. Participants in the control group were offered standard care.
Assessments were performed by appropriately trained study personnel blinded to group allocation. At baseline, participants were asked to perform spirometry, the 6-minute-walk test (33), and complete the study questionnaires. Measures were repeated at 6-month, 1-year, and 2-year follow-ups. Data on hospitalizations and ED presentations were collected from health records or registries after the end of the trial. Data on deaths, transplantations, dropouts, and adverse events were collected systematically during the trial and at each follow-up. Participants also received information on self-management of exacerbations (see the online supplement).
The primary outcome was the combined number of hospitalizations and ED presentations occurring in the three groups during the entire 2-year duration of the trial. These data were collected from health records (Australia) and registries (Denmark and Norway) at the end of the trial. Secondary outcomes included: hospitalizations and ED presentations (analyzed separately), time free from the first event, functional exercise capacity measured with the 6-minute-walk distance (6MWD) (33), dyspnea measured with the mMRC (modified Medical Research Council) dyspnea scale, health status measured with the COPD assessment test (CAT) (34), health-related quality of life measured with the EQ-5D (EuroQol 5 dimensions) questionnaire (35), anxiety and depression measured with HADS (Hospital Anxiety and Depression Scale) (36), self-efficacy measured with GSES (Generalized Self-Efficacy Scale) (37), and subjective impression of overall change measured with PGIC (Patient Global Impression of Change) scale (38). Results on the remaining secondary outcomes, including degrees of physical activity, cost-effectiveness, and experiences in telerehabilitation, will be reported separately.
The sample size requirements were intended to provide adequate power for the analysis of the primary outcome. From studies with participants with similar characteristics, we estimated an incidence density used as a null hypothesis of two events per person-year and a 40% relative reduction in the primary outcome (31). Allowing for a 20% dropout, we calculated that a sample size of 40 participants per group would allow a power of 95% to detect an incidence rate ratio of 0.60, with a type-I error (α) of 0.05.
Descriptive statistics at baseline are reported as mean and standard deviation for continuous variables and count and percentage for categorical variables. An intention-to-treat analysis was performed on all randomized subjects. The primary outcome and related secondary outcomes were measured with the incidence density, defined as the number of events in a group divided by the total person-time accumulated during the study in that group. Differences between study groups were tested by the comparison of incidence rates. A two-sided test and a significance level of α = 0.05 were used. All events from the day after randomization to participant exit or death were included. Linear mixed models were used to measure changes from baseline to all assessment time points in 6MWD, mMRC scale, CAT score, EQ-5D scores, and GSES. The minimal important difference (MID) used for the 6MWD was 30 meters (33). Baseline variables with differences among groups were also added as covariates to the comparison of incidence rates and mixed models. Kaplan-Meier curves and the log-rank test were used to determine if there were differences in the survival distribution of the time free from the first event for the telerehabilitation, unsupervised training, and control groups. The Wald test computed using binary logistic regression was used for the HADS (score less than eight = no case; a score of at least eight = case). The chi-square test was used for the PGIC. Differences in mortality rates between study groups were tested by the comparison of incidence rates. A P value less than 0.05 was considered significant for all tests. Statistical analyses were performed using IBM SPSS Statistics (Version 25; IBM Corp).
Between October 2014 and December 2016, 502 individuals were assessed for eligibility, and 120 (24%) were recruited and randomized (Figure 1). At the end of the study, data were available for the primary outcome and related secondary outcomes for 115 (96%) participants, comprising 37 (93%) in the telerehabilitation group, 40 (100%) in the unsupervised training group, and 38 (95%) in the control group. Details of the number of participants with complete data for each outcome at all assessment time points are reported in Table E1.
Figure 1. CONSORT (Consolidated Standards of Reporting Trials) diagram for study flow. COPD = chronic obstructive pulmonary disease.
[More] [Minimize]Demographic and clinical characteristics of the study participants were similar between study groups at baseline (Table 1). There were slightly more participants on long-term oxygen therapy (LTOT) in the telerehabilitation group (30%) than in the unsupervised training group (22.5%) and control group (15%), and more current smokers in the control group (37.5%) than in the telerehabilitation group (20%) and unsupervised training group (27.5%).
| Telerehabilitation | Unsupervised Training | Control | |
|---|---|---|---|
| Participants, n | 40 | 40 | 40 |
| Age (yr), mean ± SD | 64.9 ± 7.1 | 64.0 ± 7.7 | 63.5 ± 8.0 |
| Male, n (%) | 23 (57.5) | 20 (50) | 23 (57.5) |
| Stratification | |||
| FEV1 (% predicted) < 50%, n (%) | 28 (70) | 28 (70) | 28 (70) |
| FEV1 (% predicted) ⩾ 50%, n (%) | 12 (30) | 12 (30) | 12 (30) |
| COPD diagnosis (yr), mean ± SD | 8 ± 7 | 10 ± 8 | 7 ± 7 |
| LTOT, n (%) | 12 (30) | 9 (22.5) | 6 (15) |
| mMRC scale, n (%) | |||
| 0 | 0 (0) | 2 (5) | 4 (10) |
| 1 | 14 (35) | 12 (30) | 12 (30) |
| 2 | 13 (32.5) | 13 (32.5) | 10 (25) |
| 3 | 10 (25) | 12 (30) | 12 (30) |
| 4 | 3 (7.5) | 1 (2.5) | 2 (5) |
| BODE index (points), n (%) | |||
| 0–2 | 11 (27.5) | 11 (27.5) | 11 (27.5) |
| 3–4 | 13 (32.5) | 17 (42.5) | 16 (40) |
| 5–6 | 12 (30) | 9 (22.5) | 7 (17.5) |
| 7–10 | 4 (10) | 3 (7.5) | 6 (15) |
| Smoking history | |||
| Current smoker, n (%) | 8 (20) | 11 (27.5) | 15 (37.5) |
| Ex-smoker, n (%) | 31 (77.5) | 28 (70) | 24 (60) |
| Never smoked, n (%) | 1 (2.5) | 1 (2.5) | 1 (2.5) |
| Pack-year, mean ± SD | 31 ± 17 | 38 ± 21 | 35 ± 16 |
| FEV1 (L), mean ± SD | 1.18 ± 0.61 | 1.21 ± 0.52 | 1.14 ± 0.52 |
| FEV1 (% predicted), mean ± SD | 40.4 ± 16.5 | 44.6 ± 17.1 | 40.3 ± 16.1 |
| FVC (L), mean ± SD | 2.48 ± 0.89 | 2.60 ± 0.75 | 2.63 ± 0.91 |
| FVC (% predicted), mean ± SD | 68.9 ± 19.1 | 75.4 ± 18.2 | 74.4 ± 23.7 |
| FEV1/FVC (%), mean ± SD | 50.1 ± 14.8 | 49.5 ± 12.8 | 46.7 ± 14.9 |
| Comorbidities, n | 2.9 ± 1.6 | 2.7 ± 1.7 | 2.6 ± 2.1 |
| BMI (kg/m2), mean ± SD | 27 ± 6 | 28 ± 7 | 26 ± 6 |
| Living arrangements, n (%) | |||
| Alone | 19 (47.5) | 20 (50) | 17 (42.5) |
| With spouse | 15 (37.5) | 16 (40) | 15 (37.5) |
| With family | 5 (12.5) | 4 (10) | 7 (17.5) |
| With friends | 1 (2.5) | 0 (0) | 1 (2.5) |
| Supported accommodation | 0 (0) | 0 (0) | 0 (0) |
| Social status | |||
| Working, n (%) | 5 (12.5) | 8 (20) | 11 (27.5) |
| Retired, n (%) | 35 (87.5) | 32 (80) | 29 (72.5) |
| Distance to outpatient clinic (km), mean ± SD | 37 ± 67 | 37 ± 59 | 21 ± 35 |
| Digital competence, n (%) | |||
| Daily user or nearly every day | 25 (62.5) | 26 (65) | 28 (70) |
| At least once a week, but not every day | 8 (20) | 9 (22.5) | 5 (12.5) |
| No experience | 7 (17.5) | 5 (12.5) | 7 (17.5) |
No treadmill-related injuries were reported during the study period (Table E2). Adverse events included problems with the study equipment, most frequently the incline function on the treadmill, and medical problems that prevented participants from exercising (e.g., cancer, surgery, and arthritis).
For the assessment of the incidence rate of hospitalizations and ED presentations, there were 71.05 person-years in the telerehabilitation group, 76.93 person-years in the unsupervised training group, and 74.59 person-years in the control group (Table 2). By the end of the study, a total of 312 events (combined number of hospitalizations and ED presentations) occurred in the study population. Specifically, 84 events were reported in the telerehabilitation group, 88 in the unsupervised training group, and 140 in the control group. The incidence rate for the primary outcome was lower in both the telerehabilitation group (1.18 events per person-year; 95% confidence interval [CI], 0.94–1.46; P = 0.0007) and the unsupervised training group (1.14 events per person-year; 95% CI, 0.92–1.41; P = 0.0002) compared with the control group (1.88 events per person-year; 95% CI, 1.58–2.21). Similarly, the difference in the incidence rate for hospitalizations and ED presentations analyzed separately was significantly lower in both telerehabilitation and unsupervised training groups compared with the control group (Table 2). Adding smoking status and LTOT as covariates to the model did not change the results. There was a larger proportion of participants without hospital presentations (consisting of hospitalizations and ED presentations) occurring during the study period in the telerehabilitation (40.6%) and unsupervised training group (45.0%) compared with the control group (28.9%) (Table 3 and Table E3). In addition, the control group has a higher proportion of participants with recurrent (at least two) hospital presentations (55.3%) compared with telerehabilitation (35.1%) and unsupervised training group (35.0%).
| Outcome Measure | Telerehabilitation | Unsupervised Training | Control |
|---|---|---|---|
| Hospitalizations and ED presentations* (combined), n | 84 | 88 | 140 |
| Person-years, n | 71.05 | 76.93 | 74.59 |
| Incidence rate (per person-year) (95% CI) | 1.18 (0.94–1.46) | 1.14 (0.92–1.41) | 1.88 (1.58–2.21) |
| Incidence rate ratio (95% CI) | 0.63 (0.48–0.83) | 0.61 (0.46–0.79) | 1 [reference] |
| P value† | 0.0008 | 0.0002 | — |
| Hospitalizations, n | 68 | 74 | 126 |
| Person-years, n | 71.05 | 76.93 | 74.59 |
| Incidence rate (per person-year) (95% CI) | 0.96 (0.74–1.21) | 0.96 (0.76–1.21) | 1.69 (1.41–2.01) |
| Incidence rate ratio (95% CI) | 0.57 (0.42–0.76) | 0.57 (0.43–0.76) | 1 [reference] |
| P value† | 0.0002 | 0.0001 | — |
| ED presentations,‡ n | 71 | 75 | 118 |
| Person-years, n | 71.05 | 76.93 | 74.59 |
| Incidence rate (per person-year) (95% CI) | 1.00 (0.78–1.26) | 0.97 (0.77–1.22) | 1.58 (1.31–1.89) |
| Incidence rate ratio (95% CI) | 0.63 (0.47–0.85) | 0.61 (0.46–0.82) | 1 [reference] |
| P value† | 0.0022 | 0.0009 | — |
| Hospitalizations and ED Presentations | Telerehabilitation (%) | Unsupervised Training (%) | Control (%) |
|---|---|---|---|
| 0 | 40.6 | 45.0 | 28.9 |
| 1 | 24.3 | 20.0 | 15.8 |
| ⩾2 (recurrent hospital presentations) | 35.1 | 35.0 | 55.3 |
| 2–5 | 21.6 | 22.5 | 36.9 |
| 6–10 | 10.8 | 10 | 7.9 |
| ⩾10 | 2.7 | 2.5 | 10.5 |
The survival distributions of the time-to-first hospitalization or ED presentation in the three groups were not significantly different [χ2(2) = 2.345; P = 0.310] (Figure 2A). Similar results were obtained for the time-to-first hospitalization (χ2(2) = 2.946; P = 0.229) (Figure 2B) and time-to-first ED presentation [χ2(2) = 2.545; P = 0.280] (Figure 2C).
Figure 2. Time-to-first event: (A) hospitalization or emergency department presentation, (B) hospitalization, and (C) emergency department presentation. ED = emergency department.
[More] [Minimize]The telerehabilitation group experienced statistically significant changes at 6 months in CAT score (P = 0.037) and mMRC scale (P = 0.037) compared with the control group (Table 4). The gains in health status and dyspnea were not maintained after 2 years. On average, participants had improvements in 6MWD that exceeded the MID at all time points. In contrast, participants in the control group experienced a decline in the 6MWD. A considerably higher proportion of participants in the telerehabilitation group (53.1%) experienced a significant, favorable change in the PGIC at 6 months compared with the unsupervised training group (24.2%) and the control group (13.3%, P = 0.001). No differences between groups were detected for self-efficacy, anxiety, and depression. Adding smoking status and LTOT as covariates to the model did not change the results.
| Telerehabilitation | Unsupervised Training | Control | |||
|---|---|---|---|---|---|
| Outcome Measure | Mean ± SD or n (%) | P Value | Mean ± SD or n (%) | P Value | Mean ± SD or n (%) |
| 6MWD (m) | — | 0.380* | — | 0.065* | — |
| Baseline | 367 ± 125 | — | 367 ± 111 | — | 384 ± 111 |
| 6 mo | 420 ± 126 | 0.126 | 406 ± 114 | 0.332 | 389 ± 101 |
| 1 yr | 415 ± 146 | 0.209 | 431 ± 117 | 0.057 | 374 ± 116 |
| 2 yr | 400 ± 142 | 0.235 | 460 ± 126 | 0.009 | 357 ± 102 |
| CAT (total score) | — | 0.189* | — | 0.023* | — |
| Baseline | 19.6 ± 6.2 | — | 20.1 ± 6.3 | — | 19.7 ± 8.1 |
| 6 mo | 18.2 ± 6.9 | 0.037 | 15.2 ± 7.6 | 0.002 | 20.8 ± 7.2 |
| 1 yr | 18.7 ± 6.9 | 0.086 | 17.5 ± 7.6 | 0.047 | 20.8 ± 7.0 |
| 2 yr | 19.0 ± 7.1 | 0.373 | 18.4 ± 8.6 | 0.272 | 19.8 ± 6.8 |
| mMRC (score) | — | 0.131* | — | 0.033* | — |
| Baseline | 2.1 ± 1.0 | — | 1.9 ± 1.0 | — | 1.9 ± 1.1 |
| 6 mo | 1.7 ± 1.2 | 0.037 | 1.5 ± 1.0 | 0.027 | 2.2 ± 0.8 |
| 1 yr | 1.8 ± 1.2 | 0.089 | 1.5 ± 1.0 | 0.012 | 2.2 ± 1.1 |
| 2 yr | 1.9 ± 1.2 | 0.105 | 1.5 ± 1.1 | 0.008 | 2.3 ± 1.1 |
| EQ-5D (utility index) | — | 0.280* | — | 0.119* | — |
| Baseline | 0.739 ± 0.110 | — | 0.744 ± 0.155 | — | 0.759 ± 0.180 |
| 6 mo | 0.728 ± 0.154 | 0.089 | 0.768 ± 0.184 | 0.036 | 0.685 ± 0.190 |
| 1 yr | 0.671 ± 0.215 | 0.903 | 0.747 ± 0.171 | 0.373 | 0.674 ± 0.236 |
| 2 yr | 0.725 ± 0.153 | 0.259 | 0.686 ± 0.280 | 0.740 | 0.673 ± 0.228 |
| EQ-5D (EQ-VAS) | — | 0.654* | — | 0.208* | — |
| Baseline | 51.9 ± 21.0 | — | 52.0 ± 17.7 | — | 52.4 ± 19.6 |
| 6 mo | 58.7 ± 16.4 | 0.299 | 55.4 ± 21.6 | 0.735 | 55.1 ± 16.8 |
| 1 yr | 56.3 ± 18.9 | 0.653 | 58.0 ± 19.1 | 0.381 | 53.7 ± 19.5 |
| 2 yr | 54.9 ± 21.4 | 0.295 | 58.4 ± 21.2 | 0.040 | 50.0 ± 20.8 |
| GSES (total score) | — | 0.70* | — | 0.160* | — |
| Baseline | 30.7 ± 5.4 | — | 31.4 ± 5.3 | — | 32.0 ± 5.8 |
| 6 mo | 30.9 ± 5.4 | 0.165 | 31.1 ± 4.6 | 0.263 | 30.3 ± 4.7 |
| 1 yr | 30.5 ± 5.5 | 0.462 | 31.5 ± 4.7 | 0.215 | 30.3 ± 7.8 |
| 2 yr | 30.4 ± 5.6 | 0.311 | 30.6 ± 5.4 | 0.576 | 32.7 ± 5.6 |
| HADS, participants free from anxiety | |||||
| Baseline | 30/40 (75.0) | — | 31/40 (77.5) | — | 28/40 (70.0) |
| 6 mo | 26/35 (74.3) | 0.599 | 29/35 (82.9) | 0.970 | 25/32 (78.1) |
| 1 yr | 22/32 (68.8) | 0.829 | 24/30 (80.0) | 0.351 | 18/30 (60.0) |
| 2 yr | 20/27 (74.1) | 0.318 | 23/31 (74.2) | 0.290 | 25/30 (83.3) |
| HADS, participants free from depression | |||||
| Baseline | 35/40 (87.5) | — | 33/40 (82.5) | — | 33/40 (82.5) |
| 6 mo | 31/35 (88.6) | 0.110 | 27/35 (77.1) | 0.362 | 22/32 (68.8) |
| 1 yr | 23/32 (71.9) | 0.208 | 24/30 (80.0) | 0.945 | 25/30 (83.3) |
| 2 yr | 22/27 (81.5) | 0.521 | 22/31 (71.0) | 0.201 | 26/30 (86.7) |
| PGIC, participants with a score at 6 mo | |||||
| PGIC < 5 | 15 (46.9) | — | 25 (75.8) | — | 26 (86.7) |
| PGIC ⩾ 5 | 17 (53.1) | 0.001 | 8 (24.2) | 0.271 | 4 (13.3) |
The unsupervised training group also experienced improved CAT score (P = 0.002) and mMRC scale (P = 0.027) at 6 months compared with the control group (Table 4). Degrees of dyspnea were maintained for 2 years, whereas the gains in health status were maintained for 1 year. Participants had improvements in 6MWD that exceeded the MID for the entire 2-year period. However, there was only a statistically significant difference between the unsupervised training group and the control group at 2 years. Participants in the control group experienced an earlier decrease in their health-related quality of life at 6 months (EQ-5D utility index) compared with the unsupervised training group (P = 0.036), with similar findings for EQ-VAS at 2 years (P = 0.040). No differences between groups were detected for self-efficacy, anxiety, and depression.
The mortality rate at the end of the trial was 7.5% (3/40 participants), 10% (4/40 participants), and 5% (2/40 participants) in the telerehabilitation, unsupervised training, and control groups, respectively, with no difference between groups.
To our knowledge, this was the first trial delivering a 2-year telerehabilitation intervention to patients with COPD. The iTrain study demonstrated that both long-term telerehabilitation and unsupervised training at home were successful in reducing the number of hospital readmissions for patients with COPD. Telerehabilitation and unsupervised training groups experienced better health status for 1 year. Intervention participants reached and maintained clinically significant improvements in exercise capacity.
A systematic review suggested a lower likelihood of acute exacerbations and hospitalizations for maintenance telerehabilitation compared with no rehabilitation (26). The evidence, however, was limited to two studies (27, 28), neither of which had an intervention lasting longer than 12 months. The iTrain study was designed assuming an incidence rate of two events per person-year and expecting a 40% relative reduction in the primary outcome. Results showed a 37.3% reduction (−0.70 events per person-year) in the telerehabilitation group and a 39.4% reduction (−0.74 events per person-year) in the unsupervised training group compared with the control group (1.88 events per person-year). There was a larger proportion of participants without hospital presentations in the intervention groups. In addition, the control group had a higher proportion of participants with recurrent (at least two) hospital presentations. Despite no significant difference in the time-to-first event, both interventions appear to be better than the control. Moreover, although incidence rates were very similar among intervention groups, participants in the telerehabilitation group presented to the ED sooner than those in the unsupervised training group (Figure 2C). One possible reason is that they were supervised regularly by a health professional, suggesting that telerehabilitation might allow earlier detection of problems.
It was expected that participants in the telerehabilitation group would gain additional benefits because of the remote supervision by a physiotherapist (39). The findings from this trial indicate that both interventions seem to work well and produce beneficial results compared with standard care. These results might be explained by the characteristics and preferences of the participants. Positive attitudes toward supervised and unsupervised maintenance programs have been reported (40). However, although some patients need ongoing support for exercise participation, others can maintain the gains of PR regardless of intervention (41). As such, unsupervised training at home is a simple intervention using minimal resources that could be offered to patients with COPD who do not access PR or maintenance programs. Telerehabilitation is likely to be more expensive, but it may be useful for patients who are unsuitable for unsupervised training because of factors such as disease severity, anxiety, depression, poor social support, or low motivation (13). Remote supervision by a physiotherapist can provide those individuals additional benefits, as confirmed by the higher proportion of participants in the telerehabilitation group who experienced a favorable change in the PGIC. These benefits can, in turn, result in better adherence to exercise. Identifying these patient groups is an important challenge for both clinicians and researchers (41). Future research should focus on adapting PR and maintenance programs to the individual needs of the participants to maximize the benefits while making good use of healthcare resources (42).
A variety of strategies have been used to sustain the clinical gains achieved in traditional center-based PR (14, 20), but outcomes have been inconsistent (41). Maintenance models in COPD are heterogeneous in terms of supervision (supervised or unsupervised), frequency (once weekly to monthly or less frequent supervision) (21), modality (in-person or remote supervision), (22) and self-management education (42). Supervised maintenance exercise can be effective in improving CAT scores at 6–12 months after PR (22). The evidence for maintaining exercise capacity and quality of life is weak (22, 42). Supervised maintenance programs of monthly or less frequent supervision seem to be insufficient to maintain the gains of PR (21). In a multicenter RCT, a weekly maintenance program was proven modestly effective in improving 6MWD and health status for 2 years after completing PR (43). Unsupervised home-based structured exercise can also help maintain 6MWD and quality of life (24). Giving brief advice to continue exercising may have similar benefits to light-touch strategies or more intensive supervised programs, at least in some patients (41). Maintenance telerehabilitation may achieve improvements in CAT score, mMRC scale, as well as exercise capacity compared with no rehabilitation (26). In an earlier study, the 6MWD was better maintained in subjects attending a 12-month maintenance program, but it returned to prerehabilitation levels by 24 months (44). The iTrain study demonstrated that long-term telerehabilitation and unsupervised training at home lead to gains in CAT score and mMRC scale at 6 months, but these were not maintained after 2 years. Moreover, participants in both intervention groups achieved and maintained clinically significant improvements in 6MWD over 2 years. In contrast, participants in the control group experienced a decline, normally attributable to low adherence to exercise, disease progression, and exacerbations (45). There were no changes in the other outcomes. CAT score and 6MWD are more responsive to PR than other patient-centered outcomes (13), and this can explain the results in the two intervention groups. The study, however, was not powered for the secondary outcomes. The lack of changes in HADS might also be explained by the low number of participants with anxiety or depression at baseline. The PRAISE (Pulmonary Rehabilitation Adapted Index of Self-Efficacy) might have been more suitable to measure changes in self-efficacy because of five additional pulmonary rehabilitation-specific questions (46). However, validated translations in Norwegian and Danish were not available.
The results from traditional maintenance programs in COPD are applicable only to individuals who attend and complete PR (41). However, because of very low rates of referral, attendance, and completion, the majority of patients with COPD do not access PR or maintenance programs (15, 17, 19). The iTrain study addressed the unmet needs of those patients by offering easily accessible home-based models. Earlier RCTs showed that home-based primary PR models (8 wk) delivered with minimal resources and little supervision (weekly telephone calls) could produce short-term clinical improvements similar to those of center-based PR (24, 47). The interventions tested in the iTrain study, which combined components of primary and maintenance rehabilitation, not only can reduce the number of hospital readmissions and lead to improvements in health status and exercise capacity but also result in better maintenance of the benefits over the long term.
We successfully conducted a complex RCT with participants recruited from three countries. The interventions were innovative models combining elements of primary and maintenance rehabilitation, and the findings are novel. Although previous studies lasted up to 1 year, making it difficult to draw conclusions about the long-term effectiveness, our study had a unique long-term follow-up of 2 years. The RCT used robust methods, including intention-to-treat analysis, blinding of assessors, sample size requirements, and adherence to CONSORT (Consolidated Standards of Reporting Trials) guidelines. The primary outcome was relevant to both patients and healthcare systems.
Recruitment lasted for 2 years. The technical setup of the interventions was challenging. However, we successfully offered a common website in three languages and the same or very similar equipment. The applicability of our rehabilitation approaches in different health systems and funding models or groups with lower digital literacy remains to be established. Although the presence of at least one hospitalization or ED admission in the previous 12 months was an inclusion criterion, we did not record the time point at which these occurred. Rehabilitation interventions may have larger effects in recently hospitalized patients, so this could have affected the study outcomes. The study was not powered for secondary outcomes. It was not possible to compare the benefits of the interventions with traditional center-based PR or maintenance programs on the basis of the study design, and it was not possible to compare intervention fidelity across groups, as few participants in the unsupervised training group returned their paper-based training diaries. Despite randomization, the number of current smokers in the control group was higher than in the intervention groups, and the number of LTOT in the control group was lower. Controlling for these factors in the analysis of secondary outcomes did not change the pattern of findings, but we cannot exclude an effect of this imbalance in demographic characteristics. Although traditional PR programs have been conducted in groups of 8 to 12 participants (48), our telerehabilitation intervention consisted of individual sessions. Peer support in the form of group-based online exercise sessions (49, 50), both supervised and unsupervised, has the potential to increase motivation, self-efficacy (39), and the ability to exercise in the long term (42).
Long-term telerehabilitation and unsupervised exercise training at home were both successful in reducing the number of hospitalizations and ED presentations for patients with COPD. Telerehabilitation and unsupervised training groups experienced better health status for 1 year. Intervention participants reached and maintained clinically significant improvements in exercise capacity. The delivery of long-term telerehabilitation or unsupervised exercise training at home has the potential to broaden the availability of PR programs and maintenance strategies, especially for those living in remote areas and with no access to center-based exercise programs.
The authors thank the University Hospital of North Norway (Tromsø and Harstad, Norway), and especially Audhild Hjalmarsen, M.D., the Finnmark Hospital (Kirkenes, Norway), LHL-klinikkene (Skibotn, Norway), the Esbjerg Healthcare Center (Esbjerg, Denmark), the Pulmonary Ward at South West Hospital (Esbjerg, Denmark), and Alfred Health (Melbourne, Australia) for recruiting participants, performing the clinical assessments, and supervising the participants.
| 1. | Iheanacho I, Zhang S, King D, Rizzo M, Ismaila AS. Economic burden of chronic obstructive pulmonary disease (COPD): a systematic literature review. Int J Chron Obstruct Pulmon Dis 2020;15:439–460. |
| 2. | Nici L. Improving the lives of individuals with chronic respiratory disease: the need for innovation. Thorax 2022;77:636–637. |
| 3. | Chenna PR, Mannino DM. Outcomes of severe COPD exacerbations requiring hospitalization. Semin Respir Crit Care Med 2010;31:286–294. |
| 4. | Garcia-Aymerich J, Serra Pons I, Mannino DM, Maas AK, Miller DP, Davis KJ. Lung function impairment, COPD hospitalisations and subsequent mortality. Thorax 2011;66:585–590. |
| 5. | Kim V, Aaron SD. What is a COPD exacerbation? Current definitions, pitfalls, challenges and opportunities for improvement. Eur Respir J 2018;52:1801261. |
| 6. | Carrasco Garrido P, de Miguel Díez J, Rejas Gutiérrez J, Centeno AM, Gobartt Vázquez E, Gil de Miguel A, et al. Negative impact of chronic obstructive pulmonary disease on the health-related quality of life of patients. Results of the EPIDEPOC study. Health Qual Life Outcomes 2006;4:31. |
| 7. | Gordon CS, Waller JW, Cook RM, Cavalera SL, Lim WT, Osadnik CR. Effect of pulmonary rehabilitation on symptoms of anxiety and depression in COPD: a systematic review and meta-analysis. Chest 2019;156:80–91. |
| 8. | GBD Chronic Respiratory Disease Collaborators. Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet Respir Med 2020;8:585–596. |
| 9. | Halpin DMG, Miravitlles M. Chronic obstructive pulmonary disease: the disease and its burden to society. Proc Am Thorac Soc 2006;3:619–623. |
| 10. | Lindenauer PK, Dharmarajan K, Qin L, Lin Z, Gershon AS, Krumholz HM. Risk trajectories of readmission and death in the first year after hospitalization for chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2018;197:1009–1017. |
| 11. | Press VG, Au DH, Bourbeau J, Dransfield MT, Gershon AS, Krishnan JA, et al. Reducing chronic obstructive pulmonary disease hospital readmissions. An official American Thoracic Society workshop report. Ann Am Thorac Soc 2019;16:161–170. |
| 12. | McCarthy B, Casey D, Devane D, Murphy K, Murphy E, Lacasse Y. Pulmonary rehabilitation for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2015;2:CD003793. |
| 13. | Clini E, Holland AE, Pitta F, Troosters T, editors. Textbook of pulmonary rehabilitation. Cham, Switzerland: Springer International Publishing AG; 2018. |
| 14. | Spruit MA, Singh SJ, Garvey C, ZuWallack R, Nici L, Rochester C, et al.; ATS/ERS Task Force on Pulmonary Rehabilitation. An official American Thoracic Society/European Respiratory Society statement: key concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Med 2013;188:e13–e64. |
| 15. | Stefan MS, Pekow PS, Priya A, ZuWallack R, Spitzer KA, Lagu TC, et al. Association between initiation of pulmonary rehabilitation and rehospitalizations in patients hospitalized with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2021;204:1015–1023. |
| 16. | Keating A, Lee A, Holland AE. What prevents people with chronic obstructive pulmonary disease from attending pulmonary rehabilitation? A systematic review. Chron Respir Dis 2011;8:89–99. |
| 17. | Nici L. Pulmonary rehabilitation after a chronic obstructive pulmonary disease exacerbation: impact on readmission risk in a real-world setting. Am J Respir Crit Care Med 2021;204:1005–1006. |
| 18. | Stickland M, Jourdain T, Wong EY, Rodgers WM, Jendzjowsky NG, Macdonald GF. Using telehealth technology to deliver pulmonary rehabilitation in chronic obstructive pulmonary disease patients. Can Respir J 2011;18:216–220. |
| 19. | Jones SE, Green SA, Clark AL, Dickson MJ, Nolan AM, Moloney C, et al. Pulmonary rehabilitation following hospitalisation for acute exacerbation of COPD: referrals, uptake and adherence. Thorax 2014;69:181–182. |
| 20. | Spruit MA, Singh SJ. Maintenance programs after pulmonary rehabilitation: how may we advance this field? Chest 2013;144:1091–1093. |
| 21. | Alison JA, McKeough ZJ, Johnston K, McNamara RJ, Spencer LM, Jenkins SC, et al.; Lung Foundation Australia and the Thoracic Society of Australia and New Zealand. Australian and New Zealand pulmonary rehabilitation guidelines. Respirology 2017;22:800–819. |
| 22. | Malaguti C, Dal Corso S, Janjua S, Holland AE. Supervised maintenance programmes following pulmonary rehabilitation compared to usual care for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2021;8:CD013569. |
| 23. | Brennan D, Tindall L, Theodoros D, Brown J, Campbell M, Christiana D, et al. A blueprint for telerehabilitation guidelines. Int J Telerehabil 2010;2:31–34. |
| 24. | Holland AE, Mahal A, Hill CJ, Lee AL, Burge AT, Cox NS, et al. Home-based rehabilitation for COPD using minimal resources: a randomised, controlled equivalence trial. Thorax 2017;72:57–65. |
| 25. | Zanaboni P, Hoaas H, Aarøen Lien L, Hjalmarsen A, Wootton R. Long-term exercise maintenance in COPD via telerehabilitation: a two-year pilot study. J Telemed Telecare 2017;23:74–82. |
| 26. | Cox NS, Dal Corso S, Hansen H, McDonald CF, Hill CJ, Zanaboni P, et al. Telerehabilitation for chronic respiratory disease. Cochrane Database Syst Rev 2021;1:CD013040. |
| 27. | Vasilopoulou M, Papaioannou AI, Kaltsakas G, Louvaris Z, Chynkiamis N, Spetsioti S, et al. Home-based maintenance tele-rehabilitation reduces the risk for acute exacerbations of COPD, hospitalisations and emergency department visits. Eur Respir J 2017;49:1602129. |
| 28. | Bernocchi P, Vitacca M, La Rovere MT, Volterrani M, Galli T, Baratti D, et al. Home-based telerehabilitation in older patients with chronic obstructive pulmonary disease and heart failure: a randomised controlled trial. Age Ageing 2018;47:82–88. |
| 29. | Barberan-Garcia A, Vogiatzis I, Solberg HS, Vilaró J, Rodríguez DA, Garåsen HM, et al.; NEXES Consortium. Effects and barriers to deployment of telehealth wellness programs for chronic patients across 3 European countries. Respir Med 2014;108:628–637. |
| 30. | Taylor D, Jenkins AR, Parrott K, Benham A, Targett S, Jones AW. Efficacy of unsupervised exercise in adults with obstructive lung disease: a systematic review and meta-analysis. Thorax 2021;76:591–600. |
| 31. | Zanaboni P, Dinesen B, Hjalmarsen A, Hoaas H, Holland AE, Oliveira CC, et al. Long-term integrated telerehabilitation of COPD patients: a multicentre randomised controlled trial (iTrain). BMC Pulm Med 2016;16:126. |
| 32. | Borg G. Borg’s perceived exertion and pain scales. Champaign, IL: Human Kinetics; 1998. |
| 33. | Holland AE, Spruit MA, Troosters T, Puhan MA, Pepin V, Saey D, et al. An official European Respiratory Society/American Thoracic Society technical standard: field walking tests in chronic respiratory disease. Eur Respir J 2014;44:1428–1446. |
| 34. | Jones PW, Harding G, Berry P, Wiklund I, Chen WH, Kline Leidy N. Development and first validation of the COPD assessment test. Eur Respir J 2009;34:648–654. |
| 35. | EuroQol Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199–208. |
| 36. | Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–370. |
| 37. | Schwarzer R, Jerusalem M. Generalized self-efficacy scale. In: Weinman J, Wright S, Johnston M, editors. Measures in health psychology: a user’s portfolio. Causal and control beliefs. Windsor, United Kingdom: NFER-NELSON; 1995. pp. 35–37. |
| 38. | Hurst H, Bolton J. Assessing the clinical significance of change scores recorded on subjective outcome measures. J Manipulative Physiol Ther 2004;27:26–35. |
| 39. | Hoaas H, Andreassen HK, Lien LA, Hjalmarsen A, Zanaboni P. Adherence and factors affecting satisfaction in long-term telerehabilitation for patients with chronic obstructive pulmonary disease: a mixed methods study. BMC Med Inform Decis Mak 2016;16:26. |
| 40. | Spencer LM, Alison JA, McKeough ZJ. A survey of opinions and attitudes toward exercise following a 12-month maintenance exercise program for people with COPD. Cardiopulm Phys Ther J 2013;24:30–35. |
| 41. | Holland AE, Cox NS. Give a little or give a lot?: The question of long-term maintenance after pulmonary rehabilitation. Chest 2021;159:893–894. |
| 42. | Spencer LM, McKeough ZJ. Maintaining the benefits following pulmonary rehabilitation: achievable or not? Respirology 2019;24:909–915. |
| 43. | Güell MR, Cejudo P, Ortega F, Puy MC, Rodríguez-Trigo G, Pijoan JI, et al. Benefits of long-term pulmonary rehabilitation maintenance program in patients with severe chronic obstructive pulmonary disease. Three-year follow-up. Am J Respir Crit Care Med 2017;195:622–629. |
| 44. | Ries AL, Kaplan RM, Myers R, Prewitt LM. Maintenance after pulmonary rehabilitation in chronic lung disease: a randomized trial. Am J Respir Crit Care Med 2003;167:880–888. |
| 45. | Carr SJ, Goldstein RS, Brooks D. Acute exacerbations of COPD in subjects completing pulmonary rehabilitation. Chest 2007;132:127–134. |
| 46. | Vincent E, Sewell L, Wagg K, Deacon S, Williams J, Singh S. Measuring a change in self-efficacy following pulmonary rehabilitation: an evaluation of the PRAISE tool. Chest 2011;140:1534–1539. |
| 47. | Maltais F, Bourbeau J, Shapiro S, Lacasse Y, Perrault H, Baltzan M, et al.; Chronic Obstructive Pulmonary Disease Axis of Respiratory Health Network, Fonds de recherche en santé du Québec. Effects of home-based pulmonary rehabilitation in patients with chronic obstructive pulmonary disease: a randomized trial. Ann Intern Med 2008;149:869–878. |
| 48. | Cox NS, McDonald CF, Alison JA, Mahal A, Wootton R, Hill CJ, et al. Telerehabilitation versus traditional centre-based pulmonary rehabilitation for people with chronic respiratory disease: protocol for a randomised controlled trial. BMC Pulm Med 2018;18:71. |
| 49. | Hansen H, Bieler T, Beyer N, Kallemose T, Wilcke JT, Østergaard LM, et al. Supervised pulmonary tele-rehabilitation versus pulmonary rehabilitation in severe COPD: a randomised multicentre trial. Thorax 2020;75:413–421. |
| 50. | Cox NS, McDonald CF, Mahal A, Alison JA, Wootton R, Hill CJ, et al. Telerehabilitation for chronic respiratory disease: a randomised controlled equivalence trial. Thorax 2022;77:643–651. |
This study was funded by Helse Nord RHF (HST1117-13 and HST1118-13) and Norges Forskningsråd (228919/H10).
Author Contributions: P.Z., B.D., R.W., and A.E.H. conceived and designed the study. H.H., J.B., T.T.J., and B.R.M. recruited the study participants. P.Z., B.D., H.H., A.T.B., R.P., C.C.O., J.B., T.T.J., B.R.M., and A.E.H. contributed to data collection. P.Z. conducted the primary analysis and prepared the first draft of the manuscript. P.Z., B.D., R.W., and A.E.H. made substantial contributions to the interpretation of data. All authors reviewed and revised the manuscript critically for important intellectual content. All authors gave final approval of the version to be published.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Originally Published in Press as DOI: 10.1164/rccm.202204-0643OC on December 8, 2022
Author disclosures are available with the text of this article at www.atsjournals.org.
