Changes in physical activity, sedentary behaviour and sleep following pulmonary rehabilitation: a systematic review and network meta-analysis

Background: The variety of innovations to traditional centre-based pulmonary rehabilitation (CBPR), including different modes of delivery and adjuncts, are likely to lead to differential responses in physical activity, sedentary behaviour and sleep. Objectives: To examine the relative effectiveness of different pulmonary rehabilitation-based interventions on physical activity, sedentary behaviour and sleep. Methods: Randomised trials in chronic respiratory disease involving pulmonary rehabilitation-based interventions were systematically searched for. Network meta-analyses compared interventions for changes in physical activity, sedentary behaviour and sleep in COPD. Results: 46 studies were included, and analyses were performed on most common outcomes: steps per day (k=24), time spent in moderate-to-vigorous physical activity (MVPA; k=12) and sedentary time (k=8). There were insufficient data on sleep outcomes (k=3). CBPR resulted in greater steps per day and MVPA and reduced sedentary time compared to usual care. CBPR+physical activity promotion resulted in greater increases in steps per day compared to both usual care and CBPR, with greater increases in MVPA and reductions in sedentary time compared to usual care, but not CBPR. Home-based pulmonary rehabilitation resulted in greater increases in steps per day and decreases in sedentary time compared to usual care. Compared to usual care, CBPR+physical activity promotion was the only intervention where the lower 95% confidence interval for steps per day surpassed the minimal important difference. No pulmonary rehabilitation-related intervention resulted in greater increases in MVPA or reductions in sedentary time compared to CBPR. Conclusion: The addition of physical activity promotion to pulmonary rehabilitation improves volume of physical activity, but not intensity, compared to CBPR. High risk of bias and low certainty of evidence suggests that these results should be viewed with caution.