Abstract
This systematic review aimed to systematise the different models used to deliver pulmonary rehabilitation (PR) during chronic respiratory diseases (CRDs) and explore which ones are the most effective in terms of dyspnea, exercise capacity, and health-related quality of life (HRQoL). The literature search strategy involved structured searches of PubMed, Web of Science, and Cochrane Library for relevant articles published from January 2013 to March 2025. The risk of bias was assessed using ROB 2.0. Descriptive analysis and meta-analysis were performed. Forest plots and the node-splitting model presented results. Network meta-analysis was conducted in R 4.3.2. 33 studies(n = 2538) were included in this review and of those, 27 studies(n = 2106) were used for meta-analysis. 22 (66.7%) studies were at high risk of bias and the certainty of evidence for all outcomes (6MWT, dyspnea, HRQoL) was rated as low due to study limitations and imprecision. Compared with usual care, PR patients have significant improvement on 6MWT as well as mMRC (both P < 0.01). The cumulative ranking probability curves and forest plot analyses revealed a hierarchical efficacy profile among rehabilitation modalities for CRDs, with outpatient presented the larger effects on mMRC (mean difference (MD)= -0.82, 95%Cl [-1.45, -0.19]), 6MWT(MD = 65.45, 95%Cl [45.06, 85.84]). Our findings suggest that PR probably improves exercise capacity and reduces dyspnea in patients with CRDs, with outpatient-based programmes generally showing the largest effects, while the high risk of bias limits interpretation of this finding.PROSPERO ID: CRD420251013615.