Pulmonology · Medical Literature
Smartphone App-Based Music-Facilitated Pulmonary Rehabilitation Program Integrating Rhythm-Guided Walking and Singing for Patients With Chronic Obstructive Pulmonary Disease: Multicenter Randomized Controlled Trial.
Abstract
Pulmonary rehabilitation (PR) is a cornerstone for the management of chronic obstructive pulmonary disease (COPD), yet global uptake remains low due to geographic and resource barriers. Digital health technologies, specifically smartphone apps, offer a promising platform for delivering accessible home-based PR. In addition, music-assisted interventions not only offer unique physiological and psychological benefits but may also serve as an innovative approach to enhancing patient engagement and improving the effectiveness of rehabilitation in home settings. This study aimed to evaluate the effectiveness of a smartphone app-based, music-facilitated multicomponent PR program (integrating rhythm-guided walking [RW] and singing) for improving exercise capacity and other clinical outcomes in patients with COPD compared with usual care (UC). This 3-arm, parallel-group, multicenter randomized controlled trial included 70 participants in China. Participants were randomized into a multimodule training (MT) group, which included a multicomponent PR program integrating RW and singing training (n=25); a RW group, which included RW training (n=23); or a UC group (n=22). The MT and RW groups received 12-week asynchronous home-based training via a smartphone app, and all arms received structured patient education. The primary outcome was the distance achieved in the incremental shuttle walking test (ISWT) at 12 weeks. The secondary outcomes included dyspnea, quality of life, and pulmonary function. The modified intention-to-treat principle was used to analyze the 70 study patients. At 12 weeks, the ISWT distance was significantly greater in the MT group than in the UC group (mean difference [MD] 56.35 m, 95% CI 6.66-106.04 m; P=.03; Cohen d=0.30). Significant improvements were observed in the MT group compared with the UC group in the modified Medical Research Council dyspnea scale (mMRC) score (MD -0.44, 95% CI -0.80 to -0.08; P=.02), COPD Assessment Test score (MD -3.23, 95% CI -6.18 to -0.29; P=.03), Hospital Anxiety and Depression Scale-anxiety subscale score (MD -2.31, 95% CI -3.99 to -0.63; P=.008), and inspiratory capacity (MD 15.98% predicted, 95% CI 4.76 to 27.21; P=.01). However, no significant differences were found between the RW and UC groups in primary or secondary outcomes. Compared with RW, MT was significantly better at decreasing the mMRC score (P=.03). The findings of this study demonstrate that our smartphone app-based music-facilitated multicomponent PR program (including tempo-guided walking and singing) caused clinically meaningful improvements in exercise capacity among patients with COPD compared to UC. Moreover, secondary outcomes, including dyspnea, quality of life, psychological status, and inspiratory capacity, showed better improvements with MT than with UC.
Clinical relevance
In patients with moderate-to-severe stable COPD who cannot access center-based pulmonary rehabilitation, does a 12-week smartphone app–based, music-facilitated multicomponent program integrating rhythm-guided walking and singing (MT) improve exercise capacity and clinical outcomes compared with rhythm-guided walking alone (RW) or usual care (UC)? __SNAPSHOT__ {"studyDesign":"Multicenter, pragmatic, 3-arm, parallel-group, single-blind pilot randomized controlled trial (COPDMELODY; ClinicalTrials.gov NCT05832814)","setting":"5 hospitals across 3 administrative regions in China; home-based intervention delivered via smartphone app","sampleSize":"n=70 (MT: n=25; RW: n=23; UC: n=22); modified intention-to-treat (mITT) analysis included all 70 enrolled participants","population":"Adults aged 40–75 years with a confirmed COPD diagnosis, postbronchodilator FEV1 <80% predicted, and FEV1/FVC ratio <0.70 (consistent with GOLD moderate-to-severe disease), with stable symptoms at enrollment; recruited from Chinese hospital outpatient settings","intervention":"Multimodule Training (MT): 12-week asynchronous, home-based pulmonary rehabilitation program delivered via the 'Qiyue' smartphone app incorporating (1) individualized tempo-guided rhythm walking (aerobic training at 75% of peak ISWT walking speed, ≥3 sessions/week of ≥30 min, with melodies at 60–120 bpm matched to prescribed stride frequency) and (2) structured singing training (respiratory muscle training using songs selected and recorded by qualified music therapists), plus systematic patient education. A wristwatch was used to capture training time and intensity. Participants also attended a face-to-face introductory session and a videoconference during their first training session.","comparator":"RW group: rhythm-guided walking training only (same aerobic protocol as MT, without singing module) via the same app over 12 weeks, plus education. UC group: structured patient education only, no active rehabilitation program.","primaryOutcome":"Incremental Shuttle Walking Test (ISWT) distance (meters) at 12 weeks; minimal clinically important difference (MCID) = 35 meters","followUp":"12 weeks (outcomes assessed at baseline, 4, 8, and 12 weeks); no post-intervention follow-up reported","fundingSource":"Not reported in the abstract or accessible full-text sections reviewed; no industry sponsorship declared"}
Key findings
The mITT analysis included all 70 randomized participants. At 12 weeks, ISWT distance was significantly greater in the MT group versus UC (mean difference [MD] +56.35 m, 95% CI 6.66–106.04 m; P=.03; Cohen d=0.30), exceeding the pre-specified MCID of 35 m and representing a clinically meaningful, albeit small-to-moderate effect size. The RW group showed a numerical trend toward greater ISWT distance versus UC that did not reach statistical significance (MD +38.91 m, 95% CI −10.91 to 88.74 m; P=.13). No significant between-group difference was observed in ISWT distance between MT and RW (P=.45). Within-group, ISWT distance increased significantly from baseline in both MT (P<.001) and RW (P=.005) but not in UC (P=.39). For secondary outcomes, the MT group demonstrated significant improvements versus UC in: mMRC dyspnea scale (MD −0.44, 95% CI −0.80 to −0.08; P=.02; MCID=0.5), COPD Assessment Test score (MD −3.23, 95% CI −6.18 to −0.29; P=.03; MCID=1.6), HADS-Anxiety subscale (MD −2.31, 95% CI −3.99 to −0.63; P=.008), and inspiratory capacity (MD +15.98% predicted, 95% CI 4.76–27.21; P=.01). The MT group was also significantly better than the RW group in reducing mMRC score. Differences between RW and UC were not statistically significant for any secondary outcome. Per-protocol analysis confirmed the primary finding (MT vs. UC, P=.03). Attrition at 12 weeks was 21% overall: UC group experienced higher dropout (8/22, 36%) than MT (5/25, 20%) or RW (not individually reported). NNT for achieving ISWT MCID was not calculable from available data. No specific adverse event data were reported in the accessible results sections; the protocol pre-specified monitoring of exacerbations, falls, and exercise injuries. Multiple secondary outcomes were tested without correction for multiplicity, as this was an explicitly exploratory pilot study.
Practice implications
This pilot RCT (COPDMELODY) established that a 12-week home-based, app-delivered, music-facilitated multicomponent pulmonary rehabilitation program (rhythm-guided walking plus singing) produced a statistically significant and nominally clinically meaningful improvement in exercise capacity versus usual care in patients with moderate-to-severe stable COPD in China. The magnitude of ISWT improvement (56.35 m) exceeded the 35-m MCID, and the effect was consistent in sensitivity analyses, lending support to the directional signal. Secondary outcomes (dyspnea, health-related quality of life via CAT, anxiety, and inspiratory capacity) also favored MT over UC. However, several important caveats limit interpretability: (1) This was explicitly a pilot study, underpowered and designed for preliminary efficacy and feasibility assessment, not definitive confirmation; (2) The study did not establish superiority of the combined program (MT) over rhythm-guided walking alone (RW) on the primary outcome—the MT vs. RW difference was non-significant (P=.45)—meaning the incremental value of adding singing training to walking cannot be confirmed from this trial; (3) No correction for multiple secondary outcome testing was applied, inflating the risk of spurious positive findings among secondary endpoints; (4) Participant blinding was not possible due to the nature of the interventions; (5) The 12-week follow-up is short and provides no data on durability of benefit, exacerbation rates, hospitalizations, or long-term outcomes; (6) The sample was drawn exclusively from Chinese hospital outpatients and enrolled only patients with stable COPD, limiting generalizability to other ethnic populations, healthcare systems, patients post-exacerbation, or those with very severe disease (GOLD 3–4 specifically); (7) UC attrition was disproportionately high (36%), which may have inflated apparent between-group differences. The app (Qiyue) was developed by the research team and has not been independently validated. Findings should be considered hypothesis-generating and warrant confirmation in a larger, adequately powered definitive RCT with longer follow-up. Only the abstract and open-access full-text sections were reviewable; complete adverse event data, full baseline demographics, and funding disclosures from the complete manuscript may refine this assessment. __LIMITATIONS__ Small pilot sample (n=70 across 3 arms), explicitly underpowered for definitive efficacy conclusions; confidence intervals for the primary outcome are wide (CI width ~99 m), reflecting substantial statistical imprecision.,No multiple-testing correction applied across multiple secondary outcomes in this exploratory pilot; false-positive secondary findings cannot be excluded.,Single-blind design: outcome assessors and statisticians were blinded, but participant blinding was not feasible due to the nature of the interventions, introducing performance and reporting bias risks.,Short follow-up (12 weeks only); durability of exercise capacity gains, impact on exacerbation rates, hospitalizations, and mortality are unknown.,High and differential dropout in the UC arm (36% vs. 20% in MT), which may have introduced bias favoring the intervention in mITT analyses.,Exclusively Chinese hospital-based patient population limits generalizability to non-Chinese populations, community- or primary care–recruited patients, or healthcare systems with different standards of 'usual care.',The smartphone app (Qiyue) was purpose-built by the investigators; independent usability validation, scalability, and real-world effectiveness outside the trial context have not been established.,MT vs. RW comparison on the primary outcome was non-significant; the added value of singing training over rhythm walking alone remains unconfirmed.,No long-term follow-up data, no exacerbation or hospitalization outcomes, and specific adverse event frequencies are not reported in accessible sections of the publication.,Funding source not reported; potential conflicts of interest related to app development cannot be fully assessed. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ The COPDMELODY pilot RCT demonstrated that a 12-week smartphone app–based, music-facilitated pulmonary rehabilitation program (rhythm-guided walking plus singing) produced a statistically significant and nominally MCID-exceeding improvement in exercise capacity versus usual care in stable moderate-to-severe COPD, with additional benefits in dyspnea, quality of life, anxiety, and inspiratory capacity. However, this was a small pilot study (n=70) with no multiple-testing correction, high UC dropout, short follow-up, no proof of superiority over rhythm walking alone, and results confined to a Chinese hospital population—so the findings are promising but hypothesis-generating only. A larger, adequately powered, and longer-duration confirmatory RCT is needed before this intervention can be recommended for broad clinical implementation.