Singing for COPD

Singing uses the lungs to provide airflow to produce musical words or sounds with the voice. Singing can require a lot of effort for muscle contraction and co-ordination. This may benefit people with chronic obstructive pulmonary disease (COPD) in a manner similar to that of breathing exercises. Singing is said to be beneficial for health but we need evidence for this before it can be recommended specifically to address health conditions. We planned to examine whether singing had any effect on quality of life or breathlessness in people with COPD. We included three studies with a total of 112 participants. Participants were randomly assigned to singing training or to a non-singing control group. The control groups were either a film workshop, handcraft work, or nothing at all. The singing was performed in groups, once to twice a week for one hour, for a minimum of six weeks. There was diversity in the results of the studies and we were unable to combine many results in 'meta-analyses'. A meta-analysis is a statistical analysis which combines the results of two or more separate studies to give a pooled result. Some studies showed improvements in some aspects of quality of life, while others showed no improvement. Breathlessness was only measured in one study and no improvement was found. The studies did not report whether any effects lasted for a long time after the singing training was completed. No studies reported any side effects from singing, so singing appears to be safe for people with COPD. The studies were of low quality due to the small number of participants and missing information about the methods and some of the outcomes. We were unable to find enough evidence to sufficiently determine the effect of singing in people with COPD. More studies are required and they should concentrate on enrolling larger numbers of people.

Authors' conclusions: 

There is low to very low-quality evidence that singing is safe for people with COPD and improves physical health (as measured by the SF-36 physical component score), but not dyspnoea or respiratory-specific quality of life. The evidence is limited due to the low number of studies and the small sample size of each study. No evidence exists examining the long-term effect of singing for people with COPD. The absence of studies examining singing performed in conjunction with pulmonary rehabilitation precludes the formulation of conclusions about the effects of singing in this context. More randomised controlled trials with larger sample sizes and long-term follow-up, and trials examining the effect of singing in addition to pulmonary rehabilitation, are required to determine the effect of singing on health-related quality of life and dyspnoea in people with COPD.

Read the full abstract...

Singing is a complex physical activity dependent on the use of the lungs for air supply to regulate airflow and create large lung volumes. In singing, exhalation is active and requires active diaphragm contraction and good posture. Chronic obstructive pulmonary disease (COPD) is a progressive, chronic lung disease characterised by airflow obstruction. Singing is an activity with potential to improve health outcomes in people with COPD.


To determine the effect of singing on health-related quality of life and dyspnoea in people with COPD.

Search strategy: 

We identified trials from the Cochrane Airways Specialised Register,, the World Health Organization trials portal and PEDro, from their inception to August 2017. We also reviewed reference lists of all primary studies and review articles for additional references.

Selection criteria: 

We included randomised controlled trials in people with stable COPD, in which structured supervised singing training of at least four sessions over four weeks' total duration was performed. The singing could be performed individually or as part of a group (choir) facilitated by a singing leader. Studies were included if they compared: 1) singing versus no intervention (usual care) or another control intervention; or 2) singing plus pulmonary rehabilitation versus pulmonary rehabilitation alone.

Data collection and analysis: 

Two review authors independently screened and selected trials for inclusion, extracted outcome data and assessed risk of bias. We contacted authors of trials for missing data. We calculated mean differences (MDs) using a random-effects model. We were only able to analyse data for the comparison of singing versus no intervention or a control group.

Main results: 

Three studies (a total of 112 participants) were included. All studies randomised participants to a singing group or a control group. The comparison groups included a film workshop, handcraft work, and no intervention. The frequency of the singing intervention in the studies ranged from 1 to 2 times a week over a 6 to 24 week period. The duration of each singing session was 60 minutes.

All studies included participants diagnosed with COPD with a mean age ranging from 67 to 72 years and a mean forced expiratory volume in one second (FEV1) ranging from 37% to 64% of predicted values. The sample size of included studies was small (33 to 43 participants) and overall study quality was low to very low. Blinding of personnel and participants was not possible due to the physical nature of the intervention, and selection and reporting bias was present in two studies.

For the primary outcome of health-related quality of life, there was no statistically significant improvement in the St George's Respiratory Questionnaire total score (mean difference (MD) -0.82, 95% confidence interval (CI) -4.67 to 3.02, 2 studies, n = 58, low-quality evidence). However, there was a statistically significant improvement in the SF-36 Physical Component Summary (PCS) score favouring the singing group (MD 12.64, 95% CI 5.50 to 19.77, 2 studies, n = 52, low-quality evidence). Only one study reported results for the other primary outcome of dyspnoea, in which the mean improvement in Baseline Dyspnoea Index (BDI) score favouring the singing group was not statistically significant (MD 0.40, 95% CI -0.65 to 1.45, 1 study, n = 30, very low-quality evidence).

No studies examined any long-term outcomes and no adverse events or side effects were reported.