An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) FINAL REPORT

Abstract

Aim The aim of this study was to explore the feasibility of weekly community singing for people with COPD and to assess impact on lung function, functional capacity, breathlessness and quality of life. Method An uncontrolled observational study of a weekly group singing programme was undertaken over the period September 2011 to June 2012. The St Georges Respiratory Questionnaire (SGRQ), MRC breathlessness scale, EQ-5D and York SF-12 were administered at baseline, mid-point and end of study, and spirometry to assess lung function at baseline and study end. Written feedback from participants was analysed for the principal themes expressed. Results Health-related quality of life assessed by SGRQ showed a 3.3 point change in the direction of health improvement. Improvements were also found in FEV1, FVC and FVC per cent. Qualitative evidence showed that the singing groups were enjoyable social events and participants reported improvements in their breathing, activity levels and wellbeing. Conclusions Health improvements and positive feedback are encouraging as COPD is a progressive illness and a decline in health would be expected over ten months. The study provides a good foundation for designing a more robust controlled community trial. Documentaries and guide Three short films and a guide to Singing for People with COPD accompany this report and illustrate the work of the project. They are available from the Sidney De Haan Research Centre for Arts and Health. See inside back cover for details.

Full text

Sidney De Haan
Research Centre for Arts and Health
An evaluation of
community singing
for people with COPD
(Chronic Obstructive
Pulmonary Disease)
FINAL REPORT
Stephen Clift, Ian Morrison, Ann Skingley, Sonia Page, Simon Coulton,
Pauline Treadwell, Trish Vella-Burrows, Isobel Salisbury, Matthew Shipton

Summary
Aim
The aim of this study was to explore the feasibility of weekly community singing for people with COPD
and to assess impact on lung function, functional capacity, breathlessness and quality of life.
Method
An uncontrolled observational study of a weekly group singing programme was undertaken over
the period September 2011 to June 2012. The St Georges Respiratory Questionnaire (SGRQ), MRC
breathlessness scale, EQ-5D and York SF-12 were administered at baseline, mid-point and end of study,
and spirometry to assess lung function at baseline and study end. Written feedback from participants
was analysed for the principal themes expressed.
Results
Health-related quality of life assessed by SGRQ showed a 3.3 point change in the direction of health
improvement. Improvements were also found in FEV1, FVC and FVC per cent. Qualitative evidence
showed that the singing groups were enjoyable social events and participants reported improvements
in their breathing, activity levels and wellbeing.
Conclusions
Health improvements and positive feedback are encouraging as COPD is a progressive illness and
a decline in health would be expected over ten months. The study provides a good foundation for
designing a more robust controlled community trial.
Documentaries and guide
Three short films and a guide to Singing for People with COPD accompany this report and illustrate the
work of the project. They are available from the Sidney De Haan Research Centre for Arts and Health.
See inside back cover for details.
Acknowledgments
Grateful thank s are due to members of the steering group for their assistance throughout the project; the facilitator team
for their dedication and expertise, and all the research participants who so kindly and enthusiastically took part.
This report presents findings from independent research commissioned by The Dunhill Medical Trust (grant ref:
R176/1110). The views expressed are those of the authors and not necessarily those of The Dunhill Medical Trust.
Authors: Stephen Clift,* Ian Morrison,* Ann Skingley,* Sonia Page,* Simon Coulton,+ Pauline Treadwell,
#
Trish Vella-Burrows,* Isobel Salisbury* and Matthew Shipton.*
* Sidney De Haan Research Centre for Arts and Health, Canterbury Christ Church University
+ Centre for Health Services Studies, University of Kent
# Kent Community Health NHS Trust
Publisher: Canterbury Christ Church University | ISBN: 9781909067127 | Published: June 2013
Sidney De Haan Research Centre for Arts and Health, University Centre Folkestone, Folkestone, Kent CT20 1JG
© Sidney De Haan Research Centre for Arts and Health
Sidney De Haan
Research Centre for Arts and Health

3
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Contents
Summary .................................................................................................................................................................................................................................... 2
Methods.......................................................................................................................................................................................................................................6
Quantitative results .......................................................................................................................................................................................... 10
Table 1: Measures of lung function at baseline and end of programme ..........................................................................10
Table 2: Measures of self-assessed health at baseline and end of programme .......................................................... 11
Quantitative discussion ............................................................................................................................................................................................12
Quantitative conclusions ........................................................................................................................................................................................15
Qualitative findings ...........................................................................................................................................................................................16
Figure 1: Conceptual framework for data analysis .................................................................................................................................17
Mechanisms for achieving health improvements ...................................................................................................................................18
Areas of perceived benefits to psychological health ...........................................................................................................................21
Facilitator qualities valued by participants .................................................................................................................................................... 24
Qualitative discussion .................................................................................................................................................................................................26
Qualitative conclusions .............................................................................................................................................................................................28
References ...........................................................................................................................................................................................................................30

Sidney De Haan
Research Centre for Arts and Health
4
An evaluation of community singing
for people with Chronic Obstructive
Pulmonary Disease (COPD)
Study background
Innovative, cost-effective initiatives are needed to help people with COPD engage in physical and
social activity to support independence and maintain quality of life. This study explores the potential
effect of regular group singing on the clinical symptoms of COPD. Surveys have shown that choral
singers believe that singing improves their breathing (Clift and Hancox 2001; Clift et al. 2009) but
comparison of lung function in professional singers versus wind and percussion players, failed to
show significant differences in standard spirometric parameters (Clift et al. 2008). There is some
evidence, however, that group singing may be beneficial for people with chronic respiratory disease
by modifying breathing patterns, reducing breathlessness, and improving quality of life. Macklem
(2010) has recently argued that basic considerations of the pathophysiology of COPD suggest that
encouraging patients to breathe slowly and deeply during exercise, and avoid rapid upper thoracic
patterns of breathing, should help to lessen dyspnoea and improve performance.
Engen (2005) recruited participants from a gerontology clinic and pulmonary rehabilitation clinic who
had a diagnosis of emphysema. Twelve participants met in small groups twice a week for six weeks.
None of the physical health and quality of life measures employed showed improvements over the six
weeks of the study, but measures of breath control and voice intensity both improved significantly. In
addition, breathing mode changed from being predominantly clavicular to being diaphragmatic in all
cases and this was maintained for two weeks after the treatment sessions ended.
Bonilha et al. (2008) reported a small randomised controlled trial assessing the impact of singing
groups on lung function and quality of life among patients diagnosed with COPD. This study
randomised 43 patients to a programme of singing or handcraft classes. Fifteen participants in
each group completed 24 sessions and were comparable at baseline in their mean forced expiratory
volumes at one second as percentage of normative values (FEV1%). The singing group showed a small
improvement in a measure of maximal expiratory pressure at the end of the study, while the control
group showed a decline, with the difference being statistically significant. Both groups showed
increased quality of life scores with no significant difference, emphasising the benefits of group
participation for perceived quality of life.
Two small trials examining the effects of singing lessons for patients with COPD have been completed
at the Royal Brompton Hospital, London. In the first (Lord et al. 2010) thirty-six COPD patients were
randomised to either 12 one-hour sessions of singing lessons over six weeks, or usual care. Following
attrition 15 patients in the singing group (mean baseline FEV1 per cent predicted 36.8) were
compared with 13 controls (mean baseline FEV1 per cent predicted 37.7). Significant improvements
were found in levels of anxiety and self-assessed physical wellbeing in the singing group. No
differences were found between the groups for single breath counting, incremental shuttle walking
test (ISWT) scores or recovery time following ISWT and intriguingly breath-hold time increased more
in the control group than the singing group. In the second study (Lord et al. 2012) 33 patients with
COPD participated in either 16 sessions of singing over eight weeks (mean FEV1 per cent predicted
44.4), or an active control condition in which participants watched films together and discussed
them (mean FEV1 per cent predicted 63.5). Although the mean FEV1 per cent predicted value was
higher for the control group, this difference was not statistically significant. Follow up assessment
showed that the singing group improved significantly in self-assessed physical health compared with
the control group, but no differences emerged in direct measures of lung function.

5
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
To date, therefore, research on singing and COPD has been limited, with small sample sizes and
short interventions in clinical settings focused on the teaching of singing. While existing research
has shown that singing is an acceptable activity for people with COPD and that it can have general
wellbeing benefits, little or no improvement in measures of lung function have been found. It may
be that the interventions have not been long enough and, indeed, the increase in singing sessions
between the two Royal Brompton studies from 12 to 16 singing sessions was motivated by a concern
that the earlier study was too short to reveal positive benefits. It may be that even the increase
to 16 sessions was still insufficient to promote measurable improvements in breathing and lung
function. In addition, the groups in the Brompton study were small and individuals may not have
experienced the support in singing that comes from being part of a larger choir, nor the impetus
to improve that comes from preparation to perform. The present study addresses the limitations
of previous research through a community-based singing initiative for people with COPD. A larger
group of participants was recruited than in previous studies, and six community singing groups were
established meeting weekly over a longer period of time. In addition to teaching good posture,
breathing techniques and engaging in singing, the groups worked towards combined performance
events in line with a widely followed model of community singing.
Trish Vella-Burrows
Leading singing to celebrate World COPD Day 2011

Sidney De Haan
Research Centre for Arts and Health
6
Methods
The study was an observational evaluation of the feasibility, acceptability, and potential effectiveness,
of regular singing for people with COPD. Ethical approval was given by the Faculty of Health and
Social Care Research Ethics Committee of Canterbury Christ Church University and Oxford C NHS
Research Ethics Committee (REC ref: 11/SC/0115 120821).
Specific objectives of the study were to provide evidence relating to:
i. The effect of participation in regular singing on clinical measures of COPD.
ii. The effect on measures of health related quality of life.
iii. Recruitment and retention rates and preferences for singing as an intervention in this population.
iv. Patient satisfaction with the intervention (measured by questionnaire and explored by an analysis
of written comments).
Design
An observational non-randomised quasi-experimental design was employed. The study was
conducted between September 2011 and July 2012 and assessments undertaken at baseline,
prior to joining a singing group and again at five and ten months. Following the National Institute
of Health and Clinical Excellence guidelines on ‘person-centred care’ in COPD (NICE 2010), this
study recruited individuals into singing groups who expressed a preference for participating in this
activity. Six singing groups were established in or near Ashford, Whitstable, Dover, Deal, Canterbury
and Ramsgate in the county of Kent, South East England. These are areas known to have a high
prevalence of COPD (Whitmore and Limentani 2009).
Recruitment
A variety of methods of recruitment were pursued: a mailed invitation to patients on the COPD
registers within GPs’ practices serving East Kent, newspaper advertisements and direct contact with
three local support groups for people with breathing difficulties (British Lung Foundation Breathe
Easy Groups).
Inclusion criteria
i. Mild, moderate, severe or very severe COPD as assessed by post-bronchodilator spirometry
at baseline.
ii. Physically mobile and able to travel to sessions independently or with the support of a carer.
iii. Able to speak and hear and willing to commit to participating in the project over the course
of 18 months (health permitting).
iv. Able to speak English and complete questionnaires in English.
v. Aged 18 years and over.
Exclusion criteria
i. Severe dementia or other cognitive or communication disabilities which render consent
problematic.
ii. Severe co-morbidities which contra-indicate participation on the advice of GPs.

7
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Sample size
The aim of the study was to explore potential effectiveness and cost-effectiveness prior to embarking
on a community-based randomised controlled trial with a contemporaneous usual care control
group. As a feasibility study no formal sample size calculation was undertaken but it was estimated
that a minimum of 50 participants followed up at the end of the study would provide information on
feasibility and potential effects. Taking into account a conservative retention of 50% we estimated we
would need to recruit 100 participants to the study.
Measures
The following measures were used to assess outcomes for lung function, COPD specific and generic
health-related quality of life:
Lung function
Spirometry tests were carried out by a qualified health professional, to assess FEV1, FEV1 per cent,
FVC, and FVC per cent both before and at the end of the project post bronchodilation.
Lung function data was obtained using a Micro Medical (Care Fusion) MicroLab machine, which
is a portable device that uses a turbine sensor to measure air-flow. It was chosen because it as
it is widely used within Primary Care and Community settings in the UK due to its portability as
well as meeting published recommendations of European Respiratory Society (ERS) and American
Thoracic Society (ATS) (Miller et al. 2005). Before each assessment session the device was
calibrated with a 3 litre calibration syringe. The stage of COPD was defined by the FEV1 per cent
following the GOLD (2010) and NICE (2010) guidance.
St George’s Respiratory Questionnaire (SGRQ)
A self-assessed measure of health impairment employed in research on chronic respiratory illness
and COPD. Four scores are produced: symptoms, activity, impacts and total (SGRQ 2008).
A participant in the project with portable oxygen

Sidney De Haan
Research Centre for Arts and Health
8
MRC Dyspnoea Scale
A 5-point self-rated breathlessness scale for patients with lung disease (Bestall, et al. 1999;
Stenton 2008).
York SF-12
A self-assessed health related quality of life measure (Iglesias et al. 2001) validated for use with older
people and for which population norms exist. The twelve multiple choice questions cover both
physical and mental domains of health.
EuroQol-5D (EQ-5D)
A short, 3-level, 5-dimensional instrument which provides a health utility score (0 – 1), and a self-
assessed overall rating of health using a 0 – 100 visual analogue scale (VAS) (Euroqol Group 1990).
SGRQ, EQ-5D and SF-12 have been shown to relate well in a study of quality of life in patients with
severe COPD hospitalised for exacerbations (Menn et al. 2010). Generic quality of life as assessed by
EQ-5D differentiates between stages of COPD severity (Rutten-van Mölken et al. 2006). The MRC
scale has been shown to provide a valid means of categorising levels of disability in COPD patients
(Bestall et al. 1999). In studies measuring quality of life of patients with COPD it is recommended
that both the clinically specific SGRQ and more generic measures are used (Daudey et al. 2010).
Participants completed spirometry and the battery of questionnaires at baseline. Questionnaires
were completed at mid-study (five months) and study end (ten months), and spirometry undertaken
at study end. On each occasion, participants also had the opportunity to write any comments they
wished to make about their health and their experience of the project (an analysis of their written
feedback is given in the second section of this report).
Statistical analysis was performed using SPSS version 19. Change in quantitative outcome measures
between baseline and post singing programme was assessed using paired t-tests. Distributional
assumptions were checked and where appropriate statistical transformations undertaken. Where no
transformation was possible alternative non-parametric equivalents were employed. No missing data
imputation was conducted.
Lizzie Stephens and Lemon Otter
Leading the Canterbury group

9
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
The intervention
Singing groups were led by skilled and experienced singing facilitators. The facilitators received
five days training and met regularly throughout the project to ensure a broadly consistent
approach. Facilitator training for the role was led by the Musical Director, with some input from
outside trainers. Singing sessions were weekly during academic term time, from September 2011
to June 2012. Sessions were held in community halls which were booked specifically for the
event, were private and afforded acceptable levels of comfort (heating, light, ambient sound,
tea making, etc) with close integral car parking and flat access to the hall on the ground floor.
Sessions were delivered to groups ranging in size from 20 to 50, including supporters (40%
supporters on average). Supporters were health staff or voluntary organisation staff, carers,
partners, friends or family members of the person with COPD. Singing groups took place over
a total of 36 weeks including workshop/performance events at the end of each term. Sessions
were a total of 90 minutes. Thirty minutes were for socialising during the ‘meeting and greeting’
phase, and clearing away after singing. The 60 minute singing session commenced with 20
minutes of relaxation, posture, breathing and vocal exercises followed by 40 minutes singing. A
wide common repertoire of familiar and new songs was available in a high quality song book.
Participants also steered the musical direction of their group according to their interests. Keeping
the programme fresh, enjoyable, stimulating and stretching is essential for a project planned
to run over the course of ten months. Songs were taught by ear and were sung mainly without
accompaniment (Robb et al. 2011).
Jane Petto and Sadie Hurley
Leading the Ramsgate group

Sidney De Haan
Research Centre for Arts and Health
10
Quantitative results
Participants involved in the project and findings from the measures employed.
Recruitment
Of the 145 GP practices contacted in June 2011 in East Kent by the South East Primary Care Research
Network (PCRN-SE), only six practices sent out letters to 499 patients, even though practices were
offered payment to do so by the PCRN-SE. Of the 106 patients recruited via all routes, 41 were
registered with one of the six practices (this would represent an 8 per cent response rate if all of
those registered with these practices volunteered on the basis of receiving a letter). The rest (65) were
recruited through additional routes: the East Kent Community Respiratory Team, newspaper advertising,
and direct contact with the three British Lung Foundation Breathe Easy Groups (BLF BEGs) in East Kent.
Sample
Of the 126 people who volunteered to participate in the study, 121 attended for baseline assessment
during which questionnaires were completed and standard spirometry administered. Fifteen (12.3 per
cent) volunteers were found not to meet the inclusion criteria for COPD, and were excluded from
the study, but not from participating in a singing group. The sample of 106 participants with COPD
varied in the stage of their COPD with 15 per cent mild, 45 per cent moderate, 30 per cent severe
and 10 per cent very severe. The mean age of the sample was 69.5 (SD 7.64) with a third being
male. The majority of participants were retired (75.1 per cent), with 14 (13.5 per cent) who retired
due to the effects of COPD. The majority of the sample were previous smokers (69.5 per cent);
11.4 per cent currently smoked, and 19.0 per cent had never smoked. Ninety-nine percent considered
themselves white, and 51.4 per cent had continued in education, with over a third holding a degree
or equivalent qualification. The majority had a joint income with partner of less than £20k (83.0 per
cent), and 36.2 per cent less than £10k jointly. Over the course of the study 34 (32.1 per cent)
participants withdrew because of competing commitments and health problems. In only three cases,
however, were the health issues related to COPD.
A network of six singing groups for patients with COPD was successfully established and maintained
over the period of ten months. Three of the singing groups formed were fairly large, with an average
of 26 members, and three were smaller, with an average of 9 members. The larger groups were in
areas where a Breathe Easy support group operated.
Assessments of lung function
Table 1 reports the results from the spirometry undertaken at baseline and the end of the singing
programme. Significant improvements were found for FEV1 per cent (Mean change 1.94; 95% CI 0.58
to 3.30; p=0.006), FVC (0.11; 0.01 to 0.20; p=0.027) and FVC per cent (3.63; 0.28 to 6.98; p=0.034).
Table 1: Measures of lung function at baseline and end of programme
Measure nBaseline End of
programme Mean difference
(95% CI) p value
FEV166 1.29 (0.49) 1.32 (0.51) 0.03 (-0.01; 0.58) 0.094
FEV1% predicted 67 54.34 (20.45) 56.28 (21.98) 1.94 (0.58; 3.30) 0.006
FVC 64 2.43 (0.75) 2.54 (0.075) 0.11 (0.01; 0.20) 0.027
FVC% predicted 65 81.72 (22.60) 85.35 (21.70) 3.63 (0.28; 6.98) 0.034

11
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Self-assessed health
Participants in the study completed a battery of questionnaires at baseline, mid-point in the study
and at the end of the programme. No significant changes were found between baseline and mid-
point assessments on any measure. Table 2 reports the results from the St George’s Respiratory
Questionnaire, MRC breathlessness scale, EQ-5D and SF-12 for baseline and the final assessment.
Significant improvements emerged for the SGRQ total (-3.29; -6.14 to -0.45; p=0.024) and impacts
scores (-3.45; -6.77 to -0.13; p=0.042) between baseline and the end of the programme. No significant
changes were found, however, for breathlessness assessed by the MRC scale, nor for generic measures
of mental and physical health-related quality of life as measured by York SF-12 and the EQ-5D.
Table 2: Measures of self-assessed health at baseline and end of programme
Measure nBaseline End of
programme Mean difference
(95% CI) p value
SGRQ total 71 48.71 (16.95) 45.42 (16.96) -3.29 (-6.14; -0.45) 0.024
SGRQ symptoms 71 59.16 (23.49) 56.04 (22.05) -3.13 (-7.35; 1.08) 0.14 3
SGRQ activities 71 65.46 (22.41) 63 . 33 (2 2.14) -2.13 (-5.4 4; 1.18 ) 0.204
SGRQ impact 70 35 . 6 5 ( 17. 56) 32.21 (15.90) -3.45 (-6.77; -0.13) 0.042
MRC dyspnoea 68 2.68 (0.98) 2.5 4 (1.03) -0.13 (-0.34; 0.08) 0.210
EQ-5D utility score 65 0.71 (0.22) 0.75 (0.22) 0.04 (-0.01; 0.08) 0.15 2
EQ-5D VAS 65 66 . 61 (17.96 ) 68.86 (18.99) 3. 24 (-1.2; 7.68 ) 0.15 0
SF -12 m e ntal 65 53.48 (9.87) 54.99 (9.06) 1.50 (-0.90; 3.91) 0.216
SF-12 physical 65 28.91 (7.98) 28.82 (7.87) 0.09 (-1.14; 1.33) 0.882
Ramsgate group members warming up their facial muscles

Sidney De Haan
Research Centre for Arts and Health
12
Quantitative discussion
Recruitment
It was possible to recruit over a hundred COPD patients to a community singing project. However,
participation by GP practices in this process was limited, even though they were offered payment to do
so. This has recruitment implications for future larger-scale controlled studies on singing and COPD, and
the ability to build an evidence base. The six practices that took part varied from large practices with
dedicated research support, to single GP practices with minimal administration support. Discussions
with Practice Managers revealed that in large Practices, it could take up to 3 months to gain agreement
from GPs to take part in research. This issue will be built into future study design timelines, including
more attention to involving GPs in the design and recruitment processes initially. This will also be
useful in explaining to GPs the expected benefits of the study to them, and gaining their support. Also
influencing recruitment was the presence of a BLF BEG in the geographic area of the larger groups. The
influence of the three BLF BEGs was due to the encouragement possible by presenting the project to
the members at their monthly meeting, and the consequent peer support received. East Kent has an
excellent Community Respiratory Team, and they informed patients on their lists, and also those taking
part in pulmonary rehabilitation (PR). Members from the BEGs visit the PR Groups to promote the
local BEG, and they also promoted the singing groups as well. Advertising in the local papers was also
successful for harder to reach people, but was relatively expensive on a per head basis.
Attrition
Over the course of the study a total of 34 participants left the study and did not complete the final
questionnaire (32.1 per cent attrition). A further six participants were unable to attend for the final
spirometry assessments due to other commitments and illness (37.7 per cent attrition). The rate of loss
to the study is less than the conservative estimate of 50% made in designing the study and considering
the length of the study compares well with attrition rates in previous studies of singing and COPD
(Bonilha et al, 2008, 30 per cent over 24 weeks; Lord et al, 2010, 22 per cent over six weeks; Lord et
al, 2012, 19 per cent over eight weeks). Comparisons were made between those dropping out and
those remaining in the study and no significant differences emerged on any measure taken at baseline,
so attrition did not introduce bias into the study. In addition, only a small number of withdrawals
(2.8 per cent) were due specifically to COPD related health issues, and only a small proportion of missed
attendances during the programme were due to COPD related health issues (1.5 per cent). This indicates
that participants seemed to have stayed generally quite well and active whilst participating in weekly
singing. Qualitative evidence also indicates that some participants had not been as ill as in previous
years during the particularly harsh winter of 2011, when GPs were warned by the UK Department of
Health to expect increased illness rates in COPD patients (DH, 2011).
Members of the Whitstable group enjoying a session

13
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Improvements in lung function
The positive improvements in the standard spirometry measures are encouraging findings,
especially given that a decline in these measures might have been expected as COPD is a
deteriorating disease. While the improvement for FEV1 was 30ml and not statistically significant,
Dewar and Curry (2006) suggest that after 25 years of age, a non-smoking adult’s FEV1 declines
each year by an average of 20 – 40ml and that in susceptible smokers this decline can be as much
as two to five times greater. Recent evidence suggests that improvements of 120ml or more in FEV1
are associated with clinically significant improvements in perceived health status measured by the
SGRQ (Jones et al, 2011; Jones, personal communication, 2012). In the present study seven people
showed deterioration of 120ml or more over the course of the study, and 16 people showed an
improvement of at least 120ml beyond baseline.
While the FEV1 change is not statistically significant, FEV1 expressed as a percentage of expected
values (taking account of age, gender, body-mass index and ethnicity) does show a clearly significant
improvement of almost 2 per cent. Interestingly, both FVC and FVC per cent have also improved. As
FEV1 cannot exceed FVC, and increased FVC can be regarded as an indicator of reduced gas-trapping
(Macklem 2010), the changes in FVC seen may be of greater clinical and functional significance than
changes in FEV1.
St George’s Respiratory Questionnaire
The finding of an improvement in the total SGRQ score was encouraging. While the change of 3.3 is
less than what is considered a clinically important change value of 4, the change was statistically
significant. Over time a rise in SGRQ of approximately two points annually is to be expected given
that COPD is a progressive illness (Jones, personal communication, 2012). The improvement in SGRQ
is also consistent with the improvements found in the lung function measures (Jones et al, 2011).
MRC Dyspnoea Scale
The mean MRC rating showed some slight improvement, but the change was not significant. The
correlations between the MRC scale and SGRQ total at baseline and end of programme were 0.68
and 0.54 respectively. While positive and significant, these correlations are moderate. Breathlessness
is part of a more comprehensive self-assessment given by the SGRQ and appears to be more sensitive
to change than the MRC scale.
York SF-12 Physical and Mental Wellbeing
SF-12 scores suggested mental health was similar to the population norm, but physical health was
poor, and no change was found in either over the period of the study. The lack of change in the
physical component of the York SF-12 contrasts with the positive changes found by Lord et al. (2012)
in the physical health component assessed by the SF-36. Previous studies on singing have shown
improvements in mental health and wellbeing for mental health service users (Clift and Morrison,
2011), and for older people over 60 years old (Skingley et al. 2011; Clift et al. 2012). However, mental
health was quite good amongst the current sample of COPD patients, and conversations with
participants when they were attending to complete questionnaires, revealed they were in good spirits
and had adjusted to their condition. This may explain the lack of change in their mental health status.
EQ-5D
As with the York SF-12, measures from the EQ-5D did not show significant change over the course
of the study. While EQ-5D utility and VAS scores can differentiate groups of patients with severe and
very severe COPD, (Rutten-van Mölken et al. 2006), they appeared to lack sensitivity in the current
study to detect changes in response to the singing intervention.

Sidney De Haan
Research Centre for Arts and Health
14
Study limitations
The current study was designed to assess the feasibility and acceptability of establishing and
running community singing groups for people with COPD over the course of almost a year, and
to gather data on changes which would allow us to explore the potential effect for consideration
for a subsequent community-based randomised controlled trial. It was clear from the study that
the singing intervention was feasible and acceptable to the population and positive effects were
observed on a number of important clinical measures of COPD including condition specific quality
of life. In the current study participants volunteered on the basis of information received through
a variety of channels and expressed an explicit preference to engage in singing groups, and while
this may not be representative of the COPD population as a whole it was representative of those
who would be willing and able to engage in group singing.
Implications
The improvement in FEV1 per cent, FVC and FVC per cent, and SGRQ suggests that the singing
programme followed, including the attention paid to posture and breathing technique, may have an
exercise training effect on lung function. Lung function is an important aspect of health, ensuring
efficient evacuation of the lungs and providing the oxygen supply to the circulation system. As
singing can be performed sitting down, it is suitable for almost anybody. This enables a training
regimen to be individually graded both by varying the position of the participant, e.g. sitting,
standing, or walking around, and the exposure to the graded singing delivery of the vocal exercises
and song difficulty. Therefore it could be classed as a form of moderate cardiovascular exercise
suitable for all. This aspect is being discussed with the Community Respiratory Team in East Kent
with a view to integrating some singing within PR provision, but also as an option for those not able
to take up PR, for those on the waiting list for PR, and also as a post-PR maintenance activity.
Trish Vella-Burrows and Phil Self
Leading the Deal group

15
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Quantitative conclusions
The improvement in FEV1 per cent, FVC and FVC per cent indicates that community singing
(including attention to posture and breathing techniques) can have an exercise training effect on
the lung function of people with mild to very severe COPD. Singing could be considered as a form
of moderate cardiovascular exercise for this group of chronically ill patients and as such is worthy
of attention from Health Service Pulmonary Rehabilitation Teams, Respiratory Nurses, Health
Promotion Services, and voluntary organisations working to support people with COPD in the
community.
The reported study suggests the need for a larger-scale controlled study on community singing
and COPD. A further planned study will pay more attention to processes of recruitment into
the trial to assess the factors associated with willingness to participate in singing groups
among people with COPD.
Regular group singing is an innovative initiative to help people with COPD engage in physical and
social activity to support independence and quality of life.
Jane Petto and Julia Baldwin
Leading the Ashford group in a vocal exercise

Sidney De Haan
Research Centre for Arts and Health
16
Qualitative findings
Participants’ experiences of the singing groups and benefits gained.
Collecting feedback from participants
As noted above, participants had the opportunity to write comments on each of the
questionnaires completed about their experience of the project and give feedback on their health.
Such information provides valuable insights into how people experienced the singing groups and
the benefits they felt they gained from participation.
Data analysis
All written comments were transcribed and imported into NVivo9® (QSR International 2010),
a qualitative data analysis programme. Analysis proceeded as a two-stage operation. Initially,
following familiarisation of the content through reading the whole text, the complete data set
was subjected to a broad content analysis, to allow reduction of data into categories linked to
simple quantification (Breakwell et al. 2006). This was guided by an analysis of word frequencies
as generated by the software, when particular terms were highlighted. For example, it was
anticipated that participants would comment on breathing, as all had respiratory problems, so
we ran a text search query for ‘breathing’ and similar matches through the use of a wildcard
(breath*) to indicate the addition of any other characters (e.g. ‘breathe’ ‘breathing’). Breathing
then became one category (or ‘node’ as it is termed in NVivo). Becoming familiar with the text
allowed the initial identification of a further four major categories: physical health (other than
breathing); psychological health; social wellbeing; and comments on the project and singing
programme. This was achieved by two researchers working independently and then resolving any
differences emerging, all of which were minor.
Following this, the categories were overlaid with a thematic analysis, which involved a more
interpretive approach to the data through careful examination of the language used. A theme
may be defined as a data extract (an individual chunk of coded data) which captures something
important in relation to the overall research question (Braun and Clarke 2006). The questions
in this qualitative part of the study related to assessing acceptability and effectiveness in
the broadest sense, of the singing programme and the research procedures as perceived by
participants. Therefore we interrogated the comments for qualitative statements (positive
or negative indicators of acceptability); for terms indicative of attribution of any change
in wellbeing to the singing intervention (indicators of effectiveness); and also for whether
perceptions changed over time, that is, over the three administrations of the questionnaire
(baseline, mid-study, end of study). A visual representation of how this was conceptualised is
provided in Figure 1. A quantitative element was retained for this stage of the analysis, since
we felt it was useful to explore any change in terms of the number of data extracts expressing
positive, negative or attribution comments over the course of the research intervention. Data
extracts were ‘counted’ if they pertained to a comment by the same individual at different data
collection periods, but not to more than one comment repeated within the same questionnaire.

END OF PROJECT
DEFINITE ATTRIBUTION
UNCERTAIN ATTRIBUTION
POSITIVE STATEMENT
NEGATIVE STATEMENT
BASELINE
QUALITATIVE
COMMENTS
MID TERM
CATEGORY
CHANGE
OVER TIME
STRENGTH OF
ATTRIBUTION
17
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Comments were received from 96 individuals (total sample = 106) over the course of the project,
with 66 at baseline, 77 at mid-study follow-up and 73 at final follow-up. As expected, baseline
comments were very much briefer than the later ones. Presentation of findings based on
these comments is guided by the broad themes as described above, and, within them, where
relevant, the dimensions of quality (positive-negative), strength of any attribution of change to
intervention, and timing.
Figure 1: Conceptual framework for data analysis

Sidney De Haan
Research Centre for Arts and Health
18
Breathing
Baseline spirometry measures confirmed that participants were diagnosed with COPD (Mild 15%,
Moderate 45%, Severe 30%, Very Severe 10%). Overall the data generated 97 extracts related to
breathing. General comments relating to on-going respiratory conditions were overwhelmingly
negatively framed and broadly distributed across all three time bands. These provided a fuller
picture of additional respiratory diagnoses (asthma, chest infection, pneumonia, bronchiectasis –
7 individuals) as well as how participants experienced their respiratory limitations, in terms of being
out of breath or wheezy, having chest problems, having poor lung power, drug side-effects or having
to use medication pumps more frequently:
“ I have noticed that my breathing is poor when walking and talking simultaneously.
A chest infection leaves me ‘voiceless’, the longest being 9½ weeks. I am under the
care of a chest consultant and have bronchiectasis.”
Female, group 2, age 76
A few individuals commented in more positive terms on their existing health. For example two
participants noted that their current medication was controlling their condition. However, the
majority of breathing-related comments from the second and third data collection periods concerned
improvements noted by participants since the singing groups started. Although most just stated that
breathing had improved, many individuals were able to identify particular mechanisms through which
these were achieved (see below).
Members of the Ramsgate group engaged in a breathing exercise
Mechanisms for achieving health improvements
• Promotes learning for breathing properly (including breath control, techniques for daily
activities, muscle control, understanding, monitoring and awareness of breathing)
• Improves posture
• Promotes relaxation
• Helps concentration/provides distraction
• Provides a good workout/more energy
• Opens lungs/increases lung capacity
• Makes physiotherapy easier
• Helps prevent panic/hyperventilation

19
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Moreover, a large number of the comments (64) expressed strong beliefs that the singing was the
reason for improvement noted:
“ I believe that the project is teaching me how to understand my breathing and how to
control it. This is very useful; it stops me hyperventilating when my breathing is under
pressure i.e. climbing a steep hill.”
Female, group 5, age 74
In two cases there was an inference that the intervention had led to reduced uptake of health
services:
“Since first time in joining the singing group I have not had to spend time in casualty
this winter or spring for COPD.”
Female, group 1, age 64
Looking at the distribution of the comments on breathing (26 at second data collection; 37 at third
data collection) it may be that participants experienced incremental gain in respiratory health over
the intervention period. Fourteen individuals commented at both of these time periods and, while
some simply reiterated statements made earlier, in others the strength of attribution appears from
the language to have increased. For example:
Female, group 2, age 71:
“ I feel that I will benefit from all the singing and breathing techniques that I have
been Taught.” (Data collection time 2).
“ I feel that the benefits I have had from the course has been great, my breathing
has improved no end.” (Data collection time 3).
Three participants made anticipatory statements in the baseline questionnaire, expressing hope that
singing would improve their COPD, looking forward to seeing improvement and in one case being
‘convinced’ that the singing would bring benefit. This suggests that some of our sample embarked
on the project with certain expectations, though the extent to which this might have affected
later responses is unknown as, too, is the number of other individuals who may have held, but not
expressed, similar views.
Not all participants were so convinced of the effectiveness of the intervention. Twelve statements
(mostly from time 2) referred to being unsure of benefits, as yet not able to judge, not noticing vast
improvement, maintaining an open mind, effectiveness being hard to evaluate, noting change may
be a coincidence or not feeling worse. For three of these individuals later comments at collection
time 3 indicated a more certain attribution:
Female, group 5, age 77:
“ As yet, any health/breathing improvement has not been apparent.”
(Data collection time 2).
“ It certainly appears to have helped with my general breathing.”
(Data collection time 3)

Sidney De Haan
Research Centre for Arts and Health
20
Physical health
A number of individuals (n=26) commented on their existing physical health status, supplementing
the baseline standardised research measures and providing a detailed picture of how this affected
their everyday lives. These were expressed in largely negative terms, indicating that, for many, co-
morbidities existed alongside their primary diagnosis. Physical health issues related to poor mobility,
general tiredness and sleep problems, episodes of flu, voice problems, pain, cancer, stroke and
general ‘poor health’. Some of these health problems, often unrelated to COPD, kept people away
from the singing groups:
“ Due to flu, ‘maybe Asian flu’ I have had to have 4 – 6 weeks away so I am still
suffering the ‘dregs’ of the infection...”
Female, group 6, age 66
A large number of comments also reported improvements to physical health (other than breathing)
at either mid-term (n=5 ) or end point (n=11). These included comments related to more positive
energy levels, vocal capacity, mobility, physiotherapy being easier post-singing and general physical
health improvement, and most of these improvements were explicitly ascribed to the singing
intervention:
“ I love coming to COPD Singing Research Project and I always feel much better
physically and emotionally afterwards.”
Female, group 5, age 70
All those commenting on the benefits to physical health appeared to have no doubt that it was the
singing that had given rise to this.
Ashford group members greeting one another at the start of a session

21
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Psychological health
A few participants commented on pre-existing psychological health issues in terms of anxiety, feeling
down and depressed and panic attacks. While for some this was unattributed, a focus among others
was on the psychological sequelae of COPD:
“ You need to understand that being severely out of breath ALL THE TIME can be
so emotionally draining, you cannot do anything energetic. You need to stay away
from any environment where the air is less than really fresh, you have to think about
everything. Fit and healthy people don’t understand, so you can’t talk to them.”
Male, group 5, age 58
For a few individuals, mental health issues were unconnected to their respiratory diagnosis:
“ I suffer from emotional and slight depression. This is due mainly because my son died
4 years ago for no obvious reason.”
Male, group 4, age 56
“ My husband is ill at the present time so I am a little preoccupied and probably not
quite as cheerful as per normal.”
Female, group 3, age 77
The fact that these comments were made across all data collection periods suggests that certain
states of health were not amenable to improvement over the timescale of the research. However,
much more numerous were comments relating to areas where psychological health was perceived
to have improved as a result of the singing groups. Ninety-one data extracts alone referred to
‘enjoyment’ or a derivative of the term, and a further 18 to ‘fun’. Notably, this sense of enjoyment
was maintained throughout the whole duration of the project and was even mentioned by a few
individuals in relation to the taster sessions, so may have served as a contributor to the retention rate
within the groups.
Of greater importance for understanding the mechanisms through which psychological health was
felt to improve were more specific comments than those relating to enjoyment alone. The range of
areas referred to is illustrated below in order of frequency.
Areas of perceived benefits to psychological health
• Lifts spirits (feel uplifted, contributes to spiritual health)
• Promotes general psychological/mental/emotional wellbeing/feeling better/therapeutic
• Boosts confidence/provides sense of achievement and pride
• Provides a feel-good factor/adrenaline buzz
• Provides a purpose in life/reason to get out of the house/something to look forward to
• Helps relaxation
• Promotes a positive attitude/feeling upbeat/counteracts feeling low
• Helps coping/dealing with illness
• Reduces anxiety and depression
• Encourages self-help

Sidney De Haan
Research Centre for Arts and Health
22
Once again, comments were incremental over the three data collection points, adding credence to
the suggestion that a longer exposure to the singing groups led to greater perception of benefit.
Once again, too, baseline comments revealed a positive expectation on the part of some individuals:
“ I have recently lost my husband and therefore have been very low and run down. I am
sure that this course will help me and I shall benefit as we progress.”
Female, group 2, age 71
In other respects, comments on psychological health differed from those related to breathing, in
the absence of any doubt that the singing groups were responsible for improvements. This may be
because there was a pre-existing suggestion (given that COPD was a major focus) that singing may
alleviate breathing problems, whereas there was no reference to potential psychological benefits
(therefore no reason to comment on lack of effect). However, looking at those who responded at
different times in relation to psychological health illustrates a confirmation of on-going benefit and
sometimes a marked tendency for greater detail of particular improvements to emerge:
Female, group 3, age 64:
“ I am looking forward to seeing an improvement with both my breathing and wellbeing.”
Baseline feedback
“ I always feel better after the singing session.”
Midpoint feedback
“ I now have 3 friends as supporters and they also enjoy singing. One friend has
dementia and he really ‘comes alive’ during the session.”
Final feedback
This final quotation provides a clue to the popularity of the groups in that the focus is not just on the
individual and his/her health status but on the fact that the singing took place in a social context. This
was remarked upon by a large number of participants.
Phil Self
Leading the Deal group

23
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
Social wellbeing
In total, 110 data extracts made reference to the social aspects of the singing groups and these
were universally positive in nature. This exceeded the number of comments related to breathing and
content broadly pertained to either friendship and company generally, or to the support gained from
meeting those similarly experiencing COPD. Distribution across data collection points suggested some
expectation at baseline (n=7) moving to establishment and sustaining of social interaction at mid and
end-points (n=50 and 53 respectively). This relative stability from time 2 to 3 suggests that the same
friendships may have been maintained over the course of the intervention period.
Looking more closely at content, the most frequently used word was ‘friend’ (or associated
words). This was followed closely by references to the social side, the company, camaraderie, the
group/’bunch’ or meeting other people:
“...the friendship, team spirit etc. is wonderful ...personally I have benefited (sic) from
seeing friends, which makes me feel cheery.”
Female, group 4, age 62
“...as I am retired, the social ‘get together’ has been wonderful.”
Male, group 5, age 77
Other comments related the social/group side to the singing itself, indicating the importance of the
fact that, through this, they were becoming part of something to be proud of:
“ ...group has become very friendly and we seem to becoming a ‘choir’ under
excellent guidance.”
Male, group 5, age 68
A total of 33 data extracts specifically mentioned the social/group aspect vis-à-vis the unifying factor
of a COPD diagnosis. Again comments were all positive and spread across the timescale, including
some expectation at baseline:
“ would be good to meet with other people to see how they cope with people that
think you are alright.”
Female, group 4, age 70
Benefits of meeting others with the same diagnosis included: sharing and discussing experiences and
learning, being with people who understand, feeling someone cares, being able to care for each
other, mutual support and bonding, receiving reassurance, meeting people in a worse condition, and
feeling less isolated:
“ COPD is socially isolating and the singing class has allowed me to share experiences
about my condition and pick up tips from others on how to cope. There’s always a
welcoming comforting atmosphere at singing, it’s the one place I don’t feel unusual or
different from other folk and you don’t get comments or looks from those around you
if you experience discomfort as they understand your condition and its limitations.”
Male, group 3, age 70

Sidney De Haan
Research Centre for Arts and Health
24
• Encouraging,
‘can do’ attitude
• Enthusiasm
• Motivational/inspirational
• Sense of humour/fun
• Caring, understanding,
kind, helpful
• Sociable, good rapport,
friendly
• Welcoming
• Dedicated
• Good leadership
• Knowledgeable
• Professional and
competent
• Able to maintain interest
• ‘No pressure’
Comments on the project and singing programme
Participants were explicitly invited to comment on positive experiences during the project as well
as areas which could be improved. Most chose to do so, with roughly equal distribution across all
three data collection points but with positive comments outweighing others in a ratio of five to one.
Comments related to the following areas: facilitation and leading; organisation and administration;
the topic of the research; the programme and content of the sessions; the venue and environment;
and the ending of the project and future plans.
While a number of comments related to facilitation simply used words such as ‘brilliant’, ‘superb’ or
‘excellent’, others provided an indication of the qualities appreciated by participants in running the
singing groups (see below).
Facilitator qualities valued by participants
Predictably the competence and knowledge of facilitators were seen as important, however equal, if
not greater value was put on social and emotional intelligence:
“ The ‘facilitators’ (sic) ... were excellent. They made sessions light-hearted as well as
instructive. In particular they did well to encourage folk like me, who hadn’t sung
since he was 8 years old except at church services, to overcome a natural reluctance
to dare to make a noise.”
Male, group 1, age 79
The majority of comments on administration and organisation appeared within the baseline
questionnaires and related to the recruitment procedures (quality of instructions, the questionnaire
itself or contacting the Centre). Three quarters of these comments were very positive:
“ not having attended any preliminary sessions, I am impressed by service when I
phoned for information at almost the last day before a singing session. Joining
instructions were concise and complete.”
Male, group 1, age 79
A small minority made less favourable comments, finding communication with the Research Centre
problematical, the questions in the questionnaire difficult, too many forms to complete or insufficient
information provided.
Some individuals commented on the general topic of the research. Not surprisingly, because they
had volunteered to take part, these comments were very supportive. Although a number just noted
finding the subject interesting or intriguing, others expressed a hope that some benefit would come
out of the project, either personally, or more generally in terms of medical research. This sense of
altruism, which has already been noted in previous research literature (Robinson et al., 2005) is worth
emphasising in recruitment strategies:
“ the more research into lung disease and also publicity will help people of the future generation.”
Female, group 4, age 66

25
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
With regard to the singing programme, there was much approval of the range of songs with both old
and new items introduced, and only a few individuals commented negatively in terms of not liking
songs which were sung at school or not coping with the pitch (too high for one, too low for another).
There was a similar disagreement about the warm-up exercises, with some specifically liking these
and some disliking them, with a corresponding split over the emphasis on breathing. Other areas of
approval were the structure of the programme, the inclusion of harmonising and the Christmas party!
Other suggestions included wanting longer sessions or the whole project run continuously rather than
in terms, while a few felt that the singing of Christmas carols began too early.
Positive comments on the environment referred to a welcoming, relaxing or friendly atmosphere and
the quality of the venue with ample parking. However four of the venues attracted comments that
they felt cold in the winter months and for three venues individuals drew attention to poor audibility.
Other comments concerned the distance travelled to the venue and a problem with parking.
Finally, in data collection times 2 and (increasingly) 3, some 40 comments referred to the
approaching ending of the research project. Some simply stated that they would be sorry,
disappointed or sadly miss the singing, but one individual wrote more negatively about being
“left out in the cold at the very end with no director or future, felt dumped” (Female, group 1,
age 64). More numerous, however, were those expressing hope that the singing groups could
continue, some volunteering to contribute to this end, either financially or in other ways:
“ I have come to regard the social get together and singing as an important part of my
life, which in other circumstances I wouldn’t have got involved in and I intend to help
in any way to keep our ‘choir’ going after the end of the project.”
Male, group 5, age 77
The ultimate end for many of those expressing such hope was that the findings of the research would
be positive and so be disseminated sufficiently widely to effect a change in the management of
people with COPD like themselves:
“ I wish the project would carry on as it has been a great help to me. I hope in future all
doctors and respiratory nurses will find a way of getting people with lung problems
to start another project in singing.”
Female, group 4, age 66
The Ashford group engaged in a physical stretching exercise

Sidney De Haan
Research Centre for Arts and Health
26
Qualitative discussion
This section of the report presents the analysis of the written comments from 97 individuals
participating in singing groups for people with COPD. Wording on the questionnaires was purposely
broad, asking about experiences of participation in the project, but with two sections steering
respondents to making positive or negative points. The main contributions from this phase of the
research were:
• To provide a more comprehensive picture of the overall health profile of the sample.
• To generate complementary, as well as confirmatory, evidence to supplement the quantitative
element of the study.
• To contribute to the evidence base already existing in this area in terms of previous research.
Participants were included in the project by virtue of their diagnosis of COPD, a long-term,
progressive and potentially disabling condition. However, as the literature suggests (Maurer et al.
2008), both physical and psychological co-morbidities were found to be present at baseline in a
number of participants who chose to share this information with researchers. While, for some, this
entailed concomitant respiratory diagnoses, the list of (largely unrelated) physical conditions was
striking. In addition and also linked by some to the COPD, there were examples of expressed anxiety
and depression. Overall, a picture of a somewhat diverse sample emerged with regard to pre-
existing health; the general profile of good mental health demonstrated in the quantitative data was
obviously subject to exceptions as demonstrated in participant perceptions, while evidence of overall
poor physical health was supported and well illustrated.
Where there was an overlap in content matter, qualitative data broadly supported that derived from
the structured measures. Spirometry showed a mean significant improvement in forced expiratory
volume in one second FEV1 and forced vital capacity FVC as a percentage of expected values at final
follow-up, along with significant improvement in the St George’s Respiratory Questionnaire (SGRQ)
scores, although minimal change was found on the SGRQ at midpoint. Participant comments,
however, suggested that improvements in respiratory symptoms were clearly experienced by the
midpoint data collection date and, increasingly, by the endpoint. They also provided evidence of
how this improvement was experienced, therefore yielding a more nuanced picture of the impact of
singing on COPD than that derived from quantitative data alone.
No change was detected in the generic mental or physical components of quality of life from the
quantitative analysis (SF-12) at the end of the study. This contrasted with the benefits, especially in
mental health, being experienced by participants and a clear attribution to the singing programme as
the cause. It may be that the instruments used were insufficiently sensitive to detect this experienced
change in health status or it may be, bearing in mind that some individuals expressed difficulty in
completing the questionnaires, that the limited response format of the instruments failed to capture
what they wished to convey. There is also the finding that, on the mental health component, scores
were close to the population mean at baseline. For the physical health component, in contrast, mean
scores were low and the lack of change may indicate that the intervention, while positive in specific
ways, was not sufficient to improve general physical health status.
One area not well reflected in the quantitative evidence related to the social benefits which
participants experienced as a result of the singing groups. This appeared as an important and
sustaining motivation for continuing to attend and a reason for the enjoyment expressed by many,
with comments outnumbering those related to breathing. This benefit is, perhaps, less readily
measurable by research instruments but clearly relates to coping with an otherwise isolating condition.
The evidence from participant comments in our study adds to the research base on singing for

27
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
breathing. Improvements in breathing, in physical health more generally and in mental wellbeing,
experienced by our sample as a result of singing, is supported in the responses to interviews conducted
in the two studies by Lord et al. (2010; 2012). In addition, the Lord studies also reported social benefits
and support from the group nature of the intervention, adding credence to our findings.
The varying degrees of certainty with which our respondents attributed improvements in physical
health and breathing to singing is echoed in a study conducted with choral singers (Clift et al.
2009). In that study, where recruitment was not based on any physical condition, a ‘tentativeness-
certainty’ spectrum of attribution was created and it was noted that those experiencing acute or
chronic health conditions (especially affecting breathing) were more likely to be convinced that
singing had been beneficial.
Sonia Page
Leading members of the Folkestone pilot group performing for World COPD Day 2011

Sidney De Haan
Research Centre for Arts and Health
28
Qualitative conclusions
The qualitative analysis of written comments reported here aimed to explore the feasibility,
acceptability and effectiveness of singing for breathing as experienced by people with COPD. The
majority of participants chose to write comments to supplement quantitative information resulting
in a broader picture of individuals’ pre-existing state of health and subjective accounts of any health
benefits accruing from the singing to supplement our quantitative measures. Findings suggest that
singing is perceived as both acceptable and beneficial to this group, not only for breathing but
also in relation to general physical, psychological and social wellbeing. Participants were able to
identify various mechanisms whereby benefits were accrued and also commented on aspects of the
programme which were more and less favoured. Such information is useful for the future planning of
such groups.
Limitations
The study took place in a limited geographical area in the south east of England, therefore is not
necessarily representative of the population of people with COPD. Results should therefore be
treated with caution when considering generalizability. Participants in the study were volunteers who
purposely chose to become involved and often with some expectation of benefit. This may limit the
validity of the findings. As with most qualitative studies, there is a potential for a social desirability
response bias – though this is less likely where, as here, data collection is anonymous, rather than
merely confidential (where the respondent known to the researcher as in an interview). Finally,
though the thematic analysis was based on a rigorous process, it necessarily involved a degree of
inference, which might have been minimised through additional measures such as member checking.
Implications for future research
Following on from this research we would suggest the inclusion of qualitative/experiential data
in studies investigating the effects of singing on wellbeing, since this has the potential to pick up
complementary areas of impact (such as the social element). There is also a need to build on existing
knowledge through a large or multi-site randomised controlled trial to identify more reliably whether
significant changes result from the singing intervention, to support external validity and to enable
a cost effectiveness calculation. Lastly, research may benefit from recruitment through a system of
referral from healthcare staff, rather than relying solely on volunteers, who may come with certain
expectations of outcome from the study and therefore lead to some bias in reporting.
Members of the Ramsgate group engaged in a laughter exercise

29
An evaluation of community singing for people with COPD (Chronic Obstructive Pulmonary Disease) | Final Repor t
A participant in the project with portable oxygen
Sonia Page
Preparing the Whitstable group to sing

Sidney De Haan
Research Centre for Arts and Health
30
References
Bestall, J.C., Paul, E.A., Garrod, R. et al. (1999)
Usefulness of the Medical Research Council (MRC)
dyspnoea scale as a measure of disability in patients
with chronic obstructive pulmonary disease. Thorax,
54, 581-586.
Bonilha, A.G., Onofre, F., Vieira, L. M., Prado M.Y.
and Martinez J.A (2008) Effects of singing Classes on
pulmonary function and quality of life of COPD patients.
International Journal of COPD, 4(1), 1-8.
Braun, V. and Clarke, V. (2006) Using thematic analysis in
psychology. Qualitative Research in Psychology 3(2), 77-101.
Breakwell, G.M., Hammond, S., Fife -Shaw, C. and
Smith, J.A. (2006) Research methods on Psychology
(3rd edition). London: Sage.
Clift, S. and Hancox, G. (2001) The Perceived benefits
of singing: findings from preliminary surveys with a
university college choral society. Journal of the Royal
Society for the Promotion of Health, 121(4), 248-256.
Clift, S., Hancox, G., Morrison, I. et al. (2009) What
do singers say about the ef fects of choral singing on
physical health? Findings from a survey of choristers in
Australia, England and Germany, European Society for
the Cognitive Sciences of Music (ESCOM) Conference,
Jyvaskyla, Finland, 12-16 August, 2009. Available at:
https://jyx.jyu.fi/dspace/handle /123456789/20854
Clift, S., Hancox, G., Staricoff, R. and Whitmore, C.
(2008) A Systematic Mapping and Review of Non-
clinical Research on Singing and Health. Sidney De Haan
Research Centre for Arts and Health, Canterbur y Christ
Church University. Available at: www.canterbury.ac.uk/
research/centres/SDHR
Clift S, M. and Morrison I (2011) Group Singing fosters
mental health and wellbeing findings from the East Kent
‘singing for health’ network project Mental Health and
Social Inclusion 15(2), 88-97.
Clift, S., Skingley, A, Coulton, S. and Rodriguez, J.
(2012) A controlled evaluation of the health benefits of
a participative community singing programme for older
people (Silver Song Clubs). Available at:
www.canterbury.ac.uk/research/centres/SDHR
Daudey, L., Peters, J.B., Molema, J. et al., (2010) Health
status in COPD cannot be measured by the St George’s
Respiratory Questionnaire alone: an evaluation of the
underlying concepts of this questionnaire, Respiratory
Research, 11, 98. Available at: http://respiratory-
research.com/content/11/1/98
Dewar, M. and Curry, R.W. (2006) Chronic obstructive
pulmonary disease: diagnostic considerations, American
Academy of Family Physicians, 73(4), 669-676.
DH (2011) Making the Case: Why cold weather
planning is essential to health and well-being, London:
Department of Health.
Engen, R. (2005) The singer’s breath: Implications for
treatment of persons with emphysema. Journal of Music
Therapy, 42, 20-48.
Euroqol Group (1990) Euroqol – a new facility for the
measurement of health-related quality of life. Health
Policy, 16, 199-208.
GOLD (2010) Spirometry for Health Care Providers.
Global Initiative for Chronic Obstructive Lung Disease.
Available from: www.goldcopd.com
Iglesias, C.P., Birks, Y.F. and Torgerson, D.J. (2001)
Improving the measurement of quality of life in older
people: the York SF-12. Quarterly Journal of Medicine,
94, 695-698.
Jones, P.W. (2009) Health status and the spiral of
decline, COPD, 6, 59-63.
Jones, P.W., Anderson, J.A., Calverley, P.M.A. et al. (2011)
Health status in the TORCH study of COPD: treatment
efficacy and other determinants of change, Respiratory
Research, 12, 71. Available at: http://respiratory-
research.com/content/12/1/71
Lord, V.M., Cave, P., Hume, V. et al. (2010) Singing
teaching as a therapy for chronic respiratory disease –
randomised controlled trial and qualitative evaluation,
BMC Pulmonary Medicine, 10, 41. Available at:
www.biomedcentral.com/1471-2466/10/41
Lord, V.M., Hume, V.J., Kelly, J.L. et al. (2012) Singing
classes for chronic obstructive pulmonary disease: a
randomized controlled trial, BMC Pulmonary Medicine,
12, 69. Available at: w ww.biomedcentral.com/1471-
2466/12/69
Macklem, P.T. (2010) Therapeutic implications of the
pathophysiology of COPD, European Respiratory Journal,
35, 676- 680.
Maurer, J., Rebbepragada, V., Borson, S., Goldstein R.,
Kunik, M.E., Yohannes, S. and Hanania, N.A. (2008)
Anxiety and depression in COPD: current understanding,
unanswered questions, and research needs. Chest,
134(4 Suppl), 43S-56S.
Menn, P., Weber, N. and Holle, R. (2010) Health-related
quality of life in patients with severe COPD hospitalised
for exacerbations – comparing EQ-5D, SF-12 and SGRQ,
Health and Quality of Life Outcomes, 8, 39, Available at:
www.hqlo.com/content /8/1/39
Miller, M.R., Hankinson, J., Brusasco, V. et al. (2005)
Standardisation of spirometry, European Respiratory
Journal, 26, 319-338.

In September 2011, with funding from the Dunhill
Medical Trust, the Sidney De Haan Research Centre for
Arts and Health established a net work of six singing
groups for people with COPD in East Kent. Over
one hundred people with mild to very seve re COPD
participated in the study which ran until June 2012.
Careful monitoring of participants with lun g function
measures and standardised health questionnaires
revealed positive benefits from r egular group singing.
In the first of three short films about the res earch and its
findings, Sonia Page, Project Musical Direc tor, describes
approaches to facilitating singing for people with
breathing difficulties adopted in this stud y.
This DVD accompanies a report on the findings from
the singing and COPD feasibility study and a guide to
Singing and people with COPD.
Cover image (courtesy of doverdesig n.co.uk):
Sonia Page (right) leading members of the
Folkestone pilot group.
© Sidney De Haan Research Centre for Art s and Health
Canterbury Christ Church University, University Centr e Folkestone,
Mill Bay, Folkestone, Kent CT20 1JG Telephone: 01303 220 870
www.canterbur y.ac.uk/research /centres/SDHR
Sidney De Haan
Research Centre for Arts and Health
with Sonia Page, Project Musical Director
Facilitating singing
for people with
breathing dif ficulties
Singing and COPD
(Chronic Obstructive Pulmonary Disease)
In association with:
In September 2011, with funding from the Dunhill
Medical Trust, the Sidney De Haan Research Centre for
Arts and Health established a net work of six singing
groups for people with COPD in East Kent. Over
one hundred people with mild to very seve re COPD
participated in the study which ran until June 2012.
Careful monitoring of participants with lun g function
measures and standardised health questionnaires
revealed positive benefits from r egular group singing.
In the second of three short films about the res earch and
its findings, Ingrid Falcke, a member of the Folkestone
pilot group who is living with COPD, describes the
benefits she has gained from regular group singing.
This DVD accompanies a report on the findings from
the singing and COPD feasibility study and a guide to
Singing and people with COPD.
Cover image (courtesy of doverdesig n.co.uk):
Members of the Folkestone group with
Ingrid Falcke (far right).
© Sidney De Haan Research Centre for Art s and Health
Canterbury Christ Church University, University Centr e Folkestone,
Mill Bay, Folkestone, Kent CT20 1JG Telephone: 01303 220 870
www.canterbur y.ac.uk/research /centres/SDHR
Sidney De Haan
Research Centre for Arts and Health
with Ingrid Falcke, who is living with COPD
The benefits of
regular group singing
for breathing
Singing and COPD
(Chronic Obstructive Pulmonary Disease)
In association with:
In September 2011, with funding from the Dunhill
Medical Trust, the Sidney De Haan Research Centre for
Arts and Health established a net work of six singing
groups for people with COPD in East Kent. Over
one hundred people with mild to very seve re COPD
participated in the study which ran until June 2012.
Careful monitoring of participants with lun g function
measures and standardised health questionnaires
revealed positive benefits from r egular group singing.
In the third of three short films about the researc h and
its findings, researchers, facilitators and particip ants
etc, talk about the study and its findings. The fil m
provides a documentary record of how the group s
gained in confidence and came together to give public
performances.
This DVD accompanies a report on the findings from
the singing and COPD feasibility study and a guide to
Singing and people with COPD.
Cover image (courtesy of doverdesig n.co.uk):
Dr Ian Morrison, Principal Investigator, with member s
of the Whitstable group at a British Lung Foundation
conference, Birmingham, June 2012.
© Sidney De Haan Research Centre for Art s and Health
Canterbury Christ Church University, University Centr e Folkestone,
Mill Bay, Folkestone, Kent CT20 1JG Telephone: 01303 220 870
www.canterbur y.ac.uk/research /centres/SDHR
Sidney De Haan
Research Centre for Arts and Health
with images and sounds of group singing
Researchers, facilitator s and
participants talk about the
value of singing for breathing
Singing and COPD (Chronic Obstructive Pulmonary Disease)
In association with:
Singing, We llbeing a nd Health:
context, evidence and practice
Ian Morrison and Stephen Clift
Sidney De Haan
Research Centre for Arts and Health
Singing and
people with COPD
Chronic Obstructive Pulmonary Disease
NICE (2010) Chronic Obstructive Pulmonar y Disease
Clinical Guidelines 101. London: National Institute for
Health and Clinical Excellence.
QSR International (2010) NVivo 9. Warrington: QSR
International.
Robb, S.L., Burns, D. S., and Carpenter J.S. (2011)
Reporting guidelines for music-based interventions,
Music and Medicine, 3, 271-279.
Robinson, E., Kerr, C., Stevens, A ., Lilford, R.,
Braunholtz, D., Edwards, S., Beck, S. and Rowley, M.
(2005) Lay public’s understanding of equipoise and
randomisation in randomised controlled trials. Health
Technology Assessment, 9(8), 1-92.
Rutten-van Mölken, M.P.M.H., Oostenbring, J.B., Tashkin,
D.P. et al. (2006) Does quality of life of COPD patients
as measured by the generic EuroQol five-dimension
questionnaire differentiate between COPD severity
st a g e s? 13 0 (4), 1117 -1128 .
SGRQ (2008) St George’s Respiratory Questionnaire
Manual, Version 2.2, March 2008. Available at:
www.healthstatus.sgul.ac.uk
Skingley, A., Clift, S.M., Coulton, S.P. et al. (2011)
The effectiveness and cost-effectiveness of a
participative community singing programme as a health
promotion initiative for older people: protocol for a
randomised controlled trial, BMC Public Health, 11, 142.
Available at: www.biomedcentral.com/content/pdf/
1471-2458 -11-142.pd f
Stenton, M. (2008) The MRC breathlessness scale,
Occupational Medicine, 58, 226-227.
Whitmore, C. and Limentani, S. (2009) East Kent COPD
Needs Assessment, Public Health Directorate, NHS
Eastern and Coastal Kent Primary Care Trust.
Documentaries accompanying this report
DVD 1: Facilitating singing for people with breathing difficulties,
with Sonia Page, Project Musical Director
DVD 2: The benefits of regular group singing for breathing, with
Ingrid Falcke who is Living with COPD
DVD 3: Researchers, facilitators and participants talk about the
value of singing for breathing, with images and sounds
of group singing
To obtain copies please contact Isobel Salisbury,
Research Administrator, Sidney De Haan Research Centre
for Arts and Health, Canterbury Christ Church University,
Folkestone, Kent CT20 1JG
Guide to Singing and COPD
Morrison, I. and Clift, S. (2012) Singing and People with
COPD, Sidney De Haan Research Centre for Arts and
Health, Canterbury Christ Church University.
This is one of a series of guides on singing and health
produced by the Centre. Further details can be found
on the Centre’s website: www.canterbury.ac.uk/
Research/Centres/SDHR

Ju ne 2 013
Sidney De Haan Research Centre for Arts and Health, Canterbury Christ Church University,
University Centre Folkestone, Mill Bay, Folkestone, Kent CT20 1JG Telephone: 01303 220 870
www.canterbury.ac.uk/research/centres/SDHR
In association with:
“This is the first winter I have not had
to call an ambulance or be on lots of
antibiotics or steroids. This maybe a
coincidence or it may be better because
of the breathing help I have received.”
“Standing to sing helps posture, you begin to think ‘upright’ automatically as
this gives maximum output from your lungs. The relaxation exercises do just
that, and learning to breathe bringing the muscles of the abdomen into play,
as well as controlled exhalation, has helped me enormously.”
“Have enjoyed being in the project and liked the
singing bit. I gave up going to the gym as found
the singing exhausting and as good as exercise.”
“ This helped mentally and physically.
It’s somewhere to go with like-minded
people. For the first time in five years
I have not been admitted to hospital
or casualty over the winter period. It
opened up doors i.e. joining the (BLF)
Breathe Easy group.”
“I believe that the project is teaching me how to
understand my breathing and how to control it. This
is very useful; it stops me hyperventilating when my
breathing is under pressure i.e. climbing a steep hill.”
“I have enjoyed the project the
singing has help me to understand
how breathing and singing can
help me to breathe better.”
PHOTOGR APHS: (above) The Whitstable Group per forming at the British Lung Foundation Conference, Solihull, June 2012
(front cover) The Whitstable Group performing at St Gregory’s Music Centre, Canterbur y Christ Church University, March 2012