The National Benefits Hub

Research that Supports Recreation

← Home

Advanced Search

A controlled evaluation of the health benefits of a participative community singing programme for...

Key Message

  • Measures of health were consistently higher among the singing group following the singing programme than among the non-singing group
  • Three months after the singing groups stopped the participants continued to be higher on measures of health
  • Participants in the singing groups reported social, emotional and physical health benefits from taking part
  • Singing groups for older people are likely to be cost-effective as a health promotion strategy


Clift, Stephen. (2011). A controlled evaluation of the health benefits of a participative community singing programme for older people (Silver Song Clubs). Canterbury, United Kingdom: Sidney De Haan Centre for Arts and Health.


The aim of the project was to assess the effectiveness and cost-effectiveness for older people of taking part in singing groups on measures of physical and mental health. We wanted to know if singing in a group made a difference when compared to an individual’s usual activities and also how the costs involved in introducing singing compared with the degree of any health improvement.

This was a pragmatic randomised controlled trial. The sample was divided up into two groups, where one group took part in the singing and the other didn’t. All other things being equal, any difference between the two groups is probably due to singing. We can also look at any difference in health before and after the singing activity.
5 new weekly singing groups in East Kent. We asked for volunteers over 60 years to take part in the research. Everyone who volunteered and consented stood an equal chance of being allocated to either the singing group (one of the 5 clubs) or the non-singing group. Through this ‘random allocation’ we could ensure that those who were singing were similar to those who were not, making any comparisons fairer. If we had let people choose, it may be that those who opted for the singing group already felt that this might improve their health while those who opted not to sing were more dubious. This might be reflected in the findings and would make the study ‘biased’.
Participants in the singing groups took part in a weekly programme for 12 weeks. This was led by experienced musicians from Sing For Your Life and was planned in advance so that all five groups had a similar experience. Those who were not singing carried on with life as normal and everyone was asked to complete a questionnaire at 3 stages during the project – right at the start (baseline), after the singing groups had finished the 12 week singing programme, and again after another 3 months when neither group was singing. This was so that we could compare health between the two groups and also before and after the singing (or an equivalent time for the non-singing group), then again after a time lapse to test whether any effects might last. Everyone was invited to attend four extra sessions in each of the venues after data collection was complete.
We measured health through the use of questionnaires which are all used frequently in clinical practice and in research. The main one (called the York SF-12) measures health-related quality of life (QoL) by asking people to rate their own health under a number of headings and has both physical and mental health components. The HADS (Hospital Anxiety and Depression Scale) measures anxiety and depression and the EQ-5D (Euroqol Five Dimensional Scale) asks people to put a value on their health, which can be used when calculating the costs of health and social care to support different health states. The service use questionnaire helps in the costing of health and social services. We also invited people to write comments on the questionnaires about their health or their participation in the project.


Findings from this study suggest a significantly greater improvement in mental health quality of life in individuals who participate in group singing for 12 weeks compared to those continuing with normal activities. In this study the effect was greatest immediately following the intervention, but was still apparent at 3 months later. Singing groups are probably cost-effective. Participatory singing may therefore present a viable additional means to promoting the mental health of older individuals.

Benefit Statements / Outcomes

Leadership Provided By:

  • Leisure Information Network (LIN)
  • Alberta Recreation and Parks Association

On Behalf Of:

  • Canadian Parks and Recreation Association (CPRAA)

Get Updates By Email