The Contribution of Community Singing Groups to the Well-Being of Older People

Abstract

Current evidence suggests that participatory arts activities, and particularly group singing, may contribute to the well-being of older people. However, there is currently a paucity of prospective research from the participant perspective. This qualitative study nested within a randomized controlled trial aimed to assess participants’ perspectives of the acceptability and effect on health and well-being of a community singing program for older people. Volunteers recruited to the intervention arm (n = 131) were invited to write comments on their experiences over three data collection points of a 14-week singing program. A subsample (n = 19) participated in a retrospective semi-structured interview. Data were subjected to content and thematic analysis. Comments and interviews from 128 individuals suggested that the singing groups led to specific, incremental benefits to physical, psychological, social, and community well-being. Benefits tended to tail off after the program ended. Suggestions were made for the future running of such groups.

Keywords

health and well-being older people singing groups

Introduction

The change to the population demographics of the United Kingdom is well known, with the number of individuals of pensionable age (historically 60 for women and 65 for men but gradually moving to 67 for all by 2028) projected to rise from 12.2 million in 2010 to 15.6 million by 2035 and 22.2 million by 2085 (Office for National Statistics, 2012). Older people are increasingly likely to account for a significant proportion of health and social care service use (King’s Fund, 2012). This presents challenges for service providers, and there is a need to develop innovative and cost-effective approaches to maintain and promote the health of this population. Research from the perspective of older people themselves has found that the most important contributors to well-being are factors such as relationships and contacts, getting out, having a range of interests, and a positive outlook (Hoban, James, Pattrick, Beresford, & Fleming, 2011). Promoting health with these emphases suggests a need to look beyond traditional health and social care services and develop interventions that can fulfill these priorities.

Recently there has been growing interest in the potential value of the arts in addressing health and well-being (Clift, 2012). Evidence from arts therapies (delivered by specially trained therapists) has yielded mixed results (Bradt, Magee, Dileo, Wheeler, & McGilloway, 2010; Crawford et al., 2012) with such health interventions being potentially costly and of limited availability, in view of the individual or small group mode of intervention delivery and need for referral. At the same time, community-based programs, especially in music, are becoming increasingly popular and accessible in the United Kingdom.

Singing is one of the most widespread art forms in many Western societies, engaging large numbers of people (Clift, Nicol, Raisbeck, Whitmore, & Morrison, 2010). A systematic review of its therapeutic benefits to the health and well-being of groups concluded that there are indications that singing can promote a sense of personal and social well-being and may be effective in promoting physical health (Clift et al., 2010). Methodologies described in the review were largely standardized self-report health measures and physiological measures, with a smaller number of studies adopting qualitative approaches.

One of the largest studies using self-report measures investigated the perceived benefits of singing in community choirs in England, Australia, and Germany (Clift et al., 2009). Questionnaire responses from more than 1,000 choristers described benefits in terms of positive affect, focused concentration, controlled breathing, social support, cognitive stimulation, and commitment. In studies examining physiological changes, the focus has been mainly on hormone levels. Salivary immunoglobulin A, a marker of immune functioning, has been found to increase with singing (Kreutz et al., 2004), whereas cortisol, a stress hormone, has been found to decrease (Grape, Sandgren, Hansson, Ericson, & Theorell, 2003). Oxytocin, considered to underpin feelings of bonding, has also been found to rise (Grape et al., 2003). Such findings may help explain the positive psychological and social benefits reported by choir members, as well as suggesting that there may be health protective mechanisms at work.

Of particular note is the growth of community singing groups for older people and an accompanying desire to evaluate these in terms of both health and economic outcomes, to inform policy at national and local level. One of a very limited number of controlled studies (Cohen et al., 2006, 2007) found benefits to health in people aged 65+ after 1 year (n = 77) and 2 years (n = 57) allocated to a singing group intervention, when assessed on a series of standardized health measures. No attempt was made in this research to collect data on the perceptions of individual participants, which would have been an appropriate way of capturing the dimensions identified above by Hoban et al. (2011). Studies that have done this, for example, Hillman (2002) and Teater and Baldwin (2012), point to benefits such as decreased social isolation, therapeutic value, and increased quality of life, even in the face of existing illnesses and bereavements. However, these studies relied on respondents’ retrospective accounts of any improvements and there is, overall, a lack of concurrent qualitative data from the perspective of older people in the research literature.

In summary, the research indicates that there are physiological mechanisms at work which may translate into the beneficial experiences from singing reported in established choirs. Less clear is whether these findings also obtain with older people who may not have previously engaged in the art form. In view of the demographics presented above and the need for health promoting activities that are both cost-effective and acceptable to this age group, this current study was considered to be timely in filling gaps in the existing evidence base.

The Sidney De Haan Research Centre for Arts and Health has pursued a progressive program of research on the health- and well-being-related benefits of singing since 2005. An initial formative evaluation of “Silver Song Clubs” for older people, set up by the charity “Sing For Your Life” (www.singforyourlife.org.uk)(Skingley & Bungay, 2010), yielded encouraging findings. This served as grounding for a randomized controlled trial (RCT) of such singing groups (Coulton, Clift, Skingley & Rodriguez, (2015); Skingley, Clift, Coulton & Rodriguez, 2011; Skingley, Bungay, Clift & Warden, 2013) to include, as well as standardized health measures, a qualitative element designed to capture the experiences of participants at different stages of the research process. This article presents findings from the qualitative component of the research project.

Method

Aim of the Study

The aim of this study is to assess participants’ perspectives on the acceptability and effects on health and well-being of a community singing program for older people.

Design

The research adopted a two-phase, nested mixed-methods design, with the qualitative nested within the quantitative. Creswell and Plano Clark (2007) have described this as useful when researchers wish to follow up on the experiences of participants where the data are predominantly quantitative. In this we based our thinking on the work of Creswell (2013). Qualitative data were collected concurrently in the baseline and follow-up phases of a RCT, but with the purpose of addressing different questions, rather than converging the two data sets, as in some triangulation designs. Thus, although the overall study adopted a post-positivist interpretive framework, with its traditional ontological (“single reality”) stance, its epistemology (how reality is known) adopted a more constructivist approach. This allowed a more explanatory and nuanced account of research outcomes at a more individualized level. Furthermore, research findings, however highly significant, are unlikely to translate successfully into practice unless they are also acceptable, and perceived as beneficial, to those for whom they are relevant.

Consenting individuals were randomized with equal probability to either one of five newly created singing groups or usual activities. Outcome measures used were the York SF12 (measures health-related quality of life; Iglesias, Birks, & Torgerson, 2001), the Euroqol EQ-5D (a health utility measure; Euroqol Group, 1990), and the Hospital Anxiety and Depression scale (HADS; Zigmond & Snaith, 1983). Measures were taken at baseline, 3 months (immediately post-intervention), and 6 months, within a postal questionnaire. All participants were also invited to make written comments at all data collection points and subsamples of both the intervention and control groups were invited to take part in an interview following completion of data collection.

The practice of randomizing individuals involved in arts and health research is infrequent and may be regarded as potentially problematical. Although RCTs are considered to be the definitive method for establishing treatment effects (Howard & Thornicroft, 2006), recently a number of criticisms of the design have emerged. Busby Grant, Mackinnon, Christensen, and Walker (2009) have argued that subjects of such studies have historically been assumed to be passive recipients of assigned conditions, while studies where details of preference to allocation have been collected have reported that most of their participants, unlike researchers, lacked equipoise, that is, they stated a preferred treatment condition in unblinded studies (Behrendt, Gölz, Roesler, Bertz, & Wünsch, 2010; Leykin et al., 2007; Snowdon, Garcia, & Elbourne, 1997). However, most of the evidence for this has emerged from clinical intervention studies (for a specific health condition) or analyses of hypothetical scenarios. Moreover, most of this research is not specifically concerned with older adults, which is the focus of the present study. We countered the possibility of “resentful demoralization” from those not allocated to singing groups through offering an opportunity to join a group for a number of singing sessions following completion of the research. We also hoped to ensure informed consent to randomization via the taster sessions (see below). Further information on the control group is reported on elsewhere ((Skingley et al., 2013).

Recruitment and Sample

Participants were volunteers who responded to publicity for the study. Advertisements were placed in local newspapers, general practices, and community venues and flyers were distributed to the homes of local residents. The information provided explained the purpose of the study and invited interested potential participants to an information session that included a “taster” of being in a singing group. All those who expressed interest, aged from around 60 years upward, were eligible to participate. The principal exclusion criterion was inability to give informed consent, though no volunteer met this criterion.

A total of 131 individuals were allocated to the intervention group, with 105 (80%) remaining to final data collection and included in the quantitative analysis. We received comments from 128 respondents over the course of the project, with 93 at baseline, 125 at midpoint, and 90 at endpoint. A total of 65 individuals commented at all three time points, allowing for some tracking of changes over the course of the project. Of this sample (n = 128 respondents, which included those subsequently interviewed), 99% described themselves as White, 75% were retired, and 64% had continued in education beyond the minimum school leaving age. The mean age was 67.3 years (range 58-91 years), 84% were female, and the modal income was in the range £10,001 to £15,000 (20% fell into this income bracket). This represents a relatively homogeneous group of third-age, White females, which is characteristic of many singing groups (Clift et al., 2008), although the education and income levels were modest. These characteristics were also representative of the overall research sample.

Recruitment to the interview phase of the study was by invitation at the time of the final questionnaire. To counter the preponderance of women among study participants, the interview sample included all men who volunteered and completed a further consent form. Women participants who volunteered to be interviewed and consented were drawn randomly and stratified across groups to match the number of men included in the sample. Total number of interviewees was 30, with 19 in the intervention group (male = 10; female = 9) and 11 in the control group (male = 5; female = 6). The difference in sample sizes is explained by the fact that some participants initially agreeing to be interviewed were either lost to follow-up (up to 5 attempts were made to contact individuals) or changed their minds about participating. Data from personal interviews and written comments from the intervention group are reported here.

The Intervention

The singing groups were located in East Kent, South East England, and those recruited took part in a 13- to 14-week program. Trained and experienced facilitators under the guidance of “Sing for Your Life,” a charitable organization specializing in the provision of singing groups for older people, devised a 14-week 90-min program comprising songs from different eras and a variety of genres and including a break for tea and biscuits. Throughout the project, facilitators met to ensure that they had a common understanding of the material and how to deliver it (e.g., accompaniment, musical key, acquiring copyright).

The program was developmental, progressing from singing single melody lines to harmonizing, layering, and singing in rounds. Chime bars were also introduced where appropriate and participants had an opportunity to request particular songs. All clubs delivered the same program concurrently. A program manager who made unannounced visits to each club 5 to 6 times during the intervention period monitored fidelity. Facilitators produced a songbook for the trial and maintained a register of attendees. Following completion of the research, participants in both the intervention and control arms of the study were invited to take part in a further program of singing for 4 weeks.

Data Collection

Written comments were solicited on the final page of each questionnaire in response to an open invitation to participants to write about their health or their participation in the project. A semi-structured interview schedule was developed, based on the aims of the research and issues arising from the written comments that might further address the research aims. This covered the following:

Motivation to take part in the research

Expectations compared with subsequent experiences

The singing program

Satisfaction with the research process

Positive experiences and things that could have been improved

Thoughts on community singing groups generally

Those agreeing to be interviewed were contacted by phone to arrange a mutually convenient time. Personal interviews were carried out either in the participants’ homes (n = 16) or via the telephone (n = 3), according to individual preference. Although it may be argued that data obtained from the two interview methods may not be comparable, studies that compare the different approaches have tended to conclude that data are of the same quality (Carr & Worth, 2001). One researcher (A.S.) who was already experienced in this method (Barriball, Christian, While, & Bergen, 1996) conducted all the telephone interviews to enhance consistency. All participants consented to interviews being recorded.

Data Analysis

All written comments and interviews (including those conducted via the telephone) were transcribed and imported into NVivo9® (QSR International, 2010), a qualitative data analysis program. We adopted a thematic analysis (Braun & Clarke, 2006), which proceeded as a multi-stage process.

Participants’ comments were read independently by two researchers to become fully conversant with the content and gain an understanding of the themes and details in the text. Guided by the research aim, the following analysis was undertaken:

Written comments (varying in length between a single sentence and two sides of A4 paper) were analyzed inductively through the allocation of text segments into categories and sub-categories (referred to as “nodes” in NVivo). An automatic analysis of word frequencies was also conducted to identify substantive areas related to health, well-being, or acceptability and to supplement the inductive component. This also permitted a broad content analysis.

Interview data were similarly analyzed initially using the categories emerging from the analysis of the written texts, supplemented with further categories based on specific interview questions.

For both types of data, dimensions which linked together substantial sets of categories were identified to form a framework from which to make analytic generalizations and address the research aim. Analytic generalization occurs—usually in qualitative research—when moving from particulars to broader constructs or theory (Polit & Beck, 2010). This contrasts with statistical generalization, which generalizes to numbers through achieving a sample representative of the population under study. The identified dimensions focused on quality (positive/negative statements), change over the course of the data collection periods, and attribution of any change to the intervention (see Figure 1). Theory was built up as categories were added to in the form of segments of text, to which subsequent texts were compared, thus supporting or questioning the emerging theory.

The construction and allocation of text to categories was cross-checked by two researchers working independently and then resolving any minor differences emerging. Abbreviations were devised to denote time points and data sources as follows: Q₁—baseline written; Q₂—midterm written; Q₃—endpoint written; Int.—Interview. Where possible, participants’ responses were followed from Q₁ to Int. to capture the effects attributed to the intervention, noting where this varied over time. Positive and negative statements about the project were used to measure the degree of acceptability of the community singing program. Theorizing was drawn from comments on community singing groups in general, and participants’ references to possible mechanisms through which identified effects from the intervention activity took place.

Figure 1.

Analytical framework.

Findings

Presentation of findings is structured around the categories, prefaced by a health profile of participants as depicted in comments at baseline and subsequent questionnaire rounds. The cross-cutting dimensions (quality statements, strength of attribution, and change over time) are integrated where appropriate, as are relevant comparisons with findings from the quantitative data. Written and interview comments from individuals are reported to illustrate findings.

Health and Well-Being Profile of Participants

At Q₁ the data suggest that our sample were in relatively good health for this age group. Seventeen individuals commented at baseline, and 14 at other points, that their health was good. This is supported in the mean baseline score on the mental component of the SF12, which was just below the U.K. population norm (though the physical component score was somewhat further below the norm). However, a large number of respondents also commented on specific health issues, the most frequently reported being arthritis (31), other joint and orthopedic problems (15), aches and pains, likely to be connected to the above (15), and respiratory/breathing problems (15). Less frequently mentioned were poor mobility (11), high blood pressure (9), cardiac issues (8), and mental health problems such as depression (8).

Comments relating to family and domestic life were overall negatively framed, reporting feelings of isolation, anxiety, and depression. These were mainly attributed to health and other problems experienced by close family members (17), participants being carers of family members (4 specifically mentioned this, but many of the above implied a caring role), and bereavement (9).

General Impact of the Singing Program on Health and Well-Being

Evidence at Q₁ showed that even before the research had started, there was positive anticipation of the effect of community singing groups on participants’ overall health and well-being in comments such as “if it does me good I would like to try” (Q₁, female age 66). Some participants reported specific health conditions that they thought the singing program would help to improve (see later categories), which suggests there existed a rationale behind such thinking.

Comments on the change in the participants’ health and well-being were largely positive at Q₂ and Q₃ and often attributed to the singing program. These findings supplement the quantitative results from structured measures where there was a significant positive difference in favor of the intervention (singing) group at both 3 and 6 months on the mean mental component of the SF12 and a non-significant difference on the physical component:

My health and general outlook and expectations have improved a great deal since participating in this project. (Q₂, female age 67)

I was surprised at how much healthier I feel and more relaxed since I started the club. I have managed to change my work so I will be there on Fridays Choir Meeting. (Q₃, female age 67)

A few individuals (n = 6) reported their general health and fitness as having been and remaining good throughout the project duration. Besides enjoyment, these individuals were not able to distinctively report on the health benefits gained from singing in community groups.

Participation in this project was a very enjoyable experience. I am lucky to have good health on the whole and certainly this has not changed. The experience has made me want to continue singing in the future. (Q₂, female age 70)

I’m very fortunate because my health has always been very good so whether it’s improved my health I don’t know because it was quite good to start with. (Int. male age 72)

Participants observed the effect of community singing on their health and well-being in a number of ways (see Table 1) broadly grouped here as physical, psychological, and social health benefits.

Table 1.

Perceptions of Impacts of Singing on Health and Well-Being.

Area of impact	No. of individuals reporting	Examples
Physical	29	Better respiratory function
		More stamina
		Ability to cope with aches and pains
		Improved posture
		Improvements in voice projection
Psychological	107	Enjoyment
		Sense of well-being
		Improved self-confidence
		Greater confidence in singing
		Cognitive stimulation
		Memory improvement
Social	59	Meeting like-minded people
		Peer support
		Getting to know local community
		Making new friends and networks
		Counteracting loneliness

Physical Benefits

The most frequent reference to physical changes concerned breathing, with 29 individuals commenting on this across Q₂ and Q₃, and 6 anticipatory statements at Q₁. These comprised learning techniques of better breathing, improvement in breath control, more stamina while carrying on daily activities, and a distinctive improvement for those who suffered from chronic obstructive pulmonary disease. Interview data (n = 8 individuals) suggest that the impact was sustained beyond the project life span. The following comments of one individual from Q₂ to Int. demonstrate the sustained improvement.

Male age 79:

It has also made me aware of the fact that I do not breathe properly and has, I think, improved the situation. (Q₂)

The singing has made me aware of my lack of training in breathing and breath control. (Q₃)

Apart from anything else it’s made me breathe better. And that was one of the reasons for wanting to take it up. (Int.)

Interestingly, only two of these eight interviewees had made written comments about breathing improvements, suggesting that the overall number presented in the written data is an underestimate of breathing-related benefits felt by participants. One interviewee observed improvements in co-participants that could be pertinent to others with breathing problems:

I don’t suffer with chest things but people that did I thought what a wonderful thing—helpful for people with catarrh and hay fever and asthma and that sort of thing—to get in there and be shown how to breathe. (Int. male “over 60”)

Beyond breathing, singing was reported to have helped some participants in other ways, for example, to cope with aches and pain through diverting thoughts away from the physical pain and improving the general functioning of the body.

Not many people reach the age of 82 without some health problems and chronic pain back problems and I’m no exception, but I have always tried to deal positively with this. I firmly believe that events such as the Silver Song Scheme are so beneficial in making one forget a lot of the pain, really better than medication. (Q₃, female age 82)

Also at the time, I was recovering from heart surgery, so to me it was something quite useful because I was having to physically increase my physical potential as much as I could and singing was a good way of doing that and I think it helped me to recover a lot quicker and I was discharged much quicker from the hospital than I should have been. (Int., male age 63)

Other physical health benefits highlighted were the opportunity to keep active, improved vocal ability and audibility, increased levels of energy, and improved posture. The phraseology used to express these benefits, typified by the above quotations, illustrates the degree of certainty with which improvements were attributed to the singing. These varied between “I think it helped” (male age 62) to “I firmly believe” (female age 82), with instances of the degree of certainty often becoming stronger over time, from “I think [it has] improved the situation” at Q₁ to “it’s made me breathe better” at Q₃ (male age 79, quoted in full earlier).

Psychological Benefits

A large number of individuals (107) reported at one or more of the time points on the effect of the singing project on their mental health and well-being, with the majority of these being positive. By far the most common sentiment expressed was one of enjoyment, with related words written or voiced by 89 respondents. Beyond simple enjoyment, singing was felt by a number of individuals (n = 59) to convey a sense of well-being in other ways, for example, improved singing and self-confidence (n = 6) and cognitive stimulation and memory (n = 7). This was often regardless of specific individual physical ailments or pre-existing psychosocial conditions. A few of those commenting (n = 7) referred to either hoping this would be the case, or noting benefits even from the taster sessions.

Apart from a bit of arthritis in my knees and fingers I feel generally reasonably well. When I did the singing at the try-out day, it made me feel quite uplifted and my wife commented on this. (Q₁ male age 63)

Where such individuals could be followed through, there is a suggestion that the psychological effect, even in the face of initial skepticism, could strengthen over the course of the singing program, but may have lost some of its impact when the singing stopped. This is also reflected in the mental health component scores on the SF12, which dropped slightly at 6 months.

Male age 63:

I would just like to say that never having done any singing before, I was a bit skeptical about any effects that this project might have on me. To my surprise I feel generally happier and calmer than previously. My wife also says that she can see a difference; especially each Friday after singing. I know that I am not alone in hoping that the group can in some way continue in the autumn. (Q₂)

I thoroughly enjoyed taking part in this research project. I felt more positive in general and my wife told me that I was a lot nicer to be around during the weeks that the project was running. Now apparently I have gone back to being a bit miserable! I am looking forward to going to the four follow-up sessions. (Q₃)

To counter this loss of impact in well-being, a few participants (3) reported at Q₃ having joined other singing groups to sustain the psychological benefits reaped from the singing program:

The singing sessions definitely had a positive effect on my mood. I felt more cheerful, more focused and more energetic. When the sessions ended, I felt compelled to join another singing group in order to stave off the withdrawal symptoms and maintain those positive feelings. (Q₃ female age 64)

Only one negative comment emerged, stating that “I found the singing made me feel miserable” (male age 69, Q₂). Despite such initial impressions, the participant continued to attend, however, “that just reinforced my belief that I wasn’t going to enjoy it” (Int.). Although in a minority, this sentiment should caution against drawing overambitious claims from the data and prompt a consideration of what Yin (2009) terms “rival explanations.” While we canvased all participants for their responses, it is likely that those with the greatest negative experiences would withdraw completely from the project, unlike this individual who disengaged from the singing program but continued to complete the questionnaires.

Social Benefits

A total of 59 individuals made reference to the social aspects of community singing groups and the feedback was largely positive across all data collection points. The immediate benefits were perceived to be the ability to meet and mix with like-minded people, peer support, getting to know the local community, and the opportunity to network and make new friends and counteract loneliness. These are all, arguably, contributors to well-being and were seen as applying to individual participants, as well as more generally having the potential to effect community change:

I have also found it beneficial to be part of a group and mixing with others, especially as I live on my own. (Q₂, female age 63)

A lot of people, once they reach sixty, sixty-five, they’ve had a healthy working life, they’ve given it up and there’s a void in their life. A lot of them are also bereaved and it is something that’s proving a very useful focal point in bringing them together. (Int. male age 62)

Comments on the Project and Singing Program

There were specific questions at the interview stage of the project that aimed to gather information to inform future planning of community singing programs. Questions were designed to stimulate positive and/or negative responses about facilitation and leading, organization and administration, the research project, variety of program content, access to the venue and environment, suggestions on areas that could be improved, and the ending of the project and future plans.

A number of factors that motivated participants to take part in the research project were identified from the interviews, with several giving more than one reason. These included the love for music and singing (10), anticipated health benefits (3), and doing something social, often in or nearing retirement (6).

Responses to interview questions on facilitation and leading were all positive, as were statements from a further 30 individuals in the questionnaires. Participants described facilitators as “brilliant” and “inspirational,” with a high “spirit of enthusiasm” that was infectious to the rest of the groups. The majority of participants commented that the project was well administered.

Although separate groups met in different venues and at different times of the day, all participants regarded the assigned time as convenient and venues as accessible. This is no doubt due partly to the advertising material that stipulated the planned times and venues, (should volunteers be allocated to the intervention arm):

Yes I thought it was quite good. It was a nice easy venue to get to, parking was easy. It wasn’t far for anyone, there was a bus route there as well for people who needed a bus. (Int. male age 62)

Positive comments were mainly about space, light, acoustics, and height of the ceiling (n = 5), with negative aspects referring to noisy heaters, cold room, room too big for a small group, and lack of day light.

Comments on the content of the singing program mostly related to the choice of songs. The majority of participants who commented (12) expressed approval of the range of songs that included both new and old material, with something for everyone. Negative statements (10) were made about older songs included in the program. Participants felt that the songs were embedded in a certain era, which made some feel older, and they did not like this. Two of these individuals however pointed to the need to achieve the balance for different tastes in music, for example,

I’m not so keen on the pop stuff—we do things like Lollipop and De Do Ron Ron Ron and I’ll go along with the flow but I could do without them. But it’s a singing group and it’s got to suit all tastes. And I think the balance is pretty good. (Int. female age 62)

A total of 71 individuals commented over Q₂ and Q₃ on the end of the research project and future plans. The overwhelming majority of participants expressed disappointment with expressions such as “the singing will sadly be missed” or “sorry it has come to an end.” However, this was at least partly compensated for by the four sessions arranged for both intervention and control groups once the data collection was complete:

I have definitely been at a loss on Mondays, so look forward to the 14th November, to start singing with you. (Q3 female age 75)

Furthermore, at Q₃ of the research project, many individuals (17) were hoping to join a community singing group. The majority of participants interviewed (10) had joined community singing groups at the time of interview. Subsequent to the research, some participants regrouped into a self-funded social enterprise, which was formed in January 2011 under the leadership of one of the music facilitators who had been involved. More than 100 individuals from the project continued to sing in four of the groups, and, in its first year, a further four groups were created. Currently, 14 groups are running across East Kent and new groups are planned in Hampshire (see: http://www.livinglively.org.uk/). This suggests a high level of acceptability of community singing groups for older people.

Eight interviewees evaluated the singing program as “perfect” and could not think of any ways to improve it. Others (n = 15), in both the questionnaire responses and at interview, made suggestions for future improvements. These included,

Allow participants to introduce themselves

Inform participants of areas to be discussed prior to the interview

Allow participants to carry the song books home

Provide the music notation as well as the words for participants

Have more men in the group!

The National Health Service (NHS) to take on board the findings

Integrate more movement into the singing cf. Rock Choir

Make a recording

Have a charity concert

Ten participants also expressed interest in knowing the outcomes of the research project:

It would be interesting to know the result of the research and also what will come from it. If projects get started as a result of the research—if it becomes part of caring for people. It would be very interesting to know that. If it’s taken up as a mainstream thing maybe in care homes or things like that. (Int. male age 67)

Discussion

This study aimed to assess the acceptability and perceived effects on well-being of community singing groups for older people. Evidence from written comments from 128 individuals, together with 19 interviews, pointed to a high level of acceptability and a number of mechanisms through which the groups were felt positively to affect on a range of well-being dimensions. Particular benefits were noted in relation to breathing, enjoyment, mental health, and social connectedness. These findings support existing literature in the area (Clift & Morrison, 2011; Cohen et al., 2006, 2007; Hillman, 2002; Morrison et al. 2013; Teater & Baldwin, 2012). Comments on the singing program also reflect previous work in the finding that some individuals disliked songs that they felt to be more suitable for those older than themselves (Bungay, Clift & Skingley, 2010).

Beyond this, the research makes an original contribution to the evidence base in three ways. First, these qualitative findings serve to complement the quantitative data from the study through methodological triangulation. Although standardized, validated health measures are able to convey an overall picture of outcomes, they fail to identify states and changes at an individual level. At baseline, for example, although the overall health profile of the participants described a sample in reasonably good health, personal testimonies revealed individual examples of significant health, family, social, or bereavement issues. This makes expressions of well-being more remarkable, and indicates that possibly the findings in terms of the SF12 score may be able to be replicated in a group where the baseline health profile is less positive. At follow-up, participants were able to identify states of health and well-being that were important to them, such as making social connections, not well captured in the quantitative data. They were also able to indicate whether they themselves attributed any improvement to the singing, and this is important in practical terms, as explanations that make sense are more likely to result in people engaging in opportunities to sing. In addition, detailed accounts from participants allow for exploration of “disconfirming cases” (Polit & Beck, 2003)—data that appear to challenge the emerging theory—which serve as a caution against over-hasty generalization. As Gridley, Astbury, Sharples, and Aguirre (2011) observed, singing is not necessarily for everyone and it is important to be aware that for some people singing may not offer beneficial outcomes.

Second, collecting concurrent qualitative data over the timescale of the project allowed evidence to emerge of incremental health gain over time for some individuals and some aspects of health, whereas for others this may have been followed by a degree of “relapse” once the groups ended. This suggestion is given further credence through the timing of the interview data, which followed the final questionnaire. A sense of disappointment at Q₃ (3 months post-intervention) was often followed by more positive comments at interview for those individuals who had joined newly formed or pre-existing groups. These findings point to the need for any research project setting up singing groups to consider sustainability of the groups after the research phase.

Third, the written responses were all unprompted in terms of content. They were also anonymized, giving permission to participants to respond negatively and/or positively, thus enhancing credibility and lessening the likelihood of a response bias. Furthermore, the prompt within the interviews to suggest areas needing improvement also served as a guard against a tendency to desirability, which could yield only positive responses. Allowing free text also produced a broader range of topic responses than a structured questionnaire alone. For example, information was volunteered on the acceptability of the singing program that was broadly approved of but with some detractors providing helpful comments.

Limitations

There are a number of limitations with this research. The sample was composed of volunteers who responded to advertising, and therefore was likely to consist of individuals already drawn to the idea that singing might benefit health (despite the fact that we sought volunteers who were not currently members of singing groups). This is borne out in the anticipatory comments and reasons given for signing up and limits transferability, though not necessarily the practical applicability of the findings. Only those who remained in the research to completion were canvased and of those, all but a minority also remained in the intervention. Thus, we have no information from those dropping out regarding acceptability or outcomes.

The research was restricted in terms of geographical location (East Kent, United Kingdom), in length of intervention (weekly over 3 months), and to a limited demographic (predominantly White, female, retired), so the findings cannot extend beyond these parameters. The age range of the sample was wide and, from the evidence suggesting that participants related song preferences to age, it may have been more appropriate to recruit within a more limited age range and devise a less eclectic program. This might yield more reliable findings in terms of replicability.

Although the analytical strategy served the project well, any data reduction and handling involves a certain degree of inference. Despite having two researchers involved, there is a chance that findings do not accurately reflect the experiences of the participants. However, in conjunction with the quantitative findings, there is reason to believe the research successfully addressed its intended aims.

Conclusion

Findings from the research suggest that participatory singing for older people can be both acceptable and beneficial for the health and well-being of those taking part. For those in pre-existing good states of health, participation appeared to maintain well-being. In this respect, the data support the findings from the quantitative measures. The evidence from individual testimonies of specific benefits further highlighted a more nuanced account of individual motivations, satisfaction, preferences, and changing experiences over time, as well as explanations of the mechanisms through which singing promotes well-being and health. Such detailed findings may help to inform the setting up of future singing groups with and for older people.

Recommendations for Future Research

A number of recommendations emerge from the experience of this research. We suggest that singing groups set up for research should include plans for sustainability to maximize health benefits post-research. Planning for intervention programs would benefit from the inclusion of lay representation to increase the potential acceptability of musical content to the target age group. Data should be sought, where possible, from those withdrawing from singing programs to identify more specifically those individuals who are more or less likely to benefit from any initiatives. Finally, to increase the transferability of findings, consideration should be given to replicating the research in a different geographical location and with a longer intervention.

Footnotes

Acknowledgements

We would like to acknowledge the contribution to this research project of the following: Simon Coulton, Professor of Health Services Studies, University of Kent; Dr John Rodriguez, then Assistant Director of Public Health, NHS Eastern and Coastal Kent; Stuart Brown (Director) and facilitators of Sing For Your Life; NHS Kent and Medway for funding the singing program; and the research participants for their time and support.

Ethical Approval

The study was approved by the Surrey National Health Service (NHS) ethics committee (ref: 10/H1109/5) and was undertaken in accordance with the Declaration of Helsinki (World Medical Association, 2001).

Authors’ Note

A.S. was project coordinator and contributed to all stages of the research and to the first and revised draft of the article. A.M. contributed to the data analysis and first and revised draft. S.C. was chief investigator and contributed to all stages, including subsequent drafts. All authors have seen the final and revised version of the article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article presents independent research commissioned by the National Institute for Health Research (NIHR), Research for Patient Benefit program ref: PB-PG-0408-16038. The views expressed herein are those of the authors and do not necessarily reflect those of the National Health Service, NIHR, or the Department of Health, United Kingdom.

Author Biographies

Ann Skingley, RN, PhD, is principal research fellow, Sidney De Haan Research Centre for Arts and Health, Canterbury Christ Church University. Past work includes contributions to community nursing research. Current research interests include the contribution of the arts (singing and dance) to the well-being of older adults and improving the care of people with dementia admitted to acute care settings.

Anne Martin, QSW, MA, is a research fellow, England Centre for Practice Development, Canterbury Christ Church University. She has made contributions to international HIV/AIDS prevention and control. Current research interests involve developing the capacity of the health and social care workforce to apply a whole systems approach to providing person-centered, safe, and effective care.

Stephen Clift, PhD, FRSPH, is professor of health education at Canterbury Christ Church University, and director of the Sidney De Haan Research Centre for Arts and Health. He has made contributions to research, practice, and training in various areas of public health. Current interests relate to arts and health and particularly the potential value of group singing for health and well-being.

References

Barriball

Christian

While

Bergen

(1996). The telephone survey method: A discussion paper. Journal of Advanced Nursing, 24, 115-121.

Behrendt

Gölz

Roesler

Bertz

Wünsch

(2010). What do patients understand about their trial participation? Assessing patients’ understanding of their informed consent consultation about randomised clinical trials. Journal of Medical Ethics, 32, 74-80. doi:10.1136/jme.2010.035485

Bradt

Magee

Dileo

Wheeler

McGilloway

(2010). Music therapy for acquired brain injury. Cochrane Database of Systematic Reviews, 7, CD006787. doi:10.1002/14651858.CD006787.pub2

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77-101.

Bungay

Clift

Skingley

(2010). The Silver Song Club Project: A sense of wellbeing through participatory singing. Journal of Applied Arts and Health, 1(2), 165-178.

Busby Grant

Mackinnon

Christensen

Walker

(2009). Participants’ perceptions of motivation, randomisation and withdrawal in a randomised controlled trial of interventions for prevention of depression. Journal of Medical Ethics, 35, 768-773.

Carr

E. C. J.

Worth

(2001). The use of the telephone interview for research. NT Research, 6(1), 511-524.

Clift

(2012). The creative arts as a public health resource: Moving from practice-based research to evidence-based practice. Perspectives in Public Health, 132, 120-127.

Clift

Hancox

Morrison

Hess

Kreutz

Stewart

(2009, August). What do singers say about the effects of choral singing on physical health? Findings from a survey of choristers in Australia, England and Germany. Proceedings of the 7th Triennial Conference of European Society for the Cognitive Sciences of Music (ESCOM), Jyväskylä, Finland.

10.

Clift

Hancox

Morrison

Hess

Stewart

Kreutz

(2008). Choral singing, wellbeing and health: Findings from a cross-national survey. Canterbury, UK: Sidney De Haan Research Centre for Arts & Health.

11.

Clift

Morrison

(2011). Group singing fosters mental health and wellbeing: findings from theEast Kent ‘singing for health’ network project. Mental Health and Social Inclusion, 15, 2, 88-97.

12.

Clift

Nicol

Raisbeck

Whitmore

Morrison

(2010). Group singing, wellbeing and health: A systematic mapping of research evidence. UNESCO e-journal, 2(1), 1-25. Retrieved from http://education.unimelb.edu.au/__data/assets/pdf_file/0007/1105927/clift-paper.pdf

13.

Cohen

Perlstein

Chapline

Kelly

Firth

Simmens

(2006). The impact of professionally conducted cultural programs on the physical health, mental health and social functioning of older adults. The Gerontologist, 46, 726-734.

14.

Cohen

Perlstein

Chapline

Kelly

Firth

Simmens

(2007). The impact of professionally conducted cultural programs on the physical health, mental health and social functioning of older adults: 2 year results. Journal of Aging, Humanities, and the Arts, 1, 5-22.

15.

Coulton

Clift

Skingley

Rodriguez

(2015). Community singing and health in the older population: A randomized controlled trial. British Journal of Psychiatry, 207, 250-255.

16.

Crawford

Killaspy

Barnes

Barrett

Byford

Clayton

. . . Waller

(2012). Group art therapy as an adjunctive treatment for people with schizophrenia: Multicentre pragmatic randomised trial. British Medical Journal. doi:10.1136/bmj.e846. Retrieved from http://www.bmj.com/content/344/bmj.e846

17.

Creswell

J. W.

(2013). Qualitative inquiry & research design: Choosing among five approaches (3rd ed.). Thousand Oaks, CA: SAGE.

18.

Creswell

J. W.

Plano Clark

V. L.

(2007). Designing and conducting mixed methods research. Thousand Oaks, CA: SAGE.

19.

Euroqol Group. (1990). Euroqol: A new facility for the measurement of health-related quality of life. Health Policy, 16, 199-208.

20.

Grape

Sandgren

Hansson

L.-O.

Ericson

Theorell

(2003). Does singing promote well-being? An empirical study of professional and amateur singers during a singing lesson. Integrative Physiological and Behavioural Science, 38, 65-74.

21.

Gridley

Astbury

Sharples

Aguirre

(2011). Benefits of group singing for community mental health and wellbeing: Survey and literature review. Carlton, Australia: Victoria Health Promotion Foundation.

22.

Hillman

(2002). Participatory singing for older people: A perception of benefit. Health Education, 102, 163-171.

23.

Hoban

James

Pattrick

Beresford

Fleming

(2011). Voices on well-being: A report of research with older people. Cardiff, UK: WRVS. Retrieved from www.wrvs.org.uk/shapingourage

24.

Howard

Thornicroft

(2006). Patient preference randomised controlled trials in mental health research. British Journal of Psychiatry, 188, 303-304.

25.

Iglesias

C. P.

Birks

Y. F.

Torgerson

D. J.

(2001). Improving the measurement of quality of life in older people: The York SF12. Quarterly Journal of Medicine, 94, 695-698.

26.

King’s Fund. (2012). Future trends. London: Author. Retrieved from www.kingsfund.org.uk/think

27.

Kreutz

Bongard

Rohrmann

Grebe

Bastian

Hodapp

(2004). Effects of choir singing or listening on secretory immunoglobulin A, cortisol and emotional state. Journal of Behavioural Medicine, 27, 623-635.

28.

Leykin

DeRubeis

Gallop

Amsterdam

Shelton

Hollon

(2007). The relation of patients’ treatment preferences to outcome in a randomised controlled trial. Behavior Therapy, 38, 209-217.

29.

Morrison

Clift

Page

Salisbury

Shipton

Skingley

Vella-Burrows

Coulton

Treadwell

(2013). A UK feasibility study on the value of singing for people with chronic obstructive pulmonary disease (COPD). UNESCO Journal 3, 3.Retrieved from http://web.education.unimelb.edu.au/UNESCO/pdfs/ejournals/vol3iss3_2013/003_MORRISON_PAPER.pdf

30.

Office for National Statistics. (2012). National population projections—Chapter 2: Results, 2010-based NPP reference volume. Newport, UK: Author.

31.

Polit

D. F.

Beck

C. T.

(2003). Nursing research: Principles and methods (7th ed.). Philadelphia, PA: Lippincott Williams and Wilkins.

32.

Polit

D. F.

Beck

C. T.

(2010). Generalization in quantitative and qualitative research: Myths and strategies. International Journal of Nursing Studies, 47, 1451-1458.

33.

QSR International. (2010). NVivo 9. Warrington, UK: Author.

34.

Skingley

Bungay

(2010). The Silver Song Club Project: singing to promote the health of older people. British Journal of Community Nursing, 15(3), 135-140.

35.

Skingley

Bungay

Clift

Warden

(2013). Experiences of being a control group: lessons from a UK-based randomized controlled trial of group singing as a health promotion initiative for older people. Health Promotion International, 29(4), 751-758. doi:10.1093/heapro/dat026

36.

Skingley

Clift

S.M.

Coulton

S.P.

Rodriguez

(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: 42 doi:10.1168/1471-2458-11-142. Retrieved from http://www.biomedcentral.com/1471-2458/11/142.

37.

Snowdon

Garcia

Elbourne

(1997). Making sense of randomisation: Responses of parents of critically ill babies to random allocation of treatment in a clinical trial. Social Science & Medicine, 45, 1337-1355.

38.

Teater

Baldwin

(2012). Singing for successful ageing: The perceived benefits of participating in the Golden Oldies Community-Arts programme. British Journal of Social Work, 44, 81-99. doi:10.1093/bjsw/bcs095

39.

World Medical Association. (2001). Declaration of Helsinki. Bulletin of the World Health Organization, 79(4). Retrieved from http://www.who.int/bulletin/archives/79(4)373.pdf

40.

Yin

R. K.

(2009). Case study research: Design and methods (4th ed., Applied Social Research Methods Series, Vol. 5). Los Angeles, CA: Sage.

41.

Zigmond

Snaith

(1983). The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica, 67, 361-370.