Singing and Health:Summary of a Systematic Mapping

and Review of Non-Clinical Research

Stephen Clift, Grenville Hancox, Rosalia Staricoff and Christine Whitmore

Sidney De Haan Research Centre for Arts and HealthThe Sidney De Haan Research Centre for Arts andHealth is committed to researching thecontribution of music and other participative artsactivities in promoting the wellbeing and health ofindividuals and communities.

Current objectives include:

• Undertaking scientific research and evaluationon the potential benefits for wellbeing andhealth of active engagement in music making.

• Documenting and providing the researchevidence base for establishing ‘Singing onPrescription’ for its wellbeing and healthbenefits

• Working in partnership with health and socialcare agencies and service users in the SouthEast of England to promote the role of musicand arts in healthcare and health promotion

• Contributing to the wider development of thefield of Arts and Health research and practicethrough membership of national and regionalnetworks, publications and educationalactivities

The Sidney De Haan Research Centre for Arts andHealth is part of Canterbury Christ Church University.

http://www.canterbury.ac.uk/centres/sidney-de-haan-research/

Sidney De Haan Reports: 5© Sidney De Haan Research Centre for Arts and HealthAuthors: Stephen Clift, Grenville Hancox, Rosalia Staricoff and Christine WhitmorePublished 2008ISBN: 978-1-899253-30-2

Singing and Health:Summary of a Systematic Mapping and Review of Non-Clinical Research

Stephen Clift, Grenville Hancox, Rosalia Staricoff and Christine Whitmorewith Ian Morrison and Matthew Raisbeck

Sidney De Haan Research Centre for Arts and Health

Sidney De Haan Research Centre for Arts and Health

3

BackgroundMusic is one of the defining features of our humannature, and singing is a form of musicalparticipation and expression open to everyone. It istherefore of interest to consider the availableevidence on individuals’ experiences when theysing, and whether singing is beneficial forwellbeing and health.

The origins of the Sidney De Haan ResearchCentre for Arts and Health lie in a small–scaleexploratory study carried out by Clift and Hancox(2001) on the perceived benefits of singing held bychoral singers. In planning their initial study, theauthors conducted a literature search but foundvery little previous research on singing andwellbeing. Since then, however, interest in thisissue has grown. As part of a programme ofresearch on singing planned within the Centre, itwas considered essential to undertake asystematic review of the scientific evidence on thevalue of singing for wellbeing and health.

The aims of this study were: to systematicallyidentify existing published research on singing,wellbeing and health; to map this research in termsof the forms of singing investigated, designs andmethods employed and participants involved: tocritically appraise this body of research, and wherepossible synthesise findings. The hypothesisunderpinning this review was that singing, andparticularly group singing, has a positive impact onpersonal wellbeing and physical health.

Identifying sources for this reviewFor the purposes of this review, a search wasundertaken of ten bibliographic sources using acarefully selected set of search terms organisedinto three groups: health, music therapy andsinging. The strategy adopted was to start withMedline, Embase, Cinahl and Psychoinfo as thesources most likely to identify relevant publishedresearch. The search was confined to paperspublished in English, with no restriction on date ofpublication. A total of 233 papers were identifiedfrom these sources. It was decided at this stagethat clinical studies in which singing served as aform of therapeutic intervention for some specifichealth issue were best considered in a secondreview, and 16 non-clinical studies were judged tobe of interest on the basis of abstracts. Furthersearching of ERIC, ASSIA, RILM, Cochrane Library,Music Therapy World and Google Scholar, producedin excess of 800 references, but only fouradditional papers of interest were identified.

A further search strategy involved examining thereferences given in each of these papers, and thisidentified a further nine sources of interest. Anattempt was also made to contact the authors ofpapers identified to locate further publishedmaterial, giving a further five sources. A range ofadditional, often serendipitous routes served toidentify a further 20 papers of potential interest,giving a final total of 54 sources.

These papers were read independently for contentand quality by two members of the research team,and a process of categorisation and pruning led toa final selection of 35 papers for systematicmapping and review (these are listed in thereferences section). Studies were grouped asfollows: qualitative studies; studies usingspecifically designed questionnaires, quantitativestudies using previously validated scales; studiesmaking objective measures of physiologicalchanges associated with singing (e.g. hormonalchanges); studies employing well establishedmeasures of ‘mental and physical health’; researchassessing physical performance (i.e. hearing andlung function) of singers, and finally, large scaleepidemiological studies in which choral singing wasincluded as a potential determinant of health.

4

An overview of the corpus of studiesAt this stage, it was very apparent that researchinterest in singing, wellbeing and health isrelatively recent. The earliest paper identified waspublished in 1960, and most appeared in the late1990s and beyond. Also, the papers identified arehighly variable in terms of problem addressed,character of singing investigated, participantsinvolved, and methods of data gathering andanalysis. Many of the qualitative studies areexploratory in character and involve very smallsamples. This is especially true of the early studiesreported by Joyce (1996), Smithrim (1998), Adams(2000) and Patteson (2000). Most of the morequantitative studies involving surveys or quasi-experimental designs are also small in scale. Inaddition, the quality of reporting varies with somereports lacking essential details.

Citation mapping within the corpus of sourcesshows little evidence of progressive developmentin the sense that few researchers refer to and buildupon previously published research. Some of themost recently published studies, for example,make little reference to previous studies, evenwhere they are clearly relevant (e.g. Louhivouri,Salminen and Lebaka, 2005; Cohen, Perlstein,Chapline, Kelly, Firth and Simmens, 2006, 2007).

The character of singing investigated in the studiesunder review also shows wide diversity, with somestudies concerned with group or choral singing andothers with individual singers; some studies withamateur or professional singers or both, and somewith existing singing groups of long standing orwith singing groups set up specifically for researchpurposes. Interestingly, studies vary considerablyin the level of detail provided of the singinginvestigated, with some reports giving little or noaccount and others offering quite detaileddescriptions.

Twenty reports have a focus on singing in groups.An interesting aspect of these studies is that whilea few are concerned with single sex groups(Ashley, 2002; Bailey and Davidson, 2002; Silber,2005) in most, female participants out numbermales by approximately 3:1, and in some studiesthe imbalance is even greater. Interestingly, onlyone study specifically considers the factor of‘gender’ in relation to its findings (Clift and Hancox,2001) and reports a significant sex difference inperceived benefits associated with singing. It isclear, therefore, that research in this field isaffected by a considerable bias towards females,which should be taken into account in consideringany of the findings highlighted in this review.

5

Benefits associated with singingWith respect to the possible benefits associatedwith singing, the studies reviewed fall into twomain groups.

Firstly, there are those studies which adopt anexploratory, qualitative approach and gatherinformation by asking singers to give accounts oftheir experience of singing and its personal valuefor them, through interviews, group discussion andwritten answers on questionnaires (e.g. Joyce,1996; Smithrim, 1998; Clift and Hancox, 2001(study 1); Bailey and Davidson, 2002, 2005;Sandgren, 2002). In a few studies, such data isplaced in the context of broader ethnographicobservation and / or a working knowledge of theparticipants in the study (e.g. Ashley, 2002; Silber,2005). Closely associated with these studies arethose which gather essentially descriptive datathrough structured questionnaires (Wise, Hartmannand Fisher, 1992; Hills and Argyle, 1998a,b; Beck etal., 2000; Clift and Hancox, 2001 (study 2); Hilman,2002; Bailey and Davidson, 2003).

Secondly, there are those studies which adopt ahypothesis-testing approach based on the use ofpreviously validated measures of mood, wellbeingor health (e.g. Unwin, Kenny and Davis 2002; Kreutz,Bongard, Rohrmann, Grebe, Bastian and Hodapp,2004; Cohen et al., 2006, 2007); objective measuresof biological or physiological functioning assumedto be associated with emotional state (Galati,Costa, Rognoni and Pisterzi, 2006) or subjectiveexperiences of wellbeing (Beck, Cesario, Yousefi andEnamoto, 2000; Grape, Sandgren, Hansson,Ericson and Theorell, 2003; Beck, Gottfried, Hall,Cisler and Bozeman, 2006), and objective indicatorsof some aspect of bodily function directly affectedby singing (e.g. hearing or lung function) (Heller,Hicks and Root, 1960; Gould and Okamura, 1973;Schorr-Lesnick, Teirstein, Brown and Miller, 1985;Steurer, Simak, Denk and Kautzky, 1998).

In addition, two large-scale epidemiological surveysin Sweden and Finland have included singing inchoirs as a possible health determinant (Bygren,Konlaan and Johansson, 1995; Konlaan, Bygren andJohansson, 2000; Johannson, Konlaan and Bygren,2001; Hyyppä and Mäki, 2001).

Benefits identified by qualitative and questionnaire studiesIn relation to the first set of studies, a number ofrecurrent themes can be identified in thequalitative and descriptive data reported. Namely,that the experience of singing, and particularlysinging in a group, has the following perceivedbenefits:

• Physical relaxation and release of physicaltension

• Emotional release and reduction of feelings ofstress

• A sense of happiness, positive mood, joy,elation and feeling high

• A sense of greater personal, emotional andphysical wellbeing

• An increased sense of arousal and energy

• Stimulation of cognitive capacities – attention,concentration, memory, learning

• A sense of being absorbed in an activity whichdraws on multiple capacities of the body andthe mind

• A sense of collective bonding throughcoordinated activity following the same pulse

• The potential for personal contact with otherswho are like-minded and the development ofpersonal supportive friendships andconstructive collaborative relationships

• A sense of contributing to a product which isgreater than the sum of its parts

• A sense of personal transcendence beyondmundane and everyday realities, being put intouch with a sense of beauty and somethingbeyond words, which is moving or good for thesoul

• An increased sense of self-confidence and self-esteem

• A sense of therapeutic benefit in relation tolong-standing psychological and social

6

problems (e.g. depression, a history of abuse,problems with drugs and alcohol, socialdisadvantage)

• A sense of contributing to the widercommunity through public performance

• A sense of exercising systems of the bodythrough the physical exertion involved insinging – especially the lungs

• A sense of disciplining the skeletal-muscularsystem through the adoption of good posture

• Being engaged in a valued, meaningful,worthwhile activity that gives a sense ofpurpose and motivation

In addition to identifying a wide range of possiblebenefits, attempts have also been made toorganise such a list conceptually and empirically.The most significant conceptual work in this regardhas been undertaken by Bailey and Davidson(2002, 2005). They present a model of ‘the positiveeffects of participation in group singing’ based onstudies of members of two choirs fordisadvantaged adults and a small sample of middleclass choristers. This suggests four principal groupsof benefits: clinical-type benefits; social benefits;benefits associated with public performance, andcognitive stimulation. They argue that thetherapeutic benefits were widely reported acrossthe samples investigated, but other benefits variedto some extent according to social circumstances.

Interestingly, this model overlaps to some extentwith the results reported by Clift and Hancox(2001) from their second survey. Factor analysis ofdata from a structured questionnaire produced sixfactors, which they label as: wellbeing andrelaxation; breathing and posture; social benefits;spiritual benefits, emotional benefits and heart andimmune system benefits. The first factor (togetherwith the fourth and fifth) corresponds closely withthe first and fourth component of the Bailey andDavidson model, and the third social factor in Cliftand Hancox clearly links with the second socialcomponent in Bailey and Davidson.

Similar correspondence can be seen with theresults reported by Hills and Argyle, especially intheir 1998a paper. They identify five factors fromtheir musical experience scale, which they label as:wellbeing, mystical experience, social,entertainment and intellectual / musical. Clearly,the first factor corresponds with the firstcomponent of the Bailey and Davidson model andthe first factor in Clift and Hancox. Factors twoand three in Hills and Argyle correspond withsimilar factors in Clift and Hancox. Their factor fivealso clearly aligns with the component of cognitivestimulation suggested by Bailey and Davidson.

7

Benefits examined in quantitative hypothesis-testing studiesSinging and mood

A number of studies have explored the hypothesisthat singing generates positive feelings through theuse of previously standardised questionnairesdesigned to assess mood (Rider, Mickey, Weldinand Hawkinson, 1991; Valentine and Evans, 2001;Unwin et al., 2002; Kreutz et al., 2004). In each ofthese studies, the questionnaires were used toassess mood before a period of singing and thenimmediately afterwards. Despite the fact that thehypothesis under test is essentially the same ineach case, the studies are difficult to compare asthey differ in the mood questionnaire employed,and the characteristics of the singing andparticipants involved. In addition, the possibility ofsignificant demand characteristics and theconceptual problem of linking short-term and slightchanges in mood to broader notions of wellbeingand health raise serious questions over thescientific value of these studies.

The best designed and most relevant study is thatreported by Kreutz et al. (2004), and their resultsdo appear to show a significant effect of singing onmood compared with a listening to singingcondition. However, close scrutiny of their findingsshows that this change is small, compared with asignificant and large shift in negative mood statewhen singers are asked to listen to a recording ofchoral singing rather than to sing themselves!

Singing and biological / physiologicalvariables

An obvious attraction of studies which attempt toassess the impact of singing on biological orphysiological variables is that they offer thepromise of ‘objective’ knowledge. In addition, theymay contribute to the development of a theoreticalmodel of the underlying processes through whichthe activity of singing generates positive feelingsand also influences the physical functioning andhealth of the body. There is also a sense that suchobjective measures are less likely to be influencedby study demand characteristics, although decades

of research demonstrating the reality of placeboeffects in clinical pharmacological trials, clearlycontradicts any assumption that physiologicalparameters are immune to psychological or socialinfluence! (See Wootton, 2007)

A clear area of interest pursued in a number ofstudies is the connection between singing andstress. This actually has two aspects to it. The firstis that singing itself can be experienced asstressful, particularly when the material being sungis difficult to master, and there are the additionalpressures of public performance. Given that it iswell known that the experience of stress ismediated by the sympathetic nervous system andthe endocrine system, it follows that increases arelikely to be found in key hormones involved in thestress response when singers engage inperformance. Interestingly, however, none of thestudies surveyed which explore physiologicalparameters in relation to singing, consider thedistinction between ‘eustress’ and ‘distress’ – inother words, between ‘positive’ stress, which isimportant in motivating performance to a highstandard, and ‘negative’ stress which arises whenthe demands of a task or situation are perceived tobe greater than coping resources. The secondstance that can be taken in thinking about therelationship between singing and stress, is thatsinging can be relaxing and calming, and maycounteract pre-existing feelings of being stressed.From this perspective, one could predict thatsinging would be associated with a lowering ofstress hormones.

Both of these positions are illustrated in the studyby Beck et al. (2000) in which cortisol levels wereassayed in choral singers before and afterrehearsals and performance. For rehearsals,cortisol levels reduced, whereas underperformance conditions, cortisol levels increased.However, Kreutz et al. (2004) found no changes incortisol associated with a choral rehearsal; Beck etal. (2006) found no changes associated withindividual singing lessons or performances, andGrape et al. (2003) report that cortisol increased formen after a singing lesson, but decreased forwomen. Overall, therefore, the picture for cortisolchanges is somewhat complex, and generalconclusions are difficult to draw.

8

The picture for salivary immunoglobulin A (sIgA)appears to be a little clearer. Five studies haveassayed sIgA before and after singing (Rider et al.,1991; Beck et al, 2000, 2006; Kuhn, 2002; Kreutzet al., 2004;), and four of these studies suggestthat group singing is associated with increasedsecretion of this anti-body in saliva (the exceptionis Rider et al., 1991). As sIgA is part of the immunesystem which offers ‘frontline’ defence in mucosalmembranes against infection, these findingsappear to have some relevance for the health ofindividual singers. Such a conclusion would bepremature, however, given the limitations of thestudies. In particular, there is no evidence that suchchanges measured immediately after singing aresustained, and no evidence that they have anyclinical significance. The causal mechanismsinvolved in generating the changes are alsoobscure, and it may well be that increased levels ofimmunoglobulin are caused not by singing per se,but by the heightened level of stress involved insinging or the fact of being in close proximity ofother people who are breathing energetically, andexhaling potentially infectious organisms!

Singing and measures of mental andphysical health

Only two quasi-experimental studies have beenreported in which the effects of singing onstandardised and objective measures of wellbeingand health are assessed (Houston, McKee, Carrolland Marsh, 1998; Cohen et al., 2006, 2007). In theformer study, a four-week ‘old time sing-a-long’programme was set up in 3 residential carehomes, with 3 homes acting as non-interventioncontrols. Participants completed two well-knownhealth questionnaires before and after theintervention – the Hospital Anxiety and DepressionScale and the General Health Questionnaire-28. Itshould be noted, however, that while singing was acentral part of the activity established in theproject, the focus of the researchers’ attention wason the potential of singing funny songs ingenerating laughter, which they saw as the activeingredient in the intervention. In the Cohen et al.study, a fairly large community chorale wasestablished for retired people living independently,with a similar number of non-participatingcommunity members acting as a comparison

group. A range of measures was administeredbefore the start of the intervention, and then againafter one and two years.

In both cases, positive findings are reported whichappear to suggest that involvement in groupsinging activity had a positive impact on measuresof both mental and physical health. These studiesare among the best pieces of research available todate which have attempted directly to assess thevalue of singing for health. Nevertheless, bothstudies have limitations.

The first is the fact that both employ non-equivalent group designs. Although both researchteams present data to suggest that theintervention and control groups were comparableat baseline, it remains the case that additionalfactors could have produced the differencesobserved between the groups at post-test.

Secondly, the participants themselves were clearlyaware of whether they were participating in theintervention or were in the control arm of thestudy, and as such, they might well have developedtheir own assumptions regarding the purpose ofthe research, and the outcomes anticipated by theresearchers. Under these circumstances, the studyresults are likely to have been affected by study‘demand characteristics.’ To their credit, Houston etal. (1998) recognise this possibility, but Cohen et al.(2006, 2007) do not.

Thirdly, both studies are fairly small in size andgiven the small to medium effect sizes involved forthe measures employed, it is clear that both aresubstantially underpowered. Neither study providesany indication that issues of effect sizes and powerguided decisions about sample sizes employed.

Finally, both studies rely on tests of statisticalsignificance in drawing conclusions about theeffects of their interventions on health, but neitherseriously addresses the question of whether thechanges observed are substantively or clinicallysignificant. This issue is particularly important formeasures of depression employed in both studies,as no clear indication is given that any of theparticipants was affected by real feelings ofdepression prior to the intervention. In addition,neither study clearly demonstrates that the mean

9

changes observed in the intervention groupsrepresent a clinically significant shift away fromdepression.

Singing, lung function and hearing

Singing as an activity involves the singer usingtheir own bodies as an instrument. In commonwith all other musical instruments, this involvesthree components – a sound source, a soundresonator and a sound radiator (Titze, 2008). It issurprising, therefore, that so few studies haveexplored the implications of this fundamental factabout singing, and assessed the extent to whichthe exercise of the basic structures andmechanisms intrinsically involved in singing mightlead either to improvements in the fitness and sohealth of the systems involved, or to possibledamage.

The power for the sound source (the larynx) clearlycomes from the lungs – and it follows that one mightexpect that active involvement in singing may haveeffects on the structure and function of the lungsand the larynx. It is also obvious that singersproduce sound, and that singers in a group canproduce quite a substantial volume of sound. Thereis thus a possibility that regular exposure to highlevels of sound produced by a choral singing groupmay have effects on the structure and functioningof the auditory system. As with the issue ofphysiological changes associated with singing,research on these questions is attractive given thatobjective measures are possible. Given also thatany structural or functional changes would have tobe a result of factors operating over a substantialperiod of time, it is difficult to see how theirassessment could be affected by psychological orsocial demand characteristics in any study.

Both issues have received research attention, but itis remarkable that so little research has beenundertaken. The three earliest studies identified inthis review explored the idea that professionalsinging would lead to a measurable improvementin lung function (Heller et al, 1960; Gould andOkamura, 1973; Schorr-Lesnick et al., 1985). Thebest of these (Schorr-Lesnick et al.) appears tohave dismissed the idea as ‘a myth’ so conclusivelythat no further research has been undertaken onthis subject. Given the lack of evidence for

professional singers, it seems unlikely that amateursinging, whether on an individual or group basis,would have much impact.

Similarly, only one early study has investigatedwhether being part of a professional chorus mightaffect hearing. Steurer et al. (1998) demonstratehearing damage, in the sense of higher auditorythresholds, across a wide frequency range. Thereare some puzzling features of their study, such asthe failure to find any relationship between lengthof career and time devoted to singing and hearingthresholds, which call for further investigation. It isalso an open question whether such effects wouldbe detected among amateur choral singers.

Both the Schorr-Lesnick, et al. and Steurer et al.studies clearly give pause for thought in relation tothe general hypothesis guiding this review – thatparticipation in group singing has benefits forwellbeing and health. In the former case, effectswhich might be expected - and which as Clift andHancox (2001) found, were believed to be trueamong choral singers themselves – are not in factsupported by the evidence. And in the secondcase, the evidence of raised hearing thresholdsraises the possibility that group singing is notnecessarily entirely beneficial, but may in fact leadto bodily damage.

Epidemiological research with singing asa health determinant

Two large-scale Scandinavian studies of populationhealth (in Sweden: Bygren, Konlaan andJohansson, 1996, and in Finland: Hyyppä and Mäki,2001) were also considered relevant to the currentreview. Both explore the idea that social, culturaland leisure activities have a significant impact onpopulation health over and above a range of verywell known and powerful determinants such aseducation, employment, income, smoking andweight. Remarkably, both studies gathered data onwhether participants in their surveys were involvedin choral singing, but both failed to find anyevidence of a link between singing and themeasures of health employed.

Both surveys were examined very closely andwere found to have methodological and reportingweaknesses. As a result, the evidence reported

10

cannot provide a credible test of the hypothesisunderpinning this review. For the Bygren et al.,study, the quality of information on the extent andnature of participants’ involvement in choral singingis very weak, and the health outcome employed isextremely challenging i.e. whether a participantsurvived or died in the course of the study. It ishighly implausible that any relationships would befound, especially following statistical controls for awide range of highly influential factors. The Hyyppäand Mäki study too was judged to have substantialweaknesses – not least that the proportions ofchoral singers in the sample were so low that evensimple order relationships between involvement insinging and the self-assessed measure of healthcould not be demonstrated.

There is no question that in principle findings fromlarge scale population surveys could be relevant tothe hypothesis that group singing is beneficial forwellbeing and health. Nevertheless, the scale ofthe studies required, the care needed in gatheringappropriately detailed and sensitive data, and mostimportantly, the current lack of a convincing casefor expecting substantial effects, render it veryunlikely that specifically designed studies wouldever be mounted.

Theoretical perspectives on singing, wellbeing and healthIn the studies under review a number of theoreticalperspectives and principles developed outside thespecific field of singing and health are drawn uponto understand how singing could have benefits forwellbeing and health.

Bailey and Davidson (2002, 2005) for example,make a convincing case for the relevance ofCsikszentmihalyi’s (2002) ‘flow theory’ inunderstanding the mechanisms through whichsinging can be beneficial to health. Singing can bea demanding activity which requires considerableconcentration and practice, and so is capable ofcreating a sense of ‘absorption’ or ‘flow’ allowing thesinger to escape from other preoccupations. Baileyand Davidson also highlight the arguments putforward by Csikszentmihalyi on the importance ofsocial interaction for mental wellbeing, and link thisto the inherently social character of group singing.

Cohen et al. (2006) also draw upon generaltheoretical principles developed in the field ofGerontology, in attempting to explain why creativeactivities, including singing in groups, have potentialwellbeing and health benefits for older people. Thekey processes they suggest are that creativeactivity helps to generate a ‘sense of control’ andprovides opportunities for ‘social engagement’.Interestingly, they also add a further factor ofmotivation which they see as generated by theintrinsic aesthetic qualities associated with creativeactivity.

Conclusions and RecommendationsIn all human cultures and throughout history, musichas played a significant role in human life. Giventhe fact that group singing can be a powerful andmoving experience, and given the possibility thatsinging could contribute to quality of life, wellbeingand even health – it is remarkable that so littleresearch has explored these issues, anddisappointing that the research which has beenundertaken is so limited and so variable incharacter.

A number of studies stand out as makingimportant contributions to what is clearly a field ofresearch in an early stage of development:

• Bailey and Davidson (2002, 2005) showpowerfully that amateur group singing can havebenefits for participants across a wide socialspectrum – from homeless men singingtogether, to middle class singers in traditionalchoral societies

• Clift and Hancox (2001) explore in more detailthe range of benefits experienced by choralsingers, and have identified empirically anumber of key dimensions of benefit whichcan be measured

• Beck et al. (2000, 2006), Grape et al. (2002)and Kreutz et al. (2004) explore the possiblephysiological mechanisms which couldunderpin subjective experiences in bothindividual and choral singing

• Silber (2005) analyses the specific contributionswhich musical structures and processes canmake in processes of developing social andpersonal wellbeing in a challenging group ofwomen prisoners

• Louhivouri et al. (2005) link research on choralsinging to the wider issue of ‘social capital’ andhave explored experiences and benefits ofgroup singing in very contrasting culturalcontexts

• Cohen et al. (2006, 2007) report an ambitiousquasi-experimental study of the possibleeffects of community singing on the mentaland physical health of older people

At this stage, if work on the potential value ofgroup singing for wellbeing and health is todevelop, there is a need to create a collaborativeand progressive programme embracing conceptualclarification / theory building and empiricalresearch.

Conceptual clarification is needed with respect tothe nature of singing itself, the various forms it cantake, and the essential structures and processesinvolved (physically, physiologically andpsychologically) in the production of the ‘speakingvoice’ and the ‘singing voice.’

Conceptual clarification is also needed with respectto the nature of wellbeing and health. The WorldHealth Organisation has sponsored a large-scaleinternational collaborative programme of work onhealth-related quality of life, which has elaboratedand operationalised the WHO definition of healthas ‘a complete state of physical, mental and socialwellbeing, and not merely an absence of illness orinfirmity’ (WHO, 1946). Conceptual and theoreticaldevelopments in the field of ‘wellbeing’ have seena remarkable growth in recent years and the editedvolume by Huppert, Baylis and Keverne (2005)represents a landmark text in this energetic area ofwork. Further work on singing, wellbeing and

11

12

health needs to take note of these recentconceptual and theoretical developments.

Specific theoretical work is required to link thenature, process and experiences of ‘singing’ withthe nature, processes and experiences ofwellbeing and health. The key questions that needto be addressed are ‘What?’ and ‘How?’ Whateffects, if any, does active involvement in groupsinging have on wellbeing and health? And howdoes singing have these effects – whatmechanisms are at work? What mechanisms, forexample, link the physicality of singing with thephysical wellbeing of the body? What mechanismslink the psychological and social processes at workduring singing with our sense of emotionalwellbeing and mental health? Crucial to thetheoretical task is to identify clearly thosemechanisms which are distinctively and uniquely atwork in the process of singing – and those whichare contingent and may be brought into playequally well by a range of other social or creativeactivities.

Conceptual and theoretical work should go hand inhand with a systematic programme of empiricalresearch. Research can help to inform theorydevelopment up to a point, but a truly progressiveresearch programme requires the elaboration of arealistic theoretical framework which can generatehypotheses. A progressive programme alsorequires a critical mass of researchers and researchteams working in collaboration and competition intesting hypotheses and establishing robustresearch findings through well-designed studies,replication and synthesis of evidence. As yet the

field is a long way from establishing even thefoundation for such a progressive programme, butwithout it, future research is destined to behaphazard, unconnected and non-cumulative.

It is hoped that this review will contribute to aprocess of developing the progressive researchprogramme needed to take this area of workforward.

References35 reports identified for the review

Adams, K. (2000) How can I keep from singing: alifetime of sweet singing in the choir, in B.A.Roberts and A. Rose (eds.) Sharing the Voices: ThePhenomenon of Singing 2. Proceedings of theInternational Symposium, St. John’s,Newfoundland, Canada, 1999. St. John’s: MemorialUniversity of Newfoundland.

Ashley, M. (2002) Singing, gender and health:perspectives from boys singing in a church choir,Health Education, 102 (4), pp.180-186.

Bailey, B.A. and Davidson, J.W. (2002) Adaptivecharacteristics of group singing: perceptions frommembers of a choir for homeless men, MusicaeScientiae, VI (2), pp. 221-256.

Bailey, B.A. and Davidson, J.W. (2003) Perceivedholistic health effects of three levels of musicparticipation, in R. Kopiez, A.C. Lehmann, I. Wohther,and C. Wolf (eds.) Proceedings of the 5th TriennialESCOM Conference, 8-13 September 2003,Hanover University of Music and Drama, Germany.

Bailey, B.A. and Davidson, J.W. (2005) Effects ofgroup singing and performance for marginalizedand middle-class singers, Psychology of Music, 33(3), pp. 269-303.

Beck, R.J., Cesario, T.C., Yousefi, A. and Enamoto,H. (2000) Choral singing, performance perception,and immune system changes in salivaryimmunoglobulin A and cortisol, Music Perception,18 (1), pp. 87-106.

Beck, R.J., Gottfried, T.L., Hall, D.J., Cisler, C.A. andBozeman, K.W. (2006) Supporting the health ofcollege solo singers: the relationship of positiveemotions and stress to changes in salivary IgA andcortisol during singing, Journal of Learning throughthe Arts: A Research Journal on Arts Integration inSchools and Communities, 2 (1), article 19.

Bygren, L.O., Konlaan, B.K. and Johansson, S-E.(1996) Attendance at cultural events, reading booksor periodicals, and making music or singing in achoir as determinants for survival: Swedishinterview survey of living conditions, BritishMedical Journal, 313, 1577-1580.

Clift, S.M. and Hancox, G. (2001) The perceivedbenefits of singing: findings from preliminarysurveys of a university college choral society,Journal of the Royal Society for the Promotion ofHealth, 121 (4), pp. 248-256.

Cohen, G.D., Perlstein, S., Chapline, J., Kelly, J.,Firth, K.M. and Simmens, S. (2006) The impact ofprofessionally conducted cultural programs on thephysical health, mental health, and socialfunctioning of older adults, The Gerontologist, 46(6), pp. 726-734.

Cohen, G.D., Perlstein, S., Chapline, J., Kelly, J.,Firth, K.M. and Simmens, S. (2007) The impact ofprofessionally conducted cultural programs on thephysical health, mental health and socialfunctioning of older adults – 2-year results, Journalof Aging, Humanities and the Arts, 1, pp. 5-22.

Galati, D., Costa, T., Rognoni, E., Pisterzi, A. (2006)The effect of singing on subjective emotional stateand psychophysiological response in singers, paperpresented at the 9th International Conference onMusic Perception and Cognition, Alma MaterStudiorum, University of Bolgna, 22-26 August,2006.

Gould, W.J. and Okamura, H. (1973) Static lungvolumes in singers, Annals of Otology, Rhinologyand Laryngology, 82, pp. 89-95.

Grape, C., Sandgren, M., Hansson, L-O., Ericson,M. and Theorell, T. (2003) Does singing promotewell-being? An empirical study of professional andamateur singers during a singing lesson,Integrative Physiological and Behavioral Science,38 (1), pp. 65-74.

Heller, S.S., Hicks, W.R. and Root, W.S. (1960) Lungvolumes of singers, Journal of Applied Physiology,15 (1), pp. 40-42.

Hillman, S. (2002) Participatory singing for olderpeople: a perception of benefit, Health Education,102 (4), pp. 163-171.

Hills, P. and Argyle, M. (1998a) Musical andreligious experiences and their relationship tohappiness, Personality and Individual Differences,25, pp. 91-102.

13

14

Hills, P. and Argyle, M. (1998b) Positive moodsderived from leisure and their relationship tohappiness and personality, Personality andIndividual Differences, 25, pp. 523-535.

Houston, D.M., McKee, K.J., Carroll, L. and Marsh,H. (1998) Using humour to promote psychologicalwellbeing in residential homes for older people,Aging and Mental Health, 2 (4), pp. 328-332.

Hyyppä, M.T and Mäki, J. (2001) Individual-levelrelationships between social capital and self-ratedhealth in a bilingual community, PreventiveMedicine, 32, pp. 148-155.

Joyce, M. (1996) Turn off the radio and sing foryour lives! Women, singing, and experientialeducation, in K. Warren (ed.) Women andExperiential Education, Dubuque: Kendall/Hunt.

Johansson, S.E., Konlaan, B.B. and Bygren, L.O.(2001) Sustaining habits of attending cultural eventsand maintenance of health: a longitudinal study,Health Promotion International, 16 (3), pp. 229-234.

Konlaan, B.B., Bygren, L.O. and Johansson, S-E.(2000) Visiting the cinema, concerts, museums orart exhibitions as determinant of survival: aSwedish fourteen-year cohort follow-up,Scandinavian Journal of Public Health, 28 (3), pp.174-178.

Kreutz, G., Bongard, S., Rohrmann, S., Grebe, D.,Bastian, H.G. and Hodapp, V. (2004) Effects of choirsinging or listening on secretory immunoglobulinA, cortisol and emotional state, Journal ofBehavioral Medicine, 27 (6), pp. 623-635.

Kuhn, D. (2002) The effects of active and passiveparticipation in musical activity on the immunesystem as measured by salivary immunoglobulin A(SigA), Journal of Music Therapy, 39 (1), pp. 30-39.

Louhivuori, J., Salminen, V-M. and Lebaka, E.(2005) “Singing Together” – A cross-culturalapproach to the meaning of choirs as a community,in Campbell, P.S., Drummond, J., Dunbar-Hall, P.,Howard, K., Schippers, H. and Wiggins, T. (eds.)(2005) Cultural Diversity in Music Education:Directions and Challenges for the 21st Century,Australian Academic Press.

Patteson, A. (2000) Singing a woman’s life: howsinging lessons transformed the lives of ninewomen, in B.A. Roberts and A. Rose (eds.) Sharingthe Voices: The Phenomenon of Singing 2.Proceedings of the International Symposium, St.John’s, Newfoundland, Canada, 1999. St. John’s:Memorial University of Newfoundland.

Rider, M., Mickey, C., Weldin, C. and Hawkinson,R. (1991) The effects of toning, listening andsinging on psychophysiological responses, inC.D.Maranto (ed.) Applications of Music inMedicine, Washington DC: National Association ofMusic Therapy.

Sandgren, M. (2002) Voice, soma, and psyche: aqualitative and quantitative study of opera singers,Medical Problems of Performing Artists, 17, pp. 11-21.

Schorr-Lesnick, B., Teirstein, A.S., Brown, L.K. andMiller, A. (1985) Pulmonary function in singers andwind-instrument players, Chest, 88 (2), pp. 201-205.

Silber, L. (2005) Bars behind bars: the impact of awomen’s prison choir on social harmony, MusicEducation Research, 7 (2), pp. 251-271.

Smithrim, K.L. (1998) Still singing for our lives:singing in the everyday lives of women throughthis century, in B.A. Roberts (ed.) Sharing theVoices: The Phenomenon of Singing. Proceedingsof the International Symposium, St. John’s,Newfoundland, Canada, 1997. St. John’s: MemorialUniversity of Newfoundland.

Steurer, M., Simak, S., Denk, D.M. and Kautzky, M.(1998) Does choir singing cause noise-inducedhearing loss? Audiology, 37 (1), pp. 38-51.

Unwin, M.M., Kenny, D.T. and Davis, P.J. (2002) Theeffects of group singing on mood, Psychology ofMusic, 30, pp. 175-185.

Valentine, E. and Evans, C. (2001) The effects ofsolo singing, choral singing and swimming onmood and physiological indices, British Journal ofMedical Psychology, 74, pp. 115-120.

Wise, G.W., Hartmann, D.J. and Fisher, B.J. (1992)Exploration of the relationship between choralsinging and successful aging, PsychologicalReports, 70, pp. 1175-1183.

15

Additional references cited

Csikszentmihalyi, M. (2002) Flow, London: Rider

Huppert, F.A., Baylis, N. and Keverne, B. (2005) TheScience of Well-Being, Oxford: Oxford UniversityPress.

Titze, I.R. (2008) The human instrument, ScientificAmerican, 298 (1), pp.78-85.

WHO (1946) Preamble to the Constitution of theWorld Health Organization as adopted by theInternational Health Conference, New York, 19-22June, 1946; signed on 22 July 1946 by therepresentatives of 61 States (Official Records ofthe World Health Organization, no. 2, p. 100) andentered into force on 7 April 1948.

Wootton, D. (2007) Bad Medicine: Doctors doingharm since Hippocrates, Oxford: Oxford UniversityPress.

The research teamStephen Clift

Grenville Hancox

Rosalia Staricoff

Christine Whitmore

Ian Morrison

Matthew Raisbeck

Further information

For a copy of the full report and further informationabout the work of the Sidney De Haan ResearchCentre for Arts and Health, see:

http://www.canterbury.ac.uk/centres/sidney-de-haan-research/index.asp

16

Sidney De Haan Reports

An occasional series of reports on the work of the Sidney De Haan Research Centre for Arts and Health

General Editors: Stephen Clift and Grenville Hancox

For further information:

The AdministratorSidney De Haan Research Centre for Arts and Health

E-mail: sdhcentre@canterbury.ac.ukTel: 01303 220870

Report 1

Clift, S., Mackenzie, K. and Bushell, F. (2006) TheMusicStart Project: Evaluation of an Early YearsMusic Initiative. Canterbury: Canterbury ChristChurch University. ISBN: 978-1-899253-36-4

Report 2

Bamford, A. and Clift, S. (2006) Making Singing forHealth Happen: Reflections on a ‘Singing for theBrain’ Training Course. Canterbury: CanterburyChrist Church University. ISBN: 978-1-899253-38-8

Report 3

Bamford, A. and Clift, S. (2006) Music into Upton:Reflections on an Initiative to Bring Live Music intoa Hospital Setting. Canterbury: Canterbury ChristChurch University. ISBN: 978-1-899253-37-1

Report 4

Bamford, A. and Clift, S. (2006) Southampton SilverSong Club: Reflections on Music Making withElderly People Facilitated by Student Volunteers.Canterbury: Canterbury Christ Church University.ISBN: 978-1-899253-39-5

Report 5

Clift, S., Hancox, G., Staricoff, R. and Whitmore, C.(2008) Singing and Health: Summary of aSystematic Mapping and Review of Non-ClinicalResearch. Canterbury: Canterbury Christ ChurchUniversity. ISBN: 978-1-899253-30-2

Report 6

Bungay, H. and Skingley, A. (2008) The Silver SongClub Project: Summary of a Formative Evaluation.Canterbury: Canterbury Christ Church University.ISBN: 978-1-899253-32-6

Report 7

Clift, S., Hancox, G., Morrison, I., Hess, B.,Stewart, D. and Kreutz, G. (2008) Choral Singing,Wellbeing and Health: Summary of Findings from aCross-national Survey. Canterbury: CanterburyChrist Church University. ISBN: 978-1-899253-31-9