Sustaining community choirs for diverse older adults after completion of the Community of Voices trial: A multi-stakeholder, multi-site qualitative study

The growing and increasingly diverse U.S. aging population needs sustainable, evidence-based interventions to maintain health and independence in the community. Through a three-way collaboration between a university research partner, an aging services partner and a community music partner, we developed and tested a senior center-based choir intervention for ethnically diverse older adults that decreased loneliness and increased engagement. We now report on the key factors affecting sustainment, dened by continuation of the choirs after the completion of the trial and ongoing participation of older adults.


Trial registration:
The original trial was registered under ClinicalTrials.gov. NCT01869179. Registered 9 January 2013, https://clinicaltrials.gov/ct2/show/NCT01869179 Background The older population in the United States (U.S.) is bothincreasing and becoming more diverse (1,2). To improve thephysical, cognitive, and emotional well-being of adults age 60 andover, we need to identify cost-effective and sustainablecommunity-based programs that promote health and well-being. Inparticular, evidence-based behavioral interventions that can beeasily implemented by community-based organizations serving diverseolder adults could help them maintain their health, independenceand functioning as they age, and extend the time they live in theirhomes and communities.
For low-resource and culturally diverse community settings,issues of acceptability, practicability and scalability are ofspecial concern in considering which behavioral interventions toadopt or implement.
Complex contextual factors, such as culturalfactors, social networks, agency interactions, community settings,and limited economic resources in uence the implementation ofprograms designed to improve health. Implementation science oftenuses both qualitative and quantitative methods to gain insightsinto these issues, especially when the programs being consideredare multi-component and involve multiple community sites and healthdisparities populations (3,4).
Once a behavioral intervention has been found to be effective inan e cacy study, the next steps involve sustainment of theprograms and adoption by other communities/organizations in orderto reach a broader audience. The logical next step is to attempt tosustain an intervention immediately after it has been tested in thesettings in which it was initially established. Such a sustainmenteffort takes advantage of the fact that these organizations alreadyhave the infrastructure to deliver the intervention, with systemsin place to deliver it and recruit participants. This paper reports on the key facilitators and barriers tosustainment of an intervention immediately following the completionof a cluster-randomized trial (CRT) of a one-year community choirintervention for older adults from diverse racial/ethnic andsocioeconomic backgrounds (5). The intervention was offered in 12senior centers, and we tested its effects on health and well-being(6). At the end of this CRT, additional funding was obtained tocontinue (sustain) a subset of the choirs for six more months andexamine facilitators and barriers to sustainment.
In this paper, we explore factors affecting sustainment of these ve choirs by interviewing three types of stakeholders at the endof the "sustained" six-month time period: music professionals whodirected the choirs, administrators overseeing the senior centersand music organization and trial participants who continued in thechoirs during the sustainment period. We conducted a series ofsemi-structured interviews and focus groups to address threeoverarching research questions. 1) What facilitators were involvedin sustaining the choir intervention in the senior centers aftercompletion of the trial? 2) What barriers were experienced insustaining the choir intervention in the senior centers? 3) Didthese facilitators and barriers affect sustainment onintrapersonal, interpersonal and/or organizational/structurallevels?

Parent Study: The Community of Voices Trial
Through a partnership between the University of California, SanFrancisco, the San Francisco Community Music Center, and the SanFrancisco Department of Disability and Aging Services, includingtwelve Area on Aging-supported senior centers, we conducted a CRTto examine the effects of a community choir intervention on thehealth and well-being of older adults (age 60 and over) fromdiverse racial/ethnic and socioeconomic backgrounds. Usingcommunity-engaged research methods, the Community of Voices (COV)trial (6-8) randomized 12 senior centers to receive the choirintervention either immediately (n = 208) or after a six-monthdelay (n = 182). The community choir intervention was designed totarget cognitive, physical and psychosocial engagement componentsin 90-minute sessions delivered weekly by community-based musicprofessionals (5). In comparison to waitlist controls, theimmediate-start participants had statistically signi cantreductions in loneliness and increased interest in life, but not incognitive or physical function after six months, which was the mainrandomized comparison (6). In an accompanying qualitativeassessment of perceived bene ts, trial participants overwhelminglyreported improvements in emotional well-being that they identi edas related to decreases in loneliness and the view thatparticipation in the choir brought joy to their lives (9). Ancillary Study: Sustaining community choirs after completingthe parent study Sustainment de nitions As per Berta and colleagues, we restrict the de nition ofsustainment to the continuation of the processes and practices ofthe intervention (10), in this case including both sustainment ofthe choirs themselves and continued participation in the choir bystudy participants. We received supplemental funding to continuethe rst ve of the 12 choirs that completed the 12-month COVtrial. Financial assistance was provided to the ve senior centersto enable them to continue (sustain) the choirs for an additionalsix months, which we refer to as the sustainment period. During thesustainment period, the community music partner coordinated thechoir program, collaborating with the senior centers to manage thechoirs, and the research team no longer provided technical supportor protocol oversight. During this time, trial participants fromthe rst ve sites to complete the COV intervention were giventhe opportunity to continue in the choirs during the sustainmentperiod. During sustainment, the choirs were also opened to newparticipants.

Sample and Recruitment
We recruited four types of stakeholders from the ve sites inthe ancillary study: (i) music professionals who continued todeliver the choir intervention, (ii) administrators at both thesenior centers and the music organization, (iii) choir participantswho had completed the original one-year CRT (called "trialparticipants" in this paper) and continued the choir during thesustainment period, and (iv) choir participants who completed theoriginal CRT and elected not to continue during the sustainmentphase, that is, after the original CRT ended.
The music professionals, choir directors and accompanists, wholed the choirs at each of the ve sites (N = 6) and theadministrators at the ve senior center sites and the music center(N = 6) were invited to participate in individual interviewsfollowing the six-month sustainment period.
At the nal 12-month assessment at the conclusion of the COVtrial, we asked trial participants if they planned to continueparticipating in the choir; and if so, if they would consent toparticipate in a focus group (English or Spanish) at the end of thesix-month sustainment period. If trial participants decided not tocontinue in the choir during the sustainment period, they wereinvited to participate in a phone interview within one month. Studyprocedures were explained in person. A new consent form was signedby all participants who agreed to enroll in the ancillary study.The UCSF Institutional Review Board approved this ancillarystudy.

Music Professionals and Administrators
Music professionals were asked about facilitators and barriersto continuing to conduct and provide musical accompaniment for thesenior center choirs. Music professionals were also asked bothgeneral and music-speci c questions about how the sustainmentchoirs differed from the trial intervention.
Administrators wereasked general questions about how the sustainment choirs differedfrom the trial itself, and about facilitators and barriers tocontinuing the choir program. Semi-structured interviews with musicprofessionals and administrators were designed to last about 60minutes and to be conducted either at the senior centers, the musiccenter, or another music studio. Experienced interviewers completedthe semi-structured interviews, and the interviews with musicprofessionals were conducted by an interviewer with expertise inmusic anthropology. Interviewers recorded self-reported sex andrace/ethnicity of all participants.

Participants
Interviews with trial participants who continued the choirintervention were conducted via focus groups at the end of thesix-month sustainment period. These participants were asked generalquestions about the sustained choirs, how the sustained choirsdiffered from the trial choirs and what they perceived asfacilitators and barriers to their continued involvement in thechoir program. Interview guides for Spanish-speaking trialparticipants were translated into Spanish by bilingual-biculturalresearch staff using team reconciliation for the nal version.Focus groups were designed to last about 90 minutes and wereconducted at the senior centers hosting the choirs; 2 focus groupswere conducted in Spanish and 3 in English. Focus groups wereconducted by experienced facilitators; bilingual-biculturalfacilitators conducted the Spanish-language focus group.
Trial participants who elected not participate in thesustainment choirs but agreed to an interview were asked forinformation about the facilitators and barriers to continuing withthe choir intervention, including their reasons for discontinuingthe choir. Interviews were designed to last 15-30 minutes and wereconducted by telephone.
All ancillary study participants completed a brief survey ofdemographic characteristics including: sex, age, race/ethnicity,Spanish language (yes/no), foreign born (yes/no), and country oforigin.

Data Analysis
Focus groups, in-person interviews and telephone interviews wereaudio recorded, professionally transcribed in the originallanguage, and checked for accuracy. Transcripts in Spanish weretranslated into English and checked for accuracy bybilingual/bicultural members of the research team. We used aninductive approach to content analysis (11,12), using ATLAS.tianalytical software. In order to minimize bias, open coding wasconducted by two researchers, a PhD medical sociologist and a PhDpsychologist, who were not involved in the original COV trial. Opencodes were developed to identify speci c barriers and facilitatorsto sustainment. For example, the facilitator code "choir directorinvolvement" was tagged to each statement that identi ed choirdirector activities as key facilitators. Each coder independentlyread and coded the transcripts and identi ed themes through theprocess of axial coding, making connections among codes and writinganalytical memos (11,13). A third researcher, an MDgeriatrician/PhD music anthropologist, independently read all ofthe transcripts and engaged in further axial coding, a processcommon to both grounded theory and ethnographic analysis (11,14),until thematic saturation was reached (no new themes wereidenti ed) and an overarching thematic conceptual model emerged.For example, the open code "choir director involvement" was mergedinto the overarching theme of "social connection" as aninterpersonal facilitator. Throughout the analysis, input wassolicited from the senior researchers and differences werereconciled through discussion.

Participants
Of 94 participants who completed the trial (12-month assessment)at the ve sites, 31 opted to continue singing and participate ina focus group at their site. Of the 48 participants who completedthe trial but chose not to continue singing, 14 agreed to beinterviewed by phone within one month. All ve of the musicprofessionals who delivered the choir intervention wereinterviewed, as were all ve senior center administrators and themusic organization administrator. Table 1 summarizes thedemographic characteristics of each stakeholder group.

Conceptual framework
A conceptual framework for sustainment emerged, re ectingthemes at the intrapersonal, the intrapersonal, and theorganizational levels that served as facilitators or barriers tosustainment. These themes are grouped and presented by two broadcategories of facilitators and barriers, with subgroupings withineach of these by level.

Facilitators
Facilitators were associated with both ongoing individualparticipation in the choir during the sustainment phase as well assustainment of the choirs themselves within the senior centers. SeeTable 2 for example quotes [note: minor details have been changedto preserve anonymity].

Intrapersonal Level
Emotional wellbeing and positiveemotions were identi ed as key facilitators tocontinuation by both the trial participants who continued and thosewho discontinued following the CRT, and this was observed by themusic professionals and the administrators. One trial participantexplained why she had wanted to continue participation in the choirby saying, "the act of being in the choir, music feeds the soul asthey say, but aside from that, it relaxes you. And aside from that,it transports you to wonderful memories. It makes you happy."Positive emotions included overall enjoyment of the choir and withsinging, a strong sense of wellbeing, and enjoyment of speci caspects of participation such as visibility, connections with thepast, and the enjoyment of learning and physical exercises.
Interpersonal/Relationship Level Social connection, belonging and socialsupport were emphasized as facilitators in all of thetrial participant focus groups. As one trial participant explained,"I think it's that sense of belonging to a larger group. They'vemade friendships, again. It makes them feel empowered. They feelactive, you know, they have a sense of belonging… and so theycome, they're anxious to come together to practice, to share andnow they've become friends." The emergence of friendships withinthe choirs was observed by both music professionals andadministrators alike.
In addition to relationships among trial participants,relationships developed between the trial participants and musicprofessionals. Trial participants described deep admiration for themusic professionals through descriptions like the following: "andwith great teachers, because we have very good teachers, theydirect us well, they explain each thing to us, if we're doing well,if we're not getting the tone. They always are careful that webring out our voice and I just love that." The music professionalssimilarly highlighted the importance of the relationship betweentrial participants and their director. "What I've seen in thechoirs in general, is that that relationship between choir directorand participant and the creation of a social fabric within thegroup is very important. Because that sense of community is part ofwhat helps people to stay connected. So, it's not an isolatedactivity. In fact, it brings people together. And that's what helpsmake it fun and helps people to continue." Strong relationshipsbetween the senior center directors and the music center director,were also identi ed as key facilitators to sustainment.

Organizational/Structural Level
Organization factors related to the structure,format and programming of the choirs emerged as aprimary facilitator of choir sustainment. The "professionalism" ofthe choir directors and accompanists [music professionals] wassingled out as a critical factor, including one administrator'sobservation that the music professionals were "very patient, butalso very clear and rm. And I think the people follow their lead.I think you have to have that in order to produce results." Thisincluded both the technical aspects of leading a rehearsal and theoverall organizational skills needed to prepare the group for choirperformances. As one trial participant explained, "The schedulesthat the teachers gave us were extremely well organized, very welldone. It had the whole year, with the vacations, when we were out,when we would go in." Beyond scheduling, organization skills at the level of the choirinvolved sophisticated musical and pedagogical aspects of choirdirection. The music professionals spoke in detail about ways inwhich the quality of the music led trial participants to continueduring the sustainment period, and of the need to create newarrangements appropriate to the vocal ranges of older adults.Re ecting on the re nement of the choir in the sustainment phase,one music professional observed that "the curriculum of the choirsessions got stronger, with a more codi ed approach to customizingthese choir sessions, which really came out of a communitydevelopment model or a community arts model, to be even morecustomized towards older adults." The music professional noted thatthe sustainment period provided a time in which key stakeholders"learned more about how to deliver a choir program speci cally forolder adults." Organizational involvement emerged asan understated but critical aspect of sustainment. The involvementin and support of senior center personnel related to the choirprogram were viewed as essential for success over time. Without thewillingness of the senior center to schedule, reserve rooms andprovide space for rehearsals, the choirs could not have continuedafter completion of the trial. Strong senior center involvement,however, did more than make sustainment possible, it ensured thesuccess of the program. As one music professional explained, "Whenthe centers took over, they [the senior center administrators]pretty much nursed us through the transition because we didn't haveto worry about the room, they would help set it up, and it was abig choir." The music professionals and administrators also described theimportance of strong and supportive relationships at theorganizational level. One administrator acknowledged this saying,"I think that one nice thing that we see is, the choir members dofeel an affection for the senior center. They appreciate theenvironment." One administrator explained thatorganizational-level relationships are important "because oneorganization cannot do it by themselves. It takes a couple oforganizations getting together to make sure that these thingshappen, so we all put a little piece in the pot, and it works. Ilove it." These multiple and overlapping relationships were foundto be essential in the sustainment of the choirs.

Barriers
Unlike the facilitators, which the stakeholders identi ed ascritical for sustainment of the choirs, none of the barriers wereassociated with discontinuation of the choirs. Barriers listedbelow were either identi ed as causing di culties withattendance, or with the discontinuation of a speci c trialparticipant. See Table  3 for example quotes [note: minor detailshave been changed to preserve anonymity].

Intrapersonal Level
Health issues were identi ed as aprimary reason for trial participants to attend less frequently inthe choirs or to drop out entirely during the sustainment phase.These included serious life events such as cancer, chronic illnessmanagement requiring regular medical appointments, and intercurrentillnesses (see reasons for discontinuation below). The impact ofcold and u season on attendance was described by an administratorwho observed that, "by the holidays it [attendance] has kind ofdrizzled down a little bit, and then like I said, the weather, andyou know, them being a little under the weather."

Interpersonal Level
Negative social interactions weredescribed in the focus groups and interviews, although with lessfrequency than the positive interactions listed above. During choirrehearsals, there were isolated complaints of interruptions of thesessions by speci c trial participants. One trial participant feltthat there were insu cient opportunities for social interaction.When new members were allowed to join the choirs after the trialended and the sustainment period began, several trial participantsexpressed frustration with the singing abilities of the newcomers.For one choir, a music professional initially reported that "atleast the group… absorbed the newcomers," but then quali ed thestatement, noting that there "was a bit of friction with the newerpeople that did stay, when it opened up to community groups." Competing family obligations werefrequently cited as a reason for discontinuing the choir or asbarriers to attending rehearsals. When one trial participantobserved that "a few of our seniors, they had still caregivingpositions, or taking care of grandbabies," another trialparticipant chimed in with "taking care of grandbabies," followingwhich the rst speaker continued, "great grandbabies -great-great-grandbabies." Trial participants described in detailthe need to help adult children as well as the next generations,prioritizing familial relationships over those created throughparticipation in the choir, while expressing regret that thefamilial relationships interfered with choir participation.
Organizational/Structural Level Transportation/Access issues. As mightbe expected in an urban environment, a primary structural barrierinvolved travel to the senior center. Some trial participantsstruggled to navigate public transportation. One participantexplained his complex situation as follows: "I don't have easyaccess to a car, even though I have one. My daughter has a nanny,and she gets the car. And I could easily call a cab…. But I justhate spending money when I don't really have to. So, if I can walksomeplace, it's great. But, at night, in the dark, in thewintertime, I don't want to be walking alone in the street, to andfrom the bus stop." Stakeholders speci cally identi ed limitedparking as barriers to timely arrival for rehearsal, summed up mostsuccinctly by a music professional as "The parking, just theparking." Choir rehearsal and schedule issues.Issues related to choir rehearsal organization and schedulingsometimes undermined relationships and created barriers toparticipation. Several trial participants raised concerns aboutscheduling and communication issues. One of the music professionalsnoted that scheduling and communicating with the senior centers wascomplicated by the transition from a university to a communitypartner coordinator after the trial ended. As this transition wastaking place, one music professional noted the critical nature ofthis position in sustainment "Easier, would be to have de nitely acoordinator, someone that's not a director, not an accompanist…Otherwise it's just -it's too much for one or two persons to do."(Music professional). Another music professional noted that, oncethe trial was over, the choirs began to receive local invitationsto perform. These invitations were accepted, but the performancesled to schedule changes on short notice. Additionally, some trialparticipants identi ed the change in rehearsal time as a criticalobstacle to continuation and others disliked a summer break, whichwas added to conform to the long-standing schedule at the musiccenter. Communication issues between the music and senior centerorganizations created minor friction, but were reported rarely. Senior Center environment issuesincluded both modi able and unmodi able barriers. The clearestexample of ongoing attempts to reduce modi able barriers waspresented by a music professional who described issues of noisethat were ameliorated but not entirely xed by a change in seatingrearrangements during the sustainment phase: "So everything -thathas all improved in the extension. Now if we could just get rid ofthe fans, things would be really great. Super loud fans." Incontrast to seating arrangements, which were altered in severallocations, the size and availability of the rooms were notmodi able. In one case, the move to a smaller room causedfrustration for several trial participants, and led another trialparticipant to discontinue because the room had been used for afamily member's memorial service.
Issues due to transition of choir management fromacademic to senior center. Although the transitionwas overwhelmingly described as positive by trial participants,both the music professionals and the administrators identi edbarriers to sustainment associated with this change in management.In one case a music professional identi ed that one senior centerhad "a very di cult time at having choir retention [i.e.attendance and recruitment] in general. I don't know why." Anothermusic professional described a speci c transitional issue, theneed for water (which had been provided during the trial by theresearch staff) during the social break in the middle of therehearsals. The music professional said that initially, "I felt thepulling away of the support of UCSF around the food," but that,ultimately the issue was resolved through support of the seniorcenter.
Funding. In contrast to musicprofessionals and trial participants, all administrators identi edfunding as the critical factor underpinning sustainment of thechoirs or any other community-based program. One administratorsaid, "what happens over and over again, I mean, time and timeagain, is that there's funding that comes down for programs thatare good concepts. People get behind them, they organize them andall that and they're great for a while, and then the funding driesup. And then, there's no more programs." Every administratorexpressed the desire to support the choirs but noted that theywould require additional funding in order to do so. As onesummarized, "It all comes down to resources, because we have theframework." Reasons for discontinuing after the end of thetrial. All reasons for discontinuation are includedhere, whether or not they reached the level of thematic saturation.Among participants who completed the trial and then elected not tocontinue during the sustainment phase, the reasons for notcontinuing fell into three major categories: serious illness,competing caregiving responsibilities, and disliking singing in thechoir. Several trial participants reported more than one primaryreason for discontinuation. Nearly two thirds of thetelephone-interviewed trial participants noted that they hadenjoyed the choir and wanted to continue but were unable tocontinue during the sustainment phase due to health issues orfamilial responsibilities. One third cited health issues includingcancer, surgery, acute infection, chronic "dizziness," worseninghearing loss, arthritis, and acute bereavement issues. One thirdcited external family and caregiving relationships, such as movingout of the area to live closer to adult children, and providingdirect care to young children in the family. One third citedreasons involving the choir intervention itself. These includedpreferring to sing alone, nding the singing to be "stressful,"disliking unexpected changes in scheduling, feeling frustrated whena music professional arrived late, and disliking the musicselections.

Discussion
This study examined facilitators and barriers to sustainment bythe community music and senior center partners without thetechnical assistance of the research team after completing thelarge communitybased CRT. We identi ed three salient themesincluding (1) emotional engagement and personal health at theintrapersonal level, (2) the quality of social relationships andsocial connectedness at the intrapersonal level, and (3) On the interpersonal level, all stakeholder groups identi edrelationships as essential to the success or failure of choirsustainment. These ndings are consistent with studies of otherchoirs that found that social relationships (15) and socialconnections (16) were key to the continued engagement ofparticipants. Our ndings expand on this work by moving beyond therelationships between trial participants or between trialparticipants and music professionals, to examine the criticalnature of relationships between the music professionals leading thechoirs and the senior center administrators who provide theinfrastructure to house the choirs. When these relationships aresupportive and appreciated, the choirs are sustained. Notable wasthe issue of competing relationships as barrier, such as caregivingfor grandchildren or other family care responsibilities. This nding is new in the literature on choirs for older adults.
On the organizational/structural level, administratorappreciation of and commitment to the choirs served as a powerfulfacilitator. These ndings are consistent with another study thatfound that management support was essential to sustainment of newcommunity-based behavioral interventions (17). Funding wasidenti ed as the rate-limiting step for sustainment by all vesenior center administrators because their intent was to continuethe choirs inde nitely, while the administrative supplement onlyprovided six months of funding support. Structural barriersinvolving travel and parking were reported as affecting attendance,but were not cited as a reason for discontinuing participationunless a participant moved entirely out of the area. Issues ofscheduling and communication, functioned as facilitators when thestakeholders felt expectations were clear, and as barriers whentrial participants felt they received inadequate preparation forcancelled rehearsals.
Although the overall successful sustainment of all of the choirssuggests that the barriers were outweighed by the facilitators,this sustainment study has some limitations. The study isqualitative. All ve choirs that were supported by researchfunding for sustainment continued for the full six months, butassessment over a longer timeframe would strengthen the ndings.
Since completion of the CRT, all twelve choirs have beensuccessfully sustained and the intervention has been disseminatedto two new senior centers (18). In addition, the music partner ispiloting two Cantonese language choirs in collaboration withaffordable senior housing sites. The choirs continue to rehearseand perform regularly, without involvement of the researchpersonnel and with support from a variety of fundingmechanisms.

Conclusions
This qualitative study provides insight into the facilitatorsand barriers to sustainment of a successful, community-based choirprogram for older adults. In the case of choirs for older adults ofdiverse backgrounds, the marked sense of well-being and positiveemotion, coupled with a strong sense of social connection andsocial support, suggest that these programs are sustainable inexisting senior centers when funding can be found. The emotionalwell-being and positive relationships facilitate participation inthe choirs, but this is only one aspect of sustainment. The otheris the sustainment of the choirs themselves, which requires musicprofessional involvement, administrative expertise, funding andspace. By bringing together two types of established communityorganizations that do not usually collaborate, senior centers andmusic organizations, the trial served to create a lastingpartnership that underpins the sustainment and expansion of theprogram. The conceptual model of facilitators and barriers tosustainment can serve as a guide for the development andsustainment of future arts interventions in community settings.

List Of Abbreviations
Community of Voices (COV) Cluster-randomized trial (CRT) Declarations Ethics approval and consent to participate:Approval to conduct the study was gained from the University ofCalifornia, San Francisco Institutional Review Board, and consentwas obtained in writing from all participants Consent for publication: Not applicable.
Availability of data and materials: The dataset(which includes individual transcripts) is not publicly availabledue to con dentiality policies. De-identi ed data is availablefrom the corresponding author on reasonable request. Authors Contributions. T.A.A. helped plan thestudy, collect and analyze the data, and write the paper. J.K.J.helped plan the study, supervise the data collection and analysis,and write the paper. A.L.S. helped plan the study, supervise dataanalysis, and write the paper. E. P. helped supervise the dataanalysis and write the paper. S. M. helped plan the study, developthe community sites, review the analysis and edit the paper. S. S.helped plan the study, implement the intervention, review theanalysis and edit the paper. M. S. helped analyze the data and editthe paper. A.M.N. helped plan the study, supervise the datacollection and analysis, and write the paper. All authors have readand approved the manuscript.
Disclaimer. The contents and views in thismanuscript are those of the authors and should not be construed torepresent the views of the National Institutes of Health. Tables   Table 1   · "In one senior center case, there was a last-minute cancellation,and the communication was not executed well. So, we had to reallyhave a conversation and a way to ensure that that wouldn't happenagain." (music professional) Transition of choir from academic to senior center · Feelings that partnership was unequal · Changes in rehearsal structure · "Ithink that the choir is really great. I would just want a littlemore control over things next time out. I mean, we're hosting…we're called a partner, but we really haven't been much of apartner." (administrator) · [onchanging the snack break]: "Andeverybody just had different feelings. Like, some people wanted todo it, some people didn't, and then some people were intimidated."(music professional) Funding · Funding is challenging but critical · "Believeit or not, activities and senior center things are not as easy towrite grants for, in the foundation world."(administrator) Figure 1 Multi-level framework of factors affecting choir sustainment

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. AdditionalFile1SRQRchecklist.docx