Introduction

The response of communities of gay men to the emergence of the HIV epidemic have been linked to successful HIV prevention efforts, and the concept of ‘gay community’ has long been considered important in HIV research and sexual health promotion [1–3]. In the context of HIV behavioural surveillance, community-based organisations, venues on the commercial gay scene, and other spaces frequented and used by gay and bisexual men have played a critical role in accessing ‘community’ samples of men from the wider population [4]. Such organisations and venues have also played an important role in HIV education and prevention. It is within this context that the Gay Men's Sexual Health Survey (GMSH Survey) emerged. The GMSH Survey has been conducted every three years since 1996 with gay and bisexual men on the commercial gay scenes of Glasgow and Edinburgh. It was established as part of the evaluation of the Gay Men’s Task Force peer-led sexual health intervention and has gone on to examine changes in gay men's sexual behaviour over time [5, 6]. Since 2005, the survey has included the collection of anonymous oral fluid samples, which are tested for HIV antibodies to assess prevalence of HIV and undiagnosed infection [7]. Undiagnosed HIV infection is determined by comparing self reported HIV status with the results of the HIV antibody test based on the oral fluid sample. Survey data and HIV antibody tests are linked, however, participation in the survey (and provision of a saliva sample) is anonymous [8]. As the biological samples are collected anonymously the men taking part do not obtain any individual feedback on their results.

Research evidence suggests that changes are occurring to the structure of gay communities, patterns of socialisation, and ways of connecting socially and sexually with other men [9–13]. The growing social acceptability of same-sex relationships and homosexuality, have brought about changes to men’s relationships to the gay scene, and sense of connection to gay communities [11–13]. Research in the UK and Australia has identified shifts in societal norms around sexuality, and the increased use of online technologies for engaging with potential partners as facilitators of these changes [4, 14–16]. New technologies such as gay-specific social media, particularly GPS-based apps (e.g. Grindr and Gaydar), have enabled and facilitated changes in social and sexual practices, meaning that in many ways men are less reliant on the geographic places and spaces of commercial gay scenes when connecting with other men [17, 18]. Furthermore, it has been suggested that exclusively gay and/or lesbian venues are reducing in number, in part due to a growing homonormativity, but also the patronisation of gay venues by heterosexual ‘allies’ [17]. As such, there is growing concern that samples of men recruited through convenience sampling and bar-based surveys of gay men, such as the GMSH Survey, may not entirely reflect the diversity of the population, particularly as community norms and practices change [4, 19]. Valid data collection methods are crucial to the development, implementation and evaluation of HIV prevention interventions, and it has been suggested that different methodologies are likely to be required if behavioural surveillance in western countries is to keep pace with changing social and sexual practices [4].

Furthermore, there is a growing ethical question around the acceptability of not providing individual feedback on the results of HIV tests to those who participate in epidemiological surveillance, particularly those with undiagnosed HIV [20–22]. Although much of the research in this area has focused primarily on the use of unlinked anonymous blood samples from participants in low-income countries, issues raised are pertinent to the design of the GMSH Survey. Future behavioural surveillance research could capitalise on developments in HIV testing technologies which allow for immediate results in various settings by including HIV testing within the study design. However, this would require a very different model of data collection than the current bar-based GMSH Survey. Such a change has potential implications for participation; would men be open to this form of survey in this particular context? Although previous research specifically drawing on the GMSH Survey has explored willingness to participate in future HIV prevention studies amongst gay and bisexual men [23] currently little is known about how men understand and perceive the issue of feedback on anonymous HIV antibody tests.

Understanding why people participate in research, and their expectations of feedback, is important in terms of the design and development of effective research strategies. A number of qualitative studies have been conducted which explore the meanings people ascribe to participation in various types of health research. Such research has been conducted with a range of groups including children [24], pregnant women and mothers[25], individuals who have been involved in nursing research on emotive topics [26], and those who have taken, or are considering taking, part in longitudinal cohort studies [27, 28]. Much of this research emphasises people’s perceptions of taking part in studies as an altruistic act, as a selfless contribution to medical science for the wider benefit of society [27, 29–32]. Other key considerations around taking part in research include the potential for improvements in participants’ own health [33]. Some studies report more proximal motivations for research participation, specifically personal benefit. For example, Hallowell and colleagues’ [34] study exploring participation in genetics-related research emphasises the complex interrelation between perceived individual, familial and social benefits. They argue that it is overly simplistic to frame research participation as altruistic, and note the importance of social and familial context in shaping motivations for research participation. A similar theme emerges in the work of Sikweyiya and Jewkes [35]. In their study exploring participation in research around gender based violence in South Africa, they found that self interest and altruism can intersect. Participants in their study often sought to balance individual benefits (and risks) with potential benefits to others.

Some studies also included the perspectives of participants, like many who took part in the GMSH Survey, who had donated tissue samples [29, 30]. Dixon-woods and Tarrant [30] explored people’s perceptions of participation in research, drawing on accounts from individuals from three separate research projects, one of which involved the donation of a tissue sample. The findings of their qualitative analysis suggest that people’s participation in research is dependent on their perception of researchers’ commitment to mutual cooperation. In particular, participants assessed institutional factors and sought indicators that they were engaging with bona fide researchers who were conscientious about protecting the interests of those who contribute to their studies, thereby emphasising trust in researcher integrity. It is important to note that research exploring the issue of tissue sample donation has not previously been conducted with gay men.

Understanding factors which motivate gay and bisexual men to participate in behavioural research, and their perceptions of feedback on anonymous HIV antibody tests can provide valuable insights which could be used to inform the design of future studies. In order to develop our understanding of men’s perspectives on participation in research we developed a qualitative study to explore why men participate in the GMSH Survey.

Methods

Semi-structured telephone interviews were conducted between October 2011 and December 2011 with 29 men who had participated in the GMSH Survey. The College of Social Science Research Ethics Committee at the University of Glasgow reviewed and approved the study and consent procedure [Application no. 2011044].

Men were recruited in 13 licensed premises in Edinburgh (seven bars) and Glasgow (six bars). Following the GMSH Survey approach to recruitment [23], bars were visited at two different time points; in the early (7:00–9:00PM) and late (9:00–11:00PM) evening. Fieldwork staff (team of five with one team lead) spent half an hour in each venue and approached all men present to explain the nature of the study and invite them to participate. Men were asked if they took part in the GMSH Survey in 2011, if they provided an oral fluid sample during the survey, and if they were willing to take part in a telephone interview as part of a qualitative study exploring men’s perceptions of sexual health research. At this point participants were asked to provide written consent to be interviewed, as well as their first name and a contact telephone number. The fieldworker who recruited the participant then contacted the participant at an agreed date/time to conduct a telephone interview. Participants provided written consent to be interviewed at the time of recruitment, and consent was re-confirmed verbally (and audio recorded) at the start of the telephone interview. In total, 267 men were approached, of which 196 men were eligible to participate in the study. Of the men eligible to participate, 88 men were recruited to the study, and 46 men completed a telephone interview. For the purposes of this analysis, we explore the accounts of 29 men who participated in the 2011 GMSH Survey (S1 Table), the other 17 participants were excluded because they had not. Of these men, 19 had completed the GMSH questionnaire, and 10 had both completed the GMSH questionnaire and provided a saliva sample as part of the survey. Participants ranged in age from 18 to 61. No further demographic data, other than age, were collected as part of the study recruitment process. During the interviews participants were asked about their motivations for participating in the GMSH Survey and their views on the current study protocol which does not provide men with the results of their individual HIV tests. Interviews lasted approximately 30 minutes and were digitally recorded, transcribed verbatim and checked and anonymised.

To enable systematic comparisons to be made across the data, transcripts were imported into NVivo 9. The data were thematically coded and systemically charted, following the principles of framework analysis [36]. All authors contributed to the preparation of a coding framework, which was systematically applied to each transcript. Coding was regularly reviewed by the research team to identify and incorporate emergent themes. The data were then examined by Author A and Author B to identify typical quotes and common reasoning [37]. Throughout the analysis particular attention was paid to contradictory cases. Here, we present findings on i) men’s motivations for participation in the GMSH Survey, and ii) their views on HIV testing as part of the survey.

Results

Discussion

Broadly, our findings are in accordance with previous studies which suggest that people participate in research primarily for altruistic reasons [29–32]. The importance of contributing to ‘the greater good’ was emphasised throughout accounts, in particular the men conveyed a willingness to contribute to research as a means of furthering knowledge for health improvement. The findings of this study, also suggest, similar to Peel et al’s [31] findings, that participation in research is contingent on its convenience. The men in this study suggested that their participation involved little risk to themselves or investment of resources and therefore was ‘no skin off my nose’.

Although the participants in this study alluded to broadly altruistic motives and the convenience of participation, they defined their motives for participation primarily by the specific context of the GMSH Survey. Their ‘contribution’ was seen as being particularly important to an ‘imagined community’ [38, 39] of gay men. Benedict Anderson [38] conceptualised ‘imagined community’ as community extending beyond local, face-to-face ties, to include those that are ‘imagined’ to belong to a community. This is particularly useful in conceptualising gay communities as not all ties may be face-to-face, rather, may be engendered by a sense of belonging to a wider community of gay men. Indeed, the men’s emphasis on the importance of the GMSH Survey to health improvement for this ‘imagined community’ of gay men was cited as a key motive for participation. In line with the work of Hallowell et al [34] and Sikweyiya et al [35] this serves to emphasise the importance of social context in shaping motivations for participating in research.

A review of literature examining barriers and facilitators to research participation among minority ethnic groups in the United States [40] found that community ‘buy-in’ was particularly important to target populations whose participation was contingent on research being aligned with ‘community priorities’. The findings of this study similarly highlight the continued importance of buy-in from the population being targeted as a means of ensuring participation. Nevertheless, given the findings of this study, it is interesting to consider how shifts in men’s sense of belonging to gay communities may impact attitudes to research. Research has found that some gay men, particularly young men, may be ambivalent about the concept of gay communities [13, 41, 42]. Further research could explore how changes to men's relationships with and to gay communities may affect participation in behavioural research. Indeed, whatever form future behavioural surveillance research takes, the continued support and buy-in by gay men, and the gay communities they imagine themselves to be part of, continues to be vital to the ongoing success of future research efforts.

Our findings also align with wider assertions about the importance of gay community leadership in HIV prevention, as leaders and active stakeholders in the development of much HIV education, research and advocacy [3]. Men’s understandings of their participation in research as one of the ways they ‘make a contribution’ to HIV prevention, and health promotion, policy and practice in relation to gay men’s health reinforces the need for continued public engagement and dissemination of findings in a way which demonstrates to participants how they have contributed to relevant policy, practice and the development of future research.

The findings of our study also offer an alternative perspective on the importance of providing individual feedback on the results of HIV tests to those who participate in epidemiological surveillance, particularly those with undiagnosed HIV. Although increasingly recommendations suggest the importance of providing this feedback [20–22], our study highlighted how men’s understanding of the purpose of their participation as a means of contributing to behavioural surveillance research was distinct from their expectations of HIV testing, where they would expect feedback on HIV status. Indeed, most men viewed the relatively unobtrusive act of providing a saliva swab for HIV testing as a crucial part of the research and not a means of HIV testing. However, perhaps if they had provided a blood sample their expectations of feedback would have been more complex [43]. While ethical considerations continue to shape the design and development of studies, this study suggests that where participants’ understanding is well-developed and their consent informed by broader knowledge around the subject and purpose of research individual feedback on results may not always be necessary, particularly where testing is widely and freely available elsewhere.

A key limitation of the study was that we were unable to recruit men who did not participate in the 2011 GMSH Survey. The original research design included the collection of data from men who had declined to take part in the GMSH Survey. The aim of this was to explore factors that acted as barriers to participation. Seventeen men who had not taken part in the GMSH Survey were recruited to the study, however, during interviews it became clear that these men had not refused to take part, rather had not been approached during the 2011 study period. All 17 of the men explained that had they been approached, they would all have agreed to participate. Inclusion of men who refused to participate would have added robustness to the study. Further limitations of our study include the context in which the men were recruited to the study. Similar to previous research with men recruited to the GMSH Survey, exploring willingness to participate in future research [23], the sample was limited to gay men who were recruited on the commercial gay scene. It is perhaps not surprising that ‘community’ narratives are particularly prominent in this setting, due to of the long history of scene-based health promotion and health advocacy in urban areas of Scotland. The findings of the study are perhaps not more broadly relevant to groups of gay and bisexual men who do not frequent such venues. Given recent suggestions around the changing structure of gay communities, patterns of socialisation, and approaches to meeting other men [9–13], those men who do not frequent the commercial gay scene might have different views on participation as well as alternative understandings of the purpose of research.

Conclusion

Continuing to engage with gay and bisexual men, and practitioners working within these communities, as part of the process of developing new and innovative research strategies is critical to engendering support for, and trust in, future behavioural research. This underlines the need for continued engagement with stakeholder groups, both during the development phase of research studies, but also, in dissemination of the research findings. Further research is also needed to explore men’s perceptions of participation in research, and their perspectives on receiving feedback on testing, within wider contexts. Specifically, exploring if and how men’s social and sexual interactions with other men impact their understandings of research participation. For instance research could usefully explore the perspectives of those men, not visible on the commercial gay scene, who could perhaps be recruited through social media networks [18]. Indeed, it is crucial to continue to explore the opportunities and limitations of particular recruitment strategies to ensure they capture the heterogeneity of the population or ‘community’ they seek to engage.

Supporting Information

Acknowledgments

Many thanks to the people who took part in interviews for this study, and colleagues who helped with recruitment, particularly Jemma Lennox. Thanks also to Andressa Gadda for advising on early analysis, and Josh Bain and Ingrid Young for commenting on drafts.