Conducting sexualities research: an outline of emergent issues and case studies from ten Wellcome-funded projects

This letter seeks to synthesise methodological challenges encountered in a cohort of Wellcome Trust-funded research projects focusing on sexualities and health. The ten Wellcome Trust projects span a diversity of gender and sexual orientations and identities, settings; institutional and non-institutional contexts, lifecourse stages, and explore a range of health-related interventions. As researchers, we originate from a breadth of disciplinary traditions, use a variety of research methods and data sources. Despite this breadth, four common themes are found across the projects: (i) inclusivity, representations and representativeness, (ii) lumping together of diverse groups, (iii) institutions and closed settings (iv) ethical and governance barriers.

Any reports and responses or comments on the article can be found at the end of the article. sexual orientations and identities, settings; institutional and non-institutional contexts, lifecourse stages, and explore a range of health-related interventions. As researchers, we originate from a breadth of disciplinary traditions, use a variety of research methods and data sources. Despite this breadth, four common themes are found across the projects: (i) inclusivity, representations and representativeness, (ii) lumping together of diverse groups, (iii) institutions and closed settings (iv) ethical and governance barriers.

Introduction
This letter seeks to synthesise methodological issues encountered in a cohort of Wellcome Trust-funded research projects focusing on sexualities and health. Further details of these projects are listed here: https://wellcome.ac.uk/news/10-newprojects-exploring-how-sexuality-affects-health.
The links between sexuality and health are complex and diverse (Graugaard, 2017). Previous authors have explored future directions for sexuality research, highlighting challenges faced by those working in the field, such as a lack of suitable knowledge exchange and funding opportunities (Irvine, 2014;Parker, 2009;Weis, 1998). Our concerns here are more applied, as we consider the methodological challenges encountered in undertaking research in this field. The ten Wellcome Trust projects span a diversity of gender and sexual orientations and identities (cross-cutting binaries of male/female, gay/straight, masculine/feminine), settings (global south and north); institutional (from psychiatric wards to schools) and non-institutional, lifecourse stages (young through to older people), and explore a range of health-related interventions (abortion; mental health; and HIV pre-exposure prophylaxis (PrEP)). These challenges were identified during a Sexuality and Health meeting held during April 2018, at which all of the projects shared initial findings and discussed challenges faced in sexualities and health research.
As researchers, we originate from a breadth of disciplinary traditions (from anthropology through to biomedical studies), use a variety of research methods (including oral histories, ethnographic methods and meta-analyses) and data sources (from archival and secondary sources to the collection of primary data). Despite this breadth, common themes are found across the projects, and we group these under four headings: • Inclusivity, representations and representativeness • Lumping together of diverse groups • Institutions and closed settings

• Ethical and governance barriers
The core of this letter is the sections about each of the four methodological challenges listed above. But before detailing these challenges, we give a brief overview of the context in which current sexualities and health research takes place. We follow this by drawing together the four themes, finishing with some ideas for future directions.

Sexualities and health research
Human sexuality refers to the way in which people express themselves sexually and can include behavioural, social, relational, physical, erotic, biological, psychological, moral, and gendered dimensions of experience. While sexuality has played some part in all of our lives, the study of sexuality remains a relatively nascent field of research. From its origins, exploring public concerns around prostitution and venereal disease, sexuality researchers have increasingly critically interrogated sexuality 'as a broad social domain involving multiple fields of power, diverse systems of knowledge, and sets of institutional and political discourses' (Irvine, 2014 p635).
While there is growing recognition around the complexity and diversity of sexual expression, researchers continue to work within climates that place boundaries on sexuality or overlook the influence of sexuality on many life course domains, including health. Narrow definitions of sexuality can reproduce prevailing power dynamics that have historically privileged sexual behaviour taking place within the context of monogamous, procreating sexual encounters, and stigmatising other sexual expression and behaviour as risky or dangerous (Anderson, 2013), or overlooking these entirely. Critical sexualities research has challenged these conventions, leading to both an increased understanding and awareness of the diversity of behaviour and experience (i.e. not only procreative), as well as new understandings or re-conceptualisations of forms of sexuality, sexual expression and sexual relating (for example Dymock, 2012;Gabb & Fink, 2017;Parker, 2009;Ravenhill & de Visser, 2017). The links between sexuality and health are complex and diverse (Graugaard, 2017). However, a great deal of health research relating to sexualities can focus on the negative consequences of sexuality, and it is important to also acknowledge positive aspects of sexuality. For example, there is some evidence that sexual satisfaction, sexual self-efficacy, sexual self-esteem and pleasure can enhance dimensions of, physical and mental health, and overall wellbeing (Anderson, 2013).
The next section details the four methodological challenges we identified in the ten Wellcome-funded projects. We illustrate these challenges with excepts from each of the projects.

Inclusivity and representations
All of the studies included within the Wellcome cohort are, in part, a response to the lack of representations of different sexualities, genders, social classes, ethnicities, geographies, and other intersections within the existing body of literature. Even where previous studies have purported to represent diverse groups, this has often resulted in the privileging or silencing of the experience of particular groups within that cohort. For example, although the health and care needs of older LGBT people as a whole are underrepresented within health literature; even within studies focussed on LGBT health, the experiences of bisexual and transgender people are often underrepresented. Similar issues of representativeness are recurrent across different intersectionalities, leading to under/over representation of groups within research literature.
All three excerpts from our case studies below outline how the issue of inclusivity brings specific challenges in ensuring representation of intersectional experiences of sexuality, across gender and gender identity (HS/SC), ethnicity (EM/JG/RE), and other issues of gender, sexual, socioeconomic and disability neutrality or normativity (IY). Lumping 'Lumping' refers to combining the experiences of diverse groups into a single category without accounting for, or overlooking diversity within the group. Despite its pervasiveness across projects, the underlying motivations for lumping differed. In quantitative research, lumping of people with diverse experiences into a monolithic category occurs because of a need to increase statistical power within analyses and to try to mitigate the possibility that differences in the data are obscured because of insufficient statistical power. In quantitative and qualitative research, lumping can also occur because of a need to simplify complex data through categorising behaviours, experiences and attitudes. These categories can make the analysis of complex research data easier to communicate, but often sacrifice attention to minority experiences.
As the examples below also show, the values attached to these categories are often fixed and static, belying the complexity of the experiences of those to whom these categories are ascribed. As a result, while the motivations for 'lumping' are often intended to create more robust evidence, the practice can have the opposite effect.
Although some degree of 'lumping' may be viewed as inevitable in any research, a heightened awareness of the consequences of this tendency may encourage future researchers to approach sexualities research in a more nuanced way. As we show below, lumping is a particular challenge in research exploring LGBT people's experiences, which led to gaps in evidence and understanding around the health service experience and health needs of particular groups within the acronym, both historically (HS/SC) and present day (DK/RF). The practice of 'lumping' combines people with often very different needs and experiences, thus leading to research and practice which doesn't address the needs of a such a diverse group as 'LGBT people'.
A present-day consequence of lumping meant that research has tended to focus on umbrella identities formed within a particular social, political and historical context, rather than experiences or behaviours (see IY, AD, and NM). This has implications for understanding the links between sexuality, health and wellbeing. As described clearly in one of the case studies below, the practice of 'lumping' in sexualities research has direct implications for the design of policy and programmes (NM), potentially leading to less effective policy and practice.

Closed settings and institutions
Individuals may spend a large amount of time within institutions, from schools and workplaces to the hospitals and care settings more common to health research. Within institutions, sexuality is often controlled, limited and stigmatised. Carers, medical practitioners, teachers, and other gatekeepers within institutions can problematise sexuality and serve to reinforce issues of underrepresentation discussed earlier. Furthermore, as demonstrated in one case study below (CS/RS), our attitudes and treatment of sexuality within institutions can reveal a great deal about our conceptualisations of sexuality in broader society; for example, a desire to control, as opposed to explore and promote sexual wellbeing. This is emphasised in a further case study (DK/RF), where social care settings are described as a nexus of sexuality-based health and care inequalities among older people. In addition, trying to find out what happened to LGBT people in institutions raises particular challenges because of historically changing policies and legal frameworks (HS/SC).
Institutional spaces pose particular issues for sexualities researchers, although access to these spaces is essential to ensure that research is representative of a diversity of experiences; and to ensure that policy-makers and practitioners have access to evidence that can help to promote the wellbeing, including sexual wellbeing, of people within institutions.
Despite the difficulties of conducting sexualities research in institutions, the case studies emphasise the need for sexualities researchers to continue to challenge institutional barriers and research sexuality in institutional spaces. As discussed below in the context of mental health settings (JR/PR), if researchers are not granted access to patients to discuss issues relating to their sexuality, policy development will be impeded and prohibitive practices to managing patient sexuality that are not conducive to positive mental health are likely to remain. Similarly, understanding the way in which sexuality was 'treated' in the recent past, can help to avoid the catastrophic errors in care that many have received within institutional settings (HS/SC).
As the excerpts below demonstrate, sexuality does not stop at the door of institutions and does not stop with the onset of mental or physical ill health or infirmity. The 'closed door' nature of institutional settings may become increasingly blurred through social media and increased digital intimacy. It is important for such changes to be managed in a way that accounts for the links between sexuality and wellbeing, and progresses beyond narrow controlling of self-expression and identity that has been associated with institutional environments. LGBT people entering these environments, and because of the any recording of sexuality across all datasets (in addition to sexually transmitted infections).
In some cases, assumptions about the inherent sensitivity of data on sexuality made by data controllers can lead to inconsistency in the types of data made available to researchers. In turn these inconsistencies in data access can again compound issues of representation and inclusivity, or can lead to methodological issues around statistical power in analyses of minority populations (DK/RF, AD). Ethics and governance issues Sexualities research raises additional issues in research governance; for example, in the granting of ethical approval and collecting or obtaining data.

Case studies: research ethics and governance challenges faced in sexualities research
These challenges can begin even within our University institutions, where research exploring sexual practice and sexual identity can be viewed as 'sensitive' and subject to additional scrutiny (e.g. CS/RS) and greater legal restrictions on how such data is held, processed and shared. While we would not seek to downplay the importance of conducting ethically sound research, some of the case studies below contrast the perceptions of ethics committees with those of research participants (e.g. CP). In addition, blanket rules around the publication of quantitative data enforced by ethical committees and other bodies, such as suppressing small numbers to preserve potential breaches in confidentiality, while ethically sound, can inadvertently compound the issues around inclusivity and 'lumping' described above, leading to poorer quality evidence available for decision-makers (DK/RF).
The governance arrangements set in place around working with administrative data in sexuality and health research are particularly stringent. The legal restrictions are complex in England and Wales (similar arrangements exist in Scotland and Northern Ireland). There are two legal frameworks (i) the common law duty of confidence and (ii) data protection law (GDPR & Data Protection Act), both of which must be satisfied for processing of personal data for health and social care research. Under the common law duty of confidence, information about sexual orientation is considered 'confidential information'. To process confidential information, researchers have to rely on consent for disclosure in line with common law, or seek Section 251 support from the Confidentially Advisory Group (CAG) at the Health Research Authority (HRA) to set aside the common law duty of confidentiality. Under data protection law, sexual orientation would be considered not just personal data, which requires a legal basis for processing (typically 'public interest' for university research or 'legitimate interests' for commercial or charity-based research), but 'special category' personal data. This adds three further conditions for processing including compliance in terms of: (i) 'necessary purposes'; (ii) 'subject to appropriate safeguards'; and (iii) 'in the public interest'. The legal difficulties in processing personal data mean large publicly funded institutions which act as repositories for linked large administrative datasets are unable to retain sexuality in their datasets. The Secure Anonymised Information Linkage centre, funded in part by HDR-UK, the most advanced national data linkage repository in the UK, has excluded is important that we interrogate assumptions of risks and harm, and recognise that these narratives can in themselves be harmful. Building research and practice around an ethical framework may be a productive way of critiquing the harm narrative and an effective means of redirecting the conversation away from risk and harm and towards providing supportive spaces and interventions for young people. One means of doing that is to design inclusive research which involves young people, and the professionals who work closely with them, right from the start -too often research into young people's practices starts from the idea that 'something must be done '

Conclusions
The experience of Wellcome-funded researchers, upon which this letter is based, raises four key challenges for sexuality research: 1. Ensuring inclusivity in the representation and representativeness of research participants, especially around historically underrepresented categories and intersectionality.
2. Avoiding 'lumping' diverse experiences and identity under a single category or construct, (e.g. LGBT) with little attention to the important experiential variations within this category.
3. Developing a better understanding of the relationship between sexualities and health within institutional settings where sexuality is regulated and controlled 4. How to appropriately negotiate, or challenge research governance and ethics procedures which can serve to stigmatise common forms of sexuality, sexual behaviour and sexual expression.
Sexualities research implicitly involves asking challenging questions. Rather than shying away from historical, personal and political complexity, researchers can emphasise the experience of sexual variation in society, and explore how that relates to health. Approaching sexualities research with an expansive lens and an eye on historical context also helps researchers to avoid the stigmatisation that unwanted categorisation might bring. Nevertheless, exploring these questions has historically been regarded as an esoteric pursuit, despite the substantial implications for population health.

Future directions
The Wellcome-funded research projects cited in this letter have the potential to contribute to the development of national policies. We hope that research councils and other funders closely monitor the results and impact of these pilot projects and consider similar initiatives that allow researchers to explore relationships between sexualities and health, across different settings and contexts and which take into account underrepresented experiences and/or identities.
To facilitate future research, we make the following suggestions for improving sexualities research in the future: 1. Sexualities research implicitly involves working with diverse groups with complex behaviours and experiences. While categorising these or giving them acronyms can be a useful starting point in attempting to make sense of complexity, overreliance on these masks heterogeneity within groups, and attaches static values to these identities. As researchers, we need to find ways of honouring self-identities, experiences, and desires.
2. To ensure inclusivity within future LGBT research, Based on their research, Liz McDermott, Jacqui Gabb and Rachael Eastham emphasise the importance of: (i) developing a commitment to an ethnically diverse sample; (ii) finding out and understanding what is necessary to be successful; (iii) manage expectations/realistic goals as it is resource intensive; (iv) get advice from those who know, e.g. BAME youth workers.; (v) recruit ethnically diverse researchers; (vi) monitor sample, if you fail, reconfigure your strategy, e.g. stop recruiting white participants.
3. Sexualities research is often hampered by a perception that it involves 'risky' or 'sensitive' subjects, even though research participants do not always share this perception. This perception can impede funding opportunities and applications, ethics approval, and obtaining and collecting data. Ethical issues in sexualities research could be better considered by researchers and lay members with expertise in sexualities research, who currently may not feature on many institutions' ethical approval boards. We would also encourage those appraising sexualities research to look beyond the 'sexuality' aspect of the research and to consider the research merits of individual projects.
4. Sexualities researchers need to be mindful of linguistic and other traps which perpetuate stigma and the marginalisation of certain aspects of sexual expression. We also need to develop a better understanding of what works in informing and changing policy and practice. This could include developing case studies which have developed ways of communicating research findings which do not reinforce negative tropes and stereotypes about sex and sexuality.
5. We need to consider the impact of the legal restrictions on processing sexuality data for research. Policies around access to data about sexuality should be based on clearer assessment criteria, rather than the ad hoc criteria that appears to be imposed on many datasets currently. It is important to find the balance between protecting privacy with the potential benefits of research.
6. All researchers need to be mindful of future-proofing. Helen Spandler and Sarah Carr's research reminds us that we need to ask ourselves questions such as: what are the current ways that researchers collect and archive data that might make it difficult for future researchers? While it is impossible to completely 'future proof' research -we do not know what social changes lie ahead -we can certainly be more sensitive to the challenges highlighted here and ensure that we carefully record important characteristics for future generations of researchers. They are attempting to resolve challenges in in their research by taking their cue from feminist organisations and researchers who simultaneously seek to draw attention to the specificities of women's experiences while also being trans-inclusive. As researchers, we do not know how people may seek to express, define and understand their gender and sexuality in the future. However, we do know that it is very likely to shift and change in ways that we might not anticipate.
There are no easy solutions to these challenges, but by raising these issues, we hope to draw attention to the opportunities and barriers to developing new forms of knowledge about sexuality and provoke debate and discussion about how best to achieve this. Our collective work underscores the fluidity of sexuality and cautions against static representations and understandings of sexuality within sexual health interventions and research. Recognising this fluidity will improve our understanding of sexuality, and further knowledge about the links between sexuality, health and wellbeing.

Data availability
No data are associated with this article.

Author contributions
Dylan Kneale and Robert French are joint first authors. Helen Spandler, Ingrid Young and Carrie Purcell are joint third authors.
The letter is a useful, practically focused and accessibly written intervention into a growing area of research. What makes the letter compelling is its format as it allows the authors to bring together a variety of perspectives from scholars who are at the cutting edge of sexuality research, each working on a study with important implications for health and wellbeing. The letter effectively conveys both the significance of this work and the difficulty of doing it. The authors are honest about the various dilemmas and complexities they have faced in their projects, showing that, while there may rarely be "right" answers to methodological conundrums, there are nevertheless ways of improving the overall quality of work on sexuality in health research. Paying closer attention to the issues the letter identifies is a good way for researchers to start.
I enjoyed reading the letter and I have no major criticisms. My main suggestion would be to clarify the audience. Who exactly is the letter written for? Researchers, including graduate students, are the obvious audience, but this could still be specified at the outset. Towards the end of the letter, the authors mention research councils and other funders, and this audience could be highlighted earlier on too. Ethics board members might be another part of the readership worth mentioning. I think the importance and usefulness of the letter could be better conveyed in the abstract and the first paragraph of the introduction. Currently, the opening paragraph directs the reader's attention elsewhere (a website with further details about specific projects) and it could be more engaging (for example, by giving a brief overview of the projects' thematic scope). After all, Wellcome had identified "sexuality" as a priority research area, so it would be good to start (and finish) the letter on a stronger note to make the significance of this work more explicit.
The letter would also benefit from closer proofreading as it currently contains a number of typos and The letter would also benefit from closer proofreading as it currently contains a number of typos and grammatical/punctuation errors (especially in the case study boxes).
Overall, I very much welcome this open letter and the opportunity to comment on it. The authors should be commended for using this format to share the various insights from their work and consolidate them into succinct themes, which will hopefully guide current and future researchers studying health and sexuality.
Is the rationale for the Open Letter provided in sufficient detail? Yes

Does the article adequately reference differing views and opinions? Yes
Are all factual statements correct, and are statements and arguments made adequately supported by citations? Yes

Is the Open Letter written in accessible language? Yes
Where applicable, are recommendations and next steps explained clearly for others to follow? Yes No competing interests were disclosed.
sexuality and health. Researchers can also learn from those projects to avoid mistakes, to improve research on LGBT issues and to handle key challenges for sexuality research carefully.
The diverse research projects provide a solid basis for an extensive presentation of findings under four headings that are plausible and informative (Inclusivity, representations; Lumping together of diverse groups; Institutions and settings; Ethical and governance and barriers). This amount of portrays of projects limits the space that could have been used for some additional information which might be useful to understand challenges and future directions a little bit better. In the following, I will mention a few aspects that could be added to three of the four headings.
Under inclusivity, the researchers present interesting case studies that are useful to describe challenges in ensuring representation of intersectional experiences of sexuality. The case studies might be more understandable if at least one example could be presented. We learn that there was a specific ethnic recruitment strategy used (case study 2) but more information about the research design is missing. In order to avoid mistakes and to improve future research on that topic, the research design (or their interventions) could be such an example.
Lumping: It is an interesting chapter about problems due to combining diverse groups that can lead to a lack of diversity and representation. I wonder if qualitative research projects "simplify complex data" through categorising behaviour, experiences and attitudes the same way as quantitative research projects. It would be useful if the researchers could differentiate between those methodological schools or just mention the examples/case studies of social research in general. And even if "fixed categories" (see case study IY) are used findings more than "simply people behaving badly" are possible.
Closed settings and institutions: This is a very interesting chapter about practical relevance of research on sexuality and health. Instead of "sexuality does not stop at the door of institutions" it might be better to use a wider term such as sexual and gender identity.
Conclusions: To summarise the conclusions and suggestions for future research is an important and very useful part of the letter. I do not know the intention of an open letter very well but it might be interesting if some references could help to connect and to integrate those interesting findings/suggestions to the existing literature.

Does the article adequately reference differing views and opinions? Yes
Are all factual statements correct, and are statements and arguments made adequately supported by citations? Yes

Is the Open Letter written in accessible language? Yes
Where applicable, are recommendations and next steps explained clearly for others to follow? Yes