Experiences of men who have sex with men when initiating, implementing and persisting with HIV pre‐exposure prophylaxis

Abstract Introduction HIV pre‐exposure prophylaxis (PrEP) involves the use of antiretroviral medication in HIV‐negative individuals considered to be at risk of acquiring HIV. It has been shown to prevent HIV and has been available in Wales since July 2017. Measuring and understanding adherence to PrEP is complex as it relies on the simultaneous understanding of both PrEP use and sexual activity. We aimed to understand the experiences of men who have sex with men (MSM) living in Wales initiating, implementing and persisting with HIV PrEP. Methods We conducted semistructured interviews with MSM PrEP users in Wales who participated in a cohort study of PrEP use and sexual behaviour. Following completion of the cohort study, participants were invited to take part in a semistructured interview about their experiences of taking PrEP. We aimed to include both individuals who had persisted with and discontinued PrEP during the study. The interview topic guide was informed by the ABC taxonomy for medication adherence and the theory of planned behaviour. We analysed our data using reflexive thematic analysis. Results Twenty‐one participants were interviewed, five having discontinued PrEP during the cohort study. The developed themes focused on triggers for initiating PrEP, habitual behaviour, drivers for discontinuation and engagement with sexual health services. Stigma surrounding both PrEP and HIV permeated most topics, acting as a driver for initiating PrEP, an opportunity to reduce discrimination against people living with HIV, but also a concern around the perception of PrEP users. Conclusion This is the first study to investigate PrEP‐taking experiences incorporating established medication adherence taxonomy. We highlight key experiences regarding the initiation, implementation and persistence with PrEP and describe how taking PrEP may promote positive engagement with sexual health services. These findings may be useful for informing PrEP rollout programmes and need to be explored in other key populations. Patient and Public Contribution PrEP users, in addition to PrEP providers and representatives of HIV advocacy and policy, were involved in developing the topic guide for this study.


| INTRODUCTION
By the end of 2020, 37.6 million individuals were living with HIV globally and 690,000 people died from HIV-related causes. 1 In Wales, approximately 150 new cases of HIV diagnosed are each year, with 75% of these in men. 2 While no cure currently exists, advances in treatment, access to testing and treatment services and prevention methods mean that HIV is now a manageable chronic health condition with near-normal life expectancy. 3,4 One of the more recent HIV prevention methods is pre-exposure prophylaxis (PrEP).
PrEP involves the use of antiretroviral (ARV) medication in HIVnegative individuals considered to be at risk of acquiring HIV (e.g., through high-risk sexual behaviour or injecting drug use). [5][6][7] In Wales, tenofovir/emtricitabine (TDF-FTC) has been licensed as HIV PrEP since July 2017 and can be accessed through National Health Service (NHS) sexual health clinics free of charge by individuals considered to be at risk of acquiring HIV (before this, it was only available through unregulated, online purchase). PrEP is typically prescribed in 90-day supplies, and both daily (one pill a day around the same time each day) and event-based (two pills as a single dose 2-24 h before condomless sexual intercourse, followed by one pill a day thereafter until two sex-free days have passed) regimens are recommended by providers. PrEP users attending clinic to receive their prescription receive an sexually transmitted infection (STI) screen, have their renal function checked and are asked about the sexual history and PrEPtaking behaviours since their previous visit. 8,9 The latter aspects of the consultation are pertinent, as ensuring high levels of adherence to PrEP, in the absence of other HIV prevention methods, is important for maintaining a seronegative HIV status. 9,10 However, measuring and understanding adherence to PrEP is complex as it relies on the simultaneous understanding of both PrEP use and sexual activity. 11 Adherence to a pharmaceutical regimen refers to 'the process by which patients take their medication as prescribed', and is comprised of treatment initiation (when the patient takes their first dose), implementation (the extent to which a patient's actual dosing corresponds to the prescribed dosing regimen) and persistence (the length of time between initiation and the last dose). 12 The determinants of suboptimal adherence may differ across these three processes, and hence may be amenable to different forms of intervention. Furthermore, while evidence-based interventions exist for optimizing ARV medication prescribed as treatment, 13 these may not translate directly to settings where ARVs are prescribed as prophylaxisparticularly when 'optimal' adherence will depend on the extent to which an individual engages in risk behaviours and the PrEP regimen followed, which itself may vary over time.
The aim of this study, therefore, was to gain an in-depth understanding of the experiences and contextual factors that act as barriers and facilitators for the initiation, implementation and persistence with PrEP among individuals accessing it through the NHS in Wales.

| Study design and theoretical framework
We conducted a qualitative semistructured interview study of men who have sex with men (MSM) PrEP users in Wales. An interpretivist theoretical perspective was adopted, with the aim of understanding the subjective experiences of individuals through inductive reasoning.

| Participant selection
Participants were individuals receiving TDF-FTC as HIV PrEP through the NHS in Wales (a comprehensive, publicly funded health service) 14 and participating in an ecological momentary assessment (EMA) study investigating PrEP use and sexual behaviour over time. 15 Participants were approached consecutively upon completion of the EMA study. Those approached were sent study information via email, with SMS text message reminders sent to those who did not respond within 2 weeks. As an acknowledgement for their time, participants were offered a £20 gift voucher (with participants aware of this at the point of study approach).
We aimed to include between 20 and 30 participants in total, with this sample size informed by the information power model and GILLESPIE ET AL. | 1333 taking into consideration the relatively narrow aims of the research, the identification of well-defined strata (i.e., those who continued taking PrEP and those who discontinued), a theoretically informed topic guide (see below) and the strong emphasis placed on building trust and rapport with participants. 16

| Setting and data collection
All participants took part in semistructured interviews using the online video platform Zoom ® . Participants were supplied with an individual meeting ID and password (available to only the researcher and participant), gave informed consent before the interview was conducted and consent was audio-recorded. Consent procedures for the first four interviews were double-checked by F. W. Interviews were conducted on a one-to-one basis, with the aim for them to last 30-60 min. The ABC taxonomy for describing and defining adherence to medications 12 and components of the theory of planned behaviour 17 were used to inform the topic guide. Questions were also asked covering the relationship between PrEP use and sexual behaviour, in addition to levels of support around PrEP use, and the perceived impact that PrEP has had on the lives of interviewees.
The topic guide was reviewed and developed collaboratively amongst the research team and also with a stakeholder group. Field notes were taken during and after the interviews. Field notes taken during the interviews were primarily used as prompts to probe responses given by participants. See the Supporting Information Material for the topic guide.
Interviews were audio-recorded and data were transcribed verbatim by a professional transcription service.

| Data analysis
All transcripts were checked against the recording for accuracy by DG and anonymized. We conducted reflexive thematic analysis, outlined by Braun and Clarke, to analyse our interview data. 18,19 Following familiarization with the data, codes were developed by inspecting transcripts line by line, with an initial coding framework developed by D. G. Double coding was supported by coauthors F. W. and A. W. for the first four interviews to agree on the initial coding framework, accounting for alternative perspectives and subsequently by F. W. for a further three interviews to assess coding consistency.
The initial coding framework was refined in response to input from F. W. and A. W., and a revised framework was shared amongst the research team and stakeholder group for further input. Themes were developed using the 'One Sheet of Paper' or 'OSOP' technique 20 and were reviewed, refined and subsequently named. Direct participant quotes are presented with a Participant Identification number (PID) and these may include language that some readers may find triggering or offensive. While similarly aged as the majority of interviewees, there is a risk that the differing sexual orientation of the interviewer and interviewees, in addition to the interviewer never haven taken PrEP, may have resulted in lower-quality interview data through a lack of insight and shared experience. We attempted to minimize this through a team-based approach to data analysis that allowed a wider range of perspectives to influence both the topic guide and analysis. Furthermore, by conducting follow-ups with interviewees during their participation in the larger EMA study, the interviewer was able to gain trust and build rapport with interviewees before the interviews took place.

| Participants
Thirty-eight individuals were approached to take part in an interview. No response was received from 13, three declined participation and one who agreed did not turn up to the interview. In total, 21 participants were interviewed between 13 May 2020 and 6 November 2020. Interviews lasted 25-63 min (median duration 39 min). Participants were all cisgender males who exclusively had sex with other men. The majority were White British, the median age was 34 years (IQR: 27-43 years) and all except one adopted a daily PrEP regimen (with one participant taking event-based PrEP). Table 1 highlights that the participants interviewed were broadly representative of those included in the larger cohort study, with a slight underrepresentation of those in full-time employment at the time they entered the cohort study. Five interviewees had discontinued PrEP during the course of the cohort study. Figure 1 summarizes the themes developed as part of this study, and Table 2 illustrates a thematic matrix for the first theme.

| Triggers for initiating PrEP
The recognition and acceptance that they were at risk of acquiring  Influence of friends and partners PID 2: It was a friend of mine, … they got up to various things and, and they, he, he showed concern for stuff that I was doing and the places I was going, so he, he suggested it.
STI or worrying sexual encounter key trigger PID 25: Um I had a incident where I had to go on PEP, um and then after that then I thought it was more something I needed to get on as, as a preventative measure, so I didn't have to do it again, because it wasn't very pleasant

Personal choice
A key factor underpinning the decision to start PrEP was the individual taking ownership of the responsibility for reducing his risk of acquiring HIV. By doing this, there was an acknowledgement that they were also protecting others.

Protect self
Ownership of the responsibility for reducing their risk could also be framed as the individual exerting an element of control over the uncertain nature of HIV transmission.
PID 55: It is something I probably would have considered if I was single as well, um, just being sexually active with more than one partner, it seems like the kind of risk that PrEP would mitigate a little.
Protect others PID 8: to protect myself and in protecting myself, protecting others, you know, as well. Erm, so it was just to investigate and I thought, yeah, that's, you know, a reasonable step I can take.

Access to PrEP
While some individuals had heard about PrEP being available for online purchase (or available elsewhere), a driver for individuals initiating PrEP was its availability through the NHS in Wales, which was generally viewed as a more trustworthy source of both medication and advice, in addition to being free of charge.
PID 7: Having access to the prescription and, er, and regular treatment, and regular, because I've got friends in London who's got to pay for this and they, they haven't got the programme we've got in Wales… when I look at my friends now, they're taking it, stopping taking it, taking it, as and when they've got the money and stuff like that, so, it's affected then… For me, though personally it was a good thing 'cause I had access to the medication, I could see that I was getting that protection that which I needed…

| Short-and long-term drivers for discontinuation
Participants described situations where they entered relationships that they considered to be long term and monogamous, general periods of reduced sexual activity and side-effects outweighing benefits as key reasons supporting their decision to discontinue PrEP.
There was a general weighing up of the risks and benefits of continuing to take PrEP. The risk of acquiring HIV through sexual contact, and hence the need for PrEP, was viewed as transient by some participants and this led to some temporary pauses in PrEP use while HIV risk was perceived to be low.  and integrating PrEP into an existing routine. Interviews were conducted remotely. This approach has been remarked upon as reducing geographical constraints with regard to data collection and reducing some barriers towards participation (e.g., commitments that may make travelling to a face-to-face interview challenging). However, it has also been suggested that remote online interviews may also exclude certain populations (e.g., those without access to digital technology) and limit the ability for the researcher to build trust and rapport with a participant. 23 Themes incorporating initiating and discontinuing PrEP have been described in previous work exploring the barriers to PrEP use. [24][25][26] While habit formulation is an often-encouraged strategy to ensure high levels of medication adherence, 27