Relative safety II Risk and unprotected anal intercourse among gay men with diagnosed HIV Adam Bourne Catherine Dodds Peter Keogh Peter Weatherburn Gary Hammond Original Research Report Acknowledgements Our greatest debt is owed to those men who took part in this study. They made this research possible through their willingness to share their time, and talk honestly about their social and sexual lives. In addition, we are thankful to those the agencies that helped to recruit participants and made interview space available. These are: The Armistead Centre in Liverpool, Body Positive North West and George House Trust in Manchester, Eddystone Trust in Plymouth, GMFA in London, Yorkshire Mesmac and Skyline Service in Leeds, Staffordshire Buddies in Stoke-on-Trent, Terrence Higgins Trust Cymru in Cardiff and Swansea, and Terrence Higgins Trust West in Bristol. As ever, we are also grateful for the prompt and accurate transcription provided by Mandy Eaton and Gary Hammond. A draft of this report was circulated for consultation. Thanks to Will Nutland and Ford Hickson for valuable and thought-provoking feedback. This research was funded by the Terrence Higgins Trust as part of CHAPS, a national HIV prevention initiative funded by the Department of Health. Adam Bourne Senior Research Associate www.sigmaresearch.org.uk www.tht.org.uk This report is available to download at: www.sigmaresearch.org.uk/go.php/reports/report2009d/ Published by Sigma Research © March 2009 ISBN: 1 872956 98 X Contents 1 Introduction 2 2 Methods and sample 5 2.1 Methods 5 2.2 Sample description 6 3 Perceived harms associated with UAI 7 3.1 Physical harm 7 3.2 Harms to social and moral identity 10 3.3 Discussion 13 4 Managing the risk of infection 14 4.1 Condoms as a sole risk management tactic 15 4.2 Means of managing risk other than condom use 16 4.3 Lacking the capacity to manage risk 18 4.4 Discussion 19 5 Managing identity 20 5.1 Managing sexual pleasure 20 5.2 Managing moral and social integrity 22 5.3 Managing reactions from others 24 5.4 Managing the risk of criminal prosecution for HIV transmission 27 5.5 Discussion 28 6 Conclusions 30 References 33 1 Introduction In 1999 Sigma Research published Relative safety: an investigation of risk and unprotected anal intercourse among gay men diagnosed with HIV (Keogh et al. 1999). This study explored the social, psychological and cultural meanings associated with unprotected anal intercourse (UAI) among men with diagnosed HIV. It highlighted both the complexity of sexual interaction for men with diagnosed HIV, and the many potential costs and benefits perceived by them. Now, with more than 24,000 homosexually active men diagnosed with HIV in the UK (Health Protection Agency 2008), a figure that is set to increase in years to come, it is vital that agencies involved in HIV prevention interrogate their own beliefs about UAI and ensure that their interventions meet the needs of men with diagnosed HIV. Sexual enjoyment and satisfaction are important to men irrespective of their HIV status. However, more than two thirds (71%) of gay and bisexual men with diagnosed HIV have had problems related to sex in the previous year, and more than half (52%) are currently unhappy about their sex lives (Weatherburn et al. 2009). The most frequent problems are having no sex, or very little, often linked to poor self-image or low self-confidence and a loss of libido or interest in sex. Other related problems with sex commonly reported by men with diagnosed HIV include anxieties about passing on HIV infection to partners, rejection by potential sexual partners and difficulties disclosing to partners. In addition, some had concerns about potential prosecution for onward transmission of HIV during sex. Men with diagnosed HIV are thus faced with many obstacles to the fulfilment of their own sexual desires, a right that all people maintain regardless of their HIV status. Faced with this complex balancing act, many men with HIV avoid anal intercourse altogether, or do so only with condoms, or with other men with diagnosed HIV. Others engage in UAI with varying perceptions of the risk that such behaviour may pose. This report seeks to update and expand our understanding of how gay men with diagnosed HIV seek sexual pleasure during UAI while minimising the chances of infecting others, causing further physical harm to themselves, or acting in ways that threaten their own sense of moral integrity. Previous research has shown that when presented with the statement: “As an HIV positive man I should feel an extra responsibility not to pass on HIV to another person”, 87% of men with diagnosed HIV agreed (Stephenson et al. 2003: 9). However, when presented with the statement: “HIV positive gay men have more responsibility to practice safer sex than HIV negative men”, far fewer (35%) agreed. The mis-match between these responses suggests that while diagnosed men tend to seriously consider their own responsibility in avoiding transmission, they do not hold that undiagnosed men should be absolved of theirs. In addition to this risk of HIV transmission, Keogh et al. (1999) identified the emotional and psychological harm that could emerge if men discovered they were the source of infection for another individual, coupled with a further concern for the risk of social censure (being seen by others to be behaving irresponsibly). Taken together, this evidence demonstrates that the majority of men with diagnosed HIV wish to avoid participating in HIV transmission. The Gay Men’s Sex Survey (GMSS) indicates that men who have received a positive HIV test result are more likely to engage in UAI than those who have not tested positive, and men diagnosed positive do so with significantly more partners than negative and untested men (Weatherburn et al. 2008). Given that homosexually active men engaging in UAI with high numbers of partners are more likely to acquire HIV than other men, it is possible that men’s sexual activities after diagnosis simply mirror what they did prior to diagnosis. However, Weatherburn et al. (2008) also found that while engagement in UAI was more likely among men diagnosed with HIV compared to negative or 2 RELATIVE SAFETY II untested men, one third of men with diagnosed HIV had not engaged in UAI at all in the previous year. Such men do not form part of the current study, and they would no doubt tell a very different story of risk perception and response. Of those men with diagnosed HIV that do engage in UAI, not all risk exposing men without HIV to their infection. In GMSS 2002 (Hickson et al. 2003a), among 1133 respondents with diagnosed HIV, only 34.5% said they had probably or definitely participated in sero-discordant UAI (sdUAI) in the past year. Among those men with diagnosed HIV that do engage in UAI, the reasons for doing so are highly contingent upon a number of situational factors. Numerous studies have demonstrated the symbolic nature of condomless sex and have made clear the value that many gay men, regardless of HIV status, place on being able to have anal intercourse with their partners without the use of condoms (Schilder et al. 2008, Flowers et al. 1997). Some see the cessation of condom use as a milestone in a romantic or long-standing relationships (Flowers et al. 2006). Sex without condoms enables a greater sense of intimacy that extends beyond skin-to-skin contact to incorporate an enhanced self-comfort and comfort with one’s partner. Others have suggested that factors such as these, and their role in establishing relationship stability and satisfaction can be more important than any infection-related risk (Cusick & Rhodes 2000). These findings exist in addition to the widely reported feeling that the physical sensation is more satisfactory without condoms. Keogh et al. (1999) highlighted that engagement in UAI among men diagnosed with HIV was rarely the result of ‘regrettable accidents’ or of being ‘carried away’, but rather emerged as a result of considered risk assessment and subsequent behaviour modification. The harm-reduction tactics and strategies employed included elements of negotiated safety (Kippax et al. 1993, Hickson et al. 1992), avoiding the insertive role during intercourse, and withdrawal before ejaculation when insertive in UAI. One of the most common tactics among the Keogh et al. (1999) study respondents, was purposeful selection of sexual partners sharing the same HIV sero-status. By ‘sero-sorting’ their sexual partners in this manner, men felt they could eliminate the possibility of being the source of primary HIV infection for another person. Data from Hickson et al. (2007) show that among men with diagnosed HIV, engagement in UAI in the previous year was far more common with partners they knew to have diagnosed HIV, than with partners known to be tested negative. Research undertaken with men with diagnosed HIV in London clinics who engaged in UAI found that 58% had sought it only with other diagnosed positive men within the last 12 months, which was significantly associated with their subsequent behaviour (Elford et al. 2007). This desire to engage in UAI with men of the same sero-status is examined by Frost et al. (2008) who found that men seeking sero-concordant partners report a greater sense of intimacy when having sex with other diagnosed positive men. However, the success of sero-sorting in preventing HIV transmission relies on both parties being accurately aware of each others’ HIV status. Parsons et al. (2005) reported that the pervasive problem with sero-sorting is that men having UAI overestimate the likelihood that those they are having sex with are also diagnosed with HIV (see also Adam et al. 2008, Elford et al. 2007, Zablotska et al. 2007). Sero-concordancy can be especially hard to establish in casual, anonymous sex environments where the opportunities for communication are limited (Keogh et al. 1998, Richters 2007). Concerns about sero-sorting are sometimes raised by those working in the field, given that participation in UAI with partners of the same HIV status does not help to reduce the risk of other STI transmission. Health professionals, health promoters and researchers also highlight the possibility of acquiring an additional strain of HIV: a so called ‘superinfection’. Despite individual case reports of superinfection in the literature, there is little agreement about its prevalence, or what factors affect its acquisition (Piantadosi et al. 2007). Some researchers have demonstrated superinfection rates between 4% and 9% in particular cohorts (Kraft et al. 2008, Chohan et al. 2005, Smith et al. RELATIVE SAFETY II 3 2004). Others found no evidence of superinfection amongst 101 people whose HIV was resistant to treatment (Bezemer et al. 2008), with a further study finding no evidence of superinfection among a sample of 49 individuals with diagnosed HIV having unprotected intercourse with their sero- concordant partners (Willberg et al. 2008). This latter study also demonstrated a clear relationship between long-term exposure to a single partner’s HIV via receptive intercourse, and an enhanced immune response to that virus. It is not clear to what extent men diagnosed with HIV are aware of research developments in this area, how they perceive the risk of superinfection and how, if at all, they respond to it (although, as an exception, see Adam et al. 2005). Therefore, the current study also examines the extent to which men’s awareness of superinfection impacts upon their risk perceptions and sexual behaviours. There have been a number of other clinical developments since the Keogh et al. (1999) study that have the potential to impact upon the sex lives of men diagnosed with HIV. The advent of PEP provides a method of preventing sero-conversion not widely available at the time of the last study. Körner et al. (2005) found that PEP was viewed as an additional line of defence when trying to avoid HIV infection among negative or untested men, however there has been little work examining how men diagnosed with HIV understand PEP and if (or how) they utilise this knowledge when having sex with men of negative or unknown HIV status. In early 2008, a consensus statement on behalf of the Swiss Federal Commission for HIV/AIDS (Vernazza et al. 2008) suggested that individuals diagnosed with HIV who are on anti-retroviral therapy with an undetectable viral load cannot transmit HIV through sexual contact. This suggestion was, however, based on a review of clinical studies relating to heterosexual HIV exposure, and even then restricted by a number of explicit caveats. No similar suggestion was made that related to sex between homosexually active men, but that has not prevented a great deal of debate within the HIV sector about the significance of the statement and the implications it may have for future HIV prevention activities. It remains unclear however if, and how, gay men diagnosed with HIV have incorporated knowledge of these developments into their perceptions of, and responses to, risk when engaging in UAI. In addition to clinical developments, there have been some important criminal policy developments over the last 10 years that have the potential to impact upon the sex lives of men diagnosed with HIV in the UK. In 2001 the first prosecution for reckless transmission of HIV occurred. Similar cases have featured prominently in the gay press in subsequent years. Dodds et al.(2009) reported how men with and without diagnosed HIV viewed such prosecutions, but as yet there is little information on how such prosecutions have influenced sexual behaviour. The present study seeks to address gaps in our understanding by asking men with diagnosed HIV about their awareness of criminal prosecutions, and how such awareness has impacted on the sex they have. The following chapter explains how the study was undertaken, outlines the broad topic areas addressed during the interviews, and describes the sample of men who took part. Chapter 3 outlines the range of harms that men with HIV perceive when engaging in UAI. Chapters 4 and 5 explore the ways in which men responded to these perceived harms, firstly those relating to the risk of onward HIV infection, or superinfection, and latterly those concerning the potential for harms to their personal and social identities. Chapter 6 considers the implications of these findings for health promotion interventions targeting men with HIV, and with homosexually active men more broadly. 4 RELATIVE SAFETY II 2 Methods and sample This study explores the experience of unprotected anal intercourse amongst homosexually active men living with diagnosed HIV. It draws on interview data from forty-two homosexually active men in England and Wales, spanning a range of years since diagnosis, and living in areas of differing HIV prevalence. 2.1 MetHOds Recruitment took place in collaboration with a number of community-based HIV organisations across England and Wales (see Acknowledgements). These agencies distributed materials promoting the study to service users with diagnosed HIV either by email or in person. To be eligible men had to be diagnosed with HIV, and had to have participated in unprotected anal intercourse with a man in the last year. Approximately one hundred men volunteered to take part and, in an attempt to gain a diverse sample in terms of age and time since diagnosis, forty-five were subsequently invited to interview. An approximately even split between men living in areas of higher HIV prevalence (London and Manchester) and lower HIV prevalence was also ensured, given that men’s experiences of sero-sorting, disclosure, support and HIV-related stigma can be influenced by attitudes to HIV expressed in local gay scenes and services. Of the forty-five invited to interview, three were later excluded as they did not meet the study criteria. Participants’ confidentiality was assured and all were reimbursed expenses of £20. Ethics approval for this project was granted by the Faculty of Humanities and Social Sciences Research Ethics Committee at the University of Portsmouth. Interviews lasted 1-2 hours, were hosted at Sigma Research, at collaborating agency offices, or at participants’ homes, and took place in the following cities and towns: Bristol, Exeter, Leeds, Liverpool, London, Manchester, Stoke-on-Trent, and Swansea. Each interview was digitally recorded and transcribed verbatim. The interview covered the following topics: • the impact of HIV diagnosis on respondents’ sex life; • HIV status disclosure to sexual partners and others; • awareness of sexual partners’ sero-status; • details about the most recent experience of UAI; • awareness and experience of HIV risk reduction tactics; and • awareness and experience of HIV prevention technologies (e.g. PEP, PrEP). A full thematic content analysis was undertaken on each transcript, and then compared across transcripts by two researchers working independently. Following initial reading, and re-reading, an analysis template was developed and used as the basis for a synopsis of each transcript. The synopses were then used to identify significant, recurring themes. In the following chapters, blocks of bold text are verbatim quotes from respondents. The quotes selected give an overall sense of key issues raised by various respondents, and cannot be taken to be representative of all responses in a single theme. Quotes were selected for their clarity and descriptive purpose. We have given brief demographic descriptions after each quote. This is both to illustrate the range of men contributing their perspectives and also to show that men sharing an age, locality, and time since diagnosis do not necessarily have the same views and experiences. RELATIVE SAFETY II 5 2.2 sAMple desCRIptIOn The characteristics of the forty-two men whose data were included for analysis, are summarised below: Age Time since diagnosis Range 18 – 58 Range <1 – 23 yrs Median 37 yrs Median 6.75 yrs Area of residence Education London 15 ‘O’ Levels / GCSE or less 14 Manchester 8 ‘A’ Levels / college diploma 16 Lower prevalence areas 19 Degree or higher 12 Ethnic group Relationship status White British 33 No current regular partner 24 White other 3 Sero-discordant partner 7 Mixed 2 Sero-concordant partner 11 Black African 2 Asian British 1 Chinese 1 Term used for sexuality Number of male partners in the last year Gay 37 Range 1-562 Homosexual 4 Median 27 Queer 1 6 RELATIVE SAFETY II 3 perceived harms associated with UAI This chapter describes the potential harms that respondents associated with having unprotected anal intercourse. Men’s perceptions of the potential outcomes of UAI are clearly linked to knowledge of their own HIV infection. This knowledge presents men with an altered risk landscape – one in which they often feel a tremendous burden to consider and manage risk in ways that sharply contrast with their own experience prior to diagnosis. After diagnosis, men’s risk perceptions continue to undergo significant changes, influenced by their own emotional state, experience, and information gained as they make decisions about being sexually active individuals with HIV. Therefore during the research interview, each respondent was able to reflect only on his own feelings at that particular point in time, in relation to his own experience of HIV up to that point. Men diagnosed for longer periods of time tend to focus their attention on different components of risk compared to those who are more recently diagnosed. Just as men consider risk differently at different times after their diagnosis, so too do men find that the importance of sex in their lives can change once they know they have HIV. What the men in this study collectively express is the value they hold in sexual intercourse, and the key role it performs in achieving a sense of fulfilment and well-being in their lives. They are thus tasked with deciphering how this desire for enjoyable and fulfilling sex fits alongside their knowledge of the risk they know they pose to others, and the risks they face themselves. In addition to the potential physical harms that can arise from UAI, many men held pervasive concerns about how sexual behaviour and HIV status interacted in ways that threatened to undermine their self-regard, as well as damaging their standing among others. Sustaining an identity as a responsible gay man while simultaneously protecting oneself from HIV-related stigma was of critical importance to a substantial proportion of participants. An examination of men’s views on the non-physical risks associated with their participation in UAI forms the final part of this chapter. 3.1 pHysICAl HARM All respondents were aware that their own HIV could be transmitted to others through unprotected anal intercourse, but at the same time they held widely varying interpretations about the extent to which facilitating factors could increase or decrease the risk of transmission. Their views on their own vulnerability to new infections and HIV superinfection were even more diffuse. The following two sub-sections of this chapter describe the physical, infection-related elements of harm perceived by the men taking part in this study. 3.1.1 Physical harm to others The men taking part in this study were, on the whole, acutely aware of the potential harm they posed to others during unprotected anal intercourse. They recognised HIV as a transmissible disease and were aware of the impact it could have on a person’s life. I was HIV positive and the last thing that I wanted to do was put somebody through, you know infect somebody and let them go through what I had just been through. Of oxygen masks, intensive care and being incubated and all that stuff. I did not want to inflict that on anybody. [Early 40s, low prevalence area, diagnosed 8 years] RELATIVE SAFETY II 7 For those who had been recently diagnosed (a shorthand term we use to represent those diagnosed for two years or less), concerns about the onward infection of HIV to others permeated their narratives of risk perception and reduction. It was this concern that led many to avoid sex altogether for weeks, months or even years following first HIV diagnosis. But there was always at the back of my mind that sort of feeling of being unclean and not wanting to infect anyone else you know? I suppose that idea of why I can’t have sex with anybody ever again. [Early 40s, high prevalence area, diagnosed 2 years] Men who had been diagnosed for longer periods of time were no less aware of the physical harm they posed to others, but their concerns about transmission of the virus tended not to result in avoidance of sex. They had accepted the potential for harm as part of their sex lives and described taking routine measures to reduce transmission risk without having to exhaustively consider the consequences each time they had sex. Most were unwilling to gamble with another person’s health and well-being, even if that individual was willing to do so themselves. Among all respondents, notions of ‘gift-giving’ were uniformly rejected. And we were having sex and I was fucking him [without condoms] and he was saying ‘I want you to make me positive’. And instantly I lost all interest and I couldn’t get out of there quick enough. [Early 40s, high prevalence area, diagnosed 2 years] When considering the potential harms associated with UAI, most men referred solely to the harm they might cause to partners who were not already infected with HIV. There was only one individual who commented on the risk he felt he posed to sero-concordant sexual partners in terms of superinfection. It would always be in the back of my mind you know ... there’s a sort of one in ten billion risk that I’d give him a version of the virus that he hasn’t got. Which, you know, I mean he’s younger and I certainly wouldn’t want to damage his health you know. I’m considerably older and I’m not too concerned. [Mid 50s, high prevalence area, diagnosed 14 years] The overwhelming majority of respondents went out of their way to clarify that they aimed to avoid transmitting HIV to all of their sexual partners. All men recognised that anal intercourse without condoms carried a greater risk of transmission than intercourse with condoms, or no intercourse at all, but felt that they generally took enough care, or in some cases, aimed to take better care (through the tactics explored in detail in chapter 4) to reduce this risk to a degree acceptable to them. 3.1.2 Physical harm to self A small proportion of respondents raised concern about the risk of contracting another sexually transmitted infection when engaging in UAI. I think that you tend to forget sometimes when you are HIV positive you think ‘Oh well I have got it now and we can get it on with other guys’. But you do tend to forget that there are other things out there. [Mid 40s, low prevalence area, diagnosed 12 years] However, such views were certainly in the minority, with most men reporting that after receiving a positive HIV diagnosis, anything else paled into insignificance. Where men had acquired subsequent infections, they had been identified and treated quickly due to routine HIV clinic appointments, and they expected that this would continue to happen in future. Fewer than one fifth raised some concern about the risk posed to them by hepatitis C co-infection. 8 RELATIVE SAFETY II