Gay Men's Hiv Risk Reduction Practices: the Influence Of

GAY MEN’S HIV RISK REDUCTION HOLT

GAY MEN’S HIV RISK REDUCTION PRACTICES: THE INFLUENCE OF EPISTEMIC COMMUNITIES IN HIV SOCIAL AND BEHAVIORAL RESEARCH Martin Holt

Since the mid-1980s, Australian social researchers have investigated the sexual practices of gay men, describing those that protect men from HIV or put them at risk of infection. Ground-breaking (and controversial) publications have highlighted a variety of ways in which gay men protect themselves and their partners, including condom use and non-condom-based risk reduction strategies. HIV social research in Australia has been heavily influenced by a distinctive network of experts or epistemic community with shared principles and beliefs and a commitment to influencing policy and practice. This epistemic community has articulated a ‘social public health’ view of HIV that emphasises partnership, agency, understanding practices and reflexivity. This approach has clashed with those of other epistemic communities, notably around ideas of relapse and unsafe sex. This article uses the examples of negotiated safety and serosorting to illustrate this Australian epistemic community’s approach to HIV risk reduction among gay men.

INTRODUCTION

Know your epidemic, know your response, is the strategy advocated by United Na- tions Program on HIV/AIDS (UNAIDS), to plan a national HIV response: garner research evidence to identify which groups are most at risk of HIV and why, develop targeted programs and interventions in collaboration with affected communities, and monitor impact and effectiveness. However, even Michel Sidibé, UNAIDS Executive Director, acknowledges that generating knowledge of an HIV epidemic is insufficient for building an effective response; one also needs to understand the conditions under which research evidence affects policy and practice (Buse, Dickin- son, & Sidibé, 2008). Why do some ideas come to prominence while others remain ignored? Buse and his colleagues remind us that policy emerges from interactions between institutions that shape how decisions are made, the framing of problems

Martin Holt is with the Centre for Social Research in Health, University of New South Wales, Australia. The author thanks Peter Aggleton, Susan Kippax, and two anonymous reviewers for their constructive comments on earlier versions of this article. The Centre for Social Research in Health is supported by the Australian Government Department of Health and Ageing. Address correspondence to Martin Holt, Centre for Social Research in Health, University of New South Wales, Sydney, NSW 2052, Australia. E-mail: [email protected]

214 GAY MEN’S HIV RISK REDUCTION 215 and potential solutions and the interests of individuals and groups who are affected by change. HIV responses, including research, are highly dependent on national political conditions and international dynamics. If you look at any field of expertise, it is important to remember that it could have been otherwise; if different people had been involved, with different interests, values and training, the knowledge produced and the pre-eminence of particular ideas would almost certainly have been different (Latour, 2005). In this article, I contrast the types of knowledge generated by Australian researchers about gay men’s HIV risk reduction practices with international research on similar topics. In doing so, I suggest that the partnership model prominent in Australian social and behavioral research constitutes an epistemic community with shared values and beliefs (Haas, 1992). Through the promotion of certain types of collaborative research, the epistemic community that I describe has sought to foster a particular type of HIV response in Australia. It has also intervened in international debates about how to think about and conduct HIV prevention. When I began writing this article I wanted to provide an overview of the distinctive contribution made by Australian researchers to understanding risk reduction practices; the practices that gay men engage in to reduce the risk of HIV transmission, particularly during unprotected sex (Mao et al., 2011). Out of necessity, I have restricted my focus to the examples of negotiated safety and serosorting. My hope is that in analyzing these topics, readers will get a sense of how the Australian response to HIV has been shaped by a particular type of social research, and how the response could have been very different.

EPISTEMIC COMMUNITIES

The concept of epistemic communities is derived from political science (Haas, 1992). It is one way to account for the influence that networks of knowledge-based experts have in defining complex problems at a national or international level, framing debates and suggesting solutions. An epistemic community is defined by Haas (1992, p. 3) as “a network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain.” An epistemic community consists of professionals from a variety of disciplines and backgrounds who share normative and principled beliefs, shared causal beliefs, shared notions of validity (criteria for evaluating knowledge), and a common policy enterprise to improve the problem or issue area with which they are engaged (Haas, 1992). I have chosen this concept because it is an existing framework with which to analyze the practice of knowledge generation and influence; it is not the only framework and it is not without its limitations. My point is that HIV research can and should be analyzed for its history, assumptions, and contingencies. An analysis of an epistemic community involves describing its membership and shared beliefs, tracing its actions, and evaluating its impact, including its success in resisting policy directions which contradict its principles (Haas, 1992). The analysis offered here is retrospective, focusing on publications and related commentary to illustrate the principles, beliefs, and common policy enterprize of an HIV social research epistemic community. The approach of this epistemic community is con- trasted with that of researchers in other countries, particularly those from the United States. 216 HOLT

AUSTRALIAN HIV SOCIAL RESEARCH

The first case of HIV in Australia was diagnosed in 1982 (Mindel & Kippax, 2013). That first case, and the majority that followed, were among gay men. The Aus- tralian research response to HIV followed community activism and the funding of community-based AIDS Councils in the mid-1980s (Kippax, Connell, Dowsett, & Crawford, 1993; Mindel & Kippax, 2013). These organizations have remained cen- tral to the Australian response (Altman, 1994). In 1985, the AIDS Council of New South Wales approached social psycholo- gists and sociologists at Macquarie University to acquire better information to guide HIV prevention activities for gay and bisexual men (Kippax, Connell et al., 1993). This request initiated the first large social science project on gay men and HIV in Australia: the Social Aspects of the Prevention of AIDS (SAPA) project. The project relied on collaboration between the researchers, community organization represen- tatives and the state government. Many of those involved were gay community activ- ists. The partnership developed in SAPA laid the foundations for the epistemic community of Australian HIV social research. The project’s overarching finding was that the social and sexual involvement of gay and bisexual men with each other (what became known as gay community attachment) was associated with the adoption of safe sex practices. This finding, later duplicated in other countries, gave the nascent partnership confidence that HIV education and prevention needed to be community- based in order to be effective (Kippax, Connell et al., 1993). Partly because of SAPA’s success, when the federal government created three national HIV research centers in 1990, the Macquarie University group was invited to join the National Centre in HIV Social Research (Kippax, Connell et al., 1993; Mindel & Kippax, 2013). The creation of a HIV social science center was quite un- usual internationally, and institutionalized social research early in the epidemic. This commitment was underlined when the Australian Research Centre in Sex, Health and Society was established in Melbourne in 1993. Why did social science gain such a strong role in the Australian response? The federal health minister, Neal Blewett, was responsible for overseeing the development of Australia’s early national HIV strategy (Mindel & Kippax, 2013; Sendziuk, 2003). Blewett, a political scientist, relied heavily on his main advisor, Bill Bowtell, who argued against a medically-dominated model and supported a collaborative approach to HIV. Blewett’s 1984/85 visit to the U.S. also appears to have been in- structive. After the trip, Blewett favored the partnership approach he had seen in California. As well as medical and public health expertise, Blewett encouraged a leading role for affected communities in HIV education and prevention, supported by social science. If the health minister had been unsympathetic to social science, or had preferred the public health control model, then it is unlikely HIV social research would have been so prominent in Australia. So what are the normative and principled beliefs and common policy enterprise that characterize the epistemic community of Australian HIV social research (Haas, 1992)? The early publications of the SAPA researchers (Kippax, Crawford, Dowsett, & Connell) describe a set of beliefs and convictions that have been rehearsed and revisited over the past 20 years:

• A commitment to partnership or social action between affected communities, researchers, and governments (Dowsett, 1996; Kippax, Connell et al., 1993). GAY MEN’S HIV RISK REDUCTION 217

This came to be referred to as a “social public health” model, as opposed to a traditional, disease control public health model (Sendziuk, 2003). • Recognizing affected communities as critical active agents in the response to HIV (Dowsett, 1996; Kippax, Connell et al., 1993; Kippax & Race, 2003). • The importance of understanding practices, not just behavior, i.e., what people do, what it means to them and why it means what it does in particular contexts (Dowsett, 1996; Kippax, Connell et al., 1993; Kippax & Race, 2003). • Reflexivity about the interdependence of researchers and other actors (communities, organizations) within the HIV response (Dowsett, 1996; Kippax, Connell et al., 1993).

This list is not conclusive; I am not suggesting that those who participated in the Australian social research epistemic community uniformly adhered to these principles. Nor are these features unique to Australian social researchers, given that they are informed by a broader history of critical public health practice and activism. However, they are sufficiently distinguishing to mark a particularly Australian approach to understanding the practices of gay men in negotiating the risk of HIV. The examples of negotiated safety and serosorting illustrate this.

NEGOTIATED SAFETY

In the early 1990s, researchers from San Francisco announced a “relapse” from safe sex among gay men (Ekstrand & Coates, 1990; Stall, Ekstrand, Pollack, McKusick, & Coates, 1990). An analysis showed that three-quarters of men in the city had sustained safe sex practices from 1984 to 1989, but over that period 19% of men reported a change from low to high risk practices (Stall et al., 1990). This minor- ity was labelled as relapsing from safe sex. High risk practices were defined as unprotected anal intercourse (UAI) within a monogamous relationship, UAI between regular partners who always used condoms with other men, and UAI with casual partners. The assertion that gay men were relapsing to unsafe sex provoked sharp international debate (Hart, Boulton, Fitzpatrick, McLean, & Dawson, 1992). The term relapse was seen to imply failure, recidivism, and a return to bad habits. The clas- sification of all unprotected anal sex as unsafe was also seen as particularly problematic, given that much of the UAI classified as high risk appeared to be occurring within monogamous relationships and between men with the same HIV status (Hart et al., 1992). However, for the researchers advancing the relapse hypothesis, any unprotected anal intercourse was problematic; as Stall et al. (1990, p. 1182) then put it, “from a prevention point of view any form of unprotected anal sex is undesirable.” The Australian contribution to the debate followed key observations in Austra- lia, Canada, and the UK. Canadian and Australian researchers found that gay men were much more likely to use condoms with casual partners than regular partners (Connell et al., 1989; Schechter et al., 1988). In addition, British research described strategies used by gay men to prevent HIV transmission in nonexclusive sexual relationships. These strategies included regular partners agreeing to only have unprotected anal sex with each other and to always have safe sex with casual partners (Hickson et al., 1992). The Australian SAPA team then entered the fray, publish- ing an article in the journal AIDS (Kippax, Crawford, Davis, Rodden, & Dowsett, 1993). 218 HOLT

The Australian analysis showed that most men had maintained or adopted safe sex practices during the period between 1986/1987 and 1991 (Kippax, Crawford et al., 1993). Unprotected anal intercourse remained more common with regular partners than with casual partners, but the vast majority of UAI with regular partners (84%) was with partners believed to be the same HIV status. The majority of men in relationships reported explicit agreements with their partners about sex within and outside the relationship. Over one third of men in relationships agreed to be monogamous, another third agreed to practice safe sex with each other and with casual partners, and just under a quarter always had safe sex with casual partners but allowed unprotected sex within the relationship. When HIV-negative partners agreed to limit unprotected sex to within the relationship, the authors labeled this “negotiated safety.” The contrast between relapse and negotiated safety reveals the key principles underpinning the Australian social research epistemic community. Relapse was defined from a traditional public health perspective in which any unprotected anal intercourse was seen as risky, and a failure of individuals to adopt safe sex. The relapse perspective saw safe sex as a fixed set of practices that must be maintained. In contrast, negotiated safety emphasized the agency of gay men in deciding how to prevent HIV. Negotiated safety proposed that UAI could be practiced safely under certain conditions, allowing for a re-expansion of gay men’s sexual repertoires (Connell et al., 1989). It suggested that gay men were making use of knowledge of HIV status and transmission routes to generate protective practices in addition to condom use. The concept of negotiated safety was not without its detractors. Although Kip- pax, Crawford et al. (1993) acknowledged the strategy relied on accurate knowledge of HIV status, honesty and trust, critics were skeptical that gay men could successfully negotiate or maintain such agreements (Ekstrand et al., 1993; Gold, 1996; Ridge, 1996). Damien Ridge (1996) suggested that gay men who felt pressured to have UAI with their partners or participate in an open relationship (to maintain trust or keep a partner happy) would be incorrectly classified as having successfully and voluntarily negotiated an agreement. (In effect, Ridge was criticizing the SAPA researchers on the lack of attention to negotiated safety as a practice, because it was being inferred from survey data). U.S. researchers went further, saying that partners could make false assumptions about each other’s HIV statuses, and negotiated safety could in fact be ‘negotiated danger’ (Ekstrand et al., 1993, p. 281). The Australian proponents of negotiated safety responded with additional data showing that the majority of Australian HIV-negative men with concordant regular partners avoided UAI with casual partners, and those with explicit agreements were most likely to avoid UAI with casual partners (Crawford, Rodden, Kippax, & Van de Ven, 2001; Kippax et al., 1997). Because of the close links between the social research community and AIDS Councils in Australia, negotiated safety was taken up and promoted to gay men by the Victorian AIDS Council and the AIDS Council of New South Wales (ACON) from 1994 onwards (McNab, 2009). AIDS Vancouver in Canada also ran a campaign promoting negotiated safety in 1994, and there was international commentary in the gay press. Researchers found that knowledge of HIV status and negotiated agreements were much less common among gay men in Amsterdam and London than their Australian peers (Davidovich, de Wit, & Stroe- be, 2000; Elford, Bolding, Maguire, & Sherr, 1999). These researchers concluded that better promotion of the recommended conditions of negotiated safety to British and Dutch gay men would be helpful. In 2009, Australian researchers published an GAY MEN’S HIV RISK REDUCTION 219 analysis of cohort data, finding that HIV-negative men who consistently practiced negotiated safety were no more likely than men who avoided UAI to become infected with HIV (Jin et al., 2009). Some U.S. researchers have remained skeptical of negotiated safety, describing multiple problems in gay men trying to establish and maintain relationship agreements (Hoff & Beougher, 2010). Rises in unprotected sex and HIV incidence during the late 1990s in many countries with gay male epidemics made the promotion of negotiated safety difficult (Elford, 2006). However, its practice has been documented and incorporated into the flurry of research on serosorting, further illustrating the approach and influence of the Australian social research epistemic community.

SEROSORTING

The use of serosorting to describe a deliberate HIV prevention strategy emerged in the early 2000s. Suarez and Miller (2001) published a review of strategies used by gay men to reduce HIV transmission, particularly when engaging in UAI. Serosort- ing was defined as discussing HIV status with potential partners (particularly casual partners) and limiting UAI to seroconcordant partners. It was grouped with negotiated safety as “rational risk taking.” Suarez and Miller acknowledged that serosorting was an attempt to reduce harm when having UAI, but highlighted the strategy’s heavy reliance on accurate knowledge of HIV status, effective HIV disclosure, and honesty. Serosorting in this early formulation echoed many of the features of negotiated safety—largely based on quantitative survey data, it was conceptualized as a strategy generated by gay men to allow UAI but limit the risk of HIV transmission, reliant on knowledge of HIV status and negotiation between partners. Clearly, in the decade since the relapse debate it had become more acceptable to think about non-condom-based risk reduction strategies. This perhaps shows the influence of the Australian social research epistemic community and its European and Canadian peers. We should also note the power of naming a concept; just as negotiated safety had crystallized the idea of gay men’s risk reduction strategies within relationships, raising the visibility of Australian social research, the naming of serosorting focused attention on risk reduction by gay men with casual partners and highlighted U.S. behavioral research. The labelling of serosorting in the U.S., its quasi-epidemiological name, and its heavy reliance on accurate knowledge of HIV status encouraged U.S. researchers to investigate it at a time when the U.S. Centers for Disease Control and Prevention were pushing a serostatus approach to the epidemic that heavily emphasized mass HIV testing (Janssen et al., 2001). Serosorting has subsequently been documented in multiple, international stud- ies of gay men (e.g., Crepaz et al., 2009; Elford, 2006). Like negotiated safety, serosorting has generated debate and criticism. Given the greater certainty about HIV- positive serostatus, commentators have often been more comfortable with the idea of HIV-positive serosorting as a prevention strategy rather than HIV-negative serosorting. The Australian social research epistemic community did not initially engage with the term serosorting. The occasional use of HIV status to negotiate concordant UAI during casual sex was initially described as the “simplistic” application of the “sophisticated procedure” of negotiated safety to casual encounters (Van de Ven, Prestage, French, Knox, & Kippax, 1998, p. 817), implying that casual partners could not effectively negotiate UAI or trust in the disclosure of HIV status. However, this position later softened, acknowledging that the negotiation of condom use or 220 HOLT

UAI on the basis of HIV disclosure, particularly by HIV-positive men, could work in some casual sex contexts (Prestage et al., 2001). The first Australian publication to use the term serosorting observed that it had become more common among HIV- negative gay men in Sydney during 2002–2005 (Mao et al., 2006). It repeated the concern that serosorting could increase HIV transmission because men who engaged in the practice could incorrectly believe they were HIV-negative. A later Australian analysis picked up a different problem already highlighted in US research; up to one in six gay men who were serosorting were guessing or assuming that their casual partners were seroconcordant (Zablotska et al., 2009). The initially wary approach to serosorting by the Australian social research community informed how the Australian HIV sector responded to the issue. For example, the Australian Federation of AIDS Organisation’s 2001 resource, No Wor- ries, and ACON’s 2008 You Just Don’t Know campaign acknowledged serosorting, but both campaigns discouraged the practice. Further analysis seemed to confirm this cautious approach; while serosorting offered some protection, HIV-negative men who practiced serosorting with casual partners were about three times more likely to seroconvert than men who avoided UAI (Golden, Stekler, Hughes, & Wood, 2008; Jin et al., 2009). In recent years there has, however, been a gradual, pragmatic acceptance in Australia that serosorting must be addressed in HIV prevention aimed at gay men, although the default position still emphasises that it is much less safe than condom use. This reflects the recognition that UAI between casual partners is gradually be- coming more common among Australian gay men (Holt et al., 2012; Mao et al., 2011). Researchers engaged in the social research epistemic community have there- fore shifted position, recommending education on the effective use of risk reduction strategies. Educational organizations have echoed this move and have begun to address serosorting directly, such as in ACON’s 2012 Know Your Risk resource, which explicitly describes the different strategies that can be used in different situations.

DISCUSSION

I have suggested that the network of social researchers, community educators, and policymakers that came together in mid-1980s to investigate sexual practices among Australian gay men was the basis for a distinctive epistemic community (Haas, 1992). The epistemic community’s principles included partnership with affected communities and understanding the practices that gay men generate collectively in response to HIV. The epistemic community documented the uptake of safe sex in the 1980s and then the emergence of non-condom-based risk reduction strategies in the 1990s. The description of the negotiated safety strategy demonstrates the epistemic community’s core principles. The proposal that gay men could find safe ways to have unprotected sex, and that these could be built upon in education and prevention, was in stark contrast to a traditional public health approach which viewed all unprotected sex as problematic. While some greeted negotiated safety with skepticism, the proliferation of research on serosorting in the early 2000s shows the influence of the Australian social research community and its European and Canadian peers. The analysis offered here was restricted to published research and the examples of negotiated safety and serosorting; a more thorough study of the development of epistemic communities in HIV would need to fully assess debates within the field, their relationship to policy development and country-specific epidemics and the GAY MEN’S HIV RISK REDUCTION 221 views of stakeholders (Haas, 1992). This has been done in histories of the Austra- lian community and political responses to HIV (Sendziuk, 2003). Out of necessity, many significant contributions have been omitted from my partial account, particularly the role of qualitative research in elaborating on and challenging the findings of survey-based research. A history of HIV research, and the role of social science, is long overdue. I have suggested that analyzing the development of epistemic communities within the HIV field is one way to understand how HIV research comes to influence policy and practice (Buse et al., 2008). The Australian social research epistemic community is not a static entity, nor uncontested; some of its founding members have moved into other fields, some have retired. New researchers, educators, and policymakers have joined the epistemic community, changing its membership and focus. However, the epistemic community’s core principles have arguably endured. The study of risk reduction practices has revealed that gay men continue to generate strategies that accommodate scien- tific developments, often in ways that deviate from public health recommendations (Kippax & Race, 2003). The strategies may not be fully protective, and they may be difficult to practice effectively, but they demonstrate gay men’s agency within the constraints in which they find themselves. Current developments will inevitably generate new responses and practices among gay men. The growing utilization of biomedical prevention approaches, such as pre-exposure prophylaxis and treatment as prevention, will influence notions of safety, protection, and risk. The Australian social research epistemic community will undoubtedly be among those following and engaging gay men to understand the novel forms of risk reduction that emerge.

REFERENCES

Altman, D. (1994). Power and community: Or- sis. AIDS, 23, 1617–1629. doi:10.1097/ ganisational and cultural responses to QAD.0b013e32832effae AIDS. London: Falmer. Davidovich, U., de Wit, J. B. F., & Stroebe, W. Buse, K., Dickinson, C., & Sidibé, M. (2008). (2000). Assessing sexual risk behaviour of HIV: Know your epidemic, act on its young gay men in primary relationships: politics. Journal of the Royal Society of The incorporation of negotiated safety and Medicine, 101, 572–573. doi:10.1258/ negotiated safety compliance. AIDS, 14, jrsm.2008.08k036 701–706. Connell, R. W., Crawford, J., Kippax, S., Dowsett, Dowsett, G. W. (1996). Practicing desire: Homo- G. W., Baxter, D., Watson, L., & Berg, R. sexual sex in the era of AIDS. Stanford, CA: (1989). Facing the epidemic: Changes in Stanford University Press. the sexual lives of gay and bisexual men in Ekstrand, M., & Coates, T. J. (1990). Maintenance Australia and their implications for AIDS of safer sexual behaviors and predictors of prevention strategies. Social Problems, 36, risky sex: The San Francisco Men’s Health 384–402. doi:10.2307/800822 Study. American Journal of Public Health, Crawford, J. M., Rodden, P., Kippax, S., & Van 80, 973–977. doi:10.2105/AJPH.80.8.973 de Ven, P. (2001). Negotiated safety and Ekstrand, M., Stall, R., Kegeles, S., Hays, R., De- other agreements between men in relation- Mayo, M., & Coates, T. (1993). Safer sex ships: Risk practice redefined. International among gay men: What is the ultimate goal? Journal of STD & AIDS, 12, 164–170. AIDS, 7, 281–282. doi:10.1258/0956462011916965 Elford, J. (2006). Changing patterns of sexual be- Crepaz, N., Marks, G., Liau, A., Mullins, M. M., haviour in the era of highly active antiretro- Aupont, L. W., Marshall, K. J., … Wolitski, viral therapy. Current Opinion in Infectious R. J. (2009). Prevalence of unprotected Diseases, 19, 26–32. anal intercourse among HIV-diagnosed Elford, J., Bolding, G., Maguire, M., & Sherr, L. MSM in the United States: A meta-analy- (1999). Sexual risk behaviour among gay 222 HOLT

men in a relationship. AIDS, 13, 1407– Kippax, S., Noble, J., Prestage, G., Crawford, J. 1411. M., Campbell, D., Baxter, D., & Cooper, Gold, R. (1996). Dangerous liaisons: Negotiated D. (1997). Sexual negotiation in the AIDS safety, the safe sex culture, and AIDS educa- era: Negotiated safety revisited. AIDS, 11, tion. Venereology, 9, 102–103. 191–197. Golden, M. R., Stekler, J., Hughes, J. P., & Kippax, S., & Race, K. (2003). Sustaining safe Wood, R. W. (2008). HIV serosorting in practice: Twenty years on. Social Science & men who have sex with men: Is it safe? Medicine, 57, 1–12. Journal of Acquired Immune Deficiency Latour, B. (2005). Reassembling the social: An Syndromes, 49, 212–218. doi:10.1097/ introduction to actor network theory. Ox- QAI.0b013e31818455e8 ford: Clarendon. Haas, P. M. (1992). Introduction: Epistemic com- Mao, L., Crawford, J. M., Hospers, H. J., Prestage, munities and international policy coordina- G. P., Grulich, A. E., Kaldor, J. M., & Kip- tion. International Organization, 46, 1–35. pax, S. C. (2006). “Serosorting” in casual doi:10.1017/S0020818300001442 anal sex of HIV-negative gay men is note- Hart, G., Boulton, M., Fitzpatrick, R., McLean, worthy and is increasing in Sydney, Austra- J., & Dawson, J. (1992). ‘Relapse’ to un- lia. AIDS, 20, 1204–1206. doi:10.1097/01. safe sexual behaviour among gay men: A aids.0000226964.17966.75 critique of recent behavioural HIV/AIDS Mao, L., Kippax, S., Holt, M., Prestage, G., research. Sociology of Health and Illness, Zablotska, I., & De Wit, J. (2011). Rates 14, 216–232. doi:10.1111/1467-9566. of condom and non-condom based anal ep11343702 intercourse practices among homosexu- Hickson, F.C.I., Davies, P. M., Hunt, A. J., Weath- ally active men in Australia: Deliberate erburn, P., McManus, T. J., & Coxon, HIV risk reduction? Sexually Transmitted A.P.M. (1992). Maintenance of open gay Infections, 87, 489–493. doi:10.1136/sex- relationships: Some strategies for protec- trans-2011-050041 tion against HIV. AIDS Care, 4, 409–419. McNab, J. W. (2009). Negotiating HIV preven- doi:10.1080/09540129208253112 tion: The talk, test, trust story and beyond. Hoff, C., & Beougher, S. (2010). Sexual agreements Unpublished PhD thesis, The University of among gay male couples. Archives of Sex- New South Wales, Sydney. ual Behavior, 39, 774–787. doi:10.1007/ Mindel, A., & Kippax, S. (2013). A national stra- s10508-008-9393-2 tegic approach to improving the health of Holt, M., Lee, E., Prestage, G. P., Zablotska, I., de gay and bisexual men: Experience in Aus- Wit, J., & Mao, L. (2012). The converging tralia. In S. O. Aral, K. A. Fenton, & J. A. and diverging characteristics of HIV-posi- Lipshutz (Eds.), The new public health and tive and HIV-negative gay men in the Aus- STD/HIV prevention: Personal, public and tralian Gay Community Periodic Surveys, health system approaches (pp. 339–360). 2000–2009. AIDS Care, 25, 28–37. doi:10. New York: Springer. 1080/09540121.2012.686598 Prestage, G., Van de Ven, P., Grulich, A., Kippax, Janssen, R. S., Holtgrave, D. R., Valdiserri, R. O., S., McInnes, D., & Hendry, O. (2001). Gay Shepherd, M., Gayle, H. G., & De Cock, men’s casual sex encounters: Discussing HIV K. M. (2001). The serostatus approach to and using condoms. AIDS Care, 13, 277– fighting the HIV epidemic: Prevention strat- 284. doi:10.1080/09540120120043928 egies for infected individuals. American Ridge, D. (1996). Negotiated safety: Not negoti- Journal of Public Health, 91, 1019–1024. ated or safe? Venereology, 9, 98–100. Jin, F., Crawford, J., Prestage, G. P., Zablotska, I., Schechter, M. T., Craib, K. J., Willoughby, B., Imrie, J. C., Kippax, S. C., … Grulich, A. Douglas, B., McLeod, W. A., Maynard, E. (2009). Unprotected anal intercourse, M., … O’Shaughnessy, M. (1988). Pat- risk reduction behaviours, and subsequent terns of sexual behavior and condom use HIV infection in a cohort of homosexual in a cohort of homosexual men. American men. AIDS, 23, 243–252. doi:10.1097/ Journal of Public Health, 78, 1535–1538. QAD.0b013e32831fb51a doi:10.2105/ajph.78.12.1535 Kippax, S., Connell, R. W., Dowsett, G. W., & Sendziuk, P. (2003). Learning to trust: Australian Crawford, J. (1993). Sustaining safe sex: responses to AIDS. Sydney: UNSW Press. Gay communities respond to AIDS. Lon- Stall, R., Ekstrand, M., Pollack, L., McKusick, L., don: Falmer Press. & Coates, T. J. (1990). Relapse from safer Kippax, S., Crawford, J., Davis, M., Rodden, P., sex: The next challenge for AIDS preven- & Dowsett, G. (1993). Sustaining safe sex: tion efforts. Journal of Acquired Immune A longitudinal study of a sample of homo- Deficiency Syndromes, 3, 1181–1187. sexual men. AIDS, 7, 257–264. GAY MEN’S HIV RISK REDUCTION 223

Suarez, T., & Miller, J. (2001). Negotiating risks in tralian and New Zealand Journal of Public context: a perspective on unprotected anal Health, 22, 814–818. doi:10.1111/j.1467- intercourse and barebacking among men 842X.1998.tb01499.x who have sex with men—Where do we go Zablotska, I. B., Imrie, J., Prestage, G., Craw- from here? Archives of Sexual Behavior, 30, ford, J., Rawstorne, P., Grulich, A., … 287–300. doi:10.1023/a:1002700130455 Kippax, S. (2009). Gay men’s current Van de Ven, P., Prestage, G., French, J., Knox, S., practice of HIV seroconcordant unpro- & Kippax, S. (1998). Increase in unprotected anal intercourse: Serosorting or tected anal intercourse with casual partners seroguessing? AIDS Care, 21, 501–510. among Sydney gay men in 1996–1998. Aus- doi:10.1080/09540120802270292