Sexual risk behavior among homosexual men, as measured by reported unprotected anal intercourse (UAI) with casual partners, has increased since the mid-1990s in most developed nations, including Australia.1 The extent to which this reflects a true increase in HIV risk is uncertain, however, because homosexual men have adopted complex strategies to reduce the risk of HIV acquisition, although still practicing UAI. At least 3 practices have been reported. First, serosorting has been described as the practice of having UAI with a partner believed to be of the same HIV status.2,3 Although it was initially described as a behavior of HIV-positive gay men seeking UAI with casual partners who were also HIV-positive,2 there is evidence that it is also practiced by HIV-negative gay men and that this may be increasing.4 Second, strategic positioning is broadly defined as the practice of taking the sexual position believed to be less likely to result in HIV transmission, that is, the HIV-negative men in the insertive role and the HIV-positive man in the receptive role.5 Third, negotiation around viral load has been described, in which the HIV-negative partner is more likely to agree to UAI when he believes that his HIV-positive partner's HIV viral load is low or undetectable.6
It is uncertain how commonly homosexual men become HIV infected when engaging in these strategies to minimize the risk of infection in the context of UAI. In this report, we examine the risk behaviors that newly infected homosexual men believed led to their seroconversion. In addition, we determine what proportion of cases involved one of the risk reduction strategies described previously.
Participants were homosexual men from Sydney and Melbourne in Australia who were recently diagnosed with primary HIV infection between January 2003, and September 2006. They were enrolled in 2 primary HIV infection cohort studies: The Primary HIV, Acute and Early Research: Australian Cohort (PHAEDRA) and CORE01 studies.
The methods have been described in detail elsewhere.7 Briefly, 8 weeks after their laboratory diagnosis of primary HIV infection, participants were invited to complete a survey of behavioral risk associated with HIV acquisition. A face-to-face interview was administered by trained nurses. Participants were asked to identify the high-risk event that they believed had led to their HIV seroconversion. If multiple events were reported, the study nurses were instructed to assist the participant to identify the highest risk event, as assessed by the duration before seroconversion and behaviors reported at the event. Participants were then asked about details of their sexual behavior at the high-risk event and the characteristics of the person with whom they had sex, including knowledge of the partner's HIV status at the time of the sexual contact (ie, certain HIV-negative or -positive, suspected HIV-negative or -positive, and do not know) and the partner's viral load.
Statistical analyses were performed using STATA 8.2 (Stata Corporation, College Station, TX). Descriptive analyses were carried out for frequency distribution. The role of risk reduction behaviors was examined only in those men who reported that they believed they had become infected through UAI.
A total of 232 male participants were enrolled in the PHAEDRA/CORE01 studies, and 158 (68%) agreed to complete the survey of behavioral risk factors. The mean age of these participants was 36 years (range: 20 to 63 years), and most of them (93%) self-identified as gay or homosexual. A total of 143 men (91%) were able to identify at least 1 high-risk event, and multiple events were reported by 75 (52%) men. UAI was involved in the high-risk event reported by 102 men (71%). The analysis of risk reduction behaviors was restricted to these 102 men. Among these men, 30 believed that the source was their regular partner and 63 believed that the source was a casual partner. Four men could not determine whether the source was a casual or regular partner, and data were missing on partner type in another 5 men.
Twenty-one men (21%) reported that they were certain the source partner was HIV-negative, 18 (18%) suspected that the partner was HIV-negative, 17 (17%) reported that they were certain the partner was HIV-positive, 6 (6%) suspected that the partner was HIV-positive, and 36 (37%) reported that they did not know the partner's HIV status. Of the 21 men who reported that they were certain the partner was HIV-negative, 10 reported that this partner was their regular partner, which comprised 33% of those who believed that the source was their regular partner. The relationship was of <12 months' duration in most cases (70%). In addition, 11 men reported that the partner whom they believed to be HIV-negative was a casual partner (19% of those who reported that the source was a casual partner; Table 1).
At the high-risk event, 10 (10%) men reported that their highest risk behavior was insertive UAI (4 with a regular partner, 5 with a casual partner, and 1 with an unknown partner type). Among the 21 men who believed that the source was HIV-negative, 20 (95%) reported receptive UAI, compared with 88% (53 of 60) of those who were uncertain of the source's HIV status and 88% (15 of 17) of those who reported that they were certain the source was HIV-positive (Table 2).
Twenty-one men reported that the HIV-positive partner's viral load at the high-risk event was known to them at the time of interview (12 with a regular partner, 7 with a casual partner, and 2 who reported both partner types), and of these, 9 (43%) reported that the HIV-positive partner had an undetectable viral load. These 9 men all reported receptive UAI at the event, compared with 83% of those who reported that the partner had a low viral load and those who reported that the partner had a high viral load (see Table 2).
Summary of HIV Infections After Risk Reduction Behaviors
On the basis of the recent seroconverters' retrospective accounts, serosorting was implicated in 21 HIV infections, strategic positioning in 10 infections, and reliance on the undetectable load of an HIV-positive sexual partner in 9 infections. These 40 attributions of seroconversion to risk reduction strategies reported by 39 men represent 38% of all seroconversions in which UAI was reported at the high-risk event.
More than one third of homosexual men with primary HIV infection after UAI reported that their seroconversion occurred after a high-risk event at which UAI occurred in the context of an HIV “risk reduction” strategy. Men who used one risk reduction strategy were unlikely to employ another at the high-risk event. Men who believed that their partner was HIV-negative almost always (95%) reported receptive as opposed to insertive UAI, and all 9 men who believed that their HIV-positive sexual partner had an undetectable viral load reported receptive UAI.
Serosorting has been mostly described as a behavior that HIV-positive homosexual men use when negotiating UAI.8 For HIV-negative men, serosorting is a more risky behavior because it exposes them to the risk of HIV infection. It requires sexual partners' honest disclosure of their HIV status to one another9 and the accurate and up-to-date knowledge of HIV status. Within established regular relationships, this is part of the well-described practice of negotiated safety,10,11 and it may be an effective means of avoiding HIV infection in that setting. We did not have information on whether negotiated safety was adopted by participants in regular relationships, however. In casual encounters, and in new relationships, serosorting may be risky, because knowledge of any sexual partner's HIV status may be based on assumption.12 Our finding that men attributed their HIV infection after serosorting in regular relationships of short duration supports this.
In serodiscordant UAI, the risk of HIV transmission may be reduced if the HIV-negative partner takes the insertive role or the HIV-positive partner has an undetectable viral load.13,14 Nevertheless, insertive UAI still carries substantial risk,14 and an undetectable plasma viral load does not necessarily mean an absence of virus in semen.15
Our study relied on participants' self-report. Self-report of sexual risk behaviors in homosexual men after seroconversion has been shown to underestimate the frequency of UAI.16 The fact that 29% of seroconversions occurred after events at which UAI was not reported suggests that this underreporting was present in our study. Self-report is not only confined to behaviors but to the causes of or reasons for seroconversion, however. It is possible that in some cases, men accounted for their seroconversion by selecting the event they that believed was most likely to have led to their seroconversion rather than the event that actually did so. Our data concerned sexual behavior rather than intention; thus, we cannot know whether these behaviors were intentionally practiced to minimize the risk of acquiring HIV. By interviewing individuals within 8 weeks of their HIV diagnosis, we aimed to minimize errors of recall. The PHAEDRA/CORE01 studies recruited approximately 70% of homosexual men diagnosed with primary HIV infection in high-caseload clinics in Sydney and Melbourne. It is possible that characteristics of nonresponders may be different from those we have described.
Our data demonstrate that UAI in the context of risk reduction strategies is implicated in a substantial proportion of HIV infections in homosexual men. Knowledge of the partner's serostatus was central in the occurrence of UAI. Approximately 1 in 5 HIV seroconverters in our study wrongly perceived that the source person was HIV-negative, despite the high level of HIV testing in this setting.17 It remains possible that some of these behavioral risk reduction strategies might be effective on a population level. That depends on the degree to which these behaviors replace higher risk UAI behaviors, or replace “safe sex” behaviors. Our data demonstrate that, not infrequently, risk reduction strategies seem to fail to prevent HIV infection on an individual level. This finding should be communicated to the populations of gay men who might see these risk reduction strategies as an alternative to the more effective strategy of consistent condom use.
The authors thank the nurse interviewers who contributed their time and effort, the participating clinical sites, and the participants who shared their valuable information with us.
1. Grulich A. HIV
risk behaviour in gay men: on the rise? BMJ
2. Suarez T, Miller J. Negotiating risks in context: a perspective on unprotected anal intercourse and barebacking among men who have sex with men-where do we go from here? Arch Sex Behav
3. Suarez TP, Kelly JA, Pinkerton SD, et al. Influence of a partner's HIV
serostatus, use of highly active antiretroviral therapy, and viral load on perceptions of sexual risk behavior in a community sample of men who have sex with men. J Acquir Immune Defic Syndr
4. Mao L, Crawford JM, Hospers HJ, et al. “Serosorting” in casual anal sex of HIV
-negative gay men is noteworthy and is increasing in Sydney, Australia
5. Van de Ven P, Kippax S, Crawford J, et al. In a minority of gay men, sexual risk practice indicates strategic positioning for perceived risk reduction rather than unbridled sex. AIDS Care
6. Van de Ven P, Mao L, Fogarty A, et al. Undetectable viral load is associated with sexual risk taking in HIV
serodiscordant gay couples in Sydney. AIDS
7. Volk JE, Prestage G, Jin F, et al. Risk factors for HIV
seroconversion in homosexual men in Australia
. Sexual Health
8. Parsons JT, Schrimshaw EW, Wolitski RJ, et al. Sexual harm reduction practices of HIV
-seropositive gay and bisexual men: serosorting, strategic positioning, and withdrawal before ejaculation. AIDS
. 2005;19(Suppl 1):S13-S25.
9. Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis
10. Kippax S, Crawford J, Davis M, et al. Sustaining safe sex: a longitudinal study of a sample of homosexual men. AIDS
11. Kippax S, Noble J, Prestage G, et al. Sexual negotiation in the AIDS era: negotiated safety revisited. AIDS
12. Parsons JT, Severino J, Nanin J, et al. Positive, negative, unknown: assumptions of HIV
status among HIV
-positive men who have sex with men. AIDS Educ Prev
13. Garnett GP, Rottingen JA. Measuring the risk of HIV
14. Vittinghoff E, Douglas J, Judson F, et al. Per-contact risk of human immunodeficiency virus transmission between male
sexual partners. Am J Epidemiol
15. Zhang H, Dornadula G, Beumont M, et al. Human immunodeficiency virus type 1 in the semen of men receiving highly active antiretroviral therapy. N Engl J Med
16. Keet IP, Albrecht van Lent N, Sandfort TG, et al. Orogenital sex and the transmission of HIV
among homosexual men. AIDS
17. Jin FY, Prestage G, Law MG, et al. Predictors of recent HIV
testing in homosexual men in Australia
. HIV Med