One-in-eight diagnosed HIV-positive gay and bisexual men in London are seeking unprotected anal intercouse (UAI), according to a City University study published electronically in the July edition of Sexually Transmitted Diseases . However, although the majority of 'barebackers' intend to have UAI only with other HIV-positive men (i.e. 'serosort'), in practice many are having UAI with men of different or unknown HIV status. The investigators suggest that future prevention initiatives target these men, as well as men who do not seek UAI but also have it unintentionally.
Although there have been several studies from the United States examining intentional UAI , or 'barebacking', amongst HIV-positive gay or bisexual men, this study from Jonathan Elford and colleagues is the first to examine 'barebacking' amongst diagnosed HIV-positive gay men in the United Kingdom, looking specifically at men in London.
The study's objectives were to:
- examine the extent that diagnosed HIV-positive gay men in London intentionally seek UAI,
- estimate the contribution that 'barebacking' makes to total sexual risk-taking
- and compare data from a clinic-based survey to that obtained from the internet.
Who they asked
Between October 2002 and May 2003, the investigators asked 864 out of 1001 HIV-positive men attending London's Royal Free HIV outpatients clinic to fill out a self-administered written questionnaire. A total of 620 men completed the questionnaire, of whom 528 men defined themselves as gay or bisexual, or had had sex with another man in the previous year. A total of 481 men provided enough information to be included in the investigator's analysis.
The investigators point out that because the majority of people diagnosed with HIV receive their care at an HIV outpatient clinic, the clinic sample is "broadly representative" of gay men living with diagnosed HIV in London.
In addition, for four weeks in May and June 2003, the investigators asked exactly the same questions to men using two popular gay dating sites - gaydar and gay.com - with an additional question about HIV status. A total of 595 men living in London completed the online questionnaire, of whom 579 provided information about their HIV status. Of the 67 men who said they knew they were HIV-positive, 66 provided enough information to be included in the analysis.
The investigators had previously reported that gay men in London were no more likely to meet partners for high risk sex on-line than off-line, and wanted to find out whether previous studies that had sampled men online had over-represented the true prevalence of 'barebacking'.
Defining 'barebacking'
Although many gay men use the term 'barebacking' to describe the specific behaviour of anal sex without a condom, whether or not this was intentional, the researchers from City University London - in common with most other sexual health researchers - define 'barebacking' as "intentionally seeking or engaging in unprotected anal sex".
To avoid confusion, questions about 'barebacking' were phrased to include this definition. In order to asses their intention to 'bareback', participants were asked, "Have you...intentionally look[ed] for anal sex without a condom with a man in the last 12 months?" In order to assess 'barebacking' in practice, participants were also asked whether they had actually engaged in UAI in the previous three months.
Who wanted to 'bareback'?
The following data come from the "broadly representative" sample of HIV-positive gay men attending the HIV outpatient's clinic.
Of the 481 HIV-positive men surveyed, around one-in-eight, (59, or 12%) had intentionally sought to have UAI with another man in the previous year.
The majority – 34 men (58% of 'barebackers', and 7% of the whole sample) – said they had planned to have UAI only with another HIV-positive man. This is also known as 'serosorting', and is considered to be a controversial harm-reduction method.
Another 22 men (37% of 'barebackers', and just under 5% of the whole sample) said that their partner's HIV status had not been important. Just three men (5% of 'barebackers', and 0.6% of the whole sample) said they had intentionally been seeking to have UAI with an HIV-negative man.
Of these 25 men who sought to have UAI with HIV-negative men or men of unknown HIV status, 14 said they had also sought to have UAI with other HIV-positive men as well.
Who actually had unprotected anal sex?
In total, 85 HIV-positive men reported UAI in the previous three months with casual partners of unknown or HIV-negative status. Professor Elford tells aidsmap that the majority of the men did not know their partner's HIV status. "And of those who did not know their partner's status," he says, "more than half assumed he was also HIV-positive."
Unsurprisingly, the investigators found that the men who had intended to 'bareback' in the past year were significantly more likely to report UAI with a casual partner in the previous three months compared with the men who had not intentionally looked for 'bareback' partners in the past year. In fact, a total of 36 of the 59 men who intended to 'bareback' actually had UAI with men of unknown or HIV-negative status, compared with 49 of the 422 men who did not intentionally seek UAI (61% v. 12%; p
Importantly, there were some stark differences between intention and practice. For example, of the 34 men who said they had intended to only have UAI with another HIV-positive man in the past year (i.e. their 'barebacking' intention had a low HIV transmission risk to their partners), sixteen had actually had UAI with men of unknown or HIV-negative status in the previous three months.
Conversely, of the 25 men who had intended to have UAI with men of unknown or HIV-negative status in the past year, two actually only practiced UAI with another HIV-positive man in the previous three months.
The contribution of 'barebacking' intention to high risk behaviour
The majority of sexual risk-taking that was reported in the previous three months was attributed to men who had no intention to 'bareback'.
Of the 85 men who reported UAI with partners of unknown or HIV-negative status in the previous three months, 49 (58%) had not intended to 'bareback' in the past year. "This serves to remind us that a substantial proportion of high-risk sex reported by HIV-positive gay men cannot be attributed to barebacking," the investigators write.
Although a significant proportion of men who had intended to 'serosort' in the past year actually did only have UAI with HIV-positive partners in the previous three months (13/34; 38%), a larger proportion of these men did not consistently 'serosort': 13 (38%) had UAI with HIV-negative and unknown status partners, as well as HIV-positive partners; and 3 (9%) practiced UAI solely with HIV-negative and unknown status partners.
Of the 25 men of who had intended to 'bareback' with any partner, regardless of HIV status, in the past year, 20 actually did so in the previous three months.
The role of the internet
"Compared with the clinic sample, a substantially higher proportion of HIV-positive men in the internet sample reported barebacking," write the investigators. In fact, 32 of the 66 diagnosed HIV-positive gay men in the internet sample (48.5%) had intentionally sought to 'bareback'.
The investigators also found that all the men in the clinic sample who intended to 'bareback' had used the internet, and that the men who planned to 'serosort' met all of their partners online. "Consequently," the investigators write, "barebackers will be over-represented in an internet sample."
They also note that, rather than promoting 'barebacking', the internet "seems to facilitate serosorting among HIV-positive men who intentionally seek UAI with other HIV-positive men."
Comment
The phenomenon of 'barebacking' has been blamed for contributing to the recent rise in HIV infections and in the resurgence of other sexually transmitted infections, like syphilis, amongst gay men in the West.
“Our research shows that in terms of HIV transmission, much of the risk among HIV-positive gay men occurs in unplanned encounters," notes Professor Elford. "HIV prevention initiatives therefore need to target HIV-positive men who intentionally seek UAI as well as those who do not express such intentions”.
However, this study shows that there is a difference between an intention to 'bareback' and actually having UAI with a partner of different or unknown status.
In particular, the study suggests that HIV-positive men who are choosing to 'serosort' – a practice that is facilitated by the relative ease of disclosure on the internet via dating sites – are not doing so consistently. Is this failure to always follow-through due to an inability for both partners to disclose when they do not meet via the internet, or because men who prefer to 'bareback' are unable to practice safer sex when presented with a heat-of-the-moment opportunity?
HIV prevention intiatives should also, therefore, start to focus on the differences between intention and practice, and examine why gay men are unable to follow-through with their harm-reduction plans.
Elford J et al. Barebacking among HIV-positive gay men in London. Sexually Transmitted Diseases 33 (7): epublished July 2006.