Deciding what’s risky – Australian gay men pay more attention to the sexual act than to their partner’s HIV status or viral load

HIV-negative gay men in Sydney, Australia perceive the risk of HIV transmission to be different in different sexual contexts, according to a study by Dr Limin Mao and colleagues, reported in the May issue of AIDS & Behavior. But men pay considerably more attention to condom use, withdrawal and sexual role than they do to their partner’s reported HIV status or viral load in assessing the risk of transmission.

While understanding of the factors which affect the risk of HIV transmission has become increasingly complex, much less is known about whether this biomedical knowledge is known, interpreted or acted upon by gay men. “There is little research to explore their nuanced understandings of HIV transmission risk associated with specific practices, particularly in relation to non-condom-based risk reduction practices,” say the authors.

They therefore included a series of risk assessment questions in the 2010 Sydney Gay Community Periodic Survey. This regular, cross-sectional survey recruits men from bars, clubs, saunas, sexual health clinics and so forth. For this analysis, only the 1212 men who said that they were HIV-negative were included (men who did not know their status or who had been diagnosed with HIV were excluded).

The survey asked men to rate the likelihood of HIV transmission during sex with a new casual partner. A series of specific sexual behaviours were described, in the context of the partner describing himself as being ‘HIV negative’, ‘HIV positive with an undetectable/low viral load’ or ‘HIV positive with a detectable/medium-high viral load’.

Transmission risk was rated on a five-point scale, from ‘very unlikely’ to ‘very likely’. Individual respondents had different answers, but in terms of the average (median), activities were rated as follows.

Very unlikely

• Protected anal intercourse – whatever the partner’s HIV status or viral load

Unlikely

• (no behaviours)

Neither likely nor unlikely  

• Insertive unprotected anal intercourse and withdrawal – whatever the partner’s HIV status or viral load
• Insertive unprotected anal intercourse and ejaculation – with an HIV-negative or undetectable HIV-positive partner
• Receptive unprotected anal intercourse and withdrawal – with an HIV-negative partner

Likely

• Insertive unprotected anal intercourse and ejaculation – with a detectable HIV-positive partner
• Receptive unprotected anal intercourse and withdrawal – with an HIV-positive partner, whatever his viral load
• Receptive unprotected anal intercourse and ejaculation – with an HIV-negative or undetectable HIV-positive partner

Very likely

• Receptive unprotected anal intercourse and ejaculation – with a detectable HIV-positive partner

The results show that the respondents took account of both the specific sexual behaviour and the partner’s health status in their assessments of risk. However the sexual behaviour carried considerably more weight than the partner’s HIV status or viral load.

This was shown in a multivariate model in which an increased perceived risk of transmission is indicated by a larger ‘beta coefficient’. (A beta coefficient of 0 means no increased risk at all.)

In a model which controlled for the partner’s HIV status and viral load, and took anal intercourse with condoms as the baseline, a greater risk was perceived during insertive unprotected anal intercourse and withdrawal (beta coefficient 0.66), insertive unprotected anal intercourse and ejaculation (beta coefficient 0.82), receptive unprotected anal intercourse and withdrawal (beta coefficient 0.87) and receptive unprotected anal intercourse and ejaculation (beta coefficient 1.09).

In a model which looked at the partner’s reported health status (controlling for the sexual behaviour), the beta coefficients were smaller, indicating that the partner’s health status contributed less to men’s assessment of risk.

Taking an HIV-negative partner as the point of comparison, transmission risk was moderately raised during sex with an HIV-positive man with an undetectable viral load (beta coefficient 0.09) and an HIV-positive man with a detectable viral load (beta coefficient 0.22).

One possible explanation for this finding is that a number of men have a healthy scepticism about sexual partners who self-report to be HIV negative – they view all new partners as potentially risky. Another possible explanation is that men have limited awareness of the effect of HIV treatment and undetectable viral load on HIV transmission.

Overall, the researchers found that younger men perceived risks to be lesser than did older men. Similarly, men with less education perceived risks to be lower than men with a university education.

Moreover, men who had not had an HIV test in the past year also tended to evaluate risks as being lesser than other men.

The investigators wished to examine the relationship between risk assessments and sexual behaviour. After controlling for the other factors associated with lower risk assessments, they examined whether men who reported unprotected anal intercourse with a casual partner in the past six months had a different view of transmission risk, as compared to men who had consistently used condoms with casual partners.

This was the case – reporting recent unprotected sex was associated with lower risk assessments (beta coefficient ­–0.36), equivalent to an average decrease of about half a point on the 5-point scale.

“To the best of our knowledge, this is the first study to suggest that HIV risk is differentially perceived by HIV-negative gay men,” concluded Dr Mao. “Perceived risk of HIV infection was found to be associated with variations in anal intercourse practices and to a lesser extent, with variations in a new casual partner’s HIV status and viral load levels.”

Moreover these context-specific variations suggest a hierarchy that largely, but not exactly, resembles the variations estimated by scientists, she says.

More Australian data

A second Australian study, reported in the same issue of AIDS & Behavior, explores similar themes. These data come from an online survey of 2081 gay men from all parts of Australia, recruited via social networking and gay dating websites in 2009. Importantly, the analysis includes both HIV-positive and HIV-negative men.

A previous analysis from this survey was described on Aidsmap in 2011.

A minority of men agreed that HIV transmission was unlikely for a man who always took the insertive role (25.2%), if his partners withdraw before ejaculation (12.0%), or if his partners have undetectable viral loads (17.3%). However, a majority of men agreed that being told by a sexual partner that he had the same HIV status would make them less concerned about HIV (69.4%) and more inclined to consider not using a condom (57.0%).

Similarly to the study from Sydney, these beliefs were more common in men who reported unprotected anal intercourse with a casual partner in the past six months.

The beliefs were also more common in men with diagnosed HIV.