London study finds that gay men answering HIV questionnaire on-line less likely to have been tested

The sexual risk behaviours of gay men who are HIV-negative or untested who completed a questionnaire asking about their sexual behaviour on the internet are significantly different from those of HIV-negative or untested gay men completing the same questionnaire in community venues, according to a UK study published in the April 1^st edition of the Journal of Acquired Immune Deficiency Syndromes. The study also found that there were significant social and demographic differences between gay men surveyed on-line or in the community, and the investigators conclude that the internet is a useful tool for providing health education to difficult-to-reach populations of gay men.

In early 2002, 879 gay men using seven central London gyms popular with gay men completed a pen and paper questionnaire. At the same time, 1218 gay men in London using the chatrooms on the website gay.com and gaydar.co.uk completed an identical electronic questionnaire. In both the paper and on-line questionnaires, the men were asked to provide information on their age, ethnicity, employment, education, sexual orientation, HIV testing history, HIV treatment optimism, recreational drug use, and use of the internet to seek sex. The men were also asked if they had had unprotected anal sex in the previous three months and, if they had, whether this was with a partner of the same, different, or unknown HIV status.

The investigators found that there were significant differences between the men recruited through the gyms and their on-line sample. Men completing the electronic questionnaire were significantly less likely to have been tested for HIV (67.5% versus 80.2%, p

However, for men who were HIV-positive, the investigators found that there were no significant differences between the gym and internet samples regarding age, sexual orientation, drug use, or HIV optimism.

When the investigators analysed responses to the questions regarding high-risk sexual behaviour, they found that men completing the internet questionnaire were significantly more likely to report unprotected anal sex in the previous three months with a man of unknown or different HIV status. This was true for men who had tested HIV-negative (26.9% versus 18.6%), men who were untested (35.6% versus 19%) and HIV-positive men (47.2% versus 42%, all p

Unsurprisingly, the investigators found that the men questioned on the internet were more likely to say that they used websites to meet sexual partners. Furthermore, the investigators found that both HIV-positive and HIV-negative men who used the internet to seek sex were more likely to report high-risk sex (47% versus 40% and 28% versus 18.7%, p

In multivariate analysis, the association between high-risk sex and being surveyed on-line for HIV-positive men ceased to be significant. However, it remained significant for both HIV-negative and untested men.

The investigators comment that their investigation shows that it is possible to conduct behavioural research on-line with gay men. They add that their on-line questionnaire appeared to be particularly useful at reaching never-tested gay men, who were less likely to identify as gay than other men, but more likely to report high-risk sex. “The on-line survey appeared, therefore, to reach a group of never-tested men living in London who are at risk of HIV infection yet who may not be reached by traditional, off-line, health promotion initiatives and survey methods,” add the investigators.

The investigators conclude that their survey showed that HIV-negative and untested men surveyed on-line were more likely to report high-risk sex than men sampled off-line. They speculate that this could be because the on-line sample captured “some high-risk respondents who only use the internet to meet other men.”

Further information on this website

Internet the most commonly used venue to cruise for sex by gay men with syphilis in SF - news story

Internet cruisers more likely to have unprotected sex - news story