Web-based behavioral surveillance among men who have sex with men: a comparison of online and offline samples in London, UK.

OBJECTIVE: To compare the characteristics of men who have sex with men (MSM) surveyed online (through gay Internet chat rooms and profiles) and offline (in community venues) in London, UK.

METHODS: In February and March 2002, 879 MSM completed a self-administered pen-and-paper questionnaire distributed in central London gyms (offline sample). In May and June 2002, 1218 London MSM completed a self-administered questionnaire online, accessed through Internet chat rooms and profiles on gaydar and gay.com.

RESULTS: Compared with men surveyed offline, those surveyed online were significantly less likely to only have sex with men (89 vs. 94%), to be in a relationship with a man (44 vs. 52%), or to have been tested for HIV (68 vs. 80%) (P < 0.001). Men recruited online were also younger (mean age, 34 vs. 36 years) and less likely to have had a higher education (67 vs. 79%) (P < 0.001). However, differences between online and offline samples were less pronounced for HIV-positive men and more pronounced for HIV-negative men and those who had never been tested for HIV. Regardless of HIV status, men recruited online were more likely to report high-risk sexual behavior (i.e., unprotected anal intercourse with a partner of unknown or discordant HIV status) than men surveyed offline (32 vs. 22%, P < 0.001). Men recruited online were also significantly more likely to have used the Internet to look for sex (85 vs. 45%, P < 0.001); for HIV-positive and negative men, seeking sex on the Internet was associated with high-risk sexual behavior (P < 0.01). In multivariate analysis, after controlling for confounding factors, being surveyed online was independently associated with high-risk sexual behavior for HIV-negative and never-tested men (HIV-negative men, adjusted odds ratio for online vs. offline samples, 1.73; 95% CI, 1.23, 2.42; P < 0.01; never-tested men adjusted odds ratio 2.45; 95% CI, 1.40, 4.29; P < 0.01). This was not the case for HIV-positive men (adjusted odds ratio for online vs. offline samples, 1.32; 95% CI, 0.69, 2.50; P = 0.4).

CONCLUSION: The Internet offers valuable opportunities for conducting behavioral surveillance among MSM because it reaches some men who may not be easily accessed in the community yet who are at high risk for HIV and sexually transmitted diseases. Comparisons of the social, demographic, and behavioral characteristics of online and offline samples must, however, take into account the confounding effects of HIV status and seeking sex on the Internet.