AEGiS-15IAC: Discordant HIV test results among a community sample of gay and bisexual men: Results from the Ontario men's survey.

15th International AIDS Conference

Bangkok, Thailand - July 11-16, 2004

Discordant HIV test results among a community sample of gay and bisexual men: Results from the Ontario men's survey.

Int Conf AIDS 2004 Jul 11-16; 15:(abstract no. ThOrC1411)

Myers T, Allman D, Remis R, Calzavara L, Ennis M, Maxwell J, Swantee C, Travers R
HIV Social, Behavioural and Epidemiological Studies Unit, Dept. of Public Health Sciences, University of Toronto, Toronto, Canada

BACKGROUND: To explore predictors of discrepancies between self-reported HIV status and saliva test results focusing specifically on men who indicated that they tested HIV negative at their last HIV test.

METHODS: An anonymous, self-completed cross-sectional survey with optional collection of a saliva sample for HIV testing was conducted in 13 communities. Among men who reported their HIV status to be negative, a series of logistic regressions to predict discrepant status were used to examine the influence of groups of variables including, demographic, sexual risk and STIs, and drug use.

RESULTS: Of 5,080 men who participated 76.1% (n=3,865) self-reported results of at least one HIV test; 71.6% (n= 3,635) provided saliva. 2,406 self-reported their status to be HIV-negative and provided sufficient saliva to compare. Among these, 2.1% (n=51) tested HIV-positive. By multivariate logistic regression the independent predictors of discordant HIV status were: Toronto men vs others (OR=1.7, CI 0.9-3.1), full-employed vs other (OR=2.0, CI=0.9-4.4); bisexual vs gay (OR=2.0, CI 1.0-4.3), and non-Caucasian (OR=3.2, CI 1.6-6.5). When adjusting for demographics and selected other variables, the additional predictors included: men with >10 sexual partners (OR=3.4, CI 1.3-9.4); men who did not often attend gay dances or events (OR=3.3, CI 1.710); unprotected anal sex with casual partners (OR=3.5, CI 1.7-7.0); ever injected drugs (OR=3.0, CI 1.27.8); popper (OR=2.5, CI 1.34.7) and Viagra use in the last year (OR=2.1, CI 1.04.5).

CONCLUSIONS: A substantial number of gay and bisexual men who indicate they are HIV negative based on their last test, either may not know they are currently HIV positive, or inaccurately report. The extent to which this may reflect seroconversions since last test, self-perceived low risk, denial, barriers and abilities to access testing, unwillingness to report an HIV positive status, and self-report error warrants further exploration.

Keywords: AEGIS, Bisexuality, Homosexuality, Male, Homosexuality, Unsafe Sex, Acquired Immunodeficiency Syndrome, HIV Seropositivity, Risk-Taking, Sexual Behavior, Sexual Partners, Data Collection, Cross-Sectional Studies, Prevalence, Health Knowledge, Attitudes, Practice, Substance-Related Disorders, Questionnaires, Ontario, Humans, Male

040711
ThOrC1411