A case control study of HIV seroconversion in gay men, 1988-1993: what are the current risk factors?. Issue 3 (June 1996)

• Record Type: Journal Article
• Authors: Williams, D I; Stephenson, J M; Hart, G J; Copas, A; Johnson, A M; Williams, I G
• Is Part Of: Sexually transmitted infections. Volume 72:Issue 3(1996)
• Page Start: 193
• Page End: 196
• Publication Date: 1996-06
• Subjects: Sexually transmitted diseases -- Periodicals; HIV infections -- Periodicals; 616.951005
• DOI: 10.1136/sti.72.3.193
• Languages: English
• ISSNs: 1368-4973

Abstract

Abstract : OBJECTIVE: To investigate current risk factors for HIV seroconversion among homo/bisexual men. DESIGN: Matched case control study in a large STD clinic in central London. Data on risk factors were obtained by case note review. METHODS: Fifty-six cases who had at least one negative HIV test followed by a positive test between June 1988 and July 1993, and two homo/bisexual controls (having two or more negative HIV tests) matched to each case on age, total number of HIV tests and test interval period were identified. Univariate and multivariate odds ratio were calculated for acute STD, ano-genital intercourse, condom use and HIV status of sexual partners. RESULTS: Adjusted odds ratios (95% confidence intervals) for HIV seroconversion were 4.1 (1.3-13.3) for having an acute interval STD and 4.6 (1.4-15.4) for having a known HIV infected sexual partner. Compared with men who always used condoms, odds ratios for men who sometimes or never used condoms were 7.9 (2.2-28.9) and 16.2 (3.0-86.0) respectively. Unprotected ano-genital intercourse was commonly reported by both cases and controls, and reported condom use was no greater with a known HIV infected partner than with a partner of unknown HIV status. CONCLUSION: HIV seroconversion among homo/bisexual men attending STD clinics is strongly related to having an acute STD, a known HIV infected partner and not using condoms. Although consistent use of condoms is highly protective, knowing that a partner is HIV infected does not ensure condom use between serodiscordant men. More effective, well-evaluated interventions are needed to reduce sexual risk-taking in this population.