A case-control study of human immunodeficiency virus type 1 seroconversion and risk-related behaviors in the Chicago MACS/CCS Cohort, 1984-1992. Multicenter AIDS Cohort Study. Coping and Change Study.

Abstract

This paper focuses on 76 human immunodeficiency virus type 1 (HIV-1) seroconverters who concurrently participated in the Chicago, Illinois, component of the Multicenter AIDS Cohort Study (MACS) and the Coping and Change Study (CCS) of homosexual/bisexual men between 1984 and 1992. A nested case-control analysis was performed to assess the critical behavioral risk factors associated with incident HIV-1 infection and the consistency of these relations in early (1984-1988) versus later (1989-1992) phases of the study. Univariate results revealed strong early period associations between seroconversion and various measures of receptive anal intercourse (RAI) that became considerably weaker in the study's later period. The weaker associations reflected the overall decline in levels of RAI among the cohort during the 9 years of observation. In contrast, univariate results revealed stronger later period associations between seroconversion and measures of receptive oral intercourse and insertive anal intercourse. Subsequent multivariate testing did not support the hypothesis that receptive oral intercourse and/or insertive anal intercourse have replaced unprotected RAI as important risk behaviours in the homosexual transmission of HIV-1. In conditional logistic regression models combining intercourse measures with indices of drug and condom use, only the latter variables were consistently associated with HIV-1 seroconversion in both early and later study periods. Adjusted odds ratios (ORs) for nonuse of condoms during RAI were consistently significant throughout the study (ORs = 3.7-4.8), while adjusted odds ratios for recreational drug use variables rose dramatically during the latter half of the study (e.g., for use of cocaine, OR = 81.3 (95% confidence interval 8-824) [corrected], and for use of nitrite "poppers," OR = 9.1 (95% confidence interval 1.8-45.5)). The behavioral intervention applications of these findings, as well as their relation to data from other recent cohort studies of HIV-1 seroconversion among homosexual/bisexual men, are discussed.