The Talking Sex Project is an HIV risk reduction education project for gay and bisexual men. It was developed to evaluate two types of small discussion groups using a randomized control study design. This paper reports on the baseline measures of knowledge, attitudes and sexual practices. The 612 subjects had a mean age of 32 years, were all English speaking and were relatively highly educated. Bivariate analyses found that younger men, those in a relationship, those reporting bisexual activity and those with higher levels of substance use were significantly more likely to have unprotected anal intercourse. While overall knowledge was not associated with sexual behaviour, a significant relationship was found with knowledge of risk of anal sex. Attitudes, in particular, the belief in the efficacy of condoms and the enjoyment of safer sex were found to have the strongest associations with protected anal intercourse. The paper provides an understanding of some of the factors which may be important to behaviour change and clues for the targeting of programs.
PIP: Baseline data on a cohort of 612 homosexual and bisexual men were analyzed in order to examine the interrelationships between sociodemographic, knowledge, and attitude factors and sex behavior. Men were recruited for a randomized control study of small group AIDS risk reduction education in Toronto, Canada: The Talking Sex Project. Questionnaire validity and reliability checks were conducted. The development of knowledge and attitude scores is provided and includes reliability checks. Chi square and analysis of variance (ANOVA), Turkey's Studentized range Test, Pearson product moment correlations, and stepwise logistic regression analyses were performed. Population characteristics included a mean age of 32 years; 1.8% were teenagers, 2.8 had AIDS, and 6.2% had HIV-related diseases. Sociodemographic, health and life style, and sexual behavior characteristics were also reported. The univariate description of scaled knowledge and attitude variables is given. Knowledge is fairly high including knowledge or risk. Both scales had skewing toward the higher ends of the scale. 16.3% reported no sexual activity, but 52.6% reported 1 or more sexual activities in the last 3 months, excluding kissing or masturbation, which was the most common activity. 88.4% reported insertive oral-genital intercourse. 61.5% reported receptive of insertive anal sex. 42.2% had oral-anal sex. 4.1% participated in hand-anal sexual practices. Those with high school or less education had significantly lower scores on knowledge. Higher scores on knowledge of risk were found among men who had safer or protected sex than those who had unprotected anal intercourse. The relationship of background variables to sexual behavior showed that bisexuality was found to be associated with men under 24 years of age. 50% of bisexual men under 24 years reported unprotected anal intercourse vs. 24.4% of homosexual men. Alcohol or drug use was associated with sexual behavior, and men with reported safer sex consumed slightly less. Men practicing safer sex were more likely to report impulse control difficulties, had a stronger belief in the efficacy of condoms, and rated sexual practices safer. The logistic model correctly classified 94.5% of safer sex cases and 54.9% of unprotected sex. Efficacy of condoms was the best predictor of safer sex, as well as lower drug use, not being in a relationship, greater enjoyment of safer sex, being older, having fewer sexual partners, and lower, more cautious self-rating of knowledge of risk. Bisexuality needs further exploration.