The SafeGuards Men's Survey: Report of a Survey of Philadelphia Gay and Bisexual Men's Knowledge, Attitudes, Beliefs, and Behaviors Related to HIV Infection
- Cite this article as:
- Whyte, J., Bartlett, C., Polanksy, M. et al. Journal of the Gay and Lesbian Medical Association (1998) 2: 113. doi:10.1023/B:JOLA.0000004053.12411.e5
Objective: To characterize HIV-related knowledge, attitudes, beliefs, and behaviors in a sample of gay and bisexual men in the greater Philadelphia area, to allow rational development of HIV prevention strategies. Design: Self-administered survey. Setting: Gay social groups, bars, service organizations, sex establishments, and street outreach in the greater Philadelphia area. Subjects: Gay and bisexual men who accepted $2–$4 for completing surveys in the above venues. Measurements: The survey booklet, available in English and Spanish, was completed by each subject in a multiple-choice and yes/no format. Questions were based on prior HIV prevention surveys, with new items added to sample dimensions identified from the literature as of interest. Data were analyzed using descriptive statistics, cross-tabulations, correspondence analysis, and logistic regression. Results: Overall knowledge of HIV transmission was high in the sample, but specific details were lacking for many subjects and overall knowledge was poorest in the least educated and poorest segment of the population. Attitudes including internal locus of control, level of fear, and peer support were important determinants of specific sexual behaviors. Sexual behaviors occurred in “clusters” or profiles that were partly determined by risk level and partly by sexual role preferences. Unprotected anal sex was more common among primary partners than nonprimary partners, regardless of the duration of the primary relationship. Conclusions: These results confirm that socio-economic status is an important determinant of HIV transmission risk, but that this is at least partly mediated by important differences in knowledge, attitudes, and beliefs. High-risk sexual behaviors do not occur in isolation, but form profiles or clusters that are related to various demographic and attitudinal factors. These findings can contribute to rationally designed prevention strategies.