The aim of this dissertation research is to examine the associations of risk factors with HIV infection by comparing the distribution of the risk factors in infected runaway, homeless, and incarcerated adolescents with that in non-infected runaway, homeless, and incarcerated adolescents. Much of the research conducted to date on adolescents has relied on surrogate markers (high risk behaviors) as outcomes instead of HIV status due to the unavailability of HIV status in this population. Due to the recent emphasis on HIV testing on youth in high risk settings, HIV status is now available on many youth and can serve as an excellent outcome variable for looking at risk factors for HIV. Using a case-control research design, all adolescents, aged 12-21, who have been tested for HIV at three sites in San Francisco during 1994 will be compared using chart review. These three sites include Cole Street Clinic, Larkin Street Medical Clinic, and Youth Guidance Center Medical Clinic which are operated by the San Francisco Department of Public Health Special Programs for Youth. The clinics are specifically designed to provide care for runaway, homeless, and incarcerated youth in San Francisco. In 1990, twelve percent (12%) of the youth tested at Larkin Street Medical Clinic had evidence of HIV infection (Shalwitz, 1990). More recently, in seroprevalence surveys conducted in 1993 among 1,165 youth attending clinics at homeless youth centers in San Francisco, HlV seroprevalence was 26% among men who have sex with men. 50% among men who have sex with men and inject drugs (IDUs), 0% among heterosexual male IDUs, 2.8% among heterosexual females, 0.7% among male heterosexuals who are not IDUs, and 0.8% among female heterosexuals who are not IDUs (Givertz & Katz, 1993). The investigator presumes that approximately 6 - 12% of those tested during 1994 will be HlV positive based on the earlier studies. The null hypothesis to be tested is that there are no differences in risk factors between adolescents who are HIV negative and those who are HlV positive. Given the recent emphasis on HlV testing in this population, the investigator projects collecting data on 300-500 youth, of which approximately 30 will be HlV positive. A sample size calculation for multiple controls per case as described by Schlesselman (1982) with an alpha of 0.05, beta of 0.2, relative risk of 3, and estimated exposure rate of 0.4 resulted in a sample size of 27 cases (HlV positive) and 243 controls (HIV negative). Given the projected number of adolescents who will be tested during 1994, this sample size seems realistic.
