We estimated the rate of HIV infection in the United States in different age groups according to gender and race/ethnicity by working backwards from counts of AIDS cases. This calculation required knowledge of the incubation period between infection with HIV and full-blown AIDS. We estimated that between 630,000 and 900,000 adults and adolescents were living with HIV infection in January, 1993. The estimated incidence of infection was high among young adults, minorities, and women, and the number of Americans living with HIV was very high among minorities. We extended these results by combining information from multiple independent sources. We found that the prevalence of infection had increased from 1984 to 1992, with greater relative increases among women than men, largely because the number of persons infected heterosexually had increased faster than in any other group. Significant progress was made towards understanding the natural history of HIV infection. We demonstrated that the HIV viral level measured within 3 years of infection is a strong, age-independent predictor of long-term clinical outcome. We found no major secular change in the rate of disease progression among homosexual men infected between 1984 and 1991 based on the rate of CD4 decline. We linked AIDS and cancer registries to study cancer incidence in persons with AIDS. We found that the risk of Kaposi's Sarcoma (KS) following another AIDS-defining illness was strikingly high, especially in white men, and that the risks of both KS and particularly, non-Hodgkin's lymphoma appeared to increase with the time since AIDS onset.
