© 1990 Oxford University Press
research-article |
Factors Associated with AIDS and AIDS-Like Syndrome among Homosexual and Bisexual Men in Minas Gerais, Brazil


*Grupo de Pesquisas em Epidemiologia, Universidade Federal de Minas Gerais, Av Alfredo Balena 190 30. 130 Belo Horizonte, Minas Gerais, Brazil.
**Ambulatóriode Imunodeficiéncias do Pós Graduaç
o em Medicina Tropical, Universidade Federal de Minas Gerais.
Departamento de Parasitologia Universidade Federal de Minas Gerais
Department of Epidemiology, Johns Hopkins School of Hygiene and Public Health Baltimore, USA.
de Lima e Costa M F F (Grupo de Pesquisas em Epidemiologia, Universidade Federal de Minas Gerais, Av Alfredo Balena 190, 30. 130, Belo Horizonte, Minas Gerais, Brazil), Oliveira M R, Oliveira E I, Paulino U H M, Greco D B, Chiari C A, Guimaraes M D C, Proietti F A, Antunes C M F and Munoz A. Factors associated with AIDS and AIDS-like syndrome among homosexual and bisexual men in Minas Gerais, Brazil, International Journal of Epidemiology 1990, 19: 429434.
A case-control study to determine factors associated with AIDS andAIDS-like syndrome among homosexual/bisexual men was conductedin the State of Minas Gerais (Brazil). Eighty-three per cent (45 patients) of all AIDS/AIDS-like syndrome cases in homosexual/bisexual men reported in Minas Gerais between February, 1986 and June, 1987 were compared to 133 seronegative controls seen at the same clinic. Blood samples were tested by ELISA and confirmed by Western blot. Sex with men from the USA, sex with someone who developed AIDS, number of male partners (
100 lifetime), age (
30 years old) and ethnicity (white) were independently associated with AIDS/AIDS-like syndrome (Odds Ratios = 5.5, 4.3, 3.9, 3.5 and 2.7, respectively). Thirty-nine per cent of cases and 44% of controls reported bisexual activity during the previous two years. From these, a high proportion reported anal intercourse with women in the same period (53% of bisexual cases and 33% of bisexual controls). Bisexual men had more male partners than female partners in the previous two years (median male partners = 20 for cases and five for controls; median female partners = three for both cases and controls). This explains in part why the epidemic has increased more rapidly among men then among women in Minas Gerais, despite the large proportion of bisexuals with the disease.
Revised 1 January 1990