health center in Boston, Massachusetts.9"4 This population is of interest since ... represent the association with HIV antibody ofa doubling of ..... The American Association of Public Health Dentistry (AAPHD) has issued a call for abstracts on.
Behavioral Risk Factors for HIV Infection among Homosexual Men at a Boston Community Health Center JANE MCCUSKER, MD, DRPH, ANNE M. STODDARD, SCD, KENNETH H. MAYER, MD, DAVID N. COWAN, MPH, AND JEROME E. GROOPMAN, MD Abstract: Social and behavioral factors associated with human immunodeficiency virus (HIV) infection were analyzed using crosssectional data from homosexual and bisexual male clients of a Boston community health center. Partners from California, and a previous period of greater sexual activity (a "high period"), were independently associated with positive HIV antibody status, as were the frequency of receptive anogenital contact, both during the "high period" and during the last six months. (Am J Public Health 1988; 78:68-71.)
Introduction The epidemiology of the acquired immunodeficiency syndrome (AIDS) and human immunodeficiency virus (HIV) infection has received intensive study since the first recognition of this syndrome in 1981.1 Homosexually active men have constituted the largest risk group in North America and in Western European countries,2 although not in Africa.3 Most of the epidemiologic studies of AIDS in homosexual men have come from New York and San Francisco, where the epidemic was first recognized and where the virus appears to have been transmitted rapidly in the high-risk population.4 These studies have identified receptive anal intercourse as the primary method of transmission.-'3 In this study we report results on behavioral risk factors from a cohort seroepidemiologic study of HIV transmission in homosexual and bisexual male clients of a community health center in Boston, Massachusetts.9" 4 This population is of interest since the epidemic appears to have started later in Boston than in New York and California, and the seroprevalence among homosexually active males appears to be stabilizing at a lower level.
sexual activity than during the recent period. The test-retest reliability of these measures has been assessed and appears adequate.'5 Each subject was examined clinically and blood was drawn for a complete blood count and for HIV antibody determination. Antibody to HIV was determined by using the enzymelinked immunosorbent assay (ELISA) and fluorescent antibody techniques.'6 Sera reactive three-fold or greater over known negative controls were considered positive. Specimens found to be positive by ELISA were confirmed by a Western blot assay,'7 and reported as positive when both tests gave positive results.'8"9 Results
Sociodemographic characteristics of the cohort and HIV antibody status are shown in Table 1. Odds ratios and their
95 per cent confidence intervals (Cl)20 are given for the associations between initial HIV antibody status and selected variables in Table 2. These variables are obviously intercorrelated, therefore multiple logistic regression analysis was used to model the multivariate association with antibody status. Number of partners and frequencies of sexual practices and recreational drug use were transformed logarithmically (base 2) to reduce skewness. Zero was coded as 1/8 before logs were taken. The odds ratios for these variables represent the association with HIV antibody of a doubling of the number of partners or frequency. Intravenous drug use was coded as a dichotomous variable; a known "high TABLE 1-Soclodemographic Characteristics, Sexual History, and HIV Antibody Status of the Cohort (n = 290)
Characterstics
Methods The study population of 290 men was recruited over the period from January 1985 through April 1986 from men who had visited the health center during the previous six months. Criteria for participation included: recent homosexual activity, and no diagnosis or symptoms suggestive of AIDS, AIDS-related symptom complexes, or other serious diseases. Sociodemographic and behavioral data were gathered by self-administered questionnaire. Measures of specific sexual practices and recreational drug use included frequency during the previous six-month period as well as during any previous period during which the subject reported greater From the Division of Public Health, University of Massachusetts at Amherst (Drs. McCusker and Stoddard, and Mr. Cowan); Department of Medicine, Memorial Hospital, and Brown University, Providence, RI (Dr. Mayer); Department of Medicine, New England Deaconess Hospital, and Harvard University (Dr. Groopman); and Fenway Community Health Center, Boston (study location). Address reprint requests to Dr. Jane McCusker, Division of Public Health, School of Health Sciences, University of Massachusetts, Amherst, MA 01003. This paper, submitted to the Journal February 9, 1987, was revised and accepted for publication July 29, 1987. X 1988 American Journal of Public Health 0090-0036/88$1.50
68
n
%
35 plus
48 91 76 75
17 31 26 26
White
280 3
97 2 1
105 116 69
36 40 24
253 37
87 13
233 57
80 20
61 141 63 17
22 50 22 6
216
74 26
Age (years) 20-24
25-29
30-34
Ethnicity
Black Other Education Some college, or less
7
College graduate Graduate school Marital Status
Never married Ever married Sexual Orientation Exdusively homosexual Not exclusively homosexual Age when regular homosexual behavior began Less than 18
18-22 23-29 30 plus HIV antibody status
Negative Positive
74
AJPH January 1988, Vol. 78, No. 1
PUBLIC HEALTH BRIEFS TABLE 2-Unhvarlable Odds Ratlos and 95% CenfIdenb rntevals for Selected Behaviors and HIV Antibody Status
Variable
HIV Antibody Status N % Seropositive
Total 290 25.5 Lifetime male sexual partners 1-9 14 7.1 10-99 120 16.7 100+ 156 34.0 Years of regular homosexual activity 1-5 83 15.7 6-10 91 27.5 11+ 107 32.7 Friend or lover with AIDS/ARC/positive HIV No 194 19.1 Yes 96 38.5 Partners from California in previous five years 0-1 194 18.6 2+ 96 39.6 Partners from New York City in previous five years (n = 289) 0-1 162 20.4 2+ 127 32.3 Partners from Miami in previous five years (n = 289) 0-1 237 20.7 2+ 52 48.1 Male sexual partners in previous six months 0-1 45 17.8 2-4 86 23.3 5+ 159 28.9 Frequency of receptive orogenital contact, previous six months Never 17 5.9
