January 1996

Trends and Predictors of Human Immunodeficiency Virus Antibody Testing by Homosexual and Bisexual Adolescent Males, 1989-1994

Author Affiliations

From the Mayo Medical School, Rochester, Minn (Ms Povinelli); and the University of Minnesota Hospital and Clinics (Dr Remafedi) and School of Public Health (Mr Tao), University of Minnesota, Minneapolis.

Arch Pediatr Adolesc Med. 1996;150(1):33-38. doi:10.1001/archpedi.1996.02170260037005

Objective:  To identify temporal trends and predictors of human immunodeficiency virus (HIV) antibody testing in homosexual and bisexual youth, using the Health Belief Model as a conceptual framework.

Design:  Cross-sectional survey.

Subjects:  Five hundred one male volunteers, 13 to 21 years old, self-identified as homosexual, bisexual, or as having sex with men, were enrolled from June 1, 1989, to May 30, 1994.

Methods:  Structured interviews and written instruments, including measures of perceived susceptibility to and severity of HIV disease, benefits and barriers to testing, and cues to action. Based on significant (P<.001) bivariate associations, variables were selected for forward stepwise logistic regression analysis.

Outcome Measure:  Self-reported HIV antibody testing.

Results:  Forty-five percent of the subjects had undergone HIV antibody testing, with no significant differences between annual cohorts. Predictors of testing were having discussed same-sex feelings or experiences with a physician or counselor, a history of unprotected receptive anal intercourse, substance abuse, younger age at self-identification as bisexual or homosexual, ever having had a steady male partner, having many friends who understand sexual orientation, living away from family, and older age.

Conclusions:  Testing practices did not change significantly across time. Human immunodeficiency virus testing was related to age, risky behaviors, living situation, bisexual or homosexual acculturation, and contacts with health professionals, corresponding to Health Belief Model dimensions of perceived susceptibility, barriers, and cues to action.(Arch Pediatr Adolesc Med. 1996;150:33-38)