Objective: To develop and test a brief, reliable, and valid HIV-risk screening instrument for use in primary health care settings.
Design: A two-phase study: (1) developing a self-administered HIV-risk screening instrument, and (2) testing it with a primary care population, including testing the effect of confidentiality on disclosure of HIV-risk behaviors.
Setting: Phase 1: 3 types of sites (a blood donor center, a methadone clinic, and 2 STD clinics) representing low and high HIV-seroprevalence rates. Phase 2: 4 primary care sites.
Participants: Phase 1: 293 consecutively recruited participants. Phase 2: 459 randomly recruited primary care patients.
Main outcome measure: Phase 1: comparison of the responses of participants from low and high HIV-seroprevalence sites. Phase 2: primary care patients' rates of disclosure of HIV-risk behaviors and ratings of acceptability.
Results: Phase 1: through examining item-confirmation rates, item-total correlations, and comparison of responses from low and high HIV-seroprevalence sites, we developed a final 10-item HIV-risk Screening Instrument (HSI) with an internal consistency coefficient of .73. Phase 2: 76% of primary care patients disclosed at least 1 risky behavior and 52% disclosed 2 or more risky behaviors. Patients were willing to disclose HIV-risk behaviors even knowing that their physician would see this information. Ninety-five percent of our patient participants were comfortable with the questions on the HSI, 78% felt it was important that their doctor know their answers, and 52% wished to discuss their answers with their physician.
Conclusion: Our brief, self-administered HSI is a reliable and valid measure. The HSI can be used in health care settings to identify individuals at risk for HIV and to initiate HIV testing, early care, and risk-reduction counseling, necessary goals for effective HIV prevention efforts.