Audio computer assisted interviewing to measure HIV risk behaviours in a clinic population.

OBJECTIVES To examine whether audio computer assisted survey interviewing (ACASI) influenced responses to sensitive HIV risk behaviour questions, relative to interviewer administration of those questions (IAQ), among patients attending a sexually transmitted infection (STI) clinic and whether the impact of interview mode on reporting of risk behaviours was homogeneous across subgroups of patients (defined by age, sex, and previous STI clinic experience). METHODS 1350 clinic patients were assigned to complete a detailed behavioural survey on sexual risk practices, previous STIs and symptoms, condom use, and drug and alcohol use using either ACASI or IAQ. RESULTS Respondents assigned to ACASI were more likely to report recent risk behaviours such as sex without a condom in the past 24 hours (adjusted OR = 1.9), anal sex (adjusted OR = 2.0), and one or more new partners in the past 6 months (adjusted OR = 1.5) compared to those interviewed by IAQ. The impact of ACASI varied by sex but, contrary to expectations, not by whether the patient had previously visited an STI clinic. Mode of survey administration made little difference within this population in reports of STI knowledge, previous STIs, STI symptoms, or illicit drug use. CONCLUSION ACASI provides a useful tool for improving the quality of behavioural data in clinical environments.