RESUMO
OBJECTIVE: To determine the extent to which provincial recommendations, reported regional prevalence rates and perceived local prevalence rates of HIV in pregnancy influence a physician's decision to routinely offer prenatal screening for HIV. DESIGN AND METHODS: A random sample of 5,052 family physicians and obstetricians were surveyed by mail. Logistic regression was used to explore the relationships among the variables of interest. RESULTS: The response rate was 61%. Of these, 69.2% provided prenatal care and were included in the analysis. Physicians were more likely to routinely offer HIV testing if they practiced in provinces with recommendations that supported the universal offer of a test (O.R. = 5.80), independent of living in a region with an estimated prevalence rate exceeding 5/10,000 (O.R. = 1.76), or the perception that the infection rate in their practice justified universal counselling of pregnant women (O.R. = 10.41). CONCLUSIONS: Provincial recommendations supporting universal HIV testing in pregnancy are reflected in physician practice.