Abstract
OBJECTIVES: Screening for human immunodeficiency virus (HIV) infection in the emergency department (ED) has been proposed as an effective approach to increase early HIV diagnosis. To evaluate the potential for the implementation of routine screening, we determined the prevalence of unknown HIV infection among patients being seen in an urban public hospital ED. METHODS: We conducted a cross-sectional study among patients presenting to the San Francisco General Hospital's ED during March 2007. We reviewed patients' medical records to determine HIV infection status. In patients with unknown or negative HIV-infection status, we tested de-identified remnant blood specimens by HIV-antibody and nucleic-acid assays. We used a sensitive/less sensitive testing algorithm to determine the duration of HIV infection. RESULTS: During the study period, 1,820 patients had blood collected for clinical evaluation. Of those patients, 146 (8.0%) were known to be HIV-infected. Among the remaining 1,674 patients with unknown HIV-infection status, HIV-infection prevalence was 0.9% (15 of 1,674, 95% confidence interval [CI] 0.55, 1.47). In addition, one case of acute HIV infection (HIV-antibody negative, HIV RNA detected) was identified. Patients with unknown HIV infection vs. those who were uninfected were more likely to be homeless (odds ratio [OR] = 3.89, 95% CI 1.32, 11.45, p<0.05) and 18 to 30 years of age (OR=3.15, 95% CI 1.03, 9.61, p<0.05). CONCLUSIONS: In a sample of patients visiting a county ED, the relative prevalence of unknown HIV infection (10%) was modest and less than national estimates (25%). Acutely HIV-infected patients might account for a significant proportion of those with unknown HIV infection in an ED setting.