UCSF

Objective

To describe the prevalence and location of new and acute HIV diagnoses in a large urban medical center. Secondary objectives were to evaluate rapid HIV test performance, the added yield of acute HIV screening, and linkage-to-care outcomes.

Design

Cross-sectional study from November 1, 2008, to April 30, 2009.

Methods

The hospital laboratory performed round-the-clock rapid HIV antibody testing on venipuncture specimens from patients undergoing HIV testing in hospital and community clinics, inpatient settings, and the emergency department (ED). For patients with negative results, a public health laboratory conducted pooled HIV RNA testing for acute HIV infection. The laboratories communicated positive results from the hospital campus to a linkage team. Linkage was defined as 1 outpatient HIV-related visit.

Results

Among 7927 patients, 8550 rapid tests resulted in 137 cases of HIV infection [1.7%, 95% confidence interval (CI): 1.5% to 2.0%], of whom 46 were new HIV diagnoses (0.58%, 95% CI: 0.43% to 0.77%). Pooled HIV RNA testing of 6704 specimens (78.4%) resulted in 3 cases of acute HIV infection (0.05%, 95% CI: 0.01% to 0.14%) and increased HIV case detection by 3.5%. Half of new HIV diagnoses and two thirds of acute infections were detected in the ED and urgent care clinic. Rapid test sensitivity was 98.9% (95% CI: 93.8% to 99.8%) and the specificity 99.9% (95% CI: 99.7% to 99.9%). More than 95% of newly diagnosed and out-of-care HIV-infected patients were linked to care.

Conclusions

Patients undergoing HIV testing in EDs and urgent care clinics may benefit from being simultaneously screened for acute HIV infection.