August 2007: Volume 33, Number 8
Routine HIV Testing in the ED Catches Unidentified Cases
Routine, rapid HIV testing in hospital emergency departments (EDs) can help identify patients with the virus who otherwise might not be tested and link them to appropriate services, suggested preliminary research published in the June 22 issue of Morbidity and Mortality Weekly Report.
Researchers participating in a Centers for Disease Control and Prevention (CDC) initiative that aims to reduce barriers to early diagnosis of HIV and increase access to medical care and prevention services implemented rapid testing programs at emergency departments in Oakland, Calif.; Los Angeles, Calif.; and New York, N.Y. The teams tested oral mucosal transudate specimens or finger-stick whole-blood specimens using OraQuick Advance Rapid HIV-1/2 Antibody Tests (Orasure Technologies, Bethlehem, Pa.). Between January 2005 and March 2006, a total of 34,627 patients were offered rapid HIV tests. Nearly 60% agreed to be tested and 9,365 patients of those who agreed to receive rapid HIV tests actually got them. Patients with positive rapid results received risk-reduction counseling and were asked for a whole-blood or oral specimen for confirmatory testing by Western blot. Of those, 97 patients (1%) had a preliminary positive result for HIV infection and a majority of those newly identified as HIV positive (88%) were successfully linked to care.
The researchers also collected data on risk factors for 98% of patients with newly diagnosed HIV infection. Among them, 52% reported having at least one risk factor for HIV transmission during the previous 12 months. “If a risk-based approach to testing had been used in these three ED demonstration projects, 48% of the persons with newly diagnosed HIV infection would not have been offered testing,” noted an editorial comment accompanying the data. A revised CDC recommendation issued in September 2006 calls for routine, voluntary HIV testing policy that allows patients to opt out. While the CDC program employed an opt-in approach, results of the program “provide insight into methods that could be used to implement testing using an opt-out approach,” the editorial notes.