Routine, opt-out HIV tests in ED detect modestly more cases
AMA survey shows impact of consult code elimination
Screening all patients in the emergency department on a routine, opt-out basis modestly detected more patients with HIV, most of whom were late in the course of disease, reports a study that appears in the July 21 Journal of the American Medical Association.
Researchers alternated routine opt-out rapid HIV screening and physician-directed diagnostic rapid HIV testing in four-month spans at Denver Health Medical Center, an urban public safety-net hospital with an approximate annual emergency department census of 55,000 patient visits between April 2007 and April 2009.
The opt-out phase included 28,043 patients, of whom 6,933 patients (25%) completed HIV testing. (Doctors could perform diagnostic testing in this phase so they could still provide standard emergency medical care.) Ten of 6,702 patients (0.15%; 95% CI, 0.07% to 0.27%) had new HIV diagnoses, as did five of 231 patients (2.2%; 95% CI, 0.7% to 5.0%) diagnostically tested in this phase. The diagnostic phase included 29,925 eligible patients, of whom 243 (0.8%) completed HIV testing. Four patients (1.6%; 95% CI, 0.5% to 4.2%) had new diagnoses.
The prevalence of new HIV diagnoses (including those diagnostically tested) was 15 in 28,043 (0.05%; 95% CI, 0.03% to 0.09%) in the opt-out phase and four in 29,925 (0.01%; 95% CI, 0.004% to 0.03%) in the diagnostic phase. Non-targeted opt-out HIV screening was independently associated with new HIV diagnoses (risk ratio, 3.6; 95% CI, 1.2 to 10.8) when adjusting for patient demographics, insurance status and whether diagnostic testing was performed in the opt-out phase.
The median CD4 cell count for those with new HIV diagnoses in the opt-out phase (including those diagnostically tested) and in the diagnostic phase was 69/µL (interquartile range [IQR], 17 to 430) and 13/µL (IQR, 11 to 15), respectively (P=0.02).
Emergency department crowding was not affected by HIV screening. Waiting, length of stay and boarding times differed slightly between the study phases, but were not clinically meaningful. A validated composite measure of emergency department crowding showed no differences between routine and physician-directed screening.