In 2006, the CDC released recommendations for universal HIV testing in all healthcare settings, including emergency departments (EDs). However, many patients are not screened due to perceived low HIV infection risk (HIR) and lack of screening programs in EDs, demonstrating missed opportunities in identifying unknown HIV infections.
In 2015, 351 patients who presented to an urban academic ED in Washington, D.C. completed a self-administered survey assessing demographics, healthcare encounters in the prior year, and HIV risk factors. HIR was calculated using the Denver HIV Risk Score (DHRS), a composite score based on age, gender, race/ethnicity, sex of partners, injection drug use, and prior HIV testing. Patients were scored (6-item range: 0-76) and classified as low HIR (<40, n=304) or increased HIR (40, n=47) based on a validated algorithm. Logistic regression was performed to determine whether lack of healthcare engagement was associated with increased DHRS score.
The sample included 351 patients; 288 (82.1%) accepted a free HIV test prior to participation, and 63 (17.9%) refused the test. Median age of patients was 38; 57.4% were female; 40.7% were non-Hispanic (NH) white, 46.7% were NH black, and 6.6% were Hispanic/Latino. Logistic regression analyses demonstrated that patients who did not see a healthcare provider in the prior year (pOR 5.03; 95% CI 2.58-9.83) or were uninsured (pOR 3.95; 95% CI 1.57-9.93) had a higher risk of HIV infection. Conversely, patients who visited a primary care provider in the past year (pOR 0.24; 95% CI 0.13-0.45 a clinic or other health center (pOR 0.38; 95% CI 0.20 – 0.72), or an ED (pOR 0.48; 95% CI 0.25-0.90) were associated with having decreased odds of high HIR.
The DHRS categorization suggests that patients with an increased risk of HIV were less likely to have regular healthcare encounters and were more likely to be uninsured. Patients with increased risk of HIV may be more likely to have missed opportunities for HIV testing due to decreased healthcare engagement, and should be targeted for more prompt, frequent HIV screening beyond the general CDC recommendations. Insurance status may serve as a proxy variable for identifying this population in the ED.
A. Pehar, None
B. Furlong, None
P. Kumar, None