HIV Testing in the Emergency Department

Unfortunately, EDs have been slow to embrace routine HIV screening. In 2004, the Health Research & Educational Trust conducted a national survey of HIV testing practices in hospitals and found that nearly half of all hospitals surveyed had policies prohibiting routine ED-based screening.⁴ In 2006, only 13% of EDs at academic medical centers had policies recommending routine testing.⁵

In that same year, the CDC recommended routine “opt-out” HIV testing in all healthcare settings.⁶ Opt-out screening means performing an HIV test after notifying the patient that (a) the test will be performed and (b) the patient may elect to decline or defer testing. In contrast, traditional “opt-in” screening approaches involved asking for permission to test the patient, usually after giving written informed consent. The number of EDs now providing HIV testing has increased significantly over the last few years, perhaps in response to the 2006 CDC recommendation.

Impediments to routine HIV testing include the increased work burden imposed on already overcrowded facilities, the limited ability of EDs to provide linkages to follow-up care, the concern for privacy and confidentiality, and cost.⁷ HIV testing is not routinely reimbursed making funding a major issue in the implementation of routine testing. Many existing programs are supported in part by transient external funding, including resources from the CDC and the pharmaceutical industry.8 In a recent editorial, Kelen and Rothman⁸ questioned the practicality of implementing an HIV screening program: “Most EDs in which routine screening is indicated and would likely be most effective are the very ones straining to ensure an appropriate patient safety environment and can least afford to add a new initiative.”

Yet, it seems imperative that health insurers, including our national public healthcare system, help as part of the overall strategy of HIV prevention and early detection. Not only do an estimated 21% of HIV-positive persons not know they are infected,⁹ but diagnosis often occurs late in the course of infection: up to 40% to 50% of patients have CDC-defined AIDS at the time of their HIV diagnosis.^10,11 Treatment results in dramatic reductions in morbidity and mortality, and awareness of infection status has been shown to lead to behavior changes,¹² reducing further transmission of the virus.

In the meantime, how can a successful, cost-effective, and sustainable HIV testing program be developed in the ED? The CDC, in partnership with Francois-Xavier Bagnoud Center (at the University of Medicine & Dentistry of New Jersey), has offered strategic workshops in different states on this subject. An annual ED HIV Testing Consortium meeting is ongoing, and various groups from ED programs throughout the United States are starting to communicate with one another. On the Internet, one can find the helpful HIV Testing in the Emergency Department: A Practical Guide (www.edhivtestguide.org).

In addition to identifying funding, another key question is determining which type of screening program works best given the prevalence in each local community. Existing laws can preclude implementation of opt-out testing; a recent review of state statutes identified 16 states that were still inconsistent with the CDC recommendations.¹³ Not surprisingly, there is considerable variability among HIV testing programs in the United States. Some offer universal testing using a nontargeted strategy, while others still target “at-risk” populations or test on the basis of specific signs or symptoms.¹⁴

One example of a successful, large-scale HIV testing program is at the Alameda County Medical Center ED in northern California. Since 2007, persons have been notified at registration that opt-out HIV screening is being conducted. From August 2007 through March 2008, there were 49,187 ED visits by 30,418 unique persons. Based on their eligibility criteria (age 13 years or older, normal mental status, able to opt-out, and not known to be HIV-infected), 23,159 (76.1%) were eligible for opt-out screening; of these, 8,472 (36.3%) accepted testing, although only 5,275 were actually tested due to staffing issues. Despite a relatively low prevalence among those tested (0.5%), 25 previously undiagnosed persons were identified.¹⁵ The authors did question whether universal HIV screening using existing ED staff was possible, given how few tests were completed despite the widespread acceptance of testing.¹⁶

Given the realities of constrained resources, which population is most appropriate for HIV screening in the ED? Investigators from Temple University recently reported results from an interesting study in which they attempted to identify the true seroprevalence rate in the ED (not just among those offered testing).¹⁷ During a 2-week period, there were 2,427 admissions to their ED, of which 2,019 charts were reviewed. Blood samples from 942 patients were available for HIV testing. After being de-identified, 133 (14%) blood specimens were found to be reactive by a rapid testing strategy. Only 28 patients had reported a history of HIV, suggesting that most of the test results represented new diagnoses. This rate was markedly higher than their previously reported seroprevalence of 1% when rapid testing was offered. Could patients who refuse testing be the most important group to test? Are strategies to increase rates of test acceptance needed?

A critical aspect of a successful ED program is securing linkages to HIV care following diagnosis. Placing social workers in the ED and HIV clinics can be helpful. Some programs even have someone who serves as the patient’s “bridge to care” to maximize rates of follow-up. Samples to be used for confirmatory tests (eg, enzyme immunoassay and Western blots) should be obtained before the patient leaves the ED, since false-positive screening results can occur and patients may be lost to follow-up after their ED visit.

In a demonstration project at George Washington University Medical Center, in which patients were referred to the hospital or community resources for confirmatory testing, 13 of the 26 patients who tested positive for HIV were lost to follow-up.¹⁸ Collecting accurate contact information is crucial for retaining patients in care. These issues will require financial resources that are often not available.

At Duke University Medical Center, we recently implemented a nontargeted rapid testing program in the ED using rapid test kits (OraQuick, OraSure Technologies, Inc, Bethlehem, Penn.) provided through a grant from the CDC. With institutional and industry support, testing was initiated in January 2009, and about 600 patients have been approached for testing, with a 63% acceptance rate (McKellar, unpublished data, 2009). Research hypotheses have been incorporated into the program, which partners HIV providers with their ED counterparts.

We hope that the results of this program will help ED providers become more supportive of HIV testing as a standard of care. As with many large-scale preventative health care initiatives, HIV screening in the ED requires a significant up-front investment of resources. Only when such a commitment is made will the larger issue of underdiagnosed HIV infection begin to be resolved.