HIV screening in the emergency department undoubtedly can identify many HIV-positive people who might otherwise remained unaware of their status. A three-city demonstration project on routine ED HIV testing, conducted by the Centers for Disease Control and Prevention in 2005–2006, found that 48 percent of those with newly diagnosed HIV infection would not have been offered testing under previous models.
The CDC recommended several years ago that all U.S. health care settings, including EDs, adopt a program of universal opt-out HIV screening of all 13- to 64-year-olds who seek medical care at their facilities regardless of apparent risk.
“There's a real importance for emergency departments to take on this screening role because they are the safety net system,” said Bernard Branson, MD, the associate director for laboratory diagnostics in the CDC's division of HIV/AIDS Prevention and the lead author for the CDC's revised recommendations for HIV testing. “People who don't have any other opportunity for or access to health care end up going to the emergency department.”
These populations may also be more likely to be infected with HIV. A first-of-its-kind analysis released by the CDC in 2010 found that 2.1 percent of low-income heterosexual individuals living in high-poverty urban areas were HIV-positive, a figure that's exponentially larger than the national average prevalence rate of 0.45 percent.
But is routine HIV screening of all comers within a certain age group realistic for today's resource-depleted, overburdened emergency departments? And how can it be implemented?
Many models exist, but they can be roughly divided into point-of-care testing done on-site in the ED and laboratory testing in which blood samples are sent to the hospital's lab for analysis.
The laboratory model is now being used on a large scale at several Harris County, Texas, hospitals, including Houston's Ben Taub General Hospital and Lyndon B. Johnson General Hospital. Houston's HIV infection rate is twice the national average.
“Anyone who is getting blood drawn receives an HIV test [unless they opt out],” Dr. Branson said. Since adopting the program in 2007–2008, the hospitals have tested some 320,000 patients.
The technology that many hospital labs now use to generate blood test results is a multiplatform analyzer containing some 30 different assays, including HIV as well as hepatitis B and C. “You can get the results of these tests in as little as half an hour,” Dr. Branson said. “In Houston, they then employ a service linkage worker stationed in the ED who connects people who test positive with the resources they need to manage their care. This takes the burden off the ED staff.”
Other hospitals, including some in New York with which Dr. Branson is now working, are adopting a similar model. But he said that many other institutions serving low-income populations in cities like Washington, D.C., are still at the point-of-care rapid testing phase, using tests like OraSure or Uni-Gold. “They're doing an oral fluid rapid test with an incubation time of 20 minutes,” he said. “That means if you have one person doing the testing in the ED, they can test three people in an hour. It does work; you do identify patients who are positive, but it's not as efficient as we need to be in order to do this kind of screening.”
Douglas White, MD, an associate clinical professor of emergency medicine at the University of California-San Francisco and the director of ED HIV Screening at Oakland's Alameda County Medical Center-Highland Hospital, worked on the CDC pilot project in 2005. “We've had experience with pretty much all approaches to ED HIV screening: point-of-care, lab-based, using existing staff resources, and using supplemental screeners.”
Dr. White said he tried universal screening in his ED, but such an approach required ongoing supplemental grant funding to support. Instead, his institution has now adopted a physician-initiated HIV testing approach that is completely funded by the medical center without supplemental funding. “I believe this is a sustainable model for ED-based rapid testing,” he said.
On a regular basis, emergency physicians receive in-service education on high-risk subpopulations (people who are homeless, incarcerated, and injection drug users) and suggestive clinical findings (such as a rash suggestive of thrush, recurrent infections, and pneumonias). “They are urged to integrate opt-out HIV testing into their routine care of these patients,” said Dr. White, “but it's all at the physician's clinical discretion. There are no hard-and-fast rules who they can test as long as they meet the basic eligibility requirements: they must be alert and able to legally consent.”
The testing is a standard electronic order just like any other, and is integrated into the laboratory budget. “By integrating all rapid tests into the lab, they assume oversight of quality control, which they were happy to do,” Dr. White said. “Turnaround time is about one to two hours, but that's on par with all the other lab tests we're ordering in the ED. For the most part, turnaround times are not prohibitive because the patients receiving HIV testing are usually undergoing additional laboratory tests as well.”
Physicians at the Alameda ED are now ordering about 6,000 HIV tests per year, and about 0.7 percent of those tests identify HIV-positive patients who were previously unaware of their status. The test, an Abbott assay, costs about $6 vs. nearly twice as much for a rapid point-of-care test. And doing testing on only those identified as at risk by a physician rather than full universal screening is far less burdensome, Dr. White said. “Implementing true universal HIV screening in an ED setting is challenging, and for most EDs, simply not feasible. For example, a universal screening program at our ED would involve upwards of 50,000 to 60,000 tests annually, and would require full-time administrative support, along with round-the-clock testing staff. It would cost hundreds of thousands of dollars to maintain.”
Dr. White conceded that an ED testing program used at the physician's discretion will miss cases of HIV that a true universal screening program will catch. “But even when we did universal screening, we found that patients were being missed. In fact, sometimes the highest risk patients declined screening,” he said. “I'm trying to figure out how we empower our physicians so that we don't miss the HIV among the highest risk patients, the HIV that's right in front of us. I feel like we have to first dedicate our limited resources to the highest risk subpopulations with the highest testing yield before we can even approach screening.”
Dr. Branson acknowledged that the CDC's universal screening recommendations are ambitious. “Despite the recommendations, I don't think that we ever believed an ED could test everyone who walked through the door,” he said. “But we can test a large number of people by making the testing a part of routine practice in a way that doesn't interfere with patient flow and isn't a substantial burden on the ED staff.”
* Read the CDC's universal HIV screening recommendations at http://1.usa.gov/HIV-CDC.
* Comments about this article? Write to EMN at firstname.lastname@example.org.