Use of HIV testing in ED hits snags

Almost everyone should be offered an HIV test.

At least that's the take-home message of the Centers for Disease Control and Prevention guidelines issued in September 2006 with the support of many medical societies, including the American Medical Association.

Now, emerging data from the earliest attempts to study the implementation of this directive suggest that many patients will take advantage of the opportunity to be tested. Studies also are showing, though, that fully integrating this policy into medical practice can be difficult logistically.

Several presentations at the Society for Academic Emergency Medicine meeting in Chicago last month documented that a significant percentage of patients presenting to emergency departments for care said "yes" when offered a rapid HIV test.

"Patients think this is a great idea. People coming in for sprained ankles and cuts on their hand who didn't think they were going to go to the hospital that day -- let alone get an HIV test -- were willing," said Dr. Jeremy Brown, research director of the Dept. of Emergency Medicine at George Washington University Medical School in Washington, D.C.

His study found that about 60% of those offered the test between September and December 2006 consented. Of the 2,476 who agreed, 26 had preliminary positive results. Nine of these were verified as positive by later testing. But significant problems emerged with regard to confirming preliminary results or connecting patients to appropriate care. For various reasons, 13 patients were lost to follow-up.

As a result, this institution has since changed its protocol. All patients with a result suggesting that they might be HIV-positive receive confirmatory testing and consultation with an infectious disease fellow at the time of the initial visit. Contact details are also confirmed.

"We have got to have a linkage to care," Dr. Brown said.

Other studies identified additional barriers to the widespread testing advocated by the CDC. Most notably, staff may not have time to administer the test, get the results and deliver them to the patient in the course of providing other services.

A study observing HIV testing in the emergency department and urgent care center at the Alameda County Medical Center in Oakland, Calif., found that 52% of patients in the emergency setting and 48% in the urgent care center who were offered the rapid test agreed to have it done. But only 39% of those in the emergency department and 40% of those in the urgent setting actually were tested.

"Many patients were willing, but it was very difficult for the nurses to integrate it into care and not hold up a patient's discharge," said Douglas White, MD, the paper's author and the HIV project lead investigator at the medical center.

Also, despite the fact that a majority of patients opt for the test, a significant percentage don't. Researchers say gaining insights into these refusals is necessary to achieve the CDC's objective of having more people know their status. Dr. Brown's study found that those younger than 25 were more willing to be tested than those who were older. African-Americans were more willing than Caucasians, and local residents were more amenable than those who were from elsewhere.

Meanwhile, a study conducted at the University of Cincinnati College of Medicine suggested that at-risk patients might not realize the reality of their situations, and a lack of insurance could be a barrier, even when testing is free. "We need to understand patients who refuse better than we do," said Ali Raja, MD, lead author on the paper and a third-year resident at Cincinnati.

The question of how these recommendations will be funded is also an important issue. The testing carried out in Washington, D.C., was offered by additional research staff using testing kits provided by the local department of health. One confirmed case of HIV cost $4,900 to detect, but this expense is expected to be higher without such support. Another study, this one done at St. John Hospital and Medical Center in Detroit, found that unreimbursed routine HIV testing could cost hospitals more than a half-million dollars a year.

These studies also took place in locations with high HIV prevalence, and it is unknown how this will play out elsewhere. Another study, for instance, found evidence that the HIV prevalence in a health care setting might not necessarily reflect what is found in the community.

"Undiagnosed HIV can still be an issue in a relatively affluent area, although the magnitude of the phenomenon is going to be different," said Michael Lyons, MD, lead author of the Cincinnati paper and an assistant professor in the college's Emergency Medicine Dept. "HIV is more of a problem in some settings, but that doesn't mean that it's no problem whatsoever."

His study estimated that the Cincinnati metropolitan area in 2002 had an HIV case rate of 1.7 per 100,000 but in the recent study, the academic emergency department had a positive-test rate of 7.8 per 100,000. The urban ED had 6 per 100,000, and the suburban one had 3.5.