Patients administer HIV tests as accurately as pros

In what could be a boon in the fight against HIV, a new study of 402 people in an ER reveals that the untrained can administer HIV tests as well as health care workers.

Elizabeth Armstrong Moore
Elizabeth Armstrong Moore is based in Portland, Oregon, and has written for Wired, The Christian Science Monitor, and public radio. Her semi-obscure hobbies include climbing, billiards, board games that take up a lot of space, and piano.
Elizabeth Armstrong Moore
3 min read
HIV self tests work similarly to pregnancy self tests, and could become just as common. Daquella manera/Flickr

Studies suggest that anywhere from 2 to 13 percent of patients in emergency rooms are HIV positive, according to Charlotte Gaydos, a clinical microbiologist at Johns Hopkins University. So the emergency room seemed like a logical place to test whether an untrained person is able to self-administer his or her own HIV test, and then accurately read the results (one line means negative, two lines mean positive).

In an urban hospital, researchers from Johns Hopkins offered people in the emergency room the option to test themselves for HIV while they waited. More than 90 percent of the people they asked agreed, resulting in 402 people performing a test that required following written instructions and either a pin prick or mouth swab. The participants then put their blood or saliva samples into a tube and, in just 20 minutes, saw a positive or negative result.

Separately, hospital officials trained to administer the same tests replicated them on each of the 402 participants. The self tests matched the hospital tests in 400 of the 402 cases, the only two strays being the result of what looks like a lab error in one case and not enough of a blood sample in the other.

Additionally, of the 402 people tested, only two people tested positive, and they both tested positive in both tests. In other words, researchers did not detect a single false positive or false negative. The study, still ongoing, will be published this winter in the Annals of Emergency Medicine.

Gaydos says that with further study, self-testing for HIV could get regulatory approval and become routine, just as pregnancy testing has become. This could greatly improve early detection, which is important for not only preventing the spread of the virus, but also leading to longer life spans among those who do test positive.

Normal CD4 cell counts range from 400 to 1,200 cells per cubic millimeter of blood, and people with counts less than 200 are far more likely to die within 10 years, according to Michael Saag, an infectious disease physician at the University of Alabama at Birmingham. "Early testing enables people to live a relatively normal life span," Saag tells Science News.

When I called the Center for Point-of-Care Tests for Sexually Transmitted Diseases at Johns Hopkins this morning, I was told that researchers are not currently tracking patient reaction, which seems like a potentially big part of how successful self-testing can be.

The researcher I spoke to says that of the two people who did test positive in their study, one was a man who already knew, and the other was a woman who did not. There was obviously some shock involved, but the researchers did not officially record her reactions, and have no plans at present to incorporate this into the study.

Of course, someone who opts to administer an HIV test for the benefit of a scientific study while waiting in the ER is probably going to be more shocked by a positive result than someone who actively goes out and procures that kit based on some suspicion. And being able to track one's HIV status privately, at home, could help keep more people in the know, which could in turn result in a slower spread of the virus.

But researchers writing laminate instructions for how to take this test might be wise to include, at the very least, a hotline for people to call should they be staring at very bad news.