HIV COURT CASE RAISES QUESTIONS ON TEST

By Randy Dotinga

Health Scout News 23 Jan. 2002

In a case that raises questions about the accuracy of HIV tests, an Oklahoma man has won a $1.4 million settlement nine years after a health clinic mistakenly told him he was infected with the virus that causes AIDS.

Medical experts say it's nearly impossible for an HIV test to give an incorrect -- or "false positive" -- result. However, it's widely known that negative results can be wrong if a test is given too soon after infection, before the virus can be detected. The weakest link in the test process appears to be medical personnel who can mix up blood samples or misplace paperwork.

The Centers for Disease Control and Prevention(CDC) says that, in "ideal" testing circumstances, the odds of a false positive HIV antibody test are less than five in 100,000, according to a report in USA Today. However, human errors can occur.

"No test is perfect, and that goes for cancer tests, X-rays, MRIs and blood tests," says Dr. Jeffrey D. Klausner, director of sexually transmitted disease control services with the San Francisco Department of Public Health.

Even so, HIV test errors are considered rare, and experts say it shouldn't take long for any alert doctor to notice a person with a supposed HIV infection isn't getting sick.

That didn't happen in the Oklahoma case, apparently because the plaintiff didn't receive extensive medical attention.

According to news reports, 40-year-old Anthony Northcutt had an HIV test in 1993 at a clinic operated by the Oklahoma City-County Health Department. He was told he was HIV-positive.

Four years later, Northcutt obtained his file and learned he had received the wrong test results. He was actually HIV-negative at the time of the first test.

Two months after learning of the wrong results in 1997, Northcutt took another test and learned he was HIV-positive. He reportedly had engaged in unprotected sex with HIV-positive partners during the years he thought he was infected.

In his lawsuit, Northcutt claimed he struggled with depression, suicide attempts and a drinking problem.

According to The Daily Oklahoman, a jury found in favor of Northcutt, but also determined he was partially negligent himself. Attorneys for Northcutt could not be reached for comment.

HIV tests determine if a person has developed antibodies to the virus, a sign that the immune system has kicked into action. However, it may take several months after infection for enough antibodies to appear.

An ordinary HIV test actually consists of three or more separate blood tests, at least in California, says Michael Allerton, who oversees HIV policy in Northern California for the Kaiser Permanente Health Plan.

If the first test produces a "false positive," the others should catch the error, he says. In some cases, however, the tests may conflict with each other, and more tests are needed.

Even if an HIV-positive result is wrong, doctors should notice when a patient doesn't show signs of the virus in his or her blood, says Dr. Michael Horberg, a Kaiser Permanente Health Plan physician who treats AIDS patients in Northern California.

"They'd have no detectable virus at all without any treatment," Horberg says.

In one of his own cases a few years ago, he says, a woman insisted she could not have contracted HIV. She turned out to be right. She wasn't infected, despite a positive test result.

"We did a mega work-up, kept following her every six months, and she was still negative," he says. "She never showed any sign of the virus, and her [immune system strength] was better than half the U.S. population."

Powerful AIDS drugs can reduce the level of HIV in the blood to zero, although the virus will continue to lurk in other parts of the body. However, doctors wouldn't put someone on the drugs if he or she weren't already showing symptoms of HIV infection, Horberg explains.

Some people do worry that they got the wrong test results, Horberg says.

Following the Oklahoma verdict, "I had tons of people calling and saying they wanted to be re-tested. You re-test them if that's what it takes to convince them that it's accurate," he says.