False positives in screening tests

In many jurisdictions, tests for human immunodeficiency virus (HIV) are required screening for pregnant women. Some even suggest screening the entire population for HIV. The rationale is to administer powerful antiretroviral drugs to healthy individuals to "prevent transmission."

But what if healthy people take expensive, toxic drugs for a condition they cannot transmit because they don't have it?

In the spring 2010 issue of the Journal of American Physicians and Surgeons, Henry H. Bauer, Ph.D., writes that "HIV tests are not HIV tests." A positive test does not mean that a person has or will get AIDS. It doesn't even mean that he has HIV. Rather, it is a test for antibodies, either to HIV or to something that cross-reacts with it.

The FDA said in 1987 that "the significance of antibodies in an asymptomatic individual is not known." It is still not known, as positive tests have been reported in many illnesses, including multiple sclerosis, tuberculosis, malaria, and aplastic anemia. They have also been reported after immunizations to tetanus or influenza, and may even be caused by pregnancy itself.

"The problem of false positive tests applies to all screening tests, not just HIV," states Jane M. Orient, M.D., executive director of the Association of American Physicians and Surgeons (AAPS). "If a population with a very low likelihood of a disease is tested, there will be very few positive tests. But even with an excellent test, a positive is more likely than not to be a false positive."

If only one in 1,000 tested individuals has HIV, then only one out of every six positive tests is a true positive, even if the HIV test is as good as is claimed.

This is why a test that is a very good for screening blood donors can do much more harm than good if used to screen the entire population, Dr. Orient stated.

Dr. Bauer's article on HIV testing can be accessed at http://www.jpands.org/vol15no1/bauer.pdf.