Understanding the costs of testing – false-positive HIV tests

False Positives From H.I.V. Test

Diagnostic tests have several characteristics. Some tests give a spread of results, but others give a simple dichotomous results – positive or negative. When faced with a dichotomous test, we have 4 possibilities: true positive, true negative, false positive and false negative. While we often worry about false negative tests, false positives can lead to major problems.

Health officials in New York and San Francisco said yesterday that a widely used rapid test for the virus that causes AIDS had been producing too many false-positive results, frightening healthy people into thinking they might be infected.

The test, called the OraQuick Advance H.I.V. test, is the same one the Food and Drug Administration has said it will consider approving for sale to the public for home use without a prescription.

But officials in the two cities and at the drug agency also emphasized that the test, which requires a mouth swab instead of a blood sample, should not be abandoned, because its convenience made it a valuable tool in fighting AIDS.

Rather, they said, the problem needs to be investigated, and clinics and patients need to be aware that a positive result must be confirmed by another type of test.

Difficulties with the test were first reported yesterday in The San Francisco Chronicle.

In New York, Dr. Susan Blank, an assistant health commissioner, said that city clinics performed 3,600 to 3,700 tests for the virus, H.I.V., each month, largely using OraQuick, and until recently they had about five false positives a month, a rate well within the maker’s prediction.

But in November, Dr. Blank said, the number shot up to 30, which was cause for concern.

“We have made several lot switches to make sure we have the most recent specimen kits,” Dr. Blank said, adding that the city laboratories had also reviewed their own procedures to make sure they were using the test kits correctly.

She added, “We have spoken with the company several times and written to the company to try to resolve the problem, and have taken steps to inform the F.D.A.”

The city also informed the Centers for Disease Control and Prevention, Dr. Blank said. Rapid testing “remains an important tool in stopping the spread of H.I.V.,” she said, adding, “I think we need to be careful that we don’t throw out the baby with the bath water.”

What rate of false positives would you consider acceptable? Having to tell someone that they have a positive HIV test is difficult enough without having confirmatory testing. With this rapid test, one would have to put the person with a positive test into testing purgatory. As a result of the test, some without HIV must go through the angst of considering that they may have the infection.

So the trade off here is explicit. What percentage of false positive tests should we accept in the name of rapidity? This question has no easy answer. What will happen to false positive rates if people use the test at home? Who will be available to counsel the person with the positive result?

I am not posing philosophical arguments here, rather I am asking that we think carefully through the advantages and disadvantages of rapid tesing. Some would argue the advantages of speed outweigh the angst of those with false positive tests. I do not see this problem through large numbers. I must think about how those with false positive tests feel. What are those costs?