Over the weekend I read a story about a reporter who drove to Orange County, remained in his car in line for 2 hours and received an antibody test purported to be of 90% accuracy. The reported noted that “90% seemed like a good deal to me.” However, 90% is not a good deal as there are two errors that can be made in testing – a falsely positive test and a falsely negative test. An HIV test for example is 99.9% and 99.5% accurate in these measures.

 

The FDA has now issued a letter warning healthcare providers about the limits of serological tests to detect SARS-CoV-2 antibodies. The agency is urging clinicians to “not use serological (antibody) tests as the sole basis to diagnose COVID-19 but instead as information about whether a person may have been exposed.”

 

In mid-March, the FDA “provided regulatory flexibility” (I think this is a generous description but I applaud the author for their linguistic creativity) for test developers. This has resulted in scores of antibody tests quickly hitting the market without the agency’s usual review. As of April 18, just four antibody assays had received emergency use authorization from the FDA.

 

A New York Times story details the low accuracy seen with many of the tests on the market — sometimes as low as 20–30%. One infectious disease expert told the Times: “People don’t understand how dangerous this test is. We sacrificed quality for speed.”

 

I am hopeful that within these tests is one that is effective and accurate. Now we just need to take the time to identify which one that is.