Arkansas public health lab coronavirus tests overstate positives

As I recently pointed out, an important recent piece in The New York Times revealed that the coronavirus tests being used in New York and Massachusetts are too sensitive. The coronavirus tests are based on RT-PCR technology. The "PCR" portion of the test involves dozens of cycles of amplification of target viral DNA from biological material collected with the nasal swab, as I detailed. 

PCR cycle cutoffs in the low-to-mid-30s appear to be the most accurate levels to stop and declare a negative. Tests that involve as many as 40 cycles, as New York and Massachusetts did, according to the NYT analysis, significantly increase the likelihood that the tests will produce false positives due to the detection of very small amounts of live virus or only fragments of dead virus. The reviewers found that as many as 70-90% of positives that were based on cycles exceeding 30 should be considered negatives based on the small amounts of live virus being detected. 

Today the Arkansas Department of Health (ADH) confirmed to me that the ADH Public Health Lab uses a PCR cycle cutoff of 42. One researcher's review of commercial tests listed on the Food and Drug Administration's list of approved tests found that most manufacturers that listed a recommended cycle cutoff at all recommended cutoffs in the 30s or, at most, 40. The ADH Public Health Lab's cycle cutoff of 42 is, therefor, exceptionally high, and exceptionally troublesome. 

Based on information released daily by ADH, it appears that at least a quarter of testing results reported by the governor on any given day (and on some days as many as half the results) are from the ADH Public Health Lab. That means that at least that portion of the results are even more questionable than the ones analyzed by the NYT in New York and Massachusetts, and are likely producing a significant number of false positives.

The number of daily COVID-19 PCR tests processed by the ADH Public Health Lab is indicated in white. Positives produced by this lab are extremely suspect given a PCR cycle cutoff of 42, significantly higher than the 40-cycle cutoffs cited by The New York Times as producing far too many false positives in New York and Massachusetts. Source: Arkansas Department of Health

The vast majority of Arkansans who see alarming headlines about the daily numbers of new "positive" SARS-CoV-2 cases in state don't know that the state's own public health lab uses a test whose manufacturer recommends so many cycles before cutoff that the chances of a false positive are significantly increased. Reporters from mainstream news sources seem either too ill-informed or too complacent to ask the governor or ADH about this number and its implications. It's time for them to stop being stenographers and start being journalists.

Richie Graham is based in Little Rock Arkansas USA and writes from a free-market libertarian, anti-interventionist perspective.