Replies

The requirements from CDC make testing difficult to obtain. Ridiculous. I don’t blame the politicians, I blame the unelected bureaucrats at the CDC.

In away, it’s stupider than that, but also more arcane.

The declaration of an emergency Jan 31 brought more powers to bear, but also invoked new rules. The Association of Public Health Laboratories (APHL) had to figure out and then submit a request to the FDA to get around these rules (which the CDC wasn’t allowed to bypass) which didn’t normally exist, which wasn’t actually done until Feb 24. The APHL finally asked the FDA, and the FDA agreed to ignore the rules.

Feb 28th:
https://www.sciencemag.org/news/2020/02/united-states-badly-bungled-coronavirus-testing-things-may-soon-improve

In principle, many hospital and academic labs around the country have the capability to carry out tests themselves. The PCR reaction uses so-called primers, short stretches of DNA, to find viral sequences. The CDC website posts the primers used in its test, and WHO publicly catalogs other primers and protocols, too. Well-equipped state or local labs can use these—or come up with their own—to produce what are known as a “laboratory-developed tests” for in-house use.

But at the moment, they’re not allowed to do that without FDA approval. When the United States declared the outbreak a public health emergency on 31 January, a bureaucratic process kicked in that requires FDA’s “emergency use approval” for any tests. “The declaration of a public health emergency did exactly what it shouldn’t have. It limited the diagnostic capacity of this country,” Mina says. “It’s insane.”

On 24 February, APHL asked FDA Commissioner Stephen Hahn for “enforcement discretion” to sidestep the emergency process and allow APHL members labs to use their own tests. On 26 February, Hahn replied that the CDC test could be modified to use just the primers that specifically detect SARS-CoV-2, essentially ignoring the faulty portion of the kits. FDA, in other words, would look the other way to make more widespread testing possible.

PCR-based assays are very sensitive (if sample preparation is appropriate), but any contamination can lead to a lot of false positives. This is a particular consideration at this point because it appears there will be limited numbers of labs running the PCR reactions, and thus those labs will have exposure to multiple samples - increasing the possibility of contamination and thus false positives.

Labs will undoubtedly run control samples with each PCR run, but it's still not foolproof with regards false positives. Plus, this virus is a RNA virus, and thus testing will require a reverse transcription step - which is another potential exposure to error.