Yesterday, after some searching, I found a private-sector entity that provided near-real-time information on the growth of the U.K. COVID-19 variant in the U.S. That was only after finding the U.S. CDC to be absolutely (and shockingly) worthless in that regard (Post #1021).
I was shocked because the U.K. variant is, after all, the next great challenge that the U.S. faces for COVID. And unless we somehow luck out, basic epidemiology (and the experience in Great Britain and now elsewhere) tells us that we will almost inevitably have to face up to this challenge next month. (Post #1007).
Great Britain dealt with this by another round of fairly restrictive economic shutdowns. That’s the only way they could think of to deal with a new coronavirus variant that’s 40% more transmissible than the prior variants. And that’s what’s looming here, now, in the U.S.
So, from a policy perspective, knowing where we stand on the U.K. variant should be a very big deal. And yet, for the CDC, it’s not.
Isn’t that odd?
After I wrote Post #1021, the NY Times published this article on the CDC and tracking COVID-19 variants:
“C.D.C. Announces $200 Million ‘Down Payment’ to Track Virus Variants
Scientists say the new investment will help in the next couple of months but hope that the stimulus package will provide much more.”
OK, now the situation is clear.
- As I inferred from their website yesterday, the CDC is in fact useless as a source of information on the key U.K. coronavirus variant. Instead, they are focused on gearing up to be able to provide something, sometime in the far future. Which is precisely what it looked like, yesterday. But right now, there is zero information on the CDC website to allow you to infer the incidence of the U.K. variant in the U.S.
- Instead of figuring out how best to get this key information in real time, the CDC bureaucracy has apparently been concentrating on how best to leverage this into a permanently larger budget for the CDC. They’re fixated on expanding their budget to include a super-duper (“gold standard”) nationwide monitoring system — that they’ll put in place over the next couple of budget cycles. And the sole internal debate seems to be on just how heavy the gold plating needs to be, on this new gold standard system.
Let me explain exactly what’s so heinous about this.
First, it should take almost zero effort and cost to pin this key information in near-real time. If, as I estimated yesterday, the U.K. variant accounts for 15% of all new cases currently, you would need to test maybe 200 samples per week, to get a usable estimate. Depending on how much accuracy you think you need. And, given how fast this is growing, you don’t need much accuracy. You just need to validate the prior prediction of fast growth.
Briefly, for those of you who aren’t well-versed in statistics. The accuracy of an estimate — in this case, the proportion of new cases that are the U.K. variant — is based on the size of the sample of tests that you look at, not on the fraction of the population that you sample. A sample of (e.g.) 200 tests will give you the exact same accuracy, regarding that U.K. variant, whether or not that’s 200 out of 20,000 or 200 out of 20,000,000.
You can find on-line calculators that will tell you the uncertainty of your estimate, for a “binomial” situation like this. Like this one. And you will find that with an estimated 15% sample proportion, and 200 observations, you’d get a 50% confidence interval of something like 15% +/- 2.5%. Which is plenty good enough to tell whether or not the U.K. variant is doing what it was predicted to do. If you up the sample to 500, you can get to 15% +/- 1.5%.
Just read that NY Times article, and see that all the debate within CDC is about just how spectacular this new system should (eventually) be. “Minimal gold standard” of 5% of tests sequenced. Some call for 15%. And of course, we need to have many more Americans tested, as the third wave of the pandemic winds down. (??)
And at root, all of that push for the absolute best — no matter how long it takes, no matter how many taxpayer dollars need to be spent — does little more than cover up for the fact that the CDC fumbled this from the outset. And that other countries, such as Great Britain, have done a far better job than we have.
This is what you get when you have bureaucrats run the show, not policy experts. What we need right now is some reasonably accurate estimate of where we stand. What the CDC bureaucracy is fixated on right now is getting funding for a system that best serves the needs of science, down the road. That’s the wrong goal.
Second, this is classic bureaucratic imperative allowed to run wild. Instead of answering the question that we need to have answered right now, the CDC is focused on expanding the budget over the long term. They are thinking like bureaucrats whose mission is to serve scientists, not like federal officials who should be playing a key role in making government policy right now.
Sure, there are times when the focus has to be on expanding the size and cost of the bureaucracy. I’d say that right now is not one of them. But, when the Federal government is in a mood to hand out free money, with few questions asked, I can hardly blame them for shifting their focus away from the immediate need, to getting their slice of the pie.
Third, this is of-a-piece with the fundamentally unhelpful output of the CDC over the course of this pandemic. From bungling the initial COVID-19 PCR test, to telling Americans that social distancing alone is adequate to protect you from viral transmission (“don’t wear a mask”), to “wear a cloth mask” with zero guidance on what that means (but no indication that you should use an N95), from their stubborn unwillingness to recognize aerosol transmission of the disease and the garbled public guidance that still generates, to their absolutely incomprehensible public guidance regarding fomites, the CDC’s role in this has just been a slow-motion train wreck.
And so, if you want to know the state of the single most important pandemic issue in the U.S. today, you have to infer it from data provided by Helix corporation (whoever they are). Because, apparently, the bureaucrats who run the U.S. CDC can’t manage to arrange for quick genetic sequencing of 200 randomly-chosen test specimens per week. Or can’t be bothered to, as they focus on the apparently more important issue of taking advantage of the COVID relief bill to enlarge the CDC budget.
Meanwhile, the CDC bureaucrats focus on creating a “gold standard” system, that might, if we are lucky, be able to provide us some information, at some point before the pandemic ends. But provides zero information right now. In the few weeks we have left before the likely start of the U.S. fourth wave of the pandemic.
This is where leadership matters. I expected the bureaucracy to run wild under the last administration, because there was no interest at the top in any serious efforts to control the pandemic. But now? Somebody needs to yank CDC’s chain and get them focused on providing the information we need in the here-and-now. Not the budget they think they want for the future.
Addendum: If I were king of the world.
I’d have somebody beat on the CDC bureaucracy to get them to change their mask guidance to “everybody who can get one should wear an N95 or equivalent”. And blame that bureaucratic 180 on the U.K. variant.
Heck, I think we ought to do what the Germans are doing, and have the Federal government take an active role in distributing N95s to the general population.
Just refer back to the NY Times article on U.S. N95 mask manufacturers who have tens of millions of masks on hand, and no buyers. Just look on Amazon for non-medically-certified (but NIOSH-certified) N95s that are so plentiful that Amazon has had to put them on sale.
All that high-filtration mask inventory sitting around. A new and far-more-infectious COVID-19 variant is almost inevitably coming. But instead of using those surplus masks to address that coming problem, the Federal government just sits on its collective ass, waiting for things to happen.
Meet the new boss. Same as the old boss.
And the other thing I’d do? I’d incentivize the CDC bureaucrats. If I were king, they’d get not one cent for their “gold standard” system until they figured out a way to tell us, right now, where we stand on the U.K. variant, on a real-time basis.
Sure, the bureaucrats would splutter about the impossibility of it all. They’d try to call your bluff. But I’d also bet that the clever lads and lasses of the U.S. CDC would figure out a way to get it done. We just need real leadership to provide the right incentive.