The Other COVID Statistic We Still Need

There has been a lot of excitement over wastewater testing for COVID. Even pundits who know no microbiology are excited about it! Wastewater testing is useful, but there are a couple of caveats. First, the way the CDC reports the data is essentially useless. The percent change over the last fifteen days doesn’t really help. Not only doesn’t it tell us the absolute level of COVID and thus can’t be compared to earlier periods, but it is extremely difficult to interpret. If we drop from a 455% increase to a 148% 15-day increase, we can’t really determine what that means. Some states are reporting the data using actual measurement values–of course, D.C. Water and DC Health aren’t (good job, Mayor Bowser!).

The second issue, and more serious, is that it’s clear policy makers won’t do anything with these data–they haven’t really done so to date, and there’s no reason to think they’ll start. Even with reporting of absolute levels, I wouldn’t know when to take additional precautions (though it might be useful for hospitals so they can prepare).

Speaking of hospitals, that brings us to the other data we need, which I’ve been calling for since 2020: the percentage of asymptomatic hospital patients who test positive for COVID. It’s not perfect, but it gives us a real-time indication of what the prevalence of COVID is in the community. If you need to know what the prevalence–which is to say, the risk–is, this would certainly provide what you need: is the prevalence ten percent, one percent, less than a tenth of a percent? That’s useful information for individuals, and since our ‘national’ policy is to leave it up to individuals to protect themselves, at the very least, give us the damn data to do so.

Wastewater is good, but we also (still) need the asymptomatic patient data too.

This entry was posted in COVID-19. Bookmark the permalink.

3 Responses to The Other COVID Statistic We Still Need

  1. becca says:

    I have a colleague who does Covid wastewater testing, and I believe the behind-the-scenes issue is that the methods for detection are heterogenous enough that absolute quantification using the same standards is a bit of a pipe dream for now. Some of this may be because wastewater is a *variable* substrate- I know from other sources that the amount of rain matters enormously if your sewer system is all combined, and there are a bunch of things like that. At least the detection she does is all NGS based, which is pretty useful for variant dynamics (assuming all variants are equally stable, I suppose), but complicated for absolute quantitation.
    I do know different site approaches very a lot, and that there are some attempts to normalize to DNA content of dietary components (peppers, apparently?), which sounded like it would turn out exactly as anyone who has sat through multiple lab meetings where people debated whether beta actin or GAPDH was a better Western blot control would expect.
    To be 100% honest, I was a bit incredulous when I first learned we *could* do Covid detection with wastewater. Turns out, I was *super* wrong about RNA stability. But also, it IS harder than all those people who never cared about microbiology before will ever know. But then, so is a Western blot normalized to GAPDH- that’s a Them problem. But even within the field, it can be helpful to remember it’s kind of great this works even a little bit (for rapid variant shifts, ect.).

    Having different asymptomatic populations who are routinely monitored (my nominations would be not only hospitals, but long term care facility residents and workers!!!) is SUPER important data as long as we are comparing longitudinally for a given site. It’s why I’m Extra Big Mad at the colleges and employers like mine who have just pulled the plug on regular testing. It’s just. So. Stupid.

  2. Pingback: “The Chance of Rain Is 10-99% Higher Than Yesterday” | Mike the Mad Biologist

  3. Pingback: Links 4/17/22 | Mike the Mad Biologist

Comments are closed.