Let me start with an anecdote.
I went to two different farmers’ markets over the weekend. These were open-air markets, sparsely attended. And it was a breezy day, to boot.
In neither case did I think there was any reason whatsoever to wear a mask. In both cases, I ended up wearing a mask. Not because I thought it made sense, but because I wanted to fit into the crowd. In both cases, the overwhelming majority of people in the marketplace were masked.
I’ve now gone full circle on pandemic mask use. Early on, I couldn’t fathom why people weren’t using face masks. Now, at least in some cases, I can’t fathom why they are.
In today’s news, I see that two East Coast Democratic governors have set a rough timetable for rolling back K-12 school mask mandates (per this reporting.) This is in response to the declining new case counts for Omicron. Right now, it looks like both of them are shooting for a March end of their respective school mask mandates.
In my humble opinion, that’s how it should be done. They are accepting responsibility for this key part of school safety during the pandemic and they are actively managing it. They are planning an orderly transition from masked to non-masked in-person K-12 instruction, based on what I hope are public health objectives. In any case, for this group activity we call school, everyone in the group will be given the same signals regarding mask use, and when it’s time to take off the masks, everyone can be assured that the decision was done with some forethought as to the common risks involved in that.
This is good, if only because this approach avoids chaos and strife. Within the large-group activity we call public school, everyone will get the same set of instructions.
Contrast this with the Virginia approach, where the Governor took no ownership at all over this key school safety issue. Instead of managing the transition for the benefit of all, he issued an executive order to create individual parental-based exceptions to local mask mandates. It allowed parents to exempt their children, one-by-one, based on their opinions (i.e., political leanings) rather than any public health criteria for the student body as a whole. As an extra added bonus, it fairly clearly conflicted with existing Virginia statute, and as a result it has now ended up in court.
Which is bad, unless generating chaos and strife is part of your political agenda.
At any rate, because K-12 school is a group activity, run by the government, there likely will be some form of guidance in most places. Some sort of mask use guidance. If only because in most places, somebody responsible is supposed to be looking out for the health and welfare of the students as a whole.
We adults are not so lucky. We’re each going to have to make our own individual decisions about mask use. And as far as I can see, so far, that’s going to be based on the same amorphous social norms that governed masking up in the first place. To the extent that dropping the masks will be enforced, it’ll be through peer pressure, not through any explicit advice from any government agency.
In short, as an adult, nobody’s going to tell you that masks are no longer needed.
Not in the U.S, at any rate. Other countries seem to be testing the waters for treating Omicron like seasonal flu, e.g., Spain. But here in the U.S., I doubt the CDC is ever going to come out and say that masks should no longer be used.
Yet a third way to compare Omicron risk versus typical flu risk: Brief background.
Let me not belabor this. This is the third in a series of posts that asks the following question:
How low does the U.S. Omicron case load need to get, before the risk posed by Omicron is no higher than that posed by typical seasonal flu?
In Post #1400-3, I did the crude calculation for all persons pooled together, and came up with 16 new cases per 100K population per day. That calculation was absolutely straightforward and easy to check.
In Post #1400-4, I refined that by generating a separate estimate for the boostered population alone. That gave me a benchmark of 40 new COVID-19 cases per 100K population per day. The number is higher because boosters provide significant protection against Omicrion.
But that calculation was anything but transparent. And it was an extremely conservative estimate, in that I only account for the impact of vaccine and booster on the likelihood of getting infected, not on any further reduction in hospitalizations or deaths once infected.
A duh-piphany, or the most obvious way to compare Omicron risk to flu risk.
The main point of this exercise is to compare the risk of hospitalization or death under Omicron, to hospitalization or death from seasonal flu.
In which case, why don’t I .. uh … just compare those rates directly? If I want to compare them based on deaths and hospitalizations, then simply do that. Tabulate Omicron hospitalization and death rates on a 100K population basis, and compare those to typical seasonal flu.
It’s an eclat d’oh.
I mean, when the Omicron hospitalizations per 100K gets down to the level of flu hospitalizations per 100K, then by definition, the average person’s risk of getting hospitalized for Omicron matches the risk of being hospitalized for flu. No further calculation needed, unless you want to try to separate out the boostered, vaccinated, and un-vaccinated populations.
Disease burden of flu in terms of deaths and hospitalizations per day and per 100K population.
I want to compare Omicron risk to risk from flu on a typical day during “flu season”.
The first issue is that I could not find any hard-and-fast CDC definition of flu season. It’s just defined as the months — typically winter through early spring — around the peak of this curve. Typically, somewhere around five months out of the year.
Source: Calculated from CDC burden of flu, 2017-2028 season, assuming 30M total symptomatic flu cases for the entire year.
I’m going to define “flu season” as those weeks with an estimated 500K symptomatic flu cases or more. In the example above, “flu season” lasted 18 weeks, and accounted for just about 70 percent of all flu cases during the year.
Source: CDC disease burden of flu.
Based on that, and rounding the numbers, I come up with the following table comparing hospitalization and mortality rates for typical U.S. seasonal flu and the current levels of Omicron:
On a typical day in flu season, the U.S. sees 2100 flu hospitalizations. Currently, with Omicron, we’re seeing an estimated 12,000 hospitalizations per day. Based on that, for the U.S. as a whole, Omicron cases would have to fall to about 17.5 per day before the hospitalization risk from Omicron matched that of typical seasonal flu, for the average American.
(The mortality data are harder to use because a) deaths lag cases by a couple of weeks, and b) we’re only a few weeks past the Omicron peak. So, compare to the current case count, we’re looking at far too many deaths. And, accordingly, the ratio of current Omicron deaths to typical flu deaths is much larger than the current ratio of Omicron hospitalizations to typical flu hospitalizations.)
As you can see, all I have really done is re-create my first analysis. Pooling all individuals together, you’ll have the same hospitalization risk for Omicron as for flu if Omicron gets down to 17.5 new cases / 100K / day. (My initial analysis came out with 16 new cases / 100K / day).
The only value-added here is that this now directly translates into a COVID-19 daily hospitalization count. That information is available on a timely basis for all states, via the CDC COVID data tracker.
For now, I’m just going to leave it at that. Without being very precise about it, this is just another way of saying that at some point when Omicron cases get into the 10’s per 100K per day, your risk of severe illness from Omicron is no higher than your risk of severe illness from flu.
Tomorrow, I’ll take the final step in this process. I’m going to combine and clean up all the results, and translate them into a set of state-level thresholds comparable to the data publicly available on the CDC COVID data tracker. With that, you should be able to take those thresholds, bring up the CDC data from your state, and identify the time (if any) at which Omicron risk is below typical flu risk for the average resident, and for the fully-boostered resident, of your state.
