Author: chogan@directresearch.com

Post #1323: COVID-19 winter wave accelerates

 

This year’s U.S. winter COVID-19 wave is not quite looking like an exact repeat of last year, but it’s looking fairly close.  It’s running a couple of weeks late, probably owing to warmer weather this year.  It’s starting from a higher base rate of infections, owing to the tail-end of the U.S. summer (Delta) wave.

But it’s now starting to parallel last year’s wave.  Like so:

Data source for this and other graphs of new case counts:  Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 11/16/2021, from https://github.com/nytimes/covid-19-data.”  The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html.

It’s not a straight-up rerun.  Last year the winter wave was led by the Midwest and Mountain states.  The Midwest seems right on track, but the Mountain states are not.  That’s probably worth look at, at some point.

That said, the Mountain states behaved oddly last year as well.  Where the Midwest had a single, well-defined peak (blue line above), the Mountain states peaked and receded three times before the wave was finally over for them last year (gray line).  So it’s not clear how solid an indicator they are.

By eye, this seems to be shaping up to be more “compressed” this year compared to last year.  By that I mean that the lag between the leader (Midwest) and followers (all other regions) looks to be shorter than last year.  For example, the South in general (South Atlantic, South Central regions) peaked about two months after the Midwest, last year.  By contrast, this year they appear to be turning upward just three weeks after the Midwest.

If true, that would give the overall U.S. average curve a steeper profile than last year.

FWIW, my house finally dipped well below 50% relative humidity this morning (northern Virginia, USDA hardiness zone 7).  I’d guess my house is average size and construction for this area.  It’s a fair bet that most of the indoor spaces in this area are making that transition right about now, with increased rates of disease transmission soon to follow.  We’re not quite at the 40% relative humidity level where risk of transmitting respiratory diseases increases sharply.  But it should only take a few more days of cold weather to get this part of the country to that state.  I’m prepared (Post #895, humidifiers).

Have we learned anything important?

I think the answer is pretty clearly no. 

Obvious point #1:  Masks.

Let’s start with the most obvious one first.

Last year, the entire first-world northern hemisphere had a severe winter wave of COVID.  Except for the Asian countries (Japan, Korea, China).  The worst of the three was South Korea, which topped out at about 1000 cases per day.  Or, on a per-capita basis, Korea’s winter wave peak was about 2.5% as high as the U.S. winter wave peak.

This year, the entire first-world northern hemisphere is in the process of having a severe winter wave of COVID.  Except for the Asian countries.  Again.  Japan and China have virtually no cases.  South Korea is again the outlier there, with about 2000 new cases per day, or, on a per-capita basis about 14% of the current U.S. average.

I wonder what might possibly explain that. What behavior could those first-world Asian countries have in common what we somehow lack in America and Europe? /s

Here’s a Forbes article documenting better than 90% mask use rate in public spaces in Japan.  That’s based on both survey data and from on-the-ground observation.  The comparable survey-based data from the U.S. would put us in the mid-30-percents (for “always wear a mask”).

It’s not as if this is a secret.  It’s just that, as a country, we are amazingly stupid and stubborn about this particular point.  When the U.S. CDC initially declared that Americans didn’t need to wear masks, it took a rebuke from no less than the head of the Chinese CDC to make them see the light (Post #590).  I think that’s well worth repeating, even if the head of the Chinese CDC said it more than a year and a half ago.  Note “The big mistake”.

Q: What mistakes are other countries making? 

A: The big mistake in the U.S. and Europe, in my opinion, is that people aren’t wearing masks. This virus is transmitted by droplets and close contact. Droplets play a very important role—you’ve got to wear a mask, because when you speak, there are always droplets coming out of your mouth. Many people have asymptomatic or presymptomatic infections. If they are wearing face masks, it can prevent droplets that carry the virus from escaping and infecting others.

Source:  Science, the magazine of the American Academy for the Advancement of Science. March 2020.

Obvious Point #2:  Lags in data reporting.

The second point that I think will never seep into the American consciousness is that today’s case counts reflect the infections that were occurring a couple of weeks ago.   That’s due to the time it takes for symptoms to develop, for symptomatic individuals to decide to seek care, to get tested, to have the labs turn those tests around, and to have pubic health agencies tally and report those test results.

Which means, obviously enough, that you have to be proactive.  You need to put your mask back on and change your behavior when cases start to rise, not once the hospital beds are full.

At this point, it’s beyond optimistic to think that Americans are going to be proactive.  So we are once again going to see preventive steps taken by the average American only after the hospitals are overflowing, and that gets reported in the popular press.  That seemed to be the point at which things started to sink in last time around.  I’m guessing it’ll be the same this time.

Point #3:  Putting 1 and 2 together.

Obviously, the right thing to do, given how this is shaping up, would be to increase COVID-19 hygiene now.  Ahead of time.  Put your mask back on now, before case counts go up.

It’s too late for vaccination, I think.  Right now, 59% of the U.S. population is fully vaccinated.  That compares to 70% to 76% for the Asian countries above, 75% of Canada, and so on.  We aren’t quite down with the third world, but the list of countries with higher COVID-19 vaccination rates than the U.S. is kind of embarrassing.  Brazil, Argentina.  Mongolia?  Mongolia has a higher COVID vaccination rate than the U.S.?  Yep.

But at this point, it looks like, with a few exceptions (e.g., California), our collective state strategy is going to be to ignore this until the hospitals are full.  And maybe even after that point.

The only bright spot I’m seeing is that if we have one big blow-out of a winter wave in the U.S., then that ought to be the end of it.  At that point, between vaccination and infection, enough people ought to have enough immunity that this can recede into the background.  Looks like it’s going to be around forever.  But there won’t be enough potential carriers left to carry it on at pandemic levels in the U.S.  At least, that how it appears when I do my calculations.  YMMV.

Post #1322: Changing Town of Vienna elections: Avoiding high voter turnout?

 

There’s a public hearing tomorrow (Monday 11/15/2021) regarding proposals to change Town of Vienna elections.  This is in response to legislation that requires all Virginia towns to hold elections on the standard first-Tuesday-in-November election day, instead of in May.   According to the sponsors of that legislation, the point is to boost voter participation in those local elections.

I foolishly thought that the transition from May to November elections was cut-and-dried, because Virginia statute lays out a simple process for doing that.  As you move the election date back six month (from May to November), you give each incumbent roughly an additional six months in office.  After that transition, no other change is needed.

Then I took a look at the complex wording of the options the Town is considering.  And I could not for the life of me figure out why they’ve opted to make the proposed changes.  Why take something as potential simple as this, all laid out neatly in statute, and complicate it?

I had to put all the options in a table to realize that the only thing NOT up for discussion in the officially-sanctioned options is even-versus-odd-year elections.  It looks like the citizens of Vienna can have any November election schedule they want — as long as all elections are held in the off-years. 

This choice runs  contrary to what the overwhelming majority of November-voting Virginia Towns have chosen.  Choosing the odd years means not just being in the minority on this issue, it means minimizing voter turnout for Town of Vienna elections into the foreseeable future.  I think that’s a bad choice.

Background:  This could be straightforward.

If you need some background on the general issue of the switch to November Town elections in Virginia, you can read through these prior posts:

  • Start with Post #340, which introduces the benefits of moving Town elections to the standard first-Tuesday-in-November election day.  That’s where I first learned that Herndon’s election turnout tripled when they did that.
  • Post #1095 introduces recently passed legislation in Virginia that moves all Town elections to the standard first-Tuesday-in-November date.
  • Post #1135 was about our more recent town election, which, barring new legislation, will have been the last May election in Vienna.

Throughout my discussion of the change, I assumed that the Town would follow the transition to November elections as laid out in Virginia statute.  The transition from May to November elections is laid out in “§ 24.2-222.1. Alternative election of mayor and council at November general election in cities and towns“.

C. ... Mayors and members of council who were elected at a May general election and whose terms are to expire as of June 30 shall continue in office until their successors have been elected at the November general election and have been qualified to serve.

In effect, you give the incumbents another half-year in office, for the transition from May to November, and you’re done.

And so, this transition could easily be handled by holding elections in November and making no other changes, other than lengthening the terms of the incumbents as specified in Virginia statute.  Right now, Town Council has staggered two-year terms, with three seats up every year, and the Mayorship in every even year.  All they have to do is extend terms as outlined in the law, and they’d be done.  Elections would occur every November, and that would be the only change.

If you want to see the likely impact on voter participation, just look at nearby Herndon (above).  They switched to November voting in (the even year of ) 2016.  And voter participation in their Town elections roughly tripled.

But for some reason, that’s not what’s on the table in the Town of Vienna.

I had (foolishly) assumed the Town would do just that.  Keep it simple, and change the elections as outlined in statute.  But I was wrong.  As with much of what the Town does, it’s all but impossible to take the written documents and understand what’s being proposed.

Aside:  It shouldn’t be this hard.

You can find some limited popular-press reporting of this issue.  That reporting is the only place where it is clearly stated that you, as a citizen, may suggest your own alternative to the Town’s proposed changes, at this hearing.  But when you try to find out what the actual language of the proposals is, that’s when it gets difficult.

Let me put aside just how hard it is to know that this is in play.  Chrome on a PC won’t even open up the Town’s website due to a misconfigured security certificate message.   Firefox on a PC or Chrome on a phone will open it, but the splash page is so misconfigured that you can’t actually read what the public hearing is supposed to be about.  The text bleeds off the page.  And if you click on the public hearing link, you are informed that there is, in fact, a public hearing on Monday.   But with no clue as to what the hearing is about.  As pictured below:

Long story short, if you already know that this is happening and you already know where to look, you can find that out.  You have to look at the Town’s Legistar page, find the link to the meeting, and find the particular agenda item.  (And realize there are, I think, at least two separate public hearings scheduled for Monday).

If you read the Town’s official notice, and you’ve read the popular press reporting, then you might be able to infer that this phrase ” The Town Council will consider the listed options, options as modified, as well as additional options suggested at the Public Hearing.” means that you have been invited to suggest alternatives.  But, honestly, I read that and I didn’t grasp the fact that “additional options suggested at the Public Hearing” was an invitation for the public to suggest alternatives. 

Maybe I’m just slow, but I had a hard time grasping what all the complication was about.  Particularly because, as noted above, Commonwealth statute lays out a simple and obvious transition.

Instead, the Town is considering the following options: Changing to staggered four-year terms (three Town Council seats up for re-election every two years), or to un-staggered two year terms (all Town Council seats up for re-election every two years).  They’re considering some combination of giving some incumbents an additional year-and-a-half on their existing term, skipping one year’s election, and inserting some three-year terms, in order to make that all work out.

Once again, the complexity of that struck me as odd.  It’s almost as if they’ll consider anything but the system we’ve used for decades (half of Town Council is up for re-election every other year).  I would have assumed that the baseline was to do the simple thing (each existing term is about six months longer, no other changes.)  But that’s not even on the table.

It wasn’t until I put the three options in a table that I finally figured it out:  You can have any voting you like, as long as the vote is held in the off years (odd years) only.  Based on some earlier reporting on this issue, this apparently is what the majority of current Town Council wants.  That seems to be based on the fear that holding Town elections in the even years will somehow taint them with partisanship, or reduce attention on Town issues.  (N.B., as is true in almost all states, local elections of this type are non-partisan.  Since 1870, Virginia has barred the listing of party affiliation on ballots for local elections.)

Let me put the table of options here, because without that, based solely on the text descriptions, you may have a hard time seeing this plainly.

The key point is circled in red.  Much of the rest of the complexity, highlighted in the notes section, is a consequence of moving the current even-year elections to an odd year.  By simple arithmetic, doing that is going to require one term with an odd number of years.  By law, you are not allowed to shorten anybody’s existing term to make this transition.  So, practically speaking (barring having a transition election for a one-year term of office), moving the current even-year election to an odd year is going to involve a three-year something for the incumbents up for election in 2022.  Either give them a three-year term when they are re-elected in 2022 (Options 1 and 3), or convert the existing terms to three years by skipping an election (Option 2)..

I’ll make the obvious point that you could just as easily configure this so that elections are held in the even (high-turnout) years.  Or just leave it as it is, do the simplest possible transition, and have half the elections in the even years, and half in the odd years.

The data

Source:  Virginia Department of Elections

It’s no secret that election turnout is higher when there’s a high-stakes national election.  Above you see Virginia’s turnout (as percent of registered voters) for the past 45 years.  Turnout is highest in the (even) presidential election years, and lowest in the (odd) year just prior to a presidential election.

Let me formalize that by actually calculating the averages instead of just eyeballing it.

:

Source:  Virginia Department of Elections, and, separately, calculated from Fairfax County Office of Elections data for the Town of Vienna (TOV).

Above, first two bars, in Virginia, you get about one-third more voters (about 16 percentage points higher turnout) in even years.  Above, second set of bars shows that participation runs coincident with the U.S. Presidential elections.  Above, third set of bars shows the dismal turnout for a typical Town of Vienna election, and the mediocre turnout even in a hotly-contested election.  (It isn’t unusual for all seats in the Town elections to go uncontested, so low average turnout isn’t unexpected.  For those elections, there’s no practical point to voting.  For the uncontested elections, my wife votes, I don’t.  But even for a contested election, less than one-quarter of the electorate votes.)  I documented Town of Vienna election turnout in Post #266.

The obvious implications here are that, in terms of maximizing voter turnout, any November election is better than the May Town election.  And that November elections in even years are superior to November elections in odd years.

That said, we can ask one final empirical question:  What do other Virginia Towns do?  In particular, what do the Towns that already have a November election date do?  As I noted in Post #340, almost half of Virginia Towns have already opted for November elections.  So it’s not as if we lack for data on the typical choice.

Source:  Tediously calculated from the .pdf supplied at this page, by the Virginia Department of Elections.

Virginia Towns with local elections in November have opted to hold those elections in the higher-turnout even years, by a 4.5 to 1 margin.  Town Council seems to be suffering from some free-floating fear of partisan taint of local elections held in even years.  The clear point of this table is that a) a lot of Towns manage to survive, and b) if you want to base the decision on facts, there are seventy-odd Virginia towns that should be able to answer the question “are local elections tainted if held in November of even years”.

In any case, that vague and un-documented fear aside, the whole point of moving the elections to November is to increase voter participation in local elections.  If that’s the goal,  then to me — along with the clear majority of Virginia Towns so far — even years are clearly superior to odd years.

Vienna could leave things much as they are, and have half of the local elections in even years, half in odd years.  They could follow the example of the vast majority of Virginia Towns with current November elections and go for the higher-turnout even years.  Or they could go out of their way to pick the years with lower voter turnout, based on the un-documented fear of partisanship in local elections.

Seems like if you’re going to do that, minimum due diligence would be to call up a few of the Towns with November elections and try to benefit from their experience.  Heck, take a field trip:  Herndon, Dumfries, and Leesburg are nearby towns with even-year November elections.  (Per the Virginia Department of Elections).  Why not ask them?  In short, do anything but lock in lower voter participation, for all eternity, based on what amounts to an undocumented fear, when you could easily put that fear to rest (or find out that it’s real!) with a few phone calls.

Post #1321: William and Mary COVID-19 trend to 11/12/2021 — one more case this week, and the U.S. winter wave in the South Atlantic states.

Source:  Calculated from the William and Mary COVID-19 dashboard, accessed 11/13/2021

2021 Winter wave in brief.

Things look great at William and Mary, but for the U.S. as a whole, it looks like the winter wave has started.

Last year, the winter wave started in the Midwest states.  Eventually, all areas of the country saw a mid-winter increase in daily new cases.  Last year, the South Atlantic states ran a month or two behind the Midwest.  The Midwest peaked around Thanksgiving, the South Atlantic did not peak until mid-January 2021.

This year, new cases turned upward in the Midwest around Halloween. And now, about six weeks later, it looks like cases are just starting to rise in the South Atlantic states.

That’s later than last year, plausibly due to much warmer weather in the middle of the country this year, compared to last.  Below are maps for October 2020 and October 2021.  Last year (top) was much colder than normal in the middle of the country, this year (bottom) has been much warmer than normal.

In any case, below, that little upturn in cases at the right edge of the graph may not look like much.  But it’s more-or-less right on time to be the start of the winter wave in Virginia this year.  The fact that several adjacent states (NC, VA, WV, MD, DE) all show the same pattern suggests that it’s weather-related, and not just a statistical fluke in (say) Virginia’s case counts.

Data source for this and other graphs of new case counts:  Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 11/13/2021, from https://github.com/nytimes/covid-19-data.”  The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html.

My best guess is that the little upturn above is, in fact, the start of the winter wave in Virginia.  At this point, there’s no telling where that’s going to go.

I guess I’ll mark the start of the U.S. winter wave at 10/25/2021.  That’s when the U.S. daily new case count reached that most recent minimum.  In the past seven days, new case counts rose 9% for the U.S. as a whole, and were  rising in all regions except the Pacific region.

You’d think we wouldn’t have a winter wave this year, with all the people who’ve been vaccinated or have recovered from infection.  But the example of Europe pretty clearly shows us that a winter wave is possible.

When I look at the numbers, I estimate (guess) that we’re in roughly the same situation as we were last winter, in terms of the ability of the virus to spread (the “R-effective”).

Compared to last winter, we have:

  • Higher levels of population immunity (best guess 71% now versus about 10% in October 2020).
  • A far more infectious variant of COVID-19 (best guess, R-nought of 5 for Delta, versus maybe 2.5 for last year’s native strain, meaning that without any preventive measures, each person infected with the Delta variant would have infected an average of five others.)
  • Much lower levels of COVID-19 hygiene (no restrictions on public gatherings, mask-wearing down from 95% in the middle of last year’s winter wave to 60% now).

When I run that through a crude formula, including my own estimate for the impact of peak COVID-19 hygiene, and an assumption that current COVID-19 hygiene is, on net, half as effective, I come up with:

  • R-effective last winter 1.035
  • R-effective this winter 1.075

I wouldn’t put a lot of faith in either number.  I’m just saying that the magnitudes of these effects are in the ballpark of cancelling each other out.  It’s entirely possible that we’ll have a winter wave despite having 59% of the population fully immunized, and a further large percentage with some immunity due to prior infection.

Source:  CDC COVID data tracker.

To recap:  Best guess, that level of vaccination, plus all the prior infections, is just about enough immunity to offset the greater infectiousness of Delta and the reduction in COVID-19 hygiene.

Finally, FWIW, in the popular press you’ll hear the rise in cases attributed to people spending more time indoors.  Near as I can tell, that’s more-or-less nonsense.  In the modern world, there’s maybe one hour’s difference in total indoor time, winter-versus-summer, for the average adult.  There’s far more evidence to suggest that flu season is a result of dry indoor air.  Low relative humidity reduces the body’s ability to clear way inhaled pathogens (“mucocilliary clearance”) and otherwise reduces the effectiveness of the immune system at fighting respiratory illnesses.  I lay out the evidence for various hypotheses regarding why we have a winter flu (and now, coronavirus) season in Post #894 about relative humidity and flu.

Post #1320: More crisis standards of care? Yawn.

Source:  Colorado Department of Health, accessed 11/11/2021

“Vaccine hesitancy” gets my vote for the most misleading phrase in the popular press of late.

I don’t perceive the least bit of hesitancy or ambiguity among those who refuse to get a COVID-19 vaccine.  How does “hell no, I won’t get vaccinated under any circumstances” get characterized as “hesitancy”?

I blame liberals.  Liberals remain the naïve children of the Enlightenment.  Despite strong evidence to the contrary, they cling to this crazy belief that people are, at heart, fundamentally good and reasonably rational.  That they will, by and large, make informed choices for the benefit of themselves and society at large.  Within that fictional world-view, those who turn down a free, more-or-less harmless shot that reduces odds of an early death and could end the pandemic for society at large — those people must merely have different values and beliefs.  Perhaps they’ve assessed the facts differently, and have some reasonable doubts or fears regarding the efficacy of the vaccine or likelihood of true side effects.  And so they hesitate to get vaccinated.

In this liberal fantasy world, the unvaccinated are just unenlightened, or have made alternative, rational choices. They hesitate to get vaccinated.  Perhaps they can be led to see the light with proper guidance and education.

But back in the real world, nope, they’re just stupid and stubborn.  No amount of appeal to reason can fix that.  So if you want to get the entire population vaccinated, you’re going to have to resort to some level of coercion.

It’s actually completely unfair to blame “liberals” for this wishy-washy misleading term for the unvaccinated.  The true story is that “vaccine hesitancy” has been a term-of-art in public health circles for decades.  The earliest Wikipedia page on it (under the current Wikipedia page on vaccine hesitancy) dates back to 2007.  A quick search using date ranges on Google shows that the phrase appears to have emerged in the 1990s (and/or any references to it prior to 1990 aren’t captured by a Google search).

The true story isn’t that “vaccine hesitancy” is some polite new buzz-phrase, cooked up by the liberal media.  It’s just a mis-application of a bit of standard public health jargon. 

That said, surely we need something that’s a more accurate description of the current circumstances.  “Hesitancy” just doesn’t cut it.

At the very least, it should be “vaccine refusal”.  Just to make it clear that these people have been given a choice, and they’ve turned it down flat.  And that they own the consequences of their decisions.

Colorado declares crisis standards of care for hospitals.  Nobody notices.

But why bother with the whole vaccine thing, at this point?  I mean, everything’s OK, right?  More-or-less?

Well, no.  That brings me to crisis standards of care.  That’s the formal, legal declaration  by the governor of a state that, owing to a shortage of hospital beds, physicians may triage patients/allocate hospital care based not on need, but on likelihood of survival.  It provides the legal cover for physicians to allow individuals to die for lack of hospital care, because there simply aren’t enough hospital beds (or ICU beds or respirators or whatever) to meet current demand.

The potential for governors to invoke crisis standards of care has been on the books for years.  It’s a reasonable and rational part of medicare emergency preparedness.  It went hand-in-hand with pre-established CDC rules for the substitution of sub-standard PPE for normal hospital PPE in the event of a shortage. Rules that (I believe) were never invoked prior to the COVID-19 pandemic.

When Alaska and Idaho made formal declarations of crisis standards of care, that made the news.  It was judged to be a fairly significant event, that a U.S. state had reached the point of letting people die for want of hospital beds.

Only in Alaska was the effect of the hospital bed shortage obvious enough to be clearly visible on a graph (Post #1269).  You can see that in the two-month-old graph below.  Cases spiking (red), admissions declining (yellow).  In Idaho, by contrast, they largely managed to slough the problem off onto hospitals in eastern Washington.

Source:  CDC COVID data tracker, accessed 9/19/2021.

But at least that made the headlines.

Now it’s reaching the point that when another Western state runs out of beds, people hardly notice.  And the case in point today is Colorado.  They declared something like crisis standards of care last week.  And if my wife hadn’t picked up on it, based on a single New York Times article, I surely wouldn’t have noticed.  Near as I can tell, that’s the sole reference in mainstream media.

Running out of hospital beds has become the new normal.

There’s an oddity, in that the wording of the Colorado executive order is different from others.  (I’d better provide a link to the actual executive order, because in the nut-o-verse this has been characterized as requiring hospitals to refuse admission to unvaccinated individuals.  Whereas the actual executive order says nothing of the sort.)

It says ” … Order authorizing the Colorado Department of Public Health and Environment (CDPHE) to order hospitals and freestanding emergency departments to transfer or cease the admission of patients to respond to the current disaster emergency due to coronavirus disease 2019 (COVID-19) in Colorado.”

And, to be clear, it not only empowers the Colorado DPHE to stop admissions at some hospitals, it requires other hospitals to take those admissions if DPHE so directs it.  In effect, it gives the Colorado DPHE the right limit admissions and to re-allocate hospital admissions across the entire Colorado hospital system.  Not just to deny admission to certain hospitals, but to require other hospitals to accept those admissions.

If you read further, you’ll see that this particular language is taken directly from Colorado state statute.  And, as an extra for experts, this is all to get around the requirements of EMTALA, the Federal law that prevents hospitals from “dumping” undesirable (that is, uninsured) patients.  Hence the “transfer or admission” phrasing of the executive order.  The oddity of phrasing doesn’t necessarily reflect a different view of how best to triage patients.  It’s an artifact of how Colorado state statutes were written, and in turn, how the Federal EMTALA law was written.

Just to be completely clear, the declaration doesn’t even mention COVID-19 or vaccination status.  (So the claims in the nut-o-verse that this requires hospitals to deny treatment to the unvaccinated are completely fictional).

In fact, the bulk of the document pertains to insurance issues.  (Only in America, right?)  It notes that hospitals cannot take a patient’s insurance status into account, and that patients will still be insured even if sent out-of-network on an emergency basis under the provisions of this law.

But at this point, I need to cease ragging on Republican governors.  Because, despite the surge in cases, and the fact that Colorado hospitals are full, the Democratic governor of Colorado won’t do anything more than issue this order allowing the state health department to transfer cases across hospitals.  Mask mandates appear to be local-option only.  Near as I have been able to tell, bars and indoor restaurant seating remain open with no restrictions.

With that, about half the people in Colorado report wearing masks now.  That’s unchanged since the first of September.  So a little thing like running out of hospital beds doesn’t seem to be enough to affect behavior there.

Source:  Carnegie-Mellon University COVIDcast.

In any case, unlike Alaska at its peak, there’s no indication of any outright denial of hospital care yet, in Colorado.

Source:  CDC COVID data tracker accessed 11/11/2021

 

Coda:  An un-funny anecdote about Medicare Durable Medical Equipment.

Or:  Why COVID-19 in Colorado is distinctly different from COVID-19 in Louisiana.

I used to be a self-employed consultant in the area of health economics.  One day I was tossed the following problem:

In the Medicare program, at that time, there was a more than five-fold difference across the states in spending for home oxygen.  Worse, there was no indication of any difference in need for home oxygen.  There was almost no difference in prevalence of the main disease for which these rentals would be authorized (Chronic Obstructive Pulmonary Disease (COPD), which used to be called emphysema.)

That sort of thing is a big red flag for Medicare.  When they see massive differences in service use or spending, and no differences in the underlying health of the population, they immediately investigate for waste, fraud, and abuse.  In this case, my client — a manufacturer of oxygen concentrators — was rightly worried that this was going to affect his business.

At first, the Government’s case looked pretty good.  The scatterplot of oxygen use against COPD prevalence showed only a weak association.  There really was a lot of spending variation that appeared unrelated to prevalence of the relevant illness.  In particular, prevalence of illness did nothing to explain the high spending outliers.  (That is, the dots net the top of the graph below.)

Next, that spending variation was large.  The high-spend states really did out-spend the low-spend states by a factor of five or more.

But somewhere along in this process, I managed to recall that the Denver Broncos used to play in Mile High Stadium.  And then it all fell into place.  Those states at the bottom of the list all have one thing in common:  They are Mountain region states.  There isn’t much oxygen there.

When I arranged the same set of states by mean elevation above sea level, I got a much more orderly plot.  Suddenly, those big outliers made sense.

Medicare was supplying a lot of oxygen in those states because Mother Nature wasn’t.  Here’s a quantitative estimate of the impact of elevation on the amount of oxygen available (the partial pressure of oxygen).

Source:  Highpeak.com

If you live in Aspen, CO, you’re missing about one-quarter of your oxygen, compared to life at sea level.  If your lungs are healthy, you’ll soon get used to it.  But if you’ve got COVID, and your blood oxygen saturation starts to fall, you’re going to be a lot worse off in Aspen than you would be at sea level.

And so, my guess is that for a given population of individuals severely ill with COVID-19, a higher fraction of them are going to require an inpatient level of care in Colorado, compared to (say) the U.S. Gulf Coast states.  That’s because a critical deciding factor is their 02 blood saturation levels.

The upshot is that COVID is different in Colorado.  Not because the disease is different, or the people are less healthy.  There’s just less oxygen there.

Post #1319: Use of the semi-attached figure: COVID-denial and climate-change-denial propaganda

There is an entire line of disinformation about global warming that works like this:

  1. Find any indicator of climate change that isn’t hitting a new high this year.  E.g., global average temperature.
  2. Point to that decline and and contrast it to the increase in atmospheric C02 (which reliably hits a new high every year).
  3. When that indicator begins hitting new highs again, find something else.
  4. Repeat.

For a while, it was all about “the hiatus” in global warming (Forbes, NOAA), until that turned out to be literally imaginary.  The news outlets that touted “the hiatus” simply don’t talk about global temperature now.  A common alternative target for this style of propaganda is summer arctic sea ice extent, where you will reliably see Fox News coverage in any year in which the sea ice extent increases.  (Despite a clear long-term downward trend, consistent with (duh) a much warmer Arctic).  Weirdly, for a while, when the Arctic summer ice was clearly falling, denialists focused on winter antarctic ocean ice cover (which, unlike the arctic summer ice cover, has no strong implications for the future of warming and is more-or-less unrelated to summer arctic ocean ice cover).  Focused on that, except in years when the winter antarctic ice cover is below the peak.  At which time, then turn to some other target of opportunity.

When done properly, everything said in such propaganda pieces is true.  Some years, global temperatures fall, compared to the prior record.  Every year, atmospheric C02 goes up. True facts.

But these pieces are, nevertheless, propaganda.  The are disinformation designed to persuade readers to believe something that isn’t true.  The disinformation works due to the careful cherry-picking of the data point shown and the use of the semi-attached figure of one year of C02 increase.  It’s left up to the reader to make the incorrect inference (that the theory behind global warming requires temperatures to increase in yearly lockstep with annual increases in C02, so this contrary fact disproves global warming).  Many willingly do so.

The details don’t matter here, it’s the basic technique that I’m trying to emphasize.  Any time series of data that shows temporary ups-and-downs (or seasonal changes, or cross-sectional differences) is a potential propaganda tool.  Cherry-pick the right data points, tack on the semi-attached figures, and the result is a steady stream of propaganda pieces that give the impression of something (e.g., that data are always contradicting the scientific consensus on global warming) without actually telling actionable lies about it.

And I guess it goes without saying that Fox News is the master of this tactic.  Near as I can tell, the only information about global warming that makes it onto Fox News is propaganda of that sort.  They’re silent on the vast majority of studies and information that reinforce the scientific consensus (the planet is warming rapidly, we’re causing it), and highlight any piece of information that can be made to seem as if it contradicts that consensus.

And, once you’ve seen it enough times in one context (global warming), it jumps off the page when you see it in other contexts. 

Again, to review the method:

  1. Find any relevant data series that goes up and down.
  2. Cherry-pick a time where the data are trending the way you like.
  3. Tack on the semi-attached figure.
  4. Leave it to the reader to make the incorrect causal link between the two.

With that as introduction, look at the Fox headline above and tell me what you’re supposed to believe, based on that. 

Did you come up with “well, that proves that masks don’t work, doesn’t it”?  From which you have to move on to “mask mandates don’t work”.  And so, obviously, based on that, we don’t need mask mandates.  Why would any idiot think that mask mandates had any purpose?

Let me now illustrate what you should have gotten out of that headline:

Data source for this and other graphs of new case counts:  Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 11/10/2021, from https://github.com/nytimes/covid-19-data.”  The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html.

As an Extras for Experts, note the critical timing here.  As little as two weeks ago, they couldn’t have written that headline.  Four weeks ago, they’d have to have written it the other way around.  Give it another four weeks, and I’m betting they wouldn’t be able to write that.  But for this brief time window, they can write that headline, and it’s a fact.  Four weeks from now, this will have been forgotten.  And they’ll have found another target of opportunity for their next piece of propaganda.

But why?

A lot of rational people, particularly scientists, have a hard time understanding why anyone would bother to produce (or consume) this continual stream of propaganda.

To me, having seen this time and again, and having run up against it with my ultra-conservative older brother, the answer is simple.  It’s all about faith and belief, and has nothing to do with science.

In a nutshell, for most people, when facts conflict with faith, you must deny the facts.  It really doesn’t go any deeper than that.  It’s not intrinsically different from the Catholic Church’s persecution of Galileo for the heresy of claiming that the earth moved around the sun.

In my brother’s case, anything that would require deviating from orthodox political conservatism must be dismissed.  It’s not that he is anti-science, per se.  It’s that he works backwards from the policy implications.  If the science implies policies that conflict with his political faith, then the science must be denied. 

It really is that simple.

There are two types of people.

For some, facts and reason determine what their course of action should be.  And if rationality conflicts with faith, then faith has to change in light of the facts.  For others, faith determines their course of action.  And if facts and reason get in the way of that, those facts and that reasoning must be denied, so that faith remains unchallenged.

And so, broadly speaking, that “faith based community” creates a tremendous demand for hearing only what they want to hear.  Free markets will supply anything that is demanded.  The result is entire industries devoted to satisfying that demand for “facts” that match faith.

Back in the objective world, the fact is, the earth is warming.  That’s primarily due to a buildup of an incredibly stable gas (C02) in the atmosphere, faster than the ecosphere can absorb it.  The C02 comes from burning fossil fuels at a rapid rate.  The overwhelming consensus of informed opinion is that this process isn’t going to end well for civilization.

Fact is, masks work to reduce spread of COVID-19.  (And other airborne illnesses, for that matter.)  A high-quality mask (e.g., N95 respirator) works better than a low-quality mask.  But, all other things being equal, the higher the fraction of the population wearing masks in situations with non-negligible risk of disease spread (indoor public spaces), the lower the spread of COVID-19 will be.

If your faith bars you from considering anything but individual action, motivated by individual incentives, then these facts are inconvenient.  That’s because the most effective ways to stop C02 emissions, and to stop COVID-19 spread, require coordinated action that cannot be achieved by person-level free-market incentives. (Or, at least, none that have ever been proposed as feasible.)

As with the Catholic church and a heliocentric solar system, a lot of people maintain some sort of faith that requires that they must declare such actions to be heresy.   Those people need someone to feed them the disinformation that will allow their faith to remain unchallenged.  And one way or the other, the dollars behind that demand support the sort of mainstream disinformation highlighted at the top of this post.

It’s not that people are too dumb to know the difference.  It’s that they don’t care, and they actively crave the disinformation. 

And as a result, we have a fully-developed denial industry.  Literally the same entities that helped deny that smoking causes cancer were hired to help deny that combustion of fossil fuel causes global warming.

Given that, it’s really no surprise that the same techniques keep showing up.  In this case, the technique is to take any time-varying data, and pop up with a bit of disinformation during any brief period when the numbers are running your way.  Remain silent when the numbers aren’t in your favor.  And keep changing targets, because nobody in your target audience will even hold you to task for anything you’ve ever said.

I’d like to say that people will eventually figure that out.  But they won’t.  To the contrary, it’s not that the propaganda causes their non-factual beliefs.  It’s that they actively seek the disinformation that agrees with their faith.  They take comfort in it.  And the rest of us just have to live with the results, as best we can.

Post #1318: IED

 

I enjoy crossword puzzles.  And I’m not ashamed to admit it.

My puzzle habit was formed during ten years of daily commuting from the suburbs to downtown Washington DC, via DC’s Metro system.  Now, after more than three decades of puzzle-solving, I have an appreciation for the subtle science and exact art of crossword-puzzle making.

Filling in a hard crossword puzzles requires an odd assortment of skills.  It becomes roughly equal parts of:

  • Knowing the structure of language (e.g., plurals usually end in “s”).
  • Straight-up trivia (e.g., Pierre is the capital of South Dakota).
  • Current pop culture (e.g., Grammy winners).
  • Older pop culture.
  • A good sense for puns, alternative word meanings, and the like.

Much of it has a unique crossword-puzzle slant, owing to a chronic need for vowels.  For example, ONO (Yoko), ARLO (Guthrie), OCALA (Florida) all appear in crosswords far out of proportion to their importance in the real world.  As do the many, many vowel-rich four-letter rivers of Europe (e.g., ODER, YSER, URAL, ARAL, AARE, …) .

The popular-culture aspects of crossword puzzles typically don’t age well.  It’s hard to pick up a 20-year-old book of difficult crossword puzzles and fill them in.  The world has moved on.  Pop-culture names and terms familiar to every well-read reader in 2001 are seldom on the top of the tongue two decades later.

That said, they are never truly current, either.  It takes a while for any new pop-culture phenomenon or phrase to work its way into the day’s crossword puzzles.  So what you really get in crosswords is pop culture with a lag.

Which brings me to IED.  That was in a puzzle I worked yesterday, with the clue “hazard to troops”.  It was, in that sense, a perfect crossword puzzle word.  Lots of vowels, and a term that every U.S. resident would have absorbed over the past couple of decades.

But IEDs haven’t been in the news of late.  Which is a good thing.  And I can only hope that this clue and answer will be completely mystifying to some puzzle-solver a couple of decades from now.

My point being that sometimes the news ought to be about what hasn’t happened recently.  We ought to see a great big headline stating that “No American troops died in Afghanistan over the past two months”.  Or that we failed to spend $20B propping up a corrupt and unpopular government over that same time span.

But that sort of obvious good news just isn’t what the popular press is all about.  Too many other things that are better click-bait.  U.S. casualties that didn’t occur are the sort of thing that will only sink into our collective consciousness a decade or two from now.  If then.

Meanwhile, I’ll continue to enjoy the absence of the IED from our popular press.  Even if that word is still in crossword puzzles, for the time being.

Post #1317: COVID-19 Vaccine side effects in children ages 5-11

 

I saw an article yesterday listing out all of the extremely likely side-effects you should expect if you get the COVID-19 vaccine for your 5-to-11-year-old child.  It made it look like a terrible gamble.  If I’d seen that article, and nothing else, I’m not sure I’d have gotten my kid vaccinated.

The only problem is, they showed the numbers for vaccinated children ages 5-11 in complete isolation.  They didn’t show how young children fared, relative to teens.  And they didn’t compare to the placebo group.  Or to typical self-reported side-effect rates for (e.g.) flu vaccine.

If your job had been to scare parents away from vaccinating their kids, you could not have done a better job of it than by cherry-picking that exact bit of information.  And showing zero context.

Just to be clear about how these numbers arise, if somebody had asked me if I’d felt fatigue or muscle pain in the past week, the answer would have been “yes”.  As it would, pretty much any week of the year.  If I’d been part of the study, by responding honestly to that question, I’d have been contributing to the reported potential side-effect rate of the vaccine, as shown in the popular press.

This isn’t to dismiss the side effects.  They sometimes occur.  It’s more that you need some context to make sense of the reported side-effect rates.

I thought I’d get the actual data, and show how the side-effect rates for children look in proper context. 

Here’s the original slide of results, as presented to the U.S. CDC, which I will now simplify to highlight the main findings:

Source:  Presentation to CDC, “BNT162b2 (COVID-19 Vaccine, mRNA) Vaccine in Individuals 5 to <12 Years of Age”, November 2nd 2021, Alejandra Gurtman, MD, Vice President. Vaccine Clinical Research and Development, Pfizer Inc

Point 1: The side effects … were generally milder and less frequent in 5- to 11- year olds than they were in adolescents.  Putting aside that most of the side effects disappear within one to three days, little kids actually had lower rates of side effects than teens and adults.  That’s what these pairs of bars below are showing.  So if you weren’t too worried about vaccinating your  sixteen-year-old, you should be even less worried about vaccinating your six-year-old.

 

Point 2:  Side effect rates were only modestly higher for children actually getting the vaccine, compared to placebo.   These contrasts below are between vaccine and control groups.  If I had to condense it, it would say that compared to children who got the placebo, the vaccinated children were:

  • 10% more likely to report fatigue or headache
  • 5% more likely to report fever or chills
  • 3% more likely to report muscle or joint pain.

Point 3:  And again, to be clear “If they arise, side effects generally are gone within one to three days.”

There’s another potential point of comparison, which is the rate of the same reported side effects for standard flu vaccine.  One might reasonably ask how frequently these same side-effects get reported for flu shots.  That’s a vaccine that most of use get annually without a thought in the world of having a significant adverse reaction.  (N.B., a flu shot is recommended in the U.S. for everyone over the age of six months, so it’s a vaccine that is definitely recommended for this 6-11 age group.)

It’s tough to get any hard numbers on that, probably because the flu vaccine has been considered safe for so long.  Here are a few bits and pieces.

Of the side effects tracked above for COVID-19 vaccine, the one article I found listed these rates of side-effects in adults, for flu vaccine:

  • fever (perceived) 15.2%.
  • fatigue (17%)
  • muscle pain (17.7%)

Here are the side effect rates for Fluzone in adults:

  • Muscle pain 18.9%
  • Headache 13.1%
  • Shivering 4.8%
  • Fever 0.9%

I’m not going to beat this point to death, because my point is simple.

Side effects are not some new thing that just happened with the COVID-19 vaccine for young children.  The common side-effects of the COVID-19 vaccine in children are the same as the common side effects, for the same vaccine, in teens and adults.  And the same as the common side effects of the flu vaccine. 

The rate of side-effects from the COVID-19 vaccine, in young children, is less than in teens and young adults.  And, very roughly speaking, the rate of side effects of COVID-19 vaccination in children is not grossly different from the rate of the same reported side effects from flu vaccination in adults.

If you didn’t hesitate to get your teenage kids vaccinated, there’s no new reason to hesitate about younger children.  And if you get the flu vaccine every year for you and yours, you have arguably taken on as much risk of adverse events with that as you will with the COVID-19 vaccine.

It is all-but-impossible to get those conclusions out of standard mainstream news reporting about this.  It’s just too dog-bites-man, I guess.  It doesn’t fit into the modern style of fear-oriented journalism.  Instead, what will catch your eye is some chart showing an apparently high rate of side effects.  With no way for you to know that this is completely normal, no different from the experience of other age groups, and not hugely different from the side-effect rates for flu vaccine.

Post #1316: The universal state budget surplus of FY 2021 and the economic boom of FY 2022.

 

In Post #G21-058, I stumbled across an interesting finding.  More-or-less every U.S. state had a large (often record) budget surplus for FY 2021.  As far as I can tell, this has gotten exactly zero notice in the popular press.

Reading a few reports of these surpluses, it seems like various sources of state tax receipts started to pick up around April 2021 and just haven’t quit since.  And nobody is quite exactly sure why, although the obvious suspect is all the spending power that the Federal government injected into the economy over the past 18 months.

Now here’s the weird thing, and the main conclusion that I’ve drawn so far:  We seem to be in a genuine economic boom.  I keep looking for signs that revenue growth will be petering out, now that we’re reaching the end of the pandemic.  But there’s no sign of that in sight.

At some level, it shouldn’t be a surprise.  The Federal government has just gotten through two years of the largest peacetime economic stimulus in U.S. history.  A good chunk of that was simply saved, presumably to be spent later.

And now, with all that free money burning holes in many pockets, the result is just standard Keynesian economics.  There’s a whole lot of new economic activity, with a side-order of inflation.

But you’ll have to judge for yourself.  As I say, this started out as a study of state budgets, and rapidly turned into an analysis of just how rapidly the U.S. economy seems to be heating up.

U.S. Treasury Revenues are clearly up.

Let me start with the most stable source of timely national information on economic activity that I know of:  The Monthly U.S. Treasury Statement.  If somebody’s making money from it, it’s a good bet that Uncle Sam is taxing it.  So, putting aside the big lump of revenue that arrives at tax time, Federal receipts provide a pretty good estimate of the pace of economic activity.

Source:  My plot, of data taken directly from the U.S. Monthly Treasury Statement.

No matter which perspective you take — two decades, or five years — we have clearly entered a period of rapid growth in U.S. Treasury receipts.

Flash GDP estimates are running to double-digit growth.

These get a little murkier, as they are no longer hard data, but are estimates from somebody’s economic model, fed by current data.  For this, I’m relying on the Atlanta Federal Reserve’s GDPNow estimate.

“The GDPNow model estimate for real GDP growth (seasonally adjusted annual rate) in the fourth quarter of 2021 is 8.5 percent ”

They also note that their model is well above the “blue chip consensus forecast” of real GDP.

Virginia’s general fund revenue numbers are running 10 to 15 percent above the same period last year.

Source:  Virginia monthly revenue letter, September 2021.

And, from what I can tell by casually checking a few other states, this is not unusual.  Seems like a lot of states have seen broadly-based revenue growth continuing well into FY 2022.

Whether or not state tax receipts will continue to grow is the question of the moment.

The fact that started me on this analysis — the large number of states with record FY 2021 budget surplus — has not gone unnoticed in the academic press.  Of the articles that have focused on this, the Pew Charitable Trust managed to hit the nail on the head.

Awash in Cash, State Lawmakers Ask How Long the Boom Will Last, dated July 26, 2021, by Sophie Quinton. 

By now, more-or-less everyone involved in state finance has noted the jump in revenues in FY 2021.

But here are two conclusions, based on that article, that I find remarkable.

  1. Nobody has any idea how long these higher revenues will last.
  2. A far as I can tell, other than direct federal payments to states for COVID relief, nobody has any firm idea of what caused the jump in revenues, including tax revenues.

Here’s a quote that pretty much sums it up:

“The growth trajectory—it’s higher than we expected,” said Adams of Idaho’s Division of Financial Management. “I don’t anticipate that it will continue at this pace. I don’t think anyone does, frankly.”

Kate Watkins, who leads the team that prepares revenue forecasts for the Colorado legislature, said she expects Colorado’s revenue growth to flatten out.

“In many cases,” she said, “we’re still waiting on data to validate what the story is moving forward, whether or not this is really kind of a blip or if it really is a sustainable growth trajectory.”

As I read it, the reason there’s no “smoking gun” is that revenue growth is quite broad-based.  Not only is income tax withholding up, so is sales tax, so are corporate tax payments, and so on.

Basically, we seem to be in the middle of an economic boom.  One that doesn’t seem to have gotten much attention.  But one for which the Federal and State tax data, and the flash GDP estimates, suggest is pretty substantial.

Amidst all the negative press regarding the President, I sure haven’t heard much about the U.S. being in the middle of rapid GDP growth.  The only sign of that has been the steadily falling unemployment rate.

But, as far as I can tell, that appears to be true.  I started out assuming that we were in the middle of some temporary bubble in state finances caused by direct Federal pandemic relief.  But now, that appears to be wrong.  For whatever reason — making up for lost time in the pandemic, spending all that free pandemic money, or who knows why — we’re suddenly in the middle of economic good times.

 

Post #1315: COVID-19 trend to 11/5/2021: Probable start of the U.S. winter wave.

 

Let me get right to it.  I think we haven’t had winter wave of COVID-19 yet because we haven’t had much winter yet.  Certainly not when you compare this year to the same months last year.

And now that’s changing.  Both the weather, and the trend in U.S. cases.  Let me do the U.S. COVID-19 case counts first, then talk about the weather. Continue reading Post #1315: COVID-19 trend to 11/5/2021: Probable start of the U.S. winter wave.