Post #960: Current U.S. COVID trend. Do you think anyone will notice?

Source:  U.S. (dark blue line) and six regions, new cases / 100,000 / day, seven day moving average.  Data ending 1/17/2021.  Underlying data from the NY Times Github COIVD data repository.

The U.S. seven-day moving average for new COVID cases per 100,000 is back down to where it was a month-and-a-half ago.  All six of broadly-defined regions show the same downward slope.

Note that the first two dips in the line above are due to the holidays.  But this last decline is a new thing entirely.  Given that we never had a post-holiday surge for either Thanksgiving or Christmas/New Year’s, it’s hard to believe that the current decline is somehow tied to the holidays.  A large negative post-holiday un-surge doesn’t seem plausible.  So it is what it is.

The current decline in daily new cases is so nearly universal that not only is every region trending down, just a handful of states don’t have a level-to-downward trend:  Maine, Virginia and South Carolina. And the only high-population state with a level trend is Texas.  That’s more-or-less apparent from the graph below, where almost all the lines in the right edge slope downward.

The big exception to that is Virginia.  Not sure what we’re doing wrong, or whether it’s just our turn in the barrel, but new case counts are going through the roof.  Yesterday we broke a record for daily new cases, by a factor of about 50%.  After nine months of lying low, suddenly we’re the nation’s hotspot for new case growth?  And not so much as a note of explanation of the Commonwealth’s COVID dashboard.

If that doesn’t somehow resolve with today’s update, I’ll take a deeper dive into the Virginia data to see if there’s anything obvious.

Post #959: If you’ve been vaccinated, do you still need to wear a mask?

If you’ve been vaccinated, do you still need to wear a mask?  I got this question from a reader a couple of days ago, and gave a partially-correct answer via email.  Here, I’m going to post what I believe to be the fully-correct answer.

Briefly:  Yes.  The CDC says you still need to wear a mask, social distance, and so on.  As long as the virus remains widely circulating in your community.  That’s not just the CDC being its usual fussy self.  There are sound reasons for that, involving non-trivial risk of catching and spreading disease.

In this post, I’ll try to explain that.  But, at the minimum, that “90% efficacy” you read about doesn’t include asymptomatic or mildly symptomatic infections.   They only counted severe, symptomatic cases.  The reduction in total coronavirus infections is unknown, and you can still spread the disease if you get a mild case of it.  In addition, that 90% reduction was based on all the participants maintaining their existing COVID hygiene.  If you get vaccinated, then start hanging around maskless in bars, you plausibly have greater odds of getting infected after the vaccine than you did before it.

But surely that has to prompt a few follow-on questions.  I’m not going to provide detail on these, in this post.  But I’ll give you my answers.

Well then, when can we  ditch the @#$@# masks?  I think the answer to that is “when the Governor says we can”.  We’ll get there when we get there.  I don’t think it goes any deeper that that.  It really shouldn’t be an individual-level decision, but that’s a topic for another post.

Continue reading Post #959: If you’ve been vaccinated, do you still need to wear a mask?

Post #958: About a third of U.S. residents have already had COVID

In the spirit of “it can’t get but so much worse”, here’s an estimate of the fraction of the U.S. population that has already been infected with COVID-19.

As of today (1/17/2021), about 24 million cases have been reported.  It is well- established that there are several actual infections for every one that is reported.  That’s due to asymptomatic and mildly symptomatic individuals who don’t bother to go to a doctor about their symptoms, and so don’t get tested.

Here, I am assuming a conservative five-to-one ratio between total infections and formally-diagnosed infections.  That five-to-one splits the difference between two different estimates from CDC staff, as documented in Post #940.  A CDC estimate based on likelihood of getting tested would put that ratio at closer to 8-to-1.  A CDC estimate based on presence of antibodies in a recent (non-random) sample of blood draws would put it closer to 4-to-1.  I chose 5-to-1 as a seemingly reasonable estimate, as documented in that post.

Post #957: COVID trends, but I’m not really sure what to title this one.

Trees don’t grow to the sky?

What goes up must come down?

A trend is a trend until it ceases to be a trend?

Hope springs eternal?

Maybe the right adage is “If a tree falls in the forest,and no one hears it, does it still make a sound?”  Because I don’t see any mainstream reporting on what’s happening in the sequence of graphs below.  So I figured I’d post this, harking back to my guess that we were due to peak, in Post #930.

In any case, as the news ping-pongs from local crisis to local crisis, here’s a view how the U.S. situation developed over the past week.  All graphs except the last were taken from prior posts, so the stuff in red was, in fact, done on the fly, not after-the-fact. Continue reading Post #957: COVID trends, but I’m not really sure what to title this one.

Post #956: One more day, six more states: Update to British COVID variant post #952

Yesterday, Post #952, 14 states had at least one case.

Today, 22 states have at least one case.  Not clear that it’s actually spreading that fast, as states may now be finding it because they are looking for it.  That said, it’s not going to be long before it’s been found more-or-less everywhere.

References for today’s additions:

Illinois

Louisiana

Michigan.

Oregon

Utah

Wyoming

Post #955: More people saying “get a better mask”.

This might be a case of finding what I’m looking for.   But I seem to be seeing more mentions in minstream media regarding the need to wear N95 masks (respirators).  In particular, I see more people pointing to citizen use of N95s as a rational response to the new, more contagious British variant of COVID-19.

In mid-2020, a policy of reserving N95s for health care workers made sense.  But now that domestic production has increased several-fold, and even a hard-hit state like Minnesota has a half-year supply on hand for hospital use (see below), and we’re facing a faster-spreading COVID variant, it’s more than time to rethink that, and start getting N95s into the hands of the public.

Continue reading Post #955: More people saying “get a better mask”.

Post #954: A trend is a trend until it ceases to be a trend

This is a brief update on trend in new COVID-19 cases, using data reported through 1/15/2021.

By and large, previously-identified short term trends are continuing.  Nationally, we have a little flattening of the curve.  Most states show a downward trend in new cases/ 100,000/ day.  California continues to move sideways.  The only new development is that cases counts have stopped rising in New York.

What I find most notable is how tightly correlated the movements are across states.  It’s not as if a few states are on a downward trend.  Very nearly all of them are, starting very nearly around the same time.

With the holidays now ended more than two weeks ago, it’s hard to think that this is somehow a data reporting or behavioral artifact of that period.  But it’s also hard even to guess what common factor would lead to that coordination.

For example, if you look back to the end of the Thanksgiving data artifact on the first graph below (the first dip in the blue line), the regions were going every-which-way.  Some up, some down, some sideways.  But now?  Now they’re almost all in synch.

I’ve already stated that I think the odds are that we are over the hump (Post #930).  Or so I thought at the new year.  But a trend is a trend until it ceases to be a trend.  We’ll only know the peak when we see it in our rear-view mirror.

Graphs follow.

Nationally, a little flattening of the curve.  Six regions and US (blue line), then all states since 10/1/2020, then all states since start of pandemic.

Midwest and mountain states continue the declines that started in early 2021.    There was a sharp peak in Arizona and Utah, followed by what looks like an equally sharp retreat from that peak.  North Dakota still has the lowest new case rate in the mainland U.S.

The South-Central states appear to be starting in on a coordinated decline in new cases.  Possibly even the South Atlantic (or much of it) shows a similar path.  It’s hard to tell, as the upward trends are fairly modest to begin with.

California continues to go sideways.  New York has at least a temporary plateau in new cases rates.

Post #953: I emailed my Senators and my Congressman

I’m not going to give directions in how to email your own Senator or Member of Congress.  All of them have email forms that you can use, and it’s easy enough to find them via Google.

Instead, this is the content of what I have sent to Senators Warner and Kaine,  and Congressman Connolly, regarding using Medicare to supply N95 masks (respirators) to the elderly.

As I understand it, the trick is to make it clear what you are asking for, up front.  You are trying to get the attention of the low-paid staff member who will actually be reading this.  Even so, unless they get a flood of letters asking for the same thing, this will likely do little more than add +1 to their count of messages about coronavirus.

That said, you have to try.  So here goes.

Dear Senator Warner:

In this email, I lay out the reasons why the Federal government should supply N95 masks (respirators) to the elderly via the Medicare program.  And I ask that you introduce legislation to make this happen.

You already know that N95 masks (respirators) provide substantially better protection against COVID-19 than cloth masks or surgical/procedure masks.  And that the elderly are at much higher risk of adverse outcomes.

But U.S. citizens cannot obtain N95 masks.  Those were pulled from all normal retail channels during the acute shortage of N95 masks in early 2020.  And now, even though there is no longer a shortage of N95 masks, those are still being withheld from retail sale.

The Federal government already has the infrastructure to get those effective N95 masks into the hands of the elderly.  Both Medicare Part B and Part D routinely deliver monthly supplies (DME and drugs) by mail to tens millions of Medicare beneficiaries.  It would be no strain on the system to deliver (say) three N95 masks per month to every enrollee who wanted them.

To do this, those masks would have to be a covered benefit under Medicare.  That would likely require legislation.  And doing so in a timely fashion would require allowing Medicare to skip parts of the Administrative Procedures Act. Which again would require legislation.

Hence this letter, because this can’t be done without Congressional action.

In the current climate, mask use is one of the things that strongly and visibly separates Democrats and Republicans.  It’s one thing to say “wear a mask”.  It’s a different thing entirely to go to the effort and expense to provide the most vulnerable segment of the U.S. population with an effective mask.  And that means an N95 respirator, via Medicare.

Finally, I note that Germany has already done this.  On 12/15/2020, they instituted a program to provide high-filtration respirators to every person 60 or older.  If the Germans have figured this out, surely we can too.

Sincerely,

 

Post #952: Month-and-a-half left to prepare for the B.1.1.7 British variant of COVID.

Above:  States where the new British B.1.1.7 COVID-19 variant has been detected as of 1/15/2021

Source:   Google search.

There is now a new, more easily spread variant of the COVID virus, first detected in the area around London.  Over the course of about three months, it has out-competed other variants to be come the dominant strain in and around London.  The higher transmission rate for this new virus resulted in rapid increase in cases and sent the Brits into another strict lockdown.  That lockdown now appears to be working, in terms of reducing new cases per day.

That COVID 19 variant is loose in the U.S. and elsewhere in the world.  Near as I can tell, there is no hard information available on how prevalent it is.  All we can do at this point is count the number of states that have announced that it is present in their populations.

The best guess is that this will follow the same course as it did in Britain, and so this will become the dominant strain of COVID-19 in the U.S. sometime this spring.  You can see news reporting to that effect in this articleOr this article.  But they’re making their guesses the same way I am — based on what occurred in Britain.

The consensus is that this new, more infectious strain will become the dominant U.S. strain by March.  So, we have maybe another month-and-a-half or so to get ready for that.

What this means is that if we merely maintain our current COVID-19 hygiene, we’ll see more daily infections.  All other things being equal.  So unless we are well onto the downslope toward herd immunity, we just have to anticipate more stringent measures against COVID-19 spread, a couple of months from now.

As an aside, the more we do other than shutting down commercial establishments, the less we will have to shut down commercial establishments to keep this new variant in check.  Which is why the whole mask-resistance movement makes no sense, unless your point is to inflict as much damage on the U.S. as possible.  I’d rather buy and wear a good mask than (e.g.) have all the movie theaters in the U.S. go bankrupt.  People who rail against masks and against commercial restrictions are simply confused.

The U.S. doesn’t do much to try to find these variants.  The US CDC sequences just 750 virus samples per week, and has now contracted with private labs to sequence another 1750 per week (per this CDC web page).  The CDC is also helping states pay for genetic sequencing of virus samples, but it’s not clear how many samples are sequenced by the states themselves.  That said, overall, the overall impression is that only the tiniest fraction of samples are tested for genetic variants, and so it’s a good guess that the new British (B.1.1.7) is more widespread than we currently know.

The only information we have about prevalence of the new strain is a count of the number of states where this had been identified, at least once, by genetic sequencing.  Which is, per the prior paragraph, almost certainly an undercount of even that crude measure of prevalence.

This new variant was first detected in Colorado in the U.S. (documented in Post #932, 12/20/2020).  Last time I looked (Post #943, 1/9/2020), it was present in eight states.

So it’s time to scan the news and update the count.  After searching Google News for about half an hour, I think I have all the states where it has positively been identified.  So, at present, we’re up to 14 states.  As shown at the top of this posting.

Post #951: Brief trends update

Source data for all graphs:  NY Times Github COVID data repository.  Graphs show data reported through 1/13/2021.

First, a little flattening of the curve?

Second, ND is now sole owner of second place.  Only Hawaii has a lower rate of new cases / 100,000 / day.  As goes ND, so goes the nation.  We hope. Continue reading Post #951: Brief trends update