(Do not) Beware the Ides of March

March 15, 2021 By Alan Salzberg

A month ago I wrote that as far as COVID goes, it's all good news (https://salthillstatistics.com/posts/98).  The only thing that has changed is that it's gotten better.

1) We are vaccinating even more people.  At this time in February, we were giving about 1.6 million shots a day.  Now we're at 2.4 million, and more than 1 in 5 people in the US have received a shot.  On Saturday, we had our first 4-million-shot day.  So we have the capacity to give a lot more shots right now.  Very, very quickly, probably by mid-April, we are going to shift from not having enough shots to having too many, and we are going to shift from urging patience in getting the vaccine to simply urging people to get the vaccine.  Start now--urge your eligible friends and family to find a shot, and help them do so.  I've noticed that even now there is plenty of immediate availability for the over 65, and relatively near-term appointments for other eligible people.  It's your civic duty to get vaccinated.

2) Cases are still dropping dramatically. There was brief concern in the popular press that cases had leveled off.  This is not true.  While there was a very brief plateau, cases continue to decline, even with reduced restrictions.  Part of the reason why cases appear to be dropping less now is we are forgetting about the nature of exponential decay.  In constant exponential decay, the 125,000 drop from 250,000 (the peak--using rolling daily averages by week) to 125,000 (half that) is the same as the 62,500 drop from 125,000 to 62,500.  The drop from 250,000 to 125,000 took place between January 11 to February 5 (using weekly rolling averages).  That is 25 days.  The drop from 125,000 to 62,500 took place between February 5 and February 20 (15 days). Though Sunday's are always lower due to data collection and reporting differences, yesterday's cases were 38,034, a number we haven't seen since late September.  In case you think this is due to less testing, note that the positivity rate is dropping too, from over 10% in early January to just over 4% now.

3) Deaths are dropping also.  On January 16, the average daily death rate (again, 7-day rolling average) hit a peak of 3,322.  Now, we are at an average of 1,700, and yesterday's toll (again, low-reporting Sunday) was 572.  We still have a long way to go, but keep in mind that deaths follow cases by several weeks, so we can expect a continued decline for some time even if case declines were not continuing.  How low is "normal"?  Well, one thing to consider is that the flu causes an average of around 100 deaths a day in a typical year, and, since those deaths are concentrated in the winter, the seasonal average can be 200 in a typical year and double that in a bad year.  So when are we going to be at a "typical" flu season rate of deaths?  I don't have a model for that, but I expect we will drop below 1,000 permanently at the end of March/ beginning of April, as declines continue due to vaccination, and the arrival of spring.

4) Variants are still not an issue.  Variants are not a cause for worry.  They should motivate you to get vaccinated and it would be nice if they motivated states to continue restrictions for a little longer, because some are more contagious and more deadly.  But the good news is that so far, both prior infections and vaccines are both very effective at immunizing people against the variants.  There have been virtually no cases of reinfection here and all the current vaccines appear to be quite effective against all tested and otherwise worrying variants.  Eventually, this may change, but it will hopefully when levels of virus are extraordinarily low and when we can get a booster that will cover them.

So that's it for this Ides of March!