For the fifth day running, the U.S. shows just over 9 new COVID-19 cases per 100K population per day. Over the past 7 days, the new case rate fell just 7%. My guess is, this is as good as it gets, for now.
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 3/22/2022, from https://github.com/nytimes/covid-19-data.” The NY Times U.S. tracking page 3may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html
Looking forward, we have a few countervailing forces at work now.
On the one hand, the continued spread of son-of-Omicron (BA.2) and continued decline in all sorts of COVID-19 hygiene suggest rates will rise from here.
Source: CDC COVID data tracker
Source: Carnegie-Mellon COVIDcast. Comment in red is mine.
On the other hand, I think the weather is in our favor. Everyone expect a strong seasonality to COVID-19 in the U.S., both because it has shown winter peaks two years running, and because most other coronaviruses also peak in the wintertime.
I’ve been showing a graph of this-year-versus-last year, starting with the start of the U.S. pandemic. Now let me shift that to calendar years, so you can see what I’m talking about.
It’s tough to see the pure seasonality of it because we have not reached a steady-state. Variants kept changing. Each successful new COVID-19 variant generated its own wave, overlaying any seasonal pattern that might exist. The level of population immunity keeps changing, both from vaccination and prior infection.
But let me try to abstract from all that by doing the lowest-of-the-low data analysis: fitting a polynomial trend. In this case, since my point is to try to boil this down to simple seasonality over the year, I’m going to fit a quadratic. That’s just enough to give me one peak and one trough, if that’s what the data suggest.
Here are 2020 and 2021, with a quadratic trend fitted, trying to boil the data down to a simple seasonal pattern. On this log-scale chart, you get a remarkbly similar trend line, despite major differences in the actual progress of case counts over the year. In 2022, for example, we had both the Alpha and Delta waves, and the start of the Omicron wave.
And now here’s 2022 actual data through the most current day, plotted against those two “seasonality” quadratic curves from the prior years:
My sole point here is that the apparent seasonality of COVID-19 in the U.S. should be working to depress new case counts now.
Or, more simply, for the past two years running, the lowest case counts occurred mid-June, the highest ones occurred mid-January. If that keeps up, then the forces of seasonality are in our favor.
