Post #651: Still trending.

Posted on April 17, 2020


Last 28 days of daily increase in coronavirus cases in Virginia (blue), Fairfax County (orange), and Virginia less Fairfax (yellow).

Shows seven-day moving average (dotted line) and linear trend (solid line).

Left = data through 4/16/2020, right = linear extrapolation to 4/24/2020.

Source:  Analysis of Johns Hopkins coronavirus dashboard case counts through 4/16/2020.

See prior post for explanation.  This is a graph of daily case growth, in percentage terms, along with a (linear regression) estimate of the trend. They continue to trend down.

My wife, who is somewhat math-challenged, says I need to add the following explanation of what those downward-sloping lines mean.  Briefly:

The trend looks good.

If this continues, this signals the end of (the first phase) of the pandemic, here in Virginia.

Separately, Fairfax daily percent case growth finally seems to have fallen in line with the Commonwealth, or nearly.  (Or, the smaller case count just added that much more random noise to the Fairfax numbers when I was looking at it before.  With this many observations, it’s hard to say which).

Note that, because that was percentage growth, graphed at the top of this page, we are still at a point where the daily count of new cases (as opposed to percent increase) is still at the peak.  By eye, if that count leveled off, that only occurred a week ago, at best.  Or, about ten days after the CDC told everybody to wear a mask in public.  (Where 10 days is my best guess as to how long it takes a change public behavior to work itself into the numbers.)

So the burden on local hospitals, for the sheer number of new cases, remains high.

If the trend holds, then that will be the “inflection point”, the point where the daily increase in the count of new cases begins to fall.  And if this trend continues as shown, new case growth would more-or-less cease, outside of Fairfax County, about a week from now.  And then shortly thereafter, in Fairfax County.

At that time, Virginia would have a cumulative case count of something over 8,000 cases, which is well within what the Virginia hospital system can manage.  (Best guess, it would take 30,000 statewide to exhaust the supply of hospital beds and ventilators.)  Though I suspect that our local hospitals may be stressed, due to the high concentration of cases in Fairfax County.  (We account for about a quarter of all cases in Virginia).

This is a pure exercise in curve-fitting.  There’s no underlying understanding of why we’re seeing this particular pattern.  And I have yet to find any source that can explain why we would expect to see a more-or-less linear day-by-day reduction in the percentage growth rate.  This must be what epidemiologists mean by “sub-exponential” growth, but I have yet to find a description of that that has any mathematical detail.

So, a trend is a trend until it ceases to be a trend.  That’s about as much as I can say at this point.  The trend looks good.

That said, I’ll do my next post on where this leaves us.  Basically, it leaves us with 99% of the local population with no immunity to the virus.  That is, I think, unique in the American experience, ever.  That’s not the way pandemics have ever ended up in the US, in the past.  They’ve always “run their course”, so that, at the end, the population was more-or-less naturally immune to the disease that caused the pandemic.  This one didn’t, at least locally, thanks to decisive action by the Governor.

But the fact remains, the US has never been in this situation before.  I don’t think any country in the world ever has.  Accordingly, it will be no surprise if plans for unwinding the lockdown are tentative and changeable.  That’s not because the government can’t make decisions.  It’s because no one has ever been in this situation before.  Even if we have managed to tamp this particular phase of the outbreak down, moving forward is going to entail risks until a vaccine is developed and distributed.

I really want to emphasize that this is a unique situation.  Never before observed in the history of the US, possibly in the history of the world.

And I want to keep saying that because if Virginia doesn’t end up using the emergency beds that were set up, you can be sure the nut-cases will start coming out of the woodwork, claiming that everybody over-reacted, none of this shutdown was necessary, and so on.  (Add a side-order of your favorite conspiracy theory).  And those same people will be pressing for an immediate lifting of restrictions.  And that’s probably not a very bright thing to do, give that 99% of use remain susceptible to infection.

And I just want to be clear, if this is under control at this point, the only reason for that is that the Governor adopted extreme measures, early on.  This didn’t stop on its own, with 99% of the Virginia population lacking immunity.  Pandemics don’t do that.  This stopped because we shut down public spaces, isolated, and masked up.  Anybody who says otherwise has no clue what they are talking about.