What is a case of COVID19?

If you tell me you’ve had a case of the flu, I know exactly what you mean; you were sick for a time and are now over it.  That is not at all what State Departments of Health or the press mean by ‘cases’ of COVID19.

They mean one of two things

l. Presence of the virus genetic material on a sample you give

2. Presence of antibodies to the virus in your blood.

Neither means that you’ve had any clinical illness whatsoever.   In the case of the Bronx where 20 – 50% of various areas had a positive antibody test, none were hospitalized — these were people walking about.

So the scare headlines about cases of COVID19 explode are just that.  Americans love sports.  Consider the players on the active rosters of the NFL, Major Leagues and NBA last year.  If any of them had died of COVID19 you would certainly have heard about it.

On June 29th I wrote a post ”Where are the Deaths?” making the point that if ‘cases’ surged, deaths should as well. Here’s a link — https://luysii.wordpress.com/2020/06/29/where-are-the-deaths/.   It hadn’t happened then, and hasn’t happened now.

Granted there should be a lag between diagnosis and death.  Amazingly, it’s hard to find figures on just what the lag is.  An early report from Italy said 8 days, but these were mostly very old people.

I’ve been focusing on Georgia for several reasons.  First, they were the first to loosen their restrictions of activity. Second they are very conservative in the way they report data, not regarding any number as solid until 14 days have passed and all the information has had a chance to flow in. Third, Georgia uses moving averages of cases and death for the past seven days.  This smooths out the spikes (which are largely due to how the data flows in).  For an example of going off the deep end about a one day spike in cases — see https://luysii.wordpress.com/2020/05/28/data-cherry-picking-101/ — a copy of which will be found below the ****

So now hie yourself to the following website — https://dph.georgia.gov/covid-19-daily-status-report

If you scroll down to COVID-19 testing, you will see that they’ve tested a 1.17 million people (nearly 10% of the populaton).  94,000 had the virus present and 9,000 had a positive antibody test (so asymptomatic people account for nearly 10% of the ‘cases’ of COVID19

Now keep scrolling down the site, past the map of Georgia and you get to COVID-19 cases over time.  There are 3 tabs you can click: Cases, Cumulative Cases and Death.

Each is plotted vs. the date.

There is a movable cursor allowing you to see the data from any given day.  On the day restrictions were loosened (25 April) the 7 day moving average of new cases was 736, and the 7 day moving average of daily deaths was 41.

Now click on cases and look at 24 June (using the cursor), and the 7 day moving average of daily new cases has exploded to 2,100

Click on deaths and the 7 day moving average of daily deaths on 24 June is 17.

When did the explosion in daily new cases begin?   To my eye, it happened around the first of June (when the daily death rate was 24).  So during the explosion of new cases death rates continued to drop.

Very similar data is available from Florida; look at the two graphs — side by side– of new cases and daily deaths on the following link — https://www.miamiherald.com/news/coronavirus/article242270081.html

For those of you old enough to remember, the number of new cases is as useful as the body count was during the Vietnam war.  It was something which could be precisely determined, but what was irrelevant for what was really needed to be known.

The two pieces of data we need to determine whether we’re winning or losing in the pandemic are (1) the daily death rate from COVID19 (2) the number of people currently hospitalized with COVID19.

If the marked increases in new cases is meaningful and scary, we had better see a rise in the death rates soon; but it’s already been 3 weeks for Georgia (solid numbers) and about 5 weeks for Florida, and we still haven’t seen a rise in death rates comparable to the rise in the number of ‘cases’ since 1 June.  In fact we’ve seen no rise at all.

Follow the links, look at the data and draw your own conclusions.

****

Here’s the old post warning you to beware of ‘daily spikes’ of anything.

 

Data Cherry Picking 101

A friend sent me the following link — https://www.voanews.com/covid-19-pandemic/wisconsin-reports-its-highest-daily-increase-covid-19-cases.

It starts off like this — dates in parentheses added by me.

“Health officials in the midwestern U.S. state of Wisconsin reported a record number of new COVID-19 cases Thursday, (28 May) two weeks after the state Supreme Court struck down a state-wide stay-at-home order issued by the governor and enacted by the state health department.

The Wisconsin Department of Health Services reported 599 new known COVID-19 cases Wednesday, (27 May) with 22 known deaths, the highest recorded daily rise since the pandemic began. The department reports the state had more than 16,460 known cases and 539 known deaths as of Wednesday.”

Well that proves it, doesn’t it?   Removing restrictions has clearly  been a disaster.

No it doesn’t.  This is data cherry picking par excellence — one day’s cases — after a long holiday (Memorial Day)  weekend means nothing.  The ‘spike’ is an artifact of how cases are reported.

Here are the daily new COVID-19 cases from Massachusetts (which has relaxed nothing so far)
24 May 382
25 May 281
26 May 197
27 May 688 
 
QED
Post a comment or leave a trackback: Trackback URL.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: