Monday, March 23, 2020

What About The Data?

If you've been following the numbers for a bit, you may have noticed that they change in ways that are bizarre.  As an example, here's part of a spreadsheet I did that collected numbers for Texas from the Google display:

cases (Google)deaths
02/13/20201
first reported case, 15th in nation
02/27/2020
03/03/2020691
03/05/20201
03/10/2020
03/13/2020
03/20/20204355
03/21/204175
03/22/203346

Why did the numbers go from 435 to 417 to 334?  There was a change in the reporting -- Texas got official coronavirus test kits and instead of using hospital numbers they switched (over a 2 day period) to "official diagnosis via test kit" numbers.

The time of day that the numbers are being reported also changed.  A state might have 20 cases if you took the numbers at 10 am... but 24 cases if you took it at 10 pm.

This is problematic.  The original numbers seem to be those sent to the CDC from hospitals.  The "test kit" numbers are only given AFTER someone has gone through all the procedures to get approved to take the test AND the person has taken the test.  IF they have insurance and the ability to pay for the test and can get to a testing site (not everyone has a car and here in Texas nearly 22 percent of people are uninsured.  That translates to "at a minimum, 1/5th of all people in Texas will have trouble getting the test even if they're very sick with COVID-19. so their data won't show up in the 'official count'."

 According to the Texas Health Department (https://dshs.texas.gov/) the labs can currently only process 26 samples per day.  It can take up to a week to learn the results.

So the numbers we see are numbers from several days previously.  If I'm feeling generous, I'll say that the "current number" shown on the coronavirus info page of your choice is actually for two days ago.  If I'm not feeling generous, I'd say that it's a week out of date.  The truth lies somewhere in between.  At this point we're fairly early in the pandemic (only 2 months of data) and the real morbidity and mortality numbers won't show up for quite awhile.


Rate of Spread

A tweet just today (see below) from the head of the World Health Organization highlights something that I saw in my data last night -- the rate of spread is increasing.  Some of this is due to lack of data... no real testing, people only show up at hospitals when they're really really sick, critically ill people at home who are going out in search of treatment for their symptoms (going to a pharmacy to get cough medicine, for instance) so we can only guess at the numbers who are infected with the virus.  By the time they get to the attention of the medical monitors, they may have passed the virus along to dozens (or hundreds) of people -- and we're talking about a 2 week incubation period (our best guess.)



Feeling Helpless?

This is a big scary situation, and one that epidemiologists have been warning about for awhile.  But I'd rather not leave you with a message of fear because in these times we have things that we can do... positive things.

* write about your experiences.  Keep a blog and keep notes on what you're doing -- not only the medical prevention measures but blog about changes in work and socialization.  THIS TYPE OF PANDEMIC WILL COME AGAIN (and again, and again.)  The tips and tactics you write down now will help others when another novel virus rears its ugly head.
* keep checking on the vulnerable ones you know -- people with low income, elderly, immune compromised.
* understand that Scary Data shows trends but does not show an accurate to-the-second picture.

My advice on masks:  Use one if you want (gloves, too) and don't feel embarrassed about it.  Be sure to clean (or dispose) of the mask after each use (ditto gloves.)  Likewise if you don't feel one is necessary, then don't use one... but don't shame the ones who do.  I will be using masks (and occasionally gloves) because my husband and I are in the high risk category and because he has diabetes and Parkinson's and minor health crises can escalate rapidly with those two conditions.

Notes and Sources:

"Morbidity and Mortality" is epidemiologist-speak for "too sick to go to work" and "dead", respectively.

You can read an overview of some of the problems Dallas had on the 2nd day of "drive up testing" here: https://www.dallasnews.com/news/public-health/2020/03/22/day-2-of-dallas-countys-coronavirus-mobile-testing-leaves-some-patients-frustrated-concerned/

Tweet from the head of the World Health Organization today:


No comments: