Coronavirus Cases and Context

Since the beginning of this all, I’ve consistently been frustrated with the way the media has presented numbers. Here’s a typical chart the media will often present with a headline of “Cases Continue to Rise”, or worse, “Spread Continues to Increase.” 

It looks scary. It is scary. It is real numbers. 

But it’s neglecting some very relevant context. Let’s look at total confirmed cases vs. total tests administered.

 
The two charts tell much different stories, don’t they? The increase of identified tests is directly related to the increase in testing. 

Cuomo, in his daily press briefings, warned numerous times that as Coronavirus testing ramped up, so would the number of identified cases. He wasn’t the only one who warned this. Yet, the media never shows cases against tests.

But even these two graphs only show cumulative numbers. We’re looking for trends in daily increases, so breaking down daily new confirmed cases against daily new tests give us a better sense of how quickly either metric is growing. 

 

Above compares daily identified new cases, against negative tests, as their respective shares of that days total tests given. 

Daily new cases are growing, but not as quickly as daily new tests (the trend lines for each metric continue to separate.) What we’re looking for is a decline, and these numbers don’t tell us when we’ll see that. But it does give more context and clearly shows that the daily identified cases is being driven (in part) by increase in testing. 

As of now, NY State accounts for over 20% of total tests performed in the US, and over 40% of the total confirmed cases in the US. Over 30% of the tests given in NY are coming back positive, versus 8% in the rest of the country. I’ve always wondered why the positive percentages aren’t higher across the country AND NY…because they were limiting testing to when most needed. Nevertheless, NY’s numbers, in comparison, look much worse across the board. 

We’re looking for the peak of the curve, a curve who’s beginning and breadth we don’t know (NY has still only tested 5% of it’s population.) We’re looking for a slowing in daily new case increase, or even better…a decline in number of daily new cases. But that number has a direct relationship to the number of new tests we’re giving every day too, so to neglect that data point is a bit disingenuous.

Of course new cases is being used as an indicator of growth, or spread, of the disease. That’s what we’re trying to fight with distancing and isolating ourselves. The disease has an average 5 day incubation period, and an average 7 days between onset of symptoms. So even when we get past the peak of daily cases, daily hospitalization rate could continue to increase for 12 days. The strain on the hospital system is the big fear. So watching those rates is important too. Unfortunately, NY has only been published hospitalizations for the past week or so…

 

Because these numbers trail case increases by 12 days (in theory), these trends should mirror trends in daily case increases from 12 days prior (not accounting for changes in demographics infected – age, underlying health conditions). 

These are also numbers to keep a close eye on, more on that in another post…