Inside the Numbers: What was going on with that Sunday COVID-19 spike?
You may have seen a headline or something on social media about a dramatic jump in COVID-19 cases in Virginia on Sunday.
The Virginia Department of Health COVID-19 dashboard reported 9,914 (!) cases, roughly 47 percent higher than the previous one-day mark.
Also on Sunday, the dashboard reported a massive increase in the number of Virginians hospitalized with COVID-19 – a one-day spike of 924, from 3,117 on Saturday to 4,041 on Sunday.
On Monday, things were a bit more back to where we’d been the past couple of weeks.
The number of new positives was at 7,245 on Monday, and the hospitalization number was back down to 3,149, a drop of 892.
Looking for clues as to what the heck is going on, I saw that the Virginia Public Access Project had noted in a tweet Sunday evening that there appeared to have been a change in the way cases were being reported.
We may have to change our COVID-19 dashboard to keep up w/changing way cases reported. We’ve focused on daily “confirmed” cases, as “probable” cases were a small number. But yesterday @VDHgov showed 9,914 “reported” cases – 5,375 confirmed, 4,539 probable. https://t.co/HdkVou8gQ4
— vpapupdates (@vpapupdates) January 17, 2021
Using that as a jumping-off point, I attempted a deep dive into the VDH data, and found that the number of “probable” cases had averaged 1,370 a day over the previous seven days.
After the blip on Sunday, the number of “probable” cases was back down to 1,858 in Monday’s reporting.
Whatever the change that the folks at VPAP noticed, it appears to be a one-day change, and it also appears that it was centered in Northern Virginia, which reported 1,548 “cases” – confirmed tests and probables – on Saturday, 3,678 on Sunday, and 1,854 on Monday.
Hospitalizations in Northern Virginia went up and then back down in a similar fashion – from 591 on Saturday, to 1,600 on Sunday, then back down to 635 on Monday.
The Northern Virginia COVID-19 hospitalization numbers, up and down, actually outpaced the changes from Saturday to Sunday and then Sunday to Monday statewide – the 1,009 increase in NoVa hospitalizations from Saturday to Sunday compared to 924 statewide, the 965 decrease from Sunday to Monday in NoVa compared to 892 statewide.
Overall hospital census was not impacted by any of it – statewide, the census was at 13,339 on Saturday, 13,312 on Sunday, and is at 13,267 today; and in Northern Virginia, the census was 2,745 on Saturday, 2,753 on Sunday, and 2,664 on Monday.
We’ll continue to monitor the numbers, but the lesson here would seem to be, if a number being reported is way, way out of line with what you’d been seeing, it may be wise to look for reasons why, before jumping on the number as a sign of it being a signal of something calamitous.
Story by Chris Graham