Inside the Numbers: Was it Thanksgiving, or was it seasonality?
COVID-19 positive tests and hospitalizations increased in the two weeks after Thanksgiving, but both had already been trending upward in the weeks ahead of the holiday.
There, gotcha with the headline and lede.
Not claiming to know the answer, we started the analysis that follows with the impact of Thanksgiving on COVID trends in Virginia as our guiding question.
For the purpose of this analysis, we compiled data for COVID-19 positive tests reported by the Virginia Department of Health, and tried to match that to data on COVID-like illness emergency department visits and CLI ED hospitalizations reported weekly by VDH.
The VDH data on CLI is complete through Dec. 12, a week ago, so our analysis is limited to data through Dec. 12.
With that all said, here we go.
COVID-19 positive test data
The VDH COVID-19 dashboard had the state reporting 44,996 positive COVID-19 tests in the two-week period following Thanksgiving – the reporting weeks of Dec. 5 and Dec. 12.
That was up 12,173 – 37.1 percent – from the 32,823 positives reported in the two weeks leading into Thanksgiving.
Those totals were from the reporting weeks of Nov. 21 and Nov. 28.
The preceding two weeks – the weeks of Nov. 7 and Nov. 14 – had 19,607 reported COVID-19 positive tests, according to VDH.
The increase between that two-week period and the Nov. 21 and Nov. 28 two-week period was a raw number of 13,216, and a percentage increase of 67.4 percent.
For more context: there was also a decent percentage increase from the two-week period encompassing the weeks of Oct. 24 and Oct. 31 to the two-week period encompassing Nov. 7 and Nov. 14.
The total number of positive tests in the Oct. 24 and Oct. 31 weeks was 15,953. The raw number increase wasn’t huge – 3,654 – but the percentage increase of 22.9 percent would tend to jump out at you.
CLI ED visits
CLI emergency department visits, like the positive test numbers, have also been trending upward since October, though the share of CLI visits relative to the increase in positive tests has been on the decline since mid-November.
For three two-week periods beginning Oct. 10 – the weeks of Oct. 10 and Oct. 17, Oct. 24 and Oct. 31, and Nov. 7 and Nov. 14 – the share of CLI ED visits relative to positive tests was in the range of 46.0 percent (Nov. 7 and Nov. 14) to 48.8 percent (Oct. 24 and Oct. 31) to 54.2 percent (Oct. 10 and Oct. 17).
Beginning the week of Nov. 21, the share of CLI ED visits relative to positive COVID-19 tests went from roughly half to roughly a third – 34.6 percent (Nov. 21 and Nov. 28) and 29.8 percent (Dec. 5 and Dec. 12).
The raw numbers of CLI ED visits were holding steady in October – at 7,832 for the two-week period encompassing Oct. 4-17 and 7,791 for the two-week period encompassing Oct. 18-31.
The Nov. 1-14 period had 9,025 CLI ED visits. The Nov. 15-28 period saw an uptick to 11,356 (a 25.8 percent increase), and then the Nov. 29-Dec. 12 period saw another push, to 13,427 (an 18.2 percent increase).
CLI ED hospitalizations
The raw number of CLI ED hospitalizations held relatively steady from the Oct. 10 and Oct. 17 weeks all the way through to the Nov. 21 and Nov. 28 weeks – ranging from a low of 2,084 (for the weeks of Oct. 24 and Oct. 31) to a high of 2,485 (for the weeks of Nov. 21 and 28).
From that level, there was a push to 3,032 CLI ED hospitalizations for the weeks of Dec. 5 and Dec. 12.
The share of CLI ED hospitalizations to positive tests was at 13.1 percent for the Oct. 18-31 period.
It was at 6.7 percent for the Nov. 29-Dec. 12 period.
Whatever good news we can find here, this is it. If we were seeing CLI ED hospitalizations trending at the same rate as back in October, we’d have seen 5,894 ED hospitalizations over the Nov. 29-Dec. 12 period – an increase of 2,862 patients in our hospitals over those two weeks.
This as, from data that we’ve analyzed from the COVID-19 Tracking Project,
COVID-19 positive test numbers have tripled from early October – from the total of 14,441 reported Oct. 4-17 to the 44,996 reported Nov. 29-Dec. 12.
CLI ED visits almost doubled – from 7,832 in the Oct. 4-17 period to 13,427 in the Nov. 29-Dec. 12 period.
CLI ED hospitalizations were also up – not at a rate of two or three times, like cases and ED visits.
The ED visits are a strain on hospitals. The increase in ED hospitalizations is a strain as well.
It is worth noting that the hospital census in Virginia back on Oct. 1 was 13,247, 80.4 percent of the overall adult in-patient capacity, and on Dec. 12, the census was 13,216, 80.2 percent of capacity.
Those numbers reflect total census of everybody in hospitals – not just for COVID, but for everything, your run-of-the-mill plus the COVID.
These are the numbers that should be getting the big font headlines – the overall number in hospitals, the number with COVID, the number going to emergency rooms because they think they may have COVID.
The numbers of positive tests – which are what get the big font headlines – are … just numbers.
This is a long way of saying, we’re going to keep our eye on the CLI numbers the next few weeks.
No doubt there are going to be headlines about Christmas impacting COVID numbers that we can use to hook you in when we need to.
Story by Chris Graham