Thread by @chrisprener

Here is my #Missouri #COVID 🧵 for Monday, 5/4. To see the interactive maps, head over to the MO tracking site - slu-opengis.github.io/covid_daily_vi….

Since Missouri “reopens” today, with the statewide #StayAtHome order lifted, I want to take stock of our data.

1/22

I don’t want to bury the lede, though - as of “reopening” day, we:

❌ do not have reliable, public hospitalization or symptom data,
❌ have not increased testing volume in weeks, and
❌ have significant increases in cases in multiple regions around the state.

2/22

The state’s highest rates are in Saline County ☝️☝️, where an outbreak several meatpacking plants has led to nearly 200 cases. An earlier meatpacking outbreak in Moniteau County seems to have subsided based on the data we have at hand. 3/22

We’ve also seen a recent, and quite substantial, meatpacking outbreak in Buchanan County (St. Joseph, MO). A large number of these individuals have been reported to be asymptomatic, but don’t put an asterisk next to this - each of these individuals has COVID19. Full stop. 4/22

The other recent gains outside of #StLouis (City in particular) are in southeast Missouri - Mississippi, Scott, and Pemiscot counties all are seeing growth right now. Thankfully, Perry, which had a high rate for weeks, has not added many new cases recently. 5/22

So, as we “reopen,” we’re still seeing new cases added in substantial numbers in multiple regions of the state. This recent activity is pushing our rolling average of new cases up - we’ve plateaued overall, but the “excluding STL” trend for MO is nearing its peak again. 6/22

Our 7-day average of mortality is still near its peak, as well. We’ve had 373 Missourians die of COVID19 so far, and are averaging almost 14 more deaths per day right now. New mortality has plateaued in the #StLouis metro, and has fallen substantially in the #KansasCity area. 7/

Speaking of metro areas, the real increases in terms of cases fall basically in three - the #StLouis metro, the #KansasCity metro, and the St. Joseph metro (discussed above). 8/22

The #KansasCity metro has consistently seen the highest infection rates on the Kansas side in Wyandotte and Leavenworth counties. Both have had institutional outbreaks, with the most recent being in a prison in Leavenworth County. #KCMO has also seen a recent uptick, though. 9/22

Mortality in the #KansasCity area has been concentrated in Wyandotte County, which has had significantly higher mortality rates than any other county in the metro. 10/22

In #StLouis, the city and the county continue to be the center of the regional outbreak, though numbers in Metro East continue to tick up as well. With the exception of St. Clair, however, all counties are below their peak 7-day rolling average values. 11/22

Note that #StLouis City posted its one day largest addition of new cases yesterday, which aren’t included in this trend yet. Mortality rates in metro counties are also rising - note the jumps in Clinton, Franklin, the St. Clair counties. 12/22

There has been no change in the geographic trend within #StLouis, nor have the racial disparities indicators shown any sign of diminishing. The gap in morbidity & mortality is about x3 to x3.5 as high for African Americans as it is for white residents in the city and county. 13/

Last night, I added another set of data to the site to consider - testing data. As of yesterday, #Missouri had cumulatively tested about 1500 per 100,000 residents since the outbreak began. However, our volume of new tests has not increased since early March (on average). 14/22

I also posted last week that the hospital data available through the Missouri Hospital Association was not reliable (not because of them, per se, but because individual hospitals are not reporting into national systems at high enough rates). 15/22

So, to recap, we:

❌ do not have reliable, public hospitalization or symptom data,
❌ have not increased testing volume in weeks, and
❌ have significant increases in cases in multiple regions around the state. In terms of the White House’s guidance, we’re not… 16/22

...using crisis care, but we either cannot assess with public data or do not seem to be meeting many of the other benchmarks.

Testing to me is the most frustrating - compare how IL has increased their average of new tests per to our relatively static rate. 17/22

Yet we’re reopening anyway. There are no longer #StayAtHome orders in the bootheel, in Saline County, or in Buchanan County, all of which are seeing increasing numbers of cases. Only the red counties below still have a stay at home order (map via MO HSS): 18/22

Residents of Metro East, where cases are increasing but where they will be under a stay at home order for at least the rest of May, can also now come into MO to patronize businesses that are reopening. And come to concerts and Six Flags, if those actually re-start as well. 19/22

Keep in mind that the consequences of this will take at least two weeks to materialize - if cases go up this week, they are almost certainly due to infections that occurred before reopening. So, we need to be patient in terms of how we read these data. 20/22

My standard caveats about uncertainty - these (1) are historical data that reflect infections 2-3 weeks ago, (2) are biased by limited testing, (3) may include probable but unconfirmed cases in some counties, and (4) rates are not individual probabilities of illness. 21/22

Additional maps are on my tracking site - slu-opengis.github.io/covid_daily_vi….

My #rstats code and and data are all available on @GitHub - github.com/slu-openGIS/co…. They're licensed for re-use with attribution, and feedback and improvements are always welcome. 22/22