I noticed a weird series of events in my state's CV infection data today.

[KY_EnviroGuy]

Last edited Tue Apr 21, 2020, 07:51 AM - Edit history (1)

On the DU Kentucky board, DUer Elleng posted that there was a recent disturbing increase in CV infections reported by our governor today for the whole state.

I went to this data site to see if I could localize the source: https://infection2020.com/#

(to the left, click on "KY" to get a county drop-down, then click on the "curve" symbol to see the county's case curve)

But by looking at the individual graph of reported infections for several counties, I'm seeing sudden upticks in cases, all beginning on or near the 16th or 17th and all having a quick jump in rate on that day and continuing until today. For example, I see it in Jefferson, Kenton, Daviess, Warren, and Boone counties and numerous others.

This makes me wonder if this could be the result of Easter church services and family gatherings on the 10th (Good Friday) and 12th (Easter Sunday).

It just seems odd that the same sudden increase in infection rate occurred in several diverse areas of the state with almost identical curves starting on the 16th or 17th.

Any other ideas for a possible cause?

KY...............


4/21/20 edit: added this screen shot of the graph for one county to demonstrate this change in slope of the curve. Almost identical change noted in several counties although I have no way to determine where these new cases come from (a normal change, a change in state or federal testing locations or change in test reporting methods, etc.).

Also note that I do not observe this change in very many counties in adjacent states as I would have expected if it were Easter related.

---------------------------

[Tetrachloride]

1. Can you correlate the locations of infections with the the larger church gatherings.

[KY_EnviroGuy]

4. No way to do that accurately to my knowledge because....

there was no required reporting of meetings. Therefore, it would be entirely educated guesses as to which towns were more likely to break the rules.

Another question is whether an infection is recorded at the person's county of residence or where they were diagnosed and/or hospitalized. Some many community clinics and hospitals have closed in recent years, most people are going to the nearest large city for medical services.

I suspect our governor is on this.

[KY_EnviroGuy]

5. Another thing.....

I believe most of the megachurches here did not hold public services because they did not want to get bad press and buck the governor. However states like Kentucky have thousands of small churches that probably did not care about the rules so they would represent the problem.

[Mariana]

6. Lots of churches are gathering and staying quiet about it.

The only reason we know about the ones we know about, is that they like to brag about their stupidity.

[SheltieLover]

2. Interesting!

Is Moscow Mitch playing with the numbers to make lock down look like a failure? Sad these thoughts come to mind as a result of their determination to kill masses.

I can just hrar the "logic"... "Well if your going to get it anyway, go to work."

K&R

[elleng]

3. I saw that speculation elsewhere, too;

sounds reasonable.

[lindysalsagal]

7. It wouldn't be too hard to just ASK the current infected people if they broke quarantine in the last

month...

Not too hard, if there were an organization taking care of this mess.

[clutterbox1830]

8. I have notice the all cv numbers are lower every Sunday

I just assume that some states are not working or reporting as much on this day. The number spike up on Monday. This has been happening for the past 3 weeks.

[Amishman]

15. testing center disruptions and reporting disruptions

We only get new data when people are doing the things that generate the numbers.

[IronLionZion]

9. It could also be testing

If new testing centers opened up during that time then it would increase the numbers who tested positive.

[Laura PourMeADrink]

12. Yes, testing is key. Heard a doctor say today that it

would be a full two weeks to tell what effect, if any, the Jax Beach opening would have.

In KY, did they go to church before Easter? And did they have access to immediate/quick testing. Window small, but it is Turtle's state.

[LudwigPastorius]

10. New CDC guidelines were issued on the 14th defining case counts as confirmed AND probable cases.

*As of April 14, 2020, CDC case counts and death counts include both confirmed and probable cases and deaths. This change was made to reflect an interim COVID-19 position statement issued by the Council for State and Territorial Epidemiologists on April 5, 2020. The position statement included a case definition and made COVID-19 a nationally notifiable disease.

A confirmed case or death is defined by meeting confirmatory laboratory evidence for COVID-19. A probable case or death is defined by i) meeting clinical criteria AND epidemiologic evidence with no confirmatory laboratory testing performed for COVID-19; or ii) meeting presumptive laboratory evidence AND either clinical criteria OR epidemiologic evidence; or iii) meeting vital records criteria with no confirmatory laboratory testing performed for COVID19.

This could explain the spike.

[KY_EnviroGuy]

14. What I observed isn't what I would call a spike, but rather....

a sudden change in the slope of the curve to be much more steep. Please note the screen-shot of one of the counties (Jefferson/Louisville) I added to the OP, above.

It may be there was a change made in reporting requirements or methods from the state on the 16th or 17th, or as you said changes in CDC methodology, or that more testing sites have been added around the state.

It's quite a head spinner when our data is so fluid due to so many variables and so many agencies potentially having an effect on it.


KY........

[Ms. Toad]

11. The criteria for "confirmed cases" changed on the 14th.

The CDC now defines cases (and deaths) as confirmed if:

(1) the patient has any positive test (including for antibodies - which means they previously had the case), or
(2) There is epidemiological data to suggest the disease is probably Covid 19 (i.e. living in a nursing home or prison where one person has tested positive) + symptoms + no other explanation for the symptoms.

States are reviewing prior cases to determine whether to report previously undetected cases and/or deaths. This results in temporary surges in new cases as reporting catches up to the new criteria.

Here's a complete explanation:

https://www.worldometers.info/coronavirus/us-data/

[KY_EnviroGuy]

16. See edit to the OP, above (added typical graph).

One might guess that a change in CDC methodology would result in a simple one-time step increase in cases, but then it hit me that it might takes a number of days for data to trickle in from all our counties.

I may never know without perhaps emailing our State Dept. of Health and requesting an explanation, but I doubt if they would take time to reply.

Thank you for adding that info to the conversation.

KY......

[Ms. Toad]

17. Ohio started including the numbers on the 14th

NY didn't start including them until most of a week later.

And even when they add, states have been retroactively adusting earlier cases/deaths. I've had to adapt my formula for about a week as the numbers on which I based my formula grew.

[Warpy]

13. You're right on the money IMO.

Weird stat here is about 2000 cases over a month or more, 0 recoveries. That means that tests are in such short supply they can't test people who have recovered from the illness to make sure they're not still shedding virus.

ETA: Cases are high in the 4 corners, moving to the southeast, all downwind of a huge coal fired electrical plant, poorest air quality in the state. My county has the most cases, it's where the big hospitals are. Still, I find the connection interesting.