Friday, May 22, 2020

COVID-19 Changes in the Past Week

I found a few states surprising: Georgia, South Dakota. The states with increases are about what I would expect, especially Texas, Oklahoma, Michigan and Wisconsin. Note Michigan is still under lockdown, but their number of cases is still going up. Maybe more testing, but you have to wonder if some of it isn't all the maskless protesters. (So apparently big guns and silly camo don't stop the virus.) This image isn't dynamic -- you can't hover over the states for data -- but you can at the original at Axios.



18 comments:

David in Cal said...

Thanks for a thought-provoking exhibit, David. One question is provokes is what data is best to look at. The chart is based on the change in number of new cases, which is the 2nd derivative of the total number of cases. This is a meaningful figure, but I think it should be looked at in conjunction with the value and the 1st derivative, that is the total number of cases per capita and the rate of new cases per capita.

Cheers

Ned said...

David in Cal, 23 March: "Evidence more and more shows that cloroquine is probably beneficial against corona virus, just as Trump predicted."

Now:

Huge Study Throws Cold Water on Antimalarials for COVID-19

No surprise, as quite a few new studies have been appearing showing similar results. But this one is notable for several reasons:

* Sheer size: 96,000 patients on 6 continents

* Magnitude of effect: after adjusting for covariates, "patients who received HCQ or chloroquine were about twice as likely to die compared to controls who did not receive [them]"

* Controls: demographic factors, as well as cardiovascular disease, diabetes, lung disease, smoking, immunosuppressed conditions and baseline disease severity [in other words, the results are not due to patients who receive (H)CQ being in worse condition to start with compared to other patients]

* Conclusion: both hydroxychloroquine and chloroquine, with or without azithromycin, "are not useful and could be harmful in hospitalized patients with COVID-19"

Further conclusion: listening to President Trump's press briefings can be harmful to your health, and certainly leaves you worse informed about the pandemic than if you skipped the press briefings.

David Appell said...

David, thanks. You're probably right. I just don't have the time to crunch all the numbers; I barely have find to find how other people have crunched the numbers. I imagine all the big media outlets now have teams of young Python junkies (or some fancier, newer language) working full time to find, analyze and present all these data out there. Probably a fun job.

David in Cal said...

Ned - thanks very much for the link.

Here's a article observing that the the Association of American Physicians and Surgeons (AAPS) said that "Hydroxychloroquine Has about 90 Percent Chance of Helping COVID-19 Patients"

https://aapsonline.org/hcq-90-percent-chance/

At this point, if I do come down with the disease, I don't know whether or not I'd want to try this medicine. Hopefully the many ongoing studies will resolve the question more definitely.

Cheers

OnymousGuy said...

There are problems with these data.

New Hampshire is not improving.
New Hampshire data

David Appell said...

You know the Association of American Physicians and Surgeons is a very ideological, conservative group. Read their Web site or one of their newsletters. I wouldn't trust anything they say, at all.

David Appell said...

Ony, that NH link doesn't work. Wanna try again please?

David in Cal said...

Good point, David. I had never heard of AAPS, so I wondered who they were.

The Mehra study, which Ned pointed out, has a significant weakness. My wife spotted it immediately, just from a brief description. It's not randomized. It's retrospective rather than prospective.

I think we won't know the answer until we see the results of the various randomized trials now being conducted. Some of these studies look at various things given along with the HCQ, such as zinc. They also look at applications of the medicine at various points in the state of the disease.

Cheers

Layzej said...

DiC, if you trust the Association of American Physicians and Surgeons (AAPS) then you may want to avoid a vaccine when it becomes available. According to a re-analysis of CDC Data performed by that same group, vaccines may lead to autism.

David in Cal said...

Layzej -- that's upsetting. A quick search led me to this quote,

"Asked whether vaccines increase the risk of autism, she said, 'I think that the definitive research has not been done.'" https://www.theatlantic.com/health/archive/2020/02/aaps-make-health-care-great-again/607015/

I think that quote is amazingly ignorant. There's absolutely no reason to think or suspect that thimerosal causes autism. The original study that started this craze was shown to be bogus. Also, "Between 1999 and 2001, thimerosal was removed or reduced to trace amounts in all childhood vaccines except for some flu vaccines. This was done as part of a broader national effort to reduce all types of mercury exposure in children before studies were conducted that determined that thimerosal was not harmful. It was done as a precaution."

I would be interested in that re-analysis of CDC data that you referred to. Do you have a link?

cheers

OnymousGuy said...
This comment has been removed by the author.
OnymousGuy said...

I don’t know this link is screwy but here it is:
https://covidtracking.com/data/state/new-hampshire#historical
AutoCorrupt is being difficult.

OnymousGuy said...
This comment has been removed by the author.
OnymousGuy said...

I will try again. It works on my blog.
NH data

Layzej said...

I would be interested in that re-analysis of CDC data that you referred to. Do you have a link?

My comment above has a link to a press release.

David Appell said...

OG, this last link works fine. Thanks!

David in Cal said...

Thanks, Layzej

Layzej said...

Found this great COVID dataset for anyone interested: https://ourworldindata.org/coronavirus-testing

It has test/day, cases/day, deaths/day, + much more for just about all countries.