Don’t get me wrong, there are definitely signs that things might get worse. For example, check out the positivity rate for UNC this week vs. last week:
That is very bad news.
But I’m trying very hard to be unbiased, and not looking only for evidence of bad news.
I’ve got a pain in the ass project tying up my math brain.
I saw that daily deaths dropped just below 1,000 7DRA.
I think we are mostly waiting for the OFS spike to start.
Interesting read about genetic surveillance on transmission following that Biogen conference in late Feb.
A few tidbits that I found notable:
Responsible for at over 3% of all US cases.
Superspreaders are weird. The virus had come to Boston several times already, but didn’t really see in.
Virus underwent a small mutation during the course of the conference. One person apparently had both variants simultaneously.
As a country, we are doing very little genetic surveillance of the virus. Most positive samples are being discarded without being catalogued, which makes it harder to detect spread patterns.
How does testing factor in to your numbers? Right now, testing seems like a competition for which hot spot can do the least. Ohio numbers for testing are relatively flat, but cases vary wildly by day. To me, that suggests something odd is going on with reporting.
I’ve said this before, but I think testing is a bit of a red herring. If lower rates of new cases were driven by limited testing (rather than a decline in the actual rate of new cases), we should observe an increase in the moving estimate of recent-cohort CFR: You would have approximately the true number of COVID deaths divided by an artificially low number of positive cases. But we haven’t seen that at all. As I posted above, the estimated recent-cohort CFR has fallen steadily. Until I see that number spike up (or someone presents some different compelling data), I’m not going to believe that testing capacity is distorting the trends.
I read “fastest” correctly, but assumed it was a reference to @clovis8 rushing to defend Joe Biden from criticism.
Haven’t had any issues getting a test in Iowa and result turnaround time has consistently been 48 hours or less.
Illinois starting to reimpose restrictions, indoor dining banned in a few suburban counties. It’s been spooky, we’ve been stuck on +20%/week for about 6 weeks.
I know it’s probably a decent amount of work, but have you done an individual 50 state mock up of this idea as opposed to an overall or targeting of any particular state? I personally think the overall picture is what’s distorting everything but it’s so subtle it’s hard to see how it’s being done (we are wildly down in testing overall but it’s 5k here 10k there in most places masking the effect of where it’s coming from). I also frequently see a place (like the U.S.) go 35k, 28k, 22k, 75k, 18k, 19k, 55k. What I want to see is 35k, 33k, 34k, 36k, 32k, 31k, 33k, etc. Those are just random numbers that don’t mean anything, but the point is that even when testing was quite bad in a lot of places they were always at or near a testing cap producing relatively similar daily testing numbers. Now 7DMA tests in a lot of places ‘look’ like they’re consistent but they’re actually anything but. And it’s now becoming very obvious where lags are happening, meaning there are a lot of places where cases are seemingly being missed or shifted by delayed results. We still don’t have these key stats per new case available as far as I know:
Also, I think it’s crucial when talking about the data you’re talking about as your metric related to CFR that it’s not forgotten the data you use to track that was moved to HHS from CDC and could be cooked however they wanted it to be until it was given back as of I think the end of last week. This specifically could have been done just to cook OFB and July 4 numbers to confound anyone trying to track these rates. They just magically gave it back to the CDC when OFS happened if I’m looking for a conspiracy because their assumption is kids will die at lower rates than the general population. I wouldn’t trust any hospital/death data from the time the data was moved from CDC to HHS and probably not for at least a few weeks after it goes back to the CDC (I think this happened mid-July and was only gone about 5 weeks but it could have been mid-June as time is starting to blend).
A few weeks ago I said I thought we had entered the bulls*** phase of the pandemic. Nothing looks right to me right now (except in places that still appear to be honestly reporting and doing good testing) after having tracked all this stuff (no death stuff) relatively closely since mid-March. Something’s very off. I can’t tell you what, because I don’t remotely resemble a data scientist, but I’m convinced something’s off. That being said, I do believe cases overall are moving down, just not nearly as much as they appear to be when looking at a surface level.
“I have been criticized for remarks I made Sunday night about the benefits of convalescent plasma,” said Hahn in a tweet thread on Monday night. “The criticism is entirely justified. What I should have said better is that the data show a relative risk reduction not an absolute risk reduction.”
As an apology, that’s weak. Where’s the assurance it won’t happen again? Like maybe resign, you orange-asskisser.
i hope you brought roller blades
Co-ro-na-vi-rus, come out to play-ay!
Despite the nonsense from the admin, I’m a big believer in convalescent plasma. We made a big effort to get it to anyone who was sick, especially if they were hcws. Seemed to work for us.
Not sure what you mean - have I looked at the trends in CFR for each state? No.
Kind of seems like you’ve already decided what the truth is, and you’re only willing to accept data when it fits with that pre-determined truth.
LOL
“I just found out Herman Cain died from stage 4 colon cancer they marked him as a covid-19 death,” another says. “How can we believe anything they tell us about covid-19.”
https://twitter.com/michaelmina_lab/status/1298190789667946499
https://twitter.com/michaelmina_lab/status/1298191424094179328
https://twitter.com/michaelmina_lab/status/1298191855755173888
https://twitter.com/michaelmina_lab/status/1298192412494438400
https://twitter.com/michaelmina_lab/status/1298193019582259201
https://twitter.com/michaelmina_lab/status/1298193528686870528
https://twitter.com/michaelmina_lab/status/1298193988936241152
https://twitter.com/michaelmina_lab/status/1298194304234663937
https://twitter.com/michaelmina_lab/status/1298195059272228864
Another great thread on the whole immunity/reinfection idea. There’s going to be so much confusion about this for months.
Its kind of shocking how much the NYC subway still looks like that.
I’d want to see your graphs for each state and DC, not the U.S. as a whole or one particular state. This might be worth looking into but I understand if you don’t want to due to the level of work it might be. If you were interested, the peak for cases was the period of July 12-18 and that’s probably about as far back as you’d need to look for any kind of current trending or seeing anomalies in data.
Not at all, but you seem to be trusting things where obvious politically based shenanigans are happening. Can you find a legit reason for this to have been moved to HHS and then another legit reason why it would have been moved back and so quickly during a time where our data is extremely suspect at the current peak of the pandemic?
I currently think very little data is trustworthy, especially when it comes to hospitalizations and deaths specifically because of the politicization of that data. To me, the reality appears to be that many places are just reducing testing to make it look like they have fewer cases while the positive rate largely remains the same or goes up. A great example of something very off going on is Washington hasn’t reported any negative tests since August 3. We have no idea what their testing looks like right now. You can say I’m deciding the truth and am only willing to accept the data based on that perception, but all I’m saying is that I believe we’re flying completely blind right now. I’m not trying to make the data fit anything, I’m just looking at it. Do you have any real evidence that says we’re not flying blind that doesn’t have any earmarks of manipulation (signs of this are testing decrease, slow reporting, lags in test results, and the CDC to HHS move and back)?
The cases/testing didn’t follow this trend of weirdness until about three weeks ago. If I’m to believe what I’m seeing in the data, it’s that the miracle Trump is describing is happening. The pandemic is disappearing just like that at an unbelievably fast rate with absolutely no plausible explanation other than miracle.
