Regarding long COVID: Zvi, does your model above incorporate the following findings (published after your original deep dive in Sept)? (If anyone other than Zvi has thoughts on how these affect the model laid out above, would be curious to hear your thoughts as well.)
Affects over 1⁄3 of COVID sufferers, 2x as high as for flu:
“In the study published Tuesday in the journal PLOS Medicine, researchers found that about 36% of those studied still reported COVID-like symptoms three and six months after diagnosis. Most previous studies have estimated lingering post-COVID symptoms in 10% to 30% of patients.
...Although long COVID is poorly defined, the researchers looked at such symptoms as chest/throat pain, abnormal breathing, abdominal symptoms, fatigue, depression, headaches, cognitive dysfunction and muscle pain.
… the new study concludes that the chances of getting COVID-19 symptoms months after the acute stage of the illness was more than twice as high as for influenza.”
Other points to consider:
“The Oxford-led team also found that people who had more severe COVID-19 illness were more likely to get long COVID. Likewise, female and young adult patients also had an elevated risk for the long-term symptoms, but the authors of the study found no difference between white and nonwhite patients.”
11/24/21 Reuters article noting that “COVID-19 vaccines are highly effective in protecting against serious illness, but they do not protect against “long COVID” in people who become infected despite vaccination”
(However, both the article and the study it cites note that vaccines are protective against long COVID to the extent that they prevent infection in the first place.)
And thank you for this post and everything you’ve written throughout the pandemic!
I don’t know if I’ve looked at those particular ones but I’ve looked at a bunch. At this point I’m convinced that any study with improper controls will find a lot of Long Covid, and it doesn’t mean much, because people report everything as Long Covid.
And that if a study claims that Long Covid isn’t correlated with severity of Covid, it’s not properly controlled. This was never plausible to me, and the recent finding that Germans who didn’t know they had Covid didn’t report any Long Covid symptoms either seems conclusive that it can’t be right.
Thanks for responding. The first study did find that severity is correlated with incidence of long COVID, and controlled as follows: “Propensity score 1:1 matching [19] (with greedy nearest neighbor matching, and a caliper distance of 0.1 pooled standard deviations of the logit of the propensity score) was used to create cohorts with matched baseline characteristics and carried out within the TriNetX network” My knowledge of statistical analysis is rather insufficient to evaluate the controls in that study, so hoping you or another commenter can chime in on that.
Regarding long COVID: Zvi, does your model above incorporate the following findings (published after your original deep dive in Sept)? (If anyone other than Zvi has thoughts on how these affect the model laid out above, would be curious to hear your thoughts as well.)
9/29/21 NPR article recapping a study from around that time
Affects over 1⁄3 of COVID sufferers, 2x as high as for flu:
“In the study published Tuesday in the journal PLOS Medicine, researchers found that about 36% of those studied still reported COVID-like symptoms three and six months after diagnosis. Most previous studies have estimated lingering post-COVID symptoms in 10% to 30% of patients.
...Although long COVID is poorly defined, the researchers looked at such symptoms as chest/throat pain, abnormal breathing, abdominal symptoms, fatigue, depression, headaches, cognitive dysfunction and muscle pain.
… the new study concludes that the chances of getting COVID-19 symptoms months after the acute stage of the illness was more than twice as high as for influenza.”
Other points to consider:
“The Oxford-led team also found that people who had more severe COVID-19 illness were more likely to get long COVID. Likewise, female and young adult patients also had an elevated risk for the long-term symptoms, but the authors of the study found no difference between white and nonwhite patients.”
11/24/21 Reuters article noting that “COVID-19 vaccines are highly effective in protecting against serious illness, but they do not protect against “long COVID” in people who become infected despite vaccination”
(However, both the article and the study it cites note that vaccines are protective against long COVID to the extent that they prevent infection in the first place.)
And thank you for this post and everything you’ve written throughout the pandemic!
I don’t know if I’ve looked at those particular ones but I’ve looked at a bunch. At this point I’m convinced that any study with improper controls will find a lot of Long Covid, and it doesn’t mean much, because people report everything as Long Covid.
And that if a study claims that Long Covid isn’t correlated with severity of Covid, it’s not properly controlled. This was never plausible to me, and the recent finding that Germans who didn’t know they had Covid didn’t report any Long Covid symptoms either seems conclusive that it can’t be right.
Thanks for responding. The first study did find that severity is correlated with incidence of long COVID, and controlled as follows: “Propensity score 1:1 matching [19] (with greedy nearest neighbor matching, and a caliper distance of 0.1 pooled standard deviations of the logit of the propensity score) was used to create cohorts with matched baseline characteristics and carried out within the TriNetX network” My knowledge of statistical analysis is rather insufficient to evaluate the controls in that study, so hoping you or another commenter can chime in on that.