Re: Pfizer vaccine. It's clear something big happens around day 11. Nothing of import happens around day 32. #firstdosesfirst
Pfizer does a ridiculous time split when they look at one shot vs. two. First, there's no statistically significant difference in the first 7 days after the 2nd shot. Had they used an 11 day cut-off instead, one shot would clearly have won.
And, one shot would also likely have won had they looked within a 7 day window in either direction of shot #2.
@AlbertBourla please explain;) Can't you just publish the full data so we can have a community review? Full article here. (Otherwise, you and Pfizer are well-deserving of huge props for creating this vaccine.) nejm.org/doi/full/10.10… @
Update on this: If you look at treated vs. control in first 11 days after 2nd dose, you have 2 in treatment vs. 31 in control (where I've extrapolated control), and around 94% effectiveness. Which is a touch less effective than after but very close.
Would you believe it, the Moderna vaccine is more effective the 28 days prior to its pre-registered start count date than it is after. 96.4% for days 15-42 vs. 94.4% after day 42. #FirstDosesFirst? @EricTopol
Obviously, those differences aren't significant, but the point remains, there is no sign in the data for Moderna or Pfizer that the second dose changes anything. If you weren't told when the second dose was, you'd never be able to guess based on the data.
While, I don't think these companies are guilty of any malfeasance, if one shot is enough, suddenly demand for their products has just dropped by 50%.
Apparently, Fauci & the Biden admin have already decided not to do first doses first. That's the wrong decision. Hate to say it, but the Biden admin shouldn't listen to Fauci.
I'm old enough to remember when the pandemic started and the US still had packed NBA arenas full of fans, and Fauci said they should only be limited out of "an abundance of caution".
• • •
Missing some Tweet in this thread? You can try to
force a refresh
A few things he doesn't mention is that the Stage 3 trials only took around two months to reach statistical significance and could have been approved in early Oct. instead of mid-December, particularly minus the 5-6 week data review by the FDA.
Second, one could have done the same trials, but layered them a bit so that we get through them all a lot faster. I.e., start Stage 2 before stage 1 approval is complete, and take a large control group from Stage 1 and make it into part of the Stage 3 group, etc., so we get
Good news! I just got my first "Top 5" journal publication, in the Review of Economic Studies. Top 5 journals in econ are fetishized beyond belief. This publication helps show why we might want a more balanced attitude. #Econtwitterideas.repec.org/p/abo/neswpt/w…
In our paper, we find several coding errors which overturn a seminal paper in the "Chinese competition caused a huge increase in innovation" lit by Bloom et al (BDvR). From the beginning, I knew the BDvR paper had problems. Why? nbloom.people.stanford.edu/sites/g/files/…
Well, first I should say that I am actually somewhat empathetic about the coding errors, common in published papers--a reason we need replication. Any empirical researcher will make mistakes on occasion. But, there is a lot to chew on in this case besides the coding errors.
A bit of good news, after 19 submissions spanning six years, two months, two weeks, and four days, my job market paper on the collapse in US manufacturing has been accepted for publication in the European Economic Review. Here's a thread on the topic. 1/
The original title was "Relative Prices, Hysteresis, and the Decline of American Manufacturing", which later got shortened, with the most recent draft up on Ideas/Repec here: ideas.repec.org/p/cfr/cefirw/w… 2/
The tale of US manufacturing in the 2000s itself is an old bedtime story some economists tell when trying to scare their children. So, listen up. 3/
It's official, the Review of Economic Studies does not accept comment papers (unless they point out actual factual errors), according to my correspondence with an editor. In important ways, it's not really an academic journal. @RevEconStud#EconTwitter#EconReplicationCrisis
This isn't hard. If the RESTud doesn't accept comment papers, then people who do publish there now have an incentive to sneak papers past the referees which are obviously flawed. And people don't have incentive to replicate papers, since they cannot be published. #EconTwitter
It's no mystery why the editorial board refuses to accept comment papers. The benefit of being an editor is not the pride one takes in leaving one's intellectual stamp on the field, but rather the benefit one gets from giving favors to powerful people in the field. #Econtwitter