Lmao at an exception for college football.
2 Likes
3 Likes
My sincerest hope is that governors decide to lock down for a few weeks to get Covid under control pre-vaccine. We are still looking at an extremely gruesome next 3-4 months if not regardless of a vaccine. Dying from covid right now would be ultra dumb with a vaccine imminent.
3 Likes
If it really has 90% success, (and assuming it’s safe), does it matter if morons refuse to take it? They can all Darwin themselves for all I care.
I think if the 43000 participants were distributed properly geographically, and randomized properly for vaccine versus placebo, it’s maybe not enough to say 90% with great certainty, but it seems plenty enough to know that it works very well. But let’s let our stat nerds comment.
The 90% efficacy claim seems a bit misleading. That or I don’t understand what efficacy means. In my mind it would mean 90% of people who took it were immune. What they are actually saying is that, over a small time period and number of Covid infections (94) 90% of the infections were in the placebo group. Keep in mind this is over 43,000 participants.
So this is very good news but the headlines seem a bit misleading to me.
1 Like
Most universities are sending all of their students home for semester at Thanksgiving. Lock down for 2-3 weeks to start December, and get this thing under control before major holidays hit and everyone unwittingly kills members of their own family.
Seeing something in twitter about the vaccine needing to be frozen, too.
Hopefully Pfizer is able to back up the splashy announcement and we really have something here.
1 Like
Well I don’t know how to do the exact math, but if it didn’t work at all, you would expect the infections to be equally distributed between those getting the vaccine and those getting the placebo. (I know you get that.) How they mathematically determine efficacy based on the results, I’m not sure. But it seems reasonable to assume that Pfizer knows how to do the math.
1 Like
The vaccine needs to be stored / transported at -80c (-112f)
Pfizer believes it will be able to supply 50 million doses by the end of this year, and around 1.3 billion by the end of 2021.
The UK should get 10 million doses by the end of the year, with a further 30 million doses already ordered.
However there are logistical challenges, as the vaccine has to be kept in ultra-cold storage at below minus 80C.
Arghh, easy for NA/EU to manage cold chain these days, but kind of bad news for developing nations. Edit: actually -80 for the scale needed is no easy feat anywhere.
Hmm. My guess would be nobody’s normal supply chains are equipped to to that on a scale of hundreds of millions of doses. But I have no idea.
Oh I know. I’m just wondering if that difference is actually statistically significant. The tiny sample of infections is concerning. Which of these would be more significant(assume 43000 participants):
4% infecfed placebo, .4% infected vax
.4% infected placebo, .04% infected vax.
Obviously the first is much more statistically significant right? Because the sample of pozzed is much much higher? Or am I still sleep deprived from the election? In other words if you had 940 pozzed instead of 94 the certainty of the outcome is way way higher.
I think that’s right. But I have to believe the results are statistically significant within the design of the trial or else it would be kind of dumb to announce it, as some very smart people are going to be digging into this.
The sample size is 21.5k or so. 90 cases in the control and 10 cases in the vax group. 80 prevented infection out of 90 would-be infections is approximately 90% efficiency. The numbers are small, but that result is significant as all fuck and it’s an RCT, so there’s no need to worry about weird confounders.
EDIT: I ran it through an online calculator, and it said it was Z=8, which is pretty much p=0.000000
1 Like
My point is when you get down to these low probability events (94/43000) the variance is going to be large isn’t it? You could easily get 10/21500 or 84/21500 in a simulation if you ran it a bunch of times right?
Its good news but, and I dont want to rock anybodys world here, sometimes large corporations over promise.
1 Like
Nope. You would get a split as extreme as 60/40 (assuming no effect) 5% of the time. Never 30/70.
It’s similar to how the second 737MAX crash proved the plane was defective. One time could just mean someone used their one time. Two times is proof. Here, there are 80 extra cases in the control group to account for. It can’t just be a fluke.
3 Likes
At my university at least, this was the plan before classes convened in August. The idea was to prevent having students scatter to the four winds, be exposed to everything under the sun, then come back and infect each other along with the staff and faculty.
This is apparently the first widespread mRNA vaccine, so that’s interesting. Probably going to be a while before they can manufacture enough to be really useful.