This is actually an excellent point that is not made often enough. Most vaccine trials are done over years, so they have smaller numbers because more people get infected. The COVID-19 trials are VAST
We won't pick up on any long-term side-effects that only take place after 6-12 months, of course, because of the shorter time frame
Still, the trials are pretty amazing
Worth noting here that when I say "long-term" I really mean "only diagnosed after 6 months". By the time vaccines are rolled out, we'll have at least 6 months of follow-up data so anything that you'd expect to be picked up in that time-frame should show some signal
The real worry is chronic diseases that take years to be diagnosed, but that's not the same as serious chronic issues that start immediately after vaccination (which is what people often worry about)
And this sort of problem tends to be incredibly rare in terms of vaccines. I actually can't think off the top of my head of an example, although I'm sure they exist. Anyone?
• • •
Missing some Tweet in this thread? You can try to
force a refresh
2/n Paper is here. Basically, the authors took total COVID-19 mortality in a range of countries by August 31st and correlated them with country-level metrics such as average BMI frontiersin.org/articles/10.33…
3/n The authors found that country-level metrics compiled by the University of Oxford on the stringency of lockdowns did not correlate well with the number of deaths that a country experienced from COVID-19, but other things like BMI and latitude did
The corollary, of course, is that we won't see deaths attributable to the enormous rise in cases for another few weeks, and that hospitalizations from the record-breaking days of new cases are still coming in
This is all very rough, and since @COVID19Tracking aggregates all the different states there's a lot of variance. Different places report differently!
A really fascinating paper looking at contact tracing in England that provides some convincing evidence that testing/tracing reduces infections and deaths
The authors used the now-famous Excel spreadsheet stuff-up to perform a quasi-experimental analysis of whether places where no contacts were traced had higher levels of infections
They did
The estimates imply that, for every case that was not traced, there were an additional 25 cases down the track
In the dissection of the Year That Was COVID-19, one of the most depressing failures will probably be the vast, almost uncountable sums we've spent proving over and over that HCQ doesn't work
Glad that these results are getting published, but it's still astonishing just how much effort we threw behind a single drug despite many early indications that it was a red herring
We should never really have been in a situation where, simply because it was popular, most of the urgently needed research funds (and patients) went to research a single medication
If the U.S. maintains current infections (~200k/day) of COVID-19, and assuming about 17% of the country has already been infected, it will be about a year before the country reaches herd immunity
Compare that to vaccine rollout schedules
Calculations:
US pop: ~328mil
17% infected: ~60mil (per serology and @youyanggu)
40% (low) HIT: ~131mil
131-60 = ~71mil
~71mil/200k = 356
Add a few weeks for everyone to be ~immune~, and it's a bit more than a year
Yes, this is assuming the 200k number is reliable, which we know it is not. It is likely that there are instead 400k infections per day, which halves this number. It is also likely that the HIT is 60-70%, which doubles it 🤷♂️
You have to wonder what the "herd immunity through everyone getting sick" people are saying now that there is likely to be an effective vaccine beginning roll-out within weeks/months
I mean, the entire IDEA of letting everyone get sick was that a vaccine was a distant future that we couldn't rely on
Well, now it appears that we probably can
It would be the height of hypocrisy to not shift your message given that the foundation of the argument has now completely changed