I’m pleased to share my Commentary on the newly published *positive trial* finding metformin decreased Long Covid (Dr. Carolyn Bramante + @boulware_dr)!
“First, to our knowledge, this is the first high-quality evidence from a randomised controlled trial to show that the incidence of long COVID can be reduced by a medical intervention, metformin—an inexpensive treatment with which clinicians have ample experience…
“Second, the authors have, perhaps inadvertently, made an important contribution to medical epistemology…
This is the first randomized controlled human trial that has studied Covid-19 outcomes among *only vaccinated* people that we’ve ever seen.
That’s an absolutely astonishing fact, if you think about it.
(Assuming I am correct here. Am I wrong?)
This gives us an unusual chance to look at all-cause mortality and all-cause hospitalization among infected vaccinated people during the Omicron era...
And to compare that the pre-Omicron and pre-vaccine era...
If you aren’t seeing a disease whose true prevalence is ~1% of the population and has a 10 day course, consider the fact that if you see 20 patients in a day, your odds that none will have it is about 82%.
And yet 3m Americans would have it.
1/
This means that your small sample size is fooling your “I can’t think beyond what I can see” big brain energy.
Proceed with caution before making any declarations.
It’s to notice and even share it. It is not okay to tell people “it’s over.”
However… 2/
However, if, say, 2 or 3 of your patients (out of 20) are positive 10%-15% of your patients!) and that happens to you more than once this week, the asymmetry of anecdote becomes important.
Your anecdotal experience could have at least 3 possible meanings.
3/