Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Once you have evidence the treatment works, it becomes unethical to not offer it to the control group. This happens in a lot of medical trials.

In this specific situation, they would have been condemning through a choice to not treat, a fraction of the control group to die.



From the December FDA report: "The expected duration of study participation is approximately 25 months".

At the time they dissolved the control group, the fatality rate in control and placebo were the same. By their very own data they did not have a strong case that they would be "condemning...a fraction of the control group to die.". One would have to make a leap. A plausible leap, but a leap nonetheless.

Let's assume though that the fatality rate across the population for COVID was 1%, a very high number, to be conservative. So in that case you'd expect ~150 preventable fatalities in the control group. We now have ~that many fatalities, in a single state (Massachusetts), from COVID-19, in people who were "fully vaccinated".

You can do the math and extrapolate that there are now orders of magnitude more deaths from "breakthrough infections" than might have been in the placebo control group. If we had known that the efficacy of these vaccines was going to fall significantly, might some of those deaths have been prevented? I think there's an argument to be made there.

Now we have statisticians scrambling over themselves with complex methods to figure out what the hell is going on, when we could have had the gold standard which would be quite simple to interpret, if we had kept those control groups.

Now some argue it would have been completely impractical to maintain that control group amidst a mass vax campaign (I think we should have tried, regardless). I think there's room to debate that. But in that case I think we need a vast upgrade to our data infra, because giving up the placebo control group was a huge loss (for truth and science, at least).


> Once you have evidence the treatment works

That doesn’t actually seem right though. You have control groups at least through all phases of trials to make sure “the treatment works”. That’s literally the point of the control.

Especially now that we know the vaccines don’t completely work to stop infection or prevent symptoms or spread.

I’m all for the vaccines, but it seems crazy that they didn’t even wait until phase3 was over for a new type that has never been tested on humans before.

If the argument is “the control petiole were at risk”, they should have been told their vaccine was placebo and ask for volunteers to remain in control, seems like flat out offering them real shot defeats the point, doesn’t it?


But a good chunk of them would survive... and it would be really valuable to know why they survived without a vaccine.


But that's the crux of medical ethics of human trials. You can't treat people like mice, no matter how valuable the data gained would have been.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: