I finished reading the Remdesivir trial

If you want my thoughts you will have to read this thread

[Thread]
The issue of outcome switching is laid to rest

It is a non-issue
Folks who have asserted (repeatedly and without evidence) that getting the drug early is vital will have difficulty with this bit
The drug is successful

Primary endpoint is a meaningful clinical endpoint and it is statistically persuasive.
It appears beneficial irrespective of baseline severity

Folks in this subgroup (7) will also of course ALL get the drug;

what doctor will deny a medication based on a subgroup analysis that could be severely underpowered?

ans - no one; even if there is a shortage
Mortality is not the primary endpoint, appears favorable to date, given the time horizon here, a little more follow up will be great.

Truly a situation where the tail will matter most.
We will need to see where it ends up in follow up
Most important, although the drug has efficacy and although that is great, the benefit is not miraculous.

It is still much better to not get COVID in the first place, or to not become very sick with COVID.

We still need more therapies, better therapies and a vaccine
The price better be fair

[end]
One addendum:
I would not assert that any subgroup doesn't benefit with a forest plot like this with broad overlapping CI, and a NS interaction test.

This fig. is c/w the idea that no subgroup clearly does not benefit
We have already seen ICU docs willing to give many totally unproven drugs to folks critically ill, if anyone seriously thinks patients on a vent will not remdesivir, they have another thing coming.

Until more data PROVES no benefit in any subgroup it will be given
Addendum 2:
Does this drug lower viral load?

That is a question I don't spend too much energy on. Inactivated or "dead" virus can be detected via PCR, but more to pt, who says efficacious coronavirus therapies generally also lower viral load? it is an unvalidated surrogate
Addendum 3:
Be sure to look at this way to consider the drugs impact. Blue good, red bad. Pretty consistent to my eye
Addendum 4:
@dvgbiostat pooled the chinese RCT (under-enrolled) and the NIAID study, and the combined result remains persuasive
You can follow @VPrasadMDMPH.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: