Our paper (ty @medrxivpreprint): Remdesivir @GileadSciences pricing justification & plausibility of Rem reducing hosp length of stay by 4 days (LOS). While ACTT-1 study showed 4 day faster recovery for Rem (median 11 vs. 15 days), does that translate 1/n? https://www.medrxiv.org/content/10.1101/2020.08.10.20171637v1

Does not seem so as we show. First, ACTT-1 time to recovery calculation excludes pts expected to stay < 72 hours, and patients who died. But EUA of Rem doesn't exclude these ppl. Median recovery time in placebo arm of 15 days just seemed too long to us. 2/n

We asked in a cohort that did not get Rem, just how many pts stay less than 5 days, and how many 5-8 days. The first would need a LONGER stay to get 5 day Rem course that is authorized, second cannot get both full course and reduced LOS by 4 days or more. 3/n

And, in our cohort, median LOS 7 days (not 15). Majority of pts (across ages) have a LOS that would either need to be lengthened or instead is not long enough to be shortened by 4 days. $GILD's justification for Rem pricing does not hold up to actual severe Covid LOS data.

cc the first author (whom I just realized is on Twitter and worth a follow): @RestivoAnderson