Table S6 in the Supplementary Appendix to Molnupiravir for Oral Treatment of Covid-19 in Nonhospitalized Patients article in NEJM (Dec 16, 2021) appears fabricated.
It shows nearly the same values of the mean viral change in the Molnupiravir and Placebo groups, about 100x lower than might be expected from the respective standard deviations (columns Mean change (SD)). Apparently, Merck has done this just enough to claim in the body of the article:
“On the basis of the available data, molnupiravir treatment was associated with greater reductions from baseline in mean viral load than placebo at days 3, 5 (end-of-treatment visit), and 10 (Fig. S6 and Table S6). Results at other time points were similar in the two groups.”
Work in progress
via Science Defies Politics
December 17, 2021 at 09:25AM