MrCynical
Deity
Because of the blind credulity by some here about pharma industry sponsored studies, I don't want to miss pointing out a recent example of what very much looks like a deliberately harmful study - how evil they can get. The excuse of incompetence is a sorry one. And there weren't even quick tens of billions of dollars to be made on this one.
As noted above, this isn't anything to do with Covid, but it does have some relevance to earlier discussions on (mis)understanding how scientific research works. Unfortunately you've skipped over the most relevant angle. When you have a string of studies consistent with the null hypothesis, and then get an oddball one that shows a large effect, the reaction as a scientist shouldn't be "wow, it works after all!" it's instead "what's off with the oddball study?". Sound familiar?
Now, I haven't read up a ton on fish oil, and honestly the whole supplement industry's relationship to science is sketchy at best. The linked article's suggestion that the placebo used in the lone positive study was actively harmful is one which should really be easy to verify by statistical analysis. Indeed, implausibly poor placebo arm results are something which should really have been called out in a competent peer review. We had a similar issue show up in an earlier thread on proposed Covid treatment proxalutamide.
From what's printed in the article, it ought to be possible to get clear numbers from already existing data on the various placebos used, vs general population, and see if the mineral oil is inconsistent with the rest. Or, as suggested, just re-run with a different placebo. I'd agree that lone studies sponsored by companies as efficacy tests for their sole product should be regarded as dubious, especially when they give results that are "against the grain" compared to more independent studies. Do you have any studies relevant to Covid that follow this pattern, yet are carrying any scientific weight?