There has been a lot of buzz around a new cholesterol-lowering drug called bempedoic acid. In the US, the drug is approved for people who are intolerant to statin therapy. Earlier this year, a major trial published in the New England Journal of Medicine
They will do anything to sell their drugs.
So I guess the lesson is that having a publication in JAMA is more beneficial to one's career/funding than the damage to one's reputation from flouting good statistical practise?
The pitfalls of subgroup analysis was demonstrated in the Bangladesh mask study which showed a larger variance in outcomes (Covid cases) due to differences in mask colour than there was with mask versus no mask groups.
Big Pharma is big business.
Vertical integration ("owning" the universities, researchers, medical journals and the media which publish the "findings" without any scrutiny) is the favoured business model.
Thanks for including the "absolute risk reduction "as well as the "relative risk reduction" in your reporting of this controversial trial. Do these statistics have a "confidence interval"?
The surprising point in your report is not that the "cherry picking" of results was not "picked up" by peer review but that the FDA gave approval for this medication based on a partially reported trial.
The lesson here is that approval of a "new medication" should wait until auditors [who have no conflict of interest] have themselves reviewed the raw data. Peer review should be concerned with the interest that readers might have in the article. It should not determine the decision of the regulator who should have access the the raw data and opinion of non conflicted auditors.
Its as if we ask car manufacturers to conduct multiple crash tests on a new model [as many as they like and in a way that they determine], keep all the results private, and report only the tests [or parts of tests] that they think are relevant. It's clearly ridiculous and a scandal that this is allowed to continue.
Our lawmakers need to "wake up."
Grateful for information warning us about issues like this. Thankfully I don’t have to take statins.
I would love to know if the LDL hypothesis is accurate. I have heard so many contradictory things about LDL.
McCormack said, “I think this puts Nissen in a corner. He does not address this in the paper because it requires an explanation that goes against the whole LDL hypothesis which is difficult for some researchers to accept.”
They should be required to advertise the relative risk and absolute risk increases for the side effects in the same fashion as the treatment goal. So for this compound--according to Wikipedia mind you--the increased relative risks of limb pain and gout are quite high. 😉. Imagine that commercial.
The product label summarises the FDA’s evidence base for its approval (In terms of raw data, the FDA doesn’t release it) https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/211616s000lbl.pdf
This subgroup analysis was probably an attempt to expand the use of the drug so that people at lower risk (primary prevention) take it…
Oh I see. So how did they decide to give approval? Was it based on the whole data sat?