Read on Twitter
Some thoughts about conflict of interests in medicine. Let's imagine two drugs. Drug A is promoted by a pharmaceutical firm and received an approval based on a subgroup analysis of a subgroup of trials. Drug B is a repurposed drug with almost no evidence at all.
Manufacturer of drug A provides funding for a consensus meeting with all the "Key Opinion Leaders". This meeting will try to see wether drug A or B should be used in day to day clinical practice. https://www.bmj.com/content/336/7658/1402.long
The answer should be no for both drugs, because none has convincing evidence for recommending its use...
But, the "Key Opinion Leaders" were somewhat involved in the development of the product and in the subsequent "seeding trials" (e.g. ADVANTAGE) about drug A. Some have received grants and/or have other ties with the manufacturer.
https://www.acpjournals.org/doi/10.7326/0003-4819-149-4-200808190-00006#:~:text=Seeding%20trials%20like%20ADVANTAGE%2C%20in,patient%20injury%20for%20marketing%20purposes.
https://www.acpjournals.org/doi/10.7326/0003-4819-149-4-200808190-00006#:~:text=Seeding%20trials%20like%20ADVANTAGE%2C%20in,patient%20injury%20for%20marketing%20purposes.
By the way they have also "signed" gosht-written papers stating that drug A works well (in subgroups, on a surrogate outcome, etc.) and were invited to various "conferences" to present these results.
Of course, few of them have sufficient statistical literacy as "conventional training in statistics is insufficient, or even counterproductive, because it might give the user misplaced confidence." They are too often "led astray by confirmation bias". https://www.nature.com/articles/d41586-020-02275-8
As a consequence, drug A ends by being recommended for clinical practice by a panel of "experts" with a the highest Grade of evidence. But remember Drug A is just a bullshit.
To avoid this, it could be easy (ahaha):
- Do not lower the bar for approvals ;
- Make sure that experts are independent of pharmaceutical firms (such a list exists) ;
- Increase statistical literacy of experts ; https://jeannelenzer.com/list-independent-experts
- Do not lower the bar for approvals ;
- Make sure that experts are independent of pharmaceutical firms (such a list exists) ;
- Increase statistical literacy of experts ; https://jeannelenzer.com/list-independent-experts
Now let's see what happens to drug B. This repurposed drug is as bad as drug A. It is heavily pushed by certain doctors, and or patient groups with a very "seducing narrative": there is a conspiracy from "big-pharma" to make sure that this drug won't be approved.
In this narrative, of course, if used, Drug B (as it is stellar) would totally hamper the market of Drug A and the manufacturer of Drug B won't be happy at all. And of course, in this narrative, Drug A is a total bullshit (...but indeed this is true).
To understand what follow, you must understand that Key Opinion Leaders do not like when somebody notes that they are not independent and that their COIs may influence their judgement. See a response to our paper about conflicts of interests. https://www.bmj.com/content/367/bmj.l6015
e.g. In his rapid response Pr Schiele omitted to declare his own COIs.
Of course, we asked him to update his COIs and he did so. Importantly, he notes that "these links with the industry did not influence [his] judgment".
![Of course, we asked him to update his COIs and he did so. Importantly, he notes that "these links with the industry did not influence [his] judgment".](https://pbs.twimg.com/media/EgApYX-X0AEhnfq.jpg "Of course, we asked him to update his COIs and he did so. Importantly, he notes that \"these links with the industry did not influence [his] judgment\".")
I noticed that KOL frequently say that "one cannot be an expert if one doesn't have ties" and that "as one has ties with all industrialists, one will not favor one among the others". This is a nice narrative indeed... (ahaha)
Anyway, because they supported the use of drug A it is now difficult for them to say that drug B is a deep bullshit (even if it is indeed). Because they will be answered that they are too conflicted to talk about drug B.
Drugs B ends to be recommended with a lower grade of evidence that drug A. As drug A it should not have been recommended at all. And by the way, partisans of drug B still believe that there is a conspiracy because it has not received a high grade recommendation...
In other words, COIs are being used by partisans of dug B to make sure that almost all experts are neutralized.
This is the general example, but is seems to apply in many situations, with certain nuance :
- Drug A: Nalmefene
- Drug B: Baclofen
Other example :
- Drug A: Remdisivir
- Drug B: Hydroxychloroquine
Other examples are welcome.
- Drug A: Nalmefene
- Drug B: Baclofen
Other example :
- Drug A: Remdisivir
- Drug B: Hydroxychloroquine
Other examples are welcome.
But in my opinion, there is no conspiracy... There is only marketing, directed to small groups of selected physicians that are indeed very very easy to target.
We need independent experts who are able to tell that both DRUG A and DRUG B have not enough evidence for being used.
We need independent experts who are able to tell that both DRUG A and DRUG B have not enough evidence for being used.
Disclosure : I'm on @JeanneLenzer1 list of independent experts: https://jeannelenzer.com/list-independent-experts.
