Comment by krisoft
3 years ago
> A simple way to stem most snake oil is to put all payments for snake oil treatments into escrow till the treatment is approved.
Even simpler is to prohibit payments for these experimental treatments, but let people who whould die anyway participate in them as long as it is free for them. In return the drug companies could use the outcomes as data points towards their aproval applications.
Why would a drug manufacturer do that? Why would they offeir their advanced but as of yet unaproved drugs for free? It is true it would cost them the manufacturing cost of the drug, but what they get in return is data on how efficient/safe the drug is. Assuming the drug works well they could use that data to prove their case and get approval. And if it is safe and has the desired effect they get their aproval and they could then charge for their drug as usual.
This sets the right incentives for the drug company. If they know they are selling snake oil they would be crazy to offer it for free, especially because all it would do is to prove that it does not work. If they think it can work they gamble with their own money. Which is good because the drug company knows the most about their own drug. The patient in general is not knowledgeable enough to evaluate the chances of an experimental therapy, but the drug company is.
> Even simpler is to prohibit payments for these experimental treatments, but let people who whould die anyway participate in them as long as it is free for them. In return the drug companies could use the outcomes as data points towards their aproval applications.
That would never happen for good reason. Drug companies would have every incentive to bias the results towards approval by giving the drugs to detain patients. I’m not exaggerating when I say this might be the worst idea I’ve read in this entire thread.
> Drug companies would have every incentive to bias the results towards approval by giving the drugs to detain patients.
What do you mean “detain” patients? Remember we are talking about people who have a terminal illness with no known cure. How can drug companies “bias the results”? Obviously I’m not saying we take anything they say as gospel, but in short order these patients are expected to die. If a drug managed to cure them that would be quite hard to falsify don’t you think?
> I’m not exaggerating when I say this might be the worst idea I’ve read in this entire thread.
Well, thank you. I aim to please. Are you sure that you understood the idea well? If you are, could you please explain with more words what is the misbehaviour you are worried the drug companies would commit?
> What do you mean “detain” patients?
I meant to write "certain".
> How can drug companies “bias the results”?
Choosing appropriate patients will lead to longer survival rate unrelated to the drug in question. The whole point of double-blind studies is to avoid this.
> Well, thank you. I aim to please. Are you sure that you understood the idea well? If you are, could you please explain with more words what is the misbehaviour you are worried the drug companies would commit?
I did understood you. You, however, don't seem to understand statistics or economics. Your idea directly incentivizes drug companies to choose patients more likely to make their drugs look better. It is a truly horrible idea.
3 replies →
What do you mean with "by giving the drugs to detain patients"? Is that a typo? I don't see how companies holding on to terminally-ill patients makes the trial results look more favourable. Are you suggesting that companies might not release the bodies of the patients that died during the trial?
I meant to write “certain”.
This plan seems to have excellent incentive alignment and basically mirrors the system in place, but without the need to be approved for clinical trials.
Experimental treatments being available to people with no options (there must be no applicable option with a proven probability of success or this becomes very questionable) could save drug companies millions-billions in the off chance that it proves to be incompatible with human testing or fails to produce the expected effects in humans.
I would take this up a notch or two for patients in a short-range situation that is certainly fatal, opening the door for them to try treatments that have just barely crossed en-vidrio and seem to have acceptable toxicity in mouse models. This would essentially be donating one’s body to science but in a much more useful form, with the side benefit of a slim chance of life extension.
This could benefit humanity by significantly reducing time to market for some drugs and cutting costs on eventual dead-ends.
I don’t think these people would be useful. Pharma needs controlled double blind studies to actually tell if the therapies work.
Of course, for actual efficacy, but you can tell if a treatment has a negative reaction or fails to effect the expected physiology in the expected way with a very small number of samples.
If everyone you give it to dies from reactions, or the expected effects seem to be happening on a physiological level, you don’t know much about efficacy but you do know a lot more about the eventual prospects for an actual clinical trial. It’s basically an extension of en-vidrio testing.
No, that does not set the right incentives. You have completely written off the harm of dangerous or deadly drugs.
The HepC drug Sovaldi brought in around $50B-100B in sales, but probably costs less than $50 to administer. Under your scheme it would make economic sense for a company to kill 1,000,000,000 people testing variants until they chance upon the right one because they incur no costs for administering deadly experiments.
The problem with your scheme is that human lives are valuable. A unsafe drug that causes harm or death is immensely destructive. If we want to be completely reductive, the actuarial value of a human life is around $10,000,000. A unsafe drug could kill the person resulting in a $10M loss to society. Therefore, the company should be required to escrow $10M per person until the treatment is done, paid out proportionally to any potential death or harm in excess of the gold standard treatment. Only then would the incentives be reasonably aligned.
This is obviously a very reductive treatment, but it helps illustrate the problems with unrestricted testing.
> You have completely written off the harm of dangerous or deadly drugs.
I was only talking about testing on people who we know are going to die fast because of their condition, and we do not have a way to save them in any other way. And naturally only with their consent. People who have a ticking clock and therefore willing to try anything.
This is the context the article in question talks about.
I think you forgot to price in the fact that some people live with what is effectively a time limit sometimes counted in months or weeks (like the person from the linked article).
> Even simpler is to prohibit payments for these experimental treatments, but let people who whould die anyway participate in them as long as it is free for them. In return the drug companies could use the outcomes as data points towards their aproval applications.
Hell I'd even approve my tax money to go to those companies which end up saving lives as reward.