Discussion in 'Other Health News and Research' started by merylg, Jun 4, 2013.
Please listen to what he says after the question at 5:05.
It's the same problem over and over again. In the end we never can be sure but compared to other approved drugs, this drug seems to have an outstanding profile. What happens next?
The FDA requires Phase 3 trials, which require the new drug to be tested against existing treatment. This will take several years and you can hear how annoyed this doctor is. No patient will have access to this drug. Even if you are dying, you won't get access to the drug. You will be treated with all the crap from the last century and if these drugs don't work...who cares...these are the rules. It's not only time to get rid of the Phase III criterium but it's also time to enable the patient to decide, what drugs he takes and what not.
Hi Waverunner, I don't agree with getting rid of Phase 3 trials, instead I prefer to take them out of the hands of pharma (who are highly biased) and into an official organization that does such research independently, with drug companies paying for it. This will also simplify the job of the FDA and so maybe help streamline the approval process. Indeed the FDA could take oversight of such a research organization right from the start.
However I do think that patients should be allowed to take drugs that have only passed phase 2 trials, provide the patient is aware of the risks and other options have failed.
So many drugs have passed, been approved, and sold, and then turned out to be killing people in hundreds of thousands to millions. Its usually not discussed, but its a big reason behind the push for evidence based medicine.
However if someone is dying from something that cannot be cured then any claim that denying the drug is for their safety is irrational. If I were dying from cancer I would want the option. What sane person wouldn't?
Yet evidence based medicine is quite weak in terms of assessing safety. David Healy has written a series of blogs talking about the issues associated with safety in RCTs http://davidhealy.org/ . I think he points out the one of the early RCTs was looked at thalimomide. One of his arguments is around side effects that are also symptoms (he talks of suiside rates for anti depresents). I would also worry about safety committees also dismissing adverse events as unconnected because they don't yet know about complex causal relationships.
I thought a lot of cancer patients end up on clinical trials particularly where other drugs have failed to work.
There are huge criticisms of EBM, and thats just one. Its a driving force, but EBM is badly flawed. A section of my book will be on EBM, its failures and possibilities.
Clinical trials of cancer patiets with new drugs are usually localized, involve a small number of patients, and drug access is restricted. Its not like we can just order the drug from the drug compay. The use is still highly restricted.
To me the biggest problem with EBM is it has no underlying conceptual or mathematical framework to talk specify what a piece of evidence is or how to combine pieces of evidence. Hence we get gross over generalisations, meta analysis comparing different things and individual pieces of evidence dismissed as anacdote (which is particularly significant from a safety perspective).
But there never can be. Such things are generally too difficult to reduce to numbers like that. In fact I think the mathematical issues are a lot less important than the other things. Statistics tells you how repeatable something is, but it does not tell you if the entire something is nonsense or not. EBM will get more rigorous over time, but will this be enough? I do not think so.
Lets presume, which is not the case but just for argument, that the PACE trial had overwhelming success instead of dismal failure. This would show (hypothetically) that CBT/GET worked ... IF the results were based on objective evidence, if it were replicable, and if the study design were sound. However it can NEVER show the underlying hypothesis is sound. That wasn't tested. Such cognitive hypotheses cannot ever be tested or proven ... and it matters naught if you improve the mathematics, or better specify the conceptual framework. Its a null question. Tackling with such issues is very hard to do in a formal conceptual framework, though there are simple tests that can be performed that help. Is the underlying hypothesis tested? Clearly NO. Therefore its not a sound study with respect to the underlying hypothesis. This is easy enough to say, but in many cases this is not going to be so easy to demonstrate, a point I am hoping to look into in the next few years.
My point is mainly that there are issues in medicine, particularly psychiatry, that cannot ever be formally specified or tested with any methodology to date. They still need to be dealt with ... usually by rejecting them I would hope. If you do this though most of psychiatry would disappear in a puff of logic, precipitating a global medical crisis.
Let me give a further example. Suppose they had a drug that could be shown to moderately treat conversion disorder in a phase 3 clinical trial. The trial was large, and the statistical significance good. Is this a valid test? How do you assess a therapy if the underlying condition itself is hypothetical, untested and objectively unvalidated? Do you even know what you are treating?
Some imprecision is always going to be required I think, but that does not mean it is immune to reason. Intense rational investigation is still going to be required. Further we still have to be able to assess the old evidence for many things, and sometimes that evidence is low grade, sporadic or suspect. How do we quantify that? We cannot reject half of medicine because its unproven, which is a common estimate of how much is suspect.
I have only just begun investigating EBM. I will be saying a lot as I investigate further.
Of course there is the thorny issue that almost nothing can be proved with science, only disproved. That is a whole other problem.
And I have to disagree with your first statement as well. In fact, your idea might even increase the cost of developing drugs and therefore push down the development rate further. The only ones who can afford these trials are big pharma companies, which lack competition because of high entry barriers for smaller companies.
If someone wants to take only Phase III approved drugs, well, no problem for me. But the folks who require effective drugs neglect the fact, that many people suffer and die because the drugs they need, never get developed because of too much requirements. A patient should decide, if he buys a phase II or a phase III drug. If a patient dies because he takes a risky phase II drug, it was his choice, if a patient dies because others prevented him from trying new drugs, it's not his choice. In this case other people potentially force death upon him. This should not be tolerated.
Waverunner, the biggest cost in drug development is failed drugs. The biggest cause of that is possibly entrenched bias in drug company research. By the time they realize they have a lemon huge sums of money have been lost. Putting the research in hands of people who do not stand to lose or gain a fortune from a drug, but who are otherwise competent scientists, could be a good move, and we could better trust that the drug trial results are at least fully transparent once they are released. Many drug results from big pharma are more spin than substance, and do not survive scrutiny many years later by independent researchers.
The notion of informed choice is nice, but its impossible in an environment when much of the information is biased. Its basically a licence for drug developers to market any old rubbish, and kill millions. As long as they follow the procedures, they would be covered ... genocide by drug pushing.
alex3619: There are so many lemons because we still lack the knowledge and means to develop highly effective drugs. In my eyes this has a lot to do with genetics and the molecular understanding of disease. This gets better right now, so I doubt that we need 10 years of development time (because of required trials).
A drug company has no interest to kill people, it would destroy its reputation and lead to bankcruptcy. Moreover you always forget the patient. The patient has to decide if he takes a phase II drug or not. He can stay on phase III drugs, that's no problem, or he can wait and see what happens to people who try it. Without government intervention, the market always has a solution because no patient would have interests to take bad drugs. New companies would be created, which only would check for the safety and effectivness of new drugs and supply patients and/or doctors with their information. We need responsible citizens, who start to get involved with their health, we don't need patients who act like sheep, just because they are too lazy to inform themselves.
I think without some kind of formal or semi-formal description of a trial then the hypothesis and the interpretation of result become very unclear. You say about statistics telling us if things are repeatable but as well as looking at the result we also need to look at the conditions under which the results were obtained (for example were doses the same, was the diagnostic criteria for choosing patients the same...). Without a degree of formality it is easy to ignore inconvenient issues.
On an aside and slightly more technical note where we use statistics we generally assume we know the distribution. There are evidential theories based on possibility theory which assumes bounds on our knowledge of the underlying distribution but not actual knowledge. Such work is consistent with bayes but allows for uncertainty in the distributions. Hence may provide a better way for dealing with things like measurement error or comparing experiments with slightly different hypothesises. To my mind one of the biggest problems with EBM is the inability to compare across different pieces of information.
I've generally found that the process of producing a formal model of a situation (ok never tried EBM) helps people understand what the actual situation they are trying to understand is. Prior to doing that people become unclear about their assumptions or are vague about the outcomes they are trying to reason about. In the PACE example it is unclear what the exact hypothesis is and hence what value any of the tests being proposed have in confirming the hypothesis.
We also have the situation where inappropriate abstractions occur. For example in PACE (and alot of other psyciatry) they abstract a questionaire into a scale without validating the properties they are relying on. (I know they do certain stats tests but I don't think they are enough),. Hence we get to the situation where they use the mean and standard deviation to describe results from the CFQ and SF36 but at no point within their work do we see the underlying assumptions and also they fail to talk about things like measurement error. I did come across a nice graph theory paper that talks about problems in going between different levels of abstraction and how many system failures happen this way.
Formalising things never solves problems what it does is introduce an element of rigour which makes it harder to gloss over issues and make huge leaps of faith in a chain of reasoning. All to often I read papers and think thats not a valid inference or I'm unclear about the exact hypothesis because vague language is used (I suspect on purpose to provide wriggle room to reinterpret results).
EBM does need to become more rigorous, I agree user9876. One of the issues though is the same as we find with ME definitions: there are too many viewpoints, agendas, goals, definitions. Its not really one thing, its a hodgepodge itself. There are the people who want more statistical rigour. As you pointed out the circumstances, including the nature of the data itself, need to be evaluated ... which is a big problem with PACE since at least some of the data is not a standard distribution. Further there are those who want this applied to medical management rather than diagnosis and treatment or fundamental research. Then there are those who want the whole thing to be put on a solid foundation of regular science. There are different agendas here, ranging from objective science to waffle, and reason to politics.
My guess is that EBM is going to be come Evidence Based Medical Management, Evidence Based Medical Science, and so on. It will fracture as soon as people realize that the different agendas are not reconcilable.
Wiggle room in vague hypotheses is of course the bane of much of psychiatry, especially psychogenic psych. It makes a mockery of statistical analysis and "evidence" based approaches to the whole EBM process. It makes a mockery of medical "science" as well. Indeed, all that is left is a need to mock psychogenic medicine itself.
Drug companies have already killed millions of people, and nobody said boo. I do mean millions. They made billions doing it too. Its par for the course. Allowing the current scientific rigour (which I think is still insufficient) gives them financial incentive to cut corners. The argument that they wont is flawed in that they already have and are unaccountable. It will just get worse. Its different if the company does something deliberately ... they have been repeatedly sued or fined for doing that, they just pay the fine and carry on with business. Drug companies even control the content of medical conferences, because they make sure (I have evidence of at least one such case) that the organizers know that certain topics are off limits.
This presumption that individuals always know best or will if they take responsibility is unfounded. First, people are not entirely rational. Second, they have trouble understanding the evidence. Even their doctors often don't understand it a lot of the time. Further, much of the evidence is hidden by drug companies themselves. The stuff that is published is what they want published. Nobody will pay for secondary companies to test these things, we can't get that done even when its a priority. This is privatization gone crazy ... there are limits because the system optimizes toward profit not public good. We are facing that at the moment, and the chance that democracy will survive this century is diminishing. A fully market driven state is not a democracy, nor can it be. The two big agendas in the western world, democracy and a free market, are both good agendas but they need balance.
Freedom to choose is not freedom if the game is rigged. Its the illusion of freedom with the reality of slavery to organizations and businesses powerful enough to set the agenda. This is where politics crosses the free market, rationality and freedom. You can't do all these things at the same time.
"Freedom to choose is not freedom if the game is rigged"
If only more folks understood this.....
alex3619: I fully agree, that millions of people died of drugs and medication. Question one is, were they forced to take these drugs? No, they weren't. And question two is, were there alternatives? I doubt, that there were many alternatives and I doubt, that without looking at genetics, it is possible, to know if a drug works for the individual patient or not.
I somehow believe, that you think, that with the right form of regulation, the drug companies are forced to offer good drugs. Is this correct?
No, that is not correct. Regulation is less important than transparency. Regulation presumes accountability, but without transparency accountability is a dream. Its not about regulation, its about organizational structure. The way things are operating I think they are not efficient and highly problematic. Short term profit is getting in the way of long term advantage, both to the companies and to the public.
Imposing regulation without doing the appropriate analysis is often a mistake. I think that the entire industry needs to be under more scrutiny. I also think that some of the existing regulation is bonkers. Its not a question of more regulation, or the "right" regulation, but about finding out what is the best form of regulation. That might in some cases be less regulation, or different regulation, or more regulation, depending on the specific case.
What you are describing with regards to patients taking drugs and dying, no they were not forced to take them obviously. It is likely that in many cases they were coerced to though, either through desperation, biased advice, or incomplete or wrong data. What I want to do is increase the reliability and reduce the bias in that data. That will lead to better information and better choice.
If all the doctors tell the same story, along drug marketing lines, and the published science is limited by much of the data being withheld, or the analysis biased by drug company marketing (which happens right now) then patients do not have a choice in any real sense. Its not much different to a con game, except its legal and the status quo. One of the things drug companies have to do, and are slowly being forced to do, is publish or at least release all their drug trials. I think there is resistance to this though. They cannot be allowed to run many studies and only release the favourable ones, which is currently legal.
Let me reiterate, regulation is much less important than organizational structure. The way the system is set up is not working. It needs to be changed. Changing it the wrong way could be a disaster. Analysis must preceed change.
You can also try a Google Site Search
Separate names with a comma.