Bear with me on this, I've only just invented this thought experiment, so it's quite likely to go pear shaped. A new drug is invented to treat ME/CFS. The people who invented it are sure it will work, and they have a biological rationale for the drug which they are proud of and have written papers about even before testing it on anyone. They are pinning their status and careeers on this drug. They design a clinical trial. Unfortunately it can't be blinded like most drug trials because every time someone takes a dose their big toe turns green for half an hour. They set up a pilot study. It's designed to last for 6 weeks. They explain the rationale to the patients and tell them it's a great new invention and they are lucky to be given the drug for free for 6 weeks, as there's a good chance if they follow the dosing regime of two pills a day carefully they will recover. 6 weeks later, on average the 10 patients questionnaire results show that they are a bit less fatigued and are able to walk a little further. The researchers publish this result, with the conclusion that they are sure taking the drug for longer will lead to even better results. They are confident they have found the cure for ME/CFS. So their government medical funding agencies give them lots of money to set up a bigger trial, and design it to last for 6 months this time, and they have a control group who are on the waiting list for the treatment. After about 6 weeks the results look mildly promising again, and they are confident patients will continue to improve. After 3 months some patients are complaining they are having more episodes of PEM and taking longer to recover from them. The researchers tell the patients not to worry, it's just a bit of a glitch, leave off the pills for a day then build up again to the full dose. By 4 months some patients are quietly cutting back on their pills, trying half doses without daring to tell the researchers, more and more are hitting the buffers with worse crashes. The researchers tell them they must be doing something wrong. Perhaps they are eating the wrong food, or mixing the pills with other medications. It can't be the pills stopping working. So most of the patients persevere, but a few are dropping out. By 6 months, those patients who have stuck rigidly to the full dose regime are feeling a whole lot worse, but most have quietly cut right back on their pills and are not so badly affected. No better than they were before the trial, but not much worse either. They are glad the trial is over, and worry they must have done something wrong, so they fill in the questionniares saying they are a bit better. Those who are much worse from following the regime rigidly aren't well enough to fill in the final questionnaires, so their results are never recorded. The researchers look at the results. They realise they need to make some adjustments to their paper on the study in order to be able to claim success, so they fiddle around with the figures and find by using a different way of calculating the results of the fatigue questionnaire, and adjusting downwards their limits for 'success' they can still claim the treatment works. They publish a paper with great public fanfare reporting they have the cure for ME/CFS. All patients need is two of their 'green toe' pills a day. They report no adverse effects, and don't include any data on patient compliance with the dosage regime. A couple of years after the release of the drug, patient organisations do a large survey of patients experiences with the drug and find that the vast majority report an initial period of a few weeks of mild improvement followed by serious deterioration. And even worse, those who stuck to the drug regime for more than 3 months are still very sick a year or more after stopping taking it. Cochrane has written a review supporting the use of the drug, and now refuses to withdraw it, saying there is not clinical trial evidence of harm. End of thought experiment. GrEen Toes doesn't work and harms patients.
It's hard to make a good thought experiment on this because the main factor is medicine's widespread and infinite belief that rehabilitation works since there's "nothing wrong (that they can see) with the patient", whereas in such a thought experiment for a drug there would not be that unwavering allegiance to it, mostly because most MDs genuinely accept that since they cannot find what's wrong with the patients, then there is nothing wrong with the patients. In truth it doesn't even matter what the treatment is, it can be group singing, arranging flowers, walking in a virtual forest, talking to your symptoms, homeopathy or synchronized astrology. So in this thought experiment it wouldn't even matter what is in the pill. It could be empty for all that it matters. It could be positively talking to yourself in the mirror while mimicking eating an empty pill. It does not matter. It's the infinite belief in the power of the treatment model that is the treatment. Of course making a good comparison here highlights the absurdity of the situation we are facing, but it's entirely modulated by the infinite belief in the conclusion-driven evidence process that decided decades ago, before any currently practicing MD was born, that generic rehabilitation, whatever works for the patient, is what 'works', after having defined 'works' as nothing significant, since they cannot find what's wrong with the patients, and to them it means a single Truth: there is nothing wrong and nothing will ever be found to disprove that. The whole model is a sham, driven by mass delusion in a culture that is unable to acknowledge its own failures, no matter how many times it fails the same way. You can't replicate that in a thought experiment, not any more than we could convert an entire highly religious nation to a different religion, or no religion, just to compare how it would change. And of course the nature of beliefs is that when you do believe in something, you usually cannot accept that what you are doing is merely believing. Instead it's The Truth, or obvious, or rational, or equivalent to reason. So even though psychosomatic medicine is entirely a belief system, the medical profession is not able to see it this way, because it was taught to them in medical school alongside the number of bones in the body or how respiration works. It's The Truth. You can't reason with The Truth. It's The Truth.
You make my point well, @rvallee. The thought experiment using a drug fails because the absurd situation we find ourselves in with GET would not happen in the same way with a drug treatment. If it were a drug, it would surely be taken off the market if it showed so little efficacy and there were multiple patient reports of long lasting harm. And Cochrane would surely withdraw it's review saying the treatment worked. But, as you say, nobody at Cochrane and in large parts of the medical world want to believe exercise can be harmful to people they don't believe are really sick anyway.
phenylephrine as a nasal decongestant. Looks like your green toe pill needs the FDA. I agree with your analogy pointing out that a therapy really does need as much rigour as a pill.
If I had invented the green toes drug and really wanted to know if it worked, I'd arrange to have a control cohort. I'd have people in both cohorts have their toes covered with an adhesive bandage that showed if it was tampered with, and have them video themselves each time they take the pill, to show that the bandage is there for the half hour afterwards. The real and control pills would be identical. If I wasn't that fussed whether it really worked and just wanted to increase the company share price, of course I wouldn't bother with that. A key issue is the willingness to tolerate a trial design with a lack of blinding combined with no rigorously objective outcome. If you understand that the trial is mostly useless if you have that design and you genuinely want to know if it works, you try a lot harder to make a better design that is not so subject to bias. People whose future income or reputations depend on a positive outcome may not be naturally willing to make a better design. Which is why funders and people who do meta-analyses have to clearly signal that sloppy trial design is not okay, that it doesn't result in rewards of fame and fortune. I still don't understand why many associated with Cochrane and the new review process don't seem to be prepared to say that an outcome that is subjective in an unblinded trial is unreliable. I've tried asking why they think this way, and still don't have answers.
And here's the thing: CBT is just a delivery mode, the excuse that it is anywhere near the green toe issue for the BPS is nonsense given how deliberately ambiguous they, and indeed all those pushing proliferation of dodgy versions or just selling-hard on it, have tended to be. Most on the street (and indeed probably most doctors - as they are just told 'it helps you cope' nonsense, which I find outrageous they don't enquire further on) wouldn't know the difference between experiencing counselling, a psychologist or CBT. Heck they could be half-hypnotising people to give the right answers. And as it is actually about the model underneath there is absolutely no reason at all that they couldn't use a pretend delivery mode of CBT to just provide something benign. Given what these people have done for years and that they care less than to check for harms the idea they could use the excuse of 'it's potentially harmful or a waste of time' is nonsense. I would assert it would need to include a load of bluster that directed at spoon-feeding primed answers on the questionnaire - say the Chalder Fatigue Scale, and a bit of instruction about sucking up to make people feel they need to 'give the nice lady a point for trying'. I would however care that it wasn't them who came up with it, but there is a generic version compiled by psychologists as a control version because we all know what they did with the pacing arm making them fill out intrusive over-detailed diaries etc. SO it could easily be manipulated to make anything seem better if someone were allowed to create their own.
I definitely want to know more about the synchronised astrology! You all make good (and serious) points, though.
