A general thread on the PACE trial!

https://twitter.com/user/status/1137455869958991878

 

I wonder if the topic of "not cheating to get the outcome you were seeking" will come up. Or maybe the topic of "our cherry-picked 'evidence' shows the exact opposite of what hundreds of thousands say but we'll stick to our beliefs, thank you very much".

The Replicable part is a bit amusing, since you can evidently replicate fraud many times without being caught. Very relevant to Bastien's post on Cochrane: "I've been doing this wrong for 30 years and by gawd I will continue doing it the same way no matter what because I'm right".

It's getting pretty clear that EBM is a massive failure because it still operates in the echo chamber where patients are irrelevant to the point of being excluded entirely, especially when it contradicts the researchers' beliefs. You can replicate all you want and show the appearance of success if you just cherry-pick what you like. In the end it's all supply-side medicine: by physicians, for physicians.

A whole paradigm shift is needed. What a waste.

 

There is some good evidence for this. This study below, from our Science library at this link, was a metanalysis of clinical trials that contained both blinded and non-blinded phases (all kinds of trials). They found that treatment-related self-reported improvements were greatly exaggerated in the non-blinded arms, when compared to the blinded arms. But there was no significant effect of blinding on observer-rated outcomes or objectively measurable outcomes.

This is a very important point, because many psychologists try to claim that the placebo effect is "real", that is, the patient is actually better. The're not. They're just biased in favour of noticing ad remembering things that align with what they know about their treatment. If you treat patients by using a placebo, not only is that dishonest and patronising, but you are writing a cheque you cannot cash. That initial feeling of hope that something might have actually worked eventually gets replaced with the crushing despair of realisation that it didn't.

 

Thanks @Woolie. That study is of relevance to my expert testimony to UK NICE. I could have made use of it I guess, although I sort of like the a priori approach to this. That is to say unblinded studies with subjective outcomes are by definition unreliable in the context of human nature. Nevertheless, having formal demonstration is useful to back up arguments if challenged.

 

I hope you won't be banging your head against a brick wall of people who have decided ME is subjective and that talk of objectivity is irrelevant, or possibly even worse,who accept objective talk for tactical reasons only.

 

I do find this interesting...

Moderator note: The rest of this post and the following discussion have been moved to a new thread: Placebo effect - discussion

 

So pleased that Caroline Struthers is going to this. My older son may well go for some of it as well.

 

I have pointed out in my testimony, and can reiterate it, that objective endpoints are not in themselves best used as primary endpoints, since subjective symptoms are what actually matter, but as objective corroborators of the subjective outcomes. I have given an example of how that can be done and ha not been done in ME trials.

 

Fair enough, I just hope they care.

 

That worries me a bit. To me and my wife one of the most crucial measures of her recovery would be to be able to walk 8~10 miles across Dartmoor, or the South Downs, or the Lake District, etc, each day, as she once could. Very objective and very valid surely.

 

To be 'able to walk' I would regard as a subjective outcome. To have actually walked would be the objective corroboration.

 

So long as those you need to convince are indeed convinced by this very fine-lined distinction, then hopefully OK. I'm finding it hard to see how being able to walk is itself a subjective outcome.

 

Being 'able to walk' is a a potential capacity that you judge and report - presumably self-report. Things like Chalder fatigue score include judged abilities like this. I probably was not that clear originally. Subjective outcomes are not only symptoms like pain, they are often things reported rather than measured. Of course there are further layers to subjectivity if there is an effect of motivation on actual execution of the walking, which is why we can expect a 6 minute walking time to be influenced by unblinding to test/control a bit. But a Fitbit that shows regular long periods of walking would be pretty objective evidence that the report reflected real improvement.

 

Yes, but not enough on its own. If she then reported that she had to spend the rest of the week in bed, it wouldn't exactly be considered a success.

 

Good point and one of the problems is that people do not see us when we are in rest-recovery or relapse.

 

My experience is that increased walking over years did not lead to a maintained improvement, and in that sense the "improvement" I showed over several years was not "real". Nor did execution of this competence abolish intervening hours of fatigue, generalised fatigue or neuromuscular symptoms, feelings of being on a"waking coma", foggy-flu etc. The self reported and hard to quantify remained largely the same but was/is real and confounding despite apparent "real" improvement over several years, which has not lasted.

Corroboration of self reported general improvement by an unintimidated/unpropagandised patient is fine. Use by authorities of fulfilment of certain competences to impute general improvement is quite different, especially if the patient is under pressure to concur that they have generally improved The corroborators can only suggest. They do not necessarily imply recovery and recovery must not be inferred on their basis, or be assumed to be lasting.

How any of this stuff plays out in terms of NHS provision can be so dependent on the benevolence/malevolence/honesty/budget of those involved. "You wanted emphasis on objectivity, fine. You walked X distance, there's OUR objective corroboration, YOU can't have ME if you did that, especially since we have just defined ME as an illness where you can't do that, but you may haver CFS............"

Dangers lurk. Whatever sword we may take up, there are still those who will point it back at us.

 

What is now clear to me is that I was insufficiently precise in my original post #650, and so things have veered away from what I was trying to get at with my original statement. So to step back and correct that, am repeating it here, but rephrased to avoid the ambiguity I inadvertently introduced ...

That worries me a bit. To me and my wife one of the most crucial measures of her recovery would be if, at the end of a holiday, she had been able to walk 8~10 miles a day across Dartmoor for a good few of the days we stayed there, with no more than normal recovery needed. Very objective and very valid surely as an outcome in itself.

 

To be fair, it does take years to acquire object permanence.

It's a tough requirement, perhaps, but I really would like research into my disease to take into account object permanence. I guess it's just too darn much to ask of some people, but I will keep on insisting for that requirement until it is actually taken into account. One day, perhaps.

 

Agreed my terminology was ambiguous. I was meaning it in the other well accepted sense - actually able to walk albeit only proven once successfully undertaken; not a preemptive perception. I didn't realise the ambiguity when I wrote it, hence my correction in earlier post.

 

I had to look the term OP up! Sorry for ignorance. I suppose all chronic, fluctuating conditions present this problem.
Another complicating factor for ME/CFS is that some do recover, but the pathways to recovery are unclear and exactly when/how/why a recovery has occurred is not usually clear. It would be interesting in pedeatric cohorts, where recovery is not unknown to do e.g. the 2 day CPET in illness, repeat in self reported recovery/remission, check in relapse etc. (Though I don't know whether 2 day CPETS are generally abnormal in kids to start with). This might also be a possible more robust test for CPET, to determine whether/to what degree a normal result in remission might predict stable recovery or whether relapse might occur.