theConversation.com: we've been testing therapy like it's a pill and patients are paying the price

B

bobbler

Senior Member (Voting Rights)
theconversation.com

We’ve been testing therapy like it’s a pill – and some patients are paying the price

Why the NHS keeps offering you the same therapy. And why the science behind that decision is shakier than you think.
theconversation.com theconversation.com

This isn't about ME/CFS and I have placed it here because this is where the IAPT (now 'talking therapies' or something like that) thread is located and because it is implicitly referenced in this and in the fb post a lot of comments mention IAPT

Click to expand...


For those not on fb the only text with the link is:
Why the NHS keeps offering you the same therapy. And why the science behind that decision is shakier than you think.
Click to expand...

But I think it is really worth us having here because it doesn't necessarily exclude ME/CFS and noting other threads some seem to be caught up in this bigger fallacy of 'these things must work' etc.

The author is:
Sahanika Ratnayake
Postdoctoral Researcher, Evidence Synthesis, University of Manchester

And here is where I want to make my point which is that the argument taken is that the issue is that they have been doing randomised controlled trials - I'm pretty sure the standards used in other conditions for CBT 'trials' are little better than what we see as an ongoing treadmill gravy train here. SO someone might want to get in touch with the conversation or the author (and maybe some commenters)....?


PLus also..... please do discuss the article itself and why it has been written too of course! - because I think it might be interesting to see what comes up in this from the perspective of other angles
 
ME/CFS Science Blog said:
Thanks but I think this is a step in the wrong direction. It mainly argues that treatments like CBT shouldn't be tested in randomised controlled trials.
Click to expand...
Precisely what I’m trying to point out. Either it’s a switch and bait or they’ve not spotted the real issue (that the rct shows the wrong thing because it’s not the same as drug trisls)

If it’s the latter then their move to using iapts dsta under the justification they claim what’s been done a rct and are bad - well we’ve seen how gerrymandered that can be. And maybe pointing them here to the roll call of common issues bps perfected might help them (as they realise they are not ‘done like drug trials’ and the issue is the standards and oversight etc)

If it’s the former of course it feels like when eg Pete glad well and others tried to get the go ahead for a PROm being needed by saying the Chalder fatigue scale was bad (but we’re actually mainly using it to replace the physical function measure instead - because that’s the one they couldn’t get results for) in order to argue later that meant having something even more faulty on the same points was ‘a solution to their straw man’

But whichever way the author leans the fb comments are from a lot of professionals not just patients - some of whom might be interested in a follow-up to point this out going by what they write
 
Last edited:
An interesting aspect is that badly controlled or uncontrolled studies can provide very reliable evidence for what a treatment approach does not do. PACE was great at showing CBT and GET didn't work usefully. This essay, in itself, shows that the psychotherapy approach does not lead to self-understanding in terms of motives and goals - at least not for the therapist!!
 
ME/CFS Science Blog said:
Thanks but I think this is a step in the wrong direction. It mainly argues that treatments like CBT shouldn't be tested in randomised controlled trials.
Click to expand...

That reminds me about this article about psychology research:

Liie

Post in thread 'Miscellaneous Research Thread'

Some psychology-bashing which the forum might find interesting:

The Conceptual, Cunning, and Conclusive Experiment in Psychology Debrouwere, Stijn; Rosseel, Yves

The ideal experiment in physics must be conceptual, cunning, and conclusive. Adoption of these same standards in psychology has led to experiments that are uninformative and frivolous. We explain why we believe that psychology is fundamentally incompatible with hypothesis-driven theoretical science and conclude that this erodes the logic behind recent proposals to improve psychological research, such as...
  • Like

The article said:
psychology can focus on the descriptive, taxonomic science that forms the basis of disciplines like zoology, botany, mycology, and even meteorology.
...
In order to learn about the world we may explore and document the great variety of phenomena, to organize them, to see whether there are any obvious regularities.
...
try to find the conditions under which a particular phenomenon or mechanism will or will not show up, what strengthens and weakens it
Click to expand...

And this old quote by psychology professor Vincent Deary, cited in a bit by Brian Hughes:

The problem may well be that some of our treatments are too evidence based:

thesciencebit.net

"The problem may well be that some of our treatments are too evidence based"

Professors of psychology can be a strange breed. The problem they see is not bad evidence, but bad interpretation of the evidence (by *other* people).
thesciencebit.net thesciencebit.net
 
The idea that there is a better way to do this is obviously incorrect, but the issue of why some forms of therapy, like CBT, have become popular because they are so generic and formulaic that they are easier to loosely adapt into something that could be framed as a randomized trial is interesting. If only the author had not taken the wrong conclusion out of it, that standards need to be tightened, not loosened, that the application of this discipline needs to be scaled back, not expanded further, as in the case of IAPT.

Of course the notion that therapy has been tested like pills is incorrect. If only they were doing that, but they would mostly find that very little of what they do is actually useful. The methodologies of rigorous drug trials are simply not feasible in psychology, and even when it comes to some drugs it has a very flimsy use case, such as with psychotropic drugs, which rank at the bottom of the drug trial methodological hierarchy.

All of which makes even more suspicious the trend of forcing all of this into health care, to "treat" health issues, rather than behavioral or psychological disorders. All it takes is to arbitrarily redefine illness as a psychological disorder, which is too easy to do, and way too error-prone.

Really what this article is arguing is that scientific standards are impossible to meet in psychology, and that's true, but that should lead to a massive downgrading and removal of the creep into health care that has been poorly argued to be valid based on an incorrect application of an error-prone methodology.

Psychology is much closer to sociology. It's a statistical discipline that can define a basic average, models, but is mostly useless in particular cases. The problems all begin with the idea that it can be imposed anyway, that it doesn't matter if it meets those standards, simply because it can't. Downgrading standards further is one way of looking at it. The right way is to accept that very little of this meets quality standards sufficient for many of its applications, and that it's its use that should be downgraded.

Psychological counselling makes sense, in some cases. It can advise, but it shouldn't prescribe or be used to deny medical care, which has pretty much become its most important application.
 
'We can't meet robust methodological standards, so we must be allowed to lower them until we can.'​

Those standards were paid for in patient's suffering, blood, and lives. And still are.

They are simply admitting that they have failed to deliver by robust standards and have nothing, without admitting it, and demanding that they get to arbitrarily rewrite the rules and lower methodological standards until they have 'something'.

Oppose this with everything you have. Nothing will do more damage to us and this world than allowing this self-serving drivel to become the standard.

:mad:
 
Last edited:
Ferry said:
Psychotherapy research usually seems to be low quality, but there is apparently evidence that psychology as a whole is much closer to biology than sociology.
https://journals.sagepub.com/doi/abs/10.1037/1089-2680.8.1.59
Click to expand...
There are clearly parts that overlap, mostly the useful parts, and those are already covered within medicine.

Clinical psychology appears to be mostly the less useful area. The parts where they deal with behavior, personality, models, and so on. Basically what's applied to health care. Technically, psychosomatic ideology fits far more within psychology than medicine, even psychiatry, so it's not surprising that this part is now often called clinical psychology, since it doesn't benefit from knowledge of biology.

Not much different from economics. Some of it is closer to physics and natural systems. Other areas, not so much.
 
Merged thread

Summary​

Although randomised controlled trials (RCTs) have many merits and are suitable for interventions such as pharmaceuticals, there is a mismatch between RCTs and therapy. RCTs presuppose interventions that are standardised, specific, time-limited and measurable, characteristics that are less straightforward for therapy. Instead, a pluralistic evidence base is required for therapy.

A philosopher asks whether therapy should be evaluated like drugs | The British Journal of Psychiatry | Cambridge Core

A philosopher asks whether therapy should be evaluated like drugs
www.cambridge.org www.cambridge.org
 
Last edited by a moderator:
article:

We’ve been testing therapy like it’s a pill – and some patients are paying the price​

If you’ve had therapy, particularly if you got it through a public healthcare system like the NHS in the UK or Medicare in Australia, there’s a good chance it was cognitive behavioural therapy (CBT). Even with private health insurance, if you want therapy, the one you are most likely to be recommended is CBT.

This is quite strange when you consider how many types of therapy there are. Psychoanalysis is well known, but there’s also humanistic therapy, existential therapy and body therapy, to name a few.
Click to expand...
However, research suggests that CBT is not a always a good fit for people from black, Asian and ethnic minority backgrounds, those with learning difficulties and those with complex needs. Yet other types of therapy are usually sidelined – the NHS Talking Therapies programme explicitly aims to make 70-90% of their therapy provision CBT.

I believe the prevalence of CBT is not because it is uncontroversially better than the others. It’s a complicated story about how we evaluate whether something works, what it even means to say it “works”, how medical research is funded, and how decisions about which interventions to fund are made.

The randomised controlled trial​

So how do we decide whether a therapy works? The answer, in most health systems, is a type of study called a randomised controlled trial. These involve taking two groups of people who have a particular condition researchers are trying to treat or address in some way and giving only one group the treatment.

The people in the groups are chosen to be similar in terms of age, gender, race and other characteristics, and are randomly allocated to an “active” group that receives the treatment or a control group that doesn’t. The idea is that if you have two similar groups of people and the group that gets the treatment improves, then you can be more confident that it is because of the treatment rather than other factors.

In an article for the British Journal of Psychiatry, I argue that though randomised controlled trials might make sense for certain kinds of things, like testing new drugs, they are not always the right tool for evaluating whether therapy works. Randomised trials work best for treatments that are brief, replicable, measurable and where it is clear what you are targeting.
Click to expand...
theconversation.com

We’ve been testing therapy like it’s a pill – and some patients are paying the price

Why the NHS keeps offering you the same therapy. And why the science behind that decision is shakier than you think.
theconversation.com theconversation.com
 
This is plain nonsense. RCTs do not presuppose anything much, other than that a treatment is clearly enough defined for the trial to make a useful prediction about further use of the treatment as defined. You can have RCTs of surgical treatment, which can vary widely in what is actually done, with nothing very 'measurable'. You can have CRTs of flexible therapy. Whatever you like.

This is just another case of a humanities busybody talking through their hat.
 
I think it’s fair to say the sort of trial described is not the only way to test effectiveness of an intervention but we do need to test effectiveness and I’m not sure about some of the framing here. A lot of money is put into to talking therapies in the NHS and I think they have a place and a use, but equally lots of other things which need resources don’t get funding they need. This feels more like it’s trying to make a pitch for resources or a carve out for their special thing than a convincing case on different methods of assessing efficacy.
 
Last edited:
If the argument is that what matters is the personal interaction with the therapist then it is essential to take the identity of the therapist into account in any trial intended to be predictive. Moreover, what such a trial would predict is whixh therapist produced the best results.

So the sort of trials that therapists should be doing are, yes, randomised controlled trials of one therapist against another. A hundred patient should be randomised, 50 to each of two therapists and the improvement reported recorded. If one therapist comes out significantly better they should be designated 'best current treatment' and compared with a series of other therapists in turn. Once a group of top scoring therapists had been established they could then be used to standardise the whole profession.

Of course they wouldn't like this much.

I seem to remember that there was a trial that compared Trudie Chalder's unit with a unit in Holland or somewhere and came out rather less good. That is the sort of trial needed.
 
You must log in or register to reply here.
Back
Top Bottom