Bruce Wampold: How psychotherapy works / The dodo bird verdict

MSEsperanza

MSEsperanza

Senior Member (Voting Rights)
Stumbled across the 'dodo bird verdict' while searching for how academic psychology has been acknowledging and handling the difficulties to assess benefits and harms from psychotherapies (due to the fact those interventions can't be blinded).

The claim is that all psychotherapies are equally effective.

Not able to post relevant literature at the moment so just leave some links here, to the wikipedia pages on the 'dodo bird verdict' and on Bruce Wampold who currently seems the most prominent protagonist of that verdict.

I could not find out how Wampold believes the alleged (equal) benefits of all psychotherapies could be measured as if I understood properly, he also states that RCTs are not helpful to generate evidence in the field of psychotherapy.

Sadly, of all the scientific papers relevant to the topic (and referenced in the Wikipedia article) I could not find a single one that's free access.

Not up to write / post more now but from skimming some paper abstracts I get the impression that even though psychologists participating in the discussion acknowledge that non-blinding would not be accepted in drug trials, there seems to be a lack of sensible conclusions from that acknowledgment.

To be fair, there are also some trials that tried to include 'official' placebo psychotherapies. Sadly again, the trial reports I found all were paywalled. From the abstracts and references I skimmed it seems there’s a consensus that, for psychotherapy interventions, ‘convincing’ placebos/ sham interventions don’t exist.

The idea though that trials that can't be blinded could at least use objective outcomes in order to mitigate difficulties in psychological research and improve both research and its assessment, seems completely absent in the discussion.

Would be interested to know whether 'the dodo bird verdict' discussion is well-known among health care professionals and especially professionals in the field of evidence assessment?

If so, is this something we need to take account and refer to when criticizing the poor quality of most treatment trials in the field of psychological interventions?

If not, are there still useful findings that need to be interpreted a bit differently than in the favor of 'genuine' psychotherapy as always being beneficial?


Some links:

Interview with Wampold (free access):
American Psychological Association (Ed): How Psychotherapy Works. 22. Dezember 2009
https://www.apa.org/news/press/releases/2009/12/wampold


For more sources, including academic literature, see wikipedia:

https://en.wikipedia.org/wiki/Dodo_bird_verdict


https://de.wikipedia.org/wiki/Bruce_Edward_Wampold (German language, has some additional sources).

Google translate of German wikipedia page on Wampold
 
Last edited:
MSEsperanza said:
The idea though that trials that can't be blinded could at least use objective outcomes in order to mitigate difficulties in psychological research and improve both research and its assessment, seems completely absent in the discussion.
Click to expand...
I have been saying for a long time that psychiatry needs to grow up and do this. Otherwise much of it has no scientific credibility.
 
Sean said:
he also states that RCTs are not helpful to generate evidence in the field of psychotherapy.

Yes, all that rigour and falsification is bad for one's eminence.
Click to expand...

But it all does not make sense to me as Wampold and his supporters seem to refer to RCTs themselves for the claim that all types of psychotherapy are more or less equally effective.

Their point is that it's not the treatment's specificity that's beneficial but common features like therapeutic alliance, patients' as well as therapists’ expectations etc.

Which strikes me as being features that are usually exactly those that are deemed to cause placebo effects and various other bias.

No idea whether the findings about common ‘contextual’ elements in psychotherapy are robust but they seem to be basically acknowledged also by the opponents of the dodo verdict – the opponents just don’t share the conclusion but refer to the high amount of RCTs that do report a superiority of specific therapies – mostly CBT – to back their position that specific content of the therapy and the targeted condition also matter.

So I don't understand how people on both side of the debate don't consider what seems obvious to S4ME forum people -- that all those findings most likely point to a bias due to using only subjective outcomes in treatments that can’t be blinded and not necessarily to an actual benefit.

I really should cite the sources to make my points -- again apologies for not being able to do that at the moment. So better you don’t rely on my musings and do have a look at the wikipedia pages and references.


Will try to add some sources later.
 
Last edited:
MSEsperanza said:
But it all does not make sense to me as Wampold and his supporters seem to refer to RCTs themselves for the claim that all types of psychotherapy are more or less equally effective.

Their point is that it's not the treatment's specificity that's beneficial but common features like therapeutic alliance, patients' as well as therapists’ expectations etc.

Which strikes me as being features that are usually exactly those that are deemed to cause placebo effects and various other bias.

No idea whether the findings about common ‘contextual’ elements in psychotherapy are robust but they seem to be basically acknowledged also by the opponents of the dodo verdict – the opponents just don’t share the conclusion but refer to the high amount of RCTs that do report a superiority of specific therapies – mostly CBT – to back their position that specific content of the therapy and the targeted condition also matter.

So I don't understand how people on both side of the debate don't consider what seems obvious to S4ME forum people -- that all those findings most likely point to a bias due to using only subjective outcomes in treatments that can’t be blinded and not necessarily to an actual benefit.

I really should cite the sources to make my points -- again apologies for not being able to do that at the moment. So better you don’t rely on my musings and do have a look at the wikipedia pages and references.


Will try to add some sources later.
Click to expand...

You are 100% correct in pointing this out.

I guess it all comes back to what they are actually measuring with the way they do their RCTs in this area (and the allowing the methodology to mean teaching to the test and drop-outs and so on) vs what they claim/think/believe they are measuring.

The fact that they all come back with the same results is probably due to leveraging of tools like biasing etc that they infer is 'the patient' in placebo effect, but is just building ability to bias things into trials.

Also, originally, and in their actual more scientific form, these therapies weren't supposed to be given to everyone even with certain conditions - but 'cause-specific' and tailored even within that. SO the nonsense of transdiagnostic stuff and claiming you don't need to diagnose properly and match good targeed therapies to the cause has always been silly and means they'll get lowest-common denominator+ whatever you can scratch out through building-in bias and limiting effectivenss of questions to walk your way to claiming effect by actually putting words in people's mouths.

I've come to the conclusion that certainly CBT has now become something used merely as a sales-person-type manipulate what people say tool (how they feel is different to what they are taught to say they feel, it is very hard to access the latter former when someone is spoon-feeding you the former latter, and for ME with its fluctuation, strange array of symptoms and adrenaline hinking maybe you feel better when you are actually over-threshold, combined with the risky situation most were/probably are still in re: vulnerability of health system and vulnerability to have words put in mouth and saying them just to get out of there, home and to lie down).

I have a horrible feeling this style has leeched into all areas embedding the handshake of 'so that's great you'll feel much better now, it's all just how you look at it etc' type con-tricks (closing the deal) into whatever therapy given they probably if not having individual therapist doing different types have trainers who do and therefore have allowed one to leech into the other re: the 'surrounding the therapy customer management techniques'.

I've seen this in how other HCPs have been taught to interact, obvlivious to the fact that post-appointment, ie when someone has actually got home, the domineering control of others' communication to remove ability to be honest and walk them into saying certain things leads to worse satisfaction and effeciveness vs the old being interested and honest. But at least those of the personality type or who are ill themselves, tired or worn and don't want to actually listen anymore will probably find that it removes that hassle and stress and 'sanitises' the patient interaction to just 'doing the motions' of the therapy - to me it is box-ticking - by not really hearing, and at least boundary-setting and limiting what someone can report back when asked 'how did it go after last week' etc.
 
Last edited:
Nightsong said:
Very few recent publications, I think. A more recent (2019) critique - this systematic review: "The misleading Dodo Bird verdict. How much of the outcome variance is explained by common and specific factors?"

https://www.sciencedirect.com/science/article/abs/pii/S0732118X18301351
Click to expand...
Thanks, had a look at and it mostly argues that because common and specific effects are correlated, it is incorrect to say how much % of the treatment is due to one over the other.

Even if that is true, it doesn't refute the dodo bird effect or why it has been so difficult to demonstrate that one type of psycho-behavioral intervention is more effect than another. I suppose the most likely explanation would be that trials on psycho-behavioral intervention are mostly measuring placebo effects, reporting biases and other things that have little to do with clinical effectiveness.
 
Also in the last 10 years - some data essentially in support of the effect, the 2017 IMPACT trial and associated commentary in Lancet Psychiatry:

https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(16)30404-7/fulltext

https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(16)30378-9/fulltext
Results indicate improvements over time for all three treatment conditions. However, the primary outcome of self-reported depressive symptoms, measured with the Mood and Feelings Questionnaire, did not differ significantly between groups at 36, 52, or 86 weeks post-randomisation, nor did the two hypothesised effective treatments (CBT and short-term psychoanalytical psychotherapy) lead to significantly greater benefits compared with the brief psychological intervention. Total costs of the trial interventions did not differ significantly between groups.
Click to expand...
The results of IMPACT raise questions about the effectiveness of all three treatments (combined with SSRIs in 36–41% of adolescents in each treatment group), and suggest that most psychotherapies will be “winners” on the basis of non-specific treatment factors, or alternatively that each treatment might contain distinct specific factors that contribute to recovery.
Click to expand...
 
ME/CFS Skeptic said:
Thanks for sharing. Doesn't seem to be much consideration that all three interventions may be losers rather than winners.
Click to expand...
They will consider any hypothesis or possibility. Except the ones where they're wrong. Having built a system in which the mirror always tells them that they're the prettiest in the land, never showing a real reflection and risk dispelling the illusion.
 
You must log in or register to reply here.
Back
Top Bottom