Efficacy of therapist-delivered transdiagnostic CBT for patients with persistent physical symptoms in secondary care: an RCT, 2021, Chalder et al

The spin on this is quite shocking:

https://www.cambridge.org/core/jour...rolled-trial/CEDD9B7597902C283BBB7CA7B43A81C7

Efficacy of therapist-delivered transdiagnostic CBT for patients with persistent physical symptoms in secondary care: a randomised controlled trial
Published online by Cambridge University Press: 31 May 2021

Trudie Chalder
Open the ORCID record for Trudie Chalder [Opens in a new window]
,
Meenal Patel
,
Matthew Hotopf
,
Rona Moss-Morris
,
Mark Ashworth
,
Katie Watts
,
Anthony S. David
,
Paul McCrone
,
Mujtaba Husain
,
Toby Garrood
,
Kirsty James
and
Sabine Landau

Abstract
Background
Medically unexplained symptoms otherwise referred to as persistent physical symptoms (PPS) are debilitating to patients. As many specific PPS syndromes share common behavioural, cognitive, and affective influences, transdiagnostic treatments might be effective for this patient group. We evaluated the clinical efficacy and cost-effectiveness of a therapist-delivered, transdiagnostic cognitive behavioural intervention (TDT-CBT) plus (+) standard medical care (SMC) v. SMC alone for the treatment of patients with PPS in secondary medical care.

Methods
A two-arm randomised controlled trial, with measurements taken at baseline and at 9, 20, 40- and 52-weeks post randomisation. The primary outcome measure was the Work and Social Adjustment Scale (WSAS) at 52 weeks. Secondary outcomes included mood (PHQ-9 and GAD-7), symptom severity (PHQ-15), global measure of change (CGI), and the Persistent Physical Symptoms Questionnaire (PPSQ).

Results
We randomised 324 patients and 74% were followed up at 52 weeks. The difference between groups was not statistically significant for the primary outcome (WSAS at 52 weeks: estimated difference −1.48 points, 95% confidence interval from −3.44 to 0.48, p = 0.139). However, the results indicated that some secondary outcomes had a treatment effect in favour of TDT-CBT + SMC with three outcomes showing a statistically significant difference between groups. These were WSAS at 20 weeks (p = 0.016) at the end of treatment and the PHQ-15 (p = 0.013) and CGI at 52 weeks (p = 0.011).

Conclusion
We have preliminary evidence that TDT-CBT + SMC may be helpful for people with a range of PPS. However, further study is required to maximise or maintain effects seen at end of treatment.

Keywords
Cognitive behavioural therapy (CBT)
medically unexplained symptoms
persistent physical symptoms
randomised controlled trial (RCT)
secondary medical care
transdiagnostic

Open access paper.

 

So this is related to the PRINCE trial, but not the PRINCE trial?: https://www.centerwatch.com/clinica...physical-symptoms-pps-prince-secondary-trial/

https://clinicaltrials.gov/ct2/show/NCT02444520

 

I think there are two PRINCE trials?

PRINCE primary (as in primary care) and the one above, PRINCE secondary.

 

That makes sense - I just read it as meaning a 'secondary' part of the PRINCE trial.

 

Still lumping different groups together. "It's all one condition, eh?" By what definition?

 

Defined by symptoms, symptoms aren't even the primary outcome and are only asked generically. What a bunch of crap. Any process that approves such ridiculous studies needs to be rebuilt from scratch, is completely unfit for purpose and frankly everyone involved has no business working in such roles.

 

good catch. classic outcome-switching

 

Started looking to see what the Journal’s policy on after the fact deviations from the original protocol is, but not up to reading the full text of their publication ethics policy and the COPE policies which they claim to be bound by ( https://www.cambridge.org/core/journals/psychological-medicine/information/publishing-ethics ), though I am assuming that the failure of peer review to pick up the outcome swapping resulted in the editors breaching their own publication ethics policy by publishing the article as it stands.

 

I agree. They are savvy and competent enough to know what needs to be in a proposal for it to be accepted. So they write something acceptable and then go off and do the work (not sure at this stage if they do the protocol and ignore parts for the write up or do half the protocol).

Either way what makes it to the report stage is what was 'meant' to be there all along.

 

https://twitter.com/user/status/1402280830542794760

 

I don't know. but it is certainly worth a letter bringing it to the journal's attention. These people seem incapable of doing studies without outcome-switching, made up end. points, and the like. Or claiming success based on failed primary outcomes but marginal secondary outcome measures. It really is mind-blowing.

 

absolutely.

 

Waw, this is really a textbook example of how to misrepresent results: highlight a couple in a long list of secondary outcomes that reached statistical significance, even although differences were very minor and not clinically significant.

Also: this study had an A versus A + B design. Patients in the control got no intervention, therapists were not blinded and subjective outcomes were used and still, there was no meaningful difference between groups.

 

And the WSAS is a 40-point scale. So even if the 1.5 difference for the primary outcome were statistically significant, it would be unlikely to beclinically significant. This is just trash. I'm going to write to the journal after I finish my post about Peter White's most recent GET defense in the Journal of Psychosomatic Research.

 

And yet more research is needed because they have some indications of improvement in vague secondary outcomes.

 

The authors themselves have defined a minimum clinically important difference for the WSAS at 3.6 points. So 1.5 isn't even close. They measured the WSAS at 9 weeks (difference of 0.19), at 20 weeks (difference of 2.41), at 40 weeks (difference of 1.32), and at 52 weeks (difference of 1.48).

At no point was the difference close to what they defined as the minimum clinically important difference (a difference of 3.6 points). Yet the authors were allowed to highlight the difference of 2.41 at 20 weeks because it reached statistical significance. I don't understand why the peer reviewers allowed this.