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

The results are actually very clear that the intervention didn't work. the confidence interval for the primary outcome, WSAS measured at 52 weeks had a 95% confidence interval of −3.44 to 0.48. It excludes what the authors themselves have defined as the minimum clinically important difference. None of the secondary outcomes showed a significant difference, even though patients in the control group received no intervention. Yet the authors claim: "We have preliminary evidence that TDT-CBT + SMC may be helpful for people with a range of PPS."

 

The linguistic problems seem odd. "May" indicates conditionality. Surely it is incumbent upon researchers to indicate in what conditions it "is".

 

The point is also this was a major trial funded after other trials provided "preliminary evidence." It wasn't I assume supposed to produce preliminary evidence but actual actionable definitive evidence. Which of course it didn't produce.

Has anyone seen a mininal clinically important difference figure for the PHQ-15 measure?

 

Yes, the protocol says ...

Makes no suggestion of WSAS being a secondary outcome at any point.

 

[my bold]

This seems to be their standard get-out-of-jail card - we didn't prove anything useful but we'll spin it so it sounds like we only ever intended to lay the ground for another (worthless?) study anyway.

 

They were always going to measure WSAS at the other time-points but designated the 52-week as the primary outcome. I'm not sure if the other time points would automatically be inferred to be secondary outcomes? Or just to have no defined category? And I'm not sure oft he same about the other time-points for the designated secondary outcomes. Were they all secondary outcomes? Or just the 52-week time point? Researchers sometimes seem loosey-goosey about whether something is a primary outcome measure (WSAS) or the primary outcome at a specific time point (WSAS at 52 weeks).

 

I thought a major objective of a protocol was to pin down such detail and so avoid ambiguity. Especially something so crucial as whether something is a primary or secondary outcome at whatever point in the process. If anything so important can be reinterpreted or misinterpreted (possibly deliberately) after a trial has started, then the protocol has failed in its objective surely.

 

that is, in fact, the intent. but it seems the format can still leave it hard to pin people down on some of these things.

 

Trial By Error: Null Outcomes Presented as Success in Yet Another CBT Trial from Prof Trudie Chalder
https://www.virology.ws/2021/06/19/...t-another-cbt-trial-from-prof-trudie-chalder/

 

Wow, quick! I literally posted that a minute ago.

 

Like you've never had a stalker before...

 

Honestly whoever gave approval for WSAS (or whatever acronym I won't even bother getting it right) being a primary outcome deserves to be fired, from a cannon, into the Sun, and also mostly out of a job, Sun cannon or not.

Seriously who are these clowns? Have they no pride in their work at all? They are just there day-in-day-out going through the motions as if none of this mattered? As if none of this had any real-life impacts? Who are these clowns and what is wrong with not only them but the people who decided they should be making any decisions of more importance than which condiment goes first in the sandwich someone ordered?

 

that's kind of important, don't you think? does the training require you to learn how to change things if something doesn't work out over and over? that might be difficult for some people.

 

I think this is their post-PACE grifting strategy. Everything is now 'preliminary' and 'needs more research'. Being 'definitive' is a huge problem.

PACE backfired because it was hyped and sold as definitive. As such, too many important people and institutions took note of it being obliterated and it has turned into a grinding, comprehensive, and public loss for them. Wouldn't want to repeat that with any other conditions.

Better to fly under the radar and collect funding for studies you never intend to mean anything. Clearly there is no shortage of gullible funders. You don't even need to manufacture a positive result, apparently.

It's less than they had hoped for, but it will pay the bills.

 

I think this needs to be set up on a thread of its own particularly given the latest developments in SNOMED and ICD11 and introduction of 'Somatic symptom disorder'.
https://www.s4me.info/threads/updat...-terminology-systems.3912/page-24#post-352019

 

from @dave30th 's blog

Why does the scientific establishment allow researchers to behave like estate agents? (thats real estate agents in the US i believe) The abstract's conclusion is like reading an estate agent's spun description of a dump that is literally falling down, as 'needs new wallpaper'. What the hell are they playing at? i feel furious with them for letting all this crap slide past, for YEARS, without most of them doing more than raising an eyebrow, even AFTER the problems are pointed out to them.

Thank God for the Dave Tullers & Jonathan Edwards', without whom we would be royally, comprehensively screwed, but sometimes it feels like they are King Canute I just despair at the ever flowing tide of crap that keeps flowing and drowning us all.

 

Interesting theory. They later claimed at some point that they didn't mean to claim that PACE was "definitive," and that they'd only said "definitive" one time--that it wasn't their overall opinion. Or something like that. No one funds major trials after small trials to get "preliminary results." It is very disturbing that the journals allow them to make these claims when the trial obviously did not produce the results they expected.

 

Riiight. £5m on a trial that wasn't meant to be"definitive", funded in part by the DWP.

If that's the case ,someone needs to be asking why the heck they were shelling out that kind of dosh.

 

Yeah I think after PACE their best strategy is to avoid any showdowns. As long as funders are gullible enough, they may well be able to get funding even for 'major' trials; then they just tell the funder what they want to hear, but otherwise keep pretty quiet - just publish the paper rather than the whole media campaign that came after PACE.

 

Found a really great Twitter thread by a Norwegian Psychology Professor called Jonas Kunst. He's got some very good points regarding the authors weak adjustment for multiple comparisons:

https://twitter.com/user/status/1406531704261595138