Thanks
@Jenny TipsforME
Your findings re gin are noted for future reference. I'm also making a note of that other study you mention - it looks useful when countering the view that CBT is helpful for ME.
The following is from 'Conclusions' and is talking about the primary outcome, the SF-36.
At 12 months, the physical score was in the normal range for 46% of the CBT group, 26% of the EAS group and 44% of SMC patients. For mental health score the percentages were CBT 74%, EAS 67% and SMC 70%. Of the CBT group, 32% showed at least a 15% increase in physical function and 64% achieved a similar improvement in their mental health. For the EAS and SMC groups, this improvement in physical and mental health was achieved for 40 and 60% (EAS) and 49 and 53% (SMC), respectively. The cost-effectiveness of the intervention proved very difficult to assess and did not yield reliable conclusions.
ie
......................................................................................CBT...................EAS.....................SMC
% with at least 15% increase in physical function.........32%..................40%.......................49%
% with at least 15% increase in mental health..............64%...................60%......................53%
So, although I like this idea,
Yes especially with CBT which could in a sense be seen as training to answer the questionnaire differently, whilst doing nothing about the underlying disease (or potentially making it worse with graded activity).
in this case the researchers weren't highly successful with the training in questionnaire answering.
We don't get to see P values testing for real differences between the treatment arms for these outcomes in the abstract. Probably it's mostly pretty random. But, if I had to make a story out of this, I think the evidence points to some people having a mental health benefit when they meet with other ME patients (feeling less alone and sharing tips for coping) but that CBT with a GET component hurts rather than helps physical function.
Tom has added a comment worth looking at on PubMed below the abstract pointing out some issues with the supposed improvement in the objective outcome (walking performance) seen in the CBT arm.
https://www.ncbi.nlm.nih.gov/pubmed/17014748