Phoenix Rising tells QMUL: release the PACE trial data
Mark Berry, Acting CEO of Phoenix Rising, presents the Board of Directors’ open letter to Queen Mary University of London (QMUL) urging them to release the PACE trial data, and hopes that other non-UK organisations will join British charities in the same request...
Discuss the article on the Forums.

Randomised Controlled Trial of Cognitive Behaviour Therapy Delivered in Groups of Patients with CFS

Discussion in 'Latest ME/CFS Research' started by leokitten, Sep 27, 2015.

  1. leokitten

    leokitten Senior Member

    Messages:
    579
    Likes:
    1,875
    Maryland
    http://www.ncbi.nlm.nih.gov/pubmed/26402868

    For the umpteenth time I would like for researchers to do an RCT of CBT in other physiological chronic and fatiguing illnesses. They will find the exact same "improvement" when doing CBT on SLE, MS, RA, etc.
     
    Last edited: Sep 27, 2015
    Hutan, ukxmrv, Cheshire and 8 others like this.
  2. John Mac

    John Mac Senior Member

    Messages:
    308
    Likes:
    1,471
    Liverpool UK
    "We conducted a pragmatic randomised controlled trial with 204 adult CFS patients from our routine clinical practice who were willing to receive group therapy."

    So they ruled out non believers.

    "Thirty-four (17%) patients were lost to follow-up during the course of the trial. Missing data were imputed using mean proportions of improvement based on the outcome scores of similar patients with a second assessment."

    Then they ruled out those that became non believers, but replaced their results with positive ones.
     
    cigana, Sea, mango and 8 others like this.
  3. Dolphin

    Dolphin Senior Member

    Messages:
    10,557
    Likes:
    27,551
    *For the group who had multiple sclerosis and were fatigued before treatment, the results were statistically better than the non-fatigued healthy comparison group after treatment and at 3 and 6 months posttreatment!
     
    Hutan, Mij, Valentijn and 6 others like this.
  4. Sidereal

    Sidereal Senior Member

    Messages:
    3,074
    Likes:
    17,021
    Partying like it's 1995 not 2015. This method of handling missing data is based on the unacceptable assumption that the 17% of the sample who dropped out of the study would have had the same level of improvement as "similar patients" had they stayed in the study. What if they were getting no better or deteriorating and dropped out as a result? I take it they don't have linear mixed effects or multiple imputation in the Netherlands? :lol: If you're going to do a shitty obsolete imputation method they could have at least had the decency to use last observation carried forward (not that I would ever recommend doing this) which in this case would have given a more conservative (realistic) estimate of treatment effect.

    I wanna know if this was measured by ticking boxes on questionnaires or actual physical functioning in the standard dictionary definition of the word functioning reflecting, you know, real-world activities like being able to go out of the house, working, getting off welfare etc.
     
  5. Valentijn

    Valentijn The Diabolic Logic

    Messages:
    14,075
    Likes:
    43,811
    With these people, it's always a questionnaire. They've learned not to use objective measurements, because it reveals their therapies for the useless bullshit that they are.
     
    Marky90, jimells, Hutan and 6 others like this.
  6. Sean

    Sean Senior Member

    Messages:
    3,177
    Likes:
    17,268
    This is the critical bit of knowledge they've gained from all these studies: how to selectively massage methodology
     
    Last edited: Sep 28, 2015
    jimells, ukxmrv, Chrisb and 3 others like this.
  7. Valentijn

    Valentijn The Diabolic Logic

    Messages:
    14,075
    Likes:
    43,811
    The trial lasted 6 months, and was largely focused on brain-washing which targeted the same questions that are asked on the primary questionnaire (Checklist Individual Strength):
    And to boost the brainwashing power:
    "Clinically Significant Improvement" was defined as CIS fatigue score less than 35, SF36 Physical Function score (PF) of 65 or greater (out of 100), and reducing Sickness Impact Profile (SIP) score by at least 1 point. In the case of the CIS, entry score could be 35 or higher, so only one point improvement is needed in some cases, out of 56 possible points. The SIP entry score of 700+ had to fall by at least one point to be under 700, out of 5,799 possible points. No SF36-PF threshold was established for recruitment, so patients may have already started with a score of 65 or higher.

    I think they've repeated a PACE error here in applying standard deviations to scores which are not normally distributed (not a nice bell curve result in the general population):
    They've previously defined CIS scores of 8-26 to be normal, 27-34 as moderate fatigue, and 35-56 as severe fatigue. I'm not sure which SIP scores indicate normalcy, since I can't find any data regarding the weighting which they use.
     
    Last edited: Oct 6, 2015
    Dolphin, Esther12 and Sidereal like this.
  8. Valentijn

    Valentijn The Diabolic Logic

    Messages:
    14,075
    Likes:
    43,811
    Compared to waiting list controls, 35.3% of treated patients improved and 13.9% recovered. They're not telling us if those who recovered already had SF36-PF scores at or above 80 at the start of the trial.

    The biggest problem with the study, discussed a bit previously in this thread, is that the scores from drop-outs were imputed based on the scores of "similar" patients. They cite http://www.bmj.com/content/342/bmj.d40 for using this method, but that paper focuses on biological measurements carried forward to a final endpoint based on other patient endpoints when the missing patient displayed a similar progression at earlier assessment points. But in the current study, there were no intermediate assessment time points - just at baseline and at 6 months when treatment was finished.

    So the data could not be extrapolated in the manner described by the paper which they cite. Yet they still manage to manufacture data for 15 patients who were never treated at all, after being randomly assigned to treatment groups. This is beyond bizarre ... who needs a sufficient number of patients to get statistically significant results when you can throw in some imaginary patients and pretend that they do the same as similar patients? Hell, why not just use 20 actual patients, and give them each 10 imaginary patients to generate identical scores for? :confused:

    In total it looks like they lost 60 out of 136 patients (44.1%) in the treatment groups between randomization and follow-up. 34 of those were lost before treatment finished. 10 of 68 waiting list controls (14.7%) were lost between randomization and followup. Yet the full number of originally randomized patients was used in all calculations.

    They also don't report any follow-up data, because the waiting list control group got treated in the meantime. A really stupid mistake, or a deliberate attempt to bury data indicating deterioration after the brainwashing wore off and reality couldn't be ignored any longer.
     
    Last edited: Oct 6, 2015
    Dolphin, Esther12, Scarecrow and 5 others like this.
  9. Sidereal

    Sidereal Senior Member

    Messages:
    3,074
    Likes:
    17,021
    It's even worse than I thought.
     
    Dolphin and Valentijn like this.
  10. Dolphin

    Dolphin Senior Member

    Messages:
    10,557
    Likes:
    27,551
    No objective measures reported e.g. no actigraphy, employment data, etc.
     
    Sidereal and Valentijn like this.
  11. Dolphin

    Dolphin Senior Member

    Messages:
    10,557
    Likes:
    27,551
    A post hoc analysis, for those that don't know, means that the analysis was not preplanned i.e. included in the protocol. They could have decided at any stage to do it including after they saw the data (as far as I know).
     
    mango and Valentijn like this.
  12. Dolphin

    Dolphin Senior Member

    Messages:
    10,557
    Likes:
    27,551
    Here's the protocol they refer to. On a quick skim, it looks like they have reported everything.

    http://www.isrctn.com/ISRCTN1582371...=1&page=1&pageSize=10&searchType=basic-search

    ISRCTN15823716 DOI 10.1186/ISRCTN15823716
    The effectiveness of cognitive behavioural therapy in groups for patients with Chronic Fatigue Syndrome (CFS): a randomised controlled study
     
    Valentijn likes this.
  13. Research 1st

    Research 1st Severe ME, POTS & MCAS.

    Messages:
    681
    Likes:
    2,188
    Psych CFS papers from self proclaimed 'experts' always lift my mood.

    The hypothesis is always a work of comedy literature, as is the outcome.

    With no objective measures necessary to prove their 'therapies' work (Science) in these clinical centres of excellence of fatigue, at least there won't be any shortage of comedians in the area out of work.
     
    Valentijn likes this.
  14. Dolphin

    Dolphin Senior Member

    Messages:
    10,557
    Likes:
    27,551
    An example of where the authors are coming from:

     
  15. Marky90

    Marky90 Science breeds knowledge, opinion breeds ignorance

    Messages:
    1,156
    Likes:
    4,203
    This junk makes me angry.
     
  16. Tom Kindlon

    Tom Kindlon Senior Member

    Messages:
    1,358
    Likes:
    8,111
    For what it's worth, I just posted a very quick comment on PubMed Commons: http://www.ncbi.nlm.nih.gov/pubmed/26402868#cm26402868_12202


     
    Last edited: Oct 6, 2015
    mango, Sidereal, Valentijn and 3 others like this.
  17. ScottTriGuy

    ScottTriGuy Stop the harm. Start the research and treatment.

    Messages:
    1,064
    Likes:
    4,250
    Toronto, Canada
    Is there already a handy list of disorders 'diagnosed' by allo doctors as psychological and later proven to be biomedical diseases / disorders?

    Off the top of my head:

    Autism - was the mother's fault
    Ulcer - was by psychological stress

    others?
     
  18. Research 1st

    Research 1st Severe ME, POTS & MCAS.

    Messages:
    681
    Likes:
    2,188
    On a serious note there are two tragedies here in psych CFS research self promoted as 'evidence based'.

    Tragedy one:
    Patients who don't have psych or Fukuda CFS but an organic inflammatory neuro disease. They will never recover on innappropriate therapies.

    Tragedy one:
    CFS being diagnosed as 'fact' in people mentally ill. Mentally ill who then don't respond to CBT and GET (see PACE trial), thus the psych therapy is actually useless for them too.

    So we have two sets of patients with ruined lives yet the insurance industry profits massively from this, by denying disabled people disability claims.

    As long as the 'evidence' on Psych CFS is published, promoted and utilised in clinics then the disability insurance never has to pay out to people in their homes, and the companies who would have paid out to patients who now need looking after (but can't get care). This saves billions of dollars as do the socialised medical health care systems in Europe who can all treat people with a therapy based on ideas, not science.

    It's a win for those with money, and a loss for disabled people with no power told to be silent by people who pay them a pittance to survive (welfare payments). Meanwhile there is no effective treatment as there are no large scale biomedical research and deleting Fukuda CFS in 'research' is taboo.

    I think we could try stroking kittens next for Diabetes. Propose a theory that listening to purrs lowers blood glucose levels, and thus patients won't need insulin anymore if cats can be let free to roam endocrine units. Would it be funded?
     
  19. Effi

    Effi Senior Member

    Messages:
    1,491
    Likes:
    4,548
    Europe
    MS - aka hysterical housewives disease aka faker's disease
     

See more popular forum discussions.

Share This Page