Invest in ME Conference 12: First Class in Every Way
OverTheHills wraps up our series of articles on this year's 12th Invest in ME International Conference (IIMEC12) in London with some reflections on her experience as a patient attending the conference for the first time.
Discuss the article on the Forums.

Clinical and cost-effectiveness of the Lightning Process for chronic fatigue syndrome

Discussion in 'Latest ME/CFS Research' started by Dolphin, Sep 20, 2017.

  1. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
    Free full text: http://adc.bmj.com/content/early/2017/09/20/archdischild-2017-313375

     
    MEMum, Woolie, snowathlete and 9 others like this.
  2. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
  3. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
    MEMum, Jo Best, Simon and 5 others like this.
  4. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
  5. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
    http://adc.bmj.com/pages/authors/#correspondence
     
  6. lnester7

    lnester7 Seven

    Messages:
    2,340
    Likes:
    3,318
    USA
    They came back with a vengence.
     
    MEMum and Orla like this.
  7. alkt

    alkt Senior Member

    Messages:
    180
    Likes:
    321
    uk
    how the hell do these frauds get taxpayers money for this non science I will say this once again weighted questions on biased documents will never give a true overvue of any trial/treatment once again they left out the use of any objective measurements . the people who sign of on these waste of time studies should have to justify why they give taxpayers money to enhance the egos of such individuals. fuming is an understatement whenever I read about so called studies based on the use of such ineffective data gathering. gps do not believe how ill their patients are but for some reason they will believe any old tosh if its based on one of these poor studies.
     
    Jan, MEMum, TeresaBCN and 8 others like this.
  8. Dolphin

    Dolphin Senior Member

    Messages:
    10,637
    Likes:
    28,067
    For those who are confused about the term "Trust" in this context, it refers to a charitable body.
     
  9. Esther12

    Esther12

    Messages:
    8,453
    Likes:
    28,550
    There's also this from the SMC:

    http://www.sciencemediacentre.org/t...t-for-children-with-chronic-fatigue-syndrome/

    eg:

     
  10. deleder2k

    deleder2k Senior Member

    Messages:
    1,104
    Likes:
    4,742
    The study will never reflect ME patient when three months of fatigue some light dizziness is enough to get a diagnosis...?

    Low study uptake. Did patients know that this was a LP study at first when they got recruited? My theory is that many of the patients that wanted to opt-out had read about LP, and though "this is not for me, I don't have depression, I have no psychological problems, I just suffer from fatigue, PEM and pain".

    Why not also add Jason (2006)? I simply don't understand why they call this a ME study.
     
    Jan, MEMum, HowToEscape? and 7 others like this.
  11. Esther12

    Esther12

    Messages:
    8,453
    Likes:
    28,550
    It seems likely that Tuller, Coyne, etc are going to be looking at this, so I've not tried to get myself to go through this in datail, but here are my notes from the other thread.

    I've really just ignored the subjective self-report outcomes. They seem pointless for a trial of LP.

    Dropout rates were lower than I expected, with them getting some data from a decent percentage of participants at 1 year follow-up:

    SMC: 49 participants ->37
    SMC+LP: 51 partcipants -> 44

    http://adc.bmj.com/content/archdisc.../F1.large.jpg?width=800&height=600&carousel=1

    But Table 3 has the data for school attendance, and missing data is more of a problem here. This is the closest thing that they have to an objective outcome:

    SMC
    six months: 37 participants
    twelve months: 36 participants

    SMC+LP
    six months: 41 participants
    twelve months: 34

    The difference was only significant at twelve months, when data was missing for 17 of 51 participants from the LP group.

    http://adc.bmj.com/highwire/markup/200532/expansion?width=1000&height=500&iframe=true&postprocessors=highwire_figures,highwire_math,highwire_inline_linked_media,highwire_embed

    For school attendace at six months there was no significant difference between groups, and I think that this was intended to be SMILE's primary outcome.

    This paper mentions the change in outcome:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4235039/

    They say this on SMC:

    I couldn't see any info on how many participants in the different arms made use of CBT or GET. That seems really important. Could anyone else see this information anywhere?

    It's possible that those in the LP arm were less likely to receive CBT/GET, so got all the positive narrative/response bias stuff, but avoided a lot of the worst stuff from CBT/GET - like how Tuller pointed out that FITNET could have been better than the control as the control often involved face to face CBT/GET.

    I've not been able to find out what was going on with CBT/GET provision in the different arms.

    The uptake was relatively low, so this could be an unusual sub-group of teens with CFS.

    Summary: This trial look pretty worthless to me, as we expected from the protocol. But it's going to be great for letting Phil Parker make lots of money from the desperate parents of sick children. Good work Mary Jane Willows, you deserve an OBE.
     
    Last edited: Sep 20, 2017
    Sea, Jan, MEMum and 17 others like this.
  12. Esther12

    Esther12

    Messages:
    8,453
    Likes:
    28,550
    Also... there's no open peer-review is there? I'd have loved access to that.
     
    Jan, MEMum, JohntheJack and 8 others like this.
  13. SleepyDormouse

    SleepyDormouse

    Messages:
    5
    Likes:
    34
    Is there any data on school attendance before treatment? I also note they only look at attendance in the week before which doesn't give a very accurate view. Especially as participants chose when they filled in the survey and have been given sessions on positive thinking. They could have selected a good week to write about.

    There should have been an ongoing record of school attendance
     
    Jan, MEMum, Dolphin and 11 others like this.
  14. IreneF

    IreneF Senior Member

    Messages:
    1,550
    Likes:
    2,551
    San Francisco
    I didn't know LP is related to NLP. Plus the LP cohort probably knew they were the "special" group. I wonder if they gathered any data on the dropouts.
     
  15. IreneF

    IreneF Senior Member

    Messages:
    1,550
    Likes:
    2,551
    San Francisco
    Creepy history related to LP:

    From what I've read, LP is highly manipulative. You are always supposed to say you're doing great, for example, and if you're not doing well, it's because you haven't been doing the process correctly, or not doing it enough. I.e., blaming the victim. So people who've spent a lot of money--or maybe a relative's money--or are simply true believers will push themselves, and if they actually have cfs/me they will crash. (Which may be why 1/3 of the kids in the LP arm of the trial dropped out.)

    I did some reading about NLP today. (It gets complex; I'm not going into all the details.) One of the influences on NLP was a speech pathologist named Wendell Johnson. When he was teaching at the University of Iowa in 1939 he and grad student Mary Tudor conducted an experiment on children who stuttered. Some of the children suffered from speech and psychological problems for the rest of their lives. Here's what Wikipedia says about it:

    "It was dubbed the "Monster Study" as some of Johnson's peers were horrified that he would experiment on orphan children to confirm a hypothesis. The experiment was kept hidden for fear Johnson's reputation would be tarnished in the wake of human experiments conducted by the Nazis during World War II. Because the results of the study were never published in any peer-reviewed journal, Tudor's thesis is the only official record of the details of the experiment.[1]"
    https://en.m.wikipedia.org/wiki/Monster_Study

    I don't entirely know what to make of this. I don't how much influence the Monster Study had on Johnson's later work, such as his book People in Quandaries, which is very, very similar to NLP; or at least the embryonic version of NLP I knew about because I was friends with a linguistics student who took classes from NLP’s developers in the 1970s.

    I can't imagine how Crawley's study got approved. Anyone with a free couple of hours could've discovered the Monster Study and the dangers of psychological experimentation (read: manipulation) on children. I suppose good intentions were had by all.
     
    MEMum, Jan, Woolie and 14 others like this.
  16. Valentijn

    Valentijn WE ARE KINA

    Messages:
    14,283
    Likes:
    45,804
    Perhaps the years of delay in publishing were down to having trouble finding quacks willing to endorse an LP study to any extent? :D
     
    Jan, MEMum, Hilary and 6 others like this.
  17. Philipp

    Philipp

    Messages:
    68
    Likes:
    470
    Couple of thoughts:

    - What is SMC? There is no known, reliable treatment for ME, so any treatment someone receives will probably be symptom-based and might involve a bit of guesswork. We know that even in groups that are less heterogeneous outcomes with medications vary quite a lot, and with just ~40 people at follow up it is really hard to know how the SMC arm 'should' look (i.e., how it would look over a larger sample size).

    - like Esther12 already mentioned, if you do not have the data on who received additional CBT/GET sessions this becomes even more of a mess. On a side note, I am not at all convinced physiotherapists are good at delivering exercise programs that have other goals than restoring function after injury and similar events. Training is actually remarkably complicated and extremely easy to do wrong in sick people, especially when there is no agreed upon protocol to follow and no one actually knows how the responses 'should' look (do we go beyond anaerobic threshold? do we restore movement patterns, reducing pain? do we just go for walks for 5 minutes so we learn that it is our horrible fear of going for walks that induces a constant state of panic? without good data, a lot can happen here that we will not know about).

    - The SMC release says:

    Not even this is true here if one were to take the sentence literally. This a bit more nitpicking than necessary I guess, and I know what they mean is that this is the only conclusion the trial design structurally allows (which is true), but since it is not actually known how the SMC arm was put together and it is, from what we seem to have available, theoretically possible that they made everyone worse in the non-LP arm by torturing patients with GET and give the people from the LP arm some form of medication that actually improved their symptoms a bit the trial is uninterpretable without a look at the actual raw data, assuming they even bothered to collect everything meaningful. If one has to control for even more splitting into subgroups because SMC as they defined it may be very random, the statistical power becomes laughable. This is a similar problem as every 'CBT'-trial ever has because talking therapies cannot be standardized the same way dosages of medication can. The sample size needed to control for fluctuations in different compositions of this 'SMC' would be huge.

    - the very possible pre-selection bias in recruiting for LP and recruiting 'ME-patients' with the criteria they used in general might skew the results

    - how do the methods they apparently used for drop-out work? We know ME is a relapsing/remitting kind of disease in many people. We cannot pre-select patients for this because we do not know whose illness would swing hard in the next 12 months and whose would not.

    - they changed their primary outcome measure from something kinda objective to something subjective in an unblinded trial. We've had that discussion, but it is especially bad here because the LP appears to be primarily about people rephrasing how they think about what is happening and, in my opinion, threatening them with actual psychological violence if they dare not to do so.

    - even school attendance is not that objective, because you can be convinced or force yourself to go even when you normally would not want to. Without somehow controlling for differences while at school this does not say much. Even comparing grades would not tell the whole story because the kids might put in a lot more effort to stay at the same level (this study excluded subjects who were too sick to attend the sessions, so probably no one was even at the true house-bound level in the first place. They do say mild/moderate.)

    - I am no aware if we actually know how the illness progression in mild/moderate adolescent patients with unspecified onset would look like over time without intervention. E.g., people with glandular fever induced ME-type symptoms are said to improve quite a lot anyway, but I am not aware if different triggers/subsets do so as well. So if one cannot control for this, especially with just about a hundred subjects, randomly assigning them to groups will skew the results in an unpredictable manner.

    - I suppose patients knew they were receiving a treatment that is supposed to cost almost a grand. This might increase placebo responses and induce a tendency to misreport actual outcomes based on how humans generally react to these things.

    - the effect size would need to be huge to infer anything here if you consider all of the above, even if one was assuming that the recruited population had the same underlying medical conditions and response to coercion was comparable among groups which is, as stated, already a leap of faith.



    and the big one:

    This is literally a trial about telling adolescents to ignore their symptoms and claim recovery while ignoring actual changes in health. They measure how well they did in an unblinded trial by asking them to fill out a questionnaire. This does not measure how well people do, this measures how well one is able to convince kids to say things one told them to say. If you want to draw any conclusion on actual changes in health from this, the effect size would need to be gigantic to be even remotely confident there might be any causation here at all.
     
    Last edited: Sep 21, 2017
    Jan, HowToEscape?, Hilary and 14 others like this.
  18. Jenny TipsforME

    Jenny TipsforME Senior Member

    Messages:
    1,124
    Likes:
    3,766
    Bristol
    Exactly. If it wasn't significant all it would show is you're really bad at delivering LP :rolleyes:

    And activity trackers are so cheap now, it almost makes you think they don't want to know about objective improvements ;)

    A shame we can't show LP is better than GET though, that would have been an interesting development.
     
    MEMum, Dolphin, Orla and 7 others like this.
  19. trishrhymes

    trishrhymes Save PR. Sack the President of the Board.

    Messages:
    2,155
    Likes:
    17,894
    The kids in the SMC group had GET or other things like activity management. The LP group had these plus LP. Such a mish-mash of treatments is yet another reason to say the study is meaningless.
     
    Sea, Jan, MEMum and 16 others like this.
  20. Sean

    Sean Senior Member

    Messages:
    3,248
    Likes:
    17,945
    The funding and ethics committees are just as liable for this as the BPS researchers.

    There is something very rotten indeed at the heart of the UK medical profession.
     
    Sea, Jan, MEMum and 17 others like this.

See more popular forum discussions.

Share This Page