GETSET (white) in Lancet 22/06/17

[SilverbladeTE]

Just more evidence of deliberate, wilful, premeditated lying, fraud, abuse and systematic murder of ME/CFS patients by a group of corrupt, arrogant inhuman bastards with links to medical insurance companies and the government offices in charge of disabled welfare who have an economic and ideological desire to get rid of the Welfare state and thus GET RID OF THE LIVES OF THE DISABLED.

after the Grenfell disaster, and how it was an act of criminal negligence of the most obscene kind, and more to follow, it's about time the people of this nation rose up and dealt with these obscene, out of touch, greedy, sociopathic, evil, dangerous bastards once and for all.

 

[Esther12]

BMJ coverage of this is behind a paywall:

http://www.bmj.com/content/357/bmj.j3057

 

[RogerBlack]

[side question] we're currently working on a research survey for the Chronic Illness Inclusion Project. Which questionnaires/scales of do you think are worth the paper (or pixels) they're written on?.

Probably best to spin this off into its own thread.
It is for example, reasonable to try to measure depression at the same time, but many/most of the standard instruments for depression will read incorrectly for someone who is otherwise ill.
'Do you go out with friends as often as you did', where a no is coded as positive for depression.

Trying to assess 'normal' behaviour in very long lived illness can be very hard. It is only after 20 years I realised that my behaviour at university was not 'normal' - in that I was subconsciously avoiding talking to people or partying, because it greatly tired me and meant I couldn't carry on doing the course.

Any scales used need to try to assess change in behaviour, and will rely at least somewhat on the patient guessing what their state would be if they tried 'normal' activity.
Or indeed, guessing what normal activity is.
I had someone over to help move some rubbish, and was subconsciously assuming they'd need lots of rests, one bag at a time, ...
The goals shift.

 

[RogerBlack]

BMJ coverage of this is behind a paywall:

http://www.bmj.com/content/357/bmj.j3057

You can make comments on BMJ articles.
Keep it factual, references are best for any points you make from prior clinical studies.

 

[CFS_for_19_years]

BMJ coverage of this is behind a paywall:

http://www.bmj.com/content/357/bmj.j3057

http://sci-hub.cc/10.1136/bmj.j3057

 

[Jo Best]

Comment from AfME - for them I thought it was surprisingly strong.

https://www.actionforme.org.uk/news/action-for-me-comment-on-getset-study-in-lancet/

Don't be fooled by Action for ME, this style is par for the course over the years, a bit of righteous indignation and keeps their members onside in the belief that their charity is speaking up for them in a robust manner, and if they're lucky gets a few people thinking maybe, just maybe, this umpteenth time, they're coming around to the right way of thinking.

It may sound stronger this time for several reasons; e.g. they want patients on board with the MEGA project; the merger with AYME and now having a Children's Service, they are already being challenged by some clued up parents and they know the BBC Radio 4 programme is airing next week, they want to sound like fearless advocates for patients.

Another telling 'read between the lines' is their refrain about investment in research; if they genuinely believed this they would wholeheartedly support what Invest in ME Research has been doing since 2006. Not too late yet.

 

[Esther12]

http://sci-hub.cc/10.1136/bmj.j3057

Thanks - it wasn't available when I looked!

They largely just repeat the researcher's hype, but do have the comment from Simon Day on the trial not being blinded at the end.

 

[JaimeS]

we're currently working on a research survey for the Chronic Illness Inclusion Project. Which questionnaires/scales of do you think are worth the paper (or pixels) they're written on?

The Bell scale, the DePaul questionnaire

BTW it will be across energy limiting illness disability

Ah, then just the Bell's.

 

[Daisymay]

They say:


This appears to be it. It's from 2007 and seems to list things that the patient might have that would make you "suspicious" of "CFS/ME." It also lists things that should make you question a diagnosis of CFS/ME if they are not present.
https://www.nice.org.uk/guidance/cg53

It's been around a while, but I don't recall seeing it before (but maybe that's just me)..

Probably best to spin this off into its own thread.
It is for example, reasonable to try to measure depression at the same time, but many/most of the standard instruments for depression will read incorrectly for someone who is otherwise ill.
'Do you go out with friends as often as you did', where a no is coded as positive for depression.

Trying to assess 'normal' behaviour in very long lived illness can be very hard. It is only after 20 years I realised that my behaviour at university was not 'normal' - in that I was subconsciously avoiding talking to people or partying, because it greatly tired me and meant I couldn't carry on doing the course.

Any scales used need to try to assess change in behaviour, and will rely at least somewhat on the patient guessing what their state would be if they tried 'normal' activity.
Or indeed, guessing what normal activity is.
I had someone over to help move some rubbish, and was subconsciously assuming they'd need lots of rests, one bag at a time, ...
The goals shift.

Very well said, and you would have hoped it would be blatantly obvious to researchers doctors but it's not, but then it to their advantage that it's not as they can claim higher levels of PWME have co morbid depression etc

 

[user9876]

The study protocol has been published.12

The study protocol was published last year 8th June 2016
http://www.researchprotocols.org/2016/2/e70/

I don't think it is good practice to publish the protocol after the trial.

The trial itself was retrospectively registered in 24/05/2012 when it started on 16/05/2012 but with ethical permission on 23/11/2011.

The trial finished on 01/12/2015 so well before the protocol was published.

Also the primary end points were changed twice:

Added 21/07/2015:
2. Chalder fatigue scale measured at 12 weeks and 1 year

Updated 28/02/2017:
2. Chalder Fatigue Scale measured at 12 weeks

Where as the paper implies a single change on June 20, 2013 but this is when they claim to have made the decision. The recorded change is much later.

 

[alex3619]

So it's really merely changing their perspective on their illness rather than actually improving it.

Isn't that the very definition of CBT therapy?

 

[Barry53]

Isn't that the very definition of CBT therapy?

If your underlying issues are behavioural ones (I know this from personal experience 3/4 of a lifetime ago) then a programme of changing how you think about yourself can change how you behave. Worked a treat for me over a lot of years. But never a single psychiatrist involved, all my own work - suspect I was doing my own version of CBT long before it became the fashion.

 

[lilpink]

The Chalder fatigue scale is less tied to real world abilities

It also has a glass ceiling... which they must know about and which they might find highly useful...

 

[lilpink]

It was London Bridge Ethics committee.

Thanks for that.

 

[user9876]

In a previous trial of chronic fatigue syndrome and myalgic encephalomyelitis in secondary care (n=641),7 mean SF-36 PF scores were 37 (SD 15) at baseline and 48 (21) after 12 weeks of practitioner-led GET (ie, an 11-point increase).7 On the basis of these findings7 and our estimate that GES would be less effective than therapist-delivered GET, our sample size calculations were based on the assumption that a mean difference of 8 (SD 18) points between study groups at 12 week follow-up would be a clinically meaningful difference on the SF-36 PF scale.

There initial claims here are wrong for n which would be about 161 as they are referring to the PACE trial and specifically the GET arm but they give a value of n for the whole trial.

But then their reasoning gets strange. They seem to be saying that they think that GETSET will give worse results than PACE so they think clinically meaningful should be less than the PACE outcomes. Rather than having clinically meaningful as something that means a noticeable improvement in patients above the general noise and fluctuations.

Here they have basically declared a change in two answers on the SF36-pf scale will be considered clinically meaningful.

Items with missing data on the primary and secondary outcome variables at baseline and follow-up were imputed with mean replacement (prorating) when less than 20% of item responses per scale were missing.

I think this is dodgy for the sf36 where they say that a number of missing items can be ignored and adjusted for using an alternative denominator. I think that is what they have also done with the CFQ in the past. But maybe I am misunderstanding their procedure or where it applies.

All randomised participants with outcome data at follow-up (ie, the modified intention-to-treat population) were included in analyses, regardless of any departure from the allocated intervention.

They have modified the intention-to-treat so that it is no longer an intention to treat but the set of people who gave follow up data. That is not intention to treat at all and highly misleading.

 

[user9876]

It also has a glass ceiling... which they must know about and which they might find highly useful...

I would worry less about the ceilings and more about clumping of questions at certain points of the scale for both the CFQ and SF36 which mean that small changes in what they are measuring at some points (often the middle) could lead to larger changes in the questions than other areas.

 

[user9876]

On the entry criteria they used NICE but they did an assessment for Oxford and CDC. What seems a little strange is that only 83 and 87% that met the NICE guidelines met the Oxford ones.

 

[Snowdrop]

I read S Chowdhury's comments to the Telegraph as a push for MEGA. And Chris Ponting comes out with a statement hardly in support of the study and he would be on the MEGA team.

The thing is-- however genuine the desire to produce (whatever they might consider) real research I don't trust them to know how to do that effectively and honestly. And I do not trust whoever might be in charge not to fiddle with the interpretation of the results. For starters the MEGA bio research cannot happen without a lot of money and the Psych arm of the research will most definitely be given priority no matter the earnest discussion on how much the bio is needed.

It's a matter of power and influence.

Apologies, seems off topic but related to SC's response to study in Telegraph.

 

[greeneagledown]

The study's primary problems, in brief:

- As exercise is often helpful in depression, and as depression causes fatigue, any study on CFS patients that does not screen for depression in the subjects beforehand is completely useless, as there is no way to tell if the exercise is helping by treating the CFS or, instead, by treating depression.
- The combination of subjective endpoints and the unblinded nature of the study renders it useless.
- The obvious selection bias -- all CFS patients already know whether exercise helps or worsens their symptoms, and if it worsens their symptoms, they won't sign up for a study like this.

The third problem is probably unavoidable, but the first two could be fairly easily dealt with in study design.

 

[Keith Geraghty]

I'll add to that:

a study that shows that a group of self-selecting patients with fatigue for 4 months and 1 other symptom
- 20% had useful benefit
- 80% did not
recovery not measured
- a good number of pts were at recovery on physical health SF-36 at trial entry
- outcomes switched mid trial
-12 week evaluation point meaningless, end of trial, strongest place for expectancy impact
-an exercise trial run by a physo team with no measures of exercise tolerance, physical activity, muscle power or stamina
- unblinded
-no objective measures
- 10% dropped out in GET group and not sure their reasons or whether they are included in data presented on benefit
- lead PI has major conflicts of interest

Discount the SMC improvers from the GET improvers and discount placebo, the strong bias in the treatment arm, ie think positive reinforcement and that 1 extra person (the 10% added benefit over do nothing, can easily be accounted for by bias and design)

so much more, but these are the main ones