PACE Trial and PACE Trial Protocol

[Dolphin]

Prof. Hooper, the Lancet & the PACE Trial

I thought this was interesting:

There are a staggering number of flaws in the PACE Trial article published in The Lancet (Comparison of adaptive pacing therapy, cognitive behaviour therapy, graded exercise therapy, and specialist medical care for chronic fatigue syndrome (PACE): a randomised trial. Peter D White et al. The Lancet, 18 February 2011 doi:10.1016/S0140-6736(11)60096-2), not one of which was mentioned in the press conference.

These flaws and errors have been identified in a detailed complaint/statistical analysis sent by Professor Malcolm Hooper to The Lancet on 28th March 2011, upon which The Lancet has asked Professor Peter White to comment (a response with which Professor White has apparently not complied within the time allotted for its receipt by The Lancet).

It is understood that under the Elsevier complaints policy, Professor Hooper will be asked to respond to Professor Whites reply when it is received by The Lancet; it is also understood that the PACE Trial article was to be sent for re-review by different reviewers and statisticians whilst The Lancet was awaiting Professor Whites comments on Professor Hoopers complaint.

Professor Hoopers analysis will shortly be placed in the public domain; he had agreed with The Lancet to withhold his complaint from publication during the time allotted by The Lancet to Professor White to respond to it, but this agreed time limit has now expired.

from:
http://www.meactionuk.org.uk/The-Media-and-ME.htm
or
http://forums.phoenixrising.me/showthread.php?11114-The-Media-and-ME

 

[Dolphin]

Rejection letters are starting to be sent out for the Lancet letters.

I think it'd be good if they could be used in some way.

If people don't want to post theirs with their name on it for some reason, I'd be happy to post it e.g. here. Also could be useful for the Wiki, perhaps to show that the Lancet didn't publish all the criticisms, etc.

 

[Dolphin]

I've now heard of 9 rejections including the MEA.
-----
Paul Winters has just posted on the Lancet's Facebook Wall about the rejections (it can take a bit of fiddling to see the most recent posts rather than just the Lancet posts).

 

[Marco]

Now 10 Dolphin.

I'll add it to your repository in due course.

 

[Dolphin]

Now 10 Dolphin.

I'll add it to your repository in due course.

Sorry to hear that, Marco. You shared yours with me and I thought it contained some good points.

It's now 11 with Peter Kemp.

 

[Dolphin]

New thread: PACE Trial - letters that were not accepted by the Lancet

New thread: PACE Trial - letters that were not accepted by the Lancet

http://forums.phoenixrising.me/show...-letters-that-were-not-accepted-by-the-Lancet

 

[alex3619]

Something I missed during my earlier reading but which Hooper points out in his latest, is the CFS patients who were selected according to CDC criteria were selected using 2003 Reeves criteria, not Fukuda 1994. That was a very WTF moment for me. Bye, Alex

 

[Dolphin]

Something I missed during my earlier reading but which Hooper points out in his latest, is the CFS patients who were selected according to CDC criteria were selected using 2003 Reeves criteria, not Fukuda 1994. That was a very WTF moment for me. Bye, Alex

It's the 2005 Reeves criteria that are the big problem really/usually (the "empiric" criteria) - just in case they're being mixed up.
The 2003 criteria are not a huge amount different from the Fukuda usually - they are used by quite a few researchers.

Although we don't know how exactly the criteria were used here [symptoms were asked on a 0-4 scale - and seem to be asked over the last week (according to the full (unpublished) criteria - we don't know (I think it's fair to say?) whether somebody qualified if they said a symptom was "present a little")].

 

[Angela Kennedy]

It's the 2005 Reeves criteria that are the big problem really/usually (the "empiric" criteria) - just in case they're being mixed up.
The 2003 criteria are not a huge amount different from the Fukuda usually - they are used by quite a few researchers.

Although we don't know how exactly the criteria were used here (symptoms were asked on a 0-4 scale - and seem to be asked over the last week (according to the full (unpublished) criteria - we don't know (I think it's fair to say?) whether somebody qualfied if they said a symptom was "present a little".

Actually - the one thing Reeves 2003 DO do (dodo? sorry) is make it clear that any SIGN associated with a symptoms means 'alternative diagnoses' must be sought, and in a research context, that means excluding a patient from a CFS diagnosis! So - if a patient has ACTUAL swollen lymph glands (palpable, for example), inflamed throat and/or tonsils, swelling or redness of joints, they are excluded! This is different to Fukuda, which I believe doesn't address those pesky clinical sign issues.

 

[oceanblue]

It's the 2005 Reeves criteria that are the big problem really/usually (the "empiric" criteria) - just in case they're being mixed up. The 2003 criteria are not a huge amount different from the Fukuda usually - they are used by quite a few researchers.

Although we don't know how exactly the criteria were used here (symptoms were asked on a 0-4 scale - and seem to be asked over the last week (according to the full (unpublished) criteria - we don't know (I think it's fair to say?) whether somebody qualfied if they said a symptom was "present a little".

I think this is potentially a huge issue, because if they scored symptoms based on the last week it invalidates their 'International criteria' defintion and with it their claim that the results apply equally to patients defined by these criteria.

The problem is we can't be sure how PACE implemented the criteria, though the unpublished protocol suggests they got it wrong. Are there any friendly researchers that might be able to contact the PACE gang for clarification, or are there any other ways we might get a defintitive answer of how the patients were diagnosed?

 

[oceanblue]

105 CBT/GET sessions = 1 patient marginally improved

I thought I'd add some analysis to Peter Kemp's great comment in one of his unpublished letters:

Therefore only 15% of participants in the CBT group improved over and above the control group. For GET the figure is 13%... CBT and GET appear to be ineffective for most Chronic Fatigue Syndrome patients.

This difference is best described by the 'Number Needed to Treat- (NNT) measure:
the number of patients that need to be treated for one to benefit compared with a control. Those figures Peter quotes translate to a NNT of 7 for CBT and 8 for GET.

Or put another way you would need to treat 7 CFS patients to get just one patient up to a 'clinically useful difference' (the smallest worthwhile improvement). That's the equivalent of 105 CBT sessions. For GET it's 8 patients and 120 GET sessions. This probably doesn't look too clever in a cost effectiveness analysis.

For comparison, the Protocol predicted a 60% improvement for CBT vs only 10% for SMC, a difference of 50%, giving a NNT of 2.

Of course, an NNT of 7 might be worthwhile if it applies to patients making a full recovery (a real one) rather than a small improvement. However, it appears the PACE authors know their 'back to normal' trick doesn't hold water which is presumably why they are going to have another bash at defining 'recovery' - to come up with better cost-effectiveness figures.

 

[anciendaze]

I think this is potentially a huge issue, because if they scored symptoms based on the last week it invalidates their 'International criteria' defintion and with it their claim that the results apply equally to patients defined by these criteria.

The problem is we can't be sure how PACE implemented the criteria, though the unpublished protocol suggests they got it wrong. Are there any friendly researchers that might be able to contact the PACE gang for clarification, or are there any other ways we might get a defintitive answer of how the patients were diagnosed?

Reeves 2005 or earlier 'empirical' definitions have a bizarre characteristic. Studies conducted on the basis of these criteria appear to have been done years before the criteria were published. What they meant at any particular time remains open to question. Arguments defending this practice basically state that Reeves=Fukuda=Oxford. In other words, definitions are unimportant because WE know what CFS is. The same (royal?) WE appears active in PACE.

 

[Dolphin]

Actually - the one thing Reeves 2003 DO do (dodo? sorry) is make it clear that any SIGN associated with a symptoms means 'alternative diagnoses' must be sought, and in a research context, that means excluding a patient from a CFS diagnosis! So - if a patient has ACTUAL swollen lymph glands (palpable, for example), inflamed throat and/or tonsils, swelling or redness of joints, they are excluded! This is different to Fukuda, which I believe doesn't address those pesky clinical sign issues.

That's interesting.

Definition and Evaluation of Accompanying Symptoms
The 1994 case definition defines CFS by the presence of debilitating fatigue accompanied by at least 4 of 8 designated symptoms. Accompanying symptoms must have persisted or recurred during 6 or more consecutive months of illness and cannot have predated the fatigue. Designated accompanying symptoms include the following: post-exertional malaise lasting more than 24 hours; unrefreshing sleep; impaired short-term memory or concentration severe enough to cause substantial reduction in previous levels of occupational, educational, social, or personal activities; headaches of a new type, pattern, or severity; muscle pain; multi-joint pain without swelling or redness; sore throat; and tender cervical/axillary lymph nodes. It is important to stress that these are symptoms not signs. Signs such as inflamed tonsils or swollen lymph nodes should prompt the search for alternative diagnoses.

These symptoms are non-specific and variable in both nature and severity over time. They were selected on the basis of consensus clinical opinion and were not identified empirically. We recommend that research studies use the SPHERE (discussed below) to query subjects (cases and controls) about the occurrence, duration, and severity of the 8 case defining symptoms and other potentially accompanying symptoms.

It makes me think that it was done for ideological reasons - so Peter White and co. could say "there is no signs with this condition". It might even have been spun in meetings to say that requiring "swollen" lymph nodes or visible sore throats may mean some people won't have them. Also maybe some people were scared that Cancer cases could be missed.

It looks to me that the Holmes definition deals with it better:

Physical Criteria:

3. Palpable or tender anterior or posterior cervical or axillary lymph nodes. (Note: lymph nodes greater than 2 cm in diameter suggest other causes. Further evaluation is warranted.)

i.e. gives an idea that while the nodes might swell a bit, look for other causes if they are very swollen.

 

[Sean]

i.e. gives an idea that while the nodes might swell a bit, look for other causes if they are very swollen.

Always been my understanding of that sign. A little bit is CFS, a lot is look further.

 

[Angela Kennedy]

Always been my understanding of that sign. A little bit is CFS, a lot is look further.

Any references for that?

I think we have to be spot on with our evidence on this. I say this particularly because swollen lymph glands, on palpation, would be devilishly difficult to quantify as 'a little' or 'a lot', as it depends on the subjective interpretation of the doctor!!

But in a research context (i.e. PACE), using Reeves in this way gets rid of people, again, with clinical signs, little or large.

All part of the crazy world of ME/CFS diagnosis...

 

[Sean]

Any references for that?

Wasn't making a factual physiological claim so got no direct ref for that, just saying that is how that sign/symptom is seen by medicos, according to Fukuda at least. Certainly was the kind of stuff I got checked for when I was first getting diagnosed in the late 80s. Still get that symptom myself, though less than before.

 

[Angela Kennedy]

Wasn't making a factual physiological claim so got no direct ref for that, just saying that is how that sign/symptom is seen by medicos, according to Fukuda at least. Certainly was the kind of stuff I got checked for when I was first getting diagnosed in the late 80s. Still get that symptom myself, though less than before.

That's interesting in the context of doctor subjectivity when trying to elucidate clinical signs, and how Reeves et al are so desperate to remove even this 'small' sign from a diagnosis of 'CFS' for research purposes, rendering the condition 'medically unexplained' and 'diagnosed by exclusion'. This has extremely adverse implications for clinical populations and the way they are treated, especially with the medical tendency to inappropriate extropolation from research cohorts, as we have seen. That both Nancy Klimas and Leonard Jason are part of Reeves 2003 is of concern in that context.

 

[Dolphin]

New thread: PACE Trial - letters that were not accepted by the Lancet

http://forums.phoenixrising.me/show...-letters-that-were-not-accepted-by-the-Lancet

15 letters have now been posted.

 

[wdb]

15 letters have now been posted.

I noticed a lot of these letters describe the 6MWD as an objective test but is it truly objective in the way that for example a machine measured specific blood count would be ? There must be factors such as reckless determination, willingness to risk relapse, or blind denial of ramifications that could affect the result independently of any actual objective change.

In a sense it is factual they did walk X meters in 6 minutes, but in a sense it is also an opinion, it is a demonstration of the participants opinion of how far they believe they can safely walk in 6 minutes.

What if there had been 1million 500 meters away which they could keep if they made it there in 6 minutes, would it have changed the results ? would it have improved anyone's health ?

 

[oceanblue]

Michael Sharpe gives more honest appraisal of PACE results in ABC radio interview

Just been reading the transcript of the ABC interview with Michael Sharpe (& Richard Horton) and was suprised by what he said about the trial results, as it's a big improvement on the '30% back to normal' we had from Trudie Chalder, the '60% improved' from Peter White or even the 'moderately effective' in conjunction iwth SMC in the paper itself:

Michael Sharpe: So then the next question is, 'How much better did you do?' And we set a threshold for what was a clinically meaningful improvement. About 6 out of 10 of the people who had CBT or GET made a clinically meaningful improvement compared with about 4 out of 10 of the people who had the pacing and medical care [actually 4.5 for SMC]. So the differences we're looking at here are clinically useful, but they're not actually enormous.

Norman Swan: GET is the graded exercise therapy. How many people had to receive graded exercise or CBT for one person to improve?

Michael Sharpe: We have a number needed to treat; I think it's about seven to get a clinically important treatment benefit with CBT and GET [actually 8 for GET]. What this trial isn't able to answer is how much better are these treatments than really not having very much treatment at all.

Would have been nice to hear that at the media launch, or even read it in the paper itself (clearly no accident that we didn't). Meanwhile, Richard Horton doesn't seem to have taken this on board:

Yeah, I mean adaptive pacing therapy essentially believes that chronic fatigue is an organic disease which is not reversible by changes in behaviour. Whereas cognitive behaviour therapy obviously believes that chronic fatigue is entirely reversible. And these two philosophies are kind of facing off against one another in the patient community and what these scientists were trying to do is to say, 'Well, let's see. Which one is right?'

Well, Richard, on the basis of what Michael Sharpe has just said, the CBT model is clearly wrong. The rest of Horton's comments are in similar vein.