PACE Trial and PACE Trial Protocol

[Dolphin]

ETA: This thread has turned out to be huge. I have set up another thread ["PACE Trial summaries/critiques/links thread - no discussion please"]
which should be more manageable: http://forums.phoenixrising.me/show...s-critiques-links-thread-no-discussion-please


The 5m (US$7.5m) PACE Trial hasn't been published yet.

However, its protocol has:
http://www.biomedcentral.com/1471-2377/7/6/

BMC Neurol. 2007 Mar 8;7:6.

Protocol for the PACE trial: a randomised controlled trial of adaptive pacing, cognitive behaviour therapy, and graded exercise, as supplements to standardised specialist medical care versus standardised specialist medical care alone for patients with the chronic fatigue syndrome/myalgic encephalomyelitis or encephalopathy.
White PD, Sharpe MC, Chalder T, DeCesare JC, Walwyn R; PACE trial group.

Department of Psychological Medicine, Queen Mary School of Medicine and Dentistry, St Bartholomew's Hospital, London, UK. p.d.white@qmul.ac.uk <p.d.white@qmul.ac.uk>

Lots of comments on this can be read at: http://www.biomedcentral.com/1471-2377/7/6/comments including my latest one:

New paper lists 3 CFS studies where there was no improvement in the actometer readings but an improvement was reported in subjective outcome measures

Tom Kindlon (12 May 2010) Irish ME/CFS Association - for Information, Support & Research

I know it might perhaps have seemed to some who have read these posts that I might be concerned about something that was not important (when I was calling for actometers to be used if possible for at least some of the patients at the end of the trial). So I feel "vindicated" in a way by a review[1] that has just been published. It found that in the three Dutch CFS trials looked at, studies which all found improvements in fatigue[2-4], there was no statistically significant increase in physical activity levels as measured by actometers.

The review also found that "changes in physical activity were not related to changes in fatigue."

The authors of the review (who include people who were involved in all the studies) say that, in the three studies, "treatment was based on the manual of CBT for CFS described in detail by Bleijenberg et al. (2003)" [5]. This form of CBT is comparable to the form of CBT being assessed in the PACE Trial [6].

It is useful to point out that fatigue wasn't the only subjective outcome measure that was said to have improved (in these trials where there was no increase in physical activity).

In all of the three studies [2-4], improvements were reported in functional impairment (as measured by questionnaires). In two of the studies [2,3], improvements in physical functioning as measured by the SF-36 physical functioning subscale were reported (this questionnaire does was not used in the third study[4]). So the patients reported improvements in "physical functioning" (as measured by the SF-36 physical functioning subscale) but there was no improvement in physical activity as measured by the actometers. The SF-36 physical functioning subscale is one of the primary outcome measures in the PACE Trial[6].

This discrepancy between objective measures of activity and questionnaire is similar to data I have previously drawn attention to[7] in a study by Friedberg and Sohl[8].

References:

[1] Wiborg JF, Knoop H, Stulemeijer M, Prins JB, Bleijenberg G. How does cognitive behaviour therapy reduce fatigue in patients with chronic fatigue syndrome? The role of physical activity. Psychol Med. 2010 Jan 5:1 -7. [Epub ahead of print]

[2] Stulemeijer M, de Jong LW, Fiselier TJ, Hoogveld SW, Bleijenberg G (2005). Cognitive behaviour therapy for adolescents with chronic fatigue syndrome: randomised controlled trial. British Medical Journal 330. Published online : 7 December 2004. doi:10.1136/bmj.38301.587106.63.

[3] Knoop H, van der Meer JW, Bleijenberg G (2008). Guided self-instructions for people with chronic fatigue syndrome: randomised controlled trial. British Journal of Psychiatry 193, 340341.

[4] Prins JB, Bleijenberg G, Bazelmans E, Elving LD, de Boo TM, Severens JL, van der Wilt GJ, Spinhoven P, van der Meer JW (2001). Cognitive behaviour therapy for chronic fatigue syndrome: a multicentre randomised controlled trial. Lancet 357, 841847.

[5] Bleijenberg G, Prins JB, Bazelmans E (2003). Cognitive-behavioral therapies. In Handbook of Chronic Fatigue Syndrome (ed. L. A. Jason, P. A. Fennell and R. R. Taylor), pp. 493526. Wiley: New York.

[6] White PD, Sharpe MC,Chalder T, DeCesare JC and Walwyn R for the PACE trial group. Protocol for the PACE trial: A randomised controlled trial of adaptive pacing, cognitive behaviour therapy, and graded exercise as supplements to standardised specialist medical care versus standardised specialist medical care alone for patients with the chronic fatigue syndrome/myalgic encephalomyelitis or encephalopathy. BMC Neurology 2007, 7:6

[7] Kindlon T. Further evidence showing why objective measures are preferable in CFS trials particularly where cognitions could be changed following the intervention http://www.biomedcentral.com/1471-2377/7/6/comments#333618

[8] Friedberg F, Sohl S. Cognitive-behavior therapy in chronic fatigue syndrome: is improvement related to increased physical activity? J Clin Psychol. 2009 Feb 11.

Competing interests

No competing interests

 

[oceanblue]

Tom, do you have a link to the review paper?

I've been following your BMJ comments on actigraphs/actometers and agree with you that they would be a much better, objective, measure of patient activity levels. PDW's reason for dropping actigraphs for post-treatment evaluation make no sense to me; it was a restating of his position rather than a response to your points.

 

[Dolphin]

Tom, do you have a link to the review paper?

The full text isn't available for free (but I have read it).
The abstract doesn't make the issue as clear as it should - but that's because these are the proponents of CBT who didn't include the actometer data when they initially published their studies.

Psychol Med. 2010 Jan 5:1-7. [Epub ahead of print]

How does cognitive behaviour therapy reduce fatigue in patients with chronic fatigue syndrome? The role of physical activity.

Wiborg JF, Knoop H, Stulemeijer M, Prins JB, Bleijenberg G.

Expert Centre Chronic Fatigue and Department of Medical Psychology, Radboud University Nijmegen Medical Centre, The Netherlands.

Abstract

BACKGROUND: Cognitive behaviour therapy (CBT) is known to reduce fatigue severity in chronic fatigue syndrome (CFS). How this change in symptomatology is accomplished is not yet understood. The purpose of the present study was to determine whether the effect of CBT on fatigue is mediated by an increase in physical activity.

Method: Three randomized controlled trials were reanalysed, previously conducted to evaluate the efficacy of CBT for CFS. In all samples, actigraphy was used to assess the level of physical activity prior and subsequent to treatment or a control group period. The mediation hypothesis was analysed according to guidelines of Baron & Kenny [Journal of Personality and Social Psychology (1986)51, 1173-1182]. A non-parametric bootstrap approach was used to test statistical significance of the mediation effect.

RESULTS: Although CBT effectively reduced fatigue, it did not change the level of physical activity. Furthermore, changes in physical activity were not related to changes in fatigue. Across the samples, the mean mediation effect of physical activity averaged about 1% of the total treatment effect. This effect did not yield significance in any of the samples.

CONCLUSIONS: The effect of CBT on fatigue in CFS is not mediated by a persistent increase in physical activity.

PMID: 20047707 [PubMed - as supplied by publisher]

I've been following your BMJ comments on actigraphs/actometers and agree with you that they would be a much better, objective, measure of patient activity levels. PDW's reason for dropping actigraphs for post-treatment evaluation make no sense to me; it was a restating of his position rather than a response to your points.

Yes.
And of course, this trial costs a mere 5m of taxpayers' money.

 

[oceanblue]

Thanks for the abstract.

RESULTS: Although CBT effectively reduced fatigue, it did not change the level of physical activity.

Well that is pretty damning. What was the combined CFS sample size for all 3 studies they looked at?

 

[Dolphin]

RESULTS: Although CBT effectively reduced fatigue, it did not change the level of physical activity.

Well that is pretty damning.

Yes, esp. considering they are of the view that CBT can lead to full recovery, etc!

[What was the combined CFS sample size for all 3 studies they looked at?

401 (156 in CBT groups, 245 in control groups).

They probably have data from other studies but we're not getting to see it at the moment, it seems. One of these studies (Prins et al.) was published in 2001 - they took their time publishing the data! Don't think people would be happy if a drug company did this.

 

[flex]

They are trying to say that they reduced the "fatigue" yet we are still too non complient to engage in physical activity. No doubt "proving" we are under the influence of "faulty illness beliefs" and a burden on the state.

 

[oceanblue]

Yes, esp. considering they are of the view that CBT can lead to full recovery, etc!

I've just found this quote from Bleijenberg about these findings: "patients can recover without an increase in their physical activity" (from slide presentation).
That's a very curious definition of recovery!

Thanks for all the info. With 156 CFS patients, that's a fairly robust sample, and all the more damning for that.

 

[Min]

They might as well have shoved 5million down a drain.

 

[Dolphin]

I've just found this quote from Bleijenberg about these findings: "patients can recover without an increase in their physical activity" (from slide presentation).
That's a very curious definition of recovery!

Thanks for all the info. With 156 CFS patients, that's a fairly robust sample, and all the more damning for that.

Thanks oceanblue.

I'd be interested in a link to that slide.
I know their definitions of recovery have been rubbish but not sure I've seen it said that explicitly by them.

 

[DysautonomiaXMRV]

They might as well have shoved 5million down a drain.

Well said, I wonder how far the WPI would have got with $7.2 million?
It's as if they don't want anyone to 'discover' XMRV because it legitimises people with CFS label.

I don't see the shame in having XMRV and getting treatment for it. Why prevent us?
I see the 2009 UK NICE guidelines on CFS/ME forbid routine testing for viruses/bacterial infections and
a TILT test (to diagnose Dysautonomia).

Again, I don't see the shame in having Dysautonomia. Aren't doctors meant to futher medical science?

So sad, it's tax payers money they are wasting. In a sick twist of fate one could argue UK CFS patients
are paying for their own demise.

 

[oceanblue]

Tom: that Bleijenberg quote come from a slide he gave at the MRC CFS/ME Workshop in Nov '09 (see page 30). The workshop was part of an initiative by Prof Stephen Holgate to get biomedical research up and running in the UK. Don't read too much into Bleijenberg's presence there - a pal of mine who was also there tells me his views made little impact.

RESULTS: Although CBT effectively reduced fatigue, it did not change the level of physical activity

Thinking about it, isn't that the minimum you'd expect from pacing: same level of physical activity but with reduced fatigue?

 

[oceanblue]

Questionable way of measuring Harm?

In the absence of the PACE trial itself I've resorted to reading the whopping 20 page protocol in full. It's remarkably thorough in places, but I was concerned by the way they plan to measure and evalutate Harms:

Adverse outcomes
Adverse outcomes (score of 5–7 of the self-rated CGI [ie not 'much improved']) will
be monitored by examining the CGI at all follow-up
assessment interviews. An adverse outcome will be
considered to have occurred if the physical function score
of the SF-36 has dropped by 20 points from the previous
measurement.

20 points seems to me like a big drop in the SF36 PF subscale, particularly since this is measured only at the assessment interviews at 3, 6 and 12 months post-randomisation. this could be months after the adverse event so that some recovery may have taken place, further reducing the chance that the fall off in the PF score counts as 'adverse'.

Only this 20 point+ deterioration will be measured alongside improvements in the primary outcomes.

There are other measures of Harms, but it's not clear how these will be measured or reported.
1) At the assessment interview the Nurse will ask about specific adverse events, but it doesn't say how any such adverse events mentioned by the patients will be reported
2) There is a separate system for measuring 'adverse events', 'Serious Adverse Events' and 'Serious Adverse Reactions'. 'Serious Adverse Events' have to be reported to the appropriate authorities, but according to the protocol:

If there is any
doubt in the minds of the Research Nurse and the centre leader as to
whether the AE is a serious AE, the centre leader will
obtain a second opinion from one of the PIs.

Which makes you wonder if the PI will decide that it isn't, after all, serious...
3) Patients dropping out or missing sessions ('DNA') are another way adverse events might surface, but again it's not clear how this will be reported.

Overall, I'm concerned that the number and scale of Harms might be underreported.

I think this could be one for the Harms King.

 

[Dolphin]

I think this could be one for the Harms King.

Does he have bat phone or any way to be contacted if needed?

20 points seems to me like a big drop in the SF36 PF subscale, particularly since this is measured only at the assessment interviews at 3, 6 and 12 months post-randomisation. this could be months after the adverse event so that some recovery may have taken place, further reducing the chance that the fall off in the PF score counts as 'adverse'.

Good point.
I think there is a floor with the SF-36 and most people who are able to attend for a therapy (GET/CBT/APT) are unlikely to drop down below a certain level
e.g.
one can score 10 for each of the following questions:

Bathing or dressing yourself
Bending, kneeling, or stooping

So most people are going to score 20 anyway. And many will probably pick up some points with other things e.g. "Climbing one flight of stairs" - say "limited a little" and one gets 5 points. Similarly "Lifting or carrying groceries" - say "limited a little" and one gets 5 points. Yet most/a lot of people starting will probably be 30-60 points so "hard" to drop 20 points.
Actometers might have been better e.g. if one dropped 20% on one result or 10% (??) long-term i.e. on more than one occasion compared to when one started, that's not a good result (although getting more specific it does depend on other symptoms for APT e.g. fewer steps but fewer/less severe symptoms might be good for APT; with CBT and GET - perhaps there might be fewer steps but not less/fewer symptoms). Oceanblue did find out that actometers may not be perfect measures of course.
An increase in pain long-term wouldn't be good either.

In Nez et al., statistically significant deteriorations compared to before the intervention were found for: Number of comorbidities, Pain, Weakness, SF-36 Physical functioning and SF-36 Bodily Pain. Some other things were close-ish - nearly trends - people were worse on more other items than the number where there were slight improvements.

 

[oceanblue]

Does he have bat mobile or any way to be contacted if needed?

Concerned citizens post on Phoenix Forums then look hopeuflly to the skies

Oceanblue did find out that actometers may not be perfect measures of course.
An increase in pain long-term wouldn't be good either.

In Nez et al., statistically significant deteriorations compared to before the intervention were found for: Number of comorbidities, Pain, Weakness, SF-36 Physical functioning and SF-36 Bodily Pain. Some other things were close-ish - nearly trends - people were worse on more other items than the number where there were slight improvements.

oceanblue is a party-poooper

But the fundamental problem re Harms seems to be the use of a high threshold: the PACE trial based the 20 point PF decline on one Standard Deviation of PF scores, as estimated from previous trials. And as you say an increase in long-term pain or other symptoms wouldn't be picked up either.

 

[oceanblue]

Yet most/a lot of people starting will probably be 30-60 points so "hard" to drop 20 points.

Agreed, especially as the assessment requires doing a 6-minute walking test, so it's unlikely all but a very few patients score under, say, 35 points:
- Bathing or dressing yourself = 10
- Bending, kneeling, or stooping = 10
- Walking one block @5 or 10
- Climbing one flight of stairs @ 5 or 10 points
- Lifting or carrying groceries @ 0 or 5 points

But a 20 point drop to qualify for harms from 35 points to 15 points is huge, 55%. Even a 15 point drop would probably leave such patients really incapacitated (and probably unable to continue in the trial), yet would not count as an adverse outcome. Even losing 15 points from 50 points would be a loss of 30% of what you can do and still not count as 'Harm'.

The PACE trial aimed to specifically address the issue of Harms, yet it seems to be set up not to find them.

 

[Esther12]

Not sure how related this is, but I was just looking at the Chalder scale for the first time and thought I'd add my impressions here:

Strange they phrase it as 'less the usual' when the introduction ("If you have been tired for a long time, we want you to compare the way you feel now to how you felt when you were last well") implies that what they mean is 'less than before I got ill'. That intro does make it much less bad than I thought it was though.

I wonder if by having four:

0. Less than usual ____

1. No more than usual ____

2. More than usual ____

3. Much more than usual ____



They get more people going for 1 than if it was just 3:

0. Less than usual ____

1. No more than usual ____

2. More than usual ____



Especially if a lot of CBT is about encouraging patients to view their fatigue as being within the (very big) normal range.

 

[oceanblue]

Strange they phrase it as 'less the usual' when the introduction ("If you have been tired for a long time, we want you to compare the way you feel now to how you felt when you were last well") implies that what they mean is 'less than before I got ill'. That intro does make it much less bad than I thought it was though.

I wonder if by having four:

0. Less than usual ____

1. No more than usual ____

2. More than usual ____

3. Much more than usual ____



They get more people going for 1 than if it was just 3:

0. Less than usual ____

1. No more than usual ____

2. More than usual ____



Especially if a lot of CBT is about encouraging patients to view their fatigue as being within the (very big) normal range.

Yes, odd way of phrasing things for people who are chronically fatigued!
In fact, I don't think that the scoring in 4 parts rather than 3 will make any difference in terms of the primary outcome as it's scored bimodally, so 'more than usual' or 'much more than usual' score 1, less than usual/no more than score 0. However, the Chalder scale is scored 0,1,2,3 for use in secondary outcomes.

 

[Esther12]

Yes, odd way of phrasing things for people who are chronically fatigued!
In fact, I don't think that the scoring in 4 parts rather than 3 will make any difference in terms of the primary outcome as it's scored bimodally, so 'more than usual' or 'much more than usual' score 1, less than usual/no more than score 0. However, the Chalder scale is scored 0,1,2,3 for use in secondary outcomes.

Yeah... but I thought that 'no more than usual' could still be pretty bad... if CBT had told patients that it was more healthy to view extreme fatigue as being a part of usual human experience rather than a medical condition - yet it's currently scored as 'not a problem'. I'm not sure... I feel like this is all still a bit alien to me and I could be missing something obvious. Ta.

 

[oceanblue]

Oxford Criteria - as implemented by PACE trial

I'm trying to get a feel for how the 3 different criteria used by PACE - Oxford, CDC/Fukuda (nb not the 2005 'Emprical' CDC criteria) and 'London' will differ in practice and I'm summing up my interpretation in a series of hopefully worthy but inevitably rather dull posts. However, case defiition differences may be crucial in understanding the results of PACE whenever they finally appear.

The Oxford Criteria were used for recruitment, which means that patients can only get a Fukuda or London diagnosis if they also meet the Oxford Criteria. In practice, I don't think any CDC or London patients will be excluded by the Oxford Criteria.

There are 5 elements to the diagnosis

1. Fatigue threshold
2. Disability threshold
3. Medical exclusions
4. Psychiatric exclusions
5. Other specific requirements

The first 4 of these are the same for both the Oxford Criteria and CDC and are based on the 2003 "Identification of ambiguities in the 1994 CFS Research Case Defintion" paper by an international group including Leonard Jason and Nancy Klimas, as well as the usual suspects. In pratice, all this definition does is tidy up the original 1994 case definition. It was the later 2005 'Empirical' definition that really messed things up.

Fatigue Threshold
Set as a Chalder Fatigue Scale (CFQ) of 6 or more at recruitment - scored bimodally.
The maximum score is 11, minimum is 0 and up to 3 is considered 'normal'.

Disability Threshold
Set as a score of 65 or less on the SF-36 Physical Function subscale.
However, there is effectively a minimum score too, probably around 35-40 (see below).

Trial participants 'must convince the Research Nurse that they will be able to attend hospital regularly and reliably' and must be able to 'complete all trial measures', which includes a 20-step test and a 6-minute walking test. To be able to do this, participants presumably will score:
- 10 points for Walking one hundred yards not limited at all (even walking slowly someone would cover 250 yards in 6 minutes)
- 5 or 10 points for Climbing one flight of stairs (say 16 steps) either limited a little or not limited at all
This is in addition for 20 points for no problems with dressing and bathing, or bending/kneeling. Which adds up to a probable minimum score of 35-40.

The maximum score of 65 corresponds to, for example, someone who can't manage vigorous or moderate activity or walk a mile, but can climb several flights of steps, walk a few hundred yards without problems and be only limited a little in carrying groceries.

Medical Exclusions

1. Organ failure (eg emphysemia)
2. Chronic infection (eg Hepatitis B, not XMRV!)
3. Rheumatic & chronic inflammatory diseases
4. Major neurological diseases eg MS
5. Diseases requiring systematic treatment eg chemotherapy
6. Major endocrine diseases eg hypopituitarism

Full details here (paragraph 2 under Results).

Psychiatric Exclusions
As diagnosed by the SCID interview

Permanent psychiatric exclusions include lifetime diagnoses of bipolar affective disorders, schizophrenia of any subtype, delusional disorders of any subtype, dementias of any subtype, organic brain disorders, and alcohol or substance abuse within 2 years before onset of the fatiguing illness.
Full details (paragraph 4 under Results).

Note that this leads a whole load of depressive and anxiety disorders allowed for both the Oxford and CDC criteria.

Specific to the Oxford Criteria

1. Fatigue present for the last 6 months for more than half the time.
2. Illness affects both physical ability and mental functioning
3. Fatigue is the principal symptom
4. Fatigue out of proportion 'to what you would expect as normal for this level of exertion'
5. As judged by the Research Nurse: can the fatigue be distinguished from low mood, sleepiness and low motivation?

Actually, I was surprised by 5, in that it makes some effort to exclude fatigue purely as a result of mood factors.

Apologies for any mistakes, I'm now exhausted.

 

[oceanblue]

CDC criteria - as used by PACE trial

Basically, take Oxford Criteria patients and select those that have 4 or more of the 8 symptoms defined in the CDC/Fukuda definition:

* difficulties with memory and concentration
* problems with sleep
* persistent muscle pain
* joint pain (without redness or swelling)
* headaches
* tender lymph nodes
* increased malaise (fatigue and sickness) following exertion
* sore throat

According to the 1994 case definition

all of which must have persisted or recurred during 6 or more consecutive months and must not have predated the fatigue.

However, it's not quite clear how this is implemented in the PACE trial. The full PACE Protocol includes a Case Report form simply titles "CDC" that asks patients if they have had "any of the following symptoms in the last week". That implies patients must have 4 or more symptoms at the time of recruitment, rather than some of the 'recurring in the previous 6 months'. It's also not clear if patients are told to ignore any symptoms that patients might have from other illnesses eg a cold, which could make a big difference if there isn't a requirement to have symptoms long-term.

Also, any patients with either physical fatigue or mental fatigue, but not both won'te have made it through the Oxford Criteria selection so won't be in the trial: I don't know if we'd expect there to be a significant number of these.

The PACE protocol form also asks patients to rate any symptoms they have from "present a little" to "present all the time" and it's not clear if there's a cut off eg 'present most of the time' to qualify as a genuine symptom - I couldn't see anything on this in the protocol.

The CDC criteria patients are pretty important as, while they are not perfect criteria, 1) the Oxford Criteria are hopelessly broad and 2) they allow comparison with other research, which almost always uses the CDC criteria.