PACE Trial and PACE Trial Protocol

[Dolphin]

Info. on a different sort of 6 min test

Bedtime here so don't have time to check it further right now, but there seems to be two different tests: a 6 minute walking test (6MWT), and a 6 minute walking distance (6MWD) test. If so then we had better be careful not to mix them up.

Anyone able to explain the difference?

Here is talk of a 6 minute test where they make a distinction:

O'Dowd H, Gladwell P, Rogers CA, Hollinghurst S, Gregory A. Cognitive behavioural therapy in chronic fatigue syndrome: a randomised controlled trial of an outpatient group programme. Health Technol Assess. 2006 Oct;10(37):iii-iv, ix-x, 1-121. Free full text at: http://www.hta.ac.uk/execsumm/summ1037.htm

Sharpe and colleagues19 used a 6-minute walk test,
which measured the distance walked when the
patient was asked to walk as quickly as possible
within a 6-minute period. This test had previously
been validated in populations with chronic airways
obstruction. The shuttle walk test,65 or the
incremental shuttle walk test (ISWT), is thought to
have advantages over the 6-minute walk test,
including the achievement of a greater pulse rate
during the test and a reduced influence of bias
due to the reinforcement of the observer.66 The
ISWT is validated as an outcome measure for
chronic obstructive airways disease, low back pain
and rheumatoid arthritis.

One important consideration when choosing a
physical outcome measure for CFS/ME is that the
patient may be capable of performing during the
measure, but may suffer an unacceptable increase
in symptoms later as a result. This information is
not captured by any outcome measure known to
the authors, either for this condition or for the
somewhat related clinical area of chronic pain. In
the present study, an attempt was made to capture
this information by asking the subjects to state
their rate of perceived fatigue (RPF) at the end of
the ISWT using the modified (10-point) perceived
exertion scale – the category ratio scale CR1067
(see Appendix 13). Subjects were also asked to
inform the tester when they had reached their
normal walking speed, to gain further information
about the functional ability of the subject.

The ISWT, used as a physical performance
measure, has normative reference data described
by Taylor and colleagues.75 Their sample of 122
healthy subjects (mixed gender and age) walked a
mean of 67 10-m shuttles. By comparison, the
baseline mean in the current study was 24.6
10-m shuttles, reflecting a level of incapacity
similar to that found in a group of back pain
sufferers also sampled by Taylor and colleagues.
Although the CBT group showed a statistically
significant improvement in walking speed
compared with the EAS and SMC groups, it is not
clear whether the improvement observed is
clinically significant. The 46% increase over
baseline in median shuttles walked in the CBT
condition suggests that a useful change might
have taken place for some individuals, which did
not reach statistical significance at a group level.
Walking performance remains an important
variable for the evaluation of CFS/ME
interventions. The authors are not aware of any
research which has attempted to estimate the level
of clinically significant change for the ISWT in any
population. Further validation work is needed with
the ISWT for this population, including estimation
of what constitutes a clinically meaningful change.

This was a group CBT program based on graded exercise.
The final result was a long way from 67 shuttles.

Shuttles walked
CBT EAS SMC
Baseline 24.3 23.3 26.2
6 months 28.5 25.6 23.6
12 months 28.9 24.1 24.2

(EAS=education and support, SMC=standard medical care)

 

[Dolphin]

This is something i've been trying to get my head around...
I haven't read the whole paper, or the exact methodology, but it looks to me like they haven't included the dropouts in the figures, because in the Chalder scale figures for CBT, for example, the baseline number of patients was n=161, and then at 52 weeks n=148. So the results were recorded for n=148, not n=161.

I think that they might also have added more patients to each group, as patients dropped out, so the difference between 161 and 148 might not be the true drop-out rate. I'm not clear on that. Does anyone know about this?

I've worked out for the chalder scores in the GET group, that if there were only 6 drops outs, as is apparent from the figures in the table 3 (I'm not clear on the number due ot the point i made above), and if we were to assume that each of these drop-out patients recorded a maximum score of 33, then this only changes the total average score by 0.465. But the chalder scale is skewed in favour of the recovering patients, because 33 is the maximum score that extremely ill patients can get anyway, so they only increase their scores by a couple of points if they relapse, whereas those who improve can drop by many points.

I'm a bit confused about this myself. Table 1 (bottom) might help.

 

[Dolphin]

Data on training

If healthy people did a training program for 12 months, I imagine there scores should be quite good at the end.
I wonder are there any data out there where the 6-minute walking test was used as an outcome measure e.g. on overweight/obese people for example who might have simple deconditioning.

 

[Artstu]

I wonder are there any data out there where the 6-minute walking test was used as an outcome measure e.g. on overweight/obese people for example who might have simple deconditioning.

They beat us lot http://www.ncbi.nlm.nih.gov/pubmed/18446882

For individual evaluation, however, an improved walking distance of at least 80 m was required to make the difference clinically significant.

 

[Dolphin]

They beat us lot http://www.ncbi.nlm.nih.gov/pubmed/18446882

Thanks.
I think I'll copy it in to show the data:

The six-minute walk test in outpatients with obesity: reproducibility and known group validity.
Larsson UE, Reynisdottir S. Physiother Res Int. 2008 Jun;13(2):84-93.

Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Huddinge, Sweden. Ulla.Evers.Larsson@ki.se

Abstract
BACKGROUND AND PURPOSE: To assess the reproducibility and validity of the six-minute walk test (6MWT) in men and women with obesity in order to facilitate evaluation of treatment outcome.

METHOD: A test--retest design was used to test reproducibility and a comparative design to test known group validity. Forty-three obese outpatients (16 male), mean age 47 (21-62) years, mean body mass index (BMI) 40 (3-62)kg-m(-2) performed the 6MWT twice within one week. Intraclass correlation (ICC1.1) and measurement error (S(w)) were calculated from the mean square values derived from a one-way repeated-measures ANOVA (fixed effect model). The reproducibility was also analysed by means of coefficient of variation (CV) and the Bland Altman method including 95% limits of agreement. The variance of the distance walked was analysed by means of regressions. The known group validity of the 6MWT (distance walked and the work of walking) in obese participants was shown by comparisons with 41 lean participants (18 male), mean age 47 (24-65) years, mean BMI 22.7 kg-m(-2) (19-25).

RESULTS: The obese group walked 534 m (confidence interval [CI] 508-560 the first and 552 m (CI 523-580) the second walk (p < 0.001). S(w) was 25 m, CV 4.7%, ICC1.1 was 0.96. The limits of agreement were -46 m+80 m. The validity tests showed that they walked 162 m shorter (p < 0.001) and performed much heavier work (p < 0.001) than the lean group. In the obese group, BMI alone explained 38% of the variance of the distance walked.

CONCLUSIONS: The 6MWT showed good reproducibility and known group validity and can be recommended for evaluating walking ability in subjects with obesity. For individual evaluation, however, an improved walking distance of at least 80 m was required to make the difference clinically significant. Despite shorter walking distance the obese participants performed heavier work than the lean.

(c) 2008 John Wiley & Sons, Ltd.
PMID: 18446882 [PubMed - indexed for MEDLINE]

It would have been interesting if they'd walked the CFS patients a second time on the day day like a lot of these studies. The sum of the two scores could have been compared.

 

[anciendaze]

For individual evaluation, however, an improved walking distance of at least 80 m was required to make the difference clinically significant.

Note that, once again, we have a change greater than the largest group improvement being required to assess a clinically significant positive change for another illness. We can't see the individual changes in the PACE trial, so we don't really know how many would meet this standard. After finding moved goalposts elsewhere we really aren't inclined to accept "trust us" when it comes to determining significant individual improvement.

What was touted as "moderately effective" might better be called marginally effective, and even this is not statistically strong. The major result appears to be the discovery that GET doesn't cause too much harm, as long as it is confined to carefully supervised research. Expanding the program would lose this caution.

In addition to figuring out what happened to change numbers in trials, we need to understand what took place when those adverse events, serious adverse events and adverse outcomes occurred. If I still trusted my memory, I would say the hypothesis that research response to these was more significant than ostensible treatment looks good.

Just to show that I am not being entirely negative, I would add that, were this trial extended for 100 generations, we might well evolve more responsive patients.

 

[Bob]

The major result appears to be the discovery that GET doesn't cause too much harm, as long as it is confined to carefully supervised research. Expanding the program would lose this caution.

I'm not sure that's the case...

Firstly the deterioration rates, using an equivalent measure as the improvement rates, have not been published, so we do not know the level of deterioration. (The reported deterioration required a change of 20 points on the SF-36 physical function scale recorded on two consecutive assessment appointments.)

Also, 93% of participants in both the GET group and the SMC group experienced 'non-serious adverse events’. A pre-tested benign placebo control intervention was not used in the trial but, instead, SMC was used as a treatment comparison. If both SMC and GET were harmful then this was shown in these results.

And I'm not sure if these figures include the patients who dropped out of the study altogether, whose numbers aren't clear.

 

[anciendaze]

I'm not sure that's the case...
Also, in the GET group, 93% of patients experienced 'non-serious adverse events...

Sorry Bob, I'm afraid you need a better irony detector. My last sentence should have clued you in that some of that was severely tongue-in-check. Maybe I should have said "not too, too much harm".

My serious point was about the effect of researchers' actions to deal with adverse events in patients.

While the protocol says a great deal about who writes reports to whom about adverse events, and suggests responsible people making clinical judgments when these occur, the whole area of interactions with patients during the process of investigating and evaluating adverse events remains a blur.

With 93% of GET patients experiencing some such, these interactions could have a major bearing on study results. 'Rolling out' this treatment from presumed successful research to practice would eliminate a significant safeguard, and might even destroy the very limited 'demonstrated effectiveness' of treatment.

I don't need advanced degrees in psychology to understand that sick people care if someone checks on them when things go wrong in their lives. The highly subjective nature of all published results, even when dressed up with numbers, makes this a major issue.

 

[Bob]

Sorry Bob, I'm afraid you need a better irony detector. My last sentence should have clued you in that some of that was severely tongue-in-check. Maybe I should have said "not too, too much harm".

Sorry about that! I'll blame my poor brain!

 

[anciendaze]

Restatement

This is an attempt to restate some previous thinking clearly. I will admit that I can be obscure and confusing. I am not consciously trying. People generally get more complicated as they get older, and I have enough years behind me.

For comparison with results of the 6MWD test, I deliberately concentrated on acknowledged serious diseases where it is difficult to gain even modest improvements. Most of these patients will decline in condition without effective medical intervention. If I used a random walk model for them, it would have to be a biased random walk. (Instead of flipping a fair coin at each step they would use one weighted toward decline.) Aside from the bias toward deterioration, a cohort of such patients is more likely to remain clustered together. Random variance will be smaller. In such a situation, even small gains due to medical intervention are highly prized. This is in contrast to a remitting/relapsing disease like MS where even large improvements may take place for reasons outside medical control.

I found that a gain for an individual patient similar to the maximum group mean gain from the PACE study would not be considered clinically significant. You could change the criteria and select a smaller number of improved patients, but this would lower the statistical significance. The PACE trial gains are either not clinically significant or not statistically significant, choose one.

All those measures of significance and probability of error ultimately rest on hidden assumptions about random variables with normal distributions. We have found multiple ways bounds, and actions to prevent patients from crossing bounds, may bias results toward improvement. This is a case where those hidden assumptions are violated. The impressive numbers are no more than ciphers.

To my mind the null hypothesis that this trial has caused meaningless changes in subjective scores without any lasting clinical effect has not been rejected. This is true in advance of any follow-up study. The only way follow-up could alter the situation would be to show unexpected long-term gains or losses. I predict little long-term effect whatsoever.

I had expected to find something behind all the hoopla, not a zero. The patients recruited must have been more optimistic and suggestible than those who declined to participate. They were willing to accept mainstream medical authority. They must have been persistent to remain in the study for a year. With such a selection effect, this result is especially striking.

 

[Dolphin]

To my mind the null hypothesis that this trial has caused meaningless changes in subjective scores without any lasting clinical effect has not been rejected.

This is a good way of putting it.

 

[Dolphin]

Statement by two Dutch ME/CFS groups on PACE Trial, etc.

[Google translation of the statement which was originally in Dutch (copied at the end) - it's at:
http://www.me-cvs-stichting.nl/nieuws/items/?newsId=2059 . The English translation is a bit stilted but if one reads the full sentence one can usually get an idea what is being said]

Treatment of CFS - Gap between science and patient experience
25-02-2011

The ME / CFS Foundation Netherlands and the Support Group ME and Disability note that the claim that cognitive behavior therapy (CBT) and Graded Exercise Therapy (GET) safe and effective treatment methods for people with Chronic Fatigue Syndrome (CFS), is contrary to experience ME / CFS patients themselves and other research on treatment methods. They fear that this publication in the British journal The Lancet of February 18, 2011 one-sided and thus contributes to misperceptions about the disease and the patient will make it more difficult for them to find appropriate treatment. Both patient groups believe that the complexity of the disease and state of the science is not currently justified in clinical practice to apply a uniform treatment, mainly on psycho-social research is based. They call for more biomedical research into ME / CFS, also in the Netherlands.

Study is not representative
The patients in the study were selected based on very broad criteria.
Previous research has shown that many patients by these criteria are selected, who (also) have a psychiatric disorder. In the British study shows that nearly half of the patients the case. It is well known that (chronic) fatigue is a common symptom in psychiatric disorders. As a further condition was that patients could travel to the place where the therapy was given were severely ill ME / CFS patients excluded from the study. It can be stated that the group studied was not representative of the entire group of CFS patients.

"Moderate results'
There is within the British study concentrated on the effects of a number of different therapies on fatigue and physical functioning. Objective measures, for example, the activity level of patients are not used. Moreover, the results of CBT and GET according to the researchers themselves moderate.
More than half of the patients had no benefit of these treatments. The researchers believe further research into effective treatments are imperative.

Patients own experiences
A total of twelve surveys several thousand patients worldwide appear correct CBT and GET to score worse than most other treatment options (1). A significant proportion even reported negative effects of these methods.
An inverted image can be found at pacing (2). Patients self-reported good experiences with this strategy correct. The British study used form of pacing, called adaptive pacing therapy differs greatly from what this means patients themselves (3).

ME / CFS: a complex biomedical disorder
More than 4000 scientific articles about ME / CFS biomedical point to a cause and disease mechanism. Directed by medical examination of CFS patients, many physical defects found. Recent medical-biological research lays a possible link between a number of diseases, including CFS, and retroviruses. The results of two recent studies (4.5) with over 85% of CFS patients were positive on a retrovirus gave rise to international concern.
There is currently worldwide continued research into these viruses in CFS.
It also made Dutch patients.

More money for biomedical research
ME / CFS is a complex disease and its symptoms vary greatly from one person.
The current state of science shows that scientists themselves disagree on the best treatment for the disease. Patients do their best to get better and it does not matter to them most exact method. CBT and GET are also taking many patients have tried, mostly without much result. The ME / CFS Support Group Foundation of the Netherlands and the ME and Disability believe that psychological research, such as CBT and GET, disproportionate levels of attention in the Netherlands. This contributes to both a one-sided perception of CFS as one-sided development of treatment methods. A multidisciplinary approach is given all the scientific developments in the field of ME / CFS in place. The organizations therefore urge again to the Netherlands to give more attention and money to invest in biomedical research. If not then continue to CFS patients in the Netherlands currently the victim of an inadequate treatment offer, which also will cost the company much money.

The ME / CFS Netherlands Foundation represents the interests of people with CFS in terms of socio-medical and social care, by providing information and providing peer contact. More information www.me-cvs-stichting.nl or phone 035 6211290

Me and Disability Support Group represents the interests of CFS patients in the areas of employment, income, school / study and amenities. More
information: www.steungroep.nl or phone 050 549 29 06

Notes:
1.* http://sacfs.asn.au/news/2009/09/09_20_adverse_reactions_to_get.htm#
http://www.meassociation.org.uk/?p=951
http://www.afme.org.uk/news.asp?newsid=1045
2. http://freespace.virgin.net/david.axford/me-pace.htm
3. The Lancet adaptive pacing "solution. This is another form of pacing beyond what the vast majority of patients below means. This latter form is not assumed that the disease is irreversible and that activities should be planned using a diary. Pacing is not taking therapy, but a strategy.
4. Vincent C. Lombardi, Francis W. Ruscetti, Jaydip Das Gupta, Max A. Pfost, Kathryn S. Hagen, Daniel L. Peterson, Sandra K. Ruscetti, Rachel K. Bagni, Cari Petrov-Sadowski, Bert Gold, Michael Dean, Robert H. Silverman and Judy A. Mikovits. Detection of an Infectious Retrovirus XMRV, in Blood Cells of Patients with Chronic Fatigue Syndrome. Science Vol. 326 No. 5952 pp.
585-589;
http://lannieinthelymelight.blogspot.com/2011/01/part-1-11711-xmrv-presentat
ion-by-dr.html
5. Shyh-Ching Lo, Natalia Pripuzova, Bingjie Li, Anthony L. Komaroff, Guo-Chiuan Hung, Richard Wang, and J. Harvey Alter. Detection of MLV-related virus gene sequences in Blood of Patients with chronic fatigue syndrome and healthy blood donors. PNAS, August 23, 2010

[* Note 1 is missing on the page but was given in a yahoogroup version of the article]




============================================



http://www.me-cvs-stichting.nl/nieuws/items/?newsId=2059

Behandeling van ME/CVS - Kloof tussen wetenschap en ervaring patint
25-02-2011

De ME/CVS-Stichting Nederland en de Steungroep ME en Arbeidsongeschiktheid constateren dat de stelling dat Cognitieve Gedragstherapie (CGT) en Graded Exercise Therapy (GET) veilige en effectieve behandelmethodes zijn voor mensen met het Chronisch Vermoeidheid Syndroom (CVS), ingaat tegen de ervaringen van ME/CVS-patinten zelf en ander onderzoek naar behandelmethoden. Zij vrezen dat de publicatie hierover in het Britse blad The Lancet van 18 februari 2011 bijdraagt aan een eenzijdige en daarmee onjuiste beeldvorming over de ziekte en het voor patinten nog moeilijker zal maken om een voor hen passende behandeling te vinden. Beide patintenorganisaties zijn van mening dat de complexiteit van de ziekte en stand van de wetenschap het op dit moment niet rechtvaardigen om in de klinische praktijk n uniforme behandeling toe te passen, die vooral op psycho-sociaal onderzoek is gebaseerd. Zij pleiten voor meer biomedisch onderzoek naar ME/CVS; ook in Nederland.

Onderzoek niet representatief
De patinten voor het onderzoek zijn geselecteerd op basis van zeer ruime criteria. Uit eerder onderzoek is bekend dat door deze criteria veel patinten worden geselecteerd, die (ook) een psychiatrische aandoening hebben. In de Britse studie blijkt dat bij bijna de helft van de patinten het geval. Het is algemeen bekend dat (chronische) vermoeidheid een veelvoorkomend symptoom is bij psychiatrische aandoeningen. Omdat verder als voorwaarde werd gesteld dat patinten konden reizen naar de plaats waar de therapie werd gegeven werden ernstig zieke ME/CVS-patinten uitgesloten van het onderzoek. Er kan dus gesteld worden dat de onderzochte groep niet representatief was voor de totale groep van ME/CVS-patinten.

Matige resultaten
Er is binnen het Britse onderzoek vooral gekeken naar de effecten van een aantal verschillende therapien op vermoeidheid en fysiek functioneren.
Objectieve maten voor bijvoorbeeld het activiteitenniveau van de patinten zijn hierbij niet gebruikt. Bovendien zijn de resultaten van CGT en GET volgens de onderzoekers zelf slechts matig. Meer dan de helft van de patinten had geen baat bij deze behandelingen. De onderzoekers vinden verder onderzoek naar effectievere behandelmethoden dan ook noodzakelijk.

Eigen ervaringen patinten
Uit twaalf enqutes onder in totaal enkele duizenden patinten wereldwijd blijken CGT en GET juist slechter te scoren dan de meeste andere behandelopties (1). Een aanzienlijk deel meldt zelfs negatieve effecten van deze methoden.
Een omgekeerd beeld is te vinden bij pacing (2). Patinten zelf melden met deze strategie juist goede ervaringen. De in het Britse onderzoek gebruikte vorm van pacing, zogenaamde adaptive pacing therapy, wijkt echter sterk af van wat patinten zelf hieronder verstaan (3).

ME/CVS: een complexe biomedische aandoening Meer dan 4000 wetenschappelijk gepubliceerde artikelen over ME/CVS wijzen naar een biomedische oorzaak en ziektemechanisme. Bij gericht medisch onderzoek van ME/CVS-patinten worden vele lichamelijke afwijkingen gevonden. Recent medisch biologisch onderzoek legt een mogelijk verband tussen een aantal ziekten, waaronder ME/CVS, en retrovirussen. De uitkomsten van twee recente studies (4,5) waarbij meer dan 85% van de ME/CVS-patinten positief werd bevonden op een retrovirus gaven internationaal aanleiding tot zorg. Er wordt momenteel wereldwijd vervolgonderzoek gedaan naar dergelijke virussen bij ME/CVS. Daarbij zijn ook Nederlandse patinten betrokken.

Meer geld voor biomedisch onderzoek nodig ME/CVS is een complexe ziekte en de symptomen variren sterk per persoon. De huidige stand van de wetenschap laat zien dat wetenschappers onderling van mening verschillen over de beste behandeling van de ziekte. Patinten doen hun uiterste best om beter te worden en het maakt hen meestal niet uit via welke methode. Ook CGT en GET zijn daarbij door veel patinten al geprobeerd, meestal zonder veel resultaat. De ME/CVS-Stichting Nederland en de Steungroep ME en Arbeidsongeschiktheid zijn van mening dat psychologisch onderzoek, zoals CGT en GET, buitenproportioneel veel aandacht krijgt in Nederland. Dit draagt bij aan zowel een eenzijdige beeldvorming over ME/CVS als een eenzijdige ontwikkeling van behandelmethoden. Een multidisciplinaire aanpak is gezien alle wetenschappelijke ontwikkelingen op het gebied van ME/CVS op zijn plaats. De organisaties dringen daarom opnieuw aan om in Nederland meer aandacht te geven aan en geld te investeren in biomedisch onderzoek. Gebeurt dat niet dan blijven ME/CVS-patinten in Nederland vooralsnog de dupe van een ontoereikend behandelaanbod, die ook de maatschappij veel geld blijft kosten.

De ME/CVS-Stichting Nederland behartigt de belangen van mensen met ME/CVS op het gebied van medisch en sociaal-maatschappelijke zorg, door het geven van voorlichting en het verzorgen van lotgenoten contact. Meer informatie www.me-cvs-stichting.nl of telefoon 035 6211290

De Steungroep ME en Arbeidsongeschiktheid behartigt de belangen van ME/CVS-patinten op het gebied van werk, inkomen, school/studie en voorzieningen. Meer informatie: www.steungroep.nl of telefoon 050 549 29 06
Noten:

2. http://freespace.virgin.net/david.axford/me-pace.htm
3. In de Lancet is adaptive pacing onderzocht. Dit is een andere vorm van pacing dan wat het overgrote deel van de patinten hier onder verstaat. In deze laatste vorm wordt niet verondersteld dat de ziekte onomkeerbaar is of dat activiteiten gepland moeten worden met behulp van een dagboek. Pacing is daarbij ook geen therapie, maar een strategie.
4. Vincent C. Lombardi, Francis W. Ruscetti, Jaydip Das Gupta, Max A. Pfost, Kathryn S. Hagen, Daniel L. Peterson, Sandra K. Ruscetti, Rachel K. Bagni, Cari Petrow-Sadowski, Bert Gold, Michael Dean, Robert H. Silverman and Judy A. Mikovits. Detection of an Infectious Retrovirus, XMRV, in Blood Cells of Patients with Chronic Fatigue Syndrome. Science Vol. 326 no. 5952 pp.
585-589; http://lannieinthelymelight.blogspot.com/2011/01/part-1-11711-xmrv-presentation-by-dr.html
5. Shyh-Ching Lo, Natalia Pripuzova, Bingjie Li, Anthony L. Komaroff, Guo-Chiuan Hung, Richard Wang, and Harvey J. Alter. Detection of MLV-related virus gene sequences in blood of patients with chronic fatigue syndrome and healthy blood donors. PNAS, August 23, 2010

 

[oceanblue]

Controls and placebo effect

Apologies if this point has been made already, but I thought it was worth pointing out explicitly that there isn't a placebo-controlled group in the PACE trial. It's not CBT vs SMC, but CBT+SMC vs SMC. In other words there is no control for the therapeutic effect of up to 15 one-hour therapy sessions, as there would have been if, say, they had 15 sessions of relaxation 'therapy' for the control group.

Given the rather modest changes found, and the fact that these changes are in subjective measures, the placebo effect could be very important.

This may be pertinent to the line of enquiry being pursued by anciendaze, which I suspect is both important and wise, but which I'm afraid I don't quite understand.

 

[Angela Kennedy]

anciendaze said: "To my mind the null hypothesis that this trial has caused meaningless changes in subjective scores without any lasting clinical effect has not been rejected."

This is a good way of putting it.

It's a VERY good way of putting it. Out of interest (haven't got access to my PACE article to hand) were P values calculated at all?

 

[Angela Kennedy]

Apologies if this point has been made already, but I thought it was worth pointing out explicitly that there isn't a placebo-controlled group in the PACE trial. It's not CBT vs SMC, but CBT+SMC vs SMC. In other words there is no control for the therapeutic effect of up to 15 one-hour therapy sessions, as there would have been if, say, they had 15 sessions of relaxation 'therapy' for the control group.

Given the rather modest changes found, and the fact that these changes are in subjective measures, the placebo effect could be very important.

This may be pertinent to the line of enquiry being pursued by anciendaze, which I suspect is both important and wise, but which I'm afraid I don't quite understand.

Well- caveats about non-tenable claims of 'placebo effect' aside- the lack of a placebo arm (and therefore measurable placebo RESPONSE) in this trial is important I would say.

 

[Dolphin]

GET participants had a training effect for 6 minute walking test?

We saw from 1+ (can't remember number) of the papers posted that when people with other conditions (or indeed healthy people?) did the 6 minute walking test again on the same day/similar, on average that they improved on the first "run".

This made me think of something which I'm going to have to explain in a slightly rambling way: before I was ill, I was very fit from playing lots of sports. Occasionally I'd enter running races including cross-country "on the side" (just turn up on the day as part of the school team or whatever). However what I would often find was that towards the end I'd sprint past lots of people but not end up near the top: the basic problem was that I was going at an easy, too easy, jog/pace at the start rather than pushing myself as I didn't really know what sort of pace to go at or I could cope with. If I'd been more accustomed to running, I could have pushed myself harder at a more even pace. I'm not saying I would have won these races but I was underperforming as I wasn't sure how my body would react - I was going at a gentler pace than I needed to.

Then if one looks at the 6MWT, those who didn't go for walks e.g. many/most in the SMC and APT groups, were probably in a similar situation. They wouldn't necessarily know how their body would react to walking continuously. While the GET group would have much more experience of doing a 6 minute walk. Also because it was a walk, unlike when I was doing a run, the people who found they still had some energy left with a minture or more to go probably couldn't really increase their pace that much as it's a walking test so they wouldn't make up much ground from "underperforming" at the start.

Just a thought that came to mind.

 

[Doogle]

If there is a place on the WIKI for people to copy what they have submitted to the Lancet so others can preview and submit accordingly, would it be helpful, or would it jepardize what the Lancet picks for submission?

 

[Dolphin]

If there is a place on the WIKI for people to copy what they have submitted to the Lancet so others can preview and submit accordingly, would it be helpful, or would it jepardize what the Lancet picks for submission?

I don't know but wherever it is done should ideally be private.
Perhaps a private group could be set up for people to discuss draft letters they have as well as to post any submitted letters to avoid too much duplication. A few people have been showing me their letters but other people may be able to give other input on such letters. I'll join if one is set up but don't think I'll set it up myself as pretty stretched at the moment (probably wouldn't take much work).

One isn't really supposed to post submissions to journals on the internet as the journals generally own the copyright (although it's most likely less of an issue with letters, especially as with this journal, the letters are open access).

A small private group would be different. I think one restrict who joins such private groups? i.e. somebody would have to be approved?

 

[anciendaze]

It's a VERY good way of putting it. Out of interest (haven't got access to my PACE article to hand) were P values calculated at all?

Yes, they were. However, the underlying hidden assumption of dealing with normal distributions appears to be violated. This is a constant problem when psychologists who don't understand the derivation of these tests use them to validate their own beliefs. (There is a long chain of mathematical deductions between the assumptions and numbers you can easily calculate.)

The alternative to presumed ignorance is deliberate fraud. My own experience causes me to cast my vote for incompetence. One problem here is that they can cite any number of respectable people who have gotten away with similar mistakes.

Just to provide a specific example of a serious problem, I'm going to pretend these are normally distributed random distributions and ask a question vital in parametric statistics: which standard deviation are we to use? This appears to jump between different values during the course of this study. If such a fundamental parameter is this uncertain, (in an ordinary sense of uncertain,) what does that say about measures derived from it?

 

[Bob]

Apologies if this point has been made already, but I thought it was worth pointing out explicitly that there isn't a placebo-controlled group in the PACE trial. It's not CBT vs SMC, but CBT+SMC vs SMC. In other words there is no control for the therapeutic effect of up to 15 one-hour therapy sessions, as there would have been if, say, they had 15 sessions of relaxation 'therapy' for the control group.

Given the rather modest changes found, and the fact that these changes are in subjective measures, the placebo effect could be very important.

This may be pertinent to the line of enquiry being pursued by anciendaze, which I suspect is both important and wise, but which I'm afraid I don't quite understand.

That's an inspired observation... Why had that never occured to me before?!
I had thought that the SMC was the control group, but you're right... it's not.
They are very cunnning these people, aren't they!