Tackling fear about exercise produces long term benefit in chronic fatigue syndromeBMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h5771 (Published 28 October 2015) Cite this as: BMJ 2015;351:h5771
All rapid responses
In a BMJ article , Ingrid Torjesen claims that a follow-up study of the PACE trial has shown that: "Two treatments for chronic fatigue syndrome [...] continue to offer long term benefits". The headline boasts that: "Tackling fear about exercise produces long term benefit in chronic fatigue syndrome." The two treatments referred to are cognitive behavioural therapy (CBT) and graded exercise therapy (GET).
It is difficult to understand how the editors of BMJ can justify these assertions, considering that the study reported no useful difference in outcomes between treatment arms . The study clearly shows that CBT and GET had no treatment benefit at 2.5 year follow-up, when compared to an intervention resembling usual care.
The PACE trial was a large randomised trial that compared four interventions for chronic fatigue syndrome: CBT; GET; adaptive pacing therapy (APT); and standardised medical care (SMC). Participants in one trial arm received SMC alone, and the other three groups received either CBT, GET or APT in addition to SMC. SMC resembled usual care and typically provided pharmaceutical symptomatic relief when needed, such as pain and sleep medication, and advice to moderate activity levels.
In 2011 the Lancet reported modest differences at 52 weeks between some treatment arms when assessed using the self-rated primary outcomes measures; the Chalder fatigue scale and SF-36 physical function subscale .
Recently the Lancet Psychiatry reported the primary outcomes at 2.5 year follow-up  which showed that the differences between trial arms seen at 52 weeks were not sustained, and that therefore CBT and GET offered no treatment benefit at long-term follow-up.
The BMJ article acknowledges this, in a contradiction to its headline: "At the end of the follow-up period [...] there was little difference in outcomes among patients in any of the original four treatment groups."
Either CBT and GET "continue to offer long term benefits" over SMC and APT, or there were no useful differences in outcomes between the trial arms. It is not possible for both to be true, and the data in the Lancet Psychiatry follow-up paper clearly support the latter interpretation.
The BMJ article also discusses the issue of SMC and APT group participants receiving CBT or GET after 52 weeks and before the long-term assessment, and how this may have biased the overall outcomes. However, the accompanying commentary in Lancet Psychiatry , clarifies this issue and states that there was no significant difference between patients who received CBT or GET after 52 weeks and those who did not.
The commentary states: “The authors hypothesise that the improvement in the APT and SMC only groups might be attributed to the effects of post-trial CBT or GET, because more people from these groups accessed these therapies during follow-up. However, improvement was observed in these groups irrespective of whether these treatments were received, and thus this hypothesis remains unproven.” 
1. Torjesen I. Tackling fear about exercise produces long term benefit in chronic fatigue syndrome. BMJ 2015;351:h5771.
2. Sharpe M, Goldsmith KA, Johnson AL, Chalder T, Walker J, White PD. Rehabilitative treatments for chronic fatigue syndrome: long-term follow-up from the PACE trial. Lancet Psychiatry 2015.
3. White PD, Goldsmith KA, Johnson AL et al. Comparison of adaptive pacing therapy, cognitive behaviour therapy, graded exercise therapy, and specialist medical care for chronic fatigue syndrome (PACE): a randomised trial. Lancet. 2011; 377:823-36.
4. Moylanemail S, Eyre HA, Berk M. Chronic fatigue syndrome: what is it and how to treat? Comment. Lancet Psychiatry 2015.
Competing interests: No competing interests
There has been considerable concern expressed about the methodology behind this trial which should lead to huge caveats around the reliability of the results. David Tuller has written an excellent long article on the issues as well as a corresponding reply from the PACE investigators  and David's follow up response . I would recommend those looking to recommend or make decisions about these treatment read these articles first along with the latest ME Association patient survey  that documents may instances of harm caused by GET (one of the treatments recommended in this article).
To summarize two particular points of concern about the trial
1) There are many changes from the published protocol which is particularly serious given the trial was not blinded. Changes apparently occurred after the trial had finished but prior to looking at the data. But given the non-blinded nature of the trial it is likely that the investigators had an idea of the results. Detailed reasons beyond 'we believe this was better' statements were not given. Equally the effects of the changes could have been mitigated by publishing the original data along side the changes so that readers could see the effects. However, there has been a refusal to publish data as defined in the original protocol.
2) The results quoted rely on subjective measures of improvement. Part of the basis and rational of the CBT and GET protocols is to change how patients thing about symptoms. So improvements in subjective scores should not be considered reliable. The improvements were not demonstrated across more objective measures. At the start of this year, some 4 years after the first results were published, a graph showing the results of a fitness test was published. This showed very similar results for all groups with patients in the CBT and GET group performing slightly worse. Unfortunately the figures behind the graph are kept secret .
From the coverage of this latest long term follow up readers would believe that the trial suggested CBT and GET were the best long term treatments. However, the results of the primary long term hypothesis comparing the different groups showed no significant difference between all four treatments. This should be the main headline of this article anything else is misleading. Coyne has published a detailed analysis of this paper  - again I would suggest this is essential reading for anyone making decisions about these treatments.
Given the BMJ has called for better reporting of clinical trials, publishing of protocols and ensuring that they are enforced it is a pity they do not enforce such standards in their news stories.
 D. Tuller - TRIAL BY ERROR: The Troubling Case of the PACE Chronic Fatigue Syndrome Study - http://www.virology.ws/2015/10/21/trial-by-error-i/
 P.D. White et al - PACE trial investigators respond to David Tuller - http://www.virology.ws/2015/10/30/pace-trial-investigators-respond-to-da...
 D. Tuller - David Tuller responds to the PACE investigators - http://www.virology.ws/2015/10/30/david-tuller-responds-to-the-pace-inve...
 ME Association - ME/CFS Illness Management Survey results - http://www.meassociation.org.uk/wp-content/uploads/2015-ME-Association-I...
 G. McPhee - Fitness Data for PACE trial - https://www.whatdotheyknow.com/request/fitness_data_for_pace_trial
 J. Coyne - Uninterpretable: Fatal flaws in PACE Chronic Fatigue Syndrome follow-up study - http://blogs.plos.org/mindthebrain/2015/10/29/uninterpretable-fatal-flaw...
Competing interests: No competing interests