Intended for healthcare professionals

Rapid response to:


NICE understates role of exercise and CBT in managing ME/CFS, say medical leaders

BMJ 2021; 375 doi: (Published 29 October 2021) Cite this as: BMJ 2021;375:n2647

Rapid Response:

Are the new NICE guidelines for ME/CFS at odds with the research evidence?

Dear Editor

Medical leaders have raised concerns over the processes NICE used in developing their new guideline for ME/CFS, and for downplaying the importance of activity and exercise in managing the illness. [“NICE understates role of exercise and CBT in managing ME/CFS, say medical leaders.”]

The research evidence provided below contradicts these views. It also calls into question whether the medical leaders and the Royal Colleges views, are based on scrupulous analysis of published research, rather than on misleading media reports.

On October 29th 2021, NICE published their new guideline for the diagnosis and management of ME/CFS,[1] and the Science Media Centre published, ‘expert reactions’ to the new guideline.[2] One of the experts quoted is Professor Trudie Chalder, Professor of Cognitive Behavioural Psychotherapy, who remarks:

“The NICE guidelines for CFS/ME are at odds with the research evidence. Researchers from different Institutions in different countries have found graded exercise therapy and cognitive behaviour therapy to be effective for some patients with CFS. Evidence has shown they reduce fatigue and improve functioning without harm, if delivered by trained therapists in specialist clinics. Being a clinician and researcher in this field I can’t help but think clinicians will be confused by this message from a respected organisation.”

The suggestion that the NICE guideline is ‘at odds with the research evidence’, is a serious charge. It would be unworthy of NICE or any other respected and trusted source to confuse clinicians, or indeed, to confuse patients, the public or the media with information that is contradicted by reliable research evidence.

Professor Chalder was a principal investigator and author of the PACE Trial.[3] This was the largest controlled clinical trial testing adaptive pacing therapy (APT), cognitive behaviour therapy (CBT), graded exercise therapy (GET), and standardised specialist medical care (SMC, the control group) for ‘CFS/ME’. It continues to be referred to by supporters of GET and CBT as providing evidence of their efficacy, so it seems likely to be a significant source of the ‘evidence’ referred to in Professor Chalder’s comment.

The PACE Trial and the Science Media Centre
On Feb 17th 2011, the Science Media Centre hosted a press conference for the PACE Trial investigators to present their findings to the media.[4] The media duly reported the information provided, including the quotes shown below:

The Daily Mail: Got ME? Fatigued patients who go out and exercise have best hope of recovery, finds study
The Independent: Got ME? Just get out and exercise, say scientists
The Guardian: Study finds therapy and exercise best for ME
The Telegraph: Exercise and therapy can help ME sufferers, study claims
The Daily Record: Exercise and therapy can reverse effects of ME
Reuters: Pushing limits can help chronic fatigue patients

Quotes from media coverage of the PACE Trial:
-- With cognitive behavioural therapy, 30% of patients returned to normal levels of fatigue and physical function.
-- The results showed that CBT and GET benefited up to 60 percent of patients
-- Researchers found six in 10 patients reported significant improvements after undergoing either cognitive behavioural therapy (CBT)
-- With cognitive behavioural therapy, 30% of patients returned to normal levels of fatigue and physical function.
-- Overall, 60 per cent of patients who received CBT or GET made progress and 30 per cent recovered sufficiently to resume normal lives.
-- Six in 10 patients reported significant improvements, with half reporting a return to normal energy levels.

The consistency of statistics reported in the media suggests that they originated from the Science Media Centre event or ‘expert reactions’.

Claims that, “60 per cent of patients who received CBT or GET made progress and 30 per cent recovered sufficiently to resume normal lives”, are highly misleading and are in fact, “at odds with the research evidence”. The PACE Trial investigators omitted to inform the media that of the 60% who ‘made progress’ and 30% who ‘recovered sufficiently to resume normal lives’, only around 15% could be attributed to either GET or CBT therapies.

The true figure of a 15% having a treatment effect with the therapies is supported by another PACE Trial principal investigator, Professor Michael Sharpe, who stated on ABC radio Australia, on April 18th 2011:[5] “We have a Number Needed to Treat; I think it's about seven to get a clinically important treatment benefit with CBT and GET.”

A Number Needed to Treat (NNT) of 7 corresponds with 15% of participants reaching target thresholds. That is, it would be necessary to treat 7 patients with GET or CBT in order for one patient to reach a given threshold of improvement, or alternatively, to prevent one patient from reaching a given threshold of deterioration. If the media figures of 30% and 60% were actually true, then the Number Needed to Treat would have been, 4 and 2 respectively.

Normal Range – a contradictory construct
The claims that, “30% of patients returned to normal levels of fatigue and physical function”, and, “30 per cent recovered sufficiently to resume normal lives”, appears to be a misrepresentation of a construct which the PACE investigators called ‘Normal Range’. Far from indicating normal health, the thresholds for Normal Range represent severe disability.

The Normal Range outcome criteria were defined in the Lancet paper as participant ratings for the Primary Outcome Measures of:
• Chalder Fatigue Questionnaire (CFQ) ≤18 points (scale: 30 to 0, 0=no fatigue)
• Short Form 36 physical function subscale (SF36pf) ≥60 points (scale: 0 to 100, 100=no limitations)

Numbers meeting Normal Range criteria were: CBT 44 (29∙7%), GET 42 (27∙3%), and Specialist Medical Care (the control group) 22 (14∙5%), which produce NNT’s of 7 for CBT and 8 for GET.

Disturbingly, the PACE Trial data shows that 69 (53%) of participants who were Normal Range at outcome, still met the Oxford CFS research criteria required for enrolment.[6] Therefore more than half of the Normal Range participants were still suffering from fatigue that was: “severe, disabling, and affects physical and mental functioning”. Yet these participants were represented in the media as having, “got back to normal” or “recovered sufficiently to resume normal lives”. These contradictions show that the Normal Range construct is spurious.

The short-form 36 physical function subscale
The PACE Trial Normal Range criteria required an SF36pf rating of 60 or more. Yet any rating of 70 or less corresponds with authoritative published descriptions indicating severe impairment:

SF36pf of 70: diagnosis of CFS. Reeves,[7] Wearden,[8] White,[9]
SF36pf of 70: “severely disabled”. Knoop,[10]
SF36pf of 65: “abnormal levels of physical function”. White,[9]
SF36pf of 65: “severe functional impairment”. Bleijenberg,[11]

And for comparison
SF36pf of 58 = the average for 75 to 84 years old. Bowling,[12]

Therefore suggesting that the 25% of Normal Range participants who had SF36pf ratings of 60, 65 or 70 had, “got back to normal”, is patently false.

Controlled versus uncontrolled outcome data.
Control group adjusted and unadjusted figures for participants reaching ‘Normal Range’ or a ‘clinically useful difference’ are shown below.

“Normal Range” % of participants:
GET control adjusted: 13∙4, unadjusted: 27∙9
CBT control adjusted: 15∙3, unadjusted: 29∙7

“Clinically useful difference” % of participants:
GET control adjusted: 16∙3, unadjusted: 61∙0
CBT control adjusted: 14∙0, unadjusted: 58∙8

(A ‘clinically useful difference’ was defined as improvement by ≥2 points for the CFQ (Likert scoring) and ≥8 points for the SF36pf. Normal Range was defined as, CFQ ≤18 points and SF36pf ≥60)

By publicising uncontrolled figures, the PACE Trial investigators effectively doubled the numbers that reached the thresholds for the Normal Range construct, and quadrupled the numbers who reached the thresholds for a ‘clinically useful difference’. i.e., claims that 30% ‘got back to normal’ and 60% ‘improved’ with GET or CBT, should actually state that GET and CBT resulted in only around 15% meeting these thresholds.

The PACE trial data shows that GET and CBT benefited around 15% of participants. Whereas 85% of participants derived no benefit that can be attributed to one of the therapies. These statistics which negate the use of GET and CBT for ‘CFS/ME’, do not even allow for factors such as lax participant selection, participation bias, dubious target thresholds, reliance on participant subjective measures or the trend for more than 50% of participants to spontaneously improve. Therefore, the PACE Trial has provided compelling statistical evidence that GET and CBT are useless as treatments for ME/CFS.

The evidence shows that media reporting of the PACE Trial was, “at odds with the research evidence”. It was false and misleading. But this may not be the fault of journalists, perhaps they were confused by information provided to them under the auspices of a respected organisation, which they believed could be trusted.

If medical leaders and the Royal Colleges gleaned their ideas about GET and CBT for ME/CFS from the media, rather than from critical examination of the largest controlled clinical trial for these therapies, no wonder they are irritated with NICE. Yet the evidence shows that NICE were entirely correct to ban GET and restrict CBT – even though they might conceivably have done it for the wrong reasons.

1. National Institute for Health and Care Excellence. 2021. NICE guideline [NG206]. Myalgic encephalomyelitis (or encephalopathy)/chronic fatigue syndrome: diagnosis and management. Published: 29 October 2021. Available online at:
2. Science Media Centre. Expert reaction to updated NICE guideline on diagnosis and management of ME/CFS. Available online at: Accessed Nov 4, 2021.
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 Mar 5;377(9768):823-36. doi:10.1016/S0140-6736(11)60096-2. Available online at: Accessed Nov 4, 2021.
4. Science Media Centre. Expert reaction to Lancet study looking at treatments for Chronic Fatigue Syndrome/ME. Available online at: Accessed Nov 4, 2021.
5. ABC Radio Australia. Comparison of treatments for chronic fatigue syndrome - the PACE trial. Health Report. April 18th 2011. Available online at: Accessed Nov 4, 2021.
6. Sharpe MC, Archard LC, Banatvala JE, et al. 1991. A report--chronic fatigue syndrome: guidelines for research. J R Soc Med. 1991 Feb;84(2):118-121. Available online at: Accessed Nov 4, 2021.
7. Reeves, W., Wagner, D., Nisenbaum, R. et al. Chronic Fatigue Syndrome - A clinically empirical approach to its definition and study. BMC Med. 2005;3:19. doi: 10.1186/1741-7015-3-19.
8. Wearden AJ, Riste L, Dowrick C, Chew-Graham C, Bentall RP, Morriss RK, Peters S, Dunn G, Richardson G, Lovell K, Powell P. FINE Trial Protocol. BMC med. 2006,4:9. doi: 10.1186/1741-7015-4-9.
9. White PD, Goldsmith K, Johnson AL, Chalder T, Sharpe M. Recovery from chronic fatigue syndrome after treatments given in the PACE trial. Psychol Med. 2013 Oct; 43(10):2227-2235. doi: 10.1017/S0033291713000020.
10. Tummers M, Knoop H, van Dam A, Bleijenberg G. Implementing a minimal intervention for chronic fatigue syndrome in a mental health centre: a randomized controlled trial. Psychol Med. 2012 Oct;42(10):2205-15. doi:10.1017/S0033291712000232.
11. Stulemeijer M, de Jong LWAM, Fiselier TJW, Hoogveld SWB, Bleijenberg G. Cognitive behaviour therapy for adolescents with chronic fatigue syndrome: randomised controlled trial. BMJ. bmj.38301.587106.63v1. doi:
12. Bowling A, Bond M, Jenkinson C, Lamping DL. Health Survey England 1996. J Public Health. (1999) 21 (3):255-270.

Competing interests: No competing interests

05 November 2021
Peter F. Kemp
counsellor, researcher
London, UK