Graded exercise therapy and pacing for ME/CFS: a contested evidence base

Myalgic encephalomyelitis (ME), also known as chronic fatigue syndrome (CFS), is a complex and often debilitating condition. For many years, certain treatments like graded exercise therapy (GET) and cognitive behavioural therapy (CBT) have been considered for ME/CFS management. However, the evidence supporting these approaches has become a subject of significant scientific and patient community debate [1, 3].

This discussion involves different conclusions drawn by patient organisations, researchers, and guideline bodies from overlapping data [1, 3]. Over time, a deeper understanding of ME/CFS, particularly a key symptom called post-exertional malaise, has led to a re-evaluation of treatment recommendations.

Understanding Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS)

ME/CFS is a serious disorder marked by persistent fatigue that worsens after exertion [2]. People with ME/CFS also experience substantial symptoms related to cognitive function, immune system changes, and how their automatic bodily functions work [2].

There is no specific diagnostic test for ME/CFS [2]. Instead, healthcare professionals use a set of diagnostic criteria to identify the condition [2]. The number of people diagnosed with ME/CFS can vary depending on which diagnostic criteria are used [2]. Current treatment strategies primarily focus on relieving symptoms and helping individuals improve their daily function [2]. One treatment option that has been explored is exercise therapy [2].

The PACE Trial: Initial Findings and Interpretations

The PACE trial, published in The Lancet in 2011, was a large study designed to assess the effectiveness and safety of four different approaches for chronic fatigue syndrome [1]. This trial aimed to compare specialist medical care (SMC) alone with SMC combined with adaptive pacing therapy (APT), cognitive behavioural therapy (CBT), or graded exercise therapy (GET) [1].

Study Design and Participants

The trial was a parallel-group randomised study, meaning participants were randomly assigned to one of the four treatment groups [1]. Researchers recruited 641 eligible patients from six secondary-care clinics across the UK, and those patients met the Oxford criteria for chronic fatigue syndrome [1]. Of those 641, 160 were assigned to the APT group, 161 to the CBT group, 160 to the GET group, and 160 to the SMC-alone group [1].

That entry criterion is worth pausing on, because it sits at the centre of the dispute. The Oxford criteria define the condition primarily by unexplained, disabling fatigue, and — unlike some other case definitions — do not require post-exertional malaise to be present. Critics argue that this can admit people with chronic fatigue from other causes alongside people with ME/CFS, so that results from an Oxford-recruited sample may not describe the population most affected by post-exertional malaise. The trial did report subgroup analyses in more narrowly defined groups: 427 participants meeting international criteria for chronic fatigue syndrome and 329 meeting London criteria for myalgic encephalomyelitis, which the authors reported yielded equivalent results [1].

The primary outcomes measured were fatigue, using scores from the Chalder fatigue questionnaire, and physical function, using scores from a subscale of the short form-36 questionnaire [1]. These outcomes were assessed up to 52 weeks after participants were randomised to a treatment group [1]. Safety was primarily assessed by recording all serious adverse events, including any serious negative reactions to the trial treatments [1]. Participants knew which treatment they were receiving, but the statistician analysing the primary outcomes did not [1].

Reported Findings

The PACE trial reported the following results when comparing the treatments:

• Fatigue Scores: Compared to specialist medical care alone, participants receiving CBT had mean fatigue scores that were 3.4 points lower at 52 weeks (with a 95% confidence interval of 1.8 to 5.0 points lower) [1]. For GET, mean fatigue scores were 3.2 points lower (95% CI 1.7 to 4.8 points lower) compared to SMC alone [1]. In contrast, APT did not show a significant difference in fatigue scores compared to SMC alone [1].
• Physical Function Scores: Compared to SMC alone, participants in the CBT group had mean physical function scores that were 7.1 points higher (95% CI 2.0 to 12.1 points higher) [1]. For GET, mean physical function scores were 9.4 points higher (95% CI 4.4 to 14.4 points higher) compared to SMC alone [1]. APT did not show a significant difference in physical function scores compared to SMC alone [1].
• Comparisons with Adaptive Pacing Therapy: When compared directly with APT, both CBT and GET were associated with less fatigue and better physical function [1].
• Subgroup Analysis: The study also performed a subgroup analysis on participants meeting international criteria for chronic fatigue syndrome and those meeting London criteria for myalgic encephalomyelitis, yielding similar results [1].
• Serious Adverse Reactions: The trial reported a low number of serious adverse reactions across all groups [1]. Two out of 159 participants (1%) in the APT group, three out of 161 (2%) in the CBT group, two out of 160 (1%) in the GET group, and two out of 160 (1%) in the SMC-alone group experienced serious adverse reactions [1].

Original Interpretation

Based on these findings, the authors of the PACE trial concluded that both CBT and GET could be "safely added to SMC to moderately improve outcomes for chronic fatigue syndrome" [1]. They also concluded that APT was not an effective addition [1].

Criticisms and Reanalysis of the PACE Trial

Despite the initial conclusions of the PACE trial, its findings and methodology became a subject of significant debate and criticism, particularly from patient communities and other researchers [1, 3]. Patient organisations, for example, had reported that treatments like CBT and GET could be harmful for some individuals, and instead favoured pacing and specialist health care [1].

The Core Concern: Protocol Deviations

A central criticism of the PACE trial was that its reported analyses did not consistently follow the procedures that had been set out in its own published protocol [3]. Researchers argued that it was unclear whether the trial's conclusions were fully supported by the evidence, given these changes [3].

The Wilshire Reanalysis

In 2018, a reanalysis of the PACE trial data was published in BMC Psychology by Wilshire and colleagues [3]. This reanalysis aimed to address the concerns about protocol deviations by attempting to closely approximate the original, protocol-specified procedures [3]. The researchers were able to do this using data obtained through a Freedom of Information request [3].

Why Outcome Definition Changes Matter

When a study is designed, researchers define specific ways to measure whether a treatment is working. These are called "outcome measures." The protocol is a detailed plan that outlines exactly how these measures will be collected and analysed. If these definitions are changed after the study has started or after data has been collected, it can affect the results and how they are interpreted. For example, changing the threshold for what counts as "recovery" or "improvement" can make a treatment appear more or less effective than it would have seemed under the original rules. This is why adhering to the original protocol is considered important for the scientific rigour and unbiased reporting of a trial.

Findings of the Reanalysis

The Wilshire reanalysis reported different results when applying the original protocol-specified procedures:

• Overall Improvement Rates: On the original protocol-specified primary outcome measure, which looked at overall improvement rates, there was a significant effect of treatment group [3]. However, the groups receiving CBT or GET did not significantly outperform the Control group after accounting for the number of comparisons specified in the trial protocol [3].
• Recovery Rates: The reanalysis found that rates of recovery were consistently low across all treatment groups and were not significantly different between them [3].
• Secondary Measures: Significant effects observed on secondary measures were almost entirely confined to self-report measures, meaning they were based on what participants reported about their own symptoms [3]. These effects also did not last beyond two years [3].

Conclusion of the Reanalysis

The authors of the Wilshire reanalysis concluded that their findings "raise serious concerns about the robustness of the claims made about the efficacy of CBT and GET" [3]. They suggested that the modest treatment effects observed on self-report measures in the PACE trial could reasonably be accounted for by biases in how participants reported their symptoms [3].

The Cochrane Review on Exercise Therapy for CFS

The Cochrane Database of Systematic Reviews publishes a review of exercise therapy for chronic fatigue syndrome; the current version of that record is the 2024 one [2]. Cochrane reviews are known for their rigorous methods in evaluating medical evidence.

Review Objectives and Methods

The main objective of this review was to determine the effects of exercise therapy for adults with CFS [2]. It compared exercise therapy to any other intervention or control group, looking at outcomes such as fatigue, adverse events, pain, physical functioning, quality of life, mood disorders, sleep, self-perceived changes in overall health, health service use, and participant dropout [2].

The review included eight randomised controlled trials (RCTs) involving a total of 1518 participants [2]. The exercise therapies in these studies typically lasted from 12 to 26 weeks [2]. Most studies used aerobic exercise, such as walking, swimming, cycling, or dancing, at varying intensity levels [2]. Control groups included passive controls (like usual care, relaxation, or flexibility exercises), cognitive behavioural therapy (CBT), cognitive therapy, supportive listening, pacing, pharmacological treatment, and combination treatments [2].

The review authors assessed the certainty of the evidence using the GRADE approach [2]. This system classifies evidence certainty as high, moderate, low, or very low, reflecting how confident the researchers are in the findings. Many studies in the review had a low risk of selection bias, but all had a high risk of performance and detection bias [2]. This means participants and researchers were often aware of who was receiving which treatment, which can influence reported outcomes.

Key Findings of the Cochrane Review

The Cochrane review presented its findings by comparing exercise therapy to different control conditions:

Exercise Therapy Compared with 'Passive' Control (e.g., Usual Care, Relaxation)

• Fatigue: Exercise therapy probably reduces fatigue at the end of treatment [2]. This was based on moderate-certainty evidence, with an estimated mean difference of -3.4 points on a 0-33 scale [2]. However, the review found that it was uncertain whether fatigue was reduced in the long term, due to very low-certainty evidence [2].
• Serious Adverse Reactions: The certainty of evidence regarding the risk of serious adverse reactions was very low, making it uncertain if exercise therapy affected this outcome [2].
• Physical Functioning: Exercise therapy may moderately improve physical functioning at the end of treatment [2]. However, the long-term effect on physical functioning was uncertain due to very low-certainty evidence [2].
• Sleep: Exercise therapy may slightly improve sleep at the end of treatment and in the long term [2].
• Other Outcomes: The effect of exercise therapy on pain, quality of life, and depression was uncertain because the evidence was either missing or of very low certainty [2].

Exercise Therapy Compared with Cognitive Behavioural Therapy (CBT)

• Fatigue: Exercise therapy may make little or no difference to fatigue when compared to CBT, both at the end of treatment (low-certainty evidence) and at long-term follow-up (moderate-certainty evidence) [2].
• Serious Adverse Reactions: The certainty of evidence for serious adverse reactions was very low, meaning it was uncertain if there was a difference between exercise therapy and CBT [2].
• Other Outcomes: The available evidence suggested there may be little or no difference between exercise therapy and CBT in physical functioning or sleep (low-certainty evidence) [2]. There was probably little or no difference in the effect on depression (moderate-certainty evidence) [2]. The review was uncertain if exercise therapy improved quality of life or reduced pain compared to CBT, as the evidence was of very low certainty [2].

Exercise Therapy Compared with Adaptive Pacing

• Fatigue: Exercise therapy may slightly reduce fatigue compared to adaptive pacing at the end of treatment (mean difference -2.00 points on a 0-33 scale) and at long-term follow-up (mean difference -2.50 points on a 0-33 scale) [2]. This was based on low-certainty evidence [2].
• Serious Adverse Reactions: The certainty of evidence for serious adverse reactions was very low, making it uncertain if there was a difference between exercise therapy and adaptive pacing [2].
• Other Outcomes: The available evidence suggested that exercise therapy may slightly improve physical functioning, depression, and sleep compared to adaptive pacing (low-certainty evidence) [2]. No studies reported on quality of life or pain in this comparison [2].

Exercise Therapy Compared with Antidepressants

• The review found that it was uncertain if exercise therapy, either alone or combined with antidepressants, reduced fatigue and depression more than antidepressants alone [2]. This was due to very low-certainty evidence [2]. The one study included in this comparison did not report on adverse reactions, pain, physical functioning, quality of life, sleep, or long-term results [2].

Authors' Conclusions and Editorial Note

The authors of the Cochrane review concluded that exercise therapy "probably has a positive effect on fatigue in adults with CFS compared to usual care or passive therapies" [2]. However, they also stated that the evidence regarding adverse effects is uncertain [2]. Due to limited evidence, they found it difficult to draw conclusions about the comparative effectiveness of exercise therapy against CBT, adaptive pacing, or other interventions [2]. The review also noted that all included studies used either the 1994 Centers for Disease Control and Prevention criteria or the Oxford criteria for diagnosis, or both, and that patients diagnosed using other criteria might experience different effects [2].

The status of this review matters as much as its findings

The current version of this Cochrane review carries an editorial note dated 19 December 2024 (amended 31 January 2025 and 25 March 2025), and it is worth reading carefully [2].

The note states that the review was published in 2019 and includes studies from searches up to 9 May 2014 [2]. It also states that a pilot project for engaging interest holders in developing an update of the review was initiated on 2 October 2019 and has now been discontinued [2].

Two things follow from that, and readers should weigh both. First, the evidence this review summarises stops more than a decade ago, which is before much of the current understanding of post-exertional malaise was reflected in guidance. Second, the update that was announced in 2019 is not underway — that project has been discontinued. This is a significant part of why the review remains contested, and it is a large part of the objection raised by patient organisations: the review's conclusions continue to be cited while the evidence base behind them has not been revisited.

Post-Exertional Malaise (PEM)

Post-exertional malaise (PEM) is a key symptom of ME/CFS and is central to the controversy surrounding graded exercise therapy. PEM is a worsening of symptoms after activity. NICE guideline NG206 describes that worsening as one that is often delayed in onset by hours or days, is disproportionate to the activity, and has a prolonged recovery time that may last hours, days, weeks or longer [4]. Because the reaction is delayed and out of proportion to the effort, it can be very hard to predict and manage energy levels. Understanding PEM is vital because it highlights why an approach like graded exercise, which involves increasing activity levels, might be problematic for some individuals with ME/CFS.

The Shifting Landscape: NICE Guideline NG206 (2021)

The ongoing scientific debate and the patient community's experiences have led to significant changes in official recommendations for ME/CFS management. The National Institute for Health and Care Excellence (NICE) in the UK, an independent body that produces evidence-based guidance for health and social care, published guideline NG206 in 2021 [4].

This guideline represents a pivotal shift in how ME/CFS is understood and managed. NG206 states that people with ME/CFS should not be offered generalised physical activity or exercise programmes, nor "any programme that does not follow the approach in recommendation 1.11.13 or that uses fixed incremental increases in physical activity or exercise, for example, graded exercise therapy" [4]. It also says people should not be offered programmes based on deconditioning and exercise-avoidance theories as perpetuating ME/CFS, and it states there is no cure for ME/CFS [4]. Instead, the guideline places post-exertional malaise and energy management at the centre of care [4].

This change reflects a growing understanding that for many people with ME/CFS, pushing through symptoms or attempting to gradually increase activity levels can worsen their condition due to PEM. The updated guidance underscores the importance of individualised care that respects the unique symptom patterns and limitations experienced by each person with ME/CFS, with a focus on avoiding exacerbations of symptoms.

What this means for you

Myalgic encephalomyelitis/chronic fatigue syndrome is a complex condition, and each person's experience with it is unique. The evidence base for treatments like graded exercise therapy and pacing has been thoroughly reviewed and debated, leading to different conclusions and shifts in professional guidance.

Making decisions about managing your ME/CFS should always involve a regulated health professional who understands your medical history and current symptoms. They can help you understand the potential benefits and risks of different approaches in the context of your individual health needs.

Any plan for physical activity or energy management should be carefully developed with your healthcare provider. This plan should be built around your own symptom response, with particular attention to avoiding and managing post-exertional malaise. It is important to listen to your body and adjust activities as needed. Do not start, continue, or stop any exercise programme or treatment without first discussing it with a qualified healthcare professional.