Review Rehabilitative efficacy of cognitive behavioral therapy-based intervention for chronic fatigue syndrome, 2026, Jiang et al

[forestglip]

Rehabilitative efficacy of cognitive behavioral therapy-based intervention for chronic fatigue syndrome

Jiang, Xiaoyu; Liu, Junnan; Li, Shulun; Pei, Qian; Huang, Qiang

Background
Cognitive behavioral therapy (CBT) has been used for chronic fatigue syndrome (CFS), but previous studies have shown inconsistencies and divergent long-term outcomes. This study aims to assess the effectiveness of CBT in improving fatigue, physical functioning, and psychological health in patients with CFS.

Methods
We searched PubMed, Web of Science, Embase, and the Cochrane Library from their inception to March 2025. Studies designed as randomized controlled trials, enrolling patients diagnosed with CFS who received CBT as a treatment intervention delivered face-to-face, in groups, or online were included. The primary outcome was the fatigue score, and the secondary outcomes were the physical function score, psychological outcomes, and adverse events. The quality of the included studies was evaluated using the Cochrane Assessment Tool for Risk of Bias Version 2 (RoB 2; The Cochrane Collaboration). Heterogeneity was detected by subgroup analysis and sensitivity analysis.

Results
Fifteen randomized controlled trials with 2270 participants were involved. Meta-analysis demonstrated that CBT yielded significant posttreatment improvements across key outcomes. Fatigue severity was notably reduced, as measured by the Chalder Fatigue Questionnaire (standardized mean difference [SMD] = −0.44, 95% confidence interval [CI]: −0.57–−0.32) and the Checklist Individual Strength (SMD = −0.88, 95% CI: −1.16–−0.60).

Physical functioning (physical function subscale of the Short Form Health Survey-36) subscale also improved significantly (SMD = 0.32, 95% CI: 0.23–0.41). Psychological health benefits were evident through reduced Symptom Checklist 90 (SMD = −0.44, 95% CI: −0.62–−0.26) and specific decreases in anxiety (Hospital Anxiety and Depression Scale–anxiety scale: SMD = −0.28, 95% CI: −0.46–−0.10) and depression (Hospital Anxiety and Depression Scale–depression scale: SMD = −0.27, 95% CI: −0.45–−0.09).

Follow-up analyses confirmed sustained benefits: short-term (≤ 6 months) showed continued fatigue reduction (Chalder Fatigue Questionnaire: SMD = −0.29, 95% CI: −0.55–−0.03) and improved physical function (physical function subscale of the Short Form Health Survey-36: SMD = 0.31, 95% CI: 0.03–0.58); long-term (6–24 months) maintained fatigue relief (SMD = −0.35, 95% CI: −0.50–−0.19) and functional gains (SMD = 0.21, 95% CI: 0.06–0.37). Safety analysis showed no statistically significant difference in adverse event rates (odds ratio = 0.61, 95% CI: 0.36–1.01, P = .060).

Conclusion
Based on our results, CBT may be an effective option for CFS, but further research is warranted.

Web | DOI | PDF | Medicine | Open Access

 

[forestglip]

Another one for the collection:

Components and Delivery Formats of [CBT] for [CFS/ME]: A Systematic Review and Component Network Meta-Analysis 2026 Wang et al

Cognitive behavioural therapy for the treatment of chronic fatigue syndrome in adults – a meta-analysis, 2025, Kolala et al

Efficacy and Acceptance of Cognitive Behavioral Therapy in Adults with Chronic Fatigue Syndrome: A Meta-analysis, 2024, Maas genannt Bermpohl et al.

Does the effect of CBT for CFS (ME/CFS) vary by patient characteristics? A systematic review and individual patient data meta-analysis, 2023, Knoop

A systematic literature review of randomized controlled trials evaluating prognosis following treatment for adults with CFS, 2022, Chalder

Cochrane Review: Cognitive behaviour therapy for chronic fatigue syndrome in adults, 2008, Price et al

 

[forestglip]

7 studies[29,40,41,45,46,48,50] implemented adequate blinding and were rated low risk in this domain.

There are all the studies that the authors say have adequate blinding:

[29] Gotaas ME, Stiles TC, Bjørngaard JH, Borchgrevink PC, Fors EA. Cognitive behavioral therapy improves physical function and fatigue in mild and moderate chronic fatigue syndrome: a consecutive randomized controlled trial of standard and short interventions. Front Psychiatry. 2021;12:580924. [Link]

The control group was a waiting-list group in which the participants waited for 16 weeks after baseline before they were post-scored and then offered 8 weeks of I-CBT outside the study.

Participants and therapists could not be masked to treatment allocation.

[40] Chalder T, Deary V, Husain K, Walwyn R. Family-focused cognitive behaviour therapy versus psycho-education for chronic fatigue syndrome in 11- to 18-year-olds: a randomized controlled treatment trial. Psychol Med. 2010;40:1269–79. [Link]

Sixty-three 11- to 18-year-olds (43 girls, 20 boys) with CFS were randomly assigned to either family-focused CBT or psycho-education delivered over 6 months.

[41] Janse A, Worm-Smeitink M, Bleijenberg G, Donders R, Knoop H. Efficacy of web-based cognitive-behavioural therapy for chronic fatigue syndrome: randomised controlled trial. Br J Psychiatry. 2018;212:112–8. [Link]

A total of 240 patients with CFS were randomised to either iCBT with protocol-driven therapist feedback or with therapist feedback on demand, or a waiting list.

[45] 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;10:iii–v, ix. [Link]

This report describes the conduct and results of a double-blind randomised controlled trial to compare group cognitive behavioural therapy (CBT) with education and support (EAS) and with standard medical care (SMC) for the treatment of patients with chronic fatigue syndrome/myalgic encephalopathy (CFS/ME).

[46] Prins JB, Bleijenberg G, Bazelmans E, et al. Cognitive behaviour therapy for chronic fatigue syndrome: a multicentre randomised controlled trial. Lancet. 2001;357:841–7. [Link]

The study was an open multicentre randomised controlled trial in which individual CBT was compared with participation in guided support groups and with the natural course, a control condition in which no treatment was offered.

[48] Sharpe M, Hawton K, Simkin S, et al. Cognitive behaviour therapy for the chronic fatigue syndrome: a randomized controlled trial. BMJ. 1996;312:22–6. [Link]

Interventions: Medical care comprised assessment, advice, and follow up in general practice. Patients who received cognitive behaviour therapy were offered 16 individual weekly sessions in addition to their medical care.

[50] 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;42:2205–15. [Link]

One hundred and twenty-three patients were randomly assigned to either guided self-instruction (n=62) or a waiting list (n=61).

I don't know how exactly the Cochrane Risk of Bias tool works, but I'm pretty sure the participants would know whether or not they are in the treatment arm when the control group is a waiting list or standard care.

 

[NelliePledge]

Oh yeah. Absolutely. Further research really is merited

 

[Sean]

Coz keeping the psycho-behavioural gravy train rolling along is the most important consideration here.

 

[Arnie Pye]

How often does CBT need to be tested for ME/CFS? Will it ever end?

 

[hotblack]

Still waiting on someone to study the psychobehavioural underpinnings of psychobehavioural proponents, let alone a meta analysis of multiple studies into this.

The study groups may not be blinded but the authors do seem blind to their lack of understanding of ME/CFS, the well described problems with these studies or the evidence that these approaches do not work. But hey, they wrote a paper and so that’s the important thing right? Not the harms it may support or trying to gain an understanding of something. I try to take a positive voew of people’s motivations but it It’s mind boggling.

 

[Utsikt]

I don't know how exactly the Cochrane Risk of Bias tool works, but I'm pretty sure the participants would know whether or not they are in the treatment arm when the control group is a waiting list or standard care.

I’ve looked at the part about blinding here. RoB and GRADE are very lenient systems with lots of room for «subjective» decisions to rate studies higher than they should be, but even then there is no way of making a coherent argument for why a study with subjective outcomes and no blinding should be rated as not high risk of bias in RoB.

 

[rvallee]

How often does CBT need to be tested for ME/CFS? Will it ever end?

That's the trick, the list of trials keep growing, but it never gets any larger. They went back decades, which would easily cover 500+ such trials, and yet, as is tradition, only picked 15.

There could be 1,000 more trials, and reviews would still only ever select 10-20, which is about the usual range for most reviews I have seen. The only exceptions are what NICE did, and to a lesser degree IQWIG, both of which reviewed most of the trials published, and is the reason someone, who is likely Wessely, tried to get the evidence review to be withheld. Not that it made any difference, this industry doesn't exist to produce anything, it's simply a self-perpetuating parasite.

So with only 15 trials, that means there's so much left to explore, basically nothing's been tried yet. Whether the real number is 500, 1000 or 5000, it doesn't matter, there are only ever 10-20 to review, or something like that.

Soon enough it will be easy to fully automate reviews like this. I'm really looking forward to do a meta analysis that spans the whole process, from trials to studies and textbooks to clinical programs. At this point, in the decades this has been going on, billions have been spent, hundreds of thousands have been made to go through this. But there's only ever about 15 studies reviewed, and more research is always needed.

I think it's probably enough for an entire Moon program with several missions. Not the whole Appollo stuff that included building everything from scratch, but this is basically a moonshot-sized boondoggle that never even lifted off a single millimetre.

 

[Sean]

I try to take a positive voew of people’s motivations

I don't anymore. Doesn't pay off. Cost/benefit ratio is too high.

Better with psycho-behavioural papers to start from the assumption that the authors are some combination incompetent, lazy, grifters, and frauds, until proven otherwise (which rarely happens). Saves a lot of time and disappointment.

There could be 1,000 more trials, and reviews would still only ever select 10-20, which is about the usual range for most reviews I have seen.

In fairness (yeah, I know), journals typically limit the number of references that can be cited. Which is no excuse for biased cherry picking, of course. If anything it makes it more important for the citations to be fairly selected and interpreted.

But it is a genuine limitation for even the most competent and honourable authors.

 

[TiredMathematician]

The PACE trial is double counted throughout this meta-analysis because Sharpe 2015 and White 2011 are reporting data from exactly the same patients (They also get the sample size wrong for Sharpe 2015 intervention group - 161 is the number of participants at baseline, not the number providing that data point.)

The studies of Gotaas 2021, Janse 2018, and Wiborg 2015 also violate the assumption of independence that exists for standard inverse variance weighted meta-analysis, as the a and b estimates are both compared to the same control group, leading to double counting of patients in the control group.

This is an elementary mistake, and also means their total participant count of 2270 (look, 12.7% greater than that in the prior meta-analysis!) overestimates by more than 300 the actual number of participants in the underlying trials.

 

[TiredMathematician]

Better with psycho-behavioural papers to start from the assumption that the authors are some combination incompetent, lazy, grifters, and frauds, until proven otherwise (which rarely happens). Saves a lot of time and disappointment.

I think this is actually a case where more likely than not the authors don't care one way or the other about CBT or ME/CFS – they've just done this because it's a relatively easy way to produce a publication (a lot easier than having to actually run a study yourself). As far as I can tell, none of these researchers have ever published a paper related to CBT or ME/CFS before.

There are thousands of meta-analyses published every year in crappy journals like this one by researchers who seem to have just picked the research question at the random (because they have no prior research in this area) and who aren't particularly competent in actually doing meta-analyses.

 

[Utsikt]

This is an elementary mistake.

Could you add a comment on pubpeer for instance? Seems like something other readers should know about!

 

[TiredMathematician]

Could you add a comment on pubpeer for instance? Seems like something other readers should know about!

Will do

Update: Done, but I expect several days until mods approve it.

PubPeer - Rehabilitative efficacy of cognitive behavioral therapy-base...

There are comments on PubPeer for publication: Rehabilitative efficacy of cognitive behavioral therapy-based intervention for chronic fatigue syndrome: A systematic review and meta-analysis (2026)

pubpeer.com

 

[rvallee]

In fairness (yeah, I know), journals typically limit the number of references that can be cited. Which is no excuse for biased cherry picking, of course. If anything it makes it more important for the citations to be fairly selected and interpreted.

For most papers, sure. For systematic reviews, it goes against the whole point of systematically reviewing what's been done.

There really is a need for meta analyses of the entire field, as there is currently no accurate way to know how many of those trials there are and general characteristics about them. This is simply not a credible way to do any of this. A small company would not even be able to function without knowing what activities it performs, it would simply go bankrupt. That there are lives decided by such a rotten systems is completely insane.

 

[Sean]

For most papers, sure. For systematic reviews, it goes against the whole point of systematically reviewing what's been done.

I agree that it is a silly rule, especially in the era of online publishing access when the limited space in the paper version of a journal is no longer a factor.

 

[Sean]

I think this is actually a case where more likely than not the authors don't care one way or the other about CBT or ME/CFS – they've just done this because it's a relatively easy way to produce a publication (a lot easier than having to actually run a study yourself). As far as I can tell, none of these researchers have ever published a paper related to CBT or ME/CFS before.

That would be the grifting element.