I have written the following for a producer of clinical guidance. It is in response to a request from a doctor (who claims expertise in managing ME/CFS) for acupuncture to be recommended as a treatment for ME/CFS in an update of clinical guidance. Suggestions for improvement welcome. If it is of use to anyone, anytime, by all means use it.
There is no credible evidence that acupuncture is useful in the management of ME/CFS. The 2021 NICE ME/CFS Guideline considered the evidence for acupuncture and did not recommend it for the treatment or management of ME/CFS. The pain management guideline it links to (for neuropathic pain) does not recommend acupuncture.
I have reviewed the 2022 review paper the doctor provides as evidence for the recommendation: Acupuncture and moxibustion for chronic fatigue syndrome: A systematic review and network meta-analysis, 2022, Fang et al.
It has a number of problems that make its conclusion that acupuncture is effective and safe for ME/CFS poorly founded.
1. The selection criteria used does not require post-exertional malaise, the diagnostic symptom of ME/CFS. Instead, it just requires studies to have selected participants who have had significant fatigue for more than 6 months. It is therefore likely that most of the participants did not have ME/CFS.
2. Some of the studies have not been published and most originate from institutions with a vested interest in proving the effectiveness of these therapies. Many of the studies cannot be found online.
3.The review noted that the reliability of some studies was affected by attrition. People who are not getting any benefit from a therapy will often not make it to the end of a treatment programme, whereas people who do feel that they are improving are more likely to be among those who participate in the final assessment.
4. The reported benefit in the main symptom investigated, fatigue, was not a clinically relevant improvement. Based on 31 studies, acupuncture achieved a mean difference improvement of scores over no treatment of 2.3. The review does not give details of the version of the Chalder Fatigue Scale and the scoring system used by individual studies. The scale range may therefore be anything from 0-11 to 0-42. However, an improvement of 2.3 on any of the scales, especially in the context of considerable experimental design and implementation issues, is not persuasive or clinically relevant.
5. Some studies used a ‘care as usual’ control, some used a sham acupuncture technique and some used non-acupuncture points. Some studies (40/51) did not have a placebo control at all, and simply compared active treatments (e.g. acupuncture vs a herbal remedy). Treatment effectiveness was triangulated, with the acupuncture versus controls studies serving as the main foundation for the analysis. Therefore, the reliability or otherwise of the average outcome in control arms is key to the credibility of the review's findings.
'Care as usual' is essentially a nocebo treatment; people feel that they are missing out on the useful treatment and tend to evaluate their health pessimistically. They may even be incentivised to report poor ongoing health in order to qualify for the desired treatment after the study. The sham acupuncture technique has been criticised as not providing robust blinding; people can often tell that their skin has not been pierced and the therapist of course knows if they are delivering a real or fake therapy. Informed participants may realise if non-acupuncture points are used, especially given the repeated sessions and access to information on the internet, and again, the therapist will know if they are providing the 'real' or 'fake' treatment and may unknowingly convey that to the participant. It is therefore very likely that the data from the control arms was not reliable and helped to overstate the benefit from the treatments.
6. The studies had multiple elements of poor experimental design. It is possible to make virtually any treatment appear successful if
a. the investigator's expectation about which treatment will work best is not hidden from the participants,
b. the measure is a subjective one, as it was in this study, e.g. has your fatigue improved?,
c. recruitment is highly selective, based on the participant’s belief in the treatment. Nearly all of the studies in the review were undertaken in China and all of the studies that I can find details for were undertaken by researchers with affiliations to Chinese institutions.
d. the participant’s belief in the credibility of the treatment is further bolstered e.g. by holding the trial in a hospital dedicated to providing the treatment. For example, an acupuncture treatment in the clinic of a University of Traditional Chinese Medicine, as many of these were,
e. a therapeutic alliance is developed, by having the participants have numerous sessions with a kind therapist committed to the treatment. Such an alliance makes it harder to report that the therapy was a waste of time. This can be even more effective if the person is reliant on the institution for ongoing health care.
f. the participants are asked if the attribute has improved immediately after the treatment ends while the therapeutic alliance and hope for improvement remain strong, rather than waiting a number of weeks for followup.
6. The acupuncture points used in the studies were not consistent. There are papers, see Li et al, 2023 for example, that suggest that acupuncture has a biological effect, and that this biological effect has some relevance to problems identified in people with ME/CFS. Neither of those claims are founded in evidence; the studies are low quality. This chain of reasoning often goes via findings from studies of so-called ‘animal models of CFS’. We have seen considerable numbers of these studies, see Zhong et al, 2023 for example, that torture rodents with extended and repeated periods of activity (e.g. swimming in a tank where if they stop they drown combined with sleep deprivation) and cherry-picking from the many biomarkers measured. There is no reason to think that such ‘animal models’ have any relevance to ME/CFS at all.
Some doctors may perhaps feel that, given there is no evidence-based treatment for ME/CFS, it is fine to suggest treatments that are no more than elaborate placebos. They may think that a placebo treatment may help someone with ME/CFS. There is no evidence that that is true. It would also be ethically unsound to knowingly prescribe a placebo without telling the patient.
The types of outcomes consistently reported by acupuncture studies in ME/CFS, short-term sub-clinical improvements, are similarly reported in acupuncture studies for other health issues. I highly recommend the editorial Acupuncture Is Theatrical Placebo in Anasthesia and Analgesia to the [guideline development] staff. Some excerpts from that editorial follow:
“There is now unanimity that the benefits, if any, of acupuncture for analgesia, are too small to be helpful to patients.
Large multicenter clinical trials conducted in Germany7–10 and the United States11 consistently revealed that verum (or true) acupuncture and sham acupuncture treatments are no different in decreasing pain levels across multiple chronic pain disorders: migraine, tension headache, low back pain, and osteoarthritis of the knee.”
“it is hard to see why acupuncture is still used. Certainly, such an accumulation of negative results would result in the withdrawal of any conventional treatment.”
“the Oxford Centre for Evidence-Based Medicine updated its analysis of acupuncture for back pain. Their verdict21 was “Clinical bottom line. Acupuncture is no better than a toothpick for treating back pain.””
If we are aiming to provide a cost-effective evidence-based health system, and I'm sure that you will agree we are, acupuncture has no place in our treatment recommendations. Doctors who currently recommend it need to carefully look at the evidence and reconsider their practice.
Recommending ineffective treatments harms patients who are often least able to afford to waste money and effort. For a person with mild to moderate ME/CFS, attending an acupuncture appointment may be the only activity they do outside the home in a week, and it may result in days in bed recovering from the exertion. Recommending ineffective treatments also harms individual doctors and the medical profession as a whole, as patients may lose faith that their doctor makes well-founded treatment decisions for them. If doctors can recommend any sort of unevidenced pseudoscience, then why should what they say about, for example, vaccines in a pandemic be trusted?
Last edited:
