Skip to contentWillowOne's PositionsTagsWhat’s newMediaStatisticsSearch
WillowOne's Positions

Acupuncture

My Positions

  1. Acupuncture has no specific effect beyond placebo for any condition. In properly blinded trials it does not matter where the needles go, and it does not matter whether they go in at all.
  2. Qi and meridians do not exist. Two centuries of anatomy and physiology have found no trace of either, and the different national traditions cannot agree where the points are.
  3. "Five thousand years of Chinese wisdom" is marketing. The practice sold in the West is largely a 1950s construction, and the Chinese imperial medical academy had banned it in 1822.
  4. It is a theatrical placebo: elaborate, attentive, and occasionally dangerous. I would not fund it from public money, and I think NICE is wrong to still recommend it for chronic pain and migraine.

Notes

  1. The sham trials. The decisive test of acupuncture is to compare it with a convincing fake, and this has now been done many times. In the German GERAC trials (2007, over a thousand patients with chronic low back pain), real acupuncture at the correct points and sham acupuncture, needled shallowly at the wrong places, produced identical results, and both beat conventional care. A 2009 trial in Seattle went further and prodded patients with toothpicks that never broke the skin; the toothpicks worked as well as the needles. The largest meta-analysis of individual patient data (Vickers et al., 2012 and 2018, about twenty thousand patients) found a statistical difference between real and sham of roughly five points on a hundred-point pain scale, which is below anything a patient could notice and well within the bias you would expect when the person holding the needle cannot be blinded. The honest summary is the title of a 2013 editorial in Anesthesia & Analgesia: "Acupuncture is theatrical placebo".
  2. There is nothing there. Qi is a pre-scientific life force, and the meridians are channels for it that no dissection has ever found. The classical count of 365 points matched the days of the year and the twelve main meridians matched the great rivers of China, which tells you the system is numerology rather than anatomy. Chinese, Japanese and Korean acupuncture place the points differently; ear acupuncture, with its map of an inverted foetus on the auricle, was invented by a French doctor in the 1950s. When practitioners of a supposedly ancient and precise art cannot agree on its basic coordinates, and the trials show the coordinates do not matter, the two facts explain each other.
  3. The history is shorter than advertised. Needling appears in Chinese texts from around the second century BC, but it was one minor practice among many, and the Daoguang Emperor removed it from the Imperial Medical Academy in 1822 as unfit for gentlemen. Its revival was political. Mao's government, short of doctors, promoted "traditional Chinese medicine" as a cheap national resource for the barefoot-doctor programme, while Mao's own physician recorded that Mao did not use it himself. The West caught it in 1971, when a New York Times reporter travelling ahead of Nixon's visit wrote about his post-operative pain being relieved by needles, and in the staged demonstrations of "acupuncture anaesthesia" that followed, which were later found to involve sedatives, local anaesthetics and carefully selected patients.
  4. Why it feels like it works. Pain is the ideal condition for a placebo: it is self-reported, it fluctuates, and people seek help when it is at its worst (Regression to the mean). Add a long consultation with an attentive practitioner, an impressive ritual, and a cultural story about ancient wisdom, and the placebo response is as large as it ever gets. None of that is nothing, but none of it needs the needles, the meridians or the pretence that something is being treated. The same is true of Homeopathy, and the same honest alternative applies: a quiet room and a kind person.
  5. The harm. Needles that go into the chest can puncture a lung, and needles that are reused transmit hepatitis; a 2010 review documented 86 deaths after acupuncture. The larger harm is the one common to the whole field, which is delay and displacement: the cancer patient who tries needles first, and the NHS commissioning budget spent on a placebo that could have bought a physiotherapist. NICE dropped its recommendation for low back pain in 2016 on exactly this evidence, then recommended acupuncture for chronic primary pain in 2021 on trials that mostly lacked a credible sham arm. The second decision should follow the first.

Sources

  • Colquhoun, D. and Novella, S. P., "Acupuncture Is Theatrical Placebo", Anesthesia & Analgesia 116 (2013)
  • Haake, M. et al., "German Acupuncture Trials (GERAC) for Chronic Low Back Pain", Archives of Internal Medicine 167 (2007)
  • Cherkin, D. C. et al., "A Randomized Trial Comparing Acupuncture, Simulated Acupuncture, and Usual Care for Chronic Low Back Pain", Archives of Internal Medicine 169 (2009)
  • Vickers, A. J. et al., "Acupuncture for Chronic Pain: Update of an Individual Patient Data Meta-Analysis", Journal of Pain 19 (2018)
  • Ernst, E., "Deaths after acupuncture: A systematic review", International Journal of Risk & Safety in Medicine 22 (2010)
  • Simon Singh and Edzard Ernst, Trick or Treatment? Alternative Medicine on Trial (2008), chapter 2
  • NICE, Low back pain and sciatica in over 16s, NG59 (2016) and Chronic pain, NG193 (2021)

Also see