Acupuncture is one of the most studied parts of Traditional Chinese Medicine (TCM), and also one of the most debated. During the three decades Sherman Lai practised acupuncture in Guelph, Ontario (1986–2017), the research went from a handful of small studies to large trials and systematic reviews. This article explains, in plain language, what some of the main reviews have concluded and where the open questions remain. It is an educational overview, not a recommendation for any treatment.
How acupuncture is studied
Testing acupuncture is harder than testing a pill. A drug can be compared against an identical-looking placebo, but it is difficult to give someone a convincing "fake" acupuncture treatment. Researchers use sham acupuncture instead: needles placed at points not used in TCM, needles that retract into the handle without piercing the skin, or very shallow needling. Each type of sham may have some physical effect of its own, which makes results harder to interpret. The 1997 NIH consensus panel described this difficulty in finding appropriate controls as a central problem for the field, and it is still discussed today.
Trials also differ in what they compare. Some compare acupuncture with no acupuncture (usual care alone), some with sham acupuncture, and some with standard drug treatment. The answer often depends on which comparison is being made.
The 1997 NIH consensus statement
In November 1997, a panel convened by the U.S. National Institutes of Health reviewed the evidence then available. It concluded that there was clear evidence of efficacy for nausea and vomiting after surgery and chemotherapy in adults, and for pain after dental surgery. It listed a longer set of conditions, including headache, low-back pain and osteoarthritis, where acupuncture might be useful as an adjunct or as part of a broader management program. The panel also noted that many studies were small or poorly designed and gave equivocal results. The statement is often cited as the moment acupuncture entered mainstream North American research.
Chronic pain: the largest analyses
The Acupuncture Trialists' Collaboration pooled raw patient data from randomized trials of acupuncture for chronic pain. Its 2018 update, published in The Journal of Pain, covered 39 trials and 20,827 patients with back or neck pain, osteoarthritis, chronic headache or shoulder pain. The authors found that acupuncture was associated with better outcomes than both no acupuncture and sham acupuncture. They also found that the difference between real and sham acupuncture was relatively modest, much smaller than the difference between acupuncture and no acupuncture.
The U.S. National Center for Complementary and Integrative Health (NCCIH) summarizes similar findings. It reports that acupuncture may help with low-back pain, neck pain and knee pain from osteoarthritis. For neck pain, the advantage over sham treatment was smaller than the advantage over no treatment.
Migraine and headache
A 2016 Cochrane review by Linde and colleagues looked at acupuncture for preventing episodic migraine. The authors concluded that acupuncture can be considered a treatment option for patients willing to undergo it, and that it appeared at least comparable to several preventive drugs in the trials available. They also noted that long-term studies lasting more than a year were lacking. Cochrane reviews are regularly updated, so their conclusions can change as new trials appear.
Safety
NCCIH notes that serious complications from acupuncture are relatively rare when it is performed by a trained practitioner using sterile needles. Reported problems have usually involved non-sterile needles or improper technique. In the United States, the Food and Drug Administration regulates acupuncture needles as medical devices. In Ontario, the practitioners themselves have been regulated since 2013, a history covered in A Short History of Acupuncture and Why Ontario Regulated Traditional Chinese Medicine.
What remains uncertain
Researchers still disagree about how much of acupuncture's effect is specific to needle placement and how much comes from the treatment setting, expectation and attention. Results also vary by condition. For many conditions, the evidence is limited or inconclusive. A review that finds a benefit for one condition says nothing about others.
Sherman Lai often described TCM as something to be understood alongside Western clinical science, not in place of it. Reading the research honestly, including its limits, is part of that. More on how TCM approaches diagnosis is in How Traditional Chinese Medicine Diagnoses Illness.
Sources
- National Center for Complementary and Integrative Health (NIH). Acupuncture: Effectiveness and Safety.
- NIH Consensus Development Program. Acupuncture. Consensus statement, November 3–5, 1997.
- Vickers AJ, Vertosick EA, Lewith G, et al. Acupuncture for chronic pain: update of an individual patient data meta-analysis. J Pain. 2018;19(5):455–474.
- Linde K, Allais G, Brinkhaus B, et al. Acupuncture for the prevention of episodic migraine. Cochrane Database of Systematic Reviews. 2016, Issue 6. CD001218.