Research · Acupuncture, Honestly
Does Acupuncture Work?
What the largest trials, reviews and medical guidelines show, condition by condition, including where it doesn't help.
Acupuncture is one of the most studied complementary therapies. Whether it works depends on what you are asking it to do.
For chronic pain (back, neck and osteoarthritis) and for preventing migraine and tension-type headaches, acupuncture does better than both sham acupuncture and no acupuncture in large trials. Its advantage over sham is small, though: about a fifth of a standard deviation, against about half a standard deviation over no acupuncture. So much of the benefit seems to come from the treatment context, such as expectation, attention and the ritual of care.1 For IVF, quitting smoking and several other uses, trials show no advantage over sham.
This article goes through the evidence condition by condition, what medical guidelines say, what "sham acupuncture" is and why it matters, and where acupuncture has not been shown to help.
Does acupuncture work for chronic pain?
The largest analysis: 39 trials, 20,827 patients
The strongest evidence comes from the Acupuncture Trialists' Collaboration, which pooled the raw data of individual patients from 39 trials, 20,827 people in all. Acupuncture did better than sham acupuncture and than no acupuncture for every pain condition studied. The difference was close to half a standard deviation against no acupuncture and close to a fifth of a standard deviation against sham. About 15% of the effect had faded at one year.1
| Condition | Acupuncture vs sham | Acupuncture vs no acupuncture |
|---|---|---|
| Back and neck pain | 0.23 | 0.55 |
| Osteoarthritis | 0.16 | 0.57 |
| Chronic headache | 0.15 | 0.42 |
Differences in standard deviations from the collaboration's earlier analysis; the authors call the differences from sham "relatively modest."2
Low back pain. A 2020 Cochrane review of 33 studies (8,270 people) found acupuncture reduced pain by 9 points on a 100-point scale more than sham, which is low-certainty evidence and below the 15 points usually counted as clinically important. Against no treatment, the difference was 20 points (moderate certainty).3 A 2025 overview of Cochrane reviews concluded that acupuncture probably improves function slightly compared with sham.4
The largest single trials show the same pattern. In the German GERAC trial of 1,162 people with chronic low back pain, sham acupuncture did almost as well as the real thing, and both did better than standard care:5
A U.S. trial found the same: individualized, standardized and simulated (non-penetrating) acupuncture each improved back-related disability by about 4.5 points, against 2.1 with usual care.6
Does acupuncture work for migraines and headaches?
Moderate-quality evidence for prevention
Migraine. A 2016 Cochrane review of 22 trials (4,985 people) found that headache frequency at least halved in 41% of people having acupuncture, against 17% with no acupuncture. Against sham acupuncture, the figures were 50% and 41%. Compared with preventive drugs, 57% improved with acupuncture against 46% with the drugs at three months. The evidence was rated moderate quality.7
Tension-type headache. A 2016 Cochrane review found that headaches at least halved in 51% of people with acupuncture against 43% with sham. In the largest trial against routine care, the figures were 48% and 19%.8
Is acupuncture just a placebo?
What sham acupuncture is, and why it matters
To test acupuncture against a placebo, researchers use sham acupuncture: needles placed at the "wrong" points, inserted only shallowly, or blunt needles that touch the skin without piercing it. How you read the evidence depends on what you think sham acupuncture is.
A 2009 review of 13 trials with three groups (real acupuncture, sham acupuncture and no acupuncture; 3,025 patients) found that real acupuncture beat sham by about 4 points on a 100-point pain scale, while sham beat no acupuncture by a larger margin. The authors concluded that the specific effect "seems to lack clinical relevance and cannot be clearly distinguished from bias."9
The other reading is that some shams are not inert. In the large pooled analysis, trials whose sham used needles that pierced the skin found smaller differences, which suggests those shams have effects of their own.1 The 2009 review, however, found no link between the type of sham and the size of the effect.9 The U.S. National Center for Complementary and Integrative Health (NCCIH) puts it plainly: nonspecific effects contribute to acupuncture's benefit.10
Expectation matters too. In four trials with 864 patients with chronic pain, people who expected acupuncture to work were more likely to improve.11 A fair summary: acupuncture appears to have a small specific effect, sitting on top of a larger effect of the whole treatment experience.
What medical guidelines say about acupuncture
They don't all agree
- American College of Physicians (2017), low back pain. For chronic low back pain, clinicians should first choose non-drug treatments, including exercise, rehabilitation, acupuncture and mindfulness-based stress reduction (strong recommendation, moderate-quality evidence).12
- NICE (England), chronic primary pain. Consider a single course of acupuncture or dry needling, within a limited number of hours of clinician time.13
- NICE, headaches. For chronic tension-type headache, consider up to 10 sessions over 5 to 8 weeks. For migraine prevention, consider acupuncture only if propranolol, topiramate and amitriptyline have not worked (updated 2025).14 How many sessions trials used, and how long results lasted, is covered in How many acupuncture sessions do you need?
- NICE, low back pain and sciatica. Do not offer acupuncture.15
The American and English guidelines reach opposite conclusions on low back pain from much the same evidence: the American one weighs the benefit over usual care, the English one the small difference from sham.
Where acupuncture doesn't work, or the evidence is weak
No advantage over sham in several large trials
- IVF. In a large trial, live births were 18.3% with acupuncture and 17.8% with sham.16
- Quitting smoking. A Cochrane review found no consistent, bias-free evidence of benefit lasting six months or more.17
- Chronic knee pain in people over 50. Neither needle nor laser acupuncture did better than sham.18
- Depression. The evidence is low or very low quality.19
- Irritable bowel syndrome and hot flashes. NCCIH reports no benefit over sham.10
- Nausea from chemotherapy. Whether acupuncture adds anything to modern anti-nausea drugs is uncertain, and a 2026 umbrella review rated most of the evidence low or very low certainty.10,20
What about the WHO's list of 28 conditions?
An old report, often misquoted
A World Health Organization report from 2002 (often cited as 2003) listed 28 conditions for which acupuncture had been "proved—through controlled trials—to be an effective treatment," including depression, stroke, high blood pressure and induction of labor. The report covers trials published up to 1998, was drafted by a single author, and did not grade trial quality the way a modern systematic review does. It also says that "only national health authorities can determine" which conditions acupuncture should be recommended for.21 It is not a WHO endorsement, and several items on its list have since been contradicted by better trials.
Is acupuncture safe?
Serious harm is rare with sterile, single-use needles
In the hands of trained practitioners using sterile, single-use needles, acupuncture has a low rate of serious side effects. Most problems are minor, such as bruising or soreness at the needle site.10 For the numbers, and who should take extra care, see Is acupuncture safe?
If you are considering acupuncture, talk with your physician or a licensed acupuncturist about whether it suits your situation.
Quick answers
Does acupuncture actually work?
For chronic pain and for preventing migraine and tension-type headaches, it does better than no treatment and slightly better than sham acupuncture. For IVF, smoking and several other uses, it has not beaten sham.
Is acupuncture a placebo?
Partly. Much of the benefit seen in trials also appears with sham acupuncture, which points to expectation and the treatment context. A small additional effect appears for real acupuncture in pain trials, and researchers disagree about how much it matters.
Is acupuncture scientifically proven?
"Proven" overstates it. There is moderate-quality evidence for chronic pain and headache prevention, and little or none for many other uses.
Does acupuncture work for back pain?
It does better than no treatment, and slightly better than sham. The American College of Physicians recommends it for chronic low back pain; England's NICE advises against it.
Does acupuncture work for sciatica?
One 2024 trial of 216 people with sciatica from a herniated disc found leg pain fell more with acupuncture than with sham, and the difference lasted a year. It needs to be repeated before anyone can be confident.
Does acupuncture help with fertility and IVF?
In a large trial, live birth rates were almost the same with acupuncture and sham acupuncture.
Sources
- Vickers AJ, et al. Acupuncture for chronic pain: Update of an individual patient data meta-analysis. Journal of Pain. 2018;19(5):455–474. doi:10.1016/j.jpain.2017.11.005. PubMed
- Vickers AJ, et al. Acupuncture for chronic pain: Individual patient data meta-analysis. Archives of Internal Medicine. 2012;172(19):1444–1453. doi:10.1001/archinternmed.2012.3654. PubMed
- Mu J, et al. Acupuncture for chronic nonspecific low back pain. Cochrane Database of Systematic Reviews. 2020;(12):CD013814. doi:10.1002/14651858.CD013814. PubMed
- Rizzo RR, et al. Non-pharmacological and non-surgical treatments for low back pain in adults: An overview of Cochrane reviews. Cochrane Database of Systematic Reviews. 2025;(3):CD014691. doi:10.1002/14651858.CD014691.pub2. PubMed
- Haake M, et al. German Acupuncture Trials (GERAC) for chronic low back pain: Randomized, multicenter, blinded, parallel-group trial with 3 groups. Archives of Internal Medicine. 2007;167(17):1892–1898. doi:10.1001/archinte.167.17.1892. PubMed
- Cherkin DC, et al. A randomized trial comparing acupuncture, simulated acupuncture, and usual care for chronic low back pain. Archives of Internal Medicine. 2009;169(9):858–866. doi:10.1001/archinternmed.2009.65. PubMed
- Linde K, et al. Acupuncture for the prevention of episodic migraine. Cochrane Database of Systematic Reviews. 2016;(6):CD001218. doi:10.1002/14651858.CD001218.pub3. PubMed
- Linde K, et al. Acupuncture for the prevention of tension-type headache. Cochrane Database of Systematic Reviews. 2016;(4):CD007587. doi:10.1002/14651858.CD007587.pub2. PubMed
- Madsen MV, Gøtzsche PC, Hróbjartsson A. Acupuncture treatment for pain: Systematic review of randomised clinical trials with acupuncture, placebo acupuncture, and no acupuncture groups. BMJ. 2009;338:a3115. doi:10.1136/bmj.a3115. PubMed
- National Center for Complementary and Integrative Health. Acupuncture: Effectiveness and Safety. Updated October 2022. nccih.nih.gov
- Linde K, et al. The impact of patient expectations on outcomes in four randomized controlled trials of acupuncture in patients with chronic pain. Pain. 2007;128(3):264–271. doi:10.1016/j.pain.2006.12.006. PubMed
- Qaseem A, et al. Noninvasive treatments for acute, subacute, and chronic low back pain: A clinical practice guideline from the American College of Physicians. Annals of Internal Medicine. 2017;166(7):514–530. doi:10.7326/M16-2367. PubMed
- National Institute for Health and Care Excellence. Chronic pain (primary and secondary) in over 16s: Assessment of all chronic pain and management of chronic primary pain (NG193). April 2021. nice.org.uk
- National Institute for Health and Care Excellence. Headaches in over 12s: Diagnosis and management (CG150). 2012, updated June 2025. nice.org.uk
- National Institute for Health and Care Excellence. Low back pain and sciatica in over 16s: Assessment and management (NG59). 2016. nice.org.uk
- Smith CA, et al. Effect of acupuncture vs sham acupuncture on live births among women undergoing in vitro fertilization: A randomized clinical trial. JAMA. 2018;319(19):1990–1998. doi:10.1001/jama.2018.5336. PubMed
- White AR, et al. Acupuncture and related interventions for smoking cessation. Cochrane Database of Systematic Reviews. 2014;(1):CD000009. doi:10.1002/14651858.CD000009.pub4. PubMed
- Hinman RS, et al. Acupuncture for chronic knee pain: A randomized clinical trial. JAMA. 2014;312(13):1313–1322. doi:10.1001/jama.2014.12660. PubMed
- Smith CA, et al. Acupuncture for depression. Cochrane Database of Systematic Reviews. 2018;(3):CD004046. doi:10.1002/14651858.CD004046.pub4. PubMed
- Cao J, et al. Acupuncture for chemotherapy-induced nausea and vomiting: An umbrella review of systematic reviews. Complementary Therapies in Medicine. 2026;102:103438. doi:10.1016/j.ctim.2026.103438. PubMed
- World Health Organization. Acupuncture: Review and Analysis of Reports on Controlled Clinical Trials. Geneva: WHO, 2002. ISBN 92-4-154543-7.
This article is for education only and is not medical advice. See our Medical Disclaimer. Last reviewed October 13, 2026.
Research Worth Reading
New articles like this one, sent to you when they are published: what the evidence actually shows and what the classical texts actually say, in plain English. Join The Naturalist Archive and you will also hear first about Archive giveaways, and receive the English translation of Nam Dược Thần Hiệu free the day it is released.