The state of the evidence on acupuncture

Acupuncture has strong evidence for some conditions, moderate for others, and weak or no evidence for many. Here's an honest condition-by-condition summary, drawn from Cochrane reviews, NIH NCCIH guidance, and major clinical society guidelines.

A lot of acupuncture content online overstates the evidence. A lot of skeptical content understates it. Here's a clear-eyed summary of where the research actually stands as of 2026, drawing on Cochrane systematic reviews, the U.S. National Center for Complementary and Integrative Health (NCCIH), and major clinical society guidelines.

The general framing: acupuncture has been studied extensively over the past three decades, and the evidence is condition-specific. Some conditions have a real and consistent body of supporting research. Others have promising early evidence. Others have been studied and shown little benefit. Treating the whole field as either "proven" or "unproven" is a category error.

Strong evidence

These conditions have multiple high-quality systematic reviews showing meaningful, consistent benefit. Acupuncture is recommended in major clinical guidelines for these conditions in several countries.

• Chronic low back pain. Cochrane reviews and major guidelines (American College of Physicians, NICE in the UK) include acupuncture as a recommended non-pharmacologic option. Medicare covers it for this indication.
• Migraine prevention. Cochrane reviews find acupuncture appears at least as effective as standard preventive medications for reducing migraine frequency, with fewer side effects. Several national headache society guidelines include it as a recommended preventive option.
• Tension-type headache. Similar to migraines — reduces frequency and severity of attacks.
• Postoperative and chemotherapy-induced nausea and vomiting. Particularly using the P6 (Neiguan) point. Strong, consistent evidence; included in some oncology supportive-care guidelines.
• Chronic neck pain. Moderate-to-strong evidence, often paired with electroacupuncture for musculoskeletal conditions.

Moderate evidence

Promising research, often positive across multiple trials, but with smaller effect sizes, mixed results, or methodological limitations that prevent definitive conclusions. Reasonable to try if you're curious; reasonable to be uncertain about the magnitude of benefit.

• Knee osteoarthritis. Multiple positive trials; some major guidelines mention it as a possible adjunct.
• Fibromyalgia. Modest improvements in pain and function in several trials; evidence is suggestive but not conclusive.
• Generalized anxiety. Modest reductions in anxiety symptoms; evidence is moderate, growing, but not yet definitive.
• Insomnia. Modest improvements in sleep quality and total sleep time; CBT-I remains the strongest non-pharmacologic option.
• Carpal tunnel syndrome. Some positive trials, particularly for mild-to-moderate cases.
• Plantar fasciitis. Moderate evidence for short-to-medium-term benefit.
• Menopausal hot flashes. Some positive trials; effect appears smaller than hormone therapy but with fewer side effects.
• Stroke recovery (motor function). Multiple positive trials; included in some rehabilitation protocols.
• Allergic rhinitis (seasonal allergies). Modest improvements in some trials.

Mixed or limited evidence

Research has been done; results are inconsistent. Some patients clearly benefit; in larger pooled analyses the picture is unclear. May be worth a trial alongside conventional care; not the first thing to lean on.

• IVF support / fertility. Some meta-analyses show small positive effects on pregnancy rates, particularly around embryo transfer; others find no significant effect. Widely used clinically; evidence remains genuinely contested.
• Sciatica. Limited but mostly positive trials.
• Depression. Some positive trials; effect appears smaller than for anxiety.
• PTSD. Limited trials with positive trends; protocols like NADA are widely used in trauma settings.
• TMJ disorders. Limited but positive evidence.
• IBS. Mixed evidence; some positive trials.
• Smoking cessation and substance use disorders. Mixed trials. NADA protocol widely used as part of recovery programs but not as a standalone intervention.
• Hypertension. Limited positive evidence; not a substitute for blood pressure medication.

Weak or insufficient evidence

Either not enough research has been done, or what's been done shows no significant benefit beyond placebo. We don't recommend choosing acupuncture primarily for these — but acupuncturists may treat them, and individual patients may benefit.

• Tinnitus. Limited evidence.
• Acute stroke (not recovery). Inconsistent results.
• Asthma. Limited and inconsistent evidence.
• Cognitive function in dementia. Insufficient evidence.
• Most autoimmune conditions (rheumatoid arthritis, MS, lupus). Acupuncture may help with associated symptoms (pain, fatigue, sleep) but does not modify the underlying disease.

What about... (the careful warnings)

Several conditions deserve specific cautions, regardless of what individual practitioners may claim:

• Cancer. Acupuncture has good evidence as supportive care — for chemotherapy-induced nausea, fatigue, pain, hot flashes, and quality of life. It is not a treatment for cancer itself. Practitioners or clinics suggesting otherwise are dangerous.
• Infectious disease. Acupuncture does not treat infections. See a doctor.
• Acute medical emergencies. Stroke symptoms, chest pain, severe injury — call 911. Acupuncture has no role here.
• Mental health crises. If you're in crisis, call 988 (US). Acupuncture is not a substitute for professional mental health care.

How to read claims you encounter

When you see acupuncture claims online, useful filters:

1. Is the source a research review or a personal practitioner site? Reviews are more reliable.
2. Is the claim specific (a particular condition, a specific effect) or sweeping ("balances your energy")? Specific is better.
3. Does it acknowledge uncertainty? Honest sources say "the evidence suggests" or "studies show modest improvement," not "guaranteed results."
4. Is the source selling something? Practitioners selling herbs, supplements, or expensive packages have a financial interest in optimistic claims.

The framing on this site is intentional: we'd rather you have realistic expectations and decide acupuncture is right for you than have inflated expectations and feel let down. Acupuncture is a real intervention with real but modest effects for many conditions. It's a useful tool. It's not magic.

Sources we trust

If you want to dig deeper:

• Cochrane Library (cochrane.org) — gold-standard systematic reviews
• NIH National Center for Complementary and Integrative Health (nccih.nih.gov)
• PubMed (pubmed.ncbi.nlm.nih.gov) — primary research literature
• National Certification Commission for Acupuncture and Oriental Medicine (nccaom.org)

Continue reading

• How to choose a practitioner
• Red flags to watch out for
• How acupuncture works