The Evidence

Does acupuncture work for chronic back pain?

Want just the essentials? Quick Answers — the four most common questions, directly answered →

TL;DR — What the Evidence Shows

Research suggests acupuncture may reduce chronic low back pain more than no treatment. Several large reviews (studies that combine results from many studies) support this.

When compared to sham acupuncture (a fake version that mimics needling), the difference is smaller and less clear. Scientists still debate what this means. In December 2023, the WHO recommended acupuncture for chronic low back pain for the first time.

How certain is the evidence? Moderate that it helps more than no treatment. Low that it helps more than sham.

How certain is the evidence?

Comparison	Certainty	What the evidence says
Vs. no treatment / usual care	Moderate Certainty	Acupuncture likely results in meaningful pain reduction.
Vs. sham acupuncture	Low Certainty	Acupuncture may result in small additional benefit beyond sham.
Vs. PT, massage, chiropractic	Very Low Certainty	The evidence is very uncertain about meaningful differences between active treatments.

Ratings follow GRADE methodology. How we rate evidence.

Fact Box

Chronic low back pain: acupuncture vs. no treatment

	Without acupuncture	With acupuncture
Experience meaningful pain reduction (≥50% from baseline)	~30 out of 100	~50 out of 100
Experience meaningful disability improvement	~35 out of 100	~50 out of 100

Certainty: MODERATE (GRADE)

Source: Vickers et al. 2018 (n=20,827 across 39 trials); Cochrane 2020

'Meaningful' = ≥50% pain reduction from baseline. Definitions vary across trials. Figures are approximate, derived from pooled effect sizes.

What do the largest reviews find?

Vickers et al. 2018 — The largest pooled analysis

The Acupuncture Trialists' Collaboration pooled individual patient data from 39 trials involving 20,827 people with chronic pain, including back and neck pain. It is the largest and most rigorous pooled analysis of acupuncture trials conducted to date.

Acupuncture vs. no treatment: approximately 0.5 standard deviation improvement in pain — a clinically significant effect.
Acupuncture vs. sham acupuncture: approximately 0.2 SD improvement — small but statistically significant.
The effect vs. sham was retained at 1-year follow-up in 85% of trials, suggesting durability.

Vickers AJ et al. "Acupuncture for Chronic Pain: Update of an Individual Patient Data Meta-Analysis." J Pain. 2018;19(5):455–474. PMID 29198932

Cochrane Review 2020

The Cochrane review found two distinct findings that must be presented together:

Vs. sham: mean difference of −9.22 points on a 0–100 pain scale. The threshold for clinical importance is typically 15 points. This difference does not meet that threshold. GRADE certainty: low.
Vs. no treatment: mean difference of −20.32 points — this crosses the 15-point clinical importance threshold. GRADE certainty: moderate.

Mu J et al. "Acupuncture for chronic low back pain." Cochrane Database Syst Rev. 2020;CD013814.

ACP Clinical Practice Guideline 2017

The American College of Physicians issued a strong recommendation for acupuncture as a first-line nonpharmacologic option for chronic low back pain, based on moderate-quality evidence. No single modality was identified as clearly superior to others.

Status note: This guideline was published in 2017 and is past its typical 5-year review window. It had not been updated as of March 2026. It is directionally consistent with subsequent evidence but should be considered potentially outdated.

Qaseem A et al. "Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain." Ann Intern Med. 2017;166(7):514–530. PMID 28192789.

WHO Guideline December 2023

The World Health Organization issued its first-ever guideline for non-specific low back pain in December 2023, with a conditional recommendation for acupuncture.

A conditional recommendation means the evidence supports use in appropriate patients, but remaining uncertainty means not every patient will benefit equally. This was the first time the WHO included acupuncture in a low back pain guideline.

WHO Guidelines on Low Back Pain. Geneva: World Health Organization; December 2023. ISBN 9789240081789.

What is the sham acupuncture problem?

The most important methodological debate in acupuncture research is whether sham acupuncture — the control condition — is a valid placebo. This affects how to interpret the smaller real-vs-sham difference. Three distinct interpretations exist in the scientific literature.

Frame 1: A specific effect exists

Real acupuncture consistently beats sham by ~0.2 SD. Biomarker research suggests real and sham produce different physiological responses. Brain imaging studies find the two are mechanistically distinguishable (Napadow et al., Hum Brain Mapp. 2013;34(10):2592–2606). The persistent real-vs-sham difference reflects genuine specificity of needle placement.

Frame 2: Non-specific effects dominate

The large effect vs. no treatment (~0.5 SD) and small effect vs. sham (~0.2 SD) suggests most benefit is non-specific — driven by context, expectation, therapeutic relationship, and attention. The sham needle also penetrates or indents skin and triggers neural responses, making it an imperfect placebo. The benefit may be real but the mechanism largely non-specific.

Frame 3: Sham is not a valid control

Both real and sham acupuncture activate sensory and endogenous opioid pathways. Comparing them may ask the wrong question — both may be active treatments with similar mechanisms. The clinically relevant question may be "does any needling help vs. no intervention?" rather than "does needle placement matter?"

This is an active area of scientific disagreement. The evidence does not resolve which frame is correct.

What do we not know?

Whether needle placement location (meridians, depth) matters for outcome. Trials show inconsistent location effects.
Long-term durability beyond 12 months. Few trials follow patients past 1 year.
Which patients are most likely to respond. No validated predictors of treatment response exist.
Whether acupuncture is superior, equivalent, or inferior to other active treatments. No adequate head-to-head RCTs exist for acupuncture vs. PT, massage, or chiropractic.
Whether the number of sessions and treatment frequency affect outcomes. Dose-response data is limited.

Discuss with your doctor before starting if you:

Take blood thinners (warfarin, heparin, direct anticoagulants)
Have a pacemaker or implantable cardiac device
Are pregnant — specific acupuncture points are contraindicated
Have a bleeding disorder or compromised immune system

Based on cited sources. This is not personalized medical advice — discuss with your healthcare provider.

Full safety profile, adverse event data, and absolute contraindications.

Key sources

Page last reviewed: March 7, 2026 · Authored by Claude (Anthropic AI) · Research methodology