Skip navigation

Acupuncture for neck and back pain

Clinical bottom line: There is no convincing evidence demonstrating that acupuncture is more effective than placebo for the relief of back or neck pain. The higher quality (double blind) studies were all negative. The trials with positive outcomes were all of low quality (not blind) with serious methodological flaws.

Systematic review:

Smith LA, Oldman AD, McQuay HJ, Moore RA. Teasing apart quality and validity in systematic reviews: an example from acupuncture trials in chronic neck and back pain. 2000. Pain; 86:119-132

Date review completed: August 1998

Number of trials included: 13 randomised controlled trials

Number of patients: 251 in active groups and 241 in sham control groups

Control groups: Inactive controls including: sham acupuncture, sham TENS, no treatment or waiting list

Main outcomes: Pain intensity, pain relief of global measures of treatment efficacy.

Inclusion criteria were randomised controlled trials comparing acupuncture, with or without electrical stimulation, or laser acupuncture with an inactive control group; acute or chronic back or neck pain; group size of at least 10 patients and pain outcomes.

Reviewers conducted a thorough search of the literature including the main databases and reference lists of retrieved reports. The Oxford Pain Validity Scale (OPVS) was used to assess the trial reports. The OPVS assesses five aspects of trial methodology known to be sources of bias and error in the field of pain: blinding, group size, outcome measures, baseline pain and internal sensitivity and data analysis. This enabled more weight to be placed on trial findings from studies of higher quality and validity than those of the lower quality studies and enabled the reviewers to make their own conclusions about trial results. A qualitative assessment of the included studies was made, as data pooling was not possible due to the disparate nature of the studies.


Trials were generally of low quality with a number of methodological flaws including lack of blinding, small group sizes, poorly defined or inappropriate outcome measures, lack of internal sensitivity and poor data analysis. The duration of acupuncture treatments varied from single to multiple sessions, and type of acupuncture included traditional Chinese acupuncture and manual or electrical stimulation of trigger or non-traditional points.

Based on a vote counting exercise, five trials concluded that acupuncture was effective and eight did not. A sensitivity analysis between the trial result (positive or negative) and OPVS score showed a significant relationship between validity of the trial and the reviewer's conclusions (p = 0.023). Trials with lower validity scores were more likely to show a benefit of acupuncture whereas trials with higher scores were more likely to show no benefit of acupuncture over placebo (Figure).

Adverse effects

Three of the 13 trials reported on adverse effects. In one trial two patients has pain at the acupuncture site and one acupuncture patient had a fever. In one trial one patient withdrew who received TENS and the other trial report stated that none of the patients experienced any adverse effects.



This review contains important methodological thoughts, especially about validity criteria in pain trials. It also demonstrates that a number of original papers come to the opposite conclusion from their data. On top of that it shows again that trials of greater validity are more likely to have negative outcomes.

Further reading

Ernst E, White AR. Acupuncture for back pain: A meta-analysis of randomised controlled trials. 1998 Arch Intern Med; 158: 2235-2241 White AR, Ernst E. A systematic review of randomised controlled trials of acupuncture for neck pain. 1999 Rheumatology; 38: 143-147

Related topics

Acupuncture and back pain