The US ‘Agency for Healthcare Research and Quality (AHRQ) have published a most comprehensive review update entitled ‘Noninvasive Nonpharmacological Treatment for Chronic Pain‘. It followed the AHRQ Methods Guide for Effectiveness and Comparative Effectiveness. The conditions included were:

  • Chronic low back pain
  • Chronic neck pain
  • Osteoarthritis (knee, hip, hand)
  • Fibromyalgia
  • Chronic tension headache

Here are the main findings related to acupuncture:


  • Acupuncture was associated with a small improvement in short-term function compared with sham acupuncture or usual care (4 trials); there was no difference between acupuncture and controls in intermediate-term (3 trials) or long-term (1 trial) function (: low). Acupuncture was associated with small improvements in short-term (5 trials) and long-term (1 trial) pain compared with sham acupuncture, usual care, an attention control, or a placebo intervention but there was no difference in intermediate-term pain (5 trials) (SOE: moderate for short term, low for intermediate term and long term).


  • Acupuncture was associated with small improvements in short-term (5 trials) and intermediate-term (3 trials) function versus sham acupuncture, a placebo (sham laser), or usual care; one trial reported no difference in function in the long term (: low for all time periods). For pain, there were no differences for acupuncture versus sham acupuncture or placebo interventions in the short (4 trials), intermediate (3 trials), or long (1 trial) term (SOE: low for all time periods).


  • No differences were seen between acupuncture and control interventions (sham acupuncture, waitlist, or usual care) for function in the short term (4 trials) or the intermediate term (4 trials) (: low for short term; moderate for intermediate term). Stratified analysis showed no differences between acupuncture and sham treatments (4 trials) but moderate improvement in function compared with usual care (2 trials) short term. For pain, there were no differences between acupuncture versus control interventions in the short term (6 trials) or clinically meaningful differences in the intermediate term (4 trials) (SOE: low for short term; moderate for intermediate term). Short-term differences in pain were significant for acupuncture versus usual care but not for acupuncture versus sham acupuncture.


  • Laser acupuncture was associated with small, short-term improvements in pain intensity and in the number of headache days per month versus sham in one trial (: low).


  • Acupuncture was associated with a small improvement in function compared with sham acupuncture at short-term (3 trials [1 new]) and intermediate-term (2 trials) follow-up (: moderate). There was no effect for acupuncture versus sham acupuncture on pain in the short term (4 trials [1 new]) or intermediate term (3 trials) (SOE: low) or based on pooled estimates across control conditions (sham or attention control, 5 trials [2 new]) SOE: low).


Treatment-related SAEs were rare (across 5 , 5 neck pain, 4 , 1 knee , and 1  trial); only one event (needle insertion site pain lasting1 month) in a LBP patient (<1%) in one trial was considered related to treatment,

SAEs not considered to be related to acupuncture or the study conditions (range 0% to 9% across 5 , 5 neck pain, 4 , 1 knee , and 1  trial). These included hospitalization (primarily) or outpatient treatment; reasons were not specified.

The most commonly reported non-serious AEs: swelling, bruising, bleeding or pain at the acupuncture site (1% to 61%, 12 RCTs; or 1% to 18% excluding an outlier trial)); numbness, discomfort, pain or increase in symptoms (1% to 14%; 11 RCTs), dizziness, nausea, fainting (1% to 7%, 7 RCTs), headache (1% to 2%; 4 RCTs), vasovagal symptoms (1% to 4%; 2 RCTs), respiratory problems, chest discomfort (1%; 2 neck pain RCTs), and infection at needle insertion site [1%; 1  (knee )]


I find this interesting, especially if we consider that chronic pain is THE domain for acupuncture (as practised in the West). It shows that, contrary to what so many enthusiasts try to tell us, the evidence for acupuncture is very weak. It also demonstrates that, contrary to what some sceptics assume, the evidence is not totally negative.

As far as harms are concerned, we need to be aware of the fact that the above conclusions are based on clinical trials. We and others have repeatedly shown that in the real of SCAM many, if not most clinical studies fail to mention adverse effects. This means two things: firstly, the trialists violate research ethics; secondly, the above information is woefully incomplete.

5 Responses to Acupuncture for chronic pain: an important up-date

  • This is an interesting review and a useful addition to the literature. However, one of the issues with their assessment approach is that they include comparisons with usual care rather than just against a recognised placebo / comparator in the evidence assessment. In fact they say in the discussion “The majority of trials compared interventions with usual care and very few trials employed pharmacological treatments or exercise as comparators”. This means that non-specific treatment effects have not been controlled for. This is an issue across all of the therapies assessed, not just acupuncture, and makes it difficult to really know whether these treatments are actually effective.

  • In Brazil, the Federal Council of Medicine (CFM), decided to elevate Acupuncture to the condition of MEDICAL SPECIALTY, similar to cardiology, rheumatology, endocrinology etc., and its directors, over the years, kept the decision.They did the same with Homeopathy!More recently, they corroborated the decision to use hydroxychloroquine in the treatment of COVID-19 in the early stages!

    • More recently, they corroborated the decision to use hydroxychloroquine in the treatment of COVID-19 in the early stages!

      I was interested to read yesterday that the WHO has just suspended its trial of hydroxychloroquine in Covid-19 after the Lancet published an analysis of over 96,000 patients from 671 hospitals worldwide, finding that the 14,888 individuals receiving chloroquine or hydroxychloroquine, alone or in combination with macrolide antibiotics, showed double the mortality compared to a matched control group who received neither.

      But maybe this isn’t surprising. Chloroquine and hydroxychloroquine well-known to be cardiotoxic, and one of their effects is prolongation of the Q-T interval on the ECG, leading to an increased risk of ventricular arrhythmias, including ventricular fibrillation (a form of cardiac arrest). Coronavirus also prolongs the Q-T interval, and is also known to cause ventricular arrhythmias and cardiac arrest (a cardiologist told me recently that sudden cardiac death in the community has increased by a factor of four in New York since the start of the pandemic; I’m afraid I don’t have the reference).

      This highlights the dangers of adopting a treatment without data to support it.

      • Thanks for the timely observation. It was for these and other reasons, perfectly known and considered, that we present these disappointing facts in this prestigious site of Prof. Edzard Ernst.

  • I also suspect non-specific treatment effects were not controlled for. A critically-important one is quality of blinding at the various links in delivering and assessing both sham and “real” treatments. So I have to wonder did these trials actually survey participants, therapists, assessors and other personnel involved in the trial to determine their beliefs around whether or not it was the “real” treatment.

    We also know from previous studies that factors such as the apparel worn by the therapist, the setting for the procedure, and the length of the procedure, as well as the entire experience around the procedure, are more closely associated with outcomes than is the quack treatment.

    When outcomes are so minor as these studies suggest, one has to immediately be suspicious that these non-specific effects were the cause.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Subscribe via email

Enter your email address to receive notifications of new blog posts by email.

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.