WHO not to be trusted with review of CAM

In Pricks I posted a list of diseases that according to Chinese herbalist and acupuncturist Shenrong Liu are some of the conditions which the World Health Organization (WHO) says are treatable by acupuncture. Here's the complete list from the WHO website. WHO lists the diseases/symptoms/conditions under four categories: 1.those for which acupuncture has been proved to be effective. 2. those for which the therapeutic effect of acupuncture has been shown but for which further proof is needed. 3. those for which there are only individual controlled trials reporting some therapeutic effects. 4. those for which acupuncture may be tried provided the practitioner has special modern medical knowledge and adequate monitoring equipment. 

1. Diseases, symptoms or conditions for which acupuncture has been proved-through controlled trials-to be an effective treatment:

    Adverse reactions to radiotherapy and/or chemotherapy
    Allergic rhinitis (including hay fever)
    Biliary colic
    Depression (including depressive neurosis and depression following stroke)
    Dysentery, acute bacillary
    Dysmenorrhoea, primary
    Epigastralgia, acute (in peptic ulcer, acute and chronic gastritis, and gastrospasm)
    Facial pain (including craniomandibular disorders)
    Headache
    Hypertension, essential
    Hypotension, primary
    Induction of labour
    Knee pain
    Leukopenia
    Low back pain
    Malposition of fetus, correction of
    Morning sickness
    Nausea and vomiting
    Neck pain
    Pain in dentistry (including dental pain and temporomandibular dysfunction)
    Periarthritis of shoulder
    Postoperative pain
    Renal colic
    Rheumatoid arthritis
    Sciatica
    Sprain
    Stroke
    Tennis elbow


2. Diseases, symptoms or conditions for which the therapeutic effect of acupuncture has been shown but for which further proof is needed:

    Abdominal pain (in acute gastroenteritis or due to gastrointestinal spasm)
    Acne vulgaris
    Alcohol dependence and detoxification
    Bell’s palsy
    Bronchial asthma
    Cancer pain
    Cardiac neurosis
    Cholecystitis, chronic, with acute exacerbation
    Cholelithiasis
    Competition stress syndrome
    Craniocerebral injury, closed
    Diabetes mellitus, non-insulin-dependent
    Earache
    Epidemic haemorrhagic fever
    Epistaxis, simple (without generalized or local disease)
    Eye pain due to subconjunctival injection
    Female infertility
    Facial spasm
    Female urethral syndrome
    Fibromyalgia and fasciitis
    Gastrokinetic disturbance
    Gouty arthritis
    Hepatitis B virus carrier status
    Herpes zoster (human (alpha) herpesvirus 3)
    Hyperlipaemia
    Hypo-ovarianism
    Insomnia
    Labour pain
    Lactation, deficiency
    Male sexual dysfunction, non-organic
    Ménière disease
    Neuralgia, post-herpetic
    Neurodermatitis
    Obesity
    Opium, cocaine and heroin dependence
    Osteoarthritis
    Pain due to endoscopic examination
    Pain in thromboangiitis obliterans
    Polycystic ovary syndrome (Stein-Leventhal syndrome)
    Postextubation in children
    Postoperative convalescence
    Premenstrual syndrome
    Prostatitis, chronic
    Pruritus
    Radicular and pseudoradicular pain syndrome
    Raynaud syndrome, primary
    Recurrent lower urinary-tract infection
    Reflex sympathetic dystrophy
    Retention of urine, traumatic
    Schizophrenia
    Sialism, drug-induced
    Sjögren syndrome
    Sore throat (including tonsillitis)
    Spine pain, acute
    Stiff neck
    Temporomandibular joint dysfunction
    Tietze syndrome
    Tobacco dependence
    Tourette syndrome
    Ulcerative colitis, chronic
    Urolithiasis
    Vascular dementia
    Whooping cough (pertussis)


3. Diseases, symptoms or conditions for which there are only individual controlled trials reporting some therapeutic effects, but for which acupuncture is worth trying because treatment by conventional and other therapies is difficult:

    Chloasma
    Choroidopathy, central serous
    Colour blindness
    Deafness
    Hypophrenia
    Irritable colon syndrome
    Neuropathic bladder in spinal cord injury
    Pulmonary heart disease, chronic
    Small airway obstruction


4. Diseases, symptoms or conditions for which acupuncture may be tried provided the practitioner has special modern medical knowledge and adequate monitoring equipment:

    Breathlessness in chronic obstructive pulmonary disease
    Coma
    Convulsions in infants
    Coronary heart disease (angina pectoris)
    Diarrhoea in infants and young children
    Encephalitis, viral, in children, late stage
    Paralysis, progressive bulbar and pseudobulbar

Just counting those in categories 1 & 2 there are a total of 91 conditions that WHO claims to be treatable by acupuncture. That by merely sticking needles in various points on the skin can treat such a wide range of diseases and conditions is already suspect. So we shouldn't be surprised that the WHO review is rather flawed. According to Singh & Ernst the WHO committed two major errors in their review. First, WHO set the bar too low for clinical trials it admitted as evidence. "[T]he WHO had taken into consideration almost every trial ever conducted, because it had set a relatively low quality threshold." And so low quality trials (which are more prone to producing positive findings) skewed the review. Secondly, WHO included a large number of studies from China. This wouldn't be a problem if not for the fact that "careful statistical analyses of all the Chinese results ... demonstrate beyond all reasonable doubt that Chinese researchers are guilty of so-called publication bias." We saw the evidence for this bias in our survey of the studies by Vickers et al., Pan et al., and Tang et al. Given the low admission threshhold and the fact that published Chinese studies are more biased against negative results, the WHO review cannot but come to a very positive conclusion of acupuncture. Garbage in, garbage out. As if that's not enough to weaken the review, the report was "drafted and revised by Dr Zhu-Fan Xie, who was Honorary Director of the Institute of Integrated Medicines in Beijing, which fully endorses the use of acupuncture for a range of disorders." This conflict of interest is more reason to be suspicious of the WHO's glowing findings. [Singh & Ernst p.71-73]

Let's now compare the WHO report with the studies by the Cochrane Collaboration. Why Cochrane? Well, because it's committed to "adhering to the principles of evidence-based medicine" [Singh & Ernst p.74].

The Cochrane Collaboration is practically obsessed with producing high quality, unbiased reviews, given that quality and the avoidance of bias are as important in conducting reviews of trials as they are in conducting the trials themselves. In fact, the organization will not add a review to its database that does not comply with its voluminous set of regulations and safeguards, the most important of which from our perspective is that the quality of the trials must be factored into a succinct, bottom-line conclusion. [W]hile organized medicine might be susceptible to accusations of bias toward CAM, the Cochrane Collaboration is not. The entire reason for the organization's existence is to provide the public and the medical profession with unbiased assessments of what works and what does not work, and it takes this mission very seriously. [Bausell p.202]

The comparative analysis is pretty straightforward. I first searched for all Cochrane reviews with "acupuncture" in the title. I then culled those reviews which had matching diseases/conditions in the WHO report. The results are summarized in the table below. Note that I did not include any findings for electroacupuncture or other forms other than manual needling. And I was tendential toward results for real versus sham acupuncture rather than other types of controls. To my mind the only way to know whether acupuncture is in fact not merely a placebo is to compare it with the results of blinded sham needling.

Disease/Condition	WHO category	Cochrane Collaboration review
Depression	1	"We found insufficient evidence to recommend the use of acupuncture for people with depression. The results are limited by the high risk of bias in the majority of trials meeting inclusion criteria." [source]
Headache	1	"[T]he available evidence suggests that acupuncture could be a valuable option for patients suffering from frequent tension-type headache." [source]  Note: this review is for tension-type headache.
Induction of labor	1	"There is insufficient evidence describing the efficacy of acupuncture to induce labour." [source]
Low back pain	1	"The data do not allow firm conclusions about the effectiveness of acupuncture for acute low-back pain. For chronic low-back pain, acupuncture is more effective for pain relief and functional improvement than no treatment or sham treatment immediately after treatment and in the short-term only." [source]
Nausea and vomiting	1	Acupuncture did not reduce the proportion of acute vomiting from chemotherapy. [source]  Note: this review is for chemotherapy-induced nausea or vomiting "Compared with sham treatment P6 acupoint stimulation significantly reduced: nausea ...; vomiting ..., and the need for rescue antiemetics." [source]  Note: this review is for post-operative nausea and vomiting.
Neck pain	1	"There is moderate evidence that acupuncture relieves pain better than some sham treatments, measured at the end of the treatment ."[source]
Rheumatoid arthritis	1	"From the little evidence that there is, acupuncture does not appear to improve the symptoms of rheumatoid arthritis." [source]
Stroke	1	"There is no clear evidence of benefit from acupuncture in acute stroke." [source] "There is no clear evidence of the effects of acupuncture on stroke rehabilitation." [source] "There is not enough evidence to make any conclusion about the therapeutic effect of acupuncture for dysphagia after acute stroke." [source]
Tennis elbow (lateral elbow pain)	1	"There is insufficient evidence to either support or refute the use of acupuncture (either needle or laser) in the treatment of lateral elbow pain." [source]
Bell's palsy	2	"The quality of the included trials was inadequate to allow any conclusion about the efficacy of acupuncture [source]
Bronchial asthma	2	"There is not enough evidence to make recommendations about the value of acupuncture in asthma treatment." [source] Note: this review is for chronic asthma.
Cocaine dependence	2	"There is currently no evidence that auricular acupuncture is effective for the treatment of cocaine dependence." [source]
Female infertility	2	"The data from this meta-analysis suggests that acupuncture does increase the live birth rate with in vitro fertilisation (IVF) treatment when performed around the time of embryo transfer. However, this could be attributed to placebo effect and the small number of trials included in the review." [source]
Insomnia	2	"The small number of randomised controlled trials, together with the poor methodological quality and significant clinical heterogeneity, means that the current evidence is not sufficiently extensive or rigorous to support the use of any form of acupuncture for the treatment of insomnia." [source]
Osteoarthritis	2	"Sham-controlled trials show statistically significant benefits; however, these benefits are small, do not meet our pre-defined thresholds for clinical relevance, and are probably due at least partially to placebo effects from incomplete blinding." [source]
Schizophrenia	2	"We found insufficient evidence to recommend the use of acupuncture for people with schizophrenia." [source]
Tobacco dependence	2	"Acupuncture and related therapies do not appear to help smokers who are trying to quit." [source]
Vascular dementia	2	"There is no evidence from randomized controlled trials to determine whether acupuncture provides any effect when treating people with vascular dementia." [source]
Irritable bowel/colon syndrome	3	"There is no evidence to support the use of acupuncture for the treatment of irritable bowel syndrome." [source]

In the above table the only conditions for which Cochrane reviews found evidence of efficacy are tension-type headache, neck pain, short term relief of chronic low back pain, and nausea and vomiting (but not those that are chemotherapy-induced). Of the nine conditions in the table above which according to the WHO are most certainly treatable by acupuncture (category 1), only four are supported by Cochrane reviews and even then the evidence is far from overwhelming.

While the Cochrane review for acupuncture for IVF does say there is evidence in its favor, the authors are cautious given the small number of trials. More importantly that 2007 review has to be tempered by the large RCT by So et al. (2009) employing 370 participants, which found that "placebo acupuncture was associated with a significantly higher overall pregnancy rate when compared with real acupuncture." In other words, real acupuncture is no better than sham. Moreover, a meta-analysis of 14 clinical trials having a total of 2670 subjects published in 2010 by Cheong et al. found "no evidence of benefit in the use of acupuncture during assisted conception." Thus, none of the category 2 conditions listed in the table have any good evidence in their favor.

Although only a minority of the diseases/conditions in the WHO report have corresponding Cochrane reviews the trend is apparent--Cochrane reviews paint a much sober picture of acupuncture's effectiveness. Except for some four conditions, there is currently a dearth of good evidence for the efficacy of acupuncture. This is further drummed in when we consider the fact that the findings of other Cochrane reviews of acupuncture which have no matching conditions in the WHO report are all negative:

Uterine fibroids: "There is no reliable proof of effectiveness of acupuncture for uterine fibroids due to lack of randomized controlled trials up to now."
Migraine prophylaxis: "There is no evidence for an effect of 'true' acupuncture over sham interventions, though this is difficult to interpret, as exact point location could be of limited importance."
Epilepsy: "The current evidence does not support acupuncture as a treatment for epilepsy."
Glaucoma: "At this time, it is impossible to draw reliable conclusions from the available data to support the use of acupuncture for the treatment of glaucoma."
Shoulder pain: "There is not enough evidence to say whether acupuncture works to treat shoulder pain or whether it is harmful."
Restless legs syndrome: "There is insufficient evidence to determine whether acupuncture is an efficacious and safe treatment for RLS."


One would think that this blunder by the WHO is an isolated case. Unfortunately, it seems that CAM has a special place in its heart. In 2005 the British medical journal Lancet reported that

a draft report on homoeopathy by the World Health Organization says the majority of peer-reviewed scientific papers published over the past 40 years have demonstrated that homeopathy is superior to placebo in placebo-controlled trials. Furthermore, it says that homoeopathy is equivalent to conventional medicines in the treatment of illnesses, both in humans and animals.

Looks as if the WHO yet again made no effort to separate chaff from grain and instead admitted any Tom, Dick and Harry study into their review of homeopathy.

We're talking about the World Health Organization here. This is suppose to be an organization which has the health of the human population in mind. Needless to say, it ought to promote and suggest treatments that work, treatments that are backed up by good science and good evidence. Patently, there is no good scientific rationale undergirding acupuncture and homeopathy. And the best evidence at hand refutes them.

There may be hope, however. In 2009 as a response to an open letter by the Voice of Young Science network WHO officials came out stating that WHO does not recommend homeopathy for the treatment of AIDS, tuberculosis, malaria, and diarrhea in infants. While laudable, I'd rather hear them announcing that the WHO does not recommend homeopathy for any and all conditions.


An interesting aside:

In the above Cochrane review of acupuncture for migraine prophylaxis, Linde et al. posit a strange hypothesis. Given that there was no statistically significant difference between real and sham acupuncture the authors muse that the "exact point location could be of limited importance," that "correct placement of needles seems to be less relevant than is usually thought by acupuncturists." What they seem to be saying is that sham acupuncture is as therapeutic and efficacious as real acupuncture. In the above study by So et al. the authors also make a similar a comment after finding that sham performed significantly better than real needling: "Placebo acupuncture may not be inert." Dr. Steven Novella has this to say:

It is just a form of special pleading to argue after a negative trial that placebos work also. In this study the authors are left trying to justify the conclusion that placing the needles in the “correct” locations had a disadvantage over placing them in random locations, which would also mean that expertise in acupuncture is a disadvantage.

One wonders how Linde, So, and their team members would react if after a new drug fails to perform any better than the placebo control the authors of the trial hypothesize that the placebo they employed may not be inert.


----

References:

R. Barker Bausell. Snake Oil Science: The Truth About Complementary and Alternative Medicine. Oxford Press, 2007.
Simon Singh & Edzard Ernst. Trick or Treatment: Alternative Medicine on Trial. Bantam, 2008.