Get Involved
About Us
Our Members
Trick or Treatment?: Alternative Medicine on Trial
Trick or Treatment?: Alternative Medicine on Trial by Simon Singh and Edzard Ernst. London, England: Bantam Press, 2008; Hardcover, 342 pages. ISBN: 9780593061299. $29.00.

The catchy title of this volume is evocative of the playful, puerile Halloween address. It may suggest to some that patients who submit to practitioners of “alternative” medicine may be victims of fraud—whether deliberate or simply based on ignorance. The alternative aspect of “treatment” addresses the effectiveness and associated dangers of the different modalities that fall under its rubric, “used supposedly to heal patients.”

Professor Edzard Ernst, MD, has been acknowledged as “the world’s first professor of complementary medicine” at Exeter University in the United Kingdom; Simon Singh holds a PhD in particle physics. The authors Singh and Ernst dedicated their book to His Royal Highness The Prince of Wales, and they claim that they “are confident that [their book] offers an unparalleled level of rigour, authority, and independence,” their “only get to the truth.”

Unlike the volume by Fugh-Berman, Alternative Medicine, What Works: A Comprehensive, Easy-to-Read Review of the Scientific Evidence, Pro and Con (Lippincott Williams & Wilkins, 1997), which examined more than a dozen modalities, Singh and Ernst confine their detailed analyses to Acupuncture, Chiropractic Therapy, Homeopathy, and Herbal Medicine, relegating relatively minor modalities to an Appendix.

Chiropractic Therapy was founded by D.D. Palmer of Ontario, Canada towards the end of the 19th century (although the authors note that Hippocrates, the so-called father of medicine, is claimed to be responsible for “the first documented account of manipulating the spine for therapeutic reasons” around 400 BCE). Slight misalignments of spinal vertebrae (subluxations) are claimed to interfere with the flow of so-called “innate intelligence,” a kind of “life force” or “vital energy.” These concepts are regarded by Singh and Ernst as mystical and as baffling as the concept of ch’i in Traditional Chinese Medicine (TCM).

Regarding back pain, the authors conclude that, compared with conventional treatment, “each is just about as effective (or ineffective) as the other,” cautioning that “it would be unwise to visit a chiropractor for other than a problem related to your back.” A litany of cautions and dangers is provided, along with graphic case histories, including death due to vertebral arterial damage from manipulation of the neck.

Singh and Ernst next provide a detailed historical background of acupuncture, the origins of which are “shrouded in the mists of time,” most likely in China.

Several reviews of acupuncture research have been conducted, including a review by the Cochrane Collaboration, whose analyses are now recognized as systematic reviews. The Cochrane reviews “suggest that there is no significant evidence to show that acupuncture is an effective treatment for most of the conditions to which it has been applied.” However, Singh and Ernst have been “cautiously optimistic” about a few, including treatment for bedwetting, some types of pain, and nausea—primarily back pain and idiopathic headache. However, Singh and Ernst posit that “the quality and amount of evidence are not fully convincing,” and they impute any positive influence to a very likely placebo effect.

“The power of placebo” is also marshalled to explain any possible beneficial effect of homeopathy. Homeopathy, developed by Samuel Hahnemann in the late 18th century, treats symptoms by administering minute or non-existent (molecule-less) doses of a substance, which in large amounts produces the same disease symptoms in healthy individuals. Ernst, a former practicing homeopath, eschewed the modality in 1991, following the observation of no effect in a repeat of Hahnemann’s experiment with cinchona bark (from Cinchona officinalis, Rubiaceae), the source of the antimalarial drug quinine, by the German pharmacologist Professor W.H. Hopff (an experiment that claimed to produce symptoms of malaria). Singh and Ernst cite a 1996 article in US News and World Report, underlining “the utter absurdity and profiteering that underpins the homeopathic industry”: the flu medicine Oscillococcinum® (produced by the French homeopathic giant Boiron), derived from the heart and liver of a solitary French duck, has been estimated to generate sales of more than $20 million, despite being a “self-declared 100% sugar pill!”

The fourth modality examined is herbal medicine, the subject probably of greatest interest to readers of this review. The gist of the assessment is summarized in 2 tables representing, respectively, the efficacy and risks of herbal medicine. Table 1 lists 35 prominent herbs, roughly rated as “good,” “medium,” or “poor,” based on the amount and quality of evidence supporting their particular therapeutic applications. Those rated as “poor” are recommended to be avoided, including chamomile (Matricaria recutita, Asteraceae), evening primrose (Oenothera biennis, Onagraceae), ginseng (Panax ginseng, Araliaceae), hops (Humulus lupulus, Moraceae), passionflower (Passiflora incarnata, Passifloraceae), and thyme (Thymus vulgaris, Lamiaceae), presumably as ineffective treatments for the listed complaints. Table 2 addresses 34 herbs.

St. John’s wort (Hypericum perforatum, Clusiaceae), the flagship herb representing the Singh and Ernst judgments, is rated “good” for mild to moderate (but not severe) forms of major depression, based on a 2005 evaluation by the Cochrane Collaboration of 37 clinical trials. However, addressing the question of “the key active ingredient,” these authors render a rather curious, confusing assessment. They claim that the constituents hyperforin and hypericin have been considered the key active ingredients but that they have not proven as effective as the plant itself when tested, and “it seems that the benefits of St. John’s wort are due to a combination of chemicals, each one working to enhance the effect of the others.” There is no clear evidence to support this latter hypothesis—much favored by herbalists—and it has been recognized for some time that neither hypericin nor hyperforin make an appreciable contribution to this plant’s anti-depressant effect. A preparation virtually devoid of both compounds has been found to exert significant activity, comparable to hyperforin- and hypericin-containing extracts.

Feverfew (Tanacetum parthenium, Asteraceae) is given a “medium” evidence rating for preventing migraine “because there have been mixed results from trials—mainly positive, but partly negative.” However, Singh and Ernst do not distinguish between tested preparations: all feverfew whole leaf treatments tested have produced positive results, and the only negative outcome involved a 90% aqueous ethanol extract obtained from a 21-day digestion of leaf, which almost certainly resulted in extensive degradation of the active principle(s). These authors properly point out trial deficiencies of small sample size and rather modest effect—common to many herbal studies—but in this case only relating to the initial feverfew trial with 17 subjects, also appropriately criticized on the basis of self-selection.

The following botanicals were graded as “good” by Ernst and Singh.

Curiously, ephedra, or ma huang (Ephedra sinica, Ephedraceae) is approved by the authors for weight loss. However, while the RAND Corporation expert panel of 2003 acknowledged significant short-term weight loss from ephedrine, especially when combined with caffeine, the herb has been banned as a dietary supplement by the US Food and Drug Administration due to a possible association between supplements and serious adverse events (myocardial infarction, cerebrovascular accidents, seizures, and a psychiatric case). Also, curiously, no safety concern regarding ephedra is registered in Table 2, despite widespread concern about its safety and bans against its use for both weight loss and enhancement of athletic performance.

Preparations from the 3 echinacea species commonly employed in therapy, namely Echinacea angustifolia (Asteraceae), E. pallida, and

E. purpurea, are recommended for the treatment and prevention of the common cold. While echinacea is widely claimed to be beneficial as an early treatment for symptoms of the cold, there is mixed, and to some unpersuasive, evidence for a prophylactic role.

Garlic (Allium sativum, Alliaceae) is likewise firmly recommended for hypercholesterolemia, despite conflicting and only moderately positive results in support. Devil’s claw (Oplopanax horridus, Araliaceae) for musculoskeletal pain, ginkgo (Ginkgo biloba, Ginkgoaceae) for dementia and poor circulation in the leg, hawthorn (Crataegus spp., Rosaceae) for congestive heart failure, kava (Piper methysticum, Piperaceae) for anxiety, and red clover (Trifolium pratense, Fabaceae) for menopausal symptoms round out the blue-ribbon group, despite contradictory evidence for the latter.

Table 2, listing health risks, is dominated by precautions against bleeding, mostly based on speculation arising from in vitro observations, involving almost half (16) of the 34 herbs examined. This is followed by 6 indictments of herbs with the claimed potential to interact with heart medications and associated with reports of liver damage. This latter condition notably involves black cohosh (Actaea racemosa, Ranunculaceae, syn. Cimicifuga racemosa) (about 10 cases) and kava (80 cases). The other 2 herbs are valerian (Valeriana officinalis, Valerianaceae) and willow (Salix spp., Salicaceae), linked with isolated reports. Regarding bleeding, except for garlic and ginkgo, no clinical significance has so far been attached to other herbs containing components shown to possess anti-coagulant properties in non-clinical settings. Excessive operative bleeding has been associated only with garlic and, possibly, ginkgo, and preoperative use of dietary garlic has been associated with increased surgical blood loss. Ginkgo extract has been linked to spontaneous and increased bleeding when combined with anticoagulants and might be expected to increase the risk of operative hemorrhage.

Since all other cited herbal remedies are associated with serious adverse possibilities, it is more than passing perplexing to consider the reason for including, as presenting appreciable risks, artichoke (Cynara scolymus, Asteraceae) for high cholesterol (dyspepsia), which is not known to exert adverse effects apart from flatulence, and lavender (Lavandula spp., Lamiaceae), for “in rare cases (causing) hormonal side effects such as swelling of the breast tissue”—the latter being an apparently uncritical reference to several case reports of use of “lavenderscented” soaps and shampoos by young men.

Preceding the final heading of the last chapter is a segment entitled “Lettermanesque.” The authors list “top ten culprits in the promotion of unproven and disproven medicine,” although not in order of demerit: celebrities, medical researchers, universities, alternative gurus, the media (twice), doctors, alternative medicine societies, governments and regulators, and the World Health Organization.

The book’s overall verdict is that “alternative medicine . . . seems to consist of treatments that are untested or unproven, or disproven, or unsafe, or placebo, or only marginally beneficial.” “The truth . . . is that there is no such thing as alternative science, just as there is no alternative biology, alternative anatomy, alternative testing, or alternative evidence,” and “any provably safe and effective alternative medicine . . . becomes a conventional medicine.”

—Dennis V.C. Awang, PhDMediplant ConsultingWhite Rock, BC, Canada