More Options

Medicines Derived from Herbs


Edzard Ernst

Skeptical Inquirer Volume 36.1, January/February 2012

medicine from herbs

Herbal medicines are currently quite popular; consumers are spending billions on them each year. Enthusiasts praise them as natural and safe, while skeptics often see them as little more than glorified placebos. The general public is frequently confused by such controversies, by a plethora of misinformation, and by the bewildering categories of medicines derived from herbs (U.S. Government Accountability Office 2010). Here I will try to clear up some of this confusion by explaining what the different categories are.

Herbal Medicines

Herbal medicines are preparations made from whole plants or whole parts of plants. They are also called botanical medicines, remedies, or supplements. Invariably they contain a mixture of ingredients, some of which may be pharmacologically active. Frequently they are marketed as dietary supplements, which are not required to have proven efficacy, safety, or quality in the United States and most other countries (Marcus and Grollman 2002, 347). Thus the spectrum is wide with both high- and low-quality products often placed side by side. Calls for tighter regulation are made regularly (e.g., U.S. Government Accountability Office 2009) but are routinely frustrated.

Herbal medicines are mostly used by consumers for self-treatment of minor symptoms. Doctors rarely employ them (except in some countries, such as Germany) and, crucially, traditional herbalists use an entirely different approach with each treatment.

The majority of herbal medicines have not been scientifically tested. But some have been adequately analyzed, standardized, and submitted to clinical trials (Ernst et al. 2006). St. John's Wort (Hypericum perforatum) is perhaps the best-investigated example. We know that this herbal antidepressant has several pharmacologically active ingredients that have been standardized in high-quality products and tested for efficacy and safety in approximately fifty clinical trials and many post-marketing surveillance studies. The results leave little doubt that St. John's Wort is efficacious for mild to moderate depression. It is also relatively safe as long as it is not combined with other drugs (Ernst et al. 2006).

When taken together with other medications, St. John's Wort can powerfully interact such that it lowers the plasma level of many drugs (Izzo and Ernst 2001, 15) which, of course, can have serious consequences. Thus the example of St. John's Wort goes some way toward demonstrating that herbal medicines can do both good and harm to patients. In other words, some herbal medicines are complicated pharmacological treatments and are biologically plausible (Schulz and Hänsel 2003).

Many other herbal medicines are not well-researched; therefore we cannot be certain about their risk-benefit profile (Ernst et al. 2006). Even the well-researched examples like St. John's Wort should be approached with healthy skepticism: the few high-quality products available are outnumbered by supplements of low quality and dubious content. Thus the market of herbal medicines is littered with products that contain little or no herbal ingredients (Sievenpiper et al. 2004, 27), are adulterated with prescription drugs (Miller and Stripp 2007, 9), or are contaminated with heavy metals (Buettner et al. 2009, 24; Cohen 2009, 361).

Synthetic Drugs Derived From Herbs

Many of our modern drugs (e.g., aspirin, Morphium, Tamoxifen, Vin­cris­tin, etc.) were originally derived from botanical material. In fact, many skeptics wonder why we cannot also extract and synthesize the active ingredients from well-researched herbal medicines such as St. John's Wort and generate single ingredients derived from that plant. This would clearly solve several problems inherent in herbal medicine, such as standardization.

While this approach of creating pure compounds does work occasionally, it fails in other instances. One reason can be the fact that herbal medicines tend to have not one but a multitude of pharmacologically active ingredients. Thus extracting only one ingredient might reduce the pharmacological activity of the whole plant extract.

Single ingredients derived from herb­al extracts can no longer be considered herbal medicines as, by definition, herbal medicines are based on the whole plant. Nevertheless, such drugs are reminders of the fact that many plants contain molecules that are pharmacologically active and can thus have both beneficial and detrimental health effects.

Traditional Herbalism

If a patient consults a Chinese, Indian, Japanese, or European herbalist, he will be diagnosed and treated according to obsolete and untested principles of diagnosis, pathophysiology, and so forth. Treatment will typically be individualized according to the characteristics of each patient and based on complex, tailor-made herbal mixtures of several (up to ten) herbal extracts. This means that ten patients suffering from depression may receive ten different, individualized concoctions, none of which might contain St. John's Wort, the only evidence-based herbal antidepressant. In other words, the biological plausibility of traditional herbalism is questionable.

Traditional herbalism is thus dramatically different from the herbal medicine described above. To scientifically test its value can be complex but it is doubtlessly possible. Few rigorous studies of this approach are currently available, and those that have been published do not support the notion that traditional herbalism is effective (Guo et al. 2007, 83).

Neither can we be certain about its safety. Because the tailor-made concoctions of traditional herbalists may contain a confusing number of active ingredients, the potential for toxicity, herb-drug interaction, contamination, and so on can be considerable. More vigorous regulation of herbalists, a subject currently being discussed in Europe (Hawkes 2010, 339), is therefore required.

Homeopathic Remedies

The public frequently confuses homeopathy with herbal medicine. The error usually arises because many homeopathic remedies are produced from "mother tinctures," which are based on herbal extracts. Thus they can carry the same (or similar) names as herbal products. The difference is that homeopathic remedies are typically highly diluted and therefore contain no active ingredients at all. Thus homeopathy lacks any biological plausibility.

Arnica is a good example. It is used as an herbal cream as well as a homeopathic remedy. Because it is toxic, Arnica should not be taken as an oral herbal medicine. Being highly diluted, homeopathic Arnica is, of course, both nontoxic and entirely ineffective (Ernst and Pittler 1998, 133).

Bach Flower Remedies

These products are currently very popular for self-medication, particularly in Europe. They are produced by placing freshly picked flowers in spring water. Thus they are also plant-derived and frequently confused with herbal medicines. After the flowers have floated for a while, the water is mixed with brandy and sold at high prices as Bach Flower Remedies.

Bach Flower Remedies were developed by the British physician Edward Bach, who had previously worked as a homeopath. His remedies have, however, little in common with homeopathy except, of course, that they are neither biologically plausible nor of proven effectiveness for any condition (Ernst 2010, 140).

Anthroposophical Medicines

Rudolf Steiner developed his anthroposophical medicines about one hundred years ago (Ernst 2008, 150). They are produced according to protocols similar to those of homeopathic remedies. Unlike homeopathy, however, anthroposophical medicine does not follow the "like cures like" principle.

As many anthroposophical medicines are based on plants, they are also often confused with herbal medicines. The best known example is Iscador®, a fermented mistletoe preparation that is a highly popular treatment for cancer in Europe. Numerous trials exist, but collectively their results do not show that this is an effective therapy (Horneber et al. 2008, 16).


Many articles on herbal medicine conclude by stating that more research is needed. Between 1999 and 2007, the National Institutes of Health has spent US $1.9 billion on research into dietary supplements (Regan, Wambogo, and Haggans 2011, 141). Not all of this money was well invested (Ernst et al. 2011). I therefore advocate not necessarily more research but better-designed studies into the few plausibly beneficial aspects of herbal medicine. n


Buettner, C., K.J. Mukamal, P. Gardiner, et al. 2009. Herbal supplement use and blood lead levels of United States adults. Journal of General Internal Medicine 24(11): 1175-82.

Cohen, P.A. 2009. American roulette: Con­taminated dietary supplements. New Eng­land Journal of Medicine 361(16): 1523-25.

Ernst, E. 2008. Anthroposophic medicine: A critical analysis [in German]. MMW Fortschritte der Medizin 150(Suppl. 1):1-6.

---. 2010. Bach flower remedies: A systematic review of randomised clinical trials. Swiss Medical Weekly 140: w13079.

Ernst, E., and M.H. Pittler, 1998. Efficacy of homeopathic arnica: A systematic review of placebo-controlled clinical trials. Archives of Surgery 133(11): 1187-90.

Ernst, E., M.H. Pittler, B. Wider, et al. 2006. The Desktop Guide to Complementary and Alternative Medicine, 2nd ed. Edinburgh: Elsevier Mosby.

Ernst, E., S.K. Hung, and Y. Clement. 2011. NCCAM-funded RCTs of herbal medicines: An important critical assessment. Perfusion 24(3) 89-102.

Guo, R., P.H. Canter, and E. Ernst. 2007. A systematic review of randomised clinical trials of individualised herbal medicine in any indication. Postgraduate Medical Journal 83(984): 633-37.

Hawkes, N. 2010. A spanner in the herbal works. BMJ 339: b5441.

Horneber, M.A., G. Bueschel, R. Huber, et al. 2008. Mistletoe therapy in oncology. Cochrane Database Systems Review 16(2): CD003297.

Izzo, A.A., and E. Ernst. 2001. Interactions between herbal medicines and prescribed drugs: A systematic review. Drugs 15: 2163-75.

Marcus, D.M., and A.P. Grollman. 2002. Botanical medicines: The need for new regulations. New England Journal of Medicine 347(25): 2073-76.

Miller, G.M., and R. Stripp. 2007. A study of western pharmaceuticals contained within samples of Chinese herbal/patent medicines collected from New York City's Chinatown. Legal Medicine 9(5): 258-64.

Regan, K.S., E.A. Wambogo, and C.J. Haggans. 2011. NIH and USDA funding of dietary supplement research, 1999-2007. Journal of Nutrition 141(1):1-3.

Schulz, V., and R. Hänsel. 2003. Rational phytotherapie: A physician's guide to herbal medicine, 5th ed. Berlin: Springer-Verlag.

Sievenpiper, J.L., J.T. Arnason, E. Vidgen, et al. 2004. A systematic quantitative analysis of the literature of the high variability in ginseng (Panax spp.): Should ginseng be trusted in diabetes? Diabetes Care 27(3): 839-40.

U.S. Government Accountability Office. 2009. Dietary Supplements: FDA Should Take Further Actions to Improve Oversight and Consumer Understanding. United States Accountability Office, January: Report to Con­gressional Requesters. GAO-09-250. Available online at

---. 2010. Herbal Dietary Supplements: Examples of Deceptive or Questionable Marketing Practices and Potentially Danger­ous Advice. United States Govern­ment Accountability Office, May 26: Testimony Before the Special Commitee on Aging, U.S. Senate. GAO-10-662T. Available online at

Edzard Ernst

Edzard Ernst, MD, PhD, is emeritus professor in the Complementary Medicine unit, Peninsula Medical School, Exeter, U.K.