Clinical Review

How to judge an herbal remedy

Author and Disclosure Information

How do you determine usefulness? It’s a matter of efficacy, safety, quality, and cost.



How many of your patients take herbal preparations? More than you think, thanks to the proliferation of products. Between 1990 and 1997, the US population increased its use of herbal medicines by 380%, and total out-of-pocket expenditures in 1997 were $5.1 billion (TABLE 1).1,2

Safety issues surrounding herbal medicine are complex: possible toxicity of herbal constituents, presence of contaminants or adulterants, and potential interactions between herbs and prescription drugs. In addition, the preparations are often poor in quality. One reason: They are inadequately regulated, a problem many experts hope to change. Cost evaluations of herbal medicines are not available.

This article offers guidelines for prescribing herbal medications, as well as advice on when they are unwise.


10 best-selling herbal medicines

1Ginkgo biloba$46
6Saw palmetto$25
7St John’s wort$24
10Black cohosh$10
* US, 2001 data

Is the herb effective for the patient’s condition?

Although data are incomplete, some treatments have shown promise (TABLE 2), and findings indicate serious adverse effects of certain treatments (TABLE 3).

Besides safety, the critical question is: Does the remedy work for the patient’s condition? Do not prescribe or recommend an herbal remedy if the answer is not a firm yes.

Herbal medicines usually contain a range of pharmacologically active compounds. In some cases, it is unclear which constituents produce the therapeutic effect. Testing for efficacy in this situation is obviously more complex than with synthetic drugs. One approach is to view the entire herbal extract as the active component.

To optimize the reproducibility of efficacy studies, extracts must be sufficiently characterized. This is often achieved by standardizing the amount of a single key constituent (eg, a pharmacologically active ingredient or a marker suitable substance if such an ingredient is unknown).

Once the dilemma of standardization is solved, herbal medicines are scrutinized in much the same way as other drugs. The literature contains several randomized, clinical trials and systematic reviews/meta-analyses of these studies.3,4 The Cochrane database includes about 30 systematic reviews of herbal medicines, and several authoritative books recently were published.3-6

Unfortunately, systematic reviews are often limited by the paucity and varied methodological quality of the primary studies,3,7 and research funds are generally scarce, in part because plants cannot be patented.

Generalizations about the efficacy of herbal medicines are not possible. Each remedy must be judged on its own merits. Some herbal products have demon strated efficacy for certain conditions, while others have not. Overall, few products have been subjected to extensive clinical testing.3

The bottom line? As a review in the New England Journal of Medicine concluded, “Clinicians should not prescribe or recommend herbal remedies without well-established efficacy.”7

Tradition is no guarantee, as in the case of kava

Consumers are attracted to herbal medicines in part because they equate “natural” with “safe.” Yet some herbal medicines pose serious risks.7

First, the active ingredients in herbal preparations can cause both desirable and undesirable effects. TABLE 3 lists examples of commonly used herbal medicines that have been associated with serious adverse effects.3 Traditional use is no guarantee of safety and no acceptable substitute for data.8

A poignant example is kava (Piper methysticum), an herbal remedy that has been used for centuries, apparently without problems. Numerous rigorous clinical trials have shown it to be a powerful anxiolytic agent,9 but it was recently associated with several cases of serious liver damage.10 As a result, it was withdrawn from the markets of several European countries, and the US Food and Drug Administration (FDA) has issued warnings about its hepatotoxic potential.

Second, the active ingredients in herbal medicines can interact with prescription drugs. For instance, extracts of St. John’s wort (Hypericum perforatum) act as an enzyme inducer on the cytochrome P450 system and increase the activity of the P-glycoprotein transmembrane transporter mechanism. Both effects lead to a reduction of the plasma level of several conventional drugs.11 Perhaps the most serious consequence would be insufficiently low cyclosporine levels in patients after organ transplantation, which jeopardize the success of this procedure.12

Third, some herbal medicines (particularly Asian herbal mixtures) are contaminated with heavy metals13; contain misidentified, toxic herbal ingredients14; or are adulterated with prescription drugs.15 Be sure an herbal medication cannot cause harm before prescribing or recommending it.

Next Article: