Dietary Supplements

Susan Smolinski
Regional Poison Control Center
Childrens Hospital of Michigan
Detroit, MI

Int J Med Toxicol 1998; 1(2): 12

The Dietary Supplement Health and Education Act (DSHEA) of 1994 made the presumption that dietary supplements “are safe within a broad range of intake, and safety problems with the supplements are relatively rare.” Coincident with this Act, Congress established a Commission on Dietary Supplement Labels to evaluate safety, consumer and health professional information needs, and labeling of health claims. The final report of this commission was published on November 24, 1997. Among other recommendations, the commission concluded that “FDA, the industry, the scientific community, and consumer groups work together to voluntarily improve passive postmarketing surveillance systems, including adverse reaction reporting systems, to ensure that any safety problems that may arise are identified and corrected promptly.” The commission also recommended that manufacturers be responsible for knowledge of potential side effects.(1)

Without patent protection, few manufacturers will be willing to conduct clinical trials to define the side effect profile for a given supplement. In the absence of a well-organized reporting system, how can the manufacturer fulfill this requirement? And, more importantly, how can the medical community be kept informed in order to evaluate the impact of alternate therapy on a patient's therapeutic regimen?

Several existing surveillance systems exist, each with its own flaws and limitations. FDA's MedWatch system is well established and utilized. By design, it focuses on severe or life-threatening reactions, thus serves primarily a sentinel role, but is less likely to answer questions about more routine side effects. It is clear that at least some botanical products, such as Ma Huang, have accumulated an extensive list of safety problems since passage of the DSHEA act, but the dissemination of this information to the general public and health practitioners via regulation of dosage and labeling is still pending final action by FDA.(2) What is wrong with this picture? Why is it that 33 cases of heart valve disease related to a drug combination resulted in voluntary withdrawal of the drugs “Phen-Fen” from the market within two months, whereas receipt of over 900 adverse reactions and at least 38 deaths from Ma Huang resulted in a proposed rule to regulate labeling and dosage several years later? It is ironic that immediately after the voluntary withdrawal of “Phen-Fen,” several herbal combinations appeared on store shelves and in weight loss clinics, containing Ma Huang in a very high dosage.

The major limitation of using the MedWatch system, assuming that FDA takes seriously the commission's suggestion to act more promptly on major concerns, is the lack of an appropriate timely method of communication of their findings to health professionals and the public. In the case of drugs, FDA has used MedWatch reports to enforce labeling changes on package inserts, which are then published in commonly used references, such as the Physician's Desk Reference. Since herbal products do not come with package inserts and there is no complete standardized compendia, most the information currently collected falls into a black hole.

A more promising surveillance tool for herbal products is the United States Pharmacopoeia's Practitioner's Reporting Network. Although the system is primarily designed to detect product quality problems, about 25% of the reports involve adverse drug reactions, which are automatically forwarded to FDA.(3) The USP has begun producing monographs on herbs and other botanicals and is preparing a database on botanicals approved in other national compendia. Three monographs have been published (valerian, ginger, comfrey), with others pending. There is thus in place a method for dissemination of botanical adverse effect information through the publication USPDI, in both consumer and health professional language. Important emergent reactions or quality problems could be reported through the USP journal, Quality Review. This information system has the advantage of credibility to both health professionals and the herbal industry, whereas FDA is perceived by the industry as a biased source with a negative agenda. The two agencies working together, may in fact, facilitate a prompt reaction to serious problems.

The TESS system used by poison control centers has a distinct advantage of comprehensive and broad surveillance. For example, exposures that result in no or minor problems are captured, which provides a more complete view of the side effect profile. There are several obstacles that must be overcome before TESS can be useful for collection of herbal adverse reactions. The effectiveness of TESS relies on the willingness and ability of Micromedex to include herbal products in the POISINDEX database. Thus far, the record is dismal. Needless to say, TESS cannot tract products that do not exist in the database. Bother generic and product AAPCC code listings within POISINDEX are inadequate and frequently miscoded, despite an effort to link herbal products to a single herbal category for the 1998 database. For instance, there is no product entry for "dietary supplements," so finding the unique code number attached to the herbal category is a challenge.

Assuming that these limitations will be addressed, there are still some limitations of the TESS system itself. TESS has been incredibly effective when drastic measurers of toxicity were examined. For example, analysis of cases involving pure calcium channel blockers showed a 10-fold greater fatality rate compared to the overall database.(4) Accumulating a side effect profile for an individual dietary supplement ingredient will be more difficult. Death should be an unusual occurrence for dietary supplements, thus will not be the primary outcome parameter. Analysis of major outcomes within the dietary supplement category may be the best approach to serve as sentinel events. Another limitation, however, is the ability of poison centers to handle an increasing number of calls, which are inherently more complex. With 50% of the population using dietary supplements and multi-herb therapy being the rule rather than the exception, each call reporting a side effect requires meticulous and time-consuming evaluation in order to conclude a cause-effect relationship, with consideration of dosage, possibility of contamination, and verification of ingredients. In addition to the time involved in making such assessments, there is no place within TESS to include these factors.

Finally, there is another possible mechanism for those willing to make the extra effort. Currently, there are 18 poison centers involved in a prospective study of dietary supplements. The primary aims are to create a registry of calls to determine the annual incidence of dietary supplement calls and to characterize the adverse reactions for individual supplements or ingredients. In addition to the TESS form, a supplemental data form is used to capture dosage, source and intent of product use, concurrent therapy, and outcomes. One immediate outcome of the study is that Micromedex has promised to include any product information gathered by the study group into the POISINDEX database. The study is scheduled to run from January 1998 to December 1998, but will most likely be continued, with modifications made based on experience from the first year. The eventual publication of the study results will add greatly to our knowledge of the side effect profile of commonly used botanicals. At our center, the study has already resulted in recording of 5 times the number of cases captured by TESS. In addition, some unique twists to adverse reactions are surfacing. In one case, the staff specialist listed a “therapeutic” exposure to a product containing Ma Huang. The patient was shaky, jittery, and nervous after previously using the product without incident. Upon careful questioning using the study form, the patient revealed also drinking 32 ounces of coffee just prior to experiencing the reaction. As with other toxicities, herbal calls are often not as initially advertised and probably more likely to be multifactorial. In another case, a young patient presented with new onset diabetic come, pancreatitis, renal failure and numerous other complications. Along with the patient were nine bottles of herbal supplements, mostly containing Echinacea species. Further history revealed that the patient had been feeling ill for two weeks and self-treating with herbal products. The delay in seeking medical attention clearly contributed to the poor outcome in this case. Although not strictly an “adverse reaction” the consequence of choosing alternative therapy was devastating. It is these kinds of documentation that the Presidential Commission needed to hear before concluding that “safety problems are relatively rare.” Continuing the study on a routine basis could add significant value to other means of postmarketing surveillance for dietary supplements, and perhaps could be explored as a benefit worthwhile of federal funding for Poison Centers.


  1. Commission on Dietary Supplement Labels, US Government Printing Office, 017-001-00531-2, Nov 24, 1997.
  2. Food and Drug Administration: Dietary Supplements containing Ephedrine Alkaloids; Proposed Rules, Fed Register 1997;62:30677-30724.
  3. United States Pharmacopoeia, DPPR At a Second Glance, Quality Review. 1996;55:1-4.
  4. Litovitz, T. The TESS database; use in Product Safety Assessment. Drug Safety 1998;18:9-19.

Int J Med Toxicol 1998; 1(2): 12

This article is located at

Quick Survey
Please rate this article:
1 2 3 4 5
(5 is best)

IJMT Home | Current Issue | Past Issues | Search | Technical Support | Send Comments to ACMTNet

Copyright 1999-2003, American College of Medical Toxicology.