More than half of the population of the United States uses "dietary supplements," as the term is defined in the Dietary Supplement Health and Education Act (DSHEA) of 1994. [Refs. 1, 2] Dietary supplements include vitamins, minerals, amino acids, enzymes, and herbs and other botanicals. Use of most dietary supplements does not appear to be associated with any serious adverse health effects, and there is evidence that supplements confer health benefits in certain cases. For example, it is well recognized that calcium helps to build strong bones and prevent osteoporosis and that folic acid helps to reduce the risk of neural tube defects. [Ref. 3]
Certain dietary supplements also include ephedra, a plant that contains six ephedrine alkaloids among which ephedrine is predominant. The dietary supplement industry estimates that approximately 12 million individuals were using ephedra in 1999. [Ref. 4] As the tragic death of Orioles' pitching prospect Steve Bechler has reminded Americans, ephedra has long posed difficult issues for clinical investigators, for regulators, and most importantly, for the health of the public, all of whom are concerned about the safety of ephedra relative to its potential benefits, if any. These concerns stem from both the sympathomimetic mechanism of action of ephedrine alkaloids, and accumulating evidence of potentially serious adverse events after use of ephedra-containing products.
FDA regulates ephedra under DSHEA, which places the burden of proof on the government if it wishes to take regulatory action against a supplement. The government must show that the supplement presents a "significant or unreasonable risk of illness or injury" under the conditions recommended or suggested in labeling (or under ordinary conditions of use if the labeling is silent). DSHEA's regulatory framework is primarily a "postmarket" program like the bulk of food regulation. Thus, there is no requirement for manufacturers to provide evidence of product safety prior to marketing ephedra-containing products. In contrast, drug regulation involves an extensive "premarket" evaluation of safety and effectiveness with explicit standards of evidence, providing opportunities for developing evidence to guide not only approval decisions but also conditions of use to manage benefits and risks. In addition, there are postmarket reporting requirements for drugs, to support product safety monitoring.
Ephedra is a complex substance that has unusual features with respect to these alternative regulatory approaches. Because it is a botanical, it meets the condition of a dietary supplement regulated under DSHEA. On the other hand, the principal active alkaloid in ephedra is ephedrine, and synthetic ephedrine and other ephedrine alkaloids are regulated as drugs, which are only marketed for indications where safety and effectiveness have been demonstrated. Synthetic ephedrine is available as a component of various over-the-counter and some prescription drug products for treating allergies, asthma, nasal congestion, and related upper respiratory symptoms, none of which include caffeine or other stimulants that may interact with its effects. All synthetic ephedrine drug products are subject to significant regulations in dosing and in manufacturing practices. There are no synthetic ephedrine products approved for long-term use.
When FDA moved to significantly restrict access to ephedra through a proposed regulation in 1997, an independent report by the General Accounting Office concluded that the available evidence did not support the FDA's proposed restrictions on dosage limits. [Ref. 5] The difficulties that FDA encountered five years ago in imposing restrictions on the availability of ephedrine alkaloid containing dietary supplements raise two critical health policy questions. Has the evidence raising concerns about ephedra's safety become significantly stronger since that time? Given the latest evidence on benefits and safety, can reasonable restrictions on ephedra labeling and availability be imposed under DSHEA?
Shekelle et al.'s review provides the best compilation of evidence yet on the safety profile of ephedra. [Ref. 6] This thorough review, including formal meta-analyses to the extent possible, covers the vast majority of the recent clinical evidence available on the safety and efficacy of ephedra and ephedrine for its two most popular uses: weight loss and athletic performance enhancement. With respect to effectiveness, the authors found a modest but relatively consistent effect of short-term weight loss from ephedrine or ephedra-caffeine combination products. However, none of these studies included treatment for more than 4 to 6 months or any follow-up after the product was stopped; there is therefore no evidence on the critical question whether there is a long-term weight loss effect that would translate into significant health outcome improvements. There is also no evidence on whether the known effects of ephedra (increase in blood pressure and long-term inotropic stimulus) have adverse long-term health consequences. In addition, it is unclear what dose is needed and whether caffeine is also required for long-term weight loss. Further, despite the widespread marketing of ephedra products as performance-enhancers, Shekelle et al. found no evidence to support the claim for ephedra and very limited evidence for ephedrine.
Shekelle et al.'s analysis also adds significantly to the literature on the safety profiles of ephedra and ephedrine. The meta-analysis concludes that ephedra and ephedrine, especially in combination with other stimulants, are associated with a higher risk of certain mild to moderate side effects that are plausible given its mechanism of action: palpitations and gastrointestinal, psychiatric, and autonomic symptoms. However, the meta-analysis only has enough statistical power to conclude that the rate of serious adverse events, including acute myocardial infarction, stroke, and death is very likely to be less than 1 in 1000. Thus, as has marked the history of inquiries into ephedra's safety, further analysis of safety risks involves case reports - the weakest form of epidemiological evidence since there are no direct "controls" for any confounding factors or even for the natural occurrence rate of these serious events. Shekelle et al.'s thorough review of the available information on adverse effects associated with ephedra or ephedrine, which included in aggregate more than 16,000 case reports, identified "two deaths, four myocardial infarctions, nine cerebrovascular accidents, one seizure, and five psychiatric cases" involving ephedra that were "sentinel events," where the available records appeared thorough and no other precipitating factors were evident. They also identified 10 such "sentinel events" involving ephedrine, and an additional 51 [43 for ephedra] where the compound might have been involved but other contributors were also present.
There has been much discussion of whether such events constitute the "smoking gun" for a causal relationship between ephedra and serious adverse events. An FDA advisory panel argued three years ago that such sentinel events, by themselves, cannot prove causality. [Ref. 7] Even if the records were completely accurate and there were no other competing explanations, heart attacks, hemorrhagic strokes, and sudden deaths occur in young, otherwise healthy persons, albeit at a very low rate. Given industry estimates of over 3 billion "servings" of ephedra sold each year, there is reason to expect some of these serious events in ephedra users, even if ephedra use did not cause them. However, adverse events related to dietary supplements are likely to be significantly underreported [Ref. 8], which would argue for taking reported events more seriously.
Subsequent to the end date of studies included in the review by Shekelle et al., additional studies on ephedra safety have been reported. [Refs. 9, 10, 11, 12] For example, Bent, et al. compared the share of reported serious adverse events for ephedra to the share of ephedra products among all dietary supplement sales - in effect using other dietary supplements to "control" for baseline rates of adverse events in the population. [Ref. 9] Ephedra accounted for 64% of all serious dietary supplement adverse events, but less than 1% of all sales. While it is possible that ephedra events are more likely to be reported because of the public attention devoted to the compound, and that ephedra users (because of their demographics and behavior, including consumption of other stimulants) are intrinsically more likely to have serious adverse events than other dietary supplement users, this is a very disproportionate representation.
The known pharmacologic properties of ephedrine alkaloids, together with these studies, suggest that concerns about serious complications from ephedra use are not unreasonable - especially in cases involving strenuous exercise, concomitant use of other stimulants, and underlying medical conditions that limit cardiovascular capacity or ability to discharge body heat. However, they do not establish definitive causal evidence of a statistically significant elevated risk of death or serious injury from ephedra. Nor do the studies firmly establish that ephedra has meaningful health benefits. The most evidence is available for weight loss, but even there, the demonstrated effect is only short term and is not more substantial than approved drugs for weight loss, whose risk profiles are better understood and which are used with physician guidance.
While additional studies of ephedra's safety and efficacy would clearly be useful, it is unlikely that any studies that could be conducted or completed, at least in the near term, would be powerful enough to resolve these safety questions. Even if a very costly, definitive large randomized clinical trial could be funded, it might be unethical to carry it out, given the risks suspected from ephedra and the likelihood that its health benefits are modest at best. Alternatively, additional valuable evidence might be obtained from a nonrandomized, prospective case-control study. For example, the Hemorrhagic Stroke Project was a critical element in the FDA's announcement of its intention to withdraw PPA from the market. It found a significantly increased stroke risk on the order of 1 in 100,000 in women (but not men) who took PPA as an appetite suppressant within a 3-day window. [Ref. 13] An equivalent study of ephedra would probably have to be done prospectively as well, because there is no existing population database that has reliably recorded both ephedra use and patient outcomes; consequently, drug histories would have to be obtained for patients (or their proxies) after the event occurs. It would take some years for such a study to accumulate enough cases to have the statistical power to detect a significant difference in serious adverse events.
However, some additional evidence related to ephedra safety comes from another source: the more restricted availability of synthetic ephedrine in products regulated as drugs, available in low doses and with specific labeling for short-term indications and mandatory warnings, has not been associated with the same magnitude of reported adverse events as ephedra. It seems plausible that a regulatory approach akin to that used for synthetic ephedrine would have a reasonable likelihood of avoiding some of the serious adverse effects that have been reported with ephedra use. Similar restrictions on marketing and access - which could be tightened or loosened depending on further scientific evidence - could provide an effective deterrent to some current practices that appear to exacerbate ephedra's potential risks.
On February 28, 2003, the Department of Health and Human Services announced a number of actions to address concerns about ephedra's safety. First, warning letters were sent to dozens of ephedra manufacturers challenging them to remove unproven claims or substantiate those claims, with a particular focus on athletic performance enhancement claims. Given the limited evidence on ephedra's benefits, FDA and the Federal Trade Commission are assessing whether further enforcement actions are warranted against other manufacturers. Second, a new, mandatory warning label for all marketed ephedra products was proposed. It would make it clear to users, via a black-box warning on the front of the product, as well as additional information in the product labeling, that serious adverse events and death have been reported after using ephedra, and that risks of adverse events are particularly high with strenuous exercise and/or use of stimulants including caffeine. Third, FDA announced that it was seeking comments from health professionals, the supplement industry, and the general public on any additional data on ephedra's safety, so that we can acquire the most complete picture possible of the product's potential risks.
Our Federal Register announcement was also intended to support comments on whether evidence of significant safety concerns, short of a definitive demonstration of a risk profile, is sufficient for new restrictions on products regulated under DSHEA. DSHEA established that a dietary supplement may not be sold if it "presents a significant or unreasonable risk of injury." We do not have definitive evidence that ephedra has caused serious injuries and deaths. Instead, "sentinel" events have been identified along with various types of additional suggestive evidence, including ephedra's known blood pressure and inotropic effects as well as an increasing number of cleverly designed observational studies. As noted above, such suggestive rather than definitive evidence seems to be all that is likely to emerge for dietary supplements with potential safety problems regulated under DSHEA. Although some may believe that the unreasonable risk standard requires FDA to prove definitively that a product presents a defined hazard if used as indicated by the manufacturer, there is another way to understand that requirement. "Unreasonable risk" clearly implies a risk-benefit calculus. Such a calculus should be able to examine the available scientific evidence and take it into account in assessing whether the product's known or suspected risks outweigh its known or suspected benefits, in light of the claims the product makes or under ordinary conditions of use.
Such a reading helps give DSHEA the meaning in practice that many of its supporters say it should have, by clarifying that public health authorities can take actions to protect the public from unreasonable but uncertain safety risks associated with ephedra. We intend to establish an up-to-date public record for further, legally sustainable regulatory actions based on the latest scientific evidence and statutory analysis, by analyzing public comments on this approach quickly. Alternatively, if DSHEA alone is not adequate, we have also asked for comments on possible legislative modifications to DSHEA to achieve this public health goal.
Finally, expanded research on the safety and effectiveness of dietary supplements is needed, particularly for certain botanicals. In recent years, the National Center for Complementary and Alternative Medicine and the Office of Dietary Supplements (components of NIH) have substantially expanded funding for studies of commonly used supplements. Such efforts would be significantly enhanced by clear standards for the labeling and production of all dietary supplements, which the FDA is committed to implementing soon. The resulting research could provide a foundation for science-based health claims by certain supplements, or, in cases where effects can be disproved, help eliminate public misunderstandings about the purported benefits of a supplement.
1. Kaufman DW, Kelly JP, Rosenberg L, Anderson TE, Mitchell AA. Recent Patterns Of Medication Use in the Ambulatory Adult Population of the United States: The Slone Survey. JAMA 2002;287(3):337-44.
2. Taylor H, Leitman R (editors). Widespread Ignorance of Regulation and Labeling of Vitamins, Minerals and Food Supplements, According to a National Harris Interactive Survey. Harris Interactive(r) Health Care News 2002; 2, (23):1-5. (available at: http://www.harrisinteractive.com/news/newsletters_healthcare.asp).
3. 21 C.F.R. 101.72 and 101.79.
4. Dietary Supplements for Weight Loss: Limited Federal Oversight Has Focused More on Marketing than on Safety (Testimony, 07/31/2002, GAO-02-985T).
5. Dietary Supplements: Uncertainties in Analyses Underlying FDA's Proposed Rule on Ephedrine Alkaloids (Letter Report, 07/02/1999, GAO/HEHS/GGD).
6. Shekelle PG.; Hardy ML, Maglione M, Morton SC; Ephedra and Ephedrine for Weight Loss and Athletic Performance Enhancement: Clinical Efficacy and Side Effects, Agency for Healthcare Research and Quality (in press).
7. Jones WK, Report on Public Meeting: "Safety of Dietary Supplements Containing Ephedrine Alkaloids", August 8-9, 2000, Washington, DC (available at: http://www.cfsan.fda.gov/~dms/ds-ephe3.html).
8. Walker A. The Relation between Voluntary Notification and Material Risk in Dietary Supplement Safety. FDA Docket 2000, 00N-1200, (41).
9. Bent S, Tiedt TN, Odden MC, Shlipak MG. The Relative Safety of Ephedra Compared with Other Herbal products. (available at url: www.acponline.org/journals/annals/ephedra.htm).
10. Morgenstern LB, Viscoli CM, Kernan WN, Brass LM, Broderick JP, Feldmann E, et al. Use of Ephedra-Containing Products and Risk for Hemorrhagic Stroke. Neurology 2003;60(1):132-5.
11. Samenuk D, Link MS, Homoud MK, Contreras R, Theohardes TC, Wang PJ, Estes NA 3d, Adverse Cardiovascular Events Temporally Associated with Ma Huang, an Herbal Source of Ephedrine, Mayo Clinic Proceedings, 2002, vol. 77(1):12-6.
12. Haller CA, Jacob P 3d, Benowitz NL, Pharmacology of Ephedra Alakaloids and Caffeine After Single-Dose Dietary Supplement Use, Clinical Pharmacology and Therpeutics, 2002, June , Vol. 71(6), pp. 421-31.
13. Kernan WN, Viscoli CM, Brass LM, Broderick JP, Brott T, et al. Phenylpropanolamine and the Risk of Hemorrhagic Stroke. NEJM 2000;343(25):1826-32.
Additional materials: