An Analysis of the Relative Safety of Ephedra
By Doug Kalman MS, Jose Antonio, Ph.D., FACSM, and Richard
B. Kreider, PhD,
In an early Internet release, the Annals of Internal
Medicine posted an upcoming brief communication concerning
the dietary supplement ephedra (1).
This study raised media frenzy concerning the regulatory
status of ephedra.
The authors utilized the Toxic Exposure Surveillance System
(TESS) report of 2001 and compared it with ephedra sales
data provided to them by SPINS, a market analysis firm.
In addition, the authors also utilized a magazine report to
approximate the total sales of ephedra within the United
States for the year 2000 (2).
There are several methodological and fundamental flaws with
the design and conclusions made by Bent et al.
The TESS raw data indicates that 55.5% of all Poison Control
Center reports related to Ma Huang (ephedra) alone or in
combination with another herb (multi-botanical) were in
people under the age of 19.
Additionally, 27.9% of all of the exposures were in children
less than 6 years of age (3).
This information is vital as in 7,927 exposures; the Poison
Control Centers deemed 14% (1,178) to be an adverse
In clinical research the guidelines set forth by the
International Committee on Harmonization (ICH) defines an
adverse reaction/event (AE) 'any untoward medical occurrence
in a patient or clinical investigation subject administered
a pharmaceutical product and which does not necessarily have
to have a causal relationship with this treatment' (4).
The TESS system defines an adverse reaction (AR) as 'an
adverse event occurring with normal, prescribed, labeled or
recommended use of the product, as opposed to overdose,
misuse or abuse'.
The TESS system also captures AR's that are 'unwanted
effects due to an allergic, hypersensitive, or idiosyncratic
response to the active or inactive ingredients, or
Thus, the definitions and establishment of clear causality
or relationship is not clear within the TESS system and when
contrasted with normal research guidelines for defining and
AE/AR appear to be questionable.
The Center for Drug Evaluation and Research (CDER) policy on
AR/AE's is that accumulated case reports (AER's) cannot be
used to calculate incidence or estimates of drug risk (5).
This misguided calculation is exactly what the authors
attempted to do.
The 2001 TESS report details that the vast amount of
exposures were unintentional (85.2%).
In the ephedra analysis, 46.7% of the exposures were of the
unintentional variety (using TESS definitions and data from
It cannot be downplayed that the TESS report only captured
data on 12 known herbs, Drs. Bent et al mistakenly state
that ephedra accounts for 64% of all herbal related adverse
reactions, however, there are hundred of herbals sold on the
U.S. market, not 12, thus their conclusion is overstated.
The sales data that Drs. Bent et al utilized in an attempt
to correlate the TESS data with sales is incomplete. The
SPINS database does not capture data by zip code nor does it
capture the true mass market (i.e., Walmart, Costco, GNC
Corporate stores), thus any data generated by the SPINS
agency is only a small snapshot of what is truly happening
in the sales of ephedra or ephedra-related products.
The Nutrition Business Journal estimates that in 2000,
ephedra and ephedra related products generated
Utilizing the NBJ market analysis, the best estimate is that
26,250,000 servings (or individual capsules/tablets) of
ephedra or ephedra related products were sold in 2000.
The sales figures are based upon retail mass market, mail
order, practitioners, Internet sales and natural food/health
chain channels (6).
In the Bent report, it is stated that an assumption was
made that ephedra related sales were one-half of all non-
retail herb sales and this accounted for 0.82% of herbal
The confliction in detail does not make sense. It appears
that the SPINS data is inaccurate when comparing it to the
more comprehensive NBJ data.
Thus, this section of the Bent paper appears to be out of
context and unreliable.
While we as scientists and health care providers need to
know the evidence (direct, not computed) concerning the
safety of ephedra or ephedra related products, we must not
fail to use the published peer-reviewed clinical studies as
the basis for an understanding.
While the clinical trials are limited in subject size as
compared to Phase III drug studies, they do give us a basis
for understanding the potential for serious adverse events
and what population is best suited for potential use of
It is clear that people under the age of 19 should not take
this herb; there simply have been no studies in that age
group (on the herbal ephedrine).
The TESS data states 55.5% of all exposures were from people
19 or younger.
The comparison of ephedra versus other herbs inherently
inaccurate as the TESS data only captured 12 total named
Given the TESS data for ephedra reporting an adverse
reaction rate of 14% (TESS conclusion) and a mortality rate
of 0.000757% (comparison of 6 deaths versus 7,927
exposures), one would expect a better comparison to be made
using this data.
For example with relation to kava, there was one death in
336 exposures (0.002976%), thus we can also conclude that
kava is 3.9 times as likely to cause death as ephedra.
It should also be noted that the adverse reaction frequency
was similar for Gingko biloba (13.7% vs 14%) as ephedra
and the AR for kava was much higher (17.5%).
Perhaps, a less negative conclusion would not serve the
purpose of the study.
The manipulative presentation of the data shared by Bent et
al viewed alongside the fact that the authors have and still
testify for plaintiff law firms on behalf of anti-ephedra
litigation, leads to speculation that this study's intent
was to establish their published paper as evidence that
ephedra is dangerous.
An informed professional audience must wonder where the
truth actually lays. Whose future and benefit does this
Douglas S. Kalman MS, RD, FACN
Miami Research Associates
6280 Sunset Drive
Miami, FL. 33143
Disclosure: Mr. Kalman has testified in cases related to
ephedra on behalf of Cytodyne Technologies, Inc.
Jose Antonio, Ph.D., FACSM
Exercise Science & Health Promotion
Florida Atlantic University
777 Glades Road
P. O. Box 3091
Boca Raton, FL 33431-0991
Richard B. Kreider, PhD, EPC, FACSM, FASEP
Professor & Chair
Exercise & Sport Nutrition Laboratory
Center for Exercise, Nutrition, and Preventive Health
Department of Health, Human Performance & Recreation
PO Box 97313
Waco, TX 76798-7313
Disclosure: Dr. Kreider has served as an expert in
litigation for Metabolife.
1) Bent S, Tiedt TN, Odden MC, Shiplak MG. The relative
safety of ephedra compared with other herbal products. Ann
Intern Med 2003;138:000-000. www.acponline.org/journals/annals/ephedra.htm
Accessed online February 5, 2003
2) Richman A, Witkowski JP. 7th Annual Herb Sales Survey.
Whole Foods Magazine. 2001:23-30.
3) Litovitz TL, Klein-Schwartz W, Rodgers GC, Cobaugh DJ,
Youniss J, Omslauer JC, May ME, Woolf AD, Benson BE. 2001
Annual report of the American Association of Poison Control
Centers Toxic Exposure Surveillance System. Amer J Emerg Med
4) Cohen A, Posner J. A Guide to Clinical Drug Research. 2nd
edition Kluwer Academic Publishers 2002. Pp XI, 34,-35, 154.
Accessed February 18, 2003.
6) NBJ's Supplement Business Report 2002. Penton Media, Inc.
Figure 5-5, Figure 5-7. Available: www.nutritionbusiness.com