[Federal Register Volume 69, Number 28 (Wednesday, February 11, 2004)]
[Rules and Regulations]
[Pages 6788-6854]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: 04-291]



[[Page 6787]]

-----------------------------------------------------------------------

Part III





Department of Health and Human Services





-----------------------------------------------------------------------



Food and Drug Administration



-----------------------------------------------------------------------



21 CFR Part 119



Final Rule Declaring Dietary Supplements Containing Ephedrine Alkaloids 
Adulterated Because They Present an Unreasonable Risk; Final Rule

Federal Register / Vol. 69, No. 28 / Wednesday, February 11, 2004 / 
Rules and Regulations

[[Page 6788]]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Part 119

[Docket No. 1995N-0304]
RIN 0910-AA59


Final Rule Declaring Dietary Supplements Containing Ephedrine 
Alkaloids Adulterated Because They Present an Unreasonable Risk

AGENCY: Food and Drug Administration, HHS.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: The Food and Drug Administration (FDA, we, our) is issuing a 
final regulation declaring dietary supplements containing ephedrine 
alkaloids adulterated under the Federal Food, Drug, and Cosmetic Act 
(the act) because they present an unreasonable risk of illness or 
injury under the conditions of use recommended or suggested in 
labeling, or if no conditions of use are suggested or recommended in 
labeling, under ordinary conditions of use. We are taking this action 
based upon the well-known pharmacology of ephedrine alkaloids, the 
peer-reviewed scientific literature on the effects of ephedrine 
alkaloids, and the adverse events reported to have occurred in 
individuals following consumption of dietary supplements containing 
ephedrine alkaloids.

DATES: This rule is effective on April 12, 2004.

FOR FURTHER INFORMATION CONTACT: Wayne Amchin, Center for Food Safety 
and Applied Nutrition (HFS-007), Food and Drug Administration, 5600 
Fishers Lane, Rockville, MD 20857, 301-827-6733.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Introduction
    A. Why Have We Concluded That Dietary Supplements Containing 
Ephedrine Alkaloids Present an Unreasonable Risk?
    B. What Are the Ephedrine Alkaloids and Where Do They Come From?
    C. What Regulatory Actions Have We Taken Regarding Dietary 
Supplements Containing Ephedrine Alkaloids?
    D. Petitions Received Relating to Dietary Supplement Containing 
Ephedrine Alkaloids
II. Summary of Letters and Comments
III. Finding of Adulteration
    A. What Does the Final Rule Do?
    B. What Products are Covered?
IV. Legal Issues
    A. What Is Our Legal Authority Under the Act?
    B. Do the Ephedrine Alkaloid-Containing Products Covered by this 
Rule Fall Within the Definition of Dietary Supplement Under the Act?
    C. Administrative Procedures
V. Scientific Evaluation
    A. How Did We Evaluate the Evidence?
    B. What Are the Known and Reasonably Likely Risks Presented by 
Dietary Supplements Containing Ephedrine Alkaloids?
    1. Pharmacology
    2. Other Safety Data
    3. Comparison with Drug Products Containing Ephedrine Alkaloids
    4. Abuse and Misuse
    5. Traditional Asian Medicine
    6. Adverse Events
    C. What Are the Known and Reasonably Likely Benefits of Dietary 
Supplements Containing Ephedrine Alkaloids?
    1. Weight Loss
    2. Enhancement of Athletic Performance
    3. Eased Breathing
    4. Other Uses
    D. Do Dietary Supplements Containing Ephedrine Alkaloids Present 
an Unreasonable Risk?
    1. What Does ``Unreasonable Risk'' Mean?
    2. Do Dietary Supplements Containing Ephedrine Alkaloids Present 
an Unreasonable Risk for Labeled or Ordinary Conditions of Use?
    3. Conclusion
VI. Why We Conclude that Other Restrictions Would Not Adequately 
Protect Consumers from the Risks Presented by Dietary Supplements 
Containing Ephedrine Alkaloids
    A. Warning Statement Alone
    B. Multiple Restrictions
    C. Self-Regulation
    D. More Education
    E. Nonbinding Guidance
    F. Targeted Enforcement Actions
VII. Miscellaneous Issues
    A. Freedom of Choice/FDA Bias
    B. Conduct of the Advisory Committee Meetings
VIII. Analysis of Impacts
    A. Benefit-Cost Analysis
    1. Introduction
    2. Regulatory Options
    3. Summary of Conclusions
    4. Option One--Take No New Regulatory Action
    5. Option Two--Remove Dietary Supplements Containing Ephedrine 
Alkaloids from the Market
    6. Option Three--Require the 2003 Proposed Warning Statement
    7. Option Four--Require the Proposed Warning Statement, But 
Modify it or Require it Only on Certain Products
    8. Option Five--Generate Additional Information or Take Some 
Action Other Than Removing Dietary Supplements Containing Ephedrine 
Alkaloids From the Market or Requiring Warning Statements
    9. Benefit-Cost Analysis: Summary
    B. Small Entity Analysis
IX. Environmental Impact
X. Paperwork Reduction Act
XI. Federalism
XII. References

I. Introduction

A. Why Have We Concluded That Dietary Supplements Containing Ephedrine 
Alkaloids Present an Unreasonable Risk?

    We conclude that dietary supplements containing ephedrine alkaloids 
are adulterated under section 402(f)(1)(A) (21 U.S.C. 342(f)(1)(A)) of 
the act because they present an unreasonable risk of illness or injury 
under the conditions of use recommended or suggested in labeling, or if 
no conditions of use are suggested or recommended in labeling, under 
ordinary conditions of use. Dietary supplements containing ephedrine 
alkaloids are most often used for weight loss, energy, or to enhance 
athletic performance.
    By its plain language, section 402(f)(1)(A) of the act requires 
evidence of ``significant or unreasonable risk'' of illness or injury. 
There is no requirement that there be evidence proving that the product 
has caused actual harm to specific individuals, only that scientific 
evidence supports the existence of risk. The Government's burden of 
proof for ``unreasonable risk'' is met when a product's risks outweigh 
its benefits in light of the claims and directions for use in the 
product's labeling or, if the labeling is silent, under ordinary 
conditions of use. ``Unreasonable risk,'' thus, represents a relative 
weighing of the product's known and reasonably likely risks against its 
known and reasonably likely benefits. In the absence of a sufficient 
benefit, the presence of even a relatively small risk of an important 
adverse health effect to a user may be unreasonable. Because it is not 
reasonable to conclude that a product is too risky in the absence of 
any significant evidence, some weight of evidence of risk is required 
to meet this standard. For example, isolated adverse events alone might 
not be expected to constitute substantiation of risk, but adverse event 
reports combined with pharmacological and other clinical evidence might 
be expected to do so.
    In considering whether dietary supplements containing ephedrine 
alkaloids present an unreasonable risk, we considered evidence from 
three principal sources: (1) The well-known, scientifically established 
pharmacology of ephedrine alkaloids; (2) peer-reviewed scientific 
literature on the effects of ephedrine alkaloids; and (3) the adverse 
events (including published case reports) reported to have occurred 
following consumption of dietary supplements containing ephedrine 
alkaloids.

[[Page 6789]]

    Ephedrine alkaloids are members of a large family of 
pharmacological compounds called sympathomimetics. Sympathomimetics 
mimic the effects of epinephrine and norepinephrine, which occur 
naturally in the human body. Multiple studies demonstrate that dietary 
supplements containing ephedrine alkaloids, like other 
sympathomimetics, raise blood pressure and increase heart rate. These 
products expose users to several risks, including the consequences of 
increased blood pressure (e.g., serious adverse events such as stroke, 
heart attack, and death) and increased morbidity and mortality from 
worsened heart failure and pro-arrhythmic effects. Based on the best 
available scientific data and the known pharmacology of ephedrine 
alkaloids and similar compounds, we conclude that dietary supplements 
containing ephedrine alkaloids pose short-term and long-term risks. 
This is clearest in long-term use, where sustained increased blood 
pressure in any population will increase the risk of stroke, heart 
attack, and death, but there is also evidence of risk from shorter-term 
use in patients with heart failure or underlying coronary artery 
disease.
    The data do not indicate that these products provide a health 
benefit sufficient to outweigh these risks. The best clinical evidence 
for a benefit is for weight loss, but even there the evidence supports 
only a modest short-term weight loss, insufficient to positively affect 
cardiovascular risk factors or health conditions associated with being 
overweight or obese. Even if long-term weight loss could be achieved 
with the use of dietary supplements containing ephedrine alkaloids, we 
believe that the risks posed by these products when used continuously 
in the long term generally could not be adequately mitigated except 
through physician supervision. Other possible benefits, such as 
enhanced athletic performance, enhanced energy, or a feeling of 
alertness, lack scientific support and/or provide only temporary 
benefits that we consider trivial compared to the risks of these 
products, which may include long-term or permanent consequences like 
heart attack, stroke, and death. Therefore, we have determined that the 
risks of dietary supplements containing ephedrine alkaloids, when used 
for their labeled indications or under ordinary conditions of use, 
outweigh the benefits of these products. We do not believe these risks 
can be adequately mitigated through other regulatory measures available 
to FDA for dietary supplements, such as warnings in labeling.
    As with other sympathomimetics, we believe that the risks posed by 
dietary supplements containing ephedrine alkaloids, when used 
continuously over the long term, generally cannot be adequately 
mitigated except through physician supervision. Similar to over-the-
counter (OTC) single ingredient ephedrine and pseudoephedrine products, 
we expect that dietary supplements containing ephedrine alkaloids could 
be marketed without physician supervision for a very temporary, 
episodic use that provides a benefit that outweighs the known and 
reasonably likely risks of these products. However, we are currently 
unaware of any such use, and our experience with ephedrine alkaloid-
containing OTC drug products suggests that such benefits will be 
demonstrable only for disease uses.

B. What Are the Ephedrine Alkaloids and Where Do They Come From?

    The ephedrine alkaloids, including, among others, ephedrine, 
pseudoephedrine, norephedrine, methylephedrine, norpseudoephedrine, 
methylpseudoephedrine, are chemical stimulants that occur naturally in 
some botanicals (Refs. 1 through 5), but can be synthetically derived. 
The ingredient sources of the ephedrine alkaloids in dietary 
supplements include raw botanicals (i.e., plants) and extracts from 
botanicals. Ma huang, Ephedra, Chinese Ephedra, and epitonin are 
several names used for botanical ingredients, primarily from Ephedra 
sinica Stapf, Ephedra equisetina Bunge, Ephedra intermedia var. 
tibetica Stapf and Ephedra distachya L. (the Ephedras), that are 
sources of ephedrine alkaloids (Refs. 1, 6, and 7). Other plant sources 
that contain ephedrine alkaloids include Sida cordifolia L. and 
Pinellia ternata (Thunb.) Makino (Refs. 8 and 9). Common names that 
have been used for the various plants that contain ephedrine alkaloids 
include sea grape, yellow horse, joint fir, popotillo, and country 
mallow. The names desert herb, squaw tea, Brigham tea, and Mormon tea 
refer to North American species of Ephedra that do not contain 
ephedrine alkaloids but have been misused to identify ephedrine 
alkaloid containing ingredients. Although the proportions of the 
various ephedrine alkaloids in botanical species vary from one species 
to another, in most species used commercially, ephedrine is typically 
the predominant alkaloid in the raw material (Ref. 10).
    Dietary supplements containing ephedrine alkaloids are widely sold 
in the United States (Refs. 11 through 13).\1\ Over the last decade, 
dietary supplements containing ephedrine alkaloids have been labeled 
and used primarily for weight loss, energy, or to enhance athletic 
performance. Additional scientific evidence, and numerous reports of 
serious adverse events, including death, following consumption of 
dietary supplements containing ephedrine alkaloids, have raised 
concerns about their safety. Consequently, we have taken a number of 
actions in an attempt to protect the public from the risks of these 
products.
---------------------------------------------------------------------------

    \1\ We use the term ``dietary supplements containing ephedrine 
alkaloids'' in this final rule to refer to dietary supplements 
containing botanical sources of ephedrine alkaloids. We use the term 
``ephedra'' to refer to botanical sources of ephedrine alkaloids, 
whether derived from a member of the Ephedra genus or another 
botanical, such as Sida cordifolia L. or Pinellia ternata (Thunb.) 
Makino. We use the term ``Ephedra'' to refer specifically to the 
Ephedra genus of plants.
---------------------------------------------------------------------------

C. What Regulatory Actions Have We Taken Regarding Dietary Supplements 
Containing Ephedrine Alkaloids?

    In the Federal Register of June 4, 1997 (62 FR 30678) (June 1997 
proposal), we published a proposed rule on dietary supplements 
containing ephedrine alkaloids. In this document, we proposed to make a 
finding, with the force and effect of law, that a dietary supplement is 
adulterated if it contains 8 milligrams (mg) or more of ephedrine 
alkaloids per serving, or if its labeling suggests or recommends 
conditions of use that would result in an intake of 8 mg or more in a 
6-hour period or a total daily intake of 24 mg or more of ephedrine 
alkaloids. The June 1997 proposal would also have required that the 
label of dietary supplements containing ephedrine alkaloids state that 
the product should not be used for more than 7 days. We also proposed 
to prohibit the use of ephedrine alkaloids in dietary supplements with 
other ingredients that have a known stimulant effect that may interact 
with ephedrine alkaloids, and to prohibit labeling claims, such as 
weight loss or body building, that require long-term intake to achieve 
the purported effect. In addition, the June 1997 proposal would have 
required a statement accompanying claims that encourage short-term 
excessive intake to enhance a purported effect, such as an increase in 
energy, that taking more than the recommended serving may result in 
serious adverse health effects. We also proposed to require that the 
labels of all dietary supplements containing ephedrine alkaloids bear a 
statement warning consumers not to use the product if they are taking 
certain drugs;

[[Page 6790]]

advising them to contact a health care professional before use if they 
have certain diseases or health conditions; and warning them to stop 
use and call a health care professional if they develop certain signs 
or symptoms. We proposed these actions in response to reports of 
serious illnesses and injuries, including a number of deaths, 
associated with the use of dietary supplements containing ephedrine 
alkaloids and our investigations and assessment of these illnesses and 
injuries. These actions were also supported by many of the 
recommendations made during the October 1995 meeting of an ad hoc 
Working Group of the FDA Advisory Committee (Working Group) and the 
August 1996 meeting of the Food Advisory Committee (FAC) and the 
Working Group concerning the potential public health problems 
associated with the use of dietary supplements containing ephedrine 
alkaloids and what action FDA should take to address the serious health 
concerns associated with their use (Refs. 14 and 15).
    The comment period for the June 4, 1997, proposed rule ended on 
August 18, 1997. In a document published in the Federal Register of 
August 20, 1997 (62 FR 44247), we announced our intent to reopen the 
comment period after we corrected a number of inadvertent omissions in 
the administrative record. Subsequently on September 18, 1997 (62 FR 
48968), we reopened the comment period until December 2, 1997.
    During this second comment period, the Commission on Dietary 
Supplement Labels (the Commission) released its final report on 
November 24, 1997. The Commission, an independent agency established by 
section 12 of the Dietary Supplement Health and Education Act of 1994 
(DSHEA) (Public Law 103-417), was charged with conducting a study on, 
and providing recommendations for, the regulation of label claims and 
statements for dietary supplements. The Commission's members included 
several scientists from academia and industry. In its report, the 
Commission divided its conclusions into three categories: findings, 
guidance, and recommendations. The Commission Report defined 
``findings'' as conclusions reached by the Commission based on 
information and data it received during its deliberations. The 
Commission defined ``guidance'' that was directed to FDA as advice that 
we should consider as we developed or implemented activities related to 
the availability of dietary supplements in the marketplace. The 
Commission defined ``recommendations'' as suggested changes to FDA 
regulations or the development of new regulations governing dietary 
supplements.
    One guidance statement in the Commission Report pertains to the 
safety of dietary supplements containing ephedrine alkaloids. In the 
report, the Commission urges FDA to use its authority under DSHEA to 
take swift enforcement action to address potential safety issues such 
as those posed recently by products containing ephedrine alkaloids. 
While it is expected that a responsible industry will avoid marketing 
unsafe products and that the industry will react promptly to remove 
products shown to be associated with significant or serious adverse 
events, in the final analysis there must be a strong and reliable 
enforcement system to back up the safety provisions of DSHEA. Failure 
by FDA to act when strong enforcement is needed undermines public 
confidence in the ability of not only the Federal Government but also 
the dietary supplement industry to ensure safety and avoid harm to the 
public (Ref. 16 at p. VII of Executive Summary).
    In a notice published in the Federal Register on April 29, 1998 (63 
FR 23633), we announced our views on the recommendations and guidance 
of the Commission, as presented in the Commission's report. In this 
notice, we stated that we take seriously our public health protection 
mission and are committed to removing unsafe dietary supplements from 
the market (63 FR 23633 at 23634). The direction taken in the current 
rulemaking on dietary supplements containing ephedrine alkaloids is 
consistent with the Commission's advice.
    In September 1998, the U.S. General Accounting Office (GAO) began a 
study on FDA's June 1997 proposal. GAO's work culminated in the 
issuance of a July 1999 report (Ref. 17). GAO concluded that the 
evidence supported concern that ephedrine alkaloid-containing 
supplements can cause serious health problems and it recommended 
further data collection and review. At the same time, GAO criticized 
FDA's reliance on adverse event reports (AERs) as the basis for the 
proposed restrictions on dosage, frequency and duration of use.
    In the Federal Register of April 3, 2000 (65 FR 17474, April 3, 
2000), we withdrew parts of the June 1997 proposal. More specifically, 
we withdrew the proposed finding that a dietary supplement is 
adulterated if it contains 8 mg or more of ephedrine alkaloids per 
serving, or if its labeling suggests or recommends conditions of use 
that would result in the intake of 8 mg or more in a 6-hour period or a 
total daily intake of 24 mg or more of ephedrine alkaloids; the 
proposed compliance procedures (regarding the analytical method FDA 
would use to determine the level of ephedrine alkaloids in a dietary 
supplement); the proposed label statement ``Do not use this product for 
more than 7 days;'' the proposed prohibition on labeling claims for 
uses that encourage long-term intake; and the proposed label statement 
to accompany claims for short-term uses (``Taking more than the 
recommended serving may cause heart attack, stroke, seizure, or 
death.'').
    We stated in our 2000 partial withdrawal of the June 1997 proposal 
that we continued to have a public health concern about the use of 
dietary supplements containing ephedrine alkaloids and that we would 
continue to monitor and provide appropriate followup on adverse events 
associated with the use of these products. We also stated that 
withdrawal of certain provisions of the June 1997 proposal did not 
limit our discretion to initiate enforcement actions with respect to 
dietary supplements containing ephedrine alkaloids.
    On the same day as the 2000 partial withdrawal of the June 1997 
proposal, we announced the availability of certain documents to update 
the administrative docket of the proposed rule (65 FR 17509, April 3, 
2000). The documents consisted of additional information about some of 
the 270 adverse event reports (AERs) received by FDA between February 
and September 1997. In a separate Federal Register notice also issued 
on April 3, 2000, we announced the availability of additional AERs and 
related information received after publication of the proposed rule. 
The additional information included the analyses of these new AERs by 
experts both inside and outside the agency; review of labels of 
products associated with these adverse events; review of the use of 
Ephedra species in traditional Asian medicine; analysis of the 
likelihood and factors affecting the reporting of adverse events; and 
summaries of the known physiological, pharmacological, and toxic 
effects of ephedrine alkaloids (Ref. 18). This announcement was made in 
part to prepare for a meeting convened by the U.S. Department of Health 
and Human Services (HHS) Office of Women's Health (OWH) in August 2000 
to discuss information about the safety of dietary supplements 
containing ephedrine alkaloids. Shortly before that meeting, FDA 
announced (65 FR 46721, July 31, 2000) that it would again reopen the 
comment period for the June 1997

[[Page 6791]]

proposal from August 10, 2000 (the day after the OWH meeting) until 
September 30, 2000. In that notice, we also announced the availability 
of a report on phenylpropanolomine and hemorrhagic stroke (Ref. 19).
    In April 2001, HHS's Office of the Inspector General issued a 
report entitled ``Adverse Event Reporting For Dietary Supplements: An 
Inadequate Safety Valve'' (Ref. 20) that assessed the effectiveness of 
FDA's Adverse Event Reporting System. This report found that adverse 
event reporting systems typically detect only a small proportion of the 
events that actually occur.
    In the Federal Register of March 5, 2003 (68 FR 10417), we 
published a notice making available new information about dietary 
supplements containing ephedrine alkaloids and requesting public 
comment on the new information and on regulation of these products (68 
FR 10417, March 5, 2003) (March 2003 notice). We specifically sought 
comments on whether, in light of current information, we should 
determine that dietary supplements containing ephedrine alkaloids are 
adulterated because they present a significant or unreasonable risk of 
illness or injury under the conditions of use recommended or suggested 
in labeling or under ordinary conditions of use if the labeling is 
silent. The notice also sought comment on a revised version of the 
warning statement first proposed on June 4, 1997. The revised warning 
statement had two components, a short warning that would be required to 
appear on the principal display panel (PDP) and a longer warning that 
could appear elsewhere in labeling. The proposed PDP warning stated 
that strokes, heart attacks, seizures, and death have been reported 
after consumption of dietary supplements containing ephedrine alkaloids 
and that the risks of adverse events increase with strenuous exercise 
and with use of other stimulants, including caffeine. The longer 
proposed warning included more detailed information about risks 
associated with the use of the product and recommended that consumers 
avoid using the product and/or consult a doctor under certain 
circumstances.
    In the March 2003 notice, we asked for public comment on all 
additional evidence developed since the publication of the June 1997 
proposal. One such study was a report by the Southern California 
Evidenced Based Practice Center (the RAND report, RAND, or RAND Corp.), 
commissioned by the National Institutes of Health (NIH) (Refs. 21 and 
22). RAND reviewed recent evidence on the risks and benefits of ephedra 
and ephedrine\2\ and found that dietary supplements containing 
ephedrine alkaloids are associated with higher risks of mild to 
moderate side effects such as heart palpitations, psychiatric effects, 
and upper gastrointestinal effects, and symptoms of autonomic 
hyperactivity such as tremor and insomnia, especially when they are 
taken with other stimulants. The RAND report identified 21 ``sentinel 
events'' among the adverse event reports it reviewed, including stroke, 
heart attack, and death.\3\ RAND also found limited evidence of an 
effect of ephedra on short-term weight loss. Furthermore, RAND found 
limited evidence that synthetic ephedrine and caffeine in combination 
have a short-term enhancement effect on athletic performance in certain 
physical activities. RAND concluded that the scientific literature does 
not support an effect of ephedrine alone on athletic performance, and 
there were no clinical trials on the effects of dietary supplements 
containing botanical ephedrine alkaloids on athletic performance. One 
of the studies reviewed by RAND, a study by Boozer, et al. (2002), 
though frequently relied on by the dietary supplement industry to 
demonstrate the safety of ephedrine alkaloids, raised additional 
concerns about the effects of dietary supplements containing ephedrine 
alkaloids on blood pressure. This evidence, discussed in section V.B of 
this document, added significantly to the evidence suggesting that 
dietary supplements containing ephedrine alkaloids as currently 
marketed are associated with unreasonable safety risks.
---------------------------------------------------------------------------

    \2\ The RAND report uses the term ``ephedra'' to refer to 
ephedrine alkaloids from botanical sources, whether or not they are 
contained in dietary supplements. RAND uses the term ``ephedrine'' 
to refer to pharmaceutical sources of ephedrine.
    \3\ RAND defined a ``sentinel event'' as a case that met all 
three of the following criteria: (1) Documentation of an adverse 
event that met the selection criteria; (2) documentation that the 
person having the adverse event took an ephedra-containing 
supplement or ephedrine within 24 hours prior to the event (for 
cases of death, myocardial infarction [heart attack], stroke, or 
seizure); and, (3) documentation that alternative explanations for 
the adverse event were investigated and were excluded with 
reasonable certainty. These criteria were subject to procedures 
which included the following (among other procedures): medical 
record documentation that an adverse event had occurred; 
documentation that the subject had consumed ephedra or ephedrine 
within 24 hours prior to the adverse event, or that a toxicological 
examination revealed ephedrine or one of its associated products in 
the blood or urine. Cases with no such documentation were not 
reviewed further. For the Metabolife cases, ephedra was assumed to 
have been used within the prior 24 hours for all but psychiatric 
events. All cases of stroke that met the criterion of having 
consumed ephedra or ephedrine within 24 hours were reviewed in more 
detail; to be classified as a ``sentinel event,'' reports of 
thrombotic stroke needed to have an assessment for a hypercoagulable 
state and vasculitis, reports of embolic stroke needed to have an 
embolic evaluation performed, and reports of hemorrhagic stroke 
required an examination to assess structural problems with the 
circulatory system of the brain.
---------------------------------------------------------------------------

    At about the same time as we published the March 2003 notice, we 
issued warning letters to 26 firms for making unsubstantiated claims 
concerning the use of dietary supplements containing ephedrine 
alkaloids to enhance athletic performance. We also issued warning 
letters to firms promoting dietary supplements containing ephedrine 
alkaloids as alternatives to illicit street drugs.
    In July 2003, GAO testified at a House Subcommittee hearing on 
issues relating to dietary supplements containing ephedrine alkaloids. 
GAO's testimony discussed and updated some of its findings from its 
prior 1999 report on dietary supplements containing ephedrine alkaloids 
(Ref. 23). The testimony provided new information, including an 
evaluation of Metabolife International's records of health-related 
calls from consumers of Metabolife 356 (Ref. 24). GAO noted that the 
types of adverse events identified in the health-related call records 
from Metabolife International were consistent with the types of adverse 
events reported to us, as well as with the scientifically documented 
physiological effects of ephedrine alkaloids. GAO also noted that 
despite the limited information contained in most of the call records, 
14,684 call records contained reports of at least one adverse event 
among consumers of Metabolife 356. The GAO testimony identified 92 
serious events that included heart attacks, strokes, seizures, and 
deaths and emphasized that these findings were similar to other reviews 
of the call records, including those done by Metabolife International 
and its consultants. The GAO testimony noted that, in those call 
records where age was documented, many of the serious adverse events 
occurred in relatively young consumers, with more than one-third being 
under the age of 30. Furthermore, for those call records in which 
quantity of use and/or frequency and duration of use were noted, most 
of the serious adverse events occurred among Metabolife 356 users who 
used the product within the recommended guidelines, i.e., they did not 
take more of the product nor consume it for a longer period of time 
than the product label recommended.

[[Page 6792]]

D. Petitions Received Relating to Dietary Supplement Containing 
Ephedrine Alkaloids

    We received three petitions relating to dietary supplements 
containing ephedrine alkaloids. The first petition, dated August 27, 
1998, was submitted by the American Obesity Association and requested 
that we issue a final rule on dietary supplements containing ephedrine 
alkaloids that adopts the regulations in the June 1997 proposal. The 
second petition, dated October 25, 2000, was filed jointly by the 
American Herbal Products Association, the Consumer Healthcare Products 
Association, the National Nutritional Foods Association, and the Utah 
Natural Products Alliance and requested that we withdraw the remaining 
portions of our June 1997 proposal and adopt and implement in its place 
an industry-developed standard for the labeling and marketing of 
dietary supplements containing ephedrine alkaloids.
    The third petition, dated September 5, 2001, was submitted by 
Public Citizen. This petition requested that we declare dietary 
supplements containing ephedrine alkaloids adulterated because they 
present a significant or unreasonable risk of illness or injury under 
section 402(f) of the act and ban, all production and sales of these 
products under section 301(a) (21 U.S.C. 331(a)) of the act. The 
petition also requested that we issue an advisory to stop the use of 
dietary supplements containing ephedrine alkaloids due to the 
established risks of injury.
    The information cited in support of this petition included:
     Summaries of the updated numbers and types of 
adverse events reported to us for ephedrine-alkaloid containing dietary 
supplements compared to the lower incidence of the same types of 
adverse events reported for all other dietary supplements;
     An FDA preliminary analysis of data collected by 
and purchased from the American Association of Poison Control Centers 
(AAPCC) that showed an increase in the number of ephedrine alkaloid-
related AERS from 211 in 1997 to 407 in 1999; and
     Adverse events reported to Public Citizen.
    The petition also cited the known pharmacological and toxicological 
properties of ephedrine alkaloids, recent published articles and case 
reports, the fact that adverse events are invariably underreported, and 
the lack of any evidence of long-term benefits for the products.
    We have considered the information submitted by these petitions, as 
well as the comments received in response to these petitions and all 
other information in the docket. For the reasons summarized in section 
I.A of this document, we have concluded that dietary supplements 
containing ephedrine alkaloids are adulterated.

II. Summary of Letters and Comments

    We have received more than 48,000 comments in three dockets 
pertaining to ephedrine alkaloids, Docket Nos. 1995N-0304, 2000N-1200, 
and 2001P-0396. These comments include all letters received prior to 
the June 1997 proposal, all comments received in response to Federal 
Register notices, and all submissions related to public meetings 
pertaining to dietary supplements containing ephedrine alkaloids. The 
48,000 comments include more than 41,000 form letters received in the 
1997 docket. Many comments submitted identical or nearly identical 
statements to more than one docket or in response to more than one 
Federal Register notice. Most of the comments were submitted by 
individual consumers who use dietary supplements containing ephedrine 
alkaloids or by independent distributors of these products. Other 
comments were received from persons who had, or who knew persons who 
had, suffered adverse events or who were reporting adverse events 
associated with the use of an ephedrine alkaloid-containing dietary 
supplement. The remaining comments included those submitted by medical 
professionals, scientists, medical or scientific associations, State or 
local health departments, Government agencies, members of Congress, 
dietary supplement manufacturers, traditional Asian medicine 
practitioners and associations, dietary supplement industry trade 
associations, public health associations, and consumer groups.
    The form letters, while not submitting substantive evidence or 
analyses, expressed strong views about our regulation of these 
products. Most of these letters opposed further federal regulation of 
dietary supplements containing ephedrine alkaloids. More than 13,000 
comments opposed a ban of these products and indicated that further 
restrictions on these products would infringe on personal choice. 
Thousands of comments requested that FDA not impose stricter 
regulations on dietary supplements containing ephedrine alkaloids than 
those imposed on OTC drugs that contain synthetic ephedrine alkaloids. 
Hundreds of comments requested that we not ban or reclassify ephedra as 
a prescription drug because, they claimed, such action would result in 
illegitimate profits for the pharmaceutical companies. Many expressed 
the view that we should only ban supplements containing excessive 
amounts of ephedrine alkaloids and those marketed to adolescents and 
children or to others who may abuse and misuse these products.
    Some form letters supported further regulation of these dietary 
supplement products. Several stated that dietary supplements containing 
ephedrine alkaloids are dangerous and asked us to ban them. Others 
requested that we impose more stringent requirements such as mandatory 
warning labels and maximum dosage levels. Thousands of form letters 
stated that DSHEA provides us with the necessary authority to protect 
the public health and that we do not need additional authority. 
Numerous comments criticized us for failing to exercise the enforcement 
powers authorized by DSHEA. Numerous form letters requested that 
ephedrine alkaloids be allowed for professional use by traditional 
Asian medicine practitioners and dispensed by licensed health care 
professionals.
    We have also received approximately 2,500 individual comments that, 
although not form letters, did not contain substantive information, 
analyses, or data. Many of these individual comments raised the same 
issues as raised in the form letters. Many comments were personal 
testimonials of how dietary supplements containing ephedrine alkaloids 
are effective for weight control, improving stamina, or treating 
medical conditions, and should not be banned or further restricted. 
Several comments stated that the June 1997 proposal lacked scientific 
basis and that there are many legitimate studies that support the 
responsible use of dietary supplements containing ephedrine alkaloids; 
however, these comments did not submit any additional scientific 
evidence. Others stated that dietary supplements containing ephedrine 
alkaloids are safe when used appropriately. Others were personal 
testimonials of adverse events related to these products that urged a 
ban or tighter restrictions of these products. Some comments criticized 
the proposed label warning as too long and ineffective.
    Other comments came from members of Congress, with many echoing the 
issues raised by the form letters. Several congressional 
representatives commented that Americans are increasingly turning to 
dietary supplements to improve their health and that Congress passed 
DSHEA to ensure that these products are regulated

[[Page 6793]]

as foods rather than drugs. They cited our own statements that DSHEA 
gives FDA sufficient authority to remove unsafe dietary supplements 
from the market. Many urged us to ensure that there was ample 
opportunity to submit scientific evidence related to dietary 
supplements containing ephedrine alkaloids. Many urged us to base our 
decisions on sound science and not rely too heavily on AERs. Some 
expressed concern about alleged FDA bias against dietary supplements 
containing ephedrine alkaloids. Others passed on concerns expressed by 
constituents about adverse health effects from these products. Several 
comments from members of Congress expressed concern about consumers' 
ability to read and properly use labels and warnings.
    Many of the substantive comments submitted data and other 
information regarding the use of ephedrine alkaloids. Some comments 
contained legal analyses of DSHEA and other provisions of the act. Many 
comments related to provisions of the June 1997 proposal that were 
withdrawn in 2000 or that have become moot as a result of the action 
taken in this final rule and, therefore, do not require a response. 
Examples of moot issues are the proposed prohibition on claims that 
encourage long-term use and the proposed label statement that the 
product should not be used for more than 7 days. Other comments 
addressed issues outside the scope of the rulemaking (e.g., comments 
about the diversion of ephedrine alkaloids for the illegal manufacture 
of methamphetamine and methcathinone) and will also not be addressed in 
this document.
    A summary of all relevant comments and our responses to those 
comments follow. To make it easier to identify comments and our 
responses, the word ``Comment,'' in parentheses, will appear before the 
comment summary and the word ``Response,'' in parentheses, will appear 
before our response. We have also numbered each comment summary to help 
distinguish between different comment summaries. The number assigned to 
each comment summary is purely for organizational purposes and does not 
signify the comments' value or importance or the order in which they 
were received.

III. Finding of Adulteration

A. What Does the Final Rule Do?

    This final rule declares dietary supplements containing ephedrine 
alkaloids to be adulterated under section 402(f)(1)(A) of the act. We 
have determined that these products present an unreasonable risk of 
illness or injury under the conditions of use recommended or suggested 
in labeling or, if no conditions of use are suggested or recommended in 
labeling, under ordinary conditions of use. We are taking this action 
based upon the well-known and scientifically established pharmacology 
of ephedrine alkaloids, the peer-reviewed scientific literature about 
the effects of ephedrine alkaloids, published case reports of adverse 
events, and the adverse events reported to us that have occurred in 
individuals using products containing ephedrine alkaloids, particularly 
dietary supplements. We have concluded that dietary supplements 
containing ephedrine alkaloids pose a risk of serious adverse events, 
including heart attack, stroke, and death, and that these risks are 
unreasonable in light of any benefits that may result from the use of 
these products under their labeled conditions of use, or under ordinary 
conditions of use if the labeling is silent. We are not addressing the 
issue of whether these products present a ``significant'' risk under 
section 402(f)(1)(A) of the act.

B. What Products are Covered?

    This final rule applies to dietary supplements containing ephedrine 
alkaloids, including, but not limited to, those from the botanical 
species Ephedra sinica Stapf, Ephedra equisetina Bunge, Ephedra 
intermedia var. tibetica Stapf, Ephedra distachya L., Sida cordifolia 
L. and Pinellia ternata (Thunb.) Makino or their extracts. The 
ingredient sources of the ephedrine alkaloids include raw botanicals 
and extracts from botanical sources. Although synthetic ephedrine (in 
the form of ephedrine hydrochloride) has been found in products labeled 
as dietary supplements, ephedrine hydrochloride was approved for use as 
a human drug as early as the late 1940s and, to the best of our 
knowledge there is no evidence that it was marketed prior to that time 
as a dietary supplement or food. Furthermore, ephedrine hydrochloride 
and other synthetic sources of ephedrine cannot be dietary ingredients 
because they are not constituents or extracts of a botanical, nor do 
they qualify as any other type of dietary ingredient. For these 
reasons, products containing synthetic ephedrine cannot be legally 
marketed as dietary supplements (See section 201(ff)(1) and 
201(ff)(3)(B) of the act (21 U.S.C. 321(ff)(1) and (ff)(3)(B))). In 
October 2001, we brought a seizure action against $2.8 million worth of 
finished drug products containing synthetic ephedrine hydrochloride 
that were labeled as dietary supplements (United States v. 1009 Cases * 
* * E'ola International AMP II), No. 2:01CV-820C (D. Utah filed October 
22, 2001)). As a result of this seizure, in 2002, the manufacturer 
signed a consent decree agreeing to the condemnation and destruction of 
the seized products and prohibiting it from manufacturing or 
distributing violative ephedrine hydrochloride products. In other 
actions, we have sent warning letters to multiple firms that were 
marketing products containing synthetic ephedrine alkaloids as dietary 
supplements, resulting in the removal of the illegal products from the 
market.
    The final rule does not apply to conventional food products that 
contain ephedrine alkaloids. Substances intentionally added to a 
conventional food are generally considered to be food additives under 
section 201(s) of the act. Ephedrine alkaloids contained in 
conventional foods would generally be considered unsafe food additives 
(see section 409 of the act (21 U.S.C. 348)). A food that contains an 
unsafe food additive is adulterated under section 402(a)(2)(C) of the 
act.
    This final rule also does not include OTC or prescription drugs 
that contain ephedrine alkaloids. The use of ephedrine or 
pseudoephedrine for the treatment of asthma, colds, allergies, or any 
other disease is beyond the scope of this final rule. Ephedrine is 
allowed as an active ingredient in oral OTC bronchodilator drugs for 
use in the treatment of medically diagnosed mild asthma (Sec. 341.16 
(21 CFR 341.16)), when used within the established dosage limits and 
when the product is labeled in accordance with the required statements 
of identity, indications, warnings, and directions for use found in 
Sec. 341.76. In the near future, we intend to propose revisions to Sec. 
341.76 to reflect current scientific information about the risks of 
ephedrine. Both ephedrine (topical) and pseudoephedrine (oral) are 
permitted as active ingredients for use as nasal decongestants (Sec. 
341.20), when they are used within the dosage limits established by and 
labeled in accordance with Sec. 341.80. The topical use of ephedrine 
will not be further discussed in this rule because it is not relevant 
to oral consumption of ephedrine in dietary supplements. The use of 
ephedrine alkaloids in drug products is discussed in more detail in 
section V.B.3 of this document.
    Several Ephedra species (including those known as ma huang) have a 
long history of use in traditional Asian medicine. These products are 
beyond the scope of this rule because they are

[[Page 6794]]

not marketed as dietary supplements. The use of ephedrine alkaloids in 
traditional Asian medicine is discussed in more detail in section V.B.5 
of this document. As we describe there, this rule does not change how 
these products are regulated under the act.
    (Comment 1) One comment stated that we coined the term ``ephedrine 
alkaloids'' to improperly broaden the scope of the published scientific 
literature and AERs cited in the June 1997 proposal. The comment 
pointed out that ephedrine, pseudoephedrine, and phenylpropanolamine 
(PPA) are all different chemical entities and stated the opinion that 
only data on ephedrine are relevant to the June 1997 proposal.
    (Response) Although we agree that the terms ephedrine, 
pseudoephedrine, and PPA refer to different chemical entities, we 
disagree with the rest of the comment and its conclusions. The term 
``ephedrine alkaloids'' refers to a class of naturally occurring 
compounds structurally related to ephedrine, and the term has been used 
in that manner in the scientific literature (Refs. 25 and 26). We chose 
this particular term, rather than several alternatives, such as 
``Ephedra bases'' and ``ephedrine type alkaloids,'' to limit the scope 
of the June 1997 proposal to those compounds that are natural 
constituents of the aerial parts of the Ephedra plant or other 
botanical sources of ephedrine and related alkaloids. We also defined 
the term by listing the six principal natural alkaloids in the June 
1997 proposal and other FDA documents (Refs. 6 and 27). The ephedrine 
alkaloids in botanicals include l-ephedrine, d-pseudoephedrine, l-
norephedrine, l-methylephedrine, d-norpseudoephedrine, d-
methylpseudoephedrine, and minor related alkaloids. All of these 
compounds are pharmacologically active substances in the plant. 
Therefore, we considered all of them in our evaluation of the risks 
associated with the use of the botanical or extracts from the 
botanical. However, as discussed in the response to comment 24 in 
section VI.B.1 of this document, we recognize that there are some 
differences between ephedrine and PPA.
    (Comment 2) Several comments asked whether North American species 
of Ephedra (e.g., Mormon Tea) are covered in this rulemaking.
    (Response) Most North American species of Ephedra (e.g., Mormon 
tea) do not contain ephedrine alkaloids (Refs. 2 and 26). Nonetheless, 
any dietary supplement that contains ephedrine alkaloids from any 
botanical source, including from a North American species of Ephedra, 
is subject to this rulemaking.

IV. Legal Issues

A. What Is Our Legal Authority Under the Act?

    We are issuing this final regulation under sections 402(f)(1)(A) 
and 701(a) of the act (21 U.S.C. 371(a)). Section 402(f)(1)(A) of the 
act deems a food to be adulterated for the following reasons:
    If it is a dietary supplement or contains a dietary ingredient 
that--
    (A) presents a significant or unreasonable risk of illness or 
injury under--
    (i) conditions of use recommended or suggested in labeling, or
    (ii) if no conditions of use are suggested or recommended in the 
labeling, under ordinary conditions of use.
    This regulation makes a finding that dietary supplements containing 
ephedrine alkaloids are adulterated because they present an 
unreasonable risk within the meaning of section 402(f)(1)(A) of the 
act. This finding is based on our conclusion that the risks of these 
products outweigh their benefits. Our legal interpretation of 
``unreasonable risk'' is discussed in detail in section V.D.1 of this 
document. This regulation does not address the meaning of ``significant 
risk'' or whether dietary supplements containing ephedrine alkaloids 
present a significant risk under section 402(f)(1(A) of the act.
    Section 701(a) of the act gives FDA authority to issue regulations 
for the efficient enforcement of the act. We are using this rulemaking 
authority for dietary supplements containing ephedrine alkaloids 
because we are articulating a standard for unreasonable risk under 
402(f)(1)(A) of the act for the first time and because it is more 
efficient to declare these products adulterated as a category than to 
remove them from the market in individual enforcement actions in which 
we would have to establish, for each individual product, that they 
present a significant or unreasonable risk.
    The March 2003 notice asked about the adequacy of FDA's authority 
to regulate dietary supplements containing ephedrine alkaloids. More 
specifically, we sought comments on ``what additional legislative 
authorities, if any, would be necessary or appropriate to enable us to 
address this issue most effectively'' (68 FR 10417 at 10420).
    (Comment 3) Many comments expressed the view that we already have 
the authority we need to take action against dietary supplements 
containing ephedrine alkaloids. These comments cited our authority to 
declare these supplement products to be a significant or unreasonable 
risk or imminent hazard under section 402(f)(1) of the act or to 
regulate the products as containing a poisonous or deleterious 
substance that may render them injurious to health under section 
402(a). The comments differed as to whether we had the necessary 
evidence to utilize these provisions. Several comments opposed any 
additional authority and criticized us for allegedly not fully 
implementing the authority we already have.
    (Response) We agree that we have the authority to take action 
against dietary supplements that contain ephedrine alkaloids. All three 
authorities mentioned by the comments are available to us when 
circumstances warrant. In this instance, we have chosen to proceed 
under the adulteration standard in section 402(f)(1)(A) of the act. We 
believe that we have sufficient evidence to meet this standard.
    (Comment 4) In contrast, other comments stated that our legal 
authority should be strengthened. Several comments expressed the view 
that DSHEA needs to be amended because it cannot adequately protect 
public health. One public interest group noted that our delay in acting 
reflects the difficulty we encounter implementing DSHEA. Several 
comments offered suggestions for amendments that would strengthen our 
legal authority, including mandatory reporting of adverse events, 
certain sales restrictions (e.g., restricting sales to behind the 
counter only, prohibiting sales to individuals under the age of 18), 
special labeling requirements for dietary supplements containing 
ephedrine alkaloids, registration and listing, premarket approval for 
safety and efficacy (particularly for all new stimulants and steroid 
substitutes), and repeal of the de novo review provision so that we 
would receive judicial deference on adulteration issues. A few comments 
suggested that dietary supplements be regulated as drugs. One comment 
suggested new legislation to classify dietary supplements according to 
a risk-based regulatory scheme.
    (Response) We must regulate dietary supplements under our existing 
authority. Accordingly, we are unable to take action regarding 
suggestions for amendments to DSHEA because any such amendments must 
result from congressional action rather than rulemaking. Therefore, we 
are not addressing those suggestions in this rule.
    (Comment 5) One comment stated that conventional food safety 
standards, i.e., the generally recognized as safe (GRAS) standard or 
the standard for

[[Page 6795]]

FDA approval as a food additive, do not apply to dietary ingredients.
    (Response) We agree that the standards referred to in this comment 
do not apply to dietary ingredients. Premarket approval is required of 
substances that are food additives as defined in section 201(s) of the 
act. Substances that would otherwise fall under the food additive 
definition but are generally recognized as safe by experts are not food 
additives and do not require premarket approval. Dietary ingredients 
contained in, or intended for use in, a dietary supplement are 
explicitly excluded from the food additive definition in section 
201(s)(6) of the act. Therefore, neither the premarket approval regime 
for food additives nor the GRAS standard applies to dietary 
ingredients. We are instead basing this final rule on the dietary 
supplement adulteration standard set forth in section 402(f)(1)(A) of 
the act.
    (Comment 6) One comment stated we are violating the First Amendment 
of the U.S. Constitution and the Administrative Procedure Act (APA) by 
requiring a much higher standard of safety for dietary supplements than 
for conventional foods. Another comment also raised concerns about the 
First Amendment limits of FDA's authority to regulate dietary 
supplements containing ephedrine alkaloids.
    (Response) We disagree with these comments. There are a number of 
different safety standards for foods (see, e.g., section 402(a)(1) and 
section 402(a)(2)(C) of the act), and whether these standards are 
higher or lower than the ``significant or unreasonable risk'' standard 
for dietary supplements in section 402(f)(1)(A) of the act is not 
relevant to the legal sufficiency of this rule. To the extent that we 
regulate dietary supplements and conventional foods differently, these 
differences are justified by the differences in the statutory 
provisions that apply to these two categories of products. Although 
some parts of the act apply to both dietary supplements and 
conventional foods, other provisions apply only to one or the other. 
Where Congress expressly provided for dietary supplements to be subject 
to a requirement or standard that does not apply to conventional foods, 
we may implement that provision without violating the APA. Further, 
this final rule does not violate the First Amendment. This rule does 
not restrict speech; rather, it makes a finding of adulteration that 
results in a prohibition on the distribution and sale of a product that 
presents unreasonable health risks. Such restrictions on purely 
commercial, nonexpressive conduct are not subject to First Amendment 
scrutiny. See, e.g., United States v. O'Brien, 391 U.S. 367, 376 
(1968).
    (Comment 7) Several comments expressed the view that these products 
should be regulated as drugs under our existing authority. Some 
comments stated that we should make these products available only by 
prescription, arguing that the potential health hazards associated with 
dietary supplements containing ephedrine alkaloids are too serious for 
OTC use and that restricting access by requiring a prescription would 
insert trained medical professionals into a case-by-case decision on 
the appropriateness of these products to an individual consumer. 
Further, one comment recommended that if the frequency of adverse 
events under prescription status does not improve, more restrictive 
action should be implemented, including the withdrawal of all products 
containing ephedrine alkaloids from the market.
    (Response) We do not agree that all dietary supplements containing 
ephedrine alkaloids may be regulated as drugs under our existing 
authority. Products are drugs only if they meet the definition of drug 
in section 201(g)(1) of the act. Products containing ephedrine 
alkaloids are regulated as drugs if they are intended to be used in the 
diagnosis, cure, mitigation, treatment, or prevention of disease 
(section 201(g)(1)(B) of the act). Without evidence of intended use for 
such purposes, the product is not a drug under the act. Some dietary 
supplements containing ephedrine alkaloids are promoted for disease 
uses, e.g., to treat obesity. In such instances, we can and have taken 
action against certain dietary supplement products as drugs. Under the 
act, considerations such as potential risks to health, need for medical 
supervision, and pharmacology of a product that meets the dietary 
supplement definition are not by themselves sufficient to subject the 
product to regulation as a drug.
    To the extent that comments suggest that these products could 
somehow remain dietary supplements but be available only by 
prescription, we note that we do not have authority to take such 
action. The act gives us the authority to restrict drugs and devices to 
prescription use; it does not give us the authority to restrict dietary 
supplements to prescription use.
    (Comment 8) One comment stated that the generally accepted 
definition of safety for a drug, i.e., a low incidence of adverse 
reactions or significant side effects under appropriate conditions of 
use, and a low potential for harm, which might result from abuse 
situations, is equally applicable to dietary supplements or food.
    (Response) We do not agree that the safety standards for drugs 
apply to dietary supplements or other foods. As explained previously, 
dietary supplements are not drugs unless they meet the definition of 
drug in section 201(g)(1) of the act. The same is true for conventional 
foods. We are basing this final rule on the dietary supplement 
adulteration standard set forth in section 402(f)(1)(A) of the act. The 
adulteration standard for dietary supplements set forth in section 
402(f)(1)(A) of the act implies a risk-benefit calculus. While we also 
use a risk-benefit evaluation in the drug evaluation process (see Sec. 
312.21(c), Sec. 314.50(c)(5)(viii), and Sec. 330.10(a)(4) (21 CFR 
312.21(c), 314.50(c)(5)(viii), and 330.10(a)(4))), the act creates 
different evidentiary standards for dietary supplements and drugs. 
Therefore, we are not applying the drug safety standard to dietary 
supplements.

B. Do the Ephedrine Alkaloid-Containing Products Covered by this Rule 
Fall Within the Definition of Dietary Supplement Under the Act?

    A threshold issue is whether the products covered by this rule meet 
the definition of a dietary supplement under section 201(ff) of the 
act.
    (Comment 9) One comment from a State department of health stated 
the opinion that dietary supplements containing ephedrine alkaloids 
present significant risks when they are consumed as a regular part of 
the diet and do not fall within section 201(ff)(1) of the act. The 
comment explained that because these products cannot be used on a daily 
basis without presenting significant risks they cannot be ``intended to 
supplement the diet'' and are not dietary supplements within the 
meaning of the act. A related comment expressed the opinion that, for a 
substance to be a dietary supplement, it must be proven that the human 
body needs the substance to establish a need for supplementation.
    (Response) We agree with these comments in part and disagree in 
part. We agree that dietary supplements containing ephedrine alkaloids 
present a risk when consumed as a regular part of the diet; as 
discussed in section V.B of this document, they present a risk to some 
users even when consumed occasionally. We do not agree, however, that 
dietary supplements containing botanical ephedrine alkaloids do not 
fall within the definition of a dietary supplement in section 201(ff) 
of the act. Section 201(ff)(1) of the act, added by

[[Page 6796]]

DSHEA, provides, in part, that the term ``dietary supplement'' means a 
product ``intended to supplement the diet'' that bears or contains one 
or more dietary ingredients. Among the dietary ingredients listed in 
section 201(ff)(1) of the act are herbs and other botanicals. 
Therefore, botanical sources of ephedrine alkaloids, such as Ephedra 
sinica Stapf and the other botanicals described in section III.B. of 
this document, are dietary ingredients. Further, we do not agree that 
the phrase ``intended to supplement the diet'' authorizes the exclusion 
of a product from the dietary supplement definition solely on the basis 
of risk. Given the explicit references to risk in section 402 of the 
act and the inclusion of botanicals as a category of dietary 
ingredients in section 201(ff)(1) of the act, it seems clear that 
Congress intended us to regulate botanical products as dietary 
supplements (provided that they are not drugs and otherwise meet the 
dietary supplement definition) and to evaluate their risks under the 
adulteration provisions in section 402 of the act.
    We also do not agree that, under the dietary supplement definition, 
it must be proven that the human body needs a particular substance to 
establish a need for supplementation. Under DSHEA, a substance does not 
necessarily have to be shown to be essential to human nutrition to be 
marketed as a dietary supplement. Although no provision in the act or 
legislative history directly addresses this issue, section 201(ff) of 
the act lists classes of dietary ingredients (e.g., botanicals) that 
are not essential for growth or to maintain good health (Ref. 28). The 
fact that Congress classified such substances as dietary ingredients is 
clear evidence that Congress did not intend to limit dietary 
ingredients to substances that have been deemed to be essential in 
human nutrition.
    (Comment 10) Several comments, including one from an industry 
medical consultant, stated that herbal products should not be regulated 
under DSHEA because they have physiologic effects and significant 
potential for toxicity. The comment encouraged us to work with industry 
to establish an appropriate regulatory category for botanicals.
    (Response) Under the act (as amended by DSHEA), botanicals can be 
marketed as dietary supplements provided that they otherwise meet the 
dietary supplement definition, and are safe and properly labeled. If 
botanicals meet the drug definition in section 201(g) of the act, they 
are properly regulated as drugs. In this regard, we published a final 
rule entitled ``Additional Criteria and Procedures for Classifying 
Over-the-Counter Drugs as Generally Recognized as Safe and Effective 
and Not Misbranded'' (67 FR 3060, January 23, 2002). This rule defines 
the term ``botanical drug substance'' and explains how to submit a time 
and extent application to request that a botanical drug substance be 
included in an OTC drug monograph (see Sec. 330.14). In addition, we 
recognize, and are addressing, the current need for guidance for 
manufacturers seeking to develop botanicals as either OTC or 
prescription drug products under the applicable statutory and 
regulatory requirements. (See Guidance for Industry: Botanical Drug 
Products (Draft Guidance) (August 2000) (available at http://
www.fda.gov/cder/guidance/1221dft.pdf).)

C. Administrative Procedures

    (Comment 11) Several comments stated that it is premature to 
request comments on whether dietary supplements containing ephedrine 
alkaloids present a significant or unreasonable risk before we define 
that standard. These comments urged us to undertake a rulemaking, or a 
guidance document, on this new standard so that it can be applied in 
the future to all dietary supplements posing health concerns. One 
comment suggested that defining ``significant or unreasonable risk'' 
may require new legislation.
    (Response) We do not agree that we must define the term 
``unreasonable risk'' standard through regulation or guidance before 
taking action against dietary supplements containing ephedrine 
alkaloids based upon this standard. An agency may interpret a statutory 
provision through rulemaking or case-by-case adjudication (SEC v. 
Chenery, 332 U.S. 194 (1947)). We conclude, based upon available 
evidence discussed in section V of this document, that dietary 
supplements containing ephedrine alkaloids present an unreasonable risk 
of illness or injury because their risks outweigh their benefits, and 
that these products are therefore adulterated under section 
402(f)(1)(A) of the act. We are using our general rulemaking authority 
to issue regulations for the efficient enforcement of the act (section 
701(a) of the act) to issue a regulation applying the standard in the 
context of a particular category of dietary supplements--those that 
contain botanical ephedrine alkaloids. We are not required to issue a 
separate rule or guidance defining the 402(f)(1)(A) standard before 
issuing such a regulation. Similarly, lack of a regulation or guidance 
defining the standard neither prevents us from taking enforcement 
action against dietary supplements that present an ``unreasonable 
risk,'' nor is it new legislation necessary for us to interpret the 
meaning of ``unreasonable risk.'' If Congress has clearly spoken to a 
question of statutory interpretation, the agency charged with 
administering the statute must implement the unambiguous intent of 
Congress (``Chevron step one'') (Chevron U.S.A., Inc. v. Natural 
Resource Defense Council, 467 U.S. 837, 842-843 (1984)). If a statute 
is silent or ambiguous on the question, however, the agency may 
interpret the ambiguous provision (``Chevron step two'') Id. at 843-
844. When such administrative interpretations are made through 
rulemaking, they will be upheld as long as they are reasonable and 
consistent with the statute's purpose and legislative history 
(Christensen v. Harris County, 529 U.S. 576, 587 (2000); Chevron 
U.S.A., Inc. v. FERC, 193 F.Supp.2d 54, 68 (D.D.C. 2002)). As discussed 
in the response to comment 59 in section V.D.1 of this document, we 
have concluded under Chevron step one that the phrase ``unreasonable 
risk'' clearly directs FDA to conduct a risk-benefit analysis. Even if 
a court were to find that phrase ambiguous, however, our interpretation 
is reasonable under Chevron step two.
    (Comment 12) Several comments urged us not to act against all 
dietary supplements containing ephedrine alkaloids because all such 
products are different and must be considered individually. The 
comments cited differences in dosages, formulations, labeling, etc., 
across products and, thus, each product must be analyzed on its own 
merits. One industry comment argued that we exceeded our statutory 
authority in trying to regulate all dietary supplements containing 
ephedrine alkaloids through notice and comment rulemaking.
    (Response) We do not agree that we may not regulate the entire 
category of dietary supplements containing ephedrine alkaloids through 
rulemaking. We recognize that there are differences between different 
dietary supplements containing ephedrine alkaloids. However, we 
conclude, based on available science, that all dietary supplements 
containing ephedrine alkaloids present an unreasonable risk of illness 
or injury, regardless of how they are formulated or labeled, because 
the risks outweigh any benefits that may result from use of the 
products. Therefore, we may issue a rule finding the entire class of 
products adulterated.
    (Comment 13) A few comments noted that we bear the burden of proof 
to show

[[Page 6797]]

dietary supplements are adulterated under section 402(f)(1) of the act.
    (Response) We agree with this comment. Section 402(f)(1) of the act 
clearly states that in any proceeding under that provision, ``the 
United States shall bear the burden on each element to show that a 
dietary supplement is adulterated.'' We have met that burden in this 
rulemaking.
    (Comment 14) Several comments discussed our ability to declare 
dietary supplements containing ephedrine alkaloids an imminent hazard 
under section 402(f)(1)(C) of the act.
    (Response) We are not addressing these comments because we have 
chosen to proceed under section 402(f)(1)(A).
    (Comment 15) One industry comment stressed that comments to the 
June 1997 proposal may not be used to authorize other final 
regulations. The comment expressed concern that comments to a proposed 
warning statement would be used as a basis for another FDA action to 
regulate these supplements.
    (Response) We disagree with this comment. FDA may issue this final 
regulation based on a finding that dietary supplements containing 
ephedrine alkaloids are adulterated because they present an 
unreasonable risk under section 402(f)(1)(A) of the act. APA requires 
agencies to provide the public with notice and an opportunity for 
comment before issuing a new regulation (5 U.S.C. 553(b) and (c)). In 
keeping with this requirement, a final rule may differ from a proposed 
rule if the final rule is a ``logical outgrowth'' of a proposed rule 
(Small Refiner Lead Phase-Down Task Force v. EPA, 705 F.2d 506, 547 
(D.C. Cir. 1983)). The inquiry into whether a final rule is a logical 
outgrowth of the proposed rule is often stated as whether the regulated 
party ``should have anticipated that such a requirement might be 
imposed'' (Small Refiner, 705 F.2d at 549). Agencies ``undoubtedly have 
authority to promulgate a final rule that differs in some particulars 
from its proposed rule* * * `[a] contrary rule would lead to the 
absurdity that * * * the agency can learn from the comments on its 
proposals only at the peril of starting a new procedural round of 
commentary''' (Small Refiner, 705 F.2d at 546-547 (quoting 
International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 632 n.51 
(D.C. Cir.1973))). The D.C. Circuit has also stated: ``The APA notice 
requirement is satisfied if the notice fairly apprises interested 
person of the subjects and issues the agency is considering; `the 
notice need not specifically identify ``every precise proposal which 
[the agency] may adopt as a final rule''' (Chemical Manufacturers 
Association Waste Mfrs. v. EPA, 870 F.2d 177, 203 (5th Cir. 1989) 
(quoting United Steelworkers of Am. v. Schuylkill Metals, 828 F.2d 314, 
317 (5th Cir. 1987) (internal citations omitted))).
    Our June 1997 proposal, along with our March 5, 2003 Federal 
Register notice, provided a sufficient basis to allow the public to 
anticipate our actions in this final rule. Through our proposed actions 
on dietary supplements containing ephedrine alkaloids, the public was 
properly notified of the possibility that we would find such products 
to be adulterated under section 402(f)(1)(A) of the act. In fact, our 
March 2003 notice specifically asked for comment on whether dietary 
supplements containing ephedrine alkaloids present a significant or 
unreasonable risk under section 402(f)(1)(A) of the act. We also sought 
comment on new evidence concerning the safety of dietary supplements 
containing ephedrine alkaloids (68 FR 10417 at 10420). In addition, the 
restriction on ephedrine alkaloid/stimulant combinations proposed in 
1997, which was unaffected by the 2000 partial withdrawal proposal, was 
based in part on a finding of adulteration under section 402(f)(1)(A) 
of the act (62 FR 30678 at 30696). Though we did not specifically 
propose to codify a finding of adulteration based on significant or 
unreasonable risk in the March 2003 notice, it was clear that we were 
contemplating the possibility that dietary supplements containing 
ephedrine alkaloids were adulterated under section 402(f)(1)(A) of the 
act. Courts have upheld final rules that contained new elements when 
the public was made aware that the agency was contemplating such a 
change (See Chem. Mfrs. Ass'n. , 870 F.2d 202-203). Furthermore, we 
received several comments regarding the possibility of a finding that 
all dietary supplements containing ephedrine alkaloids would be deemed 
adulterated under section 402(f)(1)(A) of the act. Though not 
determinative of logical outgrowth in and of themselves, comments on 
the issue are evidence that the public received adequate notice of our 
final rule (Shell Oil Co. v. EPA, 950 F.2d 741, 757 (D.C. Cir. 1991)). 
Based upon our explicit request for comments on the adulteration issue 
in our March 2003 notice, our reference to the section 402(f)(1)(A) of 
the act adulteration standard as a basis for our June 1997 proposal, 
and the fact that a number of parties commented on whether dietary 
supplements containing ephedrine alkaloids present a significant or 
unreasonable risk, there was adequate notice to the public of our 
actions in this final rule.
    (Comment 16) Several comments cited language in section 402(f)(1) 
of the act providing that courts must review any determination under 
section 402(f)(1) of the act de novo and further stated that we would 
not get judicial deference in any court review. The comments argued 
that, under this provision, it would make no difference whether we 
brought our case initially in court or whether we proceeded through 
rulemaking that was subsequently challenged in court. One trade 
association noted that such de novo review is a novel approach in that 
usually a court would just review the administrative record.
    (Response) Section 402(f)(1) of the act states that a court will 
decide any issue under that paragraph on a de novo basis. We agree that 
the de novo standard of review applies to our factual findings under 
section 402(f)(1) of the act, but do not agree that it applies to our 
conclusion under Chevron U.S.A., Inc., that ``unreasonable risk'' means 
a risk-benefit analysis (see section V.D.1 of this document). This 
interpretation of the de novo provision of section 402(f)(1) of the act 
is consistent with case law on the Toxic Substances Control Act (TSCA), 
which contains an unreasonable risk standard coupled with a 
``substantial evidence'' standard of review, analogous to the act's 
unreasonable risk standard coupled with a de novo standard of review. 
In Chem. Mfrs. Ass'n v. EPA, 859 F.2d 977 (D.C. Cir. 1988), the D.C. 
Circuit distinguished EPA's legal interpretation of unreasonable risk, 
which received deference under Chevron U.S.A., Inc. v. Natural 
Resources Defense Council, 467 U.S. 837 (1984), from its burden of 
showing with ``substantial evidence'' in the record that it has met the 
standard. The court stated: ``This fairly rigorous standard of record 
review should not * * * be confused with the substantive statutory 
standard * * * '' (859 F.2d at 992). Thus, the court in Chem. Mfrs. 
Ass'n. held that the ``substantial evidence'' standard of record review 
applied to the factual basis of EPA's decision but not to its 
interpretation of the statutory standard. In applying Chevron U.S.A., 
Inc., we have concluded that Congress unambiguously intended that 
unreasonable risk entails a risk-benefit calculus. If a court were to 
find the phrase ``unreasonable risk'' ambiguous, however, our 
interpretation of unreasonable risk as meaning a risk-benefit calculus 
should receive Chevron U.S.A., Inc. deference, like EPA's

[[Page 6798]]

interpretation of the statutory standard in Chem. Mfrs. Ass'n.. The 
requirement for de novo review should be applied only to the factual 
basis of FDA's determination.
    Regardless of which standard applies, however, our determination 
that dietary supplements containing ephedrine alkaloids present an 
unreasonable risk under section 402(f)(1)(A) of the act should be 
sustained by a court. Our conclusion that ``unreasonable risk'' entails 
a risk-benefit analysis is consistent with the express intent of 
Congress. The scientific evidence regarding the pharmacology of 
products containing ephedrine alkaloids, clinical studies showing that 
these products raise blood pressure, published case reports, and AERs, 
when compared with the evidence regarding the very modest benefits 
conferred by these supplements, forms a strong factual basis for 
finding that the known and reasonably likely risks of dietary 
supplements containing ephedrine alkaloids outweigh the known and 
reasonably likely benefits of these products. Therefore, dietary 
supplements containing ephedrine alkaloids present an unreasonable risk 
of injury or illness under section 402(f)(1)(A) of the act.
    (Comment 17) One comment submitted by a trade association noted 
that, before requesting the Department of Justice to take any civil 
action against dietary supplements containing ephedrine alkaloids, we 
must give appropriate notice and opportunity to present oral and 
written arguments at least 10 days prior to the request.
    (Response) We agree with this comment in part and disagree in part. 
Section 402(f)(2) of the act provides that ``the person against whom 
such proceeding would be initiated'' must be given notice and the 
opportunity to present views, orally and in writing, 10 days before we 
report a violation of section 402(f)(1)(A) of the act (the 
``significant or unreasonable risk'' provision) to the Department of 
Justice for a civil proceeding. By the plain language of this 
provision, it applies to proceedings against persons, not to 
proceedings against products. Thus, the requirement applies to 
injunction actions, which are brought against a corporate or individual 
person, but not to seizures, which are brought against a product. 
Therefore, if we were to refer a seizure of dietary supplements 
containing ephedrine alkaloids to the Department of Justice, the notice 
requirement would not apply. We further note that the current 
proceeding is a rulemaking, not a civil action being referred to the 
Department of Justice, and therefore the 10-day notice requirement does 
not apply.
    (Comment 18) One industry comment stated that the stringent 30-day 
timeframe allowed for comments in response to the March 2003 notice did 
not provide the industry with a fair opportunity to review the 
administrative record and fairly respond to ``any alleged new evidence 
and analyses'' by FDA. This comment urged us to allow for a comment 
period of 180 days. The comment stated that this procedural lapse would 
render the entire rulemaking process arbitrary and capricious.
    (Response) We disagree with this comment. We believe that the 30-
day comment period on the March 2003 notice provided interested persons 
with an adequate opportunity for review and comment. The information 
placed in the public docket at that time was limited, consisting of the 
RAND report plus six recent studies. APA requires only that an agency 
``give interested persons an opportunity to participate in the 
rulemaking through submission of written data, views, or arguments * * 
*'' This opportunity to participate is all that the APA requires. There 
is no statutory requirement concerning how many days we must allow for 
comment, nor is there a requirement that we extend the comment period 
at the request of an interested person (See Phillips Petroleum Co. v. 
EPA, 803 F.2d 545, 559 (10th Cir. 1986)). Moreover, given that we first 
opened a docket on the issue of dietary supplements containing 
ephedrine alkaloids in 1995 and sought comments on this issue several 
times between then and 2003 (see section I.C of this document), there 
has been ample opportunity for all those interested to submit 
information and views.

V. Scientific Evaluation

A. How Did We Evaluate the Evidence?

    To determine whether a dietary supplement presents an unreasonable 
risk of illness or injury, the agency performs a risk/benefit analysis 
to ascertain whether the risks of the product outweigh its benefits.
    The risks and benefits of a dietary supplement must be evaluated in 
light of the claims and directions for use in the product's labeling 
or, if the labeling is silent, under ordinary conditions of use 
(section 402(f)(1)(A) of the act). Labeling claims for dietary 
supplements must be substantiated. Unless the manufacturer has 
substantiation that a labeling claim promoting a dietary supplement for 
a purported benefit is truthful and non-misleading, the claim misbrands 
the product (See section 403(a)(1) and 403(r)(6) of the act. We note 
that the standards for substantiating the efficacy of a drug for a 
labeled indication (i.e., the generally recognized as effective (GRAE) 
standard for OTC monograph ingredients and the substantial evidence 
standard for new drugs) do not apply to dietary supplements.
    Substantiation of a benefit may not be necessary to lawfully market 
a dietary supplement if its labeling does not include a claim, and the 
product poses little or no risk. In weighing risks and benefits to 
determine whether dietary supplements containing ephedrine alkaloids 
present an unreasonable risk under section 402(f)(1)(A) of the act, we 
considered only known and reasonably likely benefits, not speculative 
benefits. A reasonably likely benefit is one that is supported by a 
meaningful totality of the evidence, given the current state of 
scientific knowledge, though the evidence need not necessarily meet the 
approval standard for a prescription drug.
    Although Congress placed the burden on FDA to show ``unreasonable 
risk,'' once a danger is identified, we do not believe that Congress 
intended us to delay action until double-blind, placebo-controlled 
clinical studies could be conducted or that no action be taken if such 
clinical studies are infeasible or unethical (see the response to 
comment 19 of this document). While such studies are the ``gold 
standard'' for determining effectiveness, they are not always available 
for dietary supplements because DSHEA does not require companies to 
conduct such studies before marketing a dietary supplement. DSHEA also 
does not require postmarketing safety and adverse event reporting from 
dietary supplement manufacturers. Accordingly, FDA is relying on the 
available scientific data and literature to support its conclusion that 
dietary supplements containing ephedrine alkaloids present an 
``unreasonable risk.'' The government's burden of proof for 
``unreasonable risk'' can be met with any science-based evidence of 
risk and does not require a showing that the substance has actually 
caused harm in particular cases.
    For example, there is clear scientific evidence that a sustained 
increase in blood pressure increases the risks of cardiovascular 
disease (Refs. 29, 29a, and 30). Thus, a dietary supplement that caused 
a sustained rise in blood pressure across the population would increase 
the risk of cardiovascular events including stroke, heart attack, or 
death to that population. Even risks that

[[Page 6799]]

may not be detectable in small studies or studies of short duration 
(which are not designed to detect such risks at a statistically 
significant level) could, over time, and on a population-wide basis, 
result in thousands of adverse health events.
    In making a determination, we consider studies using closely 
related products. In considering the risks of a product, such as 
dietary supplements containing ephedrine alkaloids, it is appropriate 
to consider the safety of closely related products, such as those with 
the same active ingredient (e.g., synthetic ephedrine products) or 
closely related ingredients (such as other sympathomimetics) because we 
would expect that dietary supplements containing ephedrine alkaloids 
will exhibit pharmacological effects similar to those other products 
and, therefore, pose similar risks. It is more difficult to extrapolate 
conclusions regarding the benefits between an ephedrine drug product 
and a dietary supplement containing ephedrine alkaloids since the 
ephedrine drug product is a well defined product with a known dose of 
ephedrine, while in the latter there is a complex mixture with, 
possibly, an unknown quantity of ephedrine plus other ephedrine 
alkaloids, and sometimes other active ingredients, many of which may 
not be fully characterized. We would need to know how the two products 
compare with regard to systemic delivery of ephedrine (e.g., the 
pharmacokinetics profile) to make any judgments about comparable 
benefits of the two products. If ephedrine pharmacokinetics were the 
same in a synthetic and plant-derived product and there were no 
ingredients or components other than ephedrine, one might conclude that 
the plant-derived and synthetic products would behave similarly. In 
actual fact, that is not the case because plant derived ephedra 
products contain other ephedrine alkaloids in addition to ephedrine 
itself (e.g. pseudoephedrine, methylephedrine, and others listed in 
section I.B of this document). Moreover, if there were other active and 
inactive ingredients in the plant-derived product, their properties 
would need to be explored.
    In evaluating whether dietary supplements containing ephedrine 
alkaloids present an unreasonable risk, we looked at the seriousness of 
the risks and the quality and persuasiveness of the totality of the 
evidence to support the presence of those risks. We then weighed the 
risks against the importance of the benefits and the quality and 
persuasiveness of the totality of the evidence to support the existence 
of those benefits. We give more weight to benefits that improve health 
outcomes, especially in the long term, than to benefits that are 
temporary or rely on subjective measures such as feeling or looking 
better. For example, sustained, long-term weight loss in an obese or 
overweight person is a much more important benefit than short-term 
weight loss because long-term weight loss in these individuals reduces 
the risk of serious morbidity and mortality (e.g., heart attacks and 
strokes), while short-term weight loss does not.
    In sections V.B, C, and D of this document, we describe the 
evidence FDA evaluated to reach its determination that dietary 
supplements containing ephedrine alkaloids present an unreasonable risk 
of illness or injury.
    (Comment 19) Many comments stated that any assessment of 
unreasonable risk must be based on sound science. Several comments 
stated that a conclusion about the safety and efficacy of dietary 
supplements containing ephedrine alkaloids is premature and that 
additional prospective or retrospective case controlled studies are 
needed to determine causality. A few comments recommended that FDA, 
NIH, or other parts of the federal government conduct such research to 
address unresolved issues of causation. Another trade association urged 
the government to collaborate with industry to design future controlled 
studies. Several of these comments cited RAND in support of the need 
for further research. Several comments noted that the National Center 
for Complementary and Alternative Medicine/NIH Working Group evaluated 
the RAND report and suggested a multi-site case-control study to assess 
the risks associated with these products, although it stated that such 
a study would take 4 to 8 years and cost $2 to $4 million per year 
(Ref. 31).
    In contrast, several comments asserted that conducting clinical 
trials of ephedrine alkaloids would be unethical in light of the risks 
to the human subjects. A professional association stated that FDA 
regulations that govern drug development and approval would not allow 
such research, given the absence of information to suggest a benefit 
that would outweigh the risks. A few comments suggested that any study 
that could be approved by a human subjects committee would be required 
to exclude patients at risk and therefore, would not be useful in 
evaluating risk when the products are taken by the general population 
without medical supervision. Other comments expressed concern that the 
additional research recommended by RAND would delay efforts or render 
it virtually impossible to safeguard public health.
    (Response) We recognize the value of properly conducted clinical 
trials to answer questions regarding the safety and effectiveness of 
FDA-regulated products. It is not clear, however, that clinical trials 
to evaluate the adverse effects of ephedrine alkaloids can be 
conducted. It would not be ethical to study the arrhythmogenic 
potential of ephedrine alkaloids in patients with coronary artery 
disease, the adverse effects of ephedrine alkaloids in people with 
heart failure, or the consequences of raising blood pressure in various 
populations. Moreover, there is now sufficient evidence, generated 
through multiple sources, including clinical trials, published 
literature, and other information, to reach the conclusion that dietary 
supplements containing ephedrine alkaloids have effects on blood 
pressure and other pharmacological risks that predict adverse effects 
in users. After considering the best available information, we conclude 
that these products present an unreasonable risk because the benefits 
that may result from use of these products are outweighed by the risks 
associated with such use (see discussion in section V.D of this 
document). Because of the nature of these risks, we do not believe it 
is appropriate to delay action until further clinical studies can be 
conducted to evaluate the safety of dietary supplements containing 
ephedrine alkaloids in the general population. We would, however, 
support the conduct of clinical investigations (carried out under the 
Investigational New Drug (IND) regulations with careful screening to 
exclude subjects at risk and careful safety monitoring during the 
trials) that examine the safety and efficacy of ephedrine alkaloids, 
with or without caffeine, as drugs such as for the treatment of obesity 
(see 21 CFR part 312).
    (Comment 20) Two comments stated that there is an accepted 
scientific methodology for determining whether, and at what level, a 
food additive, dietary ingredient, OTC or prescription drug, or 
biologic may be hazardous to human health. The stated components of 
this methodology include reviews of the following reports: (1) The 
existing scientific literature on the substance, to determine what is 
known about the substance's risk, particularly at the levels to be used 
in a product; (2) clinical studies involving the substance; (3) 
available animal studies on the substance and, if necessary, the 
conduct of additional studies; and (4) adverse event reports caused by 
the substance.

[[Page 6800]]

 In addition, the methodology includes a determination of whether 
individuals who consume the products suffer from a statistically 
significantly greater number of adverse (or beneficial) events than 
those who do not. One comment stated that the absence of premarket 
approval authority for dietary supplements does not preclude reliance 
on traditional methods of evaluating safety when making a decision 
about levels that are not safe.
    (Response) We do not agree with the comments stating that there is 
a single accepted method of evaluation to determine when a food 
ingredient or dietary ingredient in a dietary supplement presents a 
hazard to the public health. In any evaluation of the risks presented 
by a substance in a product in the marketplace, the method of 
evaluating the risk must be applied on a case-by-case basis that is 
based on the available data concerning the substance being evaluated. 
We believe that our method of evaluation for ephedrine alkaloids is, 
however, consistent with that used for other substances. The scientific 
methodology we used to evaluate the risks associated with the use of 
dietary supplements containing ephedrine alkaloids consisted of a 
review and evaluation of the available scientific literature (including 
literature on pharmacology), clinical studies, published case reports, 
and other data, including adverse event reports. This is the same type 
of scientific methodology that is applied in the evaluation of adverse 
effects associated with other FDA-regulated products (Ref. 32), and 
includes most of the steps listed in the comments summarized above.
    (Comment 21) A number of comments focused on FDA's obligation to 
ensure that its regulatory assessments are science-based. Two comments 
raised concern regarding our compliance with a statutory provision 
popularly known as the Data Quality Act (section 515 of the 
Consolidated Appropriations Act, 2001, Public Law 106-554, 44 U.S.C.A. 
3516 note). One comment stated that we are vulnerable to challenge 
under the Data Quality Act because there is a disconnect between our 
proposed actions and the conclusions of the RAND report. Another 
comment pointed to our related guidance entitled ``Guidelines for 
Ensuring the Quality of Information Disseminated to the Public'' 
(http://www.hhs.gov/infoquality/fda.html[numsign]i). FDA's guidance, 
which describes how we intend to meet our obligations under the Data 
Quality Act and the implementing Office of Management and Budget (OMB) 
guidelines, states that we are committed to ensuring that our 
regulatory decisions are based on objective information and notes our 
commitment to using the best available science conducted in accordance 
with sound and objective scientific practices, including peer reviewed 
science and supporting studies when available. This comment also cited 
the Center for Food Safety and Applied Nutrition's report ``Initiation 
and Conduct of All `Major' Risk Assessments within a Risk Analysis 
Framework'' (http://www.cfsan.fda.gov/dms/rafw-toc.html), which 
similarly stresses the importance of data quality and scientific 
objectivity in regulatory decisionmaking. Finally, this comment 
suggested that in evaluating the safety of dietary supplements 
containing ephedrine alkaloids, we should apply a rigorous scientific 
standard such as that used to evaluate whether a new drug application 
(NDA) should be approved or whether a health claim should be authorized 
under the significant scientific agreement standard (See Sec.Sec. 
314.125 and 314.126) (NDAs); Guidance for Industry: Significant 
Scientific Agreement in the Review of Health Claims for Conventional 
Foods and Dietary Supplements (http://www.cfsan.fda.gov/dms/
ssaguide.html) (health claims).
    (Response) We agree that we have an obligation to base regulatory 
assessments, including our regulatory assessment of the safety of 
dietary supplements containing ephedrine alkaloids, on sound science. 
We have spent a great deal of time and effort compiling and evaluating 
the best available scientific evidence relevant to this rulemaking, and 
our decision is based on a careful, objective analysis of the most 
current information, including peer reviewed studies. In considering 
whether dietary supplements containing ephedrine alkaloids present an 
unreasonable risk, we considered evidence from three principal sources: 
(1) The well-known, scientifically established pharmacology of 
ephedrine alkaloids; (2) peer-reviewed scientific literature on the 
effects of ephedrine alkaloids; and (3) the adverse events (including 
published case reports) reported to have occurred following consumption 
of dietary supplements containing ephedrine alkaloids. We believe that 
this final rule, and the data considered, are consistent with the 
principles set forth in the Data Quality Act and related guidances 
cited in the comments. We do not agree, however, that we should apply 
the same standard of scientific proof to a determination of 
adulteration under section 402(f)(1)(A) of the act, the ``significant 
or unreasonable risk'' provision, as we would apply to a decision 
whether to approve an NDA or authorize a health claim under other 
provisions of the act. Although our decision on dietary supplements 
containing ephedrine alkaloids must be based on sound science, that 
decision is not subject to, and need not meet, the very specific 
evidentiary requirements set out in the new drug and health claim 
provisions of the act (See 21 U.S.C. 355(d) and 21 U.S.C. 
343(r)(3)(B)(i)).

B. What Are the Known and Reasonably Likely Risks Presented by Dietary 
Supplements Containing Ephedrine Alkaloids?

1. Pharmacology
    We have reviewed numerous studies and other data related to the 
safety of dietary supplements containing ephedrine alkaloids. Evidence 
about the pharmacology of ephedrine alkaloids--as well as other 
evidence in the docket--shows that these products present a risk of 
serious adverse health effects. Information submitted to the docket in 
an effort to establish the safety of these products is inadequate to 
rebut the evidence of risk.
    (Comment 22) Several comments focused on the known pharmacological 
and toxicological effects of ephedrine/ephedra on the cardiovascular 
and nervous systems, explaining that ephedra contains vasopressor 
amines that excite the heart and constrict the blood vessels, which in 
turn increases heart rate and raises blood pressure. The comments 
contended that, because of these effects, adverse events such as 
hypertensive episodes, arrhythmias (abnormal heart rhythms), heart 
attacks, seizures, and strokes can be anticipated and expected when 
millions of people are exposed to such products. Various comments 
maintained that dietary supplements containing ephedrine alkaloids have 
the same pharmacological and toxicological activity as prescription and 
OTC ephedrine alkaloid drugs and, thus, present the same risks. One 
comment emphasized that Chen and Middleton (Ref. 33) warned about 
ephedrine alkaloid-induced thromboembolism (blood clots that travel in 
the body) in 1927 and thereafter, reports of toxicity appeared in the 
medical literature, accompanied by warnings against indiscriminate use 
by doctors and sale to consumers. These early reports are relevant to 
current reports of myocardial infarctions (heart attacks) and stroke 
associated with products containing ephedrine alkaloids.

[[Page 6801]]

    One comment stated that ephedra presents a danger of prolonged 
bleeding in those who undergo surgery, and that patients and doctors 
may not be aware of this potential complication. Another comment cited 
a review article (Ref. 2) that described myocardial depression 
occurring with repeated dosing of ephedrine, and cited a reference from 
a pharmacological textbook documenting ephedrine's tendencies to cause 
atrial and ventricular arrhythmias. Another comment suggested that we 
should not ignore the other ingredients commonly found in dietary 
supplements containing ephedrine alkaloids, such as caffeine, 
laxatives, and diuretics, because these ingredients can alter 
electrolyte levels and increase the risk of arrhythmias. One comment, 
citing a study by Haller et al., contended that the apparent causal 
role of ephedrine alkaloids in severe adverse effects could be related 
to the additive stimulant effects of caffeine (Ref. 34). One comment 
submitted by a manufacturer attributed the good safety record of its 
product to, among other reasons, the absence of caffeine and other 
stimulants.
    (Response) We agree that dietary supplements containing ephedrine 
alkaloids present risks of adverse physiological and pharmacological 
effects. Based on the best available scientific data and the known 
pharmacology of ephedrine alkaloids and other sympathomimetics, 
ephedrine alkaloids--including dietary supplements containing ephedrine 
alkaloids--pose short-term and long-term risks. This is clearest in 
long-term use, where increased blood pressure in any population will 
clearly increase the risk of stroke, heart attack, and death, but there 
is also evidence of increased risk from shorter-term use in patients 
with heart failure or underlying coronary artery disease.
    Ephedrine alkaloids are members of a large family of 
sympathomimetic compounds that include dobutamine and amphetamine. 
Members of this family increase blood pressure and heart rate by 
binding to alpha- and beta-adrenergic receptors present in many parts 
of the body, including the heart and blood vessels (Refs. 35, 36, and 
37). These compounds are called sympathomimetics because they mimic the 
effects of epinephrine and norepinephrine, which occur naturally in the 
human body. In addition to their direct pharmacological effects, many 
of these compounds also stimulate the release of norepinephrine from 
nerve endings. The release of norepinephrine further increases the 
sympathomimetic effects of these compounds, at least transiently. 
Sympathomimetic effects raise three concerns. First, sympathomimetics 
can induce cardiac arrhythmias in susceptible people, such as those 
with underlying coronary artery disease. Second, increased mortality 
has been observed in patients with congestive heart failure who were 
treated with sympathomimetic drugs, such as beta-agonists (early 
studies using such drugs as albuterol led to adverse outcomes) and 
xamoterol (Ref. 38), as well as phosphodiesterase inhibitors, which 
potentiate (increase the effect of) the effects of beta-agonists, 
including milrinone (Ref. 39) and enoximone (Ref. 40). The studies that 
showed these adverse effects occurred in about 3 months of product use. 
Third, sympathomimetics can raise blood pressure (Ref. 41).
    Based on clinical data, the ephedrine alkaloids present in dietary 
supplements would be expected to have the same or similar effects as 
other sympathomimetics on heart rate and blood pressure. Controlled 
clinical trials using products containing ephedrine alkaloids confirm 
their typical sympathomimetic effects. Single-dose studies of dietary 
supplements containing ephedrine alkaloids show that these products 
cause increases in both heart rate and blood pressure in healthy 
subjects (Refs. 42, 43, and 44). In one such study of a dietary 
supplement containing ephedrine alkaloids, the peak increase in blood 
pressure following a single oral dose of ephedrine alkaloids and 
caffeine (20 mg/200 mg) was 14 millimeters of mercury (mm Hg) systolic 
and 6 mm Hg diastolic, occurring about 2 hours after the single dose 
was taken (Ref. 42).
    The findings from these studies are complicated by the presence of 
caffeine in the dietary supplements used because caffeine is also known 
to have acute effects on blood pressure and heart rate. However, the 
effect of caffeine on blood pressure is transient and is lost within 2 
weeks of continued use (Refs. 45 and 46). Evidence that ephedrine 
independently causes an increase in blood pressure when coadministered 
with caffeine comes from two sources. First, there are studies in which 
ephedrine and caffeine were tested separately so that their effects 
could be compared. In a study by Jacobs et al., a group of healthy 
subjects received ephedrine (E, 0.1 mg/kilogram (kg) orally), caffeine 
(C, 4 mg/kg orally), the combination, or a placebo (P) (Ref. 47). 
Although caffeine caused a small increase in systolic blood pressure 
(average 3 to 6 mm Hg), ephedrine alone gave a 12 mm Hg effect, and 
when added to caffeine, increased systolic blood pressure by an 
additional 15 mm Hg (C+E = 156 +/- 29 mm Hg; E = 150 +/- 14; C = 141 +/
- 16; P = 138 +/- 14) (Refs. 47 and 48). Second, ephedrine has been 
shown in a clinical study to increase blood pressure and heart rate 
acutely when administered intravenously to children to maintain blood 
pressure during surgery (Ref. 37). Therefore, these studies show a 
blood pressure effect from ephedrine itself, independent of any 
additional effect from caffeine.
    In a multiple-dose controlled trial, Boozer et al. (2002) compared 
the effects of a combination of ephedrine alkaloids (from Ephedra) and 
caffeine (from kola nut) with placebo over a 6-month period in a highly 
selected population of obese and overweight individuals, who were 
carefully screened by medical history and medical evaluation to 
eliminate cardiovascular and other acute or chronic disorders (Ref. 
49). The study measured sitting blood pressure in the clinic using the 
cuff method for all 6 months (at weeks 1, 2, 3, 4, and every 4 weeks 
thereafter) of the study; these cuff measurements were not taken 
throughout the day so they reflect only a snapshot of the blood 
pressure at the time of measurement. The study also measured changes in 
blood pressure throughout the day at weeks 1, 2 and 4 using an 
automated blood pressure monitoring device (ABPM); the ABPM method 
provides more frequent measurements of blood pressure and is, 
therefore, better able to evaluate blood pressure effects over time. 
The ephedrine alkaloids and caffeine-treated subjects did not show a 
difference in the blood pressure measurements taken at the clinic, but 
did show statistically significant higher average blood pressure 
measurements over 24 hours at week 4 measured by ABPM (approximately 4 
mm Hg for both systolic and diastolic blood pressure) when compared to 
placebo treated subjects. The ABPM results are shown in a table in the 
paper. The difference in blood pressure between the two groups 
represented the sum of small downward changes in the placebo group 
(compared to baseline) and small upward changes, or no change, in the 
ephedra group. Boozer et al. reported numerous breakdowns of these data 
(e.g., 6 a.m. to midnight and midnight to 6 a.m.) and characterized the 
difference between the ephedra and placebo groups as small (about 3 mm 
Hg) but for the most common ABPM measure, 24-hour value, the difference 
was 4/4 mm Hg. The observation that this difference (shown in table 2 
of the paper) (Ref. 49) reflected a fall in blood

[[Page 6802]]

pressure in the placebo group as much as a rise in blood pressure in 
the ephedra group is not relevant. The only controlled and, therefore, 
reliable observation is the comparison of the two groups. Small changes 
from baseline can occur for a wide variety of reasons and are commonly 
observed in placebo and treated groups. Therefore, the ABPM data are 
important because they demonstrate that the effect of the ephedrine 
alkaloids, including dietary supplements containing ephedrine 
alkaloids, on blood pressure is not transient, but is still evident 
after 1 month of continued exposure (when measured by ABPM) and, 
therefore, would be expected to persist long term. The effect reported 
in the Boozer, et al. (2002) study cannot be attributed to the caffeine 
because the effect of caffeine on blood pressure (discussed previously) 
is transient, and the acute effect of caffeine to increase blood 
pressure is lost within 2 weeks of continued use (Refs. 45 and 46). 
While some effects of sympathomimetics show tachyphylaxis (i.e., 
decrease in response following repetitive administration of a 
pharmacologically active substance http://www.stedmans.com/) 
tachyphylaxis usually occurs rapidly. (FDA has verified the Web site 
address, but FDA is not responsible for any subsequent changes to the 
nonFDA Web sites after this document publishes in the Federal 
Register.) Therefore, we believe, based upon these data and our 
experience, that the blood pressure effects of ephedrine alkaloids seen 
after 4 weeks of continued use will persist.
    The Boozer et al. (2002) study (Ref. 49) was reviewed at our 
request by three outside scientific experts, Norman M. Kaplan, M.D. 
(Ref. 50), Richard L. Atkinson, M.D. (Ref. 51), and Mark Espeland, 
Ph.D. (Ref. 52). These experts were asked to give their independent, 
scientific opinion of whether the study provides adequate data to 
assess safety of ephedrine alkaloids and caffeine for weight loss--
considering, among other things, the design and duration of the trial 
and subject selection--and whether further studies are needed. In 
general, the experts concluded that the safety of ephedrine alkaloid 
and caffeine containing products could not be established by this study 
because the study used a highly selected population (i.e., carefully 
screened by medical history and medical evaluation to eliminate 
cardiovascular and other acute or chronic disorders) and had relatively 
few subjects. One of the experts also concluded that the duration of 
the study was inadequate to establish safety. In general, the reviewers 
found that the results raised safety concerns. Dr. Kaplan, one of the 
reviewers, raised the concern that the size of the change in blood 
pressure observed with ABPM, when applied to a large population, could 
translate into a significant increase in the incidence of strokes and 
heart attacks. Dr. Kaplan's concern reflects the potential consequence 
of long-term use of ephedra (i.e., the consequence of a population 
increase in blood pressure). A short-term increase (e.g., 1 to 2 
months) would not be expected to have such an effect. Approximately one 
in four adults has high blood pressure. Of those with high blood 
pressure, 31 percent are unaware that they have it (Ref. 53). A 
relative increase in blood pressure in any population, even individuals 
with ``normal'' blood pressure, will increase the risk of heart attack, 
stroke, and death in that population (Refs. 29, 29a, and 54).
    The extremely high prevalence of diagnosed and undiagnosed 
hypertension in the U.S. population and the likelihood that blood 
pressure in obese patients is already elevated make the 4 mm Hg effect 
shown by the Boozer et al. (2002) study (Ref. 49) one of great concern. 
Reductions in blood pressure of this magnitude (i.e., around 4 mm Hg 
diastolic or systolic) are clearly associated with substantial long-
term reductions in the occurrence of heart attack, stroke and death, as 
seen in meta-analyses of antihypertensive drug trials (Refs. 55 and 
56). While these trials were conducted in patients with hypertension, 
increasing blood pressure in any population, even in individuals with 
``normal'' blood pressure, will increase the risk of cardiovascular 
disease (Ref. 29).
    Epidemiological studies support a graded and continuous 
relationship between increased blood pressure and risk of stroke, heart 
attack, and sudden death, even when the increase is within the normal 
range (i.e., less than 140 mm Hg systolic and less than 90 mm Hg 
diastolic) (Refs. 29 and 30). This indicates that many people would be 
at an increased risk with long-term use of dietary supplements 
containing ephedrine alkaloids. Studies of hypertension treatments 
suggest that this increase in risk would occur fairly quickly in 
hypertensive individuals. Anti-hypertensive drugs that lower blood 
pressure by 4 to 6 mm Hg have been shown to significantly decrease the 
occurrence of cardiovascular morbidity (stroke, heart attack) and 
mortality (Refs. 55, 57, and 58). This effect is evident within 6 to 12 
months in large outcome studies (Refs. 29 and 30). FDA is concerned 
about the adverse health effects that can occur with the use of agents 
that raise blood pressure, such as dietary supplements containing 
ephedrine alkaloids, for short- or long-term use. Even in the case of a 
controlled clinical trial of a possible hypertension treatment where 
subjects are closely monitored, we advise sponsors to limit the length 
of time subjects can be in a placebo/untreated group to about 8 weeks 
to minimize their exposure to cardiovascular risks from the absence of 
treatment.
    As noted previously, the pharmacological effects of ephedrine 
alkaloids also present increased short-term risks of adverse health 
events in susceptible populations. For example, there is evidence from 
peer-reviewed scientific literature that a wide range of drugs with 
sympathomimetic activity, including beta-agonists, phosphodiesterase 
inhibitors, and dobutamine, have adverse effects (increased mortality 
due to heart failure and sudden death) in patients studied with 
congestive heart failure. These effects have been seen in relatively 
short-term studies (Refs. 59, 60, and 61) Similarly, there are studies 
that document that people with coronary artery disease are more 
susceptible to the well-known pro-arrhythmic effects of 
sympathomimetics (Refs. 62, 63, and 64) The occurrence of such an 
arrhythmic event is not one that requires prolonged exposure but would 
represent a risk associated with each use, including the first. Many 
individuals are unaware that they have coronary artery disease or early 
heart failure because these conditions may not cause prominent symptoms 
until later in the course of these conditions. As a result, we are 
concerned that such individuals will not know that they are at an 
increased risk for developing significant cardiovascular adverse events 
from even short-term use of dietary supplements containing ephedrine 
alkaloids. Overweight and obese individuals are particularly prone to 
hypertension, coronary artery disease, and/or heart failure, as 
overweight and obesity are associated with these conditions (Refs. 65 
and 66). These conditions may not manifest clinically until later in 
the course of the condition and, therefore, individuals, including 
overweight and obese individuals, may be unaware they have these 
conditions. As a population, the overweight and obese are, thus, at a 
greater risk even from short-term use of sympathomimetics.
    As summarized previously, the comments cited certain literature 
suggesting the possibility of additional adverse effects of ephedrine 
alkaloids,

[[Page 6803]]

such as prolonged bleeding in those who undergo surgery. Given the 
clear scientific evidence of this cardiovascular risks presented by 
dietary supplements containing ephedrine alkaloids, we have not relied 
on these other possible adverse effects noted in the comments in our 
determination of unreasonable risk.
    (Comment 23) Various comments did not agree that there are risks 
with products containing ephedrine alkaloids and stated the opinion 
that cardiovascular side effects associated with products containing 
ephedrine alkaloids in several blinded studies were not significantly 
different in control and treatment groups. Several comments maintained 
that there is no evidence from clinical studies that ephedrine 
``supplementation'' increases peak heart rate, peak blood pressure, or 
the prevalence of cardiac arrhythmias. Another comment contended that 
``clinically relevant doses'' of ephedra have no clinically significant 
effect on pulse or blood pressure, and produce no measurable 
alterations in myocardial function. A number of comments noted that 
changes in heart rate and blood pressure are transient and similar to 
those produced by exercise. Several comments stated that the effects of 
ephedra combined with caffeine on blood pressure are modest and 
generally subside over the first few days of use. Other comments stated 
that, although dietary supplements containing ephedrine alkaloids have 
a relatively high incidence of subjective and cardiovascular side 
effects with first use, the side effects diminish with continued use 
due to tachyphylaxis. Several comments noted that the literature, 
including the obesity studies we cited in the June 1997 proposal (Refs. 
36 and 67 through 80), indicated that tachyphylaxis sets in within a 
few days, at the most a few weeks, and results in a dramatic decrease 
in the likelihood of adverse events. Another comment suggested that 
pharmacological studies showed that peak ephedrine levels are reached 
within 1 to 4 days and that no further accumulation occurs thereafter. 
Another comment suggested that this fact means ephedrine alkaloids pose 
no risk of long-term toxicity.
    One comment noted that ephedrine alkaloids are not toxic in the 
classic sense, that is, do not cause organ changes or damage to the 
metabolism. Other comments suggested that the available pathology data 
do not show any pattern consistent with ephedrine alkaloids as a cause 
of death.
    (Response) We do not agree that ephedrine alkaloids pose no risk of 
adverse consequences. The suggestion that the cardiovascular effects of 
ephedrine alkaloids persist for only a few days is not supported by the 
Boozer et al. (2002) study (Ref. 49), which demonstrated a higher blood 
pressure (compared with placebo) at the end of 1 month of therapy (Ref. 
80a). This difference was observed when blood pressure was measured 
throughout the day, using ABPM, but not with cuff blood pressure 
measurements (a less sensitive measure). This difference in results 
using different measurement methods may have confused some readers and 
led them to conclude that ephedrine alkaloids do not have a clinically 
meaningful effect on blood pressure. The fact that an effect on blood 
pressure (as measured using ABPM, which follows measurements throughout 
the day) was still present at 1 month strongly indicates that 
tachyphylaxis to the effects of ephedrine does not occur. As discussed 
in the response to comment 22 of this document, tachyphylaxis tends to 
occur rapidly, as with caffeine, whose blood pressure raising effect is 
lost within 2 weeks. Therefore, FDA does not agree with the comments 
expressing assurances that adverse effects will disappear with 
continued use of ephedrine alkaloids because of tachyphylaxis.
    Additionally, some of the studies cited by the comments apparently 
measured cuff blood pressure only around the time of dosing, when 
minimal serum concentrations of ephedrine alkaloids and effects on 
blood pressure would be expected. Absence of an effect at this time 
cannot be seen as evidence that ephedrine alkaloids do not increase 
blood pressure.
    The suggestion that ``clinically relevant'' or ``clinically 
significant'' doses of ephedrine have no effects on blood pressure is 
unsupported by the available data. What constitutes a ``clinically 
relevant or significant'' dose is undefined (and unlikely to be 
definable given the nature of the available efficacy data for ephedrine 
alkaloids). The difficulties in using the available clinical data to 
obtain such reassurance with regard to the safe use of ephedrine are 
discussed in the response to comment 26 of this document.
    We do not agree that the clinical studies establish that ephedrine 
does not have adverse pharmacological and clinical effects. The 
published controlled studies of the use of ephedrine alkaloid products 
for weight loss cited by these comments cannot establish the safety 
profile of these products. First, many of the most serious risks, such 
as strokes or heart attacks (consequences of elevated blood pressure), 
arrhythmias, or worsened heart failure, are relatively infrequent or 
are delayed and, therefore, will not be detected in studies using small 
populations (such as under 100 patients per group) as these studies 
did. Second, these studies often had other important design 
limitations, such as lack of adequate controls (including the absence 
of placebo groups in some studies), and inadequate information about 
the causes that led to participants dropping out of the trial. In 
addition, persons with known cardiovascular disease or cardiovascular 
risks were usually excluded. Thus, these studies were not designed to 
detect serious adverse effects in susceptible individuals, nor to 
detect adverse effects that occur infrequently. As discussed in the 
following paragraphs, these studies were also not adequately designed 
to assess blood pressure effects. Given these limitations, it is not 
surprising that these published studies do not report serious adverse 
events (Refs. 21, 22, 50, 52, and 81).
    These trials also would not have been able to detect effects on 
blood pressure because of other design limitations. For example, when 
sponsors of drug products seek to detect a drug-induced decrease in 
blood pressure in patients with hypertension, the trial is specifically 
designed to perform the following functions: (1) Assess the blood 
pressure effects at both peak and trough levels of the drug in the 
blood, and (2) measure blood pressure in a consistent and reproducible 
manner. This typically requires the enrollment of at least 100 patients 
to detect a difference from placebo of around 4 to 6 mm Hg systolic, 
multiple measures at each time point and careful attention to how blood 
pressure is measured. These design features are either lacking or not 
described in the publications cited by the comments summarized above, 
significantly limiting the trials' ability to detect any differences 
between the treatment and placebo groups with regard to blood pressure 
or heart rate. With regard to the timing of the measurement, the blood 
pressure measures appear to have been made at (or shortly after) the 
administration of the product containing ephedrine for almost all of 
the published trials. Absorption of the new dose would be minimal or 
incomplete and the dose taken the day before (8 to 12 hours earlier) 
would have been substantially removed from the circulation, given 
ephedrine's approximately 4-hour half-life. Blood levels of ephedrine 
would

[[Page 6804]]

thus be at or near their lowest values of the day (``trough level''), a 
time when minimal effects on blood pressure would be anticipated. 
Measurements made only at trough level might well miss a significant 
effect on blood pressure that would have been seen at or near peak 
concentrations of ephedrine. Thus, although some published studies on 
the cardiovascular effects of ephedrine (especially blood pressure) 
over a period of weeks or months have reported little or no effect of 
ephedrine on blood pressure and a variable effect on heart rate, these 
studies are severely limited in their ability to establish safety 
because of the clinical trial design limitations (Refs. 81a, 81b, and 
81c), such that the true effects of ephedrine on heart rate and blood 
pressure cannot have been adequately assessed.
    We do not agree with the comments that state that ephedrine 
alkaloids are not toxic because they do not induce specific organ 
pathology. Persistently elevated blood pressure can result in defined 
cardiovascular toxicity (Refs. 29, 29a, and 54), as can ephedrine's 
sympathomimetic effects in people with coronary artery disease or heart 
failure, but the kinds of damage seen in humans from these effects 
would look the same as similar damage that occurs from the underlying 
disease or from raised blood pressure or arrhythmia due to another 
cause.
    (Comment 24) A number of comments discussed the relevance of PPA to 
regulatory decisions on dietary supplements containing ephedrine 
alkaloids. Several comments stated that PPA is a metabolite of 
ephedrine. Various comments contended that ephedrine and PPA are both 
partial agonists and that adverse events associated with dietary 
supplements containing ephedrine alkaloids are of the same type and 
greater in number than those associated with PPA, which was voluntarily 
withdrawn from the U.S. market for safety reasons. Other comments 
maintained that we should not use PPA data to support the hazards of 
dietary supplements containing ephedrine alkaloids. Several such 
comments stated that because PPA differs in pharmacological, 
pharmacokinetic, and pharmacotoxic effects from ephedrine or 
pseudoephedrine, it is scientifically inappropriate for us to assume 
that all ephedrine alkaloids are equivalent. Other comments asserted 
that the various isomers of ephedrine alkaloids have different actions, 
different favorable and adverse effects, different activation of 
receptors, and different effects on human tissues. Several comments 
indicated that norephedrine (an ephedrine alkaloid that makes up one 
component of PPA) is a metabolite of ephedrine and that interactions of 
the multiple ephedrine alkaloids in Ephedra and other botanicals and 
their in vivo metabolites should be considered in a safety evaluation 
of these ingredients and products containing them.
    A few comments asserted that the Hemorrhagic Stroke Project (HSP) 
(Ref. 19) was not designed to assess ephedra exposure. These comments 
maintained that the HSP is limited by significant issues relating to 
observation bias, selection bias, and confounding. One comment 
complained that we reopened the ephedra docket requesting comment on 
the HSP, but we did not place in the docket, or request comment on, the 
many published and unpublished clinical studies submitted by one trade 
organization to support PPA's safety. The comment asserted that our 
review of the pharmacology of ephedrine alkaloids did not include most 
of the pivotal information on PPA submitted to us by the Consumer 
Healthcare Products Association (CHPA). Another comment expressed the 
view that, in our review of safety data related to ephedra, we should 
avoid relying on safety data concerning other ingredients.
    (Response) The substance, l-norephedrine, also known as (-)-
norephedrine, refers to the isomeric portion of PPA that occurs 
naturally in Ephedra and as a metabolite of ephedrine in the body. We 
agree that the l-norephedrine in racemic PPA is a metabolite of 
ephedrine, and further that ephedrine and its metabolites have potent 
vasoactive properties, reinforcing the view that dietary supplements 
containing ephedrine alkaloids have the pharmacological properties 
described in the response to comment 22 of this document. These 
properties, in turn, are linked to predictable adverse clinical 
outcomes both in the general population (e.g., increased blood 
pressure) and in susceptible populations (e.g., cardiac arrhythmias). 
Although there are some similarities between PPA and ephedrine, there 
are also differences. PPA shows tachyphylaxis to rises in blood 
pressure within approximately 24 hours and usage has been linked to 
hemorrhagic strokes (bleeding strokes due to a ruptured blood vessel). 
Ephedrine does not show such tachyphylaxis. In addition, use of 
ephedrine has been associated with ischemic strokes (a blood clot 
blocking off an artery causing a lack of oxygen to portions of the 
brain), but not hemorrhagic strokes. The major alkaloid in most dietary 
supplements containing ephedrine alkaloids is generally ephedrine, and 
not norephedrine (Ref. 82).
    Therefore, we have not relied on the HSP or spontaneous reports of 
hemorrhagic stroke in patients receiving PPA for any of our conclusions 
about the risks of ephedrine alkaloids, and data regarding PPA is not 
as informative for drawing conclusions about the benefits and risks of 
dietary supplements containing ephedrine alkaloids as data on 
ephedrine. Of course, those supplements that contain meaningful amounts 
of PPA would pose additional serious risks expected from the use of 
PPA-containing products, such as hemorrhagic strokes. This adverse 
event can occur in healthy individuals with one dose of PPA. Reopening 
the docket to request comment on these data is unnecessary as we have 
not relied on the data for our determination in this final rule.
    (Comment 25) One comment stated that l-ephedrine is both a direct 
and indirect-acting isomer with both alpha- and beta-agonist activity, 
while d-pseudoephedrine acts indirectly on both receptors. PPA, which 
is racemic (i.e., contains both the (+) and (-) forms of the chemical), 
is a direct and indirect agonist for alpha-receptors but has weaker 
beta-receptor activity. The comment suggested that ephedrine, 
pseudoephedrine, and PPA elevate blood pressure, but only l-ephedrine 
and d-pseudoephedrine increase heart rate. The comment cited Chua and 
Benrimoj (Ref. 83) stating that d-pseudoephedrine has half of the 
bronchodilator activity compared to l-ephedrine and one-quarter of the 
vasopressor effect. The comment argued that we cannot use the 
pharmacokinetic and toxicokinetic properties of any isomer to predict 
that of other ephedrine isomers.
    (Response) Given that Ephedra and other botanicals used as dietary 
ingredients contain a mixture of ephedrine alkaloids, and given the 
small database on the supposed selective effects of the isomers, we 
cannot draw any reassurance from the possibility that one alkaloid has 
more or less of an effect on the vasculature (or organ systems) than 
another alkaloid. Further, the reported differences in receptor binding 
affinity or other in vitro tests cannot eliminate concern about the 
effects of ephedrine alkaloids in humans, because there is clinical 
evidence that ephedrine alkaloids have important pharmacological 
effects (e.g., increased blood pressure, heart rate) that persist, 
particularly in the case of ephedrine, through at least 1 month of use. 
As noted previously in this document, the

[[Page 6805]]

major alkaloid in most dietary supplements containing ephedrine 
alkaloids is generally ephedrine (Ref. 82). The comments pointing to 
evidence of differences in the effects of different ephedrine alkaloids 
do not provide a basis to conclude that dietary supplements containing 
ephedrine alkaloids do not present an unreasonable risk of illness or 
injury.
    (Comment 26) Some comments argued that the scientific literature 
indicates that single doses of ephedrine up to 60 mg generally do not 
increase blood pressure (Ref. 83). Other comments cited a handbook of 
intravenous drug therapy for nurses that states that ephedrine is of 
low toxicity. One comment stated that the scientific literature 
describing the effects of ephedrine in doses of 50 to 150 mg does not 
support the contention that ephedrine in dosages of 50 to 150 mg per 
day would represent a health hazard. Many comments stated that reviews 
of the literature and other data by independent experts reflect the 
scientific consensus that ephedrine alkaloids at 25 mg per dose are 
safe. One comment cited a clinical study of 98 elderly patients 
undergoing hip surgery who received 0.6mg/kg ephedrine by intramuscular 
injection. One out of 48 patients in the placebo group and two out of 
50 in the ephedrine group experienced increased heart rate or increased 
systolic blood pressure greater than 20 percent from baseline. The 
comment concluded that the dosages used are greater than the dosages 
found in any dietary supplement containing ephedrine alkaloids and that 
the results of the study are consistent with the conclusion that, as 
also asserted by other comments, no significant injury has been clearly 
associated with dietary supplements containing ephedrine alkaloids when 
used as directed.
    We received numerous other comments dealing with the issue of 
``safe'' doses for ephedrine alkaloids in dietary supplement products. 
Many expressed the view that low doses of ephedrine alkaloids in 
dietary supplements do not pose a safety concern and should remain on 
the market.
    (Response) We do not agree that the scientific literature indicates 
that there is a dose of ephedrine or ephedrine alkaloids that does not 
present a risk of adverse events. Although dosages vary in dietary 
supplements containing ephedrine alkaloids, most products are labeled 
with 20-25 mg ephedrine alkaloids per recommended serving and 100-150 
mg ephedrine alkaloids per day. Some of the doses described in the 
comments as safe (50 to 150 mg ephedrine alkaloids per day) are in the 
range studied by Boozer et al. (90 mg ephedrine alkaloids per day) 
(Ref. 49) and, thus, could cause an increase in blood pressure, a 
significant health concern (see previous discussion). We also do not 
agree that some lower dose of ephedrine has been demonstrated not to 
increase blood pressure and heart rate. The relationship between a 
given dose of ephedrine and changes in heart rate and blood pressure 
has been poorly characterized, although it is clear that ephedrine is 
capable of increasing both. As discussed in the response to comment 23 
of this document, the published studies that have found no effects on 
blood pressure and/or heart rate have had methodological deficiencies 
that limited their ability to detect such changes. With respect to the 
clinical study of 98 elderly patients, the failure to find serious 
adverse events is understandable, as the study was designed to 
demonstrate that intramuscular ephedrine was effective to prevent 
hypotension related to spinal anesthesia. The concern that led to the 
study was adverse events related to an expected decrease in blood 
pressure resulting from the anesthesia. As would be expected based on 
the pharmacology of ephedrine, the study showed that ephedrine is 
effective in maintaining blood pressure in patients receiving spinal 
anesthesia.
    We do not agree with comments that suggest that low doses of 
ephedrine alkaloids in dietary supplements do not present an 
unreasonable risk and should remain on the market. Because this issue 
was raised in comments responding to the June 1997 proposal, we 
commissioned a scientific review that was placed in the 2000 docket 
(Refs. 84 and 85). This review concluded that a ``safe dose'' of 
ephedrine alkaloids cannot be identified. The review determined that 
even ``a dose of 1.5 mg every 4 hours (a daily dose of 9 mg) would 
produce cardiovascular effects that may be dangerous alone, or in 
association with risk factors* * *'' (Ref. 84 at p. 6). We also note 
that in the 1996 FAC meeting, several committee members stated that, 
based on the available data, no safe level of ephedrine alkaloids could 
be identified for use in dietary supplements (Ref. 86). Consequently, 
they recommended removing dietary supplements containing ephedrine 
alkaloids from the market (Ref. 87). Although the CANTOX Health 
Sciences International (CANTOX) review attempted to establish a level 
of ephedrine alkaloids at which there were no adverse effects, we do 
not consider the information submitted sufficient to establish a 
``safe'' dose (see discussion of CANTOX in the response to comment 32 
of this document).
    (Comment 27) Many comments raised the issue of the safety of 
dietary supplements containing ephedrine alkaloids for use in sensitive 
or special populations. A number of comments indicated that certain 
individuals may be relatively more sensitive to the stimulant effects 
of ephedrine alkaloids, and as a result, at greater risk for adverse 
health consequences. One comment from a physician noted that he does 
not recommend the use of ephedra products by pregnant women. Another 
comment indicated a particular safety concern with the use of dietary 
supplements containing ephedrine alkaloids in older persons; according 
to the comment, many elderly persons take medications for which the use 
of dietary supplements containing ephedrine alkaloids would be 
contraindicated. Citing a survey that indicated that shift workers 
frequently use stimulants, including ephedrine alkaloids, in 
combination with coffee, depressants and/or pain relievers that contain 
caffeine, one comment expressed the view that ephedrine alkaloids pose 
a significant health risk to the shift worker population (Ref. 88). The 
comment further submitted that 69 percent of shift workers are 
overweight, that shift work is likely to involve physical labor, often 
performed in hot conditions, and that these factors increase the risks 
of adverse cardiovascular effects when shift workers use ephedrine 
alkaloids. Other comments stated that the presence or absence of a 
susceptible population cannot be determined with the available data. 
Several comments stated that dietary supplements containing ephedrine 
alkaloids are not for everyone, and consumers should consult a 
physician prior to use if they have specified preexisting health 
conditions.
    (Response) We agree with the comments that expressed concern about 
the effects of ephedrine alkaloids on susceptible populations and have 
previously discussed long-term and short-term risks to susceptible 
populations in the response to comment 22 of this document. There is 
every reason to expect that certain populations will be more 
susceptible to the adverse effects of ephedrine alkaloids and that many 
such people will not be aware of their greater susceptibility. As noted 
previously, people with coronary artery disease, early congestive heart 
failure, and high blood pressure, all of which are more

[[Page 6806]]

common in obese individuals, are often unaware of these risk factors. 
Thus, the recommendations contained in the comments regarding the 
suitability of dietary supplements containing ephedrine alkaloids for 
certain populations and the need to consult a physician if the consumer 
has certain preexisting conditions are ineffective to mitigate the risk 
that dietary supplements containing ephedrine alkaloids pose to these 
susceptible populations.
    (Comment 28) Several comments stated that warning labels on dietary 
supplements containing ephedrine alkaloids are not sufficient to 
protect the public health because many individuals are not aware they 
have medical conditions or individual sensitivities that put them at 
greater risk for experiencing serious adverse effects.
    The comments stated that warnings are ineffective for individuals 
who are not aware that they have disease conditions such as high blood 
pressure or other cardiovascular diseases, hyperactive thyroid 
function, undiagnosed cerebrovascular abnormalities, or a propensity 
for cardiac arrhythmia, seizure or certain psychiatric disorders. The 
same comments maintained that even small amounts of ephedrine alkaloids 
can be potentially dangerous to otherwise healthy individuals who may 
have a genetically predetermined sensitivity to ephedrine alkaloids or 
other sympathomimetic agents. Other comments asserted that warning 
labels are ineffective because serious adverse events have occurred 
after the initial or first few uses.
    (Response) We generally agree with the comments. Warning labels may 
be beneficial when people are able to identify the risk factors about 
which they are being warned. As explained in section V.B.3 of this 
document, OTC drug products containing ephedrine or pseudoephedrine 
bear warnings that they should not be used by certain populations. 
Despite the identified risks of these products, we have determined that 
the demonstrated health benefits for the labeled OTC drug uses outweigh 
their risks for certain temporary, episodic disease uses when 
appropriate warnings are contained in the product labeling. While 
dietary supplements containing ephedrine alkaloids present the same 
risks, there are no health benefits for the labeled uses sufficient to 
outweigh their risks (see discussion in sections V.C and V.D of this 
document). A more detailed discussion on why a warning label would be 
insufficient to make the risks of dietary supplements containing 
ephedrine alkaloids reasonable appears in section VI.A of this 
document.
    (Comment 29) A number of comments indicated that ephedrine 
alkaloids could only be used safely under the supervision of a health 
professional or that products containing ephedrine alkaloids should be 
restricted to prescription use only. Reasons given for these opinions 
included the potential for interactions between dietary supplements 
containing ephedrine alkaloids and caffeine or other commonly available 
products (predominantly drugs) that might not be identified by the 
typical consumer. Other comments stated that consumers could not self 
diagnose many of the conditions where the use of ephedrine alkaloids 
would either be contraindicated or pose a potential safety concern.
    In contrast, a physician who used dietary supplements containing 
ephedrine alkaloids in his practice stated that he was as comfortable 
with people using dietary supplements containing ephedrine alkaloids on 
their own, as he was with people using an OTC drug product on their 
own.
    (Response) We generally believe that the risks posed by dietary 
supplements containing ephedrine alkaloids when used continuously, 
particularly in obese patients who may already have underlying 
illnesses that can be aggravated by these products (such as 
hypertension), cannot be adequately mitigated without physician 
supervision. Sustained high blood pressure has significant 
consequences, including increased risk of stroke, heart attack, and 
death. As noted previously, even short-term use of dietary supplements 
containing ephedrine alkaloids poses certain risks, such as arrhythmias 
in patients with coronary artery disease. While we allow ephedrine and 
pseudoephedrine in OTC drugs for temporary, episodic uses, such as the 
temporary relief of symptoms (shortness of breath, tightness of chest, 
and wheezing) of certain diseases (e.g., colds, allergies, previously 
diagnosed bronchial asthma, colds, allergies) individuals who use 
dietary supplements containing ephedrine alkaloids for reasons other 
than to improve their health (e.g., to lose weight for improved 
appearance) obtain no health benefits and at the same time are at risk 
for the types of adverse events that can occur with both short and 
long-term use of ephedrine alkaloids. As discussed more thoroughly in 
section V.C.1 of this document, use for relatively short term weight 
loss would give, at best, a weight loss of a few pounds, which would 
not be sufficient to result in any health benefit. However, use for 
weight loss is likely to be longer term, giving a sustained increase in 
blood pressure in addition to the short-term risks. If these products 
met prescription drug standards, then it is possible that the risks of 
use for weight loss could be mitigated by a physician's evaluation of 
the patient's medical history and appropriate monitoring during 
treatment. We note that manufacturers can conduct clinical 
investigations of ephedrine alkaloids under an IND application and can 
seek approval of ephedrine alkaloid-containing products as new drugs 
for the treatment of obesity or other diseases under a NDA if 
sufficient evidence is provided to support such use. It is also 
possible that products containing ephedrine alkaloids might not present 
an unreasonable risk, even without physician supervision, if they were 
marketed as dietary supplements for a use that results in a meaningful 
health benefit and that requires only temporary, episodic use to 
achieve the benefit. However, based on the information we have now, we 
believe that it is unlikely that any such nondisease use could be 
identified.
    (Comment 30) Another comment, citing a study by Haller et al., 
contended that the apparent causal role of ephedrine alkaloids in 
severe adverse effects could be related to the additive stimulant 
effects of caffeine (Ref. 34). One comment submitted by a manufacturer 
attributed the good safety record of its product to, among other 
reasons, the absence of caffeine and other stimulants.
    (Response) While caffeine would be expected to have additive 
effects with ephedrine alkaloids, acute administration of ephedrine 
alone increases blood pressure and heart rate (Refs. 37 and 47). The 
available evidence shows that chronic use of caffeine has no effect on 
blood pressure that persists beyond 2 weeks (Refs. 45 and 46), in 
contrast to ephedrine, which does have a persistent effect (Boozer) 
(Ref. 49).
    (Comment 31) Many comments contended that we failed to consider the 
differences among ephedrine alkaloids from the raw botanical; extracts 
from the raw botanical that contain unaltered proportions of alkaloids 
and other substances; concentrated and/or otherwise manipulated ephedra 
extracts such that naturally occurring proportions and/or quantities of 
ephedrine alkaloids are changed; and synthetic or pure isolated 
ephedrine (extracted as a single entity from the plant). Because these 
products have chemical differences and differences in

[[Page 6807]]

potency, toxicity, pharmacokinetics, and pharmacological and 
physiological effects, the comments maintained they should be 
considered separately in scientific, medical, and regulatory contexts.
    Other comments, citing a study by White et al., stated that other 
natural constituents, including other alkaloids and ephedradines in the 
raw botanical, modify or attenuate the physiological and 
pharmacological effects of the ephedrine contained in dietary 
supplements (Ref. 43). Numerous comments maintained that raw Ephedra 
and/or Ephedra extracts are safer than ephedrine that is synthetic or 
that has been isolated and that serious adverse events associated with 
the appropriate use of ephedra have been rare. Several comments 
asserted that the ephedradines have hypotensive effects and are found 
in ephedra roots, rather than the aerial portions of the plant. One 
comment maintained that ephedradines are thought to occur in small 
amounts in Ephedra stems. One comment stated that ephedra extract is 
safer than pharmaceutical ephedrine based on the fact that the 
LD50 is higher for the botanical extract (5.4g/kg) when 
compared to the LD50 for pharmaceutical ephedrine (64.9 mg/
kg) (``LD50'' refers to the amount of a material that causes 
death in 50 percent of test animals).
    Several comments stated that pharmaceutical ephedrine is more 
potent than ephedrine from botanical sources because ephedrine 
comprises only 30 to 90 percent of the total alkaloids of the raw 
botanical, with the remaining portion containing potentially less 
potent stimulants such as pseudoephedrine. Several comments claimed 
that the various ephedrine alkaloids from botanical sources have a 
slower rate of absorption due to the plant matrix as compared to the 
rate of absorption for pharmaceutical ephedrine (Ref. 43). These 
comments stated that delayed effects diminish side effects and provide 
for the cardiovascular adaptation of effects, thereby diminishing 
cardiovascular response. One comment stated that except for absorption 
rate, ephedrine alkaloids from the plant have the same pharmacokinetics 
as pharmaceutical ephedrine (Ref. 43). Other comments note that 
botanical ephedrine from formulations containing whole Ephedra is 
absorbed more slowly than dietary supplements formulated with 
standardized extracts (Ref. 44). A few comments suggested that ephedra 
extract has higher neurocytotoxic (toxic effect on nerve cells) 
potential than synthetic ephedrine hydrochloride due to combinations of 
different ephedrine alkaloids or other unknown compounds found in 
ephedra extract that are not found in ephedrine hydrochloride (Ref. 
89).
    Other comments maintained that there is no difference between blood 
levels of ephedrine from botanical sources and ephedrine contained in 
OTC drugs. Comments from a State Board of Pharmacy stated that 
ephedrine from botanical sources is neither safer than, nor different 
from, pharmaceutical ephedrine. One comment objected to our including 
clinical studies using pharmaceutical ephedrine in our evaluation. A 
number of comments suggested that naturally occurring ephedrine is more 
potent than its synthetic counterpart. A few comments stated that the 
presence of varying amounts, proportions and chemical configurations of 
ephedrine alkaloids in crude Ephedra and prepared Ephedra extracts, as 
well as the presence of unknown compounds, leads to uncertainty in 
dose, purity, and composition and a greater risk for adverse effects. 
Comments noted that this variability is not an issue for synthetic or 
pure isolated ephedrine alkaloids.
    (Response) The data are wholly inadequate to demonstrate that any 
differences among forms of naturally occurring ephedrine alkaloids and 
synthetic ephedrine have a meaningful impact on risks to health. The 
overall database of clinical trials, including trials using both 
natural and synthetic ephedrine, does not lead to the conclusion that 
one form of ephedrine is safer than the other form.
    We are not persuaded by any of the available evidence that 
ephedrine from botanical sources is materially different from ephedrine 
from pharmaceuticals with respect to chemistry, potency, or 
physiological and pharmacological effects. Chemically, any isomer with 
the same conformation from one source, including botanical sources, is 
identical to the same isomer from another source. For example, (-)-
ephedrine from Ephedra (Ephedra sinica Stapf) is chemically 
indistinguishable from synthetic (-)-ephedrine manufactured by a 
pharmaceutical company.
    Regarding the ephedradines, we are not aware of any evidence in the 
scientific literature, nor were any data provided in the comments, that 
indicate that these compounds are present in Ephedra, in other 
botanical sources of ephedrine alkaloids, or in extracts from these 
botanicals. The ephedradines are known constituents of the roots of the 
species Ephedra sinica Stapf (Ref. 90). In traditional Asian medicine, 
the roots and rhizome of the plant are referred to as ``ma huang gen,'' 
while the aerial parts of the plant are referred to as ``ma huang'' 
(Ref. 3). The ephedradines are not ephedrine alkaloids. Nor are they 
present in the aerial parts of the plant that are used in dietary 
supplements. The scientific evidence, thus, does not support the 
opinion that the other ephedradrines in the raw botanical act to modify 
or attenuate the physiological and pharmacological effects of the 
ephedrine alkaloids contained in these products.
    We do not agree, therefore, that current evidence establishes that 
ephedrine alkaloids from botanical sources, including botanical 
extracts, are different from, or are any safer than, pharmaceutical 
ephedrine alkaloids. With regard to the comment asserting that ephedra 
extract is safer than pharmaceutical ephedrine because the 
LD50 is higher for the botanical extract than the 
LD50 for pharmaceutical ephedrine, we note that scientific 
views on this point differ. Another scientific reference suggests that 
a mixture of ephedrine alkaloids from a botanical extract may be more 
toxic, based on LD50 calculations, than an equal amount of 
pharmaceutical ephedrine (Ref. 91). While there is not enough 
scientific evidence to draw a conclusion, we acknowledge the 
possibility that other components in the concentrated extracts (e.g., 
tannins derived from the botanical) may affect the toxicity of 
botanical preparations of ephedrine alkaloids (Refs. 89 and 92).
2. Other Safety Data
    (Comment 32) Many comments cited multiple data and information 
sources as support for the safety of dietary supplements containing 
ephedrine alkaloids. These cited sources have been submitted to the 
docket and include the CANTOX review, RAND Report, the Ad Hoc Committee 
on the Safety of Ma Huang report and the Ad Hoc Committee on the Safety 
of Dietary Supplements, Ephedra Education Council Expert Panel Report, 
and a 6-month clinical trial by Boozer et al. (2002) (Refs. 21, 49, 93, 
94, and 95). Some comments also claimed that the toxicological database 
supports clinical evidence of safety; that no serious adverse events 
have been reported in controlled clinical trials using products 
containing ephedrine alkaloids for weight loss, and that few or no 
serious adverse events have been reported to manufacturers of dietary 
supplements containing ephedrine alkaloids.
    One trade association commented that a valid and quantitative 
scientific process is needed to identify intakes

[[Page 6808]]

and conditions of use that do not cause significant or unreasonable 
risk, and urged us to adopt scientific conclusions based on the CANTOX 
risk assessment, which was based on methods developed by the Institute 
of Medicine (IOM) (Ref. 28). A number of comments argued that the 
results of the CANTOX review established that dietary supplements 
containing ephedrine alkaloids are safe when used in accordance with 
the industry standard.
    One comment stated that the methods employed by CANTOX were not 
appropriate for use in evaluating the safety of dietary supplements 
containing ephedrine alkaloids. Several comments stated that there are 
no data that establish that ephedrine alkaloids are an ordinary 
component of food, that there is a need for ephedrine alkaloids in the 
diet, or that some deficiency state exists when ephedrine alkaloids are 
not a normal component of the diet.
    (Response) We do not agree with the methodology or conclusions of 
the risk assessment performed by CANTOX. The CANTOX review, sponsored 
by an industry trade group, was a quantitative risk assessment that 
used IOM methods to determine a safe upper level (called the No 
Observed Adverse Effect Level (NOAEL)) for botanical ephedrine 
alkaloids as used in dietary supplements. We believe that this review 
cannot be used to establish a NOAEL for ephedrine alkaloids used in 
dietary supplements because it was flawed. Its flaws include use of an 
inappropriate risk assessment model and deviation from the criteria and 
procedures established by IOM, including relying on abstracts and 
unpublished articles, using an unsuitable definition of ``Tolerable 
Upper Intake Level'' (UL), and using an overly narrow definition of 
``adverse effect.''
    The IOM model referenced by CANTOX is the Food and Nutrition 
Board's report entitled ``Dietary Reference Intakes: A Risk Assessment 
Model For Establishing Upper Intake Levels For Nutrients.'' The 
introduction to this report states that dietary reference intakes are 
being established for ``nutrients and food components'' which include 
nutrients, dietary antioxidants, micronutrients including electrolytes 
and fluid, macronutrients, ``and other food components not 
traditionally classified as ``nutrients,'' but purported to play a 
beneficial role in human diets'' (Ref. 28 at pp. 1 and 2). The IOM 
report defined dietary reference intakes, in part, as ``reference 
values that are quantitative estimates of nutrient intakes to be used 
for planning and assessing diets for healthy people. They include both 
recommended intakes and [tolerable upper intake levels] as reference 
values'' (Ref. 28 at p. 2). The report defined ``Tolerable Upper Intake 
Level'' (UL) as ``the highest level of daily nutrient intake that is 
likely to pose no risk of adverse health effects to almost all 
individuals in the general population. As intake increases above the 
UL, the risk of adverse effects increases'' (Ref. 28 at p. 3). The 
rationale for establishing such a risk assessment model is that 
nutrients are an essential part of the diet and deficiency states 
result when they are absent from the diet or are available in too low 
of a concentration.
    CANTOX claimed that the use of this model was appropriate for 
ephedrine alkaloids in dietary supplements because nutrients, like all 
chemical agents, can produce adverse health effects if intakes are 
excessive. However, ephedrine alkaloids are not nutrients. The CANTOX 
report did not include any data establishing that there is a need for 
ephedrine alkaloids in the diet, or that some deficiency state exists 
when ephedrine alkaloids are not present in the diet. Therefore, we 
conclude that the use of the IOM risk assessment method based on the 
model of a nutrient is inappropriate for the evaluation of the safety 
of dietary supplements containing ephedrine alkaloids.
    Even if the IOM dietary reference intakes model were an appropriate 
risk assessment model for dietary supplements containing ephedrine 
alkaloids, we note that CANTOX deviated from the IOM's criteria and 
procedures in several important ways. For instance, the IOM report used 
studies published in peer-reviewed journals as the principal sources of 
data for its evaluations. In contrast, while CANTOX did use some 
publications, it also relied on abstracts and unpublished studies. For 
example, CANTOX cited the study by Boozer, et al. as the pivotal study 
demonstrating the safety of dietary supplements containing ephedrine 
alkaloids and the establishment of the NOAEL. However, the Boozer (Ref. 
96) study was only available in abstract form at the time of the CANTOX 
review. Abstracts are not subject to the same rigorous peer review that 
full manuscripts go through. Further, abstracts do not contain 
sufficient information to enable a reader fully to evaluate a study's 
methodology or independently to interpret or verify a study's results. 
As a result, abstracts should not be given the same weight as the full 
reports of studies themselves. In the case of the Boozer study, the 
abstract did not provide details on the exclusion or inclusion criteria 
for the study, so a reader could not determine how the subjects were 
selected or how they were monitored during the study. The CANTOX 
authors also did not acknowledge the significance of the blood pressure 
findings in the Boozer et al. As we have discussed extensively in 
section V.B.1 of this document, this study by Boozer et al. (Ref. 49) 
clearly demonstrates a higher blood pressure in ephedra plus caffeine 
treated subjects (compared to placebo), which translates into serious 
long-term risks in the general population and serious short-term risks 
in susceptible populations. Furthermore, as stated by outside 
scientific experts who reviewed this study, the Boozer et al. (2002) 
study cannot establish the safety of dietary supplements containing 
botanical ephedrine alkaloids and caffeine because the study used a 
highly selected population, had relatively few subjects and was carried 
out for too short a period of time. Rather, the Boozer study raises 
questions about the safety of these products.
    Indeed, of the 20 studies that CANTOX considered in identifying the 
NOAEL, four were abstracts, and two were unpublished reports. Thus, 
unlike the IOM report's reliance on peer-reviewed journal articles, a 
significant proportion of the CANTOX ``studies'' were not subject to 
peer review.
    We also note a number of other deviations from the IOM's 
application of its risk assessment model (Ref. 28). Compared to the 
definition in the IOM report, CANTOX expanded the definition of the UL 
and narrowed the population to which it applies. As noted earlier, the 
IOM report defined the UL, in part, as ``the highest level of daily 
nutrient intake that is likely to pose no risk of adverse health 
effects to almost all individuals in the general population.'' The IOM 
report stated that the term ``tolerable'' was chosen ``because it 
connotes a level of intake that can, with high probability, be 
tolerated biologically by individuals; it does not imply acceptability 
of that level in any other sense.'' The IOM report also noted that 
``the UL is not intended to be a recommended level of intake'' (Ref. 28 
at pp. 3, 4, and 5). The IOM report also stated that ``the critical 
endpoint used to establish a UL is the adverse biological effect 
exhibiting the lowest NOAEL (for example, the most sensitive indicator 
of a nutrient or food toxicity). The derivation of a UL based on the 
most sensitive endpoint will ensure protection against all other 
adverse effects'' (Ref. 28 at p. 18). The IOM report also explained 
that, ``When possible, the UL is based on a NOAEL, which is the highest 
intake (or

[[Page 6809]]

experimental oral dose) of a nutrient at which no adverse effects have 
been observed in the individuals studied. This is identified for a 
specific circumstance in the hazard identification and dose-response 
assessment steps of the risk assessment'' (Ref. 28 at p. 10).
    Although CANTOX defined the UL as ``the maximum level of chronic 
daily intake of a substance judged unlikely to pose a risk to the most 
sensitive members of the health population,'' their UL determination 
was based upon the ``specified conditions of use,'' which includes 
label warnings that these products not be used by many in the general 
population (including those under 18 years, pregnant or lactating 
women, and persons with certain health conditions, including those most 
sensitive to the effects of these products, e.g., persons with 
hypertension and coronary artery disease). In contrast, the IOM concept 
of the UL is the highest level of intake likely to pose no risk of 
adverse health effects to almost all individuals in the general 
population. Thus, the CANTOX UL is less protective than the IOM UL 
because it removes from its risk assessment the members of the 
population who would be most at risk for adverse effects of dietary 
supplements containing ephedrine alkaloids.) (Ref. 93 at p. 5).
    It also appears that CANTOX deviated from the IOM model in its 
assessment of what constituted an ``adverse effect.'' Although the 
CANTOX report failed to define the endpoints (potential adverse 
effects) that were considered in the determination of a NOAEL, the 
report stated that ``the selection of 90 mg/day is an appropriate value 
for a NOAEL for ephedra in light of the evidence of no significant 
increases in frequency of adverse effects or changes in heart rate or 
blood pressure at or below this level leading to cardiac arrhythmias.'' 
Thus, it appears that CANTOX did not consider changes in heart rate or 
blood pressure to be ``adverse effects,'' although these biological 
effects can lead to serious adverse health consequences, such as 
arrhythmias and strokes. In addition, in discussing the Boozer et al. 
study, the CANTOX report described the statistically significant 4 mm 
Hg elevation in systolic blood pressure in the ephedra plus caffeine 
treated group as compared to the placebo group, as well as other self-
reported symptoms (dry mouth, heartburn and insomnia) in the treated 
group, as ``minimal side effects.'' This choice of terminology suggests 
that CANTOX did not consider the well-described pharmacological effects 
of ephedrine alkaloids to have potentially serious adverse health 
effects. This difference would affect the NOAEL, which, in turn, would 
lead to different UL determinations. We further address the 
definitional issue of adverse events versus side effects later in 
section V.B.6. of this document.
    We also note that CANTOX's stated study objective, ``to provide and 
justify a safe upper intake level for ephedrine alkaloids from ephedra 
used as a dietary supplement,'' appears to assume that such a safe dose 
exists. This assumption indicates a bias towards finding a safe dose, 
rather than an unbiased assessment of whether any safe dose exists.
    Finally, we discuss the inadequacies of the publications used by 
CANTOX to assess the safety of ephedrine alkaloids in section V.B.2 of 
this document. Whatever methods are employed, these deficiencies in the 
data used in CANTOX's analysis significantly undermine any conclusions 
reached in the CANTOX report.
    (Comment 33) Several comments objected that we did not consider 
animal studies using ephedrine alkaloids to evaluate the safety of 
ephedrine alkaloids as dietary ingredients, as several comments noted 
had been done in the CANTOX review. One comment stated that the results 
of the National Toxicology Program's long-term rodent studies on 
ephedrine showed that a lethal dose of ephedrine alkaloids for most 
animal species, translated into human consumption, was between 200 and 
400 25 mg tablets. A related comment referred to toxicity 
(LD50) studies comparing pharmaceutical ephedrine with ma 
huang in mice, emphasizing lesser toxicity of ma huang: The 
LD50 for ephedrine alkaloids from ma huang was 5300 mg/kg 
body weight versus 689 mg/kg for pharmaceutical ephedrine. A related 
point from this comment was that wild and domestic animals consume 
Ephedra shrubs and there are no reports of adverse effects in these 
animals. One comment included data from rat, mouse, and dog toxicity 
studies on a specific ephedrine alkaloid-containing dietary supplement. 
The results and their interpretation by consultants were offered as 
demonstrating a very low toxicity for the supplement. One comment 
stated that no animal study suggests that the ephedrine alkaloids would 
be harmful at human doses of 25 mg per serving. One comment stated that 
animal and laboratory testing may be informative on some issues but, in 
and of itself, cannot answer the human causation question.
    (Response) We recognize the value of animal studies in identifying 
or predicting the toxicological properties of substances for human 
exposure. In fact, animal studies do identify the sympathomimetic 
effects of ephedrine that underlie our concern. These would not be 
expected to lead to harm in healthy laboratory animals because these 
animals do not have coronary artery disease or other susceptibility to 
arrhythmias or congestive heart failure. An effect of elevated blood 
pressure, if large and sustained, might perhaps show effects in very 
large, long-term animal studies, but there is no reason to think that a 
modest effect, one that would increase hypertensive risk in humans but 
still lead to a low overall risk in any individual, would be detectable 
in animals. The animal data are, therefore, not at all reassuring. The 
discussion of the consumption of wild Ephedra species by wild and 
domestic animals contributes no relevant safety information, since 
these animals also lack pertinent human risk factors (coronary artery 
disease, heart failure, elevated blood pressure). Also, were these 
animals to have an adverse effect, there would be no way to identify 
it. However, we believe, as stated previously, that there is sufficient 
scientific evidence from multiple sources, including clinical trials 
and the published literature pertaining to use of ephedrine alkaloids 
in humans, to conclude that dietary supplements containing ephedrine 
alkaloids pose serious risks of illness or injury.
3. Comparison with Drug Products Containing Ephedrine Alkaloids
    (Comment 34) One comment asserted that our proposal to treat 
dietary supplements more restrictively than OTC drugs containing 
ephedrine and pseudoephedrine is in violation of the Administrative 
Procedure Act's prohibition on rulemaking that is arbitrary and 
capricious. According to the comment, OTC ephedrine and pseudoephedrine 
products contain higher doses of ephedrine alkaloids and therefore are 
potentially more dangerous than dietary supplements that contain these 
substances at lower levels.
    (Response) Our decision in this rulemaking to treat dietary 
supplements that contain ephedrine alkaloids differently from OTC drugs 
that contain ephedrine or pseudoephedrine is not arbitrary or 
capricious. Our decision is based on differences in the intended uses 
of these products, as well as differences in the scientific evidence 
available to support the risk-benefit ratio for the products. The risk-
benefit ratio is dependent on several factors, including the product's 
intended use, the product's benefits, if any, and the

[[Page 6810]]

availability of adequate measures to control risk.
    As discussed previously, dietary supplements containing ephedrine 
alkaloids present an unreasonable risk of illness or injury because 
their risks outweigh their benefits. Like dietary supplements 
containing ephedrine alkaloids, OTC drug products containing ephedrine 
or pseudoephedrine have risks related to these ingredients. However, 
unlike dietary supplements, such OTC drug products have demonstrated 
benefits in the treatment and mitigation of disease. Through the OTC 
drug review process, we have determined that drug products containing 
ephedrine are GRASE for OTC use as a bronchodilator for the temporary 
relief or symptomatic control of bronchial asthma (see Sec.Sec. 341.16 
and 341.76), and that drug products containing pseudoephedrine are 
GRASE for OTC use as a nasal decongestant for the temporary relief of 
nasal congestion due to the common cold or hay fever (allergic 
rhinitis) (See Sec.Sec. 341.20 and 341.80). Based on controlled 
clinical investigations (See Sec. 330.10(a)(4)(ii)), we have determined 
that the benefits associated with the use of OTC drug products 
containing ephedrine and pseudoephedrine for these disease indications 
outweigh the risks and justify the use of these products despite their 
risks. However, such uses for disease mitigation and treatment are 
beyond the scope of permissible dietary supplement uses.
    Moreover, we do not agree that dietary supplements containing 
ephedrine alkaloids are safer than OTC drugs containing ephedrine or 
pseudoephedrine based on the relative doses of ephedrine alkaloids in 
these products. We consider an OTC drug product's safety in the context 
of its conditions of use (See Sec. 330.10(a)(4)(i)). OTC drugs 
containing ephedrine and pseudoephedrine are marketed to persons with 
specific disease conditions or symptoms for temporary, episodic relief. 
In fact, OTC ephedrine bronchodilator drug products are required to 
bear a warning limiting the use of these products to persons who have 
been diagnosed with asthma by a doctor (See Sec. 341.76(c)(1)). 
Additionally, although drug products containing ephedrine and 
pseudoephedrine are permitted to be marketed OTC at specific doses, 
these doses have been determined based on the specific indications of 
these drugs. As previously discussed, the indications and benefits 
applicable to OTC drugs containing ephedrine and pseudoephedrine do not 
apply to dietary supplements. Thus, the safety of dietary supplements 
containing ephedrine alkaloids cannot be established merely by showing 
that the level of ephedrine alkaloids in these products falls within or 
under the dose ranges permitted for OTC drug products. Furthermore, 
these dietary supplements contain several ephedrine alkaloids, making 
it difficult to draw any conclusions about benefits from studies using 
OTC drug products that contain a single ephedrine alkaloid.
    (Comment 35) Several comments pointed out that we have concluded 
that the ephedrine levels permitted in OTC drugs are generally 
recognized as safe. Other comments maintained that the long-term 
marketing and favorable safety record of OTC drugs containing ephedrine 
alkaloids is evidence of the safety of dietary supplements containing 
ephedrine alkaloids. Several comments asserted that there is a lack of 
serious AERs for both traditional Asian herbal products and OTC 
ephedrine drugs with dosages based on FDA's monograph (less than or 
equal to 25 mg per serving and less than or equal to 150 mg in a 24-
hour period) and that these dosages are, thus, safe.
    One comment maintained that the nonserious events identified by 
RAND are consistent with the side effects of caffeine and OTC ephedrine 
listed in the OTC drug review and do not pose an unreasonable risk. 
Other comments referred to statements made during the 1996 FDA Food 
Advisory Committee that there are no serious adverse effects reported 
with drugs containing ephedrine alkaloids within the allowable dosage 
range and to a February 28, 2003 FDA press release relating to ephedra 
that stated there are fewer AERs linked to OTC ephedrine drug products 
than to dietary supplements containing ephedrine alkaloids.
    (Response) We do not agree that the safety of dietary supplements 
containing ephedrine alkaloids can be established by reference to the 
safety of OTC drug products containing ephedrine or pseudoephedrine, 
two ephedrine alkaloids currently included in OTC drug monographs.
    As discussed previously, all sympathomimetics may pose risks for 
adverse events even after a single dose. GRASE status does not mean 
that an OTC drug product may not cause adverse events. In fact, there 
have been adverse events reported to FDA concerning ephedrine- and 
pseudoephedrine-containing OTC drugs. There are also numerous adverse 
event reports for dietary supplements containing ephedrine alkaloids. 
The incidence and type of adverse event reports related to dietary 
supplements containing ephedrine alkaloids are discussed in section 
V.B.6 of this document, which also contains our discussion on the 
significance of these AERs in our determination of unreasonable risk.
    As part of our OTC drug review, we have determined that ephedrine 
and pseudoephedrine are GRASE OTC drug ingredients for certain 
indications. Ephedrine is GRASE for the temporary relief or symptomatic 
control of bronchial asthma (See Sec.Sec. 341.16 and 341.76). 
Pseudoephedrine is GRASE for the temporary relief of nasal congestion 
due to the common cold or hay fever (allergic rhinitis) (See Sec.Sec. 
341.20 and 341.80). OTC ephedrine and pseudoephedrine drug products 
have been studied in controlled trials that establish their safe and 
effective dose for specific disease indications (labeled uses) (41 FR 
38312 at 38371 and 38402 to 38403, September 9, 1976) (Refs. 97 and 
98). These OTC drug products provide health benefits when used by the 
population experiencing the particular disease. We note that these OTC 
drug products bear warnings that certain populations should not use 
them, and they are not risk free. However, we have determined that the 
demonstrated benefits for the labeled OTC drug uses outweigh their 
risks (See Sec. 330.10(a)(4)(iii)). The labeling of OTC ephedrine and 
pseudoephedrine drug products warns consumers not to use the products 
if they have heart disease, high blood pressure, thyroid disease, 
diabetes, or difficulty in urination due to an enlargement of the 
prostate gland unless directed by a doctor (Sec.Sec. 341.76(c)(2) and 
341.80(c)(1)(C)). In addition, OTC ephedrine bronchodilator drug 
products are labeled with a warning not to use the product unless a 
diagnosis of asthma has been made by a doctor (Sec. 341.76(c)(1)). 
Moreover, the labeling directs users not to continue to use ephedrine 
drug products but to seek medical assistance immediately if symptoms 
are not relieved within 1 hour or become worse (Sec. 341.76(c)(5)). As 
discussed in the response to comment 34 of this document, the benefits 
of ephedrine and pseudoephedrine drug products for disease claims are 
different from the benefits of dietary supplement products for 
nondisease claims, so it would be inappropriate to conclude based on 
OTC drug product information that these dietary supplements do not 
present an unreasonable risk. No data demonstrate that dietary 
supplements containing ephedrine alkaloids provide a meaningful health 
benefit to a particular

[[Page 6811]]

population for any specific use and for short periods of time, as is 
the case for OTC ephedrine or pseudoephedrine drug products. Therefore, 
we have determined that the risks presented by dietary supplements 
containing ephedrine alkaloids (including heart attack, stroke, and 
death) outweigh their benefits, and that these products are adulterated 
regardless of what warnings are included in their labeling. We note 
that dietary supplements containing ephedrine alkaloids may also 
present other, less serious risks listed in the required warnings for 
OTC drugs containing ephedrine and pseudoephedrine; however, because we 
are removing these dietary supplement products from the market based on 
their cardiovascular risks, we are not addressing these other risks in 
this rule.
    With regard to the comments that discussed safety data for OTC 
ephedrine bronchodilator drugs specifically, we note that the studies 
used to evaluate ephedrine for the treatment of asthma and those using 
ephedrine alkaloids for weight loss and other nondisease uses enrolled 
different populations and used different study designs, endpoints, and 
monitoring protocols. Therefore, comparisons across patient populations 
or indications (e.g., asthma treatment versus weight loss) for a risk 
benefit analysis is not justified. FDA's 1986 final rule finding 
ephedrine GRASE as a bronchodilator was based on the 1976 
recommendation of the Advisory Review Panel on OTC Cold, Cough, 
Allergy, Bronchodilator, and Antiasthmatic Drug Products (the Panel) 
(See 51 FR 35326, October 2, 1986 and 41 FR 38312 at 38370 to 38372, 
September 9, 1976). The Panel relied on data from studies conducted in 
1973 and 1975 (Refs. 97 and 98). These studies were designed to examine 
the efficacy of terbutaline as a bronchodilator. The patient population 
enrolled in these studies were not only clinically stable (i.e. normal 
electrocardiogram, blood pressure, and pulse) but also had no apparent 
history of adverse events related to treatment with other stimulant 
bronchodilators used at the time. These studies support the use of 
ephedrine for patients with asthma who are otherwise clinically stable 
(i.e. not found by a physician to have high blood pressure or other 
cardiovascular risk); however, they do not support the safety or 
efficacy of dietary supplements containing ephedrine alkaloids for 
weight loss or other nondisease uses.
    (Comment 36) Several comments asserted that it is misleading to 
compare the safety and efficacy of ephedra to OTC drugs because all 
drugs are toxic to some individuals and all products must be evaluated 
on the basis of their benefits relative to their risks. These comments 
expressed the view that dietary supplements containing ephedrine 
alkaloids have only limited benefit for weight loss over placebo and 
that this modest weight loss has never been shown to reduce the 
increased morbidity that is associated with obesity.
    (Response) We agree that dietary supplements containing ephedrine 
alkaloids and OTC drug products must be evaluated based on a comparison 
of their risks and benefits. It should be noted, however, that the 
evidentiary standards for evaluating these two categories of products 
are different. We have done a risk-benefit analysis for dietary 
supplements containing ephedrine alkaloids for weight loss, as well as 
other uses, and have discussed our analysis and conclusions regarding 
weight loss in section V.C.1 of this document.
    (Comment 37) Numerous comments asserted that herbal medicines, 
including ephedra, have a favorable safety record when compared to 
approved pharmaceuticals. Several comments cited the numbers of serious 
adverse events associated with approved pharmaceuticals, including 
deaths, among the U.S. population that are not due to medication 
errors. For example, various authorities estimate that more than 
100,000 deaths per annum are associated with approved pharmaceuticals 
(Refs. 99 and 100). One comment stated that the rate of severe adverse 
reactions to prescription drugs, without necessarily including misuse, 
ranks as the fourth to sixth leading cause of death in the United 
States (Ref. 100). The comment expressed the view that ephedrine 
alkaloids do not carry a significant or unreasonable risk of harm when 
compared to the high incidence of serious adverse effects with 
prescription drugs.
    (Response) While we agree that serious adverse events can occur 
with the use of prescription drugs, that fact does not change our 
determination that dietary supplements containing ephedrine alkaloids 
present an unreasonable risk. Prescription medications, although 
considered safe and effective for their labeled indications, are not 
free from all risks. However, the benefit of using prescription 
medications outweighs such risks for particular patients with 
particular disease conditions, in part because the risk is managed 
through the physician supervision required for the use of prescription 
medications. Although dietary supplements need not be free of risks to 
be lawfully marketed, the risks of using dietary supplements containing 
ephedrine alkaloids are not outweighed by any benefit. Moreover, it 
would not be surprising to see more AERs for prescription drugs than 
for dietary supplements. Healthcare professionals, who are aware of the 
drugs prescribed for their patients, are the primary source of drug 
AERs reported to us directly or through manufacturers. They may not be 
similarly aware of their patients' use of dietary supplements. In 
addition, there are no mandatory reporting requirements for dietary 
supplement manufacturers, unlike for prescription drug manufacturers. 
Finally, the comments and literature cited pertain to adverse events 
for all prescription drugs combined. This information has no meaningful 
bearing on whether dietary supplements containing ephedrine alkaloids 
present risks.
    (Comment 38) One comment contended that dietary supplements 
containing ephedrine alkaloids should be banned because we have already 
banned OTC drugs containing ephedrine in combination with caffeine. 
Numerous other comments stated that our November 18, 1983 (48 FR 
52513), prohibition of ephedrine alkaloids combined with caffeine and 
other stimulants (48 FR 52513) was due to such products' potential for 
abuse and misuse as illicit street drug alternatives and not because of 
safety issues. One comment stated that our proposal (60 FR 38643, July 
27, 1995) (July 1995 proposal) to amend the final monograph for OTC 
bronchodilator drug products to remove the ingredients ephedrine, 
ephedrine hydrochloride, ephedrine sulfate, and racephedrine 
hydrochloride and to classify these ingredients as not generally 
recognized as safe and effective for OTC use was proposed to restrict 
the OTC availability of ephedrine because of its illicit use as the 
primary precursor in the synthesis of the controlled substances 
methamphetamine and methcathinone. The comment stated that the July 
1995 proposal does not discuss the safety of the use of ephedrine and 
thus does not support our actions.
    (Response) We do not agree that our July 1995 proposal did not 
discuss the safety of OTC bronchodilator drug products containing 
ephedrine alkaloids (60 FR 38643 at 38644). In any event, comments 
about the basis and scope of our 1983 prohibition on ephedrine and 
caffeine combinations in OTC drug products and the 1995 ephedrine drug 
product proposal are not relevant to this rulemaking because we are not 
relying on those actions as a basis for the

[[Page 6812]]

removal of dietary supplements containing ephedrine alkaloids.
4. Abuse and Misuse
    (Comment 39) Many comments asserted that we must consider 
directions for use, warnings, and other labeling when making an 
assessment of significant or unreasonable risk. The comments stated 
that we cannot consider misuse or abuse of properly labeled dietary 
supplements. One comment urged that any evaluation of significant or 
unreasonable risk be based on the standards specified in the American 
Herbal Products Association's (AHPA) Ephedra Trade Recommendation, 
which recommends that dietary supplements containing ephedrine 
alkaloids be formulated to contain no more than 25 mg of ephedrine 
alkaloids per serving, that such products bear a warning statement and 
that directions for use limit consumption to 100 mg of ephedrine 
alkaloids per day (Ref. 101).
    (Response) We agree that directions for use, warnings, and other 
labeling must be considered when making an assessment of significant or 
unreasonable risk. Section 402(f)(1)(A) of the act provides that 
whether a dietary ingredient or dietary supplement presents a 
significant or unreasonable risk must be evaluated ``under conditions 
of use recommended or suggested in labeling,'' except that ordinary 
conditions of use may be considered if the labeling is silent on 
conditions of use. Thus, for purposes of the ``significant or 
unreasonable risk'' provision, unless no conditions of use are 
recommended or suggested in labeling, we must consider a dietary 
supplement's labeled use rather than its actual use. We do not agree, 
however, that our evaluation of significant or unreasonable risk should 
be based on the standards specified in AHPA's Ephedra Trade 
Recommendation (Ref. 101). These standards are voluntary 
recommendations by a trade association and are not universally 
followed. We must consider all dietary supplements containing ephedrine 
alkaloids, not just those formulated and labeled in accordance with the 
Ephedra Trade Recommendation. In this instance, we conclude that all 
dietary supplements containing ephedrine alkaloids present an 
unreasonable risk, regardless of whether they are formulated and 
labeled in accordance with the Ephedra Trade Recommendation, based on 
our evaluation of the totality of the evidence and a weighing of the 
risks and benefits of the products. As discussed in section VI.A of 
this document, the presence of a warning label or of directions 
recommending a limit on daily consumption of ephedrine alkaloids does 
not sufficiently reduce the risks of dietary supplements containing 
ephedrine alkaloids to allow them to continue to be marketed as 
currently labeled or under ordinary conditions of use, and the risks of 
these products outweigh their benefits regardless of labeling.
    (Comment 40) Several comments compared the effects of ephedra to 
other sympathomimetics such as cocaine or amphetamine. Several other 
comments stated that while ephedrine, PPA, and amphetamine are similar 
in chemical structure, they differ in physiological effect, and that 
amphetamines have much stronger reinforcing effects and a much higher 
liability for abuse than ephedrine. One comment stated that the 
subjective effects of ephedrine more closely resemble caffeine. Another 
comment stated that amphetamines do not have direct agonist properties, 
but promote release of neurotransmitters and inhibit their deactivation 
and reuptake. One comment from a manufacturer of a dietary supplement 
containing ephedrine alkaloids stated that its product label warns 
consumers not to take the product longer than 12 weeks because it can 
be habit forming and to take it longer runs the danger of ``getting 
hooked.''
    Several comments expressed the opinion that ephedrine alkaloid 
dependence is similar to amphetamine dependence, as are the 
psychological effects of abuse such as psychosis, paranoia, and the 
potential to cause mania in susceptible individuals. Comments from 
several individuals and the founder of a consumer advocacy Web site 
included anecdotal reports of individuals who reported dependence or 
apparent addiction associated with use of ephedrine and dietary 
supplements containing ephedrine alkaloids. Several other comments 
cited the German Commission E monograph's instructions to limit the use 
of ephedra preparations to short-term because of the danger of 
addiction. (The Commission E was a division of the German Federal 
Health Agency established in 1978 to evaluate the safety and efficacy 
of herbal medicines sold in Germany. It produced official monographs 
for botanicals and botanical formulations sold in German pharmacies.)
    (Response) We agree that ephedrine alkaloids and amphetamines share 
some pharmacological and physiological properties that may be 
associated with abuse and dependence. Psychostimulant effects that have 
been reported with sympathomimetic agents include drug tolerance, 
dependence, or addiction, although these psychostimulant effects are 
better recognized for cocaine and amphetamines (Refs. 102 and 103 of 
English abstract), Ephedrine alkaloids exhibit physiological effects 
common to the amphetamines, but differ in the relative intensity of 
these effects. We agree that amphetamines and cocaine have been shown 
to have much greater reinforcing effects and higher liability for abuse 
than products containing ephedrine alkaloids, but also agree that the 
development of dependence from the use of ephedrine alkaloids has been 
noted with both pharmaceutical and botanical products (Refs. 104, 105, 
and 106). The greater possibility of dependence and abuse of 
amphetamine-containing and cocaine-containing drug products marketed in 
the United States is recognized by the placement of these substances in 
Schedule II of the Controlled Substances Act (CSA). Ephedrine-
containing drug products are not scheduled under the CSA; however, 
ephedrine, its salts, optical isomers, and salts of optical isomers are 
List I chemicals under the CSA (See 21 U.S.C. 802(34)) because they are 
chemical precursors of methamphetamine (Schedule II) and are used in 
its illicit manufacture. As List I chemicals, these substances are 
subject to various Drug Enforcement Administration (DEA) requirements, 
including recordkeeping, reporting, and sale behind the counter (See 21 
CFR 1310.03 through 1310.07). While we are concerned about the 
potential for abuse, we did not rely on evidence of abuse or dependence 
to make our determination under section 402(f)(1)(A) of the act.
    (Comment 41) Some comments advocated use of ephedra as an 
alternative to more dangerous street drugs. They postulated that 
banning dietary supplements containing ephedrine alkaloids would push 
those products underground or drive consumers to seek out more 
dangerous drugs for stimulant effects.
    (Response) No data were submitted with these comments to support 
their conclusions. We have no information regarding the extent of use 
of ephedra, or dietary supplements containing ephedrine alkaloids, as 
an alternative to more dangerous street drugs, nor do we have any 
information about whether users of ephedrine alkaloids would be likely 
to use other substances were ephedra to become unavailable. Regardless, 
such information would not affect the determination we have made that 
dietary supplements containing ephedrine alkaloids present an 
unreasonable risk.

[[Page 6813]]

    (Comment 42) Several comments stated that we cannot stop the abuse 
of substances by regulation. Some comments cited tobacco and alcohol as 
examples. Another comment stated that if we regulated products that 
caused injury because of their potential for abuse, then common 
household products, such as aerosol paint, would be banned.
    (Response) Our conclusion that dietary supplements containing 
ephedrine alkaloids present an unreasonable risk is based not on abuse 
or misuse but rather on evidence supporting the presence of risks under 
conditions of use recommended or suggested in the labeling, or if the 
labeling is silent, under ordinary conditions of use. Abuse or misuse 
of other products is not relevant to our determination that dietary 
supplements containing ephedrine alkaloids present an unreasonable 
risk.
    (Comment 43) Several comments stated the opinion that we do not 
appear to distinguish between dietary supplements containing ephedrine 
alkaloids marketed for weight loss or energy from those products 
marketed as alternatives to illicit street drugs or as ``legal highs.''
    (Response) We do not agree with these comments. Beginning with the 
June 1997 proposal on dietary supplements containing ephedrine 
alkaloids, we have repeatedly warned industry and the public that we do 
not consider products marketed as street drug alternatives to be 
dietary supplements because they are intended for recreational purposes 
to affect psychological states (e.g., to get high) and are not intended 
to be used to augment the diet or to promote health (62 FR 30678 at 
30699 and 306700). Since 1997, we have issued a series of warning 
letters to firms for marketing ephedrine alkaloid-containing products 
as street drug alternatives and warned consumers not to purchase or 
consume such products. In March 2000, we issued a guidance document 
stating that street drug alternatives are unapproved and misbranded 
drugs that are subject to regulatory action, including seizure and 
injunction (available at http://www.fda.gov/cder/guidance/3602fnl.pdf). 
Our position was that street drug alternatives are drugs, not dietary 
supplements, was upheld in United States v. Undetermined Quantities of 
Articles of Drug (Street Drug Alternatives), 145 F. Supp. 2d 692 (D. 
Md. 2001). That case involved a seizure of numerous street drug 
alternatives marketed as dietary supplements, including four products 
containing botanical ephedrine alkaloids. In January 2003, we witnessed 
the voluntary destruction of $4 million worth of illegally marketed 
street drug alternative products containing ephedrine alkaloids. We 
continue to address the street drug alternatives with appropriate 
regulatory actions. We have determined that the appropriate regulatory 
action for dietary supplements containing ephedrine alkaloids--i.e., 
products marketed for weight loss, athletic performance, energy 
enhancement, or other nonstreet drug alternative uses--is to issue a 
final rule finding that these products present an unreasonable risk of 
illness or injury.
5. Traditional Asian Medicine
    (Comment 44) Many comments stated that the use of ephedrine 
alkaloids in dietary supplements is safe based on its traditional use 
in Asian medicine for thousands of years. Several comments asserted 
that few or no adverse effects have been recorded with the use of 
Ephedra in traditional Asian medicine. Numerous other comments, 
including those by traditional Asian medicine practitioners, disagreed 
with these comments about dietary supplements, highlighting the 
differences in the products themselves and how they are used from what 
is used in traditional medicine.
    Several comments suggested that the raw Ephedra and Ephedra 
extracts used in traditional Asian medicine formulae differ in potency, 
toxicity, pharmacokinetics, and pharmacological and physiological 
effects from many dietary supplements containing ephedrine alkaloids 
and, therefore, that these formulations should be considered distinct 
in scientific, medical, and regulatory contexts. Comments stated that 
``Ephedra'' properly refers to dried aerial parts of medicinal plants, 
or crude extracts thereof, not to isolated alkaloidal constituents. 
Several comments further distinguished the various products containing 
Ephedra as follows: Herb and extracts of raw herb of medicinal Ephedra 
plants containing naturally occurring alkaloids and other compounds 
without further manipulation, concentration, or adulteration; Ephedra 
extracts that are concentrated, manipulated, or adulterated such that 
naturally occurring proportions and/or quantities of ephedrine 
alkaloids are altered; products containing ephedrine alkaloids combined 
with other agents such as caffeine, caffeine-containing herbs, 
salicylate-containing herbs, synephrine, and other substances; and 
traditional Asian herbal medicinal formulae.
    Several comments asserted that traditional Asian medicine Ephedra 
formulae often deliver lower amounts of ephedrine alkaloids compared to 
other types of ephedrine alkaloid-containing products and that 
traditional formulae rarely contain more than 15 percent Ephedra in the 
herb mixture. Comments also asserted that Ephedra in traditional 
formulae is usually combined with other botanicals that typically 
modify Ephedra's inherent stimulant effects. Another comment attributed 
the relative safety of Ephedra to the mixture of ephedrine alkaloid 
isomers not present in purified or synthetic alkaloids. One comment 
suggested that the established therapeutic dose range of Ephedra sinica 
in herbal medicine formulae is 60 to 90 mg total alkaloids per day 
(adults), which falls within the dosage range established for OTC 
ephedrine/pseudoephedrine-containing drugs (150 mg and 240 mg alkaloids 
daily, respectively), and the recommendations of the Germany Commission 
E (maximum daily Ephedra alkaloid dose of 300 mg daily). Other comments 
asserted that infusions or teas of Ephedra are effective in relieving 
respiratory symptoms but have fewer side effects and are safer than 
formulations containing isolated or synthetic ephedrine alkaloids or 
prescription drugs. Another comment stated that supplements in a liquid 
tea form greatly reduce the risk of excess acute consumption by the 
public.
    In contrast, several other comments stated that the presence of 
varying amounts, proportions, and chemical configurations of ephedrine 
alkaloids in crude Ephedra and prepared Ephedra extracts, as well as 
the presence of unknown compounds, leads to uncertainty as to dose, 
purity, and composition and to a greater risk of adverse effects. 
Comments noted that this variability is not an issue for synthetic or 
pure isolated ephedrine alkaloids.
    Numerous comments, including those by traditional Asian medicine 
practitioners, also noted differences in how the products are used. 
Several comments stated that most traditional Asian uses of Ephedra are 
the same as the indications for OTC ephedrine and pseudoephedrine drugs 
(e.g., short-term use to improve respiratory function) and that few if 
any adverse effects have been recorded. Several comments stated that 
use of Ephedra (ma huang) for weight control or for its stimulating 
effects, for more than a short period of time, in combination with 
caffeine and other botanical stimulants, and without the supervision of 
a health care provider, is irresponsible and dangerous. A number of 
traditional Asian medicine practitioners maintained that many

[[Page 6814]]

consumers experienced adverse effects because of this improper use, 
over-dosage, or conflict with their illnesses.
    Because of these differences, many practitioners of traditional 
Asian medicine commented that they support our June 1997 proposal 
except to the extent that it would restrict their use of Ephedra in 
traditional Asian medicine. Several comments asserted that since most 
serious adverse effects involve use of ephedrine alkaloids and not 
whole herb or whole herb extracts of Ephedra, any rule must exempt 
whole herb Ephedra or whole herb Ephedra extracts that contain no added 
ephedrine alkaloids. Furthermore, ephedrine alkaloid-free species of 
Ephedra should also be exempted.
    Numerous comments asserted that because traditional Asian herbal 
products are prescribed by appropriate practitioners (licensed, 
certified, and registered acupuncturists, herbalists, and naturopathic 
physicians) and because these products are not associated with serious 
adverse effects, the products do not appear to constitute a public 
health risk and their use should not be prohibited. Many traditional 
Asian medicine practitioners stated that Ephedra is an essential 
medicine and requested an exemption from the final rule for use of 
Ephedra by traditional Asian medicine practitioners and acupuncturists. 
A few comments asserted that Ephedra should not be used commercially, 
but be restricted to professional use, to be dispensed by licensed 
health care professionals trained in the appropriate use of traditional 
Asian medicine.
    (Response) This final rule does not affect the use of Ephedra 
preparations in traditional Asian medicine, although we considered the 
comments' views and information on the use of Ephedra in traditional 
Asian medicine in the context of their possible relevance to the risks 
of dietary supplements containing ephedrine alkaloids. This rule 
applies only to products regulated as dietary supplements (See 62 FR 
30678 at 30691). Traditional Asian medicine practitioners do not 
typically use products marketed as dietary supplements.
    With respect to the absence of adverse effects recorded with the 
use of traditional Asian medicine, as we stated in the June 1997 
proposal, we are not aware of any systematic collection of data related 
to adverse effects occurring in individuals treated with Ephedra in 
traditional Asian medicine. The absence of recorded adverse events with 
the use of Ephedra, therefore, may be related to the lack of a 
mechanism for reporting. Under these circumstances, there are no data 
to evaluate. We note that the potential for adverse effects resulting 
from the traditional Asian use of Ephedra is implied in several 
reference texts that list precautions and contraindications for the use 
of the botanical Ephedra in traditional Asian medicine preparations 
(Refs. 3, 107, and 108). Moreover, even if we could say that the 
absence of recorded adverse events with the use of Ephedra in 
traditional Asian medicine was due to its safety for that use rather 
than due to a lack of mechanism for reporting, the history of use of 
Ephedra in traditional Asian medicine primarily for the treatment or 
mitigation of respiratory illness cannot provide assurance about the 
safety of dietary supplements containing ephedrine alkaloids for other 
uses.
6. Adverse Events
    AERs involving drugs include those submitted to us voluntarily by 
consumers or healthcare professionals and those submitted by 
manufacturers who are required to report them to us. However, there is 
no required reporting of AERs to us for dietary supplements, including 
those containing ephedrine alkaloids. Depending on other information we 
may have about the event or about the suspect product, AERs can be hard 
to interpret. AERs may raise concerns about a product, as well as 
buttress a finding that a particular dietary supplement represents an 
unreasonable risk based on other types of evidence. Some AERs can be 
reasonably persuasive on their own. For example, individual cases of 
adverse events where dechallenge (discontinued use) and rechallenge 
(restarting use) have been linked to the abatement and recurrence of 
the events, strongly support the association between exposure to the 
product and occurrence of the adverse event. FDA, and others, have 
reviewed and analyzed the AERs in depth to add to the body of evidence 
and to ensure that all relevant evidence is considered (Refs. 109 
through 115). Despite the limitations of such reports, a detailed 
review of the AERs submitted to us for dietary supplements containing 
ephedrine alkaloids and comparison of those AERs to scientific data 
about the pharmacology of these substances establishes that the AERs 
are consistent with the known and expected pharmacological effects of 
these products considered (Refs. 109, 115, and 116).
    In the preamble to the June 1997 proposal, we stated that there 
were more than 800 reports of illnesses and injuries associated with 
the use of dietary supplements containing ephedrine alkaloids. Since 
that time, we have received more than 2,200 additional AERs submitted 
directly to us plus approximately 16,000 reports from call records 
submitted by Metabolife International, one of the largest distributors 
of dietary supplements containing ephedrine alkaloids. These records 
have been placed in the record for this rulemaking in redacted form.
    A Congressional subcommittee minority report (Ref. 117), posted at 
http://www.house.gov/reform/min/pdfs/
pdf_inves/pdf_dietary_ephedra
_metabolife_rep.pdf \4\ noted that the call records from Metabolife 
International contain nearly 2,000 reports of significant AERs for its 
products, including 3 deaths, 20 heart attacks, 24 strokes, 40 
seizures, 465 episodes of chest pain, and 966 reports of heart rhythm 
disturbances. In addition to these cardiac and neurological events, 
psychiatric symptoms were also reported. These reports include 46 
reports of hospitalization following use of their products, and 82 
additional reports of emergency room care. The report stated that in 
more than 90 percent of the most serious AERs-- stroke, heart attack, 
seizure, and psychosis--where dosage information is documented in the 
call record, the consumer had followed the manufacturer's dosage 
recommendations. It also stated that among those most significant 
adverse event reports for which age was noted, 50 percent of the 
consumers were under 35 and many of the consumers were reported as 
being in good health with no prior medical problems. Despite the 
limited information provided in Metabolife International's call 
records, we note that these types of adverse events reported are 
consistent with the scientifically documented effects and potential 
risks of ephedrine alkaloids in those cases where appropriate 
information was available to make a medical evaluation of the reported 
event.
---------------------------------------------------------------------------

    \4\ FDA has verified the Web site address, but FDA is not 
responsible for any subsequent changes to the nonFDA Web sites after 
this document publishes in the Federal Register.
---------------------------------------------------------------------------

    (Comment 45) Many comments criticized our system for collecting and 
evaluating adverse events and our use of AERs. A number of comments 
criticized the reporting system, stating that many of the received 
reports were insufficiently documented and lacked critical information 
necessary for appropriate evaluation. Other comments stated that the 
reports were anecdotal

[[Page 6815]]

and that no scientific standards were used in their evaluation.
    Several comments stated that our attempt to rely on AERs for 
attributing adverse events to dietary supplements containing ephedrine 
alkaloids is in conflict with established scientific principles and FDA 
policy. The comments cited the criticism of our reliance on AER in the 
July 1999 GAO Report, our bases for regulation of Yellow No. 5 which 
included AERs and multiple clinical studies, and the opinion that our 
AER review system was biased and lacked scientific rigor.
    Several comments stated that our methods of data collection might 
have affected the integrity of the data. The comments explained that we 
included in the database AERs that had not been verified. Many of these 
comments also stated that adverse events were frequently reported by 
family members and FDA officials rather than by physicians, health care 
facilities, and dietary supplement manufacturers. Some comments stated 
that certain products that did not contain ephedrine alkaloids were 
reported to be associated with adverse events. Several comments 
expressed the opinion that the AER database must be corrected to remove 
AERs that relate to products that do not contain ephedrine alkaloids 
prior to any rulemaking.
    (Response) Because there is no mandatory requirement for submission 
of adverse event reports involving foods (including dietary 
supplements) to us, we rely on voluntary adverse event reporting from 
consumers, physicians and other health care professionals, product 
manufacturers, poison control centers, and State health agencies as a 
monitoring tool in our identification of potentially serious public 
health concerns that may be associated with a particular ingredient, 
product, or type of product. As with other passive surveillance 
systems, we acknowledge that voluntarily submitted adverse event 
reports do not always include adequate descriptions of the event and 
important elements of medical history, such as preexisting illness or 
other therapy. Our concerns about the risks of dietary supplements 
containing ephedrine alkaloids are based primarily on the known 
pharmacological effects of sympathomimetics and clinical studies using 
botanical and/or synthetic ephedrine alkaloids. Based on these 
pharmacological effects, we have identified a likelihood of potentially 
fatal arrhythmias, increased mortality in heart failure, and an 
increased rate of the consequences of elevated blood pressure, such as 
heart attack, stroke, and death. All of these events have been reported 
to be associated with consumption of dietary supplements containing 
ephedrine alkaloids. Because these events also occur spontaneously, 
specific occurrences of the events generally cannot be definitively 
attributed to dietary supplements containing ephedrine alkaloids, 
although they are compatible with the expected effects of these 
products. The AERs were, thus, only one component of our evaluation, 
which primarily relied on review of the best available scientific 
literature, such as peer-reviewed controlled clinical trials. The AERs 
are consistent with events expected from ephedrine alkaloids based on 
known pharmacological effects and other evidence in the scientific 
literature, and the AERs support our findings concerning the risks of 
dietary supplements containing ephedrine alkaloids.
    a. Definitional issues.
    (Comment 46) Some comments argued that only ``life-threatening'' 
adverse events should have been considered as the basis for the 
rulemaking. Another comment pointed out that a ``serious event'' is 
described in FDA's publication entitled ``Clinical Impact of Adverse 
Event Reporting'' (Ref. 32) as an event that is fatal, life-
threatening, permanently/significantly disabling, requires or prolongs 
hospitalization, causes a congenital anomaly, or requires intervention 
to prevent permanent impairment or damage. The comment stated that any 
event that fails to meet any of these criteria must then be nonserious, 
reasonable, or insignificant. The comment also pointed out that an 
``adverse effect'' is an unwanted effect and does not necessarily imply 
``serious.'' The comment further stated that we should define key 
terms, including ``serious,'' ``unreasonable,'' ``significant,'' 
``adverse effect,'' and ``side effect.''
    Several comments also noted that the vast majority of complaints 
received by Metabolife International were mild and common. As such, one 
comment stated that some of the complaints were more accurately termed 
``side effects,'' not ``adverse events.'' One Metabolife International 
consultant who reviewed the call records noted that there is no FDA 
guidance to define ``significant effect.''
    (Response) We do not agree that we should consider only ``serious'' 
or ``life-threatening'' adverse events in our evaluation of AERs for 
dietary supplements containing ephedrine alkaloids. In considering 
reports of adverse effects of ephedra, we have focused on the reports 
themselves and their implications, not how they were designated. Thus, 
a report of tachycardia, not necessarily serious in itself, indicates a 
sympathomimetic response that in some patients could be dangerous. 
Marked increases in blood pressure would have similar implications and 
could suggest greater sensitivity to sympathomimetic effects in 
particular individuals. Reports of serious events like stroke, death or 
ventricular tachycardia are important, of course, but as noted earlier, 
can be difficult to interpret outside of a controlled trial or 
epidemiologic investigation. Concerns about ephedra arise principally 
because it has effects known to put particular individuals at risk 
(those with coronary artery disease or heart failure) or to pose a risk 
to any individual with continued use (increased blood pressure). 
Nonserious events that suggest sympathomimetic effects of ephedra are 
therefore important and need evaluation.
    There is no real distinction between side effects and adverse 
effects. In either case, they are unwanted effects of the product. The 
description of the reported event is what is critical. Although we 
agree that the term ``adverse effect'' means there is an unwanted 
effect and does not necessarily imply that the event is serious, that 
does not mean it is insignificant. Such effects could be indicative of 
more serious cardiovascular risks if use of the product is continued. 
When considered with the scientific literature and other data, the less 
clinically significant effects may provide evidence that the use of a 
dietary supplement or dietary ingredient presents a significant or 
unreasonable risk of illness or injury.
    In the case of dietary supplements containing ephedrine alkaloids, 
our evaluation indicates that serious adverse cardiovascular effects 
(e.g., heart attack, stroke, worsened heart failure) can be expected to 
occur with the use of these products by the general population. Such 
events are relevant even if they may be expected to occur because they 
are known to be related to a substance, or combination of substances, 
contained in the product. Under section 402(f)(1)(A) of the act, a 
dietary supplement is adulterated if it presents a significant or 
unreasonable risk of illness or injury based on the conditions of use 
in its labeling (or under ordinary conditions of use if the labeling is 
silent). Therefore, if the labeled use of a dietary supplement 
containing ephedrine alkaloids would be expected to result in a risk of 
illness or injury, we must consider that risk in evaluating whether the 
dietary supplement is adulterated. For these reasons, we

[[Page 6816]]

considered all types of adverse events associated with the use of 
dietary supplements containing ephedrine alkaloids, even those that 
would not be considered ``serious'' or ``life-threatening.''
    (Comment 47) Some comments stated that the AERs were anecdotal and 
by their nature do not allow for statistical evaluation. Other comments 
stated that AERs cannot establish a causal relationship between ephedra 
use and adverse events. Some comments cited the RAND report as support 
for the view that a causal relationship has not been shown.
    Many comments stated that, without a control group, it is 
impossible to predict the number of persons who could experience the 
same type of adverse events that occur in the population not exposed to 
the product. Several comments argued that we may be detecting 
coincidental adverse events, which could have occurred whether or not 
consumers used an ephedrine alkaloid-containing dietary supplement. 
Many comments also stated, and pointed out that we have stated, that 
AERs cannot be used to calculate incidence rates of adverse events 
(i.e., the expected rate of adverse events occurring in the population 
using a product) because the actual number of persons exposed to the 
product is unknown, as is the actual number of adverse events that 
occur with use of these products.
    (Response) As noted in the comments, the rate of occurrence of 
serious adverse events associated with a particular product or 
substance cannot be calculated based simply on the number of adverse 
events reported. Furthermore, we agree that the RAND report did not 
conclude that a causal relationship between ephedra and the reported 
adverse events had been shown. Despite the limitations of AERs, 
however, they can be of value in an evaluation of whether a dietary 
supplement presents a significant or unreasonable risk. Such reports 
can be important as signals of potential problems. Moreover, they can 
be more or less persuasive as to the strength of association between 
exposure to a product and occurrence of an event, depending, in part, 
on how likely the event is in the general population in the absence of 
the product. Thus, spontaneous reports have repeatedly signaled the 
ability of drugs to cause hepatic injury (e.g., bromfenac, 
troglitizone) because the events seen were rarely witnessed in the 
absence of hepatotoxic drug or viral illness (which could be ruled 
out). Similarly, spontaneous reports have shown drug-caused torsade de 
pointes-type arrhythmias, which are also rare in the population. For 
more common events (e.g., stroke, heart attack, headache), single 
reports may be harder to interpret. As previously discussed, the AERs 
for dietary supplements containing ephedrine alkaloids are consistent 
with events expected based on the scientific evidence, and the AERs 
support our findings.
    (Comment 48) One comment urged us to disregard an e-mail memorandum 
from Dr. Paul Shekelle (Ref. 118) of the RAND Corp. that responds to 
our questions about the level of scientific proof that supports a 
causal relationship between the use of ephedrine-containing products 
and serious adverse events. The comment maintained that the opinions 
expressed in the e-mail are speculative, not objective, and not 
consistent with the peer-reviewed findings of the RAND report. The 
comment expressed concerns that we and others will interpret the e-mail 
as an extension or interpretation of the RAND report.
    (Response) We are not treating the e-mail by Dr. Shekelle as an 
extension or interpretation of the RAND report. In seeking information 
from Dr. Shekelle, we were attempting to clarify the basis for RAND's 
conclusion regarding evidence of a causal relationship between dietary 
supplements containing ephedrine alkaloids and serious adverse events. 
We do not consider the Shekelle e-mail and Dr. Shekelle's subsequent 
publication (Ref. 119) as influencing the validity or interpretation of 
the RAND report, which is the document on which we rely.
    (Comment 49) Several comments objected that we did not consider 
``denominator data'' in our evaluation. Several comments stated that 
when the number of AERs we received is compared to the number of units 
sold and the population of users, the incidence of injury is 
insignificant or below the threshold for spontaneous illness (e.g., the 
incidence of an adverse event in the general population) and that the 
level of risk is acceptable. Several related comments argued that if we 
made a statistical comparison of the number of AERs to the number of 
servings used, we could find the number of AERs to be statistically 
insignificant. Several comments made such a statistical comparison. For 
example, one comment estimated the annual number of servings of dietary 
supplements containing ephedrine alkaloids based on its own sales 
figures and an estimate of their share of the market, and concluded 
that the 800 AERs represent one adverse event occurring with every 8 
million servings. The comments concluded that if the AER rate is 
statistically insignificant, the risk would be considered to be 
``insignificant'' under the act.
    Several comments requested that we consider industry evidence of 
the safe use of dietary supplements containing ephedrine alkaloids. 
Several of these comments were from manufacturers and distributors of 
dietary supplements containing ephedrine alkaloids that discussed the 
AERs their companies had received. One comment stated that the number 
of serious adverse events that the company received was statistically 
insignificant. Other manufacturers and distributors claimed that they 
had not received reports of adverse events related to the use of their 
dietary supplements containing ephedrine alkaloids when the products 
were used according to labeled directions or that lawsuits had not been 
filed against them. Comments from several dietary supplement trade 
groups or industry committees submitted survey information about the 
number of users of particular products or the number of units sold for 
particular products and the number of adverse events that were reported 
during the survey. These comments indicated that there were no or few 
adverse events (and these were mostly of a minor nature) in contrast to 
the millions of doses sold.
    Many comments noted the experience of firms with respect to the 
number of complaints or lawsuits they had received on products 
containing particular amounts of ephedrine alkaloids, sometimes in 
conjunction with particular amounts of caffeine, and labeled for use 
for various levels of time. Some of these comments included information 
on the amount of product sold or the number of people consuming the 
product in a specified time period.
    Several comments suggested that the number of adverse events 
estimated from the AERs is inconsistent with international data. For 
example, one comment noted that the Committee on Safety of Medicine 
(U.K.) indicated that there were only 22 reported adverse events on a 
product sold in the U.K. that contains a mixture of ephedrine alkaloids 
and caffeine in the 40 years or more that the product has been 
available. Similarly, some comments noted that Danish investigators 
estimated that 9.6 million doses of a product containing a combination 
of ephedrine and caffeine had been sold in Denmark in 1991 and 1992 and 
that only 86 reportable adverse events, defined as reactions which 
necessitated stopping the therapy, had been reported to the authorities 
during that time, despite relatively ``high dosage levels''.

[[Page 6817]]

    (Response) We are not persuaded that the lack, or limited numbers, 
of adverse events reported to a limited subset of dietary supplement 
manufacturers and distributors demonstrates that the use of dietary 
supplements containing ephedrine alkaloids is safe. In contrast to the 
absence or low number of AERs described in some of the comments, we 
have received a total of more than 18,000 AERs directly, through 
dietary supplement firms, and from other sources. The AERs and 
international data discussed by the manufacturers and distributors in 
their comments are consistent with other adverse event reports we have 
received. We note that the Danish product referred to by some comments 
has been withdrawn from the market for safety reasons, including 
serious adverse event reports documenting cardiovascular and nervous 
system effects (Refs. 120 and 121).
    There is little doubt that dietary supplement adverse events are 
underreported (Ref. 20). There is no requirement that manufacturers of 
dietary supplements report such events to FDA. Moreover, the usual 
reporters of AERs, physicians, are often unaware of the events 
themselves or the person's history of dietary supplement use. We 
therefore agree with the comments that the number of AERs reported to 
us cannot be used to calculate incidence rates. To calculate the 
incidence rate of an adverse event in the general population or in a 
subgroup of the general population, both numerator (i.e., the number of 
times a specific adverse event occurred with the use of a particular 
product over a given time period) and denominator (i.e., the total 
number of persons using the product over the same time period) data are 
needed. For reasons described previously, the adverse events that are 
actually reported are likely only a small fraction of the actual number 
of adverse events that occur with the use of these products. In 
addition, we have no reliable data on the use of these products by the 
general population or subgroups of the population. We could not 
evaluate the information from industry surveys on the number of people 
who use dietary supplements containing ephedrine alkaloids or the 
number of units of these products sold because this information was in 
summary form only (e.g., the raw data were not submitted). Therefore, 
we do not know the actual number of persons who have used the product. 
In addition, because we do not have reliable information on the actual 
number of adverse events occurring with these products and on the size 
of the population exposed to dietary supplements containing ephedrine 
alkaloids, we cannot calculate the rate of adverse events occurring in 
the population using these products (i.e., incidence rate). Although we 
have done rough estimates for the purpose of calculating a potential 
economic impact, these estimates cannot be used to determine the 
precise incidence rates of adverse events for dietary supplements 
containing ephedrine alkaloids. However, we do not believe it is 
necessary to calculate the incidence rate to determine that dietary 
supplements containing ephedrine alkaloids present an unreasonable 
risk. Such a determination does not require us to find actual harm, 
only that a product's risk of illness or injury outweighs its benefits 
in light of the claims and directions for use in the product's labeling 
or, if the labeling is silent, under ordinary conditions of use.
    b. Reporting issues, including underreporting.
    (Comment 50) Although many comments agreed that the adverse events 
for dietary supplements containing ephedrine alkaloids were 
underreported, a number of comments disagreed with our estimates in the 
June 1997 proposal. Some comments believed that adverse events were 
less underreported than we estimated, while others thought they were 
more underreported. One manufacturer stated that it does not report the 
complaints it receives to us but rather keeps them for its own records.
    (Response) As discussed in the response to comment 49 of this 
document, we continue to believe that adverse events are underreported 
due to the voluntary nature of the adverse event reporting system for 
dietary supplements and other factors. The manufacturer comment 
confirms that at least some firms in the dietary supplement industry 
receive AERs that they do not share with us. We commissioned a study 
that estimated that adverse events reported to us represent less than 1 
percent of all of the adverse events associated with dietary 
supplements (Ref. 122). Our preliminary evaluation of data purchased 
from the American Association of Poison Control Centers, covering the 
years 1997 through 1999, indicated more adverse events than we had 
received for the same years (Ref. 123). In addition, the Office of the 
Inspector General of HHS determined that the number of dietary 
supplement adverse event reports we received was significantly less 
than the number of dietary supplement adverse event reports received by 
Poison Control Centers (Ref. 20 at p. 9).
    In section VIII.A.5.a.i, we discuss in detail how we estimated 
rates of adverse event reporting for purposes of our impact analysis 
for this final rule.
    (Comment 51) One comment stated that, despite underreporting, 
incomplete reports, and inadequate staff, there is no credible evidence 
that our reporting system makes errors in detection of adverse event 
signals. The comment asserted the validity of an association between 
AERs and risks presented by ephedrine alkaloids. The comment argued 
that this conclusion is confirmed by the known pharmacology of 
ephedrine alkaloids and the types of reports seen in ephedrine clinical 
trials and with drugs that have a similar pharmacological action. The 
comment noted that 26 percent of the reports over a four-year period 
documented dechallenge and 4 percent documented positive rechallenge, 
providing additional evidence supporting causation.
    (Response) We agree that our spontaneous reporting system detected 
the potential health risks associated with dietary supplement products 
containing ephedrine alkaloids and that these health risks are 
consistent with those documented in the scientific literature and with 
the known pharmacology of these products. As stated in the July 1999 
GAO report entitled ``Uncertainties in Analyses Underlying FDA's 
Proposed Rule on Ephedrine Alkaloids'' (Ref. 124), AERs surveillance 
can be important as an early alert to potential problems.
    In considering the comments that disputed our estimates of adverse 
event reporting rates, it is important to note that we are not relying 
on the number of AERs for dietary supplements containing ephedrine 
alkaloids to demonstrate quantitatively that these products present an 
unreasonable risk. Rather, we are relying on the AERs as supportive 
evidence of the risks. Although the fact that we received many AERs for 
these products is relevant, an exact count of the number of AERs 
associated with consumption of dietary supplements containing ephedrine 
alkaloids is not necessary to our determination that these products 
present an unreasonable risk.
     c. Interpretation of AERs as supporting the existence of public 
health risks.
    (Comment 52) Several comments stated that the number of AERs does 
not raise a public health concern. One comment asserted that AERs with 
appropriate use of ephedra are rare. Other comments stated that there 
is no

[[Page 6818]]

association between the use of dietary supplements containing ephedrine 
alkaloids and serious adverse events when used with appropriate 
dosages, including the American Herbal Products Association (AHPA) 
trade recommendations. One comment noted that some of the AERs appear 
to be related to high amounts of ephedrine (i.e., in excess of 500 mg/
day) and that the relationship of intake to adverse events with the use 
of lower amounts consumed is unknown.
    (Response) We disagree with these comments. Public health concerns 
were initially raised by the number of AERs following consumption of 
dietary supplements containing, or suspected to contain, ephedrine 
alkaloids in comparison to the number of AERs for all other dietary 
supplements; the type of adverse event (e.g. cardiovascular system and 
nervous system effects); and the severity of the adverse events 
associated with the use of these products. The type, severity, and 
number of adverse events reported to us prompted us to investigate 
further. In many of these AERs, including those designated as ``most 
significant'' in the Congressional minority report (Ref. 117), the 
dietary supplement products were consumed as directed on the 
manufacturer's label. Although we do not endorse any current trade 
recommendations for the use of dietary supplements containing ephedrine 
alkaloids, we note that in many of the AERs, the amounts of ephedrine 
alkaloids consumed were within the ranges listed in trade 
recommendations or in product labeling. In addition, we note that the 
ephedrine alkaloid daily dose limit recommended by AHPA (Ref. 101) is 
higher than the dose administered to the treatment group in Boozer et 
al. (2002), which resulted in significantly higher blood pressure 
measured by ABPM when compared to the placebo group.
    (Comment 53) Several comments cited the 1999 GAO report (Ref. 124) 
to support their criticisms of our the June 1997 proposal. These 
comments state that GAO criticized the validity of serious AERs 
reported for ephedra, particularly when used according to trade 
recommendations.
    (Response) We do not agree that the July 1999 GAO report found the 
serious AERs reported for ephedra to be invalid (Ref. 124). Although 
the July 1999 GAO report criticized our use of adverse event reports to 
support the serving size and duration of use limits in the June 1997 
proposal, it also emphasized that the adverse events reported to us 
were serious enough to warrant FDA's further investigation of the 
safety of dietary supplements containing ephedrine alkaloids. In 
addition, the report concluded that scientific information indicates 
that ephedrine alkaloids can affect the cardiovascular and nervous 
systems, citing (among others) published case reports that suggest 
ephedrine alkaloids can increase blood pressure in persons with normal 
and high blood pressure; predispose certain individuals to tachycardia 
(rapid heart rate), and cause cardiomyopathy (disease of the heart 
muscle), stroke, or myocardial necrosis (death of cells in the heart). 
The 1999 GAO report also noted that adverse events associated with 
dietary supplements containing ephedrine alkaloids include effects on 
the central nervous system, such as mania, paranoid psychoses, and 
seizures.
    GAO's 2003 testimony before the Subcommittee on Oversight and 
Investigation of the House Committee on Energy and Commerce discussed 
and updated some of GAO's findings from its 1999 report on dietary 
supplements containing ephedrine alkaloids and provided new 
information, including an evaluation of Metabolife International's 
records of health-related calls from consumers of Metabolife 356 (Refs. 
23 and 24). The 2003 GAO testimony noted that the types of adverse 
events identified in the health-related call records from Metabolife 
International were consistent with the types of adverse events reported 
to us, as well as with the scientifically documented pharmacological 
and physiological effects of ephedrine alkaloids. The 2003 GAO 
testimony noted that despite the limited information contained in most 
of the call records, approximately 14,684 call records contained 
reports of at least one adverse event among consumers of Metabolife 
356. The 2003 GAO testimony identified 92 serious events that included 
heart attacks, strokes, seizures, and deaths and emphasized that these 
findings were similar to other reviews of the call records, including 
those done by Metabolife International and its consultants. The 2003 
GAO testimony noted that, in those call records where age was 
documented, many of the serious adverse events occurred in relatively 
young consumers, with more than one-third of such adverse event 
occurring in individuals under the age of 30. Furthermore, for those 
call records in which quantity of use and/or frequency and duration of 
use were noted, most of the serious adverse events occurred among 
Metabolife 356 users who used the product within the recommended 
guidelines, i.e., they did not take more of the product nor consume it 
for a longer period of time than the product label recommended. These 
findings are consistent with our evaluations of AERs that we have 
received regarding dietary supplements containing ephedrine alkaloids 
(Refs. 27 and 109).
    The 2003 GAO testimony noted that the adverse event reports are 
important sources of information concerning health risks of dietary 
supplements containing ephedrine alkaloids because the regulatory 
framework for dietary supplements is basically one of postmarketing 
surveillance and does not require premarket approval. The testimony 
stressed that despite the limited information obtained from the 
Metabolife International call records, the types of adverse events 
reviewed were consistent with the known risks of ephedrine alkaloids, 
including serious adverse events such as five reports of death. 
Finally, the testimony noted that several years earlier, we had 
concluded that dietary supplements containing ephedrine alkaloids 
present a ``significant public health hazard'' based upon the adverse 
event reports received and the consistency of those reports with the 
known pharmacological effects of ephedrine alkaloids.

C. What Are the Known and Reasonably Likely Benefits of Dietary 
Supplements Containing Ephedrine Alkaloids?

1. Weight Loss
    (Comment 54) Numerous comments, including those from manufacturers 
and industry trade groups, stated that the results of the RAND report 
and other evidence, including the CANTOX review and the Boozer et al. 
clinical studies (Refs. 49 and 125), support or establish the safety 
and efficacy of dietary supplements containing ephedrine alkaloids for 
weight loss. Several comments stated that RAND concludes that dietary 
supplements containing ephedrine alkaloids have proven benefits for 
weight loss purposes. Several comments stated that RAND shows that 
dietary supplements containing ephedrine alkaloids provide a 
statistically significant increase in short-term weight loss compared 
to placebo of about 2 pounds per month for up to 6 months.
    (Response) We agree that the RAND report found evidence that 
supported an association between short-term use of ephedrine, ephedrine 
plus caffeine, or dietary supplements containing ephedrine alkaloids 
with or without botanicals containing caffeine and a statistically 
significant increase in short-term weight loss compared to placebo. 
RAND found that combinations of

[[Page 6819]]

botanical ephedrine alkaloids plus botanical sources of caffeine, or 
synthetic ephedrine plus caffeine, were more effective in promoting 
short-term weight loss than ephedra or ephedrine alone. The RAND report 
concluded that ephedrine alkaloid containing products, in combination 
with caffeine, resulted in a modest weight loss of approximately two 
pounds per month greater than that with placebo over a period of 4 to 6 
months.
    We also agree that this modest weight loss effect may be perceived 
as a benefit by consumers who seek to lose weight for nonhealth related 
purposes (e.g., to look slimmer). We do not agree, however, that these 
studies demonstrate the long-term weight loss necessary to provide 
health benefits. While the improvements in obesity/overweight and the 
accompanying risk factors may be demonstrated in as few as 1 to 2 
months, the improvements must be maintained for years to achieve a 
reduction in risk (Refs. 66, 126, 127, and 128). We note that dietary 
supplements cannot be lawfully marketed for the treatment of obesity, a 
disease with serious health consequences. From a health perspective, 
the goal of weight loss is to prevent the substantial morbidity and 
mortality associated with overweight and obesity (Refs. 66, 129, and 
130). Obesity itself adversely impacts multiple cardiovascular risk 
factors, or comorbidities, including hypertension, dyslipidemia (high 
cholesterol), and insulin resistance with glucose intolerance. Clinical 
studies have demonstrated improvements in these risk markers with even 
modest sustained weight loss (i.e., approximately 5 to 10 percent of 
initial body weight). Clinical studies have also demonstrated that both 
the weight loss and the improvements in the comorbidities take time to 
accrue (i.e., months) and that, as a rule, weight is regained and the 
comorbidities worsened when the intervention, pharmacological or 
behavioral, is discontinued. Thus, interventions necessary for 
successful weight maintenance must be long term. As discussed in 
greater detail below in the response to comment 56 of this document, 
the reasonably well-documented moderate, short-term weight loss from 
use of ephedrine alkaloids, with or without caffeine, does not prevent 
or decrease substantial, obesity-related irreversible morbidity and 
mortality. We have not found evidence that demonstrates long-term 
weight loss with these products.
    We note that, to the extent these comments raise the issue of 
safety, we address those issues in section V.B of this document.
    (Comment 55) A number of comments from manufacturers, distributors, 
industry experts, and trade groups were critical of the methodology 
used for the RAND report or the conclusions of this review. One comment 
stated that RAND does not take a sufficiently quantitative approach in 
its review of the data in contrast to the review performed by CANTOX. 
The comment also objected that RAND did not perform an efficacy 
comparison for ephedra-caffeine and that its dose-response assessment 
excludes the medium dosage range (40 to 90 mg), which includes the 6-
month Boozer et al. (2002) study. Consequently, the comment argued that 
these omissions preclude any assessment of the degree of agreement or 
disagreement between RAND and CANTOX.
    Other comments objected to RAND's criteria for study inclusion in 
the evaluation process, stating that RAND failed to consider all 
relevant and applicable trials. In particular, one comment criticized 
RAND's decision to consider only human weight loss trials that lasted 
at least 8 weeks, noting that 20 of 46 identified studies were excluded 
for this reason, and an additional six studies for other ``alleged'' 
reasons. Several comments objected to RAND's conclusions that weight 
loss research on ephedra, ephedrine, and caffeine (6-month data) is 
``short-term'' only and not sufficient to demonstrate long-term weight 
loss, and cited additional studies to support this view. One comment 
stated that 6 months is longer than the period of time recommended by 
FDA's Advisory Review Panel on OTC Miscellaneous Internal Drug Products 
with respect to evaluating weight loss ingredients used in OTC drugs. 
The comment stated that, by these standards, RAND's 6-month weight loss 
efficacy data ``exceeds the scientific requirement for evaluating OTC 
weight loss drugs recommended by FDA's advisory panel by 3 months.'' 
Other comments stated that, from a scientific perspective, there is no 
reason to believe the weight loss from dietary supplements containing 
ephedrine alkaloids would cease after a 6-month period (Refs. 70, 79, 
and 131).
    (Response) RAND, using the principles of evidence-based medicine, 
established the scope of the review and methodology used in its 
assessment of the currently available data. The RAND reviewers limited 
their evaluation to those randomized or controlled clinical trials of a 
minimum study duration (8 weeks) that provided adequate information, 
including sufficient protocol design and safety information on the 
basis that shorter treatment durations were insufficient to assess 
long-term weight loss. We believe that RAND's study selection criteria 
were appropriate. Further, we note that in the absence of statutory 
requirements for dietary supplement manufacturers to submit well-
designed, long-term, placebo-controlled studies to us, the available 
body of well-controlled clinical data is limited. We believe that RAND 
appropriately screened the available data and reviewed all relevant 
studies and adverse event reports meeting their stated minimum standard 
criteria, and thus we consider the results and conclusions of this 
assessment valid. Exclusion of studies not directed toward weight loss 
or obesity was appropriate for this evaluation in that these studies 
were designed to examine the efficacy of these agents for asthma and 
related pulmonary indications, rather than their safety.
    We have reviewed the additional studies cited in the comments to 
support the effectiveness of dietary supplements containing ephedrine 
alkaloids for long-term weight loss (Refs. 68, 79, and 131). The 
results of the Filozof study have been presented only in abstract form 
and, therefore, neither details of the protocol nor data were available 
for review. The Daly et al. study enrolled only 24 subjects for 8 weeks 
in a placebo-controlled trial. After that period, 8 subjects were 
followed in an open label study for varying durations (1 subject was 
followed for 26 months). These additional studies were not evaluated in 
the RAND assessment because they did not meet RAND's screening 
criteria, and we find these studies to be either irrelevant or 
inadequate to change the conclusions stated in the RAND report. 
Therefore, we find that the Boozer 2002 study remains the longest (6-
month) placebo-controlled study using ephedrine alkaloids. 
Consequently, we agree with RAND's conclusion that there are no studies 
showing an effect of dietary supplements containing ephedrine alkaloids 
on weight loss for more than 6 months.
    Concerning the comment that referenced the Advisory Review Panel on 
OTC Miscellaneous Internal Drug Products with respect to evaluating 
weight loss ingredients used in OTC drugs, we note that the 1979 report 
of this panel was discussed in an advance notice of proposed rulemaking 
published in the Federal Register on February 26, 1982 (47 FR 8466). 
Based on the standard of practice at that time, the Advisory Review 
Panel

[[Page 6820]]

recommended that non-monograph weight loss ingredients (i.e., those not 
classified as GRASE) be studied for a period of 12 weeks to demonstrate 
effectiveness.
    The treatment of obesity has evolved over the past 50 or so years 
(Refs. 127 and 128). In the 1960s, the mainstay of obesity treatment 
was behavioral modification and drugs were approved for short-term 
treatment to ``jump start'' patients' weight loss. There was a paradigm 
shift in the 1990s, with the realization that obesity is a chronic 
disease requiring long-term treatment, both with behavior modification 
and long-term drug therapy, when appropriate, in addition to diet and 
exercise. This shift is reflected in our draft guidance published in 
1996 recommending the performance of clinical trials with a minimum 12-
month treatment duration (see FDA Draft Guidance for The Clinical 
Evaluation of Weight-Control Drugs, Division of Metabolic and Endocrine 
Drug Products, issued on September 24, 1996) (Ref. 129). Therefore, 
because the treatment of obesity has evolved over time, the 1982 OTC 
Advisory Panel recommendations do not reflect current scientific 
understanding of effective treatment of obesity. There are currently no 
GRASE OTC drug products for weight loss or management.
    (Comment 56) Many comments stated that obesity is a disease with 
serious health consequences. Numerous comments from consumers and 
physicians contained personal testimonials regarding the efficacy of 
dietary supplements containing ephedrine alkaloids for weight loss. 
Several physicians noted that patients who used these products were 
able to achieve long-term weight loss with an overall improvement of 
health, including improved cholesterol levels and lower blood pressure. 
No data were submitted, however, to support these statements. Several 
comments stated that ephedrine alkaloids are an effective tool to fight 
obesity. Several comments expressed the view that there are health 
benefits from short-term weight loss. Several other comments stated 
that dietary supplements containing ephedrine alkaloids are as--or 
more--effective for weight loss than some prescription drugs (e.g., 
amphetamine, phentermine, sibutramine, phendimetrazine). Another 
comment stated that the evidence suggested that ephedra/ephedrine-
caffeine supplements are as effective as OTC drugs for weight 
management. One comment stated that other modalities used to promote 
weight loss are very difficult, very dangerous, or very unsuccessful.
    A comment by an industry trade group stated that the amount of 
weight loss identified by RAND for dietary supplements containing 
ephedrine alkaloids (approximately 2 pounds per month greater than 
placebo) is similar to that reported for approved obesity drugs (citing 
Ref. 128). Further, the comment asserted that ``similar to ephedra-
containing supplements, there is no long-term information [on weight 
loss] for any but the two most recently approved drugs [sibutramine and 
orlistat]'' and that few studies of drugs approved for weight loss have 
extended to 6 months or beyond. One comment stated that double-blind 
placebo-controlled studies, including Boozer et al. (2002) (Ref. 49) 
have addressed the safety and efficacy of the dietary supplements 
containing ephedrine alkaloids, and further stated that the low cost of 
these products is beneficial, especially for low income groups where 
maintenance of a good diet is a challenge.
    In contrast, other comments from physicians and medical societies, 
while acknowledging the results of the RAND report showing modest, but 
statistically significant short-term weight loss, questioned such a 
weight loss effect in light of the risks of these products. One comment 
indicated that this modest degree of ``drug-induced weight loss'' has 
never been shown to reduce the increased morbidity observed in obese 
patients. Several comments stated that there is no evidence for 
efficacy or safety of chronic treatment with ephedra. One medical 
association stated that the very modest benefits of ephedra combined 
with caffeine on short-term weight loss are far outweighed by the 
adverse effects observed in the clinical trials and the serious risks 
reported with the use of dietary supplements containing ephedrine 
alkaloids.
    Several other comments, including those from an herbalist 
association and an herbal product manufacturer, stated that the use of 
these supplements, although effective, is not a sensible or healthy 
approach to long-term, sustainable weight management. The comment from 
the herbalist association also stated that obesity, with its higher 
risk for cardiovascular disease, is more likely to be a 
contraindication rather than an indication for the use of ephedra. A 
comment from a medical association said that NIH guidelines for the 
pharmacological treatment of adult obesity state that herbal 
preparations, including ephedra-containing products, are not 
recommended as part of a weight-loss program (Ref. 66).
    Several comments, including one by a trade association and a 
medical society, while acknowledging the conclusions of the RAND report 
with regard to ephedrine alkaloids and weight loss, said that this 
effect should not be construed to imply that dietary supplements 
containing ephedrine alkaloids can treat diseases. One comment 
expressed the view that we should consistently state that obesity is a 
disease and, therefore, should only be treated with drugs that have 
been approved as safe and effective for that disease. These comments 
stated that use of dietary supplements to ``treat'' obesity is 
inappropriate.
    (Response) As stated previously, we agree that obesity is a disease 
with serious health consequences; however, as some comments noted, 
treatment of a disease is outside the scope of the uses authorized for 
dietary supplements under DSHEA. Consequently, although dietary 
supplements containing ephedrine alkaloids could, if they did not 
present an unreasonable risk of illness or injury, be labeled for 
ordinary weight loss, they are subject to regulation as drugs if 
promoted for the treatment of obesity (65 FR 1000 at 1026 and 1027, 
January 6, 2000). We agree with the comments stating that obesity 
should be treated only with drugs that have been approved as safe and 
effective for that use.
    We do not agree with the comments comparing the effectiveness of 
dietary supplements containing ephedrine alkaloids for weight loss to 
approved prescription drugs. The drugs mentioned by the comments are 
approved for the treatment of obesity, which is a use for which dietary 
supplements cannot be marketed. Furthermore, we are unaware of any data 
that have made direct comparisons between dietary supplements 
containing ephedrine alkaloids for weight loss and drugs approved for 
the treatment of obesity. As discussed previously, prescription drugs 
for the treatment of obesity are no longer approved on the basis of 
short-term data or for short-term use. Of note, the few prescription 
drugs that were approved for short-term use to ``jump-start'' weight 
loss are all stimulants and are controlled substances, the first group 
being approved in 1939 (amphetamine) and the last being approved in 
1979 (phendimetrazine). The use of the majority of these drugs has 
fallen out of favor or the drugs have been withdrawn from the U.S. 
market. Whether the remainder of these drugs with indications for 
short-term use should be withdrawn is beyond the scope of this 
rulemaking. The rationale for requiring

[[Page 6821]]

long-term studies (1 to 2 years) to evaluate drugs intended to treat 
obesity was thoroughly discussed in the 1995 FDA/Center for Drug 
Evaluation and Research (CDER) Endocrinologic and Metabolic Drugs 
Advisory Committee Meeting. In that meeting, the panel discussed the 
duration of trials for evaluating both efficacy and safety of drugs for 
the treatment of obesity and used the example of Fluoxetine as a drug 
that demonstrated efficacy for weight loss at 6 months but did not 
promote additional weight loss or maintain previous weight loss in 
longer term (1-year) studies, although the risk for experiencing 
adverse effects still persisted.
    Alleged economic benefits of these products are not considered as a 
component of our evaluation of their risks and benefits. Therefore, 
comments suggesting an economic benefit from using dietary supplements 
containing ephedrine alkaloids as an alternative to drugs for weight 
loss are not relevant to whether dietary supplements containing 
ephedrine alkaloids present an unreasonable risk. We also note that 
there are currently no stimulant-containing OTC drugs (including those 
with phenylpropanolamine) legally marketed for weight management and 
that amphetamine is no longer labeled for weight loss. There are no 
existing final OTC drug monographs for any weight control drug 
products, although one nonstimulant ingredient (benzocaine) remains to 
be evaluated for this use as part of FDA's OTC drug review and can 
continue to be marketed pending the outcome of that review.
    The comments that mentioned health benefits from short-term weight 
loss submitted no data to support this contention, and we are not aware 
of any studies that indicate any meaningful health benefit from short-
term weight loss. In the longest controlled study to date on the effect 
of ephedrine alkaloid containing products on weight loss by Boozer et 
al. (2002) (Ref. 49), subjects treated with placebo, plus diet and 
exercise recommendations, lost an average of approximately 6 pounds 
over a period of 6 months (Ref. 49). Subjects treated with a 
proprietary blend of herbal ephedra and kola nut (a source of 
caffeine), plus diet and exercise recommendations, lost an average of 
approximately 12 pounds during the same time period. As described 
previously in the response to comment 22 of this document, on balance 
this trial did not show a favorable effect on cardiovascular risk 
factors. To the contrary, there was a statistically significant 
increase in heart rate in the ephedra/kola nut (i.e., herbal ephedrine 
alkaloids/caffeine) treated subjects compared to the control group. 
Moreover, 24-hour measurements of blood pressure measured by ABPM at 1 
month showed that the ephedrine alkaloid/caffeine treated subjects had 
blood pressure that was approximately 4 mm Hg higher than the placebo-
treated subjects for both systolic and diastolic blood pressure.
    While the authors report small but statistically significant 
decreases in total cholesterol and low density lipoproteins (LDL) 
cholesterol, the clinical significance of the net 3 mg/dl and 8 mg/dl 
decreases, respectively, cannot be determined from this study. In 
studies designed to assess modifications in cardiovascular risk 
factors, cholesterol changes are reported as percentage change from 
baseline. These data are not available from the Boozer et al. (2002) 
study (Ref. 49).
    (Comment 57) A number of comments stated that the Danish experience 
using ephedrine/caffeine in a prescription drug for the treatment of 
obesity supported the use of dietary supplements containing ephedrine 
alkaloids for weight loss. One comment from a manufacturer of dietary 
supplements containing ephedrine alkaloids shared the opinion that the 
effectiveness of ephedrine alkaloids ``to support one's diet'' has been 
demonstrated in numerous studies, involving hundreds of patients in 
well-controlled environments, and that efficacy has also been 
demonstrated by extensive use data in the United States and Denmark. A 
comment from a medical association stated that, in Denmark, ephedrine 
is available to treat obesity, but only by prescription. Another 
comment stated that the Danish ephedrine-caffeine product (Letigen) has 
been banned and withdrawn from the market because of safety issues.
    (Response) We agree with the comments that the product used in 
Denmark, Letigen, was a prescription drug and that this drug has been 
withdrawn from the market for safety reasons, including serious adverse 
event reports documenting cardiovascular and nervous system effects 
(Refs. 120 and 121). We note that certain studies from Denmark using 
the ephedrine-caffeine combination found in Letigen were considered as 
part of the RAND report. We do not agree with the comment that numerous 
studies have demonstrated the effectiveness of ephedrine alkaloids to 
support weight loss for the treatment of obesity, as discussed 
previously. The use of dietary supplements containing ephedrine 
alkaloids has been shown to produce a small, short-term weight loss, 
but no studies showing long-term weight loss with accompanying benefits 
to health have been conducted. In any case, if botanical ephedrine 
alkaloid products could be shown effective in long-term treatment of 
obesity or for long-term weight loss in people who are not obese, they 
would need to be marketed as prescription drugs and meet the standards 
of safety and effectiveness legally mandated for such products because 
physician supervision would be necessary to adequately mitigate the 
risks of using these products continuously in the long term.
2. Enhancement of Athletic Performance
    (Comment 58) Several comments discussed the effects of ephedrine 
alkaloids on athletic performance. One comment noted that, while RAND 
states that ephedrine is a good surrogate for evaluation of dietary 
supplements containing ephedrine alkaloids, RAND does not make this 
extrapolation for athletic performance. Many other comments stated that 
there are few data to support the use of synthetic ephedrine alkaloids, 
and no data to support the use of dietary supplements containing 
ephedrine alkaloids to enhance athletic performance. Therefore, these 
comments do not consider the enhancement of athletic performance to be 
an appropriate use for dietary supplements containing ephedrine 
alkaloids. According to some comments, RAND concluded that there are 
insufficient data to support use for enhancement of athletic 
performance. One comment asserted that any effect on athletic 
performance is more likely due to the caffeine in ephedrine-caffeine 
dietary supplements. According to another comment, the few studies that 
have assessed the effect of ephedrine for this use support a modest 
effect of ephedrine plus caffeine on very short-term (1 to 2 hours 
after a single dose) athletic performance in a highly selected, 
physically fit population, but no studies have assessed the effect of 
dietary supplements containing ephedrine alkaloids.
    (Response) We generally agree with these comments. The RAND report 
provides the most comprehensive, currently available review of efficacy 
studies for ephedrine alkaloid containing products, focusing on two 
popular uses of these products--athletic performance and weight loss 
(see section V.C.1 of this document). (Note that the RAND report did 
not consider the effectiveness data for ephedrine alkaloid containing 
products marketed as drugs for other uses, such as to treat asthma, or 
for other dietary supplement uses of such products). The effect of 
synthetic ephedrine on athletic

[[Page 6822]]

performance was assessed in seven studies that were reviewed in the 
RAND report. The RAND report noted that the effects of ephedrine on 
exercise performance were most often studied acutely (e.g., 1 to 2 
hours after a single dose) (Refs. 21 and 22). The RAND report could 
identify no studies that assessed the effect of dietary supplements 
containing ephedrine alkaloids on athletic performance. While the RAND 
report found that existing data supported a modest effect of synthetic 
ephedrine alkaloid containing products plus caffeine on athletic 
performance enhancement in healthy males in the very short term, no 
data support a sustained improvement in athletic performance over any 
significant time period. In these studies, the performance enhancement 
effect was demonstrated only with a combination of synthetic ephedrine 
and caffeine, not with ephedrine alone. Therefore, since the available 
evidence does not indicate that ephedrine itself enhances athletic 
performance, there is no need to address the issue as to whether 
ephedrine is a good surrogate for ephedra in evaluating athletic 
performance enhancement with the use of dietary supplements containing 
ephedrine alkaloids.
    We determined that certain labeling claims made by manufacturers of 
dietary supplements containing ephedrine alkaloids for athletic 
performance enhancement were unsubstantiated in light of the findings 
in the RAND report. These claims were the subject of warning letters 
sent to various manufacturers in February and March 2003 (available at 
http://www.fda.gov/bbs/topics/NEWS/ephedra/letterslist.html (list of 
firms) and http://www.fda.gov/bbs/topics/NEWS/ephedra/warning.html 
(sample letter).
3. Eased Breathing
    We are aware that there are teas and other types of dietary 
supplements containing ephedrine alkaloids marketed with claims such as 
``eased breathing'' or ``better breathing.'' There are no data that 
support a benefit to breathing from dietary supplements containing 
ephedrine alkaloids in healthy people. Moreover, because healthy people 
are able to breathe without difficulty, we do not believe there is any 
respiratory benefit in the absence of a disease state (e.g., asthma or 
a respiratory infection). We note that claims to treat or mitigate a 
disease, or the effects of a disease, subject a product to regulation 
as a drug under the act.
4. Other Uses
    We are also aware that dietary supplements containing ephedrine 
alkaloids are promoted for other uses, such as to ``feel better,'' 
``feel more alert,'' and ``energized.'' Effects such as ``feel better'' 
are subjective in nature and difficult to quantify. The agency is 
unaware of any data substantiating these types of subjective effects. 
Effects such as ``alertness'' and ``energy'' are consistent with the 
pharmacological properties of ephedrine alkaloids, although we are not 
aware of any studies evaluating ephedrine alkaloid products for these 
uses. Effects like alertness and energy may be of modest benefit to the 
individual (if they occur), but such effects are temporary and do not 
improve health. Any such temporary benefits must be weighed against the 
health risks discussed in section V.B of this document, which can 
result in long-term or permanent, serious adverse health effects.

D. Do Dietary Supplements Containing Ephedrine Alkaloids Present an 
Unreasonable Risk?

1. What Does ``Unreasonable Risk'' Mean?
    A threshold issue is the legal standard of ``significant or 
unreasonable risk of illness or injury'' (section 402(f)(1)(A) of the 
act). By its plain language, this standard requires evidence of 
``significant or unreasonable risk of illness or injury'' (emphasis 
added).'' There is no requirement that there be evidence conclusively 
demonstrating causation of actual harm in specific individuals. In our 
evaluation of ``significant or unreasonable risk,'' we can consider any 
relevant evidence, including scientific data about the toxicological 
properties of a dietary ingredient or its mechanisms of action; 
scientific information about the well-known effects of 
pharmacologically-related compounds, including those regulated as 
drugs; the results of clinical studies, including observational 
studies; and adverse event reports that have been subject to sound 
scientific analysis. The Government's burden of proof for ``significant 
or unreasonable risk'' can be met with any science-based evidence of 
risk, without the need to prove that the substance has actually caused 
harm in particular cases.
    Thus, a dietary supplement that caused a sustained rise in blood 
pressure across the population would increase the risk of 
cardiovascular events including stroke, heart attack, or death to that 
population. Even risks that may not be detectable in small studies or 
studies of short duration could, over time, and on a population-wide 
basis, result in hundreds or thousands of adverse events. The 
Government's burden of proof for ``unreasonable risk'' is met when a 
product's risks outweigh its benefits in light of the claims and 
directions for use in the product's labeling or, if the labeling is 
silent, under ordinary conditions of use.
    (Comment 59) Most comments that articulated a view agreed with the 
general notion that we must consider a risk-benefit calculus to 
determine whether dietary supplements containing ephedrine alkaloids 
present an unreasonable risk, although the comments differed as to how 
to perform such a calculus and as to the conclusion about whether the 
risks of these products outweigh their benefits. Several comments 
agreed with our interpretation, as published in (Ref. 132), that a 
``significant or unreasonable risk'' exists when a product's risks 
outweigh its benefits, based on the available scientific evidence, in 
light of the claims the product makes and in light of the products 
being directly sold to consumers without medical supervision. One 
comment from a public interest group stated that this interpretation 
represents a reasonable and practical interpretation of the act that 
offers some protection to consumers. One comment argued that this 
interpretation is not permissible under Chevron U.S.A., Inc. because we 
have never adopted a risk-benefit calculus in assessing the safety of 
foods and because the legislative history of DSHEA does not indicate 
any Congressional intent to establish a risk-benefit analysis for 
dietary supplements. The comment stated that we should determine 
whether risks are ``unreasonable'' without resorting to an assessment 
of the benefits of the product.
    (Response) We agree with the comments stating that a risk-benefit 
calculus is appropriate to determine whether dietary supplements 
containing ephedrine alkaloids present an unreasonable risk of illness 
or injury under conditions of use recommended or suggested in the 
labeling, or if no conditions of use are suggested or recommended in 
the labeling, under ordinary conditions of use. The relevant analysis 
for evaluating an agency's interpretation of a statute is set forth in 
Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837 
(1984). Under Chevron, the first question is whether Congress has 
directly spoken to the precise question at issue (Step 1). If so, the 
agency must implement the unambiguous intent of Congress Id. at 842-
843. If Congress has

[[Page 6823]]

not directly spoken to the precise question at issue, our 
interpretation will be upheld as long as it is based on a ``permissible 
construction'' of the statute (Step 2) Id. at 843-844.
    In determining whether Congress has specifically addressed the 
question at issue, ``courts must exhaust the traditional tools of 
statutory construction, including looking at the statute's text, 
structure, and legislative history.'' Chevron v. Federal Energy 
Regulatory Commission, 193 F.Supp.2d 54, 67 (D.D.C. Cir. 2002). Section 
402(f)(1)(A) of the act states that a dietary supplement is adulterated 
if it presents a significant or unreasonable risk of illness or injury 
under the conditions of use recommended or suggested in labeling, or, 
if the labeling is silent, under ordinary conditions of use. The plain 
meaning of the statute is the starting point of statutory 
interpretation. (See 2A SUTHERLAND STATUTORY CONSTRUCTION 81 (5th ed. 
1992).) The words ``significant'' and ``unreasonable'' have two 
different meanings. ``Significant'' involves an evaluation of risk 
alone. The plain meaning of ``unreasonable,'' on the other hand, 
connotes comparison of the risks and benefits of the product. A risk 
could be significant but reasonable if the benefits were great enough 
to outweigh the risks. That ``unreasonable risk'' entails a balancing 
test in which the benefits of the product or activity are weighed 
against its dangers is well-established in tort law (See PROSSER AND 
KEETON ON THE LAW OF TORTS, Sec. 31, at 173 (5th ed. 1984).)
    In assessing whether Congress has clearly spoken to the question at 
issue, a court ``should not confine itself to examining a particular 
statutory provision in isolation. Rather, it must place the provision 
in context, interpreting the statute to create a symmetrical and 
coherent regulatory scheme'' (FDA v. Brown and Williamson Tobacco 
Corp., 529 U.S. 120, 121 (2000)). The term ``unreasonable risk'' is 
used in other provisions of the act, e.g., in the provisions related to 
medical devices. In the medical device classification provisions, Class 
III devices are distinguished from Class I and Class II devices in part 
because they present a ``potential unreasonable risk of injury or 
illness.'' The legislative history of the device provisions provides 
some indication of how Congress intended FDA to interpret the term 
``unreasonable risk in this context. The House Committee Report states: 
``the requirement that a risk be unreasonable contemplates a balancing 
of the possibility that illness or injury will occur against the 
benefits of use'' (H. Rept. 853, 94th Cong., 2d Sess. 19 (1976)). 
Therefore, ``unreasonable risk'' in the context of classification of 
medical devices is properly interpreted to require a risk-benefit 
calculus. There is nothing in the provisions of the act dealing with 
dietary with dietary supplements, or the legislative history thereof, 
that would suggest that FDA should interpret the term ``unreasonable 
risk'' in the context of dietary supplements differently than it does 
in the context of medical devices.
    An interpretation of unreasonable risk as entailing a balancing of 
the risks and benefits of the product is also consistent with the 
interpretation of other similar statutory provisions outside the act. 
The Toxic Substances Control Act contains an ``unreasonable risk'' 
standard, and legislative history indicates that Congress intended that 
this standard be evaluated through a balancing test (e.g., H. Rept. 94-
1341, 94th Cong., 2d Sess. 32 (1976)). Indeed, it is difficult to 
construct an alternative formulation for the phrase ``unreasonable 
risk.''
    Based upon the plain meaning of ``unreasonable risk,'' the judicial 
interpretation of that phrase, and legislative history interpreting 
``unreasonable risk'' in other contexts, including the device 
provisions of the act and other statutes, we conclude that Congress 
unambiguously intended that an assessment of ``unreasonable risk'' in 
the dietary supplement context should entail a risk-benefit analysis.
    In the alternative, if a court were to find that Congress has not 
directly spoken to the issue of whether ``unreasonable risk'' in the 
dietary supplement context is demonstrated by balancing risks and 
benefits, our interpretation of an ambiguous provision should receive 
deference so long as it is ``permissible'' (Chevron Step 2). In 
interpreting ambiguous statutory language, we are guided by the same 
criteria we evaluated in Step 1 of the Chevron analysis, i.e., the 
statute's text, structure, history, and purpose (See Bell Atlantic 
Telephone Cos. v. FCC, 131 F.3d 1044, 1049 (D.C. Cir. 1997); Chevron 
U.S.A., Inc. v. FERC, 193 F. Supp. 2d at 68). Our interpretation of the 
``unreasonable risk'' standard for dietary supplements as requiring a 
comparison of the risks and benefits of use is consistent with the 
purpose of the act, as amended by DSHEA, to promote public health and 
safety. This interpretation is also consistent with the legislative 
history of the medical device classification provisions. Therefore, our 
interpretation that ``unreasonable risk'' implies a weighing of the 
risks and benefits of use is, at a minimum, a ``permissible 
construction.''
    In the absence of explicit standards for the evaluation of 
``unreasonable risk,'' one comment urged us to be guided by precedent 
from other agencies. The comment highlighted the Consumer Product 
Safety Act (CPSA), its implementing regulations, and related case law. 
The comment stated that any assessment of ``unreasonable risk'' must 
include a balancing of risks and benefits, a stringent burden on us to 
demonstrate that the product poses an unreasonable risk of injury, 
evidence other than consumer complaints, and valid scientific data 
sufficient to predict how likely an injury is to occur. (Citing Gulf 
South Insulation v. CPSC, 701 F.2d 1137, 1143 (5th Cir. 1983)), (citing 
Aqua Slide `N' Dive v. CPSC, 569 F.2d 831, 838 (5th Cir. 1978)), the 
comment stated, ``[T]he ultimate question in assessing unreasonable 
risk is whether the record contains `such relevant evidence as a 
reasonable mind might accept as adequate to support a conclusion.''' 
The comment acknowledged differences in the statutes, including the 
explicit statutory requirement in CPSA that the regulation impose the 
least burdensome requirement that prevents or adequately reduces the 
risk injury for which the rule is being issued (15 U.S.C. 
2058(f)(3)(F)). The comment also cited Consumer Product Safety 
Commission (CPSC) case law stating that reliable evidence of the likely 
number of injuries is necessary to determine whether a risk is 
unreasonable (Southland Mowor Co. v. CPSC, 619 F.2d 499, 510 (5th Cir. 
1980)).
    (Response) We do not agree that our interpretation of 
``unreasonable risk'' must be confined to the view reflected in the 
CPSC case law cited by the comment. We have concluded, based on a 
Chevron analysis, that Congress expressly intended ``unreasonable 
risk'' to entail a risk-benefit analysis (see the response to comment 
59 of this document). In the alternative, if the term ``unreasonable 
risk'' is ambiguous, we may interpret its meaning under Chevron. As the 
comment noted, CPSA contains an extensive list of findings that the 
CPSC must make, based on substantial evidence, before concluding that a 
consumer product poses an unreasonable risk, including, for example: 
(1) The degree and nature of the risk of injury the rule is designed to 
eliminate or reduce; (2) the approximate number of consumer products, 
or types or classes thereof, subject to such rule; and (3) any means of 
achieving the objective of the order while minimizing adverse effects 
on competition or disruption or dislocation of

[[Page 6824]]

manufacturing and other commercial practices (15 U.S.C. 2058(f)(1) and 
(f)(3)). The requirements imposed on CPSC in the cases that the comment 
cited are based on the explicit requirements of CPSA. In contrast, the 
adulteration provision in section 402(f)(1)(A) of the act does not 
require that we make any such findings. Like section 402(f)(1)(A) of 
the act, other parts of the act that require an evaluation of 
unreasonable risk, such as the device classification and banning 
provisions, also do not require that we make the findings set forth in 
CPSA. Had Congress intended that FDA make specific findings such as the 
degree of risk of injury, it could have so directed in the act; 
however, it did not. Our conclusion that dietary supplements containing 
ephedrine alkaloids present an unreasonable risk is based upon our 
finding that the risks of heart attack, stroke, and death outweigh the 
minimal benefits conferred by the supplements. Our conclusion is 
consistent with Congress's express intent in section 402(f)(1)(A) of 
the act.
    (Comment 60) One comment by a health professional group stated that 
unreasonable risk likely exists when there is no information that 
substantiates a clear therapeutic benefit or describes a predictable 
relationship between exposure (dose) and response, and when the 
appropriate product dose is not known or achievable.
    (Response) We agree that unreasonable risk exists when a dietary 
supplement presents a risk to health, and there is no information 
substantiating a benefit sufficient to outweigh that risk. In this 
rulemaking, we base our determination that dietary supplements 
containing ephedrine alkaloids present an unreasonable risk under 
section 402(f)(1)(A) of the act on a risk-benefit analysis, finding 
that the risks of heart attack, stroke, and death outweigh the benefits 
that may result from such products. In the absence of a use that 
results in a benefit that outweighs the risks of these products, we 
conclude that all such products pose an unreasonable risk. We therefore 
need not determine whether an unreasonable risk exists when the precise 
relationship between exposure and response is not predictable or when 
the appropriate product dose is not known or achievable.
    (Comment 61) Several comments stated that proof of causation is 
required to establish unreasonable risk.
    (Response) We do not agree that proof of causation is required to 
establish unreasonable risk under section 402(f)(1)(A) of the act, and 
conclude that the plain meaning of the standard precludes such an 
interpretation. In determining whether Congress has specifically 
addressed the question at issue, ``courts must exhaust the traditional 
tools of statutory construction, including looking at the statute's 
text, structure, and legislative history'' (Chevron U.S.A., Inc. v. 
FERC, 193 F.Supp. 2d at 67). The plain meaning of the statute is the 
starting point for an analysis of legislative intent. The most 
applicable definition of the word ``risk'' in Merriam Webster's 
Collegiate Dictionary is ``possibility of loss or injury'' (Merriam 
Webster's Collegiate Dictionary, 10th ed. 1008 (2002)) (emphasis 
added). Black's Law Dictionary defines ``risk,'' in part, as follows: 
``In general, the element of uncertainty in an undertaking; the 
possibility that actual future returns will deviate from expected 
returns. Risk may be moral, physical, or economic.'' Black's Law 
Dictionary, 6th ed. 1328 (1990) (emphasis added). The words 
``possibility'' and ``uncertainty'' in these definitions indicates that 
proof of a definitive causal relationship between the product and 
illness or injury is not required under section 402(f)(1)(A) of the 
act. If Congress had intended that definitive proof that a dietary 
supplement causes harm be a requirement for a showing of adulteration, 
it would not have used the word ``risk'' in the statute, and would have 
instead provided that a dietary supplement is adulterated if it 
``causes'' illness or injury. This interpretation is consistent with 
other parts of the act, as interpreted in legislative history and case 
law. For instance, the legislative history of the medical device 
banning provisions, which require a showing of ``substantial deception 
or an unreasonable and substantial risk of illness or injury'' states 
that ``[A]ctual proof of deception or injury to an individual is [not] 
required'' (Section 516 of the act (21 U.S.C. 360f), H. Rept. 853, 94th 
Cong., 2d Sess. 19 (1976)). Case law on medical device classification 
also supports that we need not have causal evidence of harm (See Lake 
v. FDA, 1989 WL 71554 (E.D. Pa.)) (upholding FDA's finding of 
unreasonable risk where the risks were unknown and the benefits 
unproven)). Therefore, we conclude that Congress has spoken clearly and 
unambiguously that proof of causation is not required to show that a 
dietary supplement presents an ``unreasonable risk'' under section 
402(f)(1)(A) of the act.
    Our interpretation is also consistent with other statutes that 
regulate public health risks, most notably TSCA (15 U.S.C. 2601 et seq. 
(1976)). TSCA authorizes the EPA to place restrictions on chemical 
substances if it finds that ``* * * there is a reasonable basis to 
conclude that the [chemical substance] presents or will present an 
unreasonable risk of injury to health or the environment'' (Id. Sec. 
2605(a)). The legislative history of this provision states:
    This standard for taking action recognizes that factual 
certainty respecting the existence of an unreasonable risk of a 
particular harm may not be possible and the bill does not require 
it. Further, regulatory action may be taken even though there are 
uncertainties as to the threshold levels of causation.
(H. Rept. 94-1341, 94th Cong., 2d Sess. 25 (1976)).
    (Comment 62) Several comments stated that any FDA regulatory 
approach to dietary supplements containing ephedrine alkaloids must 
consider both risks and benefits, and moreover, that we should 
determine, based on scientific evidence, a risk-benefit ratio for 
assessing their safety. These comments suggested that, if we were to 
set a break-even point, a decision matrix should be established along 
the following lines: (1) A benefit-to-risk ratio below the break-even 
point would mean that the risks outweigh the benefits and this would 
justify either a decision to (a) ban dietary supplement products 
containing ephedrine alkaloids or (b) restrict access to a case-by-
case-basis, i.e., prescription; (2) a benefit-to-risk ratio in excess 
of the break-even point would mean that the benefits outweigh the risks 
and this would justify continued availability, with appropriate warning 
labels, dosage instructions, etc.; and (3) a benefit-to-risk ratio 
equal to the break-even point would mean that the risks equaled the 
benefits and this would justify either (a) continued availability under 
the present regulatory framework with appropriate labeling or (b) 
prescription-only access, whereby a medical professional would make the 
decision as to whether or not the product was appropriate for an 
individual consumer on a case-by-case basis.
    One comment by a medical association stated that, because dietary 
supplements are classified as foods, and therefore are assumed to be 
safe, it is imperative that such products have no risks and provide 
some benefit to consumers. More specifically, the comment stated that 
dietary supplements containing ephedrine alkaloids should be safer than 
drugs and should have a much higher overall benefit/risk ratio when 
compared to drugs.
    (Response) We agree that in regulating dietary supplements, we 
should

[[Page 6825]]

consider both risks and benefits. As discussed previously in this 
document, we also agree that we should weigh risks and benefits when 
evaluating the safety of dietary supplements under the adulteration 
standard in section 402(f)(1)(A) of the act. With regard to the comment 
from the medical association, we agree in part and disagree in part. 
Although the comment is correct that dietary supplements are classified 
as foods, we do not agree that they are required to have no risks at 
all. Section 402(f)(1)(A) of the act provides that a dietary supplement 
is adulterated if it ``presents a significant or unreasonable risk of 
illness or injury'' (emphasis added) as labeled, not if it presents any 
risk at all. Accordingly, risks that are insignificant and reasonable 
in light of the benefits from the supplement would not render a dietary 
supplement adulterated. Further, we note that conventional foods are 
not always risk-free. With regard to the comment's statements that 
dietary supplements should be safer than drugs and have a higher 
overall benefit/risk ratio than drugs, we do not believe it is 
necessary to reach these issues. For purposes of this rulemaking, we 
are considering whether the known and reasonably likely risks of 
dietary supplements containing ephedrine alkaloids outweigh their known 
and reasonably likely benefits. It is not necessary to determine 
generally how the risk/benefit ratio of dietary supplements should 
compare to that of drugs.
2. Do Dietary Supplements Containing Ephedrine Alkaloids Present an 
Unreasonable Risk Under Labeled or Ordinary Conditions of Use?
    (Comment 63) Several comments stated there is enough evidence, both 
scientific and anecdotal, to conclude that the risks of taking dietary 
supplements containing ephedrine alkaloids are so severe and reported 
adverse events sufficiently numerous to conclude that the risks clearly 
exceed the benefits because either there are no benefits or the 
benefits are unsubstantiated or modest for both efficacy and duration. 
These comments included references to support their conclusions. Some 
cited the RAND report's conclusions regarding the very modest benefit 
for short-term weight loss and the questionable benefit for other uses; 
according to the comments, these limited or questionable benefits are 
far outweighed by adverse events observed in clinical trials. Other 
references submitted by these comments included (Refs. 19, 34, 42, and 
133 through 136).
    Several comments argued that the harm caused by certain medical 
conditions--for example, obesity--is so severe as to render the 
unsubstantiated (in the commenter's view) risks of taking dietary 
supplements containing ephedrine alkaloids insignificant relative to 
the benefits that would accrue from use of these products. In this 
view, the weight loss benefit would exceed any potential risk from 
taking the product and the risk is not unreasonable when compared to 
the harm caused by obesity. Several comments cited the prevalence of 
obesity and an increase in obesity over time, and urged us not to take 
away one important tool for consumers to address the problem. Two 
comments cited statistics showing that 54 percent of adults are obese 
in the United States, that the prevalence of obesity increased by 30 
percent from 1980 to 1994, and that in 1997 the Centers for Disease 
Control and Prevention (CDC) attributed 42 percent of deaths to 
conditions that typically result from obesity. One comment stated that 
the risks due to obesity are a greater danger than the rare incidences 
of stroke or heart attacks attributed to dietary supplements containing 
ephedrine alkaloids.
    Other comments concluded that dietary supplements containing 
ephedrine alkaloids do not present an unreasonable risk because the 
risks do not outweigh the benefits. They argued that while the benefits 
of dietary supplements containing ephedrine alkaloids are 
substantiated, the adverse events reported are either mild, anecdotal, 
or unsubstantiated and not scientifically valid. Some comments cited 
the RAND report to support the benefit of ephedrine alkaloids for 
short-term weight loss and the lack of adverse effects in clinical 
trials. The comments assert that only a speculative risk for serious 
adverse events exists and that RAND concluded that an assessment of 
case reports is insufficient to reach conclusions regarding causality.
    (Response) We have carefully reviewed the preceding comments, and 
note that many of these issues have been addressed in more detail in 
the scientific evaluation sections V.B and C of this document. Based on 
the scientific data and information discussed in those sections, we 
have concluded that dietary supplements containing ephedrine alkaloids 
present an unreasonable risk of illness or injury under conditions of 
use recommended or suggested in their labeling, or, if no conditions of 
use are suggested or recommended in the labeling, under ordinary 
conditions of use. As discussed in the responses to comments 34 and 35 
of this document, even if we were to extrapolate from data 
demonstrating effectiveness of certain ephedrine drug products when 
considering the reasonably likely benefits of dietary supplements 
containing ephedrine alkaloids, we conclude that the known and 
reasonably likely risks would outweigh even such extrapolated benefits. 
A summary of our rationale for reaching this conclusion is presented in 
our analysis below.
    a. Summary of risks for dietary supplements with ephedrine 
alkaloids. People who use dietary supplements containing ephedrine 
alkaloids are at increased risk for serious adverse events, including 
heart attack, stroke, and death. Susceptible individuals (e.g., those 
with coronary artery disease or heart failure), many of whom may not 
know they have underlying illnesses, are at increased risk for adverse 
events because these products can cause abnormal heart rhythms (pro-
arrhythmic effect), even when the product is ingested at recommended 
doses over a short course (one or a few doses). Over longer periods of 
use, the risk for adverse health effects to the general population, 
including susceptible individuals, increases further due to a sustained 
elevation in blood pressure. This is a characteristic effect of the 
sympathomimetic class of pharmacological compounds. Moreover, the 
results of Boozer, et al. (2002) demonstrate that weight loss achieved 
with botanical ephedrine alkaloids does not produce the expected 
decrease in blood pressure (Ref. 49). The risk of experiencing harmful 
effects from elevated blood pressure increases the longer the blood 
pressure remains high, and such adverse effects are likely to occur 
sooner in individuals with hypertension, many of whom are unaware of 
their illness.
    b. Summary of known and reasonably likely benefits for dietary 
supplements containing ephedrine alkaloids. As discussed in the 
following paragraphs, we conclude, based on all available information 
and data reviewed in this rulemaking, that these products do not 
provide a meaningful health benefit. The best clinical evidence for a 
benefit is for weight loss, but even there the evidence supports only a 
modest short-term weight loss insufficient to positively affect 
cardiovascular risk factors or health conditions associated with being 
overweight or obese. Other possible benefits, such as enhanced athletic 
performance, enhanced energy, or a feeling of alertness, lack 
scientific support and/or they would provide only temporary benefits 
that are trivial in comparison to the risks of serious long-

[[Page 6826]]

term or permanent consequences like heart attack, stroke, and death.
    i. Weight loss. As discussed previously, the RAND report provides 
the most comprehensive review of efficacy studies for ephedrine 
alkaloid containing products. The RAND report found evidence that 
supported an association between short-term use of ephedrine, ephedrine 
plus caffeine, or dietary supplements that contain ephedrine alkaloids 
with or without herbs containing caffeine, and a statistically 
significant increase in short-term weight loss compared to placebo. The 
RAND report concluded that products containing ephedrine alkaloids in 
combination with caffeine resulted in a modest weight loss of 
approximately 2 pounds per month more than placebo over a period of 4 
to 6 months. RAND concluded that the use of ephedrine without caffeine 
was associated with a statistically significant increase in weight loss 
(1.3 pounds of weight loss per month) compared with that of placebo for 
up to 4 months of use. RAND identified a single trial of 3 months 
duration that assessed the effect of herbal ephedra versus placebo. 
Those in the ephedra arm lost 1.8 pounds more per month than did those 
in the placebo arm. We are unaware of any appropriate, well-designed 
studies showing an effect of dietary supplements containing ephedrine 
alkaloids on weight loss for more than 6 months. Such a long-term 
effect would be necessary to translate into health outcome 
improvements.
    Even if there were adequate substantiation that dietary supplements 
containing ephedrine alkaloids produce long-term, sustained weight loss 
in the overweight or obese population, the long-term risks posed by 
these products, particularly in obese patients who may already have 
underlying illnesses that can be aggravated by these products (such as 
hypertension), remain a serious concern. We believe that physician 
supervision is necessary to mitigate the risks associated with the use 
of sympathomimetic products in the long term for weight loss and the 
treatment of obesity, or for any other long-term use. This is achieved 
in part by monitoring patients who use these products and discontinuing 
product use if the patient develops hypertension, experiences other 
adverse health effects, or fails to achieve weight loss that would 
justify continued exposure to the risks associated with use of the 
product.
    People might choose to use a dietary supplement containing 
ephedrine alkaloids to lose weight for purposes other than to improve 
health (e.g., to look slimmer or fit into an outfit for a special 
occasion), and we do not dismiss this use as without value to the 
individual. To achieve the result of modest weight loss, however, these 
products must be used over a period of months. Individuals who use 
these dietary supplements over a period of months for weight loss are 
at risk for the adverse events that can occur with both short- and 
long-term use of these products. These risks are greater than the 
modest benefits described in the RAND report.
    In the case of both short-term and long-term use, any benefits of 
dietary supplements containing ephedrine alkaloids for weight loss are 
outweighed by their risks. Therefore, we conclude that dietary 
supplements containing ephedrine alkaloids labeled or used for weight 
loss present an unreasonable risk.
    ii. Enhancement of athletic performance. The effects of synthetic 
ephedrine on athletic performance were assessed in seven studies that 
were reviewed in the RAND report. Despite the widespread marketing of 
products containing ephedrine alkaloids as performance-enhancers, the 
RAND report found no studies involving botanical ephedrine alkaloids, 
and very limited evidence involving synthetic ephedrine, to support the 
claims. Furthermore, the RAND report concluded that, ``to show even a 
short-term effect of ephedrine, combination with caffeine was 
required.'' Therefore, there is no evidence to indicate that ephedrine 
alone enhances athletic performance. People who use dietary supplements 
containing ephedrine alkaloids for athletic performance are at risk for 
the same serious adverse events as individuals who use these products 
for other indications. As discussed previously in section V.C.2, the 
available evidence regarding a possible benefit from these products for 
enhancing athletic performance is further limited: the supporting 
evidence all comes from studies in which synthetic ephedrine and 
caffeine in combination were administered to healthy males, and the 
modest effects shown were in the very short term only. Even if one 
could disregard all the gaps in the scientific evidence and assume that 
ephedra has the same effect on athletic performance as synthetic 
ephedrine in combination with caffeine, we do not consider a modest, 
temporary enhancement of certain aspects of athletic performance to be 
a benefit sufficient to outweigh the risks of dietary supplements 
containing ephedrine alkaloids. Therefore, we conclude that the use of 
dietary supplements containing ephedrine alkaloids to enhance athletic 
performance for any duration of use present an unreasonable risk.
    iii. Eased breathing and other uses. We have long recognized the 
legitimate short-term oral use of sympathomimetics, such as ephedrine, 
in OTC bronchodilator drug products. These products are marketed for 
those who have been diagnosed with asthma by a physician. The products 
are GRASE when formulated and labeled in accordance with the 
requirements of the final monograph for OTC bronchodilators (21 CFR 
part 341). Mandatory warnings include advising the consumer not to use 
the product unless diagnosed as having asthma by a doctor and not to 
use the product if suffering from heart disease or high blood pressure.
    We are aware that there are dietary supplements containing 
ephedrine alkaloids that are marketed for uses other than weight loss 
or athletic performance enhancement, such as ``eased breathing,'' 
``better breathing,'' ``feel better,'' ``feel more alert,'' 
``energized.'' By contrast to the monograph-compliant OTC 
bronchodilators, and as discussed in section V.B.3 of this document, we 
have seen no data that support any benefit relating to eased breathing 
in healthy people from dietary supplements containing ephedrine 
alkaloids. Moreover, as also discussed in that section, because healthy 
people are able to breathe without difficulty, we do not believe there 
is any respiratory benefit in the absence of a disease state, such as 
asthma or a respiratory infection. At the same time, however, there are 
data that establish the risks of these products. We note that claims to 
treat or mitigate the effects of a disease subject a product to 
regulation as a drug under the act.
    With regard to other claims such as ``feel better,'' ``feel more 
alert,'' and ``energized,'' effects of this nature may be of modest 
benefit to the individual (if they occur), but they are temporary and 
do not improve health. Therefore, such effects would not be sufficient 
to outweigh the risks of dietary supplements containing ephedrine 
alkaloids.
    There are also dietary supplements containing ephedrine alkaloids 
that do not make any specific claims or otherwise suggest or recommend 
conditions of use in their labeling. The use of such products presents 
the same risks and can lead to the same serious adverse events as 
discussed previously for weight loss and athletic performance, even if 
the product is

[[Page 6827]]

taken under ordinary conditions of use (i.e., not abused).
    A dietary supplement labeled for a very temporary, episodic use 
might not present an unreasonable risk if there were adequate evidence 
that the use resulted in a health benefit sufficient to outweigh the 
health risks. Any new indication would still be subject to our post-
market risk evaluation as to whether it could be legally marketed. 
Conclusions regarding the benefit of dietary supplements containing 
ephedrine alkaloids for nondisease claims cannot be drawn solely from 
studies using synthetic ephedrine for specific diseases. Although we 
could require labeling for dietary supplements containing ephedrine 
alkaloids to limit the duration of use, among other things, currently 
there are no data that demonstrate that dietary supplements containing 
ephedrine alkaloids provide a benefit to a particular population when 
used temporarily or episodically (in contrast to OTC ephedrine and 
pseudoephedrine products for disease uses).
3. Conclusion
    Multiple studies demonstrate that dietary supplements containing 
ephedrine alkaloids, like other sympathomimetics, raise blood pressure 
and increase heart rate. These products expose users to several risks, 
including the consequences of a sustained increase in blood pressure 
(e.g. serious illnesses or injuries that include stroke and heart 
attack that can result in death) and increased morbidity and mortality 
from worsened heart failure and pro-arrhythmic effects. Although the 
pro-arrhythmic effects of these products typically occur only in 
susceptible individuals, the long-term risks from elevated blood 
pressure can occur even in nonsusceptible, healthy individuals. These 
risks are neither outweighed by any known or reasonably likely benefits 
when dietary supplements containing ephedrine alkaloids are used under 
conditions suggested or recommended in their labeling, such as for 
weight loss, athletic performance, increased energy or alertness, or 
eased breathing. Nor do the benefits outweigh the risks under ordinary 
conditions of use, in the absence of suggested or recommended 
conditions of use in product labeling. As discussed above in section 
V.C of this document, the best scientific evidence of benefit is for 
modest short-term weight loss; however, such benefit would be 
insufficient to bring about an improvement in health that would 
outweigh the concomitant health risks. The other possible benefits 
discussed in section V.C if this document, have less scientific 
support. Even assuming that these possible benefits in fact occur, such 
temporary benefits are also insufficient to outweigh health risks that 
can lead to serious long-term or permanent consequences like heart 
attack, stroke, and death. On the other hand, we have determined that 
there are benefits from the use of OTC and prescription drug products 
containing ephedrine alkaloids in certain populations for certain 
disease indications that outweigh their risks.
    As with other sympathomimetics, the risks posed by dietary 
supplements containing ephedrine alkaloids for continuous, long-term 
use cannot be adequately mitigated without physician supervision. 
Temporary, episodic use can be justified only if a known or reasonably 
likely benefit outweighs the known and reasonably likely risks. Similar 
to OTC single ingredient ephedrine products, dietary supplements 
containing ephedrine alkaloids could theoretically be marketed without 
physician supervision for a very temporary, episodic use if there were 
adequate evidence that the use resulted in a benefit sufficient to 
outweigh the risks of these products. However, we are currently unaware 
of any such use, and our experience with ephedrine and pseudoephedrine 
OTC drug products suggests that such benefits will be demonstrable only 
for disease uses. Therefore, we conclude that dietary supplements 
containing ephedrine alkaloids present an unreasonable risk of illness 
or injury under conditions of use recommended or suggested in labeling 
or under ordinary conditions of use, if the labeling does not suggest 
or recommend conditions of use.

VI. Why We Conclude that Other Restrictions Would Not Adequately 
Protect Consumers from the Risks Presented by Dietary Supplements 
Containing Ephedrine Alkaloids

    We considered several regulatory alternatives to this final rule. 
As discussed in section I.C of this document, we issued a proposed rule 
in 1997 that would have placed various restrictions on dietary 
supplements containing ephedrine alkaloids. Most of the proposed 
restrictions were withdrawn in 2000; only the proposed prohibition on 
combining ephedrine alkaloids with other stimulant ingredients and the 
proposed warning statement (as modified in FDA's March 2003 notice) 
remain. As discussed in the following paragraphs, we have reached the 
conclusion that those restrictions are inadequate to protect public 
health. In addition, we considered other regulatory alternatives 
presented in the comments received.

A. Warning Statement Alone

    We first proposed a warning statement in the June 1997 proposal. At 
that time, we tentatively concluded that a warning statement was 
necessary to disclose material facts about the consequences of using 
these products, and that it would help to reduce the risk of an adverse 
event after use of dietary supplements containing ephedrine alkaloids 
(62 FR 30670 at 30703). In our March 2003 notice, we reopened the 
comment period to seek, among other things, comments on a revised 
warning statement that we were considering at that time for dietary 
supplements containing ephedrine alkaloids.
    We received a number of comments on the proposed labeling 
requirements in the June 1997 proposal and on the revised warning 
statement in our March 2003 notice. Because we have decided to proceed 
under the adulteration provision in section 402(f)(1)(A) of the act 
rather than to require labeling for dietary supplements containing 
ephedrine alkaloids, these comments are moot to the extent that they 
discuss the substance or format of the warning statement. Nevertheless, 
comments regarding the sufficiency of a warning are relevant to this 
rulemaking.
    (Comment 64) Many comments supported the use of a warning label as 
an effective way to protect public health, although they differed on 
the specific language and format of the warning. Many comments urged us 
to mandate strict warning labels to inform users about the potential 
health risks that have been reported to be associated with the use of 
dietary supplements containing ephedrine alkaloids. One comment stated 
that product labeling does influence user behavior and strongly urged 
us to take action in the form of issuing a mandatory warning label for 
all dietary supplements containing ephedrine alkaloids. Several 
comments stated that there was a significant decrease in the number of 
AERs in certain States after their respective departments of health 
mandated label restrictions and strong cautionary statements. A number 
of comments stated that the warning labels voluntarily adopted and 
already used by industry are sufficient to protect the public from any 
risks. A number of comments proposed different labels to be adopted by 
the entire industry.
    In contrast, many comments maintained that warnings are 
insufficient and recommended a ban of these products. Several comments 
pointed out that serious adverse events

[[Page 6828]]

continue to occur even though most dietary supplements containing 
ephedrine alkaloids already carry warning statements, such as those 
recommended by industry trade groups. For several years, warning labels 
have also been mandated in several states by law or regulation. Many 
comments noted that, in at least 90 percent of the adverse event 
reports submitted to us, consumers reported taking dietary supplements 
containing ephedrine alkaloids as directed on the label.
    A few other comments asserted that warning labels are ineffective 
because serious adverse events have occurred after the initial use or 
after very short-term use of dietary supplements containing ephedrine 
alkaloids. As pointed out in the June 1997 proposal, about 40 percent 
of the 600 AERs reported between 1993 and 1996 occurred with the first 
use or within 1 week of first use, providing little or no warning to 
consumers of risk. Many of the adverse events occurred in individuals 
who had no apparent risk factors, or who were unaware that they were at 
risk.
    Several comments stated that warning labels on ephedrine alkaloid-
containing dietary supplements are not sufficient to protect the public 
health because many people are not aware they have medical conditions 
or individual sensitivities that put them at greater risk for 
experiencing serious adverse effects.
    (Response) We agree that warning statements cannot adequately 
protect consumers from the risks associated with dietary supplements 
containing ephedrine alkaloids. Even if all consumers read the warnings 
and the warnings thoroughly describe the risks, many using these 
products may not be aware they have medical conditions or individual 
sensitivities that put them at greater risk for experiencing serious 
adverse effects. A full discussion of the risks to sensitive 
populations appears previously in the response to comment 22 of this 
document.
    Warning labels may be beneficial when people are able themselves to 
identify the risk factors they have, or when evaluation by a physician 
prior to use can identify whether they have the risk factors and 
further supervision by a physician is not necessary for safe use of the 
product. The purpose of the physician's evaluation is to identify 
individuals with underlying conditions (such as heart failure or 
coronary artery disease) that place them at risk for serious adverse 
events (such as death) due to pro-arrhythmic effects. Such warnings can 
reduce but not eliminate the risks from episodic use of dietary 
supplements containing ephedrine alkaloids because not all susceptible 
individuals can be identified by a physician's evaluation. For example, 
people can have asymptomatic coronary artery disease or early heart 
failure that a physician would not recognize without performing tests 
that would usually be reserved for patients with signs or symptoms of a 
disease. We are not aware of a nondisease claim for which the known and 
reasonably likely benefits of dietary supplements containing ephedrine 
alkaloids would outweigh their known and reasonably likely risks when 
used episodically.
    A warning to consult your physician before use provides even less 
risk mitigation for dietary supplements containing ephedrine alkaloids 
that are used continuously because even healthy people would experience 
a rise in blood pressure and, therefore, be at increased risk for heart 
attack, stroke, and death. At a minimum, continued physician 
supervision would be a necessary risk management tool. Thus, even if 
consumers were to heed warning labels and consult their physician, the 
known and reasonably likely risks of dietary supplements containing 
ephedrine alkaloids when used episodically or continuously would still 
outweigh their known and reasonably likely benefits.
    The conclusion that warning statements are not adequate to protect 
public health is consistent with the fact that, since 1993, we have 
received more than 18,000 AERs (including both adverse events reported 
directly to FDA and the Metabolife call records). The majority of the 
products associated with these AERs contained directions for use and 
warning statements. The warning statements varied from general 
precautions, suggesting that consumers check with a health care 
professional before beginning any diet or exercise program, to more 
specific warning statements, including cautions that consumers not use 
the product if they have certain diseases or health conditions or are 
using certain drugs, and to stop the use of the product if they develop 
certain symptoms. Despite these warning statements in the product 
labeling of dietary supplements containing ephedrine alkaloids, we 
continue to receive reports of serious adverse events.
    (Comment 65) Several comments compared sensitivity to ephedrine 
alkaloids in dietary supplements to sensitivity to food allergens. One 
comment expressed the opinion that the number of individuals sensitive 
to ephedrine alkaloids in dietary supplements is either less than, or 
comparable with, those individuals who suffer from food allergies. One 
comment argued that warning statements are effective for people who 
know they are sensitive to a substance, such as peanuts. The comment 
suggested that if warning labels are considered sufficient in this 
context, they should also be considered sufficient in the context of 
dietary supplements containing ephedrine alkaloids. Another comment 
stated that, with respect to those individuals who are unaware that 
they may have one of the conditions that is contraindicated on the 
label, some misuse due to ignorance is unavoidable and occurs no matter 
what regulations are put in place.
    (Response) We do not agree that individuals sensitive to ephedrine 
alkaloids in dietary supplements are comparable to individuals who 
suffer from food allergies. In the case of food allergies, individuals 
learn that they are allergic to certain foods (e.g., shellfish and 
nuts) and, because we require that the presence of the food ingredients 
be declared on the food label (see 21 CFR 101.4), these individuals can 
then avoid the problem ingredient by reading the food label. The 
physical manifestations of the allergic reaction are usually readily 
recognized by the consumer. In the case of the ephedrine alkaloids, as 
discussed previously in the responses to comments 22 and 27 of this 
document, many individuals are not aware that they are sensitive to 
sympathomimetic agents, such as the ephedrine alkaloids, and may not 
recognize early signs of risk, such as elevated blood pressure or the 
adverse cardiovascular and nervous system effects related to the use of 
ephedrine alkaloids. In most instances, patients with nascent food 
allergies experience classic allergy symptoms, such as tingling lips, 
scratchy throat, wheezing, and shortness of breath, that alert them to 
the development of a particular food allergy, whereas with ephedrine 
alkaloids, severe, life-threatening reactions, may occur at any time, 
even with the first exposure. Therefore, an ingredient declaration or a 
warning label statement cannot assist these consumers in adequately 
reducing their risk of adverse events.

B. Multiple Restrictions

    (Comment 66) Addressing the inadequacy of a warning statement 
alone, many comments supported multiple restrictions (e.g., dosage 
limits, ingredient combination restrictions, duration of use 
restrictions, label claim restrictions, good manufacturing practices 
(GMP) requirements, and warning label statements) to reduce the risk of 
adverse events. One comment pointed out that the frequency, severity, 
and the broad cross section of the

[[Page 6829]]

population for which there are documented adverse events support at 
least this level of regulation. Some comments contended that we should 
establish more stringent regulations. Several of these comments 
recommended that we ban the use of ephedrine alkaloids in dietary 
supplements because of the serious health hazards associated with their 
use and the potential for abuse and misuse of these products.
    (Response) We do not agree that the restrictions recommended in 
these comments will eliminate the risks imposed by dietary supplements 
containing ephedrine alkaloids. As discussed in the response to comment 
26 of this document, we are not aware of any evidence that establishes 
a safe dose of ephedrine alkaloids in dietary supplements. Therefore, 
dose limitations cannot change the unfavorable risk-benefit ratio of 
these products. Similarly, a requirement for a label statement 
recommending that consumers limit the duration of product use will not 
provide adequate protection because adverse events sometimes occur 
after the first use or in the first few days. We also do not agree that 
dietary ingredient restrictions, such as limiting the presence of other 
stimulant ingredients, will eliminate the unreasonable risk associated 
with the use of dietary supplements containing ephedrine alkaloids. As 
explained in section V.B.1 of this document, ephedrine alkaloids given 
alone can be expected to cause significant increases in blood pressure, 
although the presence of other stimulants combined with ephedrine 
alkaloids may increase the risks associated with use of these products. 
Finally, while GMP requirements may ensure consistent quality across 
dietary supplement products containing ephedrine alkaloids, the risks 
attributed to ephedrine alkaloids are due to their inherent 
pharmacological and physiological effects rather than the quality of 
their manufacture, although poor manufacturing could lead to additional 
risks, such as from the introduction of toxic impurities into the 
product.

C. Self-Regulation

    (Comment 67) Other comments objected to the June 1997 proposal, 
arguing that no FDA action is necessary. Several of these comments 
recommended that we take no action but instead continue to monitor 
adverse events. A number of comments stated that the dietary supplement 
industry will self-regulate. These comments argued that several dietary 
supplement trade associations have reacted responsibly to the public 
concerns about the AERs by setting standards for the use of ephedrine 
alkaloids in dietary supplements for their members (Ref. 101).
    (Response) We disagree with the comments that state that no FDA 
action is necessary because the industry will self-regulate. It is 
incumbent upon us to respond to the serious adverse events associated 
with the use of dietary supplements containing ephedrine alkaloids and 
other information about the risks of these products. We have been aware 
for several years that a number of trade associations have policies 
concerning the formulation and labeling of dietary supplements 
containing ephedrine alkaloids. These voluntary industry standards are 
insufficient to alter the risk-benefit ratio for these products. 
Despite the fact that these industry standards are in place, we 
continue to receive reports of clinically significant adverse events 
following the consumption of dietary supplements containing ephedrine 
alkaloids. Some of these adverse events may be due to noncompliance 
with those voluntary standards; however, for the reasons stated in the 
response to comment 39 of this document, these types of standards, even 
if adhered to, would be insufficient to protect consumers from the 
risks posed by dietary supplements containing ephedrine alkaloids.

D. More Education

    (Comment 68) One comment recommended that we provide better 
education to the public on the public health concerns about dietary 
supplements containing ephedrine alkaloids.
    (Response) We do not agree that educating consumers about the 
public health concerns related to the use of dietary supplements 
containing ephedrine alkaloids is an appropriate substitute for this 
regulation. Although we have been active in, and support, consumer 
education activities about these supplements, consumer education will 
not adequately address the risks they present. For example, many 
individuals who are sensitive to sympathomimetic agents, such as the 
ephedrine alkaloids, and are therefore at an increased risk of 
experiencing an adverse event, are not aware that they are at risk. 
Therefore, consumer education would not be expected to greatly reduce 
the risk of adverse events.

E. Nonbinding Guidance

    (Comment 69) Several other comments recommended the issuance of 
nonbinding guidance providing notice to marketers as to which dietary 
supplements containing ephedrine alkaloids would most likely be the 
subject of FDA enforcement. One comment argued that a guidance document 
would conform to our good guidance practices (21 CFR 10.115) and 
provide guidance to the dietary supplement industry as to a level of 
ephedrine alkaloids that can be used in their products with some 
confidence that such products will not be subject to regulatory action. 
In arguing for a guidance document and against a regulation, the 
comment said that a Federal regulation is only appropriate and 
necessary to protect the public health when safe use of a product 
cannot be ensured absent such a regulation; the comment maintained that 
we have not made this showing. One comment stated that the major 
dietary supplement industry trade associations could exhort industry 
compliance to guidelines issued by us or by the trade associations.
    (Response) We disagree that nonbinding guidance would be an 
effective substitute for this rulemaking. As stated previously in this 
document, several industry trade associations have established policies 
concerning the formulation and labeling of dietary supplements 
containing ephedrine alkaloids. These policies are non-binding and 
manufacturers and distributors are under no obligation to comply. 
Moreover, as discussed previously in the responses to comments 39 and 
67 of this document, guidance on labeling or product formulation, even 
if adhered to, would be insufficient to protect consumers from the 
risks posed by dietary supplements containing ephedrine alkaloids.

F. Targeted Enforcement Actions

    (Comment 70) Other comments stated that enforcement actions against 
products containing extremely high levels of ephedrine alkaloids should 
be sufficient to address the problem.
    (Response) We find that individual enforcement actions against 
products containing high levels of ephedrine alkaloids are inadequate 
to protect the public health. Data from the scientific literature and 
AERs indicate that clinically significant adverse effects are not 
limited to the use of products containing high levels of ephedrine 
alkaloids (Refs. 109 and 134). Therefore, enforcement actions against 
products containing only high levels of ephedrine alkaloids would not 
be expected to eliminate the unreasonable risk presented by these 
products. We also

[[Page 6830]]

note that rulemaking is a more efficient regulatory mechanism than 
individual enforcement actions in cases where hundreds of different 
products on the market contain the same ingredient that presents a risk 
to the public health, as is the case here. Without a regulation, we 
would be required to establish our case de novo with witnesses in every 
enforcement proceeding. Multiple proceedings would require multiple 
witnesses and extensive discovery, and would be extremely time-
consuming and burdensome for both the courts and us. However, we point 
out that a regulation is not necessary to find that a dietary 
ingredient or a dietary supplement presents an unreasonable risk.

VII. Miscellaneous Issues

A. Freedom of Choice/FDA Bias

    (Comment 71) Many comments stated that our attempt to regulate 
dietary supplements containing ephedrine alkaloids would erode personal 
freedom and the public's freedom of choice, values that the comments 
maintained were established through the passage of DSHEA. Several 
comments stated that DSHEA gives the public a right to access 
affordable, natural, and effective dietary supplements. A number of 
comments alleged that we issued the June 1997 proposal because we are 
biased against dietary supplements. One industry comment accused us of 
selectively including information in the docket. Several of these 
comments alleged that our purpose for issuing the June 1997 proposal 
was to protect the business interests of the pharmaceutical industry. 
Several comments explained that, if access to dietary supplements for 
weight loss is restricted, consumers will have little choice but to use 
prescription drugs. Many comments from consumers stated that use of 
prescription drugs for weight loss is both more costly and associated 
with more adverse effects than use of products containing natural 
herbs. Many of these comments stated that dietary supplements 
containing ephedrine alkaloids from natural sources are safe and have 
no side effects. Conversely, several comments stated that the 
perception that supplements are natural and, therefore, safe and 
acceptable alternatives to prescribed medications is erroneous and that 
there are serious concerns about the safety and efficacy of these 
products.
    (Response) We deny these allegations of bias against the marketing 
and use of dietary supplements and any allegations of protecting or 
favoring the pharmaceutical industry. We support access to dietary 
supplements that are safe, properly labeled, and in compliance with 
Federal law. However, we are also obligated under DSHEA to protect the 
public against dietary supplements that are unsafe or otherwise 
adulterated. Contrary to one comment's assertion, we did not base our 
decision on selectively chosen information; instead, we considered all 
information that was submitted to the relevant dockets, including more 
than 48,000 comments and hundreds of studies submitted by the dietary 
supplement industry, trade associations, academics, health 
professionals, scientists, public health groups, and consumer groups. 
Given the scientific information about the pharmacology of ephedrine 
alkaloids, clinical studies examining their effects, and AERs, we found 
that there are serious and well-documented public health risks 
associated with the use of dietary supplements containing ephedrine 
alkaloids. Therefore, our obligation under DSHEA is to take action to 
address such risks, particularly in light of the products' lack of 
health benefits.
    Additionally, comments concerning the pharmaceutical industry's 
business interests and possible consumer use of prescription drugs are 
not relevant to our determination as to whether dietary supplements 
containing ephedrine alkaloids are adulterated under section 
402(f)(1)(A) of the act. Section 402(f)(1)(A) of the act focuses 
exclusively on whether the dietary supplement or dietary ingredient 
presents a significant or unreasonable risk; consequently, arguments 
pertaining to other industries or other products have no bearing on 
whether dietary supplements containing ephedrine alkaloids are 
adulterated under the act.

B. Conduct of the Advisory Committee Meetings

    (Comment 72) Several comments stated that we conducted the October 
1995 meeting of the Working Group and the 1996 meeting of the Food 
Advisory Committee (the Committees) in a manner that improperly 
influenced their deliberations and recommendations. These comments 
argued that we instructed the Committee members not to consider certain 
data (e.g., data concerning the use of ephedrine-containing OTC drug 
products for the treatment of asthma); misrepresented certain data 
(e.g., data concerning the AERs and data from clinical trials on the 
use of ephedrine in the treatment of obesity); failed to present data 
that industry believed to be relevant to the evaluation (e.g., number 
of units of products sold during the period of time the AERs were 
received, data regarding whether a cause and effect relationship 
existed between dietary supplements containing ephedrine alkaloids and 
the adverse events reported to us); instructed the Committee to 
evaluate safety using an interpretation of ``significant harm'' (i.e., 
either a large number of adverse events or a serious adverse event in 
one individual) that is not specified in DSHEA; and improperly asked 
the Committee to recommend action to reduce the risks associated with 
the use of these products.
    Other comments argued that the procedures we followed at the 
Committees' meetings were unfair. The comments cited several reasons, 
including the following: FDA materials were not made available to 
dietary supplement industry groups and other interested persons prior 
to the meetings; we were given unlimited time to ``influence'' the 
Committee, and the time others were given to present comments was 
limited; and interested persons were not allowed to question FDA 
officials. For these reasons, several of these comments stated that we 
must reconvene the Committee.
    (Response) We disagree with the comments. The comments concerning 
the data and information we presented or did not present during the 
meetings are without merit because the essence of these comments is 
that they disagreed with our interpretation of the data or preliminary 
conclusions. Presenting our interpretation of the data and our 
preliminary conclusions is entirely appropriate and does not constitute 
undue influence over the Committees (Ref. 137). Interested persons, 
including the dietary supplement industry, were provided with ample 
opportunity to express their views and present data they believed 
relevant to the evaluation during the public hearing portions of the 
meetings or in written comments to the Committees. To the extent that 
specific comments on the data, our interpretation of the data, and our 
preliminary conclusions are relevant to this rulemaking, they are 
addressed in other sections of this document.
    Regarding the conduct of the Committees' meetings, those meetings 
were conducted in accordance with the Federal Advisory Committee Act (5 
U.S.C. App. 2), FDA's implementing regulations (21 CFR part 14), and 
FDA guidance entitled ``Policy and Guidance Handbook for FDA Advisory 
Committees'' (1994) (Ref. 137). We also note that the procedures 
followed during these meetings were no different from the procedures 
used in conducting the numerous advisory committee

[[Page 6831]]

meetings we have held on a variety of other issues.
    We convened the Committees as a means to acquire independent 
scientific and technical advice on the public health concerns 
surrounding the use of dietary supplements containing ephedrine 
alkaloids and on specific ways to address these public health concerns. 
During the meetings, we implemented several safeguards to ensure the 
Committees' independence and fairness to all interested parties.
    First, it was made entirely clear during the meetings that the 
Committees' members were invited to express a view different than ours, 
so that our tentative conclusions could be revised, if necessary. 
During these meetings, we presented a critical and fair evaluation and 
interpretation of the available data. We also expressed our tentative 
conclusions and our concern for the public health. Again, it is 
entirely appropriate for us to state our views and interpretation of 
the data. Furthermore, individual members of the Committees took 
advantage of the many opportunities during the meetings to discuss 
their views and to question FDA officials about the available data, our 
interpretation of the data, and our tentative position.
    Second, the Committees included consumer and industry 
representatives, including two representatives from associations 
representing the dietary supplement industry. The consumer and industry 
representatives represented the views of consumers and industry 
throughout the meeting and made recommendations to us. All FDA-prepared 
materials to be considered by the Committees were sent to all members 
of the Committees, including the dietary supplement industry 
representatives, prior to the meeting.
    Third, the Committees' meetings provided a forum for public 
discussion. Interested persons, including the dietary supplement 
industry, were provided with ample opportunity to express their views 
and present data they believed relevant to the evaluation during the 
public hearing portions of the meetings or in written comments to the 
Committees. During the Committees' meetings, we provided over 2 hours 
of public hearing time, which is twice the time required by our 
regulations (21 CFR 14.29(a)).
    Thus, contrary to the comments' assertions, we provided ample 
opportunity for public participation in the meetings. The public 
hearings were conducted prior to the Committees' deliberations so that 
comments made by interested parties could be considered by the 
Committees in making their recommendations.

VIII. Analysis of Impacts

A. Benefit-Cost Analysis

1. Introduction
    We have examined the economic implications of this final rule as 
required by Executive Order 12866. Executive Order 12866 directs 
agencies to assess all costs and benefits of available regulatory 
alternatives and, when regulation is necessary, to select regulatory 
approaches that maximize net benefits (including potential economic, 
environmental, public health and safety, and other advantages; 
distributive impacts; and equity). Executive Order 12866 classifies a 
regulatory action as a significant regulatory action if it meets any 
one of a number of specified conditions, including having an annual 
effect on the economy of $100 million or more, adversely affecting a 
sector of the economy in a material way, adversely affecting 
competition, or adversely affecting jobs. Executive Order 12866 also 
classifies a regulatory action as significant if it raises novel legal 
or policy issues. We have determined that this final rule is a 
significant regulatory action as defined by Executive Order 12866 
because the benefits of the rule could exceed $100 million per year and 
because the rule raises novel legal and policy issues.
    The Small Business Regulatory Enforcement Fairness Act of 1996 
(Public Law 104-121) defines a major rule for the purpose of 
congressional review as having caused or being likely to cause one or 
more of the following: An annual effect on the economy of $100 million; 
a major increase in costs or prices; significant adverse effects on 
competition, employment, productivity, or innovation; or significant 
adverse effects on the ability of U.S.-based enterprises to compete 
with foreign-based enterprises in domestic or export markets. In 
accordance with the Small Business Regulatory Enforcement Fairness Act, 
the OMB has determined that this final rule will be a major rule for 
the purpose of congressional review because the benefits may exceed 
$100 million annually.
    Title II of the Unfunded Mandates Reform Act of 1995 (Public Law 
104-4) requires cost-benefit and other analyses before any rule making 
if the rule would include a ``Federal mandate that may result in the 
expenditure by State, local, and tribal governments, in the aggregate, 
or by the private sector, of $100,000,000 or more (adjusted annually 
for inflation) in any 1 year.'' The current inflation-adjusted 
statutory threshold is $113 million per year. We have estimated that 
the total cost of this final rule would be no more than $90 million per 
year. Therefore, we have determined that this final rule does not 
constitute a significant rule under the Unfunded Mandates Reform Act.
2. Regulatory Options
    We discussed the following seven regulatory options in the benefit-
cost analysis of the June 1997 proposal: (1) Take no action; (2) take 
no new regulatory action, but generate additional information on which 
to base a future regulatory action; (3) take the actions in the June 
1997 proposal; (4) take the proposed action, but with a higher potency 
limit; (5) remove dietary supplements that contain ephedrine alkaloids 
from the market; (6) take the proposed action, but do not require a 
warning statement; and 7) require a warning statement only (62 FR 30678 
at 30705). We later withdrew all elements of the proposed action except 
the warning statement and prohibition of dietary supplements that 
combine ephedrine alkaloids with other stimulants (65 FR 17474). In 
2003, we issued a March 2003 notice seeking comment on, among other 
things, a revised warning statement consisting of a short warning on 
the PDP and a more detailed warning elsewhere in the product labeling. 
We did not perform any economic evaluation of the revised warning 
statement at that time. We received additional comments on the revised 
warning statement. In addition, the comments on the June 1997 proposal 
suggested some additional options. Considering the options from these 
sources, we address the following options in this analysis: (1) Take no 
new regulatory action; (2) remove dietary supplements containing 
ephedrine alkaloids from the market; (3) require the proposed warning 
statement, as revised in 2003; (4) require a warning statement, but 
modify it or require it only on certain products; and (5) generate 
additional information or take some action other than removing dietary 
supplements containing ephedrine alkaloids from the market or requiring 
warning statements. Executive Order 12866 requires us to analyze 
regulatory options but recognizes that there are practical limits to 
the number of options that we can analyze. The options listed above 
encompass all or most of the significant suggestions raised in the 
comments.
3. Summary of Conclusions
    We have decided to remove dietary supplements containing ephedrine

[[Page 6832]]

alkaloids from the market, identified as option 2 in the previous 
paragraph. We estimate net effects would be between -$47 million and 
$125 million per year from this option, if consumer behavior does not 
already incorporate the health risks posed by these products, and 
between -$90 million and -$7 million per year, if consumer behavior 
already incorporates the health risks. A detailed discussion of all the 
options is provided in the following paragraphs.
4. Option One--Take No New Regulatory Action
    We use this option as the baseline for determining the costs and 
benefits of the other options. Therefore, we do not associate costs or 
benefits with this option. Instead, we discuss the costs and benefits 
of taking no action in the context of the costs and benefits of the 
other options. As we discuss more fully under the other options, the 
expected number of adverse events from these products will probably 
decline, over time, even if we take no regulatory action, for two 
reasons. First, many firms are moving away from the use of ephedrine 
alkaloids because of media coverage of adverse events associated with 
these products, the high cost of liability insurance, and the potential 
for legal actions by consumers. Second, some State and local 
governments have either banned the sale of these products or placed 
various requirements or restrictions on sales of these products.
5. Option Two--Remove Dietary Supplements Containing Ephedrine 
Alkaloids from the Market
    a. Benefits of removing dietary supplements containing ephedrine 
alkaloids from the market. The benefits of this final rule stem from 
the reduction of risks brought about by removing dietary supplements 
containing ephedrine alkaloids from the market. We measure the risk 
reduction, for the purpose of estimating benefits, as the number of 
illnesses and deaths averted. Because OMB's guidance to Executive Order 
12866 calls for quantification of risk reduction, we place special 
emphasis in this part of the document on those AERs that lend 
themselves more readily to quantification.
    As shown earlier in this document, dietary supplements containing 
ephedrine alkaloids would be expected to increase heart rate/rhythm and 
blood pressure. Increasing blood pressure in any population is 
associated with increased probabilities of heart attack, stroke, and 
death, which are the serious adverse events most commonly associated 
with ephedrine alkaloids. The known pharmacological effects of 
ephedrine alkaloids lead us to conclude that removing these dietary 
supplements from the market will reduce the incidence of these adverse 
events. Estimating the likely reduction, however, presents challenges. 
One method used in similar situations is to combine data on exposure 
with a dose-response function to generate estimates of adverse events 
prevented as exposure declines. We cannot use that method here, 
however, because we do not have sufficient data on exposure to 
ephedrine alkaloids from dietary supplements, and we do not know the 
associated dose-response function. Therefore, the best available 
approach, and the method we apply here, is to use AERs to generate 
estimates of the number of adverse events associated with dietary 
supplements containing ephedrine alkaloids.
    It is important to note that the AERs are not the principal 
scientific basis for the regulatory action we selected. Instead, the 
AERs are consistent with the known pharmacological and physiological 
effects of ephedrine alkaloids, as well as the results of clinical 
studies and, therefore, support our finding of unreasonable risk. As we 
explain in more detail later in this document, we use a high barrier 
before admitting an AER as evidence of adverse events associated with 
ephedrine alkaloids. We also use conservative methods to infer the 
total number of adverse events from the reports.
    i. Use of AERs in estimating benefits and baseline number of AERs. 
In the analysis of the June 1997 proposal, we based our estimate of the 
impact of removing dietary supplements containing ephedrine alkaloids 
from the market on the estimated annual number of adverse events caused 
by dietary supplements containing ephedrine alkaloids (62 FR 30678 at 
30705). We based the latter estimate on the average annual number of 
AERs that we received between January 1993 and June 1996, that we 
suspected of having been caused by these supplements, which we 
characterized as the ``baseline number of AERs.'' We then adjusted this 
number of AERs by a series of assumptions designed to reflect various 
sources of uncertainty over whether these supplements actually caused 
those AERs and the uncertainty over the relationship between the AERs 
and the actual number of adverse events associated with the use of 
dietary supplements containing ephedrine alkaloids (including both 
reported and unreported adverse events).
    (Comment 73) A number of comments on the June 1997 proposal 
addressed the issue of the baseline number of AERs. Some comments 
objected to adjusting the number of AERs with assumptions designed to 
reflect uncertainty over the relevance of those AERs. One comment said 
we should have used only those AERs that we were certain had been 
caused by dietary supplements containing ephedrine alkaloids. Other 
comments simply pointed out that some adverse events might not have 
been caused by dietary supplements containing ephedrine alkaloids.
    Some comments suggested that our estimate of the number of adverse 
events based on the number of AERs was inconsistent with the results of 
various studies on the safety of ephedrine alkaloids, herbal ephedra, 
or particular dietary supplements containing ephedrine alkaloids. One 
comment noted that the estimated number of adverse events, particularly 
the estimated number of deaths, was inconsistent with data collected by 
the Drug Abuse Warning Network program, which is administered by the 
Office of Applied Studies in the Substance Abuse and Mental Health 
Services Administration of HHS. Some comments made similar points with 
respect to the inconsistency of our estimated adverse events with the 
lower number of adverse events reported for ephedrine alkaloid-
containing products marketed in foreign countries.
    Several comments suggested that our estimate of the number of 
adverse events was inconsistent with their personal experience. Many 
comments included information on the amount of the product sold or 
estimates of the number of people who consumed the relevant product.
    A number of comments discussed adverse events that purportedly 
would have occurred without consumption of dietary supplements 
containing ephedrine alkaloids. These comments argued that we probably 
generated a large number of irrelevant AERs by asking consumers to 
report ubiquitous symptoms as adverse events that may have been caused 
by these products.
    Some comments criticized the report that RAND prepared for HHS on 
the safety and effectiveness of dietary supplements containing 
ephedrine alkaloids because of its attention to AERs (Ref. 21). One 
comment argued that RAND's approach was inappropriate because GAO had 
previously criticized our use of the AERs in the analysis of the June 
1997 proposal. Other comments supported RAND's attention to AERs. One 
comment argued that RAND did not

[[Page 6833]]

adequately account for preexisting health conditions when classifying 
events in the AERs as ``sentinel'' or ``possibly sentinel'' events. 
Other comments criticized RAND's review of the clinical studies 
involving ephedrine alkaloids. One comment argued that the method RAND 
used to determine which clinical studies to review was biased. Some 
comments argued that the results of RAND's review of the AERs were 
inconsistent with the results of RAND's review of the clinical studies 
because the clinical studies enrolled enough patients to uncover the 
types of adverse events that appear in the AERs, if ephedrine alkaloids 
could cause those types of events. Other comments suggested that 
sources other than the RAND report provide better assessments of the 
risks associated with dietary supplements containing ephedrine 
alkaloids.
    Other comments addressed one or more of the other articles that we 
listed in the March 2003 reopening of the comment period. Many comments 
criticized one or more of those studies on various bases. Other 
comments supported one or more of those studies. One comment argued 
that we presented a biased list of studies because we ignored four 
other articles that were published at about the same time as the 
articles that we listed. Some comments noted that RAND said that 
clinical trials that they reviewed had enrolled enough patients to 
detect serious adverse events at rates of 1 per 1,000 or higher.
    Finally, some comments addressed trends that might affect the 
estimated number of adverse events. Some comments addressed the 
apparent upward trend in the rate at which we received AERs as of 1997, 
which we mentioned in the proposed rule. Some comments suggested that 
the perceived upward trend in AERs at that time may have been caused by 
changes in publicity or in the methods we used to collect adverse 
events, rather than by changes in the number of adverse events. One 
comment noted that many firms had stopped making dietary supplements 
containing ephedrine alkaloids.
    (Response) Although uncertainty remains over the exact number of 
adverse events that are caused by dietary supplements containing 
ephedrine alkaloids, we disagree that, when estimating the number of 
adverse events, we should use only those AERs that we or others have 
proven to have been caused by dietary supplements containing ephedrine 
alkaloids. The comments appear to suggest that we should adopt a 
standard of absolute proof that a dietary supplement caused an 
individual adverse event. However, establishing absolute proof for 
individual cases is very difficult for dietary supplements or most 
other substances other than direct poisons. It is appropriate in the 
case of dietary supplements containing ephedrine alkaloids to estimate 
the number of adverse events prevented by this rule based upon 
scientifically established pharmacological effects of ephedrine 
alkaloids and the clinical and epidemiological evidence. The RAND 
report used the term ``sentinel events'' to describe adverse events 
that involved ephedrine alkaloids and for which RAND could exclude 
alternative explanations for the event with ``reasonable certainty.'' 
If other possible causes could not be excluded, then the report 
classified the cases as possible sentinel events. This level of 
certainty is unusually high in the context of identifying a public 
health risk.
    We also disagree that we should use only clinical studies when 
estimating the number of adverse events. In addition, we disagree with 
the comments that stated that because clinical studies find baseline 
rates for stroke and major cardiac events in excess of 1 per 1,000, the 
existing clinical evidence is sufficient to detect adverse events 
associated with ephedrine alkaloids. The clinical studies reviewed by 
RAND were not large enough to distinguish between effects of ephedrine 
alkaloids and the ordinary variance around the baseline. We, therefore, 
do not agree that existing clinical studies are sufficiently large to 
detect additional adverse events associated with ephedra or ephedrine. 
As discussed in section V.B of this document, the scientific evidence 
identifies the risks presented by dietary supplements containing 
ephedrine alkaloids. For example, a 6-month clinical study examining 
the efficacy and safety of ephedrine alkaloids for the treatment of 
obesity found a statistically significant association between treatment 
with ephedrine alkaloids and higher blood pressure compared to placebo 
(Ref. 49). Higher blood pressure tends to increase the likelihood of 
cardiovascular disease. Thus, the clinical evidence establishes a 
potential mechanism leading from the use of dietary supplements 
containing ephedrine alkaloids to the occurrence of serious adverse 
effects.
    We link the findings from this clinical study and the well-known 
pharmacological effects of ephedrine alkaloids to adverse events to 
establish the likelihood that at least some adverse events reported to 
be associated with the use of dietary supplements containing ephedrine 
alkaloids were in fact caused by these products. Although not as 
rigorous as an epidemiological case control study, this evidence is the 
best available to estimate the benefits of this rule.
    We agree that we should reduce the uncertainty associated with the 
AERs as much as possible and accurately express any remaining 
uncertainty. Therefore, we have replaced the baseline number of AERs 
that we used in the analysis of the proposed rule with the number of 
AERs that RAND identified as sentinel and possibly sentinel events 
involving herbal ephedra. RAND identified 20 sentinel events over a 
period of approximately 9 years from 1992 to 2001, which corresponds to 
an average of about 2 such events per year. RAND also identified 42 
possible sentinel events in this time period, which corresponds to an 
average of about five such events per year.
    We have based our revised estimate on the RAND report because it is 
the most comprehensive review of the information that is currently 
available on the safety and efficacy of dietary supplements containing 
ephedrine alkaloids. However, we acknowledge that considerable 
uncertainty continues to exist with respect to the number of adverse 
events that have been caused by ephedrine alkaloids. We have attempted 
to reflect the continuing uncertainty by updating the assumptions we 
used in the analysis of the June 1997 proposal, as we discuss in the 
following paragraphs.
    We did not attempt to forecast trends in the number of adverse 
events in the analysis of the June 1997 proposal, and we have not done 
so in this analysis. Forecasting trends in the number of adverse events 
would be difficult, and any such forecasts would be associated with 
large uncertainty ranges. Although we recognize that some firms may 
have recently discontinued the use of ephedrine alkaloids in some or 
all of their products, we have insufficient information to revise the 
results of the RAND report on that basis.
Assumptions used in analysis of the final rule
First assumption
    Ninety percent to 100 percent of the sentinel events and 50 percent 
to 100 percent of the possible sentinel events identified in the RAND 
report were caused by dietary supplements that we suspect contained 
ephedrine alkaloids.
    (Comment 74) A number of comments addressed the first assumption. 
One comment suggested that we should have set the lower bound of the 
first assumption to zero because it was possible that none of the AERs 
had been

[[Page 6834]]

caused by dietary supplements containing ephedrine alkaloids. Some 
comments provided their own estimates of the number of AERs that had 
been caused by those supplements.
    (Response) We have revised our estimate of the baseline number of 
AERs using the number of sentinel and possible sentinel cases 
identified in the RAND report in order to address the concerns that 
these comments raised about causation and the presence of ephedrine 
alkaloids with respect to some of the AERs that we used as a basis for 
our benefit estimates in the analysis of the June 1997 proposed rule. 
Although RAND stressed that it could not conclude that these events 
were definitely caused by ephedrine alkaloids and declined to make any 
probabilistic statements about causality, the definitions that it used 
for sentinel and possible sentinel events suggest that those AERs have 
a relatively high probability of having been caused by ephedrine 
alkaloids. Therefore, we have revised the assumption concerning the 
proportion of the AERs that were caused by dietary supplements from 80 
percent to a range of 90 percent to 100 percent for sentinel events and 
50 percent to 100 percent for possible sentinel events.
Second assumption
    One hundred percent of the sentinel and possible sentinel events 
that were caused by dietary supplements that we suspect contained 
ephedrine alkaloids involved dietary supplements that did, in fact, 
contain ephedrine alkaloids.
    (Comment 75) Other comments addressed the second assumption. One 
comment reported that an industry review of the 920 AERs in the docket 
found that more than 123, or 13 percent, involved products for which 
there was no indication that the product contained ephedrine alkaloids. 
One comment was from a firm that claimed it had informed us during FAC 
meetings that nearly 25 percent of the AERs that involved their 
products involved products that did not, in fact, contain ephedrine 
alkaloids.
    (Response) One of the criteria that RAND used to identify sentinel 
and possible sentinel events was documentation that the person that 
suffered the adverse event had consumed a dietary supplement containing 
ephedra within 24 hours prior to the adverse event. The assumption in 
the proposed rule that 80 percent of the AERs involved products that 
contained ephedrine alkaloids applied to the set of AERs used in that 
analysis. RAND has documented that all of the sentinel and possible 
sentinel events it reviewed involved products containing ephedrine 
alkaloids. Documentation of the presence of ephedrine alkaloids varied 
from case to case, and included blood tests of the person who suffered 
the adverse event, chemical analysis of capsules, and labeling of the 
products consumed. RAND did not consider self-reports alone to be 
sufficient documentation for sentinel and possible sentinel events. 
Because we use the RAND study as the basis for the analysis of this 
final rule, the 80 percent assumption is no longer relevant. In the 
analysis of this final rule, we assume that 100 percent of the AERs 
involved products that contained ephedrine alkaloids.
Third assumption
    AERs represented 10 percent of the actual number of adverse events.
    (Comment 76) Some comments argued that our assumption of a 10 
percent reporting rate was too low. Some comments argued that people 
are more likely to overreport than underreport adverse events involving 
dietary supplements containing ephedrine alkaloids for various reasons, 
including FDA's public statements and media coverage of this issue. One 
comment argued that people are more likely to overreport than 
underreport serious adverse events such as heart attack, stroke, 
seizure, psychotic events, and death, because people tend to consider 
any temporal connection equivalent to a causal connection. However, 
this comment suggested that people probably underreport minor adverse 
events. Some comments noted that the AERs that we discussed in the June 
1997 proposal appeared to arrive in discrete groups as though in 
response to inciting events, such as FDA press releases. One comment 
noted that, of the 22 AERs in the docket that involved their products, 
we received two-thirds of those AERs within 1 week of our April 1996 
press release, and we received the other one-third over a much longer 
period of 30 months. Some comments suggested that the 10 percent 
assumption might be appropriate for passive reporting systems, but 
argued that the reporting system that we used to generate the AERs was 
not passive because both the Texas Department of Health and FDA took 
various steps to solicit AERs. Two comments discussed estimates of 
reporting rates for a passive adverse event reporting system in 
Britain. One comment estimated the reporting rate for serious adverse 
events at 50 percent. Another comment estimated the same rate at 10 
percent. Both comments estimated that the system had a much smaller 
reporting rate of 2 percent to 4 percent for nonserious adverse events. 
Some comments noted that we assumed a 50 percent reporting rate in our 
report on Eosinophilia-Myalgia Syndrome, which was an outbreak level 
event (Ref. 138). These comments noted that this report referred to 
adverse events related to a dietary supplement, L-tryptophan, which had 
also received significant media publicity. These comments argued that 
it was, therefore, a reasonable model to use for the ephedrine alkaloid 
situation. Some comments suggested that we revise our reporting rate 
assumption from 10 percent to a range of 10 percent to 50 percent.
    Other comments argued that our assumption of a 10 percent reporting 
rate was too high. Some comments argued that people are more likely to 
underreport than overreport adverse events involving dietary 
supplements containing ephedrine alkaloids for various reasons, such as 
not wanting to acknowledge using the product. One comment noted that a 
2001 report from the Office of the Inspector General of HHS concluded 
that current surveillance systems for identifying adverse reactions 
from dietary supplements probably detect less than 1 percent of adverse 
reactions (Ref. 20). However, another comment claimed that most 
researchers consider a reporting rate of less than 1 percent to reflect 
a worst-case scenario. One comment noted that the report that suggested 
a reporting rate of less than 1 percent did not differentiate between 
serious and nonserious adverse events. This comment argued that the 
reporting rate for serious adverse events is probably higher than for 
nonserious adverse events.
    (Response) In order to express the continuing uncertainty over the 
reporting rate, we have calculated benefits based on reporting rates of 
10 percent, 50 percent, and 100 percent of sentinel and possible 
sentinel events. Although the reporting rate could be lower than 10 
percent, the severity of the adverse events under consideration and the 
level of media coverage suggest that the reporting rate may be 10 
percent or higher. The assumed 100 percent reporting rate generates a 
lower bound number of adverse events. We selected 50 percent as an 
intermediate number. We used a 10 percent reporting rate in our summary 
statements to simplify the presentation of the results and because 10 
percent reporting appears to be a reasonable point estimate, taking 
into account the seriousness and media coverage of these adverse events 
and the estimated reporting rates of 1 percent or lower for adverse 
events involving drugs (Refs. 32 and 139). The 10 percent reporting 
rate applies to serious events only, and incorporates the fact that a

[[Page 6835]]

report of a serious adverse event had to fulfill the RAND criteria in 
order to be included as a sentinel or possible sentinel event. We did 
not consider nonsentinel events in the analysis, as explained in the 
following paragraphs.
    ii. Valuing reductions in adverse events.
    (Comment 77) Some comments addressed the values that we placed on 
eliminating various types of adverse events in the analysis of the 
proposed rule. One comment objected to the value of $5 million that we 
placed on one fewer fatality per year across the affected population, 
which is sometimes called the value of a statistical life. This comment 
described this value as the value of an average life and argued that 
this figure is unrealistic because the average person does not have $5 
million.
    (Response) In its guidelines on performing economic analysis of 
federal regulations under Executive Order 12866, OMB noted that the 
term ``statistical life'' can lead to some confusion. It pointed out 
that this term refers to the sum of risk reductions expected in a 
population, as expressed in the following example: If the annual risk 
of death is reduced by one in a million for each of two million people, 
that represents two ``statistical lives'' saved per year (two million x 
one in one million = two). If the annual risk of death is reduced by 
one in 10 million for each of 20 million people, that also represents 
two statistical lives saved (Ref. 140). Similarly, the estimated value 
of a statistical life (VSL) is based on the willingness to pay for 
relatively small reductions in the risk of premature death for many 
people summed across a population. The individual risk management 
decisions on which we base estimates of the VSL must reflect the budget 
constraints of those individuals making those decisions. However, the 
resulting VSL need not reflect the budget constraints of the average 
person. We have revised the VSL in this analysis to a range of $5 
million to $6.5 million to reflect the latest estimates of this figure 
(68 FR 41433 through 41506, July 11, 2003).
    In addition, we have revised our method of estimating the values of 
avoiding the other health endpoints. For nonfatal myocardial infarction 
(MI), we used the same procedure that we used in our analysis of the 
proposed rule on trans fatty acids (64 FR 62772, November 17, 1999). 
That method was based on estimating the sum of the medical costs, the 
cost of functional disability, and the cost of pain and suffering. This 
method assumes that someone suffering a nonfatal MI will have 
functional disability or pain and suffering or both in every year after 
the year following the MI. We estimated the loss per year to be 0.2 
quality adjusted life years (QALYs) every year of life following the 
MI. We did not include any reduction in life expectancy due to the MI. 
For this rule, we based the years of disability or pain and suffering 
on the ages of those suffering nonfatal myocardial infarction in the 
RAND report (Ref. 141). RAND reported summary information on age by 
type of adverse event using three age categories (13 to 30, 31 to 50, 
and 51 to 70). We took the midpoints of the three age categories and 
constructed a weighted average based on the proportion of people 
suffering that adverse event in those categories. We then compared that 
age to an average life expectancy in the United States in 2001 of 77.2 
years to determine the years of disability or pain and suffering or 
both (Ref. 142).
    We used a similar procedure to estimate new values for strokes. To 
estimate combined functional disability and pain and suffering we used 
a 0.2 quality adjusted life year (QALY) loss per year after a stroke 
(Ref. 143). We used the same QALY losses for ``other cardiovascular'' 
events that we used for nonfatal MI. We were unable to find information 
on chronic QALY losses for acute cases of ``other neurological,'' 
``seizure,'' or ``psychiatric'' adverse events. For medical costs, we 
used 2001 National Statistics from HCUPnet (Ref. 144). We provide 
summary information on these values in table 1 of this document.
    (Comment 78) Some comments that discussed the background rates of 
expected but unexplained adverse events argued that many AERs involved 
people with underlying health conditions and that dietary supplements 
containing ephedrine alkaloids might have simply precipitated adverse 
events that would have occurred within a short time anyway.
    (Response) As we indicated previously in this document, we have 
revised our estimate of the number of relevant AERs to reflect the RAND 
report. The definition that RAND used for sentinel events involved 
investigating alternative explanations and excluding them with 
reasonable certainty. However, the definition that RAND used for 
possible sentinel events included cases where another condition by 
itself could have caused the adverse event, but for which the known 
pharmacology of ephedrine made it possible that ephedra or ephedrine 
may have helped precipitate the event. We have reflected the 
uncertainty over causality in the first of the three assumptions that 
we discussed above. We assume that dietary supplements containing 
ephedrine alkaloids caused 90 percent to 100 percent of sentinel events 
and 50 percent to 100 percent of possible sentinel events.
    iii. Serious versus minor adverse events.
    (Comment 79) Some comments suggested that some AERs that we used in 
the analysis of the June 1997 proposal involved events that we should 
not have classified as adverse events. These comments argued that these 
events involved expected side effects of ephedrine alkaloids that are 
both minor and transient.
    (Response) We discussed adverse events that we classified as ``less 
serious'' in the analysis of the proposed rule (62 FR 30678 at 30708). 
However, we indicated that the value of eliminating those adverse 
events contributed very little to total estimated benefits. RAND did 
not include these types of more minor adverse events in its sentinel 
and possible sentinel event cases. Although it did find evidence that 
products that contained both ephedrine alkaloids and caffeine increased 
the risk of certain minor adverse events, it noted that it was unable 
to distinguish the effects of the ephedrine alkaloids and the caffeine. 
Based on these considerations, we have not attempted to address adverse 
events beyond those that RAND identified as sentinel and possible 
sentinel events.
    iv. Risks of substitutes and weight regain.
    (Comment 80) Some comments argued that consumers would face similar 
or greater health risks if they switched from dietary supplements 
containing ephedrine alkaloids to alternative weight loss solutions, 
such as prescription weight-loss drugs, other dietary supplements, or 
weight loss surgery.
    Some comments discussed what would happen if consumers stopped 
using dietary supplements containing ephedrine alkaloids and did not 
switch to equally effective alternative weight loss methods. Some 
comments discussed the extent and rising trend of obesity in the United 
States. Some comments noted that obesity increases the risk for heart 
attack, stroke, diabetes, and cancer. However, other comments argued 
that any countervailing health costs that would result if people 
stopped using dietary supplements containing ephedrine alkaloids to 
lose weight would be small or nonexistent. Some comments suggested 
there were no clear health benefits from the amount of weight loss that 
the RAND report attributed to dietary supplements

[[Page 6836]]

containing ephedrine alkaloids. Other comments disagreed and argued 
that there were clear health benefits from the amount of weight loss 
that the RAND report attributed to dietary supplements containing 
ephedrine alkaloids. One comment argued that, although people often 
regain weight that they lose during a diet program, people who have 
participated in diet programs nevertheless generally maintain lower 
weights than those who have not.
    (Response) Subtracting the value of countervailing health effects 
posed by substitute products and activities from the value of the 
health benefits from removing dietary supplements containing ephedrine 
alkaloids from the market to obtain the net health benefits is 
consistent with our approach for estimating benefits. (For purposes of 
this economic impact analysis, ``health benefits'' refers to an 
improvement to health and is not synonymous to the ``benefits'' that we 
mention in our risk-benefit analysis for purposes of determining that 
these products present an unreasonable risk of illness or injury; 
``health benefits'' are a type of ``benefit'' we consider when making 
an unreasonable risk determination.) Our full conceptual model of 
benefits is as follows: (net change in risk from the reduction in 
intake of ephedrine alkaloids x value per unit change in risk) + (net 
change in risk from substitute products and activity x value per unit 
change in risk) + (net change in risk from weight gain x value per unit 
change in risk) + (any net change in risk from the small impact on 
wealth from the cost of substitute products or activity x value per 
unit change in risk).
    However, we do not have sufficient information to estimate all 
elements of this model. In the analysis of the June 1997 proposal , we 
noted one article that found that a product a firm had reformulated to 
remove ephedrine alkaloids had lost approximately 33 percent of its 
previous sales (Ref. 145). Since that time, a media report discussed 
another reformulated product that had greater sales than the original 
product (Ref. 146). Therefore, we estimate that from two-thirds to all 
of the consumers of these supplements would probably switch to other 
dietary supplements that firms market for the same purposes as dietary 
supplements containing ephedrine alkaloids. This implies that between 
one-third and none of the consumers of these products would switch to 
entirely different types of weight loss or performance enhancing 
substitutes.
    Some manufacturers have already reformulated dietary supplements so 
that products that had contained ephedrine alkaloids now contain 
alternative ingredients. Some of these reformulated products contain 
Citrus aurantium L., which is a source of synephrine, and caffeine, 
sometimes in the form of green tea extract. Synephrine is a 
sympathomimetic agent, and these agents are a class of compounds that 
also includes ephedrine alkaloids. A number of other potential herbal 
sources of sympathomimetics probably exist. These ingredients may pose 
risks that are similar to those of ephedra. If consumers switched to 
substitute products containing these ingredients, similar health risks 
might be expected as those with products containing ephedrine 
alkaloids. Some other ingredients that have been reported in 
reformulated products include cocoa beans, yerba mate, cinnamon twig, 
and galangal.
    The estimated none to one-third of the consumers of dietary 
supplements containing ephedrine alkaloids who would switch to products 
other than other dietary supplements might switch to alternatives that 
carry either health risks or benefits. Some of those who consumed these 
supplements for weight loss may seek medical care to obtain 
prescription weight loss medications or for weight loss surgery. 
However, only some of these consumers would qualify for these medical 
treatments. These treatments would carry health risks that might be 
equal to, or greater than, the risks of ephedrine alkaloids. Only the 
risks that remain after accounting for the management of risk under 
physician supervision would be relevant in this context. In addition, 
these treatments may be more expensive than dietary supplements. The 
resulting relatively small reductions in the overall wealth of those 
who switch to more expensive alternatives could also generate small 
countervailing health risks because they have less disposable income to 
spend on other risk-reducing activities.
    Other consumers interested in weight loss may switch to meal 
replacements or other diet products rather than seek medical treatment. 
Other consumers might choose to do nothing and simply forego the weight 
loss they may have obtained with ephedra products. This foregone weight 
loss could, in theory, generate health costs. The lack of health 
benefits from the weight loss associated with the use of these 
products, however, implies that these health costs, if any, would be 
negligible. Finally, some consumers might choose to reduce their 
caloric intake or increase their caloric output through additional 
exercise. These consumers would obtain additional health benefits 
beyond eliminating the risk of adverse events associated with dietary 
supplements containing ephedrine alkaloids. Those who consume 
supplements containing ephedrine alkaloids to enhance their athletic 
performance and who do not switch to other dietary supplements marketed 
for that purpose might switch to other stimulants, including black 
market products containing ephedrine alkaloids or methamphetamines. 
These products would pose health risks equal to or greater than those 
of currently marketed dietary supplements containing ephedrine 
alkaloids.
    We have insufficient information to quantify the effects of 
switching to alternative weight loss or athletic performance enhancing 
products or activities, or to quantify the health costs associated with 
the absence of weight loss that might be achieved using dietary 
supplements containing ephedrine alkaloids.
    v. Risks of certain dietary supplements containing ephedrine 
alkaloids.
    (Comment 81) A number of comments suggested that certain dietary 
supplements containing ephedrine alkaloids do not pose any health 
risks. These comments addressed this point in the context of exempting 
certain products from the proposed warning statement. However, these 
comments are also relevant to the issue of exempting certain products 
from a regulation removing dietary supplements containing ephedrine 
alkaloids from the market. Therefore, we discuss these comments under 
this option.
    Several comments argued that we should not treat ephedrine 
alkaloids in Chinese herbal formulas that are used in Chinese medicine 
treatment protocols the same as dietary supplement products containing 
ephedrine alkaloids that consumers use to lose weight or enhance 
athletic performance. One comment suggested that warning statements are 
unnecessary for herbal products that firms distribute to ``healthcare 
professionals,'' including members of the American Herbalists Guild. 
Some comments suggested that we should set different regulatory 
requirements for different products or product types because risks vary 
by product or product type.
    (Response) The RAND report found little scientific agreement on the 
dose-response relationship for ephedrine alkaloids (Refs. 21 and 22). 
Therefore, we are unable to estimate the impact of exempting products 
from this rule based on the level of ephedrine alkaloids that they 
contain. As we discussed earlier in the preamble, we have determined 
that botanical sources of ephedrine alkaloids

[[Page 6837]]

in traditional Asian herbal therapies are not covered by this rule. We 
do not have sufficient information to estimate the impact of exempting 
products based on the other considerations suggested in the comments, 
including type of product, label warnings, or directions for use.
    b. Revised benefit estimates. Based on the preceding discussion, we 
have revised our estimate of the benefits of removing dietary 
supplements containing ephedrine alkaloids from the market. The social 
benefits of removing dietary supplements containing ephedrine alkaloids 
from the market consist of the increase in consumer utility that would 
be generated by any net health benefits resulting from removing dietary 
supplements containing ephedrine alkaloids from the market. The 
following table 1 of this document provides an estimate of the number 
of the various types of serious adverse events that we might eliminate 
by removing dietary supplements containing ephedrine alkaloids from the 
market, along with an estimate of the utility loss prevented by that 
reduction. As we discussed previously, benefits could be much lower and 
potentially zero if the health risks posed by substitute weight loss or 
sports performance products, such as other dietary supplements 
containing sources of sympathomimetics, were comparable to the health 
risks posed by ephedrine alkaloids.
    We convert the number of deaths prevented into a monetary estimate 
by multiplying by the number of deaths by the VSL. We convert the 
number of nonfatal events prevented into a monetary estimate by 
multiplying the number of nonfatal events by the value of the 
appropriate change in quality QALYs. Acute events that do not have 
clear chronic effects will generate only minimal losses in terms of 
QALYs. We calculated the total benefits for each class of adverse 
events as: (Number of deaths prevented) x ($ per fatal case); and 
(number of nonfatal cases prevented) x (($ per QALY x QALY loss) + 
medical costs per case)). The benefits for the first year would be 
slightly different from the benefits in every subsequent year because 
the effective date is 60 days after the publication date of the final 
rule. By convention, we calculate benefits starting from the 
publication date of the final rule. Therefore, the benefits in the 
first year will be 5/6 (or 83 percent) of the benefits of every 
subsequent year. To simplify the discussion, we use the benefits for 
every year after the first year in all summary discussions.

    Table 1.--Annual Number of Sentinel and Possible Sentinel Events
   Prevented Under Option Two (Removing Dietary Supplements Containing
  Ephedrine Alkaloids from the Market), with QALY and Medical Cost per
                                  Case
------------------------------------------------------------------------
                     Annual Number      QALY Loss Per  Medical Costs per
      Type             Prevented            Case              Case
------------------------------------------------------------------------
Death              0.7 to 1.2           NA (used VSL)  $25,742
MI (heart          0.6 to 1.0           0.29           $30,586
 attack)
CVA (stroke)       1.5 to 2.1           0.2             $20,898
Other              0.1 to 0.2           0.29            $30,586
 Cardiovascular
 (e.g.
 Cardiomyopathy,
 Ventricular
 Tachycardia)
Other              0.1                  minimal         $13,212
 Neurological
 (e.g. Transient
 Ischemic
 Attack)
Seizure            0.5 to 0.9           minimal         $11,812
Psychiatric        0.9 to 1.3           minimal         $6,927
------------------------------------------------------------------------
Note. All dollar values in this document represent 2003 prices.


  Table 2.--Annual Benefits of Option Two (Removing Dietary Supplement
  Containing Ephedrine Alkaloids from the Market) Based on Alternative
 Assumptions of Reporting Rates and Values of Preventing Adverse Events,
                          Rounded to $ Millions
------------------------------------------------------------------------
 Value of Avoiding       Adverse Event Reporting Rate ($ in millions)
  Fatal Cases and   ----------------------------------------------------
    QALY Losses         10 percent        50 percent       100 percent
------------------------------------------------------------------------
$ per fatal case =    $43 to $73        $9 to $15         $4 to $7
 $5 million $ per
 QALY = $100,000
$ per fatal case =    $53 to $91        $11 to $18        $5 to $9
 $6.5 million $ per
 QALY = $100,000
$ per fatal case =    $56 to $93        $11 to $19        $6 to $9
 $5 million $ per
 QALY = $300,000
$ per fatal case =    $66 to $112       $13 to $22        $7 to $11
 $6.5 million $ per
 QALY = $300,000
$ per fatal case =    $80 to $132       $16 to $26        $8 to $13
 $6.5 million $ per
 QALY = $500,000
------------------------------------------------------------------------

    c. Costs of removing dietary supplements containing ephedrine 
alkaloids from the market. In the analysis of the proposed rule, we 
identified the costs that would be generated by removing dietary 
supplements containing ephedrine alkaloids from the market as the one-
time cost of reformulating and relabeling products that currently 
contain ephedrine alkaloids, plus the utility loss for those consumers 
who would need to switch from their preferred option (consuming these 
products) to their next most preferred option (consuming an alternative 
product or taking some other type of action) (62 FR 30678 at 30709). In 
that analysis we did not estimate utility losses for consumers. A 
number of comments stressed this cost but did not provide estimates of 
it. Nevertheless, we have revised the analysis by attempting to 
quantify this cost.
    Theoretically, we could measure the utility loss for consumers by 
looking at the difference between their willingness to pay for products 
containing ephedrine alkaloids and their willingness to pay for 
alternative supplements or other substitute products or activities. 
However, we do not have sufficient information to implement this 
approach, and may never have a direct measure of the utility loss in 
this market. Instead, we attempt to measure indirectly the utility loss 
for consumers of these products. We assume that the premium that these 
consumers are willing to pay to consume dietary supplements containing 
ephedrine alkaloids rather than whatever they perceive to be the next 
closest alternative is between 1

[[Page 6838]]

percent and 10 percent of the sales price of the dietary supplements 
containing ephedrine alkaloids. This range is based on the fact that 
some premium must exist if consumers prefer these products to 
alternatives. We selected 1 percent as a lower bound because we did not 
find any large price differences between products containing ephedrine 
alkaloids and those that did not contain ephedrine alkaloids. Of 
course, it is possible that current consumers place a much higher 
premium on products containing ephedrine alkaloids than consumers who 
have already switched to alternatives. To allow for that possibility, 
we selected 10 percent (a substantial premium) as the upper bound of 
the range. Current market prices do not provide sufficient information 
for a more precise estimate. This estimate of the utility loss assumes 
that consumers do not incorporate the expected utility losses from 
potential adverse events in their willingness to pay for dietary 
supplements containing ephedrine alkaloids. If consumers already 
incorporate this information in their purchasing decisions, then it 
would be inappropriate to compare the value of the health benefits to 
the estimated utility losses for consumers using willingness to pay 
because the willingness to pay would already account for any adverse 
health effects. In that case, the estimated utility loss from the 
removal of these products from the market would represent the full net 
loss of utility.
    A recent article estimated that the sales of ``herbal products'' 
containing ephedra accounted for between 4.3 percent and 13.5 percent 
of the sales for all herbal products (Ref. 135). The article did not 
define ``herbal products,'' but it noted that their use of the phrase 
``herbal products'' included products that a natural products 
information company had classified as ``vitamins/supplements'' and 
``grocery'' items rather than as ``herbal products'' (Ref. 147). 
Therefore, these estimates may have included products other than 
dietary supplements. Another source argued that the estimates presented 
in the article that we discussed previously in this paragraph did not 
include all relevant products. The source claimed that more 
comprehensive data from the Nutrition Business Journal showed that the 
sales of products containing herbal ephedra accounted for 33 percent of 
the total sales of all herbal products and 7.5 percent of the total 
sales of dietary supplements (Ref. 148). Both of these articles 
apparently dealt only with products that contained herbal ephedra. 
Ephedrine alkaloids are also contained in a number of different plants, 
including Sida cordifolia L., and Pinellia ternata (Thunb.) Makino. 
Therefore, these articles may have underestimated the number of 
products that contained ephedrine alkaloids. These articles did not 
present actual sales figures for herbal products, dietary supplements, 
or products containing ephedra. However, the Nutritional Business 
Journal estimated that the sales of all dietary supplements and all 
herbal dietary supplements in 2002 were $18 billion and $4.3 billion, 
respectively (Ref. 149). If one assumes that ``herbal dietary 
supplements'' corresponds to ``herbal products,'' then total sales of 
dietary supplements containing ephedrine alkaloids would be $185 
million to $1,419 million.
    In an effort to reduce this range, we estimated the sales of these 
products based on a recent survey that showed that 2 million consumers 
used these products at some point during a given week (Ref. 150). We 
assumed that consumers who used these products at some point during a 
given week probably used the products every day during that week, 
because most of the labels we have examined say that the product should 
be taken daily, or several times per day. We also assumed that the 
particular week under study was comparable to any other week. 
Therefore, we assumed that 2 million consumers use these supplements 
per day. We then multiplied this number of consumers by the average 
daily cost of these supplements, which we estimated from a sample of 30 
dietary supplements containing ephedrine-alkaloids that we found on the 
Internet. Based on the recommended intake levels appearing on the 
labels of these products, the corresponding estimated total sales per 
year is $559 million to $806 million. The costs in the first year after 
publication of the rule would be slightly different from the cost in 
every subsequent year because the effective date is 60 days after the 
publication date of the final rule. Therefore, the utility losses in 
the first year will be 5/6 (or 83 percent) of the losses of every 
subsequent year. To simplify the discussion, we use the benefits for 
every year after the first year in all summary discussions.
    Earlier, we assumed that the consumer utility loss from switching 
from an ephedra-based product to the next closest substitute would be 
from 1 percent to 10 percent of the sales price at the current level of 
consumption. Under this assumption and our estimate of total sales, the 
consumer utility loss associated with removing dietary supplements 
containing ephedrine alkaloids from the market would be $6 million to 
$81 million per year. The loss of consumer utility would probably 
decline over time as consumers find more substitute products and as 
producers develop new, more acceptable substitute products. Eventually, 
consumer substitutions and product development could drive this cost to 
zero. We have insufficient information to estimate the rate at which 
this cost would decline over time.
    In the analysis of the June 1997 proposal, we estimated relabeling 
costs of $3 million to $60 million and product reformulation costs of 
$0 million to $25 million, for a total cost for these two activities of 
$3 million to $85 million (62 FR 30678 at 30709). We did not receive 
any comments on these estimates. We have, however, revised the analysis 
to incorporate a new model for estimating reformulation costs that we 
developed after publication of the proposed rule (Ref. 151). According 
to that model, reformulation costs with a 12-month reformulation period 
would be $7 million to $78 million. In deriving that figure, we assume 
that reformulating dietary supplements would not be as complicated as 
reformulating most other types of food and cosmetics. In particular, we 
assume that reformulating dietary supplements would include the 
following cost generating activities: Idea generation, product 
research, analytic testing, packaging development, plant trials, 
startup, and lost inventory. We assume that reformulating dietary 
supplements would not include the following types of cost generating 
activities: Process development, coordinating activities, consumer 
tests, shelf life studies, any type of safety studies, and market 
tests. If all of these other steps were involved, then estimated 
reformulation costs for a 12-month reformulation period would be $22 
million to $142 million. We assume that 6 months is the most likely 
time period for reformulation if dietary supplements containing 
ephedrine alkaloids are removed from the market. Although the effective 
date of this rule is 60 days after the publication date, we do not 
expect that many firms will try to condense the reformulation process 
into a 60-day period. Some firms may have already done some of the 
preliminary work for reformulation. Other firms might need to withdraw 
their product from the market in the period between the effective date 
and the date at which they complete their reformulation. FDA's 
reformulation cost model does not address costs for a reformulation 
time of 6 months, so we

[[Page 6839]]

extrapolated the costs based on the proportionate change in cost that 
would result from halving the reformulation time from 24 months to 12 
months. Under that extrapolation, we estimate that reformulation costs 
for a 6-month reformulation period would be $10 million to $100 
million. We annualize these estimated costs over 20 years at an 
interest rate of 3 percent to convert these one-time costs to a yearly 
cost of $1 million to $7 million. Annualizing these costs over 20 years 
at an interest rate of 7 percent gives an annual cost of $1 million to 
$9 million.
    We summarize the annual costs of this option in table 3 of this 
document. We compare the benefits and costs of this option in table 4 
of this document. To obtain the higher bound estimate of net benefits, 
we start with the higher bound estimate of benefits and subtract the 
lower bound estimates of costs. To obtain the lower bound estimate of 
net benefits, we start with the lower bound estimate of costs and 
subtract the higher bound estimate of costs. If consumer behavior 
already incorporates health risks, then utility costs would already be 
net of health benefits. In that case, the net impact of this rule is 
simply the total costs.

    Table 3.--Annual Costs of Option Two (Removing Dietary Supplement Containing Ephedrine Alkaloids from the
                                          Market) Rounded to $ Millions
----------------------------------------------------------------------------------------------------------------
       Type of Cost                                      Cost (rounded to $ millions)
----------------------------------------------------------------------------------------------------------------
Utility Losses for          $6 to $81
 Consumers
Product Reformulation       $1 to $9
----------------------------------------------------------------------------------------------------------------


   Table 4.--Annual Social Benefits and Costs of Option Two (Removing
   Dietary Supplement Containing Ephedrine Alkaloids from the Market)
                          Rounded to $ Millions
------------------------------------------------------------------------
                                         Benefit or Cost (rounded to $
      Type of Benefit or Cost                     millions)
------------------------------------------------------------------------
Health Benefits (for 10 percent       $43 to $132
 reporting rate)
Cost of Utility Losses for            $6 to $81
 Consumers
Cost of Product Reformulation         $1 to $9
Net Effect (if consumer behavior      -$47 to $125
 does not already incorporate
 health risks)
Net Effect (if consumer behavior      -$90 to -$7
 already incorporates health risks)
------------------------------------------------------------------------

    d. Distributional issues and impact on industry. In the analysis of 
the June 1997 proposal, we estimated that removing dietary supplements 
containing ephedrine alkaloids from the market would reduce the sales 
of dietary supplements containing ephedrine alkaloids by between $200 
million and $230 million per year (62 FR 30678 at 30710). We discussed 
reduced sales because, in that analysis, we characterized a 
reformulated product as the same product as before reformulation for 
purposes of describing the impact of the proposed action (although the 
reformulated products would obviously not be the same as the products 
they replaced). We did not receive comments that would require us to 
change those estimates. However, we have revised the analysis to 
reflect the fact that the effect on accounting profit is a more 
appropriate way to conceptualize the potential distributional impact 
than reduced sales. We can use the same information that we used to 
estimate consumer utility losses to consider the likely effect on the 
profits of firms that currently produce dietary supplements containing 
ephedrine alkaloids. In 2001, the average accounting profit for all 
Fortune 500 companies was about 5 percent of revenue, and some 
pharmaceutical firms had profit rates as high as 19 percent of revenue 
(Ref. 150). Profit rates for firms in the dietary supplement industry 
are probably toward the low end of this scale because of the low 
barriers to entry for this industry. Therefore, we assume that the 
profit rate for dietary supplement manufacturers is about 5 percent of 
total sales. As we discussed previously, press accounts suggest that 
manufacturers that have reformulated their dietary supplements to 
eliminate ephedrine alkaloids have experienced declines in sales 
ranging from about one-third to no decline in sales. We previously 
estimated total sales to be $559 million to $806 million. Therefore, we 
estimate that sales may decrease by $0 to $269 million per year. 
Assuming that the profit rate is 5 percent of sales, removing dietary 
supplements containing ephedrine alkaloids from the market would 
generate accounting profit losses of $0 to $13 million per year. We 
classify this impact as a transfer and not a social cost because 
removing dietary supplements containing ephedrine alkaloids from the 
market would increase the profits of firms that produce and distribute 
substitute products. If these other firms also have an average profit 
rate of 5 percent of sales, then the profit gained by these companies 
would also equal $0 to $13 million per year.
    In addition to causing a potential reduction in profits for firms 
currently producing dietary supplements containing ephedrine alkaloids, 
removing dietary supplements containing ephedrine alkaloids from the 
market might also generate some countervailing transfers through the 
elimination of insurance costs and lawsuits associated with products 
containing ephedrine alkaloids. Eliminating legal fees and court costs 
would also generate social benefits. Of course, if reformulated 
products were eventually found to pose health risks comparable to those 
found for ephedra-based products, then these effects (i.e., the 
elimination of insurance and legal costs) would eventually decrease to 
zero. A recent press report found that ephedra manufacturers or 
distributors have settled 33 cases since 1994 and that an additional 42 
cases were pending (Ref. 152). This represents 75 cases over 9 years, 
or about 8 cases per year. Recent awards for cases that have gone to 
court have ranged from $2.5 million to $13 million (Refs. 152 and 153). 
The figures reported in the media for cases that were settled out of 
court were considerably lower. One such case was settled out of court 
for $25,000 (Ref. 152). If removing dietary supplements containing 
ephedrine alkaloids from the market eliminated 8 cases per year, then 
it would decrease transfer payments from firms to consumers by between 
$0.2 million per year, if all cases were settled out of court, and $104 
million per year, if all cases were lost in court at the high end of 
the range of legal penalties.
    One company noted in 2002 that its product-liability insurance 
increased by $2.1 million from 2001 to 2002 (Ref. 146). If all 30 
manufacturers saw this increase in insurance premiums, then the total 
increase in insurance premiums would be $60 million. Some of the 
independent distributors might also face higher insurance rates, but we 
have insufficient information to estimate those costs. Insurance rates

[[Page 6840]]

would not necessarily increase at this same rate in the future, and 
they could decrease. Therefore, we will assume that this adjustment in 
insurance rates reflects a one-time adjustment in the perceived 
liability risks associated with these products. If these higher 
premiums were unnecessary for reformulated products, then removing 
dietary supplements containing ephedrine alkaloids from the market 
would generate a one-time reduction in private costs of $60 million. 
However, if reformulated products were eventually shown to pose risks 
comparable to those for ephedra-based products, then insurance rates 
might increase to a comparable level for these products.
    The uncertainty ranges associated with the potential transfers of 
accounting profits make it impossible to estimate the impact of 
removing dietary supplements containing ephedrine alkaloids from the 
market on the firms that currently produce and distribute dietary 
supplements containing ephedrine alkaloids. Firms that are unable or 
unwilling to produce or sell substitute products would suffer losses, 
and firms that are able and willing to produce or sell substitutes 
might not suffer decreases in profits. Indeed, media reports suggest 
that many firms have already voluntarily withdrawn their ephedra-based 
products and replaced them with reformulated products to avoid the high 
legal and insurance costs associated with dietary supplements 
containing ephedrine alkaloids (Ref. 146).
6. Option Three--Require the 2003 Proposed Warning Statement
    a. Benefits of requiring the 2003 proposed warning statement 
comparison to removing dietary supplements.
    i. Containing ephedrine alkaloids from the market. In the analysis 
of the June 1997 proposal, we noted that estimating the benefit of 
limiting our regulatory action to requiring the 1997 proposed warning 
statement involved a potentially controversial value judgment about how 
one evaluates risks that consumers voluntarily accept in the presence 
of adequate warning statements (62 FR 30678 at 30711). Our analysis of 
a mandatory warning statement is further complicated by the fact that 
the labels of most dietary supplements containing ephedrine alkaloids 
already bear warning statements.
    (Comment 82) One perspective that we discussed in the analysis of 
the June 1997 proposal was that adverse events that occur despite the 
presence of adequate warning statements are not social costs but are 
instead private costs that reflect informed decisions about the private 
benefits and costs of using these products. A number of comments agreed 
with this perspective. One comment argued that consumers have a 
responsibility to read and follow warnings and instructions for use on 
products that they consume. Some comments suggested that we should 
expect consumers to read and follow warning statements, and we should 
not hold manufacturers liable if consumers fail to do so. One comment 
argued that we have adopted that viewpoint in other cases involving 
products that can produce severe adverse effects. Some comments from 
consumers argued that we should take no regulatory action other than 
requiring a warning statement because that approach would allow 
consumers to decide whether or not to assume the risks associated with 
these products. One comment pointed out that a recent report on the 
safety of ephedrine alkaloids that was sponsored by industry endorsed 
this perspective, as expressed in the following quote: ``As the law 
appropriately suggests, the FDA cannot assume responsibility for 
protecting the public from themselves, if they choose to use this or 
any other product at higher than recommended levels or otherwise misuse 
properly labeled products.''
    The other perspective on warning statements that we discussed in 
the analysis of the June 1997 proposal was that adverse events that 
occur despite the presence of adequate warning statements represent 
social costs. Under this perspective, requiring a warning statement 
would not be a sufficient regulatory action unless it actually caused 
consumers to change their behavior so as to eliminate any adverse 
events associated with these products. Some comments supported this 
perspective by arguing that warning statements are inappropriate or 
inadequate because they probably would not significantly reduce the 
number of adverse events among all or some subset of consumers.
    (Response) In the analysis of removing dietary supplements 
containing ephedrine alkaloids from the market, we concluded that 
removing dietary supplements containing ephedrine alkaloids from the 
market would generate net social benefits if consumers fail to 
incorporate the probability of adverse events into their demand for 
those products. Our assessment of the effects of a warning statement 
hinges on the same uncertainty. If consumers do not fully incorporate 
the risk of adverse events into their demand for products containing 
ephedrine alkaloids, and if the proposed warning label would cause at 
least some consumers to change their demand so as to incorporate the 
risk, then the warning label could reduce adverse events and generate 
net social benefits. The likelihood of that outcome depends on the 
effectiveness of current warning statements and of warning statements 
in general. One consideration that suggests that consumers fail to 
incorporate, at least in part, the probability of adverse events into 
their market behavior is that some consumers do not know they have the 
underlying conditions discussed in warning statements.
    ii. Comparison to existing warning statements. In economic terms, 
the benefit of changing a warning statement is the value that consumers 
place on the change in the information available on product labels. If 
we had information on how consumers value different warning statements, 
then we would not need to consider the impact of changing the warning 
statements on adverse events. Without that information, we must infer 
the value from the adverse health effects that changing the warning 
statement would eliminate. This value represents the minimum value of 
changing the warning statements: Consumers who change their behavior in 
response to the change in warning statements would presumably be 
willing to pay the amount that they saved in health costs and lost 
utility because of that change in warning statements, but some 
consumers might value the information even though they do not change 
their behavior. Because the information value for consumers who do not 
change their behavior is likely to be small, the value of the 
eliminated adverse events is probably a close approximation to the 
value of changing the warning statements. Therefore, we have based our 
analysis on estimating the impact on adverse events of changing the 
warning statements from the existing voluntary industry warning 
statements to the proposed mandatory warning statement.
    iii. Effectiveness of warning statements in eliminating adverse 
events. In the analysis of the June 1997 proposal, we estimated that 
the warning statement that we proposed in 1997 would reduce the 
estimated number of annual adverse events caused by dietary supplements 
containing ephedrine alkaloids by 0 to 15 percent (62 FR 30678 at 
30712).
    (Comment 83) A number of comments addressed this estimate. One 
comment suggested that the estimated impact was too low and noted that 
a recent study showed that almost 70 percent of adults read product 
labels every time they use

[[Page 6841]]

a product. However, another comment argued that warning statements 
would probably be ineffective because most consumers do not read 
product labels. This comment noted that there is no evidence that 
warning labels on alcohol and tobacco products reduced consumption of 
those products. Other comments simply pointed out that warning 
statements might not eliminate all adverse events, because some 
consumers might not read or follow them. One comment provided a number 
of reasons why warning statements might be ineffective at reducing 
adverse events (e.g. many consumers do not read labels for OTC drugs 
and would be even less likely to do so for dietary supplements, many 
consumers base their usage patterns on suggestions read in magazines 
rather than on label information, many consumers believe consuming more 
of a dietary supplement makes it more effective). Another comment noted 
that we appeared to infer the ostensible benefit of warning statements 
rather than demonstrating their effectiveness through carefully 
conducted clinical trials. This comment also argued that warning 
statements would not be useful for consumers with unrecognized medical 
conditions that might predispose them to adverse reactions caused by 
ephedrine alkaloids, such as hypertension, hyperthyroidism, vascular 
malformations of the brain, and subclinical cardiac arrhythmias. One 
comment suggested that the proposed warning statement was too long to 
be effective. This comment claimed that the necessary print size and 
spacing would discourage some consumers from reading the warning 
statement.
    (Response) These comments did not provide sufficient information to 
allow us to change our estimate of the effectiveness of the warning 
statement that we originally proposed in 1997 and revised in 2003. The 
comments that noted that warning statements might not eliminate all 
adverse events are consistent with the assumption that warning 
statements would eliminate 0 to 15 percent of the adverse events. The 
comment that noted a study that showed 70 percent of consumers read 
product labels every time they purchase a product did not provide a 
reference for that study, but the reported results are consistent with 
other studies. The FDA 2002 Health and Diet Survey found that 80 
percent of nonvitamin/mineral supplement users reported that they used 
product labels to find out if there were side effects or drug 
interactions associated with a product or if anyone should avoid the 
product. A survey of consumer use of dietary supplements by Prevention 
Magazine found that the following percentages of herbal remedy shoppers 
reported looking for the following types of information: 72 percent for 
possible side effects; 70 percent for warnings for people not to take 
the supplement, e.g. pregnant women; 65 percent for warnings about 
possible interactions with prescription medicines; and 59 percent for 
warnings about possible interactions with OTC products (Ref. 154). 
However, consumers who read warning statements will not necessarily 
change their behavior. A 2002 recent survey of consumers who have 
recently taken OTC pain medications found that 84 percent read at least 
some of the label the first time they took a product but that 44 
percent said they took more than the recommended dosage, despite the 
warnings on the label (Ref. 155). In general, most of the literature on 
warning statements has not focused on product purchase or use pattern 
decisions but on issues such as comprehension, awareness, and 
believability (Ref. 156). Some articles have found that alcohol warning 
statements have had little or no impact on behavior (Ref. 157). 
However, these results do not necessarily hold for the proposed warning 
statement because the effectiveness of warning statements varies with a 
number of considerations, including the content and format of the 
warning and the characteristics of the consumers reading the warning. 
Thus, this literature does not provide a basis for revising our 
assumption that the proposed warning statement will reduce adverse 
events by 0 to 15 percent. However, the fact that most dietary 
supplements already bear extensive warning statements suggests that 15 
percent is probably an upper bound and that a value closer to 0 percent 
is probably more likely.
    (Comment 84) Some comments argued that the proposed warning 
statement would probably have little effect on the number of adverse 
events because many dietary supplements that contain ephedrine 
alkaloids already bear warning statements. One comment argued that some 
existing warning statements fully and accurately describe the potential 
for adverse effects and thereby satisfy the objectives of the proposed 
warning statement. One comment argued that some existing warning 
statements are more complete than the proposed warning statement. 
However, one comment said that the proposed warning statement would 
probably be more effective than existing warning statements because 
existing warnings do not alert consumers to avoid taking multiple 
products containing ephedrine alkaloids at the same time.
    (Response) To address these comments, we reviewed and compared the 
labels of forty dietary supplements containing ephedrine alkaloids that 
we collected between March 20 and May 30, 2001, and also compared the 
number of adverse reports received during the period January 2000 to 
January 2004 as warning labels appeared on certain dietary supplements. 
(Ref. 158) All of the product labels bore some sort of warning 
statement. Most warning statements had many of the same basic elements 
as the proposed warning statement. For example, most existing warnings 
listed various conditions under which consumers should not take the 
product, various conditions under which consumers should see a health 
care provider before taking the product, and side effects or symptoms 
that should lead consumers to consult with a health care provider. 
However, the specific content of the various elements varied quite a 
bit both among existing warning statements and between existing warning 
statements and the proposed warning statement. Some elements of the 
proposed warning statement were common in existing warning statements; 
other elements were less common. For example, none of the existing 
product labels carried a PDP warning statement. In contrast, most 
product labels carried some sort of warning for people who had 
previously experienced heart problems. In addition, parts of some 
existing warnings were more strongly worded than the corresponding 
parts of the proposed warning. In other cases, parts of the proposed 
warning were more strongly worded than the corresponding parts of 
existing labels. Our label comparison did not support the notion that 
the proposed warning statement would have no effect because it was 
identical to existing warning statements. The comparison did suggest 
that the proposed warning statement is similar in many respects to 
existing warning statements, and that the proposed warning statement 
might not reduce adverse events very much. This result is consistent 
with the assumption that the proposed warning statement might eliminate 
between 0 and 15 percent of adverse events.
    (Comment 85) Some comments argued that the proposed warning 
statement would be ineffective because some States already require 
warning statements, and the presence of multiple warning statements 
would confuse consumers.

[[Page 6842]]

    (Response) Multiple warning statements might reduce the impact of 
the proposed warning statement. However, many different warning 
statements might be more effective than one or a few. The comments did 
not provide sufficient information to enable us to revise our estimate 
of the effectiveness of the proposed warning statement based on the 
possibility that some products might face multiple labeling 
requirements.
    b. Revised benefit estimates. When we revise the analysis as 
described previously, we obtain the estimated benefits shown in table 5 
of this document. The assumption underlying the table is that the 
proposed warning statement would cause some proportion of consumers to 
incorporate the risks from dietary supplements containing ephedrine 
alkaloids into their demand for these products. Some proportion of 
those consumers (0 to 15 percent) would cease using those products, 
which would reduce the number of adverse events by a like percentage. 
The benefits would therefore be some percentage (between 0 and 15 
percent) of the benefits of removing dietary supplements containing 
ephedrine alkaloids from the market. The results presented in table 5 
of this document apply to every year after the first year. Benefits for 
the first year would be lower because our proposed rule would have 
allowed firms up to 6 months to comply with the warning statement 
requirements. We do not know the actual rate at which firms would come 
into compliance during the initial 6 months after publication of a rule 
finalizing the proposed warning statement requirements. To simplify the 
analysis, we assume that it would take all firms 6 months to comply 
with such a rule. Under this assumption, the benefits in the first year 
would be half those of every year after the first year. In the summary 
of regulating options and table 8 of this document, we use the range $0 
to $20 million for annual benefits (excluding the first year) because 
it is inconsistent with the presentation of the other options.

  Table 5.--Annual Benefits of Option Three (Require the 2003 Proposed
 Warning Statement) Based on Eliminating 0 to 15 Percent of the Sentinel
                      and Possible Sentinel Events
------------------------------------------------------------------------
                                       QALY Loss Per      Medical Costs
      Type             Number              Case             Per Case
------------------------------------------------------------------------
Death                   0.0 to 0.2      NA (used VSL)            $25,742
MI (heart               0.0 to 0.2               0.29            $30,586
 attack)
CVA (stroke)            0.0 to 0.3                0.2            $20,898
Other                          0.0               0.29            $30,586
 Cardiovascular
 (e.g.
 Cardiomyopathy
 , Ventricular
 Tachycardia)
Other                          0.0            minimal            $13,212
 Neurological
 (e.g.
 Transient
 Ischemic
 Attack)
Seizure                 0.0 to 0.1            minimal            $11,812
Psychiatric             0.0 to 0.2            minimal             $6,927
------------------------------------------------------------------------


  Table 6.--Annual Benefits of Option Three (Require the 2003 Proposed
 Warning Statement) Based on Alternative Assumptions of Reporting Rates,
                          Rounded to $ Millions
------------------------------------------------------------------------
                                   Adverse Event Reporting Rate
 Value of Avoiding Fatal -----------------------------------------------
  Cases and QALY Losses      10 percent      50 percent     100 percent
------------------------------------------------------------------------
$ per fatal case = $5          $0 to $11        $0 to $2        $0 to $1
 million $ per QALY =
 $100,000
$ per fatal case = $6.5        $0 to $14        $0 to $3        $0 to $1
 million $ per QALY =
 $100,000
$ per fatal case = $5          $0 to $14        $0 to $3        $0 to $1
 million $ per QALY =
 $300,000
$ per fatal case = $6.5        $0 to $17        $0 to $3        $0 to $2
 million $ per QALY =
 $300,000
$ per fatal case = $6.5        $0 to $20        $0 to $4        $0 to $2
 million $ per QALY =
 $500,000
------------------------------------------------------------------------

    c. Costs of requiring the 2003 proposed warning statement.
    i. Label Costs.
    (Comment 86) Some comments said that the proposed PDP or nonPDP 
warning statements are too long to fit on the labels of most dietary 
supplement products. One comment noted that firms package many 
``traditional style extracts'' in containers that have a maximum label 
size of 1.75 x 3.75 inches, or about 6.6 square inches. The comment 
argued that the proposed warning statements cannot fit on a label of 
this size. One comment argued that the proposed warning statement would 
take up so much space on the label that firms would be able to provide 
very little other information on the label. One comment argued that 
there is not enough room on package labels for multiple warning 
statements and suggested that we clarify that our proposed warning 
statement would preempt any state labeling requirements.
    (Response) We reviewed the labels of the 40 dietary supplements 
containing ephedrine alkaloids that we collected between March 20 and 
May 30, 2001, to investigate label size. Most labels were wrap-around 
adhesive labels with a minimum label size of about 7.5 square inches 
and an average of about 22.8 inches. Nearly all labels already bore 
extensive warning statements, and most of the content of the existing 
warning statements was distinct from the additional warning material 
required by some States. Therefore, we conclude that the proposed 
warning statements would probably have fit on most product labels. 
However, some dietary supplements containing ephedrine alkaloids, 
possibly including traditional style extracts, might have significantly 
smaller labels than the products that we collected. If we had adopted 
this option, we would have addressed this possibility in a number of 
ways. Firms that cannot fit the proposed PDP warning statement on the 
PDP if they use the normal font size would be able to use a smaller 
font size. Firms that cannot fit the nonPDP warning statement on the 
product labels could place the warning statement on any product 
labeling that is an integral part of the outer product packaging such 
that consumers may read the warning statement at the point of purchase, 
including the rise backing, panel extension, and outsert. In some 
cases, firms may already use these packaging features. These firms 
would simply need to revise the content of existing

[[Page 6843]]

labeling. In other cases, firms might need to change the style of their 
packaging to utilize these types of labels. Rather than changing the 
style of their packaging, firms could also change the size of the 
package to increase the label space available for the warning 
statement. Changing the product packaging in one of these ways might 
require some firms to purchase new packaging machinery, which would be 
an additional cost beyond the cost of the label changes that we 
discussed in the analysis of the June 1997 proposal. We have 
insufficient information to estimate the number of products that might 
need to take these steps. Based on our review of existing product 
labels, we estimate that the number of such products is probably very 
small.
    We have reestimated labeling costs because we have new information 
on the number of dietary supplements containing ephedrine alkaloids and 
we have updated the labeling cost model that we used to estimate 
labeling costs in the analysis of the June 1997 proposed rule. The cost 
of changing labels varies with the amount of time that we give firms to 
change the labels. We previously proposed setting the effective date 
for this option to be 180 days after the publication of the final rule. 
According to the revised label cost model, the one-time cost of adding 
or revising a PDP and a nonPDP warning statement to the labels of all 
dietary supplements under a 6-month compliance period would be 
approximately $140 million to $319 million. The labeling cost model 
does not differentiate dietary supplements that contain ephedrine 
alkaloids from other dietary supplements. However, a database of 
dietary supplements compiled by Research Triangle Institute (RTI) under 
contract to FDA listed a total of 3,000 dietary supplement products in 
1999, and 49 of those products, or about 2 percent, listed ephedrine or 
one of the following sources of ephedrine alkaloids in their ingredient 
lists: Ephedra, ephedra extract, ephedra herb, Ephedra sinica Stapf., 
ma huang, ma huang extract, ma huang herb, ma huang concentrate, or ma 
huang herb extract (Ref. 159). In the absence of other information, we 
assume that the cost of changing the labels of these products would be 
about 2 percent of the cost of changing all dietary supplement product 
labels. Therefore, we estimate that the one-time cost of changing the 
labels of dietary supplements containing ephedrine alkaloids is $3 
million to $6 million. Annualizing this cost over 20 years at 3 percent 
gives an annual cost that rounds to $0 million per year; that is, less 
than $500,000 per year. Annualizing this cost over 20 years at 7 
percent gives an annual cost of $0 million to $1 million.
    ii. Risks of substitutes/absence of weight loss.
    (Comment 87) One comment noted that the proposed warning statement 
would instruct consumers not to take dietary supplements containing 
ephedrine alkaloids before or during strenuous exercise. This comment 
argued that this element of the warning statement could harm consumers 
by inhibiting weight loss because exercise is an essential component of 
a weight loss program.
    (Response) As we discussed under Option Two of this section, we 
have insufficient information to estimate countervailing health effects 
such as the health risks generated by the use of substitute products or 
by the reduction or elimination of weight loss benefits. However, for 
this option, we have calculated benefits as a range of $0 to $20 
million. This range is consistent with the existence of countervailing 
health risks from the source suggested by this comment.
    d. Effective date.
    (Comment 88) Some comments recommended that we revise the proposed 
effective date for the warning statement that we proposed in 1997 and 
revised in 2003. One comment suggested that we set the effective date 
to 12 months after publication of the final rule, rather than the 
proposed 180 days after publication of the final rule, to give industry 
more time to comply with the labeling requirements. Another comment 
suggested that we set the effective date to 60 days after publication 
of the final rule. Some comments suggested that we base the effective 
date on labeling at the manufacturing site. Under this approach, we 
would require products leaving the manufacturing site after the 
effective date to bear the warning statements, but firms could continue 
to sell existing inventory without the warning statement after that 
date.
    (Response) Setting the effective date to 12 months after 
publication of a final rule requiring the warning statement would lead 
to one time labeling costs of between $2 million and $5 million. 
Annualizing this cost over 20 years at 3 percent and 7 percent gives an 
annual cost that rounds to $0 million per year (i.e., less than 
$500,000 per year). This would also reduce benefits in the first year 
to $0 under the simplifying assumption that all firms would take 12 
months to comply with the required warning statement.
    Eliminating all costs associated with unusable label or package 
inventory by allowing firms to continue to sell product without the 
warning statement after the effective date would lead to compliance 
costs of $2 million to $6 million under the proposed 180 day compliance 
period. Annualizing this cost over 20 years at 3 percent gives an 
annual cost that rounds to $0 million per year (i.e., less than 
$500,000 per year). Annualizing this cost over 20 years at 7 percent 
gives an annual cost of $0 million to $1 million per year. In our 
summary statements, we present the cost estimates under the 7 percent 
discount rate because that range includes the range of costs that we 
estimated under a 3 percent discount rate. However, this option would 
also generate additional enforcement costs because we would need some 
way of determining that the products that firms sell without the 
warning statement were actually labeled before the effective date. In 
addition, this revision would reduce benefits over a number of years 
according to the proportion of products sold during that time that did 
not bear warning statements. The period over which benefits would be 
reduced could be quite large because firms might produce as much 
product as possible prior to the effective date to avoid having to meet 
the labeling requirements. The comments did not provide information on 
this issue, and we are unable to estimate this reduction in benefits.
    We compare costs of different effective dates for the proposed 
labeling option in table 7 of this document. We only consider first 
year net benefits because changing the effective date from 180 days to 
365 days only affects benefits in the first year. After the first year, 
annual benefits would be the same for either effective date. To obtain 
the higher bound estimate of net benefits, we start with the higher 
bound estimate of benefits and subtract the lower bound estimates of 
costs. To obtain the lower bound estimate of net benefits, we start 
with the lower bound estimate of costs and subtract the higher bound 
estimate of costs. We do not have information suggesting that any of 
these options would lead to greater net benefits than the proposed 
enforcement period of 180 days.

[[Page 6844]]



    Table 7.--Comparison of Effective Date Options for Option Three (Require the Proposed Warning Statement),
                                              Rounded to $ Millions
----------------------------------------------------------------------------------------------------------------
                                           Annualized Cost        First Year Benefits    First Year Net Benefits
            Effective Date                    (millions)               (millions)               (millions)
----------------------------------------------------------------------------------------------------------------
180 days                                              $0 to $1                $0 to $10               -$1 to $10
365 days                                                    $0                       $0                       $0
180 days at manufacturing site              $0 plus additional                       NA                       NA
                                             enforcement costs
----------------------------------------------------------------------------------------------------------------

    e. Conclusions on the benefits and costs of 2003 proposed warning 
statement. We estimate costs to include the one-time cost of changing 
the labels of dietary supplements containing ephedrine alkaloids to be 
$3 million to $6 million, which rounds to approximately $0 million per 
year (i.e. less than $500,000 per year) when annualized over 20 years 
at 3 percent and approximately $0 million to $1 million per year when 
annualized over 20 years at 7 percent. We are unable to quantify 
potential recurring countervailing health costs. We estimate the 
recurring annual benefit to be $0 to $20 million, depending on the 
reporting rate for adverse events, and the method used to value those 
events. Therefore, we estimate the annual net benefit of this option to 
be -$1 million to $20 million. In the long run, this option would 
probably generate net benefits, for two reasons: First, the benefits 
recur annually and any non-zero level of benefits will eventually 
surpass the one-time labeling cost. Second, as we discussed above, the 
recurring countervailing health costs are unlikely to exceed the 
recurring health benefits.
7. Option Four--Require the Proposed Warning Statement, But Modify it 
or Require it Only on Certain Products.
    a. Require warning only for certain products. We discussed a number 
of comments under Option Two that claimed that certain dietary 
supplements containing ephedrine alkaloids do not pose any health 
risks. That discussion is also relevant in the context of exempting 
certain products from the proposed warning statement. The summary of 
those comments and our response is the same as under Option Two in 
section VIII.A.5 of this document. For example, one comment suggested 
that warning statements are unnecessary for herbal products that firms 
distribute to ``healthcare professionals,'' including members of the 
American Herbalists Guild. We do not have sufficient information to 
estimate the impact of exempting products based on patterns of 
distribution or other product characteristics.
    b. Placement and format of warning statement.
    (Comment 89) Some comments addressed the placement of the proposed 
warning statement on product packages. Some comments suggested that we 
allow firms to use inserts, stickers, or ``peel away'' labels. One 
comment said that we should allow firms to use alternative methods of 
disseminating warning information if they dispense products that are 
part of a bulk decoction formula that lacks standard labeling, such as 
products compounded and dispensed in Chinese herbal medicine pharmacies 
or by ``qualified health professionals.''
    (Response) According to the March 2003 notice, we proposed to allow 
firms to use special labeling for the nonPDP warning statement as long 
as consumers could read the warning statement at the point of purchase.
    (Comment 90) Some comments objected to the PDP warning statement 
that was part of the revised warning statement that we proposed in 
2003. Other comments supported the 2003 proposed PDP warning statement. 
Some comments suggested that we require firms to use the PDP warning 
statement on both the product container and the outside container or 
wrapper of the retail package. One comment suggested that we require 
firms to include the PDP warning statement in any promotional 
literature and advertising.
    (Response) Eliminating the PDP warning statement but retaining the 
nonPDP warning statement would probably significantly reduce the impact 
of the proposed warning statement. The PDP warning statement was one of 
the main elements of the proposed warning statement that differed from 
most existing warning statements. Reducing the impact of the warning 
statement by eliminating the proposed PDP warning statement would 
reduce both the benefits and the distributive impacts of the warning 
label option. However, eliminating the PDP warning statement would have 
little impact on the overall cost of changing labels to comply with the 
proposed warning statement because firms would still need to change 
labels even if we did not require a PDP warning statement. Requiring 
firms to place the warning statement on both the product container and 
the outside container or wrapper and requiring firms to include it in 
any promotional literature and advertising might increase the impact of 
the warning statement, but would also increase the costs. The comments 
did not provide sufficient information to establish that the benefits 
from these revisions would outweigh the costs.
    (Comment 91) One comment argued that the PDP for mail order dietary 
supplements corresponds to the front page of any product literature 
that a firm uses to advertise its product. This comment said that the 
proposed regulation would, therefore, require some firms to change 
their pamphlets and other material. The comment argued that such a 
requirement would put mail order businesses at a competitive 
disadvantage relative to retail businesses. The comment suggested that 
we allow the warning statement to appear either above the mail order 
form that consumers use to order the product or above the toll free 
telephone number that consumers call to order the product. The comment 
argued that these locations would be more similar to the labeling 
requirements for OTC drugs.
    (Response) The PDP for mail order dietary supplements is defined in 
the same way as the PDP for supplements sold in other ways: The label 
that appears on the front of the product package. It does not 
correspond to the front page of any product literature that a firm uses 
to advertise its product.
    (Comment 92) Some comments objected to the requirement that firms 
set off the warning statement in a box graphic. One comment argued that 
the RAND report did not support the need for a black box type of 
warning statement. Some comments suggested that we give manufacturers 
greater leeway with respect to the format of the warning statement. 
Other comments supported the requirement that firms set off the warning 
statement in a box graphic. One comment suggested that we require firms 
to set off the warning

[[Page 6845]]

statement in a brightly colored or neon box instead of in a black box.
    (Response) The proposed warning statement is consistent with 
current research on effective warning statements. Eliminating the box 
graphic would probably not significantly reduce relabeling costs. 
However, it might reduce the visibility of the warning statement, which 
would reduce the distributive impacts of the rule as well as the rule's 
potential health benefits. We have no information establishing that 
colored boxes are more effective than black boxes. Depending on the 
background color of the label, colored boxes may reduce the color 
contrast between the border and the background, which would decrease 
visibility of the warning statement. In addition, requiring colored 
boxes would increase labeling costs because some existing labels are 
not printed in colors.
    c. Content of PDP warning.
    (Comment 93) Some comments suggested that we revise the proposed 
PDP warning statement in various other ways. One comment argued that 
there was no evidence that ``whole-herb products'' containing ephedrine 
alkaloids have been associated with heart attack, stroke, seizure, or 
death, so that the proposed PDP warning statement would be 
inappropriate for those products. This comment suggested that we revise 
the PDP statement so that it simply informs consumers that a product 
contains ephedrine alkaloids and directs them to a warning statement 
elsewhere on the label. A number of comments argued that shortening the 
proposed PDP warning statement would make it more effective. One 
comment noted that the proposed approach is inconsistent with the 
``signal/refer/explain'' format used for the labeling of other 
hazardous products. However, one comment suggested that we add material 
to the PDP warning statement, rather than shortening it.
    (Response) Revising the PDP warning statement for some or all 
dietary supplements that contain ephedrine alkaloids would have little 
effect on labeling costs because firms would still need to revise their 
labels even if we did not require a PDP warning statement. The comments 
did not provide sufficient information to establish that revising the 
PDP warning statement would increase net benefits.
    (Comment 94) A number of comments raised the issue of whom we 
instruct consumers to contact under various conditions. The proposed 
PDP and nonPDP warning statements suggest that consumers contact a 
``doctor'' under various conditions. Some comments suggested we use a 
more general phrase such as ``health care provider'' in order to 
include nurse practitioners and pharmacists. One comment suggested that 
we change ``doctor'' to ``licensed health care provider'' to include 
acupuncturists who are trained in traditional Chinese medicine. The 
comment noted that at least half of the states that regulate the 
practice of acupuncture include the use of herbs in the authorized 
scope of practice of acupuncturists. The comment also noted that herbal 
ephedra is used by health care providers in other disciplines, such as 
naturopathy and herbalism. This comment argued that it was important to 
protect the ability of these groups to dispense dietary supplements 
containing ephedrine alkaloids.
    (Response) Changing the specification of the person that the 
proposed warning label directs consumers to contact under various 
conditions would have little impact on labeling costs but would affect 
the benefits and distributional effects of this rule. Medical doctors 
are probably in the best position to advise consumers on the health 
implications of consuming ephedrine alkaloids under various conditions, 
but consumers might be able to get comparable advice from some other 
sources, including pharmacists and other health care providers, as well 
as some practitioners of acupuncture, herbalism, and naturopathy. On 
the other hand, obtaining advice from a medical doctor is probably more 
costly for many consumers than obtaining advice from other potential 
sources. In addition, some consumers may be unwilling to seek advice 
from medical doctors on the use of dietary supplements for reasons 
other than cost. These consumers may be less likely to follow 
directions to contact a medical doctor than they are to follow 
directions to contact a broader variety of health care providers. This 
component of the warning statement could also have distributional 
effects because directing consumers to contact a medical doctor 
increases the demand for the services of medical doctors and reduces 
the demand for the services of competing health care providers. The 
comments did not provide sufficient information to allow us to 
determine that changing the specification of the person that the label 
directs consumers to contact would increase net benefits. The comments 
also did not provide enough information for us to quantify the 
potential distributional impact of revising this component of the 
label.
    (Comment 95) Some comments noted that the PDP warning statement 
implied that ephedrine alkaloids cause heart attack, stroke, seizure, 
and death. These comments argued that this is misleading because no one 
has proven that ephedrine alkaloids cause these types of adverse 
events. One comment suggested that if we refer to these types of 
adverse events in the warning statement, then we should include a 
qualifying statement explaining that no one has established a causal 
link between these types of adverse events and ephedrine alkaloids. 
This comment also suggested that we indicate in the warning statement 
that reports of serious adverse events are extremely rare.
    (Response) Although the information in the proposed warning 
statement is factually correct because some people have reported the 
specified adverse events after consuming ephedrine alkaloids, some 
consumers might interpret the phrase ``have been reported'' to mean 
that a proven causal relationship exists between the consumption of the 
ephedrine alkaloids and the reported adverse events. This perception 
could generate additional costs in terms of lost consumer utility 
because some consumers who would choose not to consume these products 
if a proven causal relationship existed might choose to continue to 
consume these products if a causal relationship were only possible or 
even likely. One way to reduce potential misperceptions would be to add 
a disclaimer to the label, explaining that the causal relationship 
between ephedrine alkaloids and these adverse events may be uncertain. 
This additional material might either decrease or increase the demand 
for these products, and consumers are generally less likely to respond 
to a longer, qualified warning statement, than to a shorter, non-
qualified warning statement. The comments did not provide sufficient 
information to establish that adding this type of clarification to the 
warning would increase the benefits of the warning statement.
    d. Content of nonPDP warning statement.
    (Comment 96) A number of comments suggested that we revise the 
proposed nonPDP warning statement. Some comments suggested that we use 
the same warning statement that appears on OTC drug products containing 
ephedrine alkaloids. One comment suggested that we allow firms to use 
the OTC warning statement for dietary supplements that they sell 
directly to health professionals for subsequent sale to consumers. One 
comment argued that the warning statement should not instruct consumers 
to contact a doctor if they experience nausea because nausea is not 
likely to be a precursor symptom

[[Page 6846]]

of a potentially serious or life-threatening condition.
    Some comments objected to the warning that the risk of serious side 
effects increases with duration of use. One comment suggested that the 
scientific data showed that adverse effects dramatically decline with 
continued use. Some comments argued that there was no persuasive 
evidence that ephedrine alkaloids had any cumulative effect on the 
cardiovascular or central nervous systems.
    One comment suggested that we allow manufacturers to add 
contraindications beyond those listed on the required warning label. 
One comment suggested that we require a statement clarifying that we 
have not reviewed the product for safety or efficacy. Some comments 
argued that we should require warning statements to include the toll 
free telephone number and Web site address for our MedWatch program. 
Some comments recommended that we require firms to indicate the amount 
of ephedrine alkaloids present in a product on the product label.
    (Response) These comments did not provide sufficient information to 
analyze the costs and benefits of revising the proposed nonPDP warning 
statement according to their recommendation.
    e. Conclusions on benefits and costs of modifying the proposed 
warning statement or requiring it only for certain products. Requiring 
a warning statement for certain products only would reduce costs and 
distributional effects and might reduce benefits compared with Option 3 
(all comparisons in this section are with Option 3). Eliminating the 
PDP warning statement or eliminating the box graphic would have little 
effect on costs but would reduce distributional effects and probably 
also reduce benefits. Requiring a colored box graphic instead of a 
black and white box graphic would increase costs and possibly increase 
distributional effects and benefits. Revising the content of the 
warning statements would have little effect on costs but might increase 
or decrease distributional effects and benefits, depending on the 
revision. We have insufficient information to quantify these possible 
impacts, so we are unable to provide a summary estimate of the costs 
and benefits of this option.
8. Option Five--Generate Additional Information or Take Some Action 
Other Than Removing Dietary Supplements Containing Ephedrine Alkaloids 
From the Market or Requiring Warning Statements
    (Comment 97) One comment argued that we have no controlled 
epidemiological studies that support an association between ephedrine 
alkaloids and stroke, seizure, or myocardial infarction. Other comments 
noted that RAND said in its report that it was unable to establish that 
ephedrine alkaloids caused adverse events and that RAND recommended 
that someone perform a controlled clinical study to address the issue. 
Another comment noted that Haller and Benowitz (2000) said that their 
approach did not establish that ephedrine alkaloids caused adverse 
events and suggested that someone do a large scale case control study 
to quantitatively determine the risks associated with ephedrine 
alkaloids (Ref. 34). One comment noted that the NIH National Advisory 
Council for Complementary and Alternative Medicine Working Group on 
Ephedra suggested that someone perform a multi-site prospective case-
control study to assess the risks associated with taking ephedra. This 
comment suggested that such a study would require 4 to 8 years to 
complete and cost $2 million to $4 million per year. Another comment 
argued that even if someone were to establish that ephedrine alkaloids 
increased cardiovascular risk by raising blood pressure, someone would 
still need to do a controlled research study to determine whether that 
effect outweighed the reduction in cardiovascular risk resulting from 
any weight loss generated by these products. One comment argued that a 
retrospective case control study is the correct study design for rare 
events. This comment argued that someone could do multiple studies of 
this type because they are quick, relatively inexpensive, and because 
the population exposure level is relatively high at 1 percent, 
according to a multistate survey on reported use of ephedra products 
from 1996 to 1998. Some comments suggested that we not take regulatory 
action until we determine that the adverse events that we suspect are 
caused by these supplements are due to ephedrine alkaloids rather than 
due to inconsistent and inaccurate formulations.
    Some comments argued that we do not need to generate additional 
information because we already have sufficient information to remove 
dietary supplements containing ephedrine alkaloids from the market or 
require warning statements. Other comments argued that we do not need 
to generate additional information because we already have sufficient 
information to establish that these restrictions are unnecessary. Some 
of these comments argued that Morgenstern et al., which was published 
after the RAND report, was just the type of case control study that the 
RAND report recommended (Ref. 136). These comments noted that this 
study found that ephedra did not raise the risk for hemorrhagic stroke. 
However, other comments argued that this study found that ephedra did 
raise the risk for hemorrhagic stroke. Some comments criticized various 
aspects of that study. A number of comments argued that the only 
additional studies that would be worthwhile to perform at this point 
would be unethical. These comments suggested that a human subjects 
committee would not allow a prospective study of the safety of 
ephedrine alkaloids without medical screening. They also suggested that 
a cohort study would be difficult because ephedrine alkaloids do not 
generate significant health benefits and also because of the ethical 
requirements to effectively inform participants of the risks.
    (Response) Generating additional information might reduce the 
remaining uncertainty associated with the benefits of this rule or it 
might not. Generating additional information may be difficult, time 
consuming, and expensive. In addition, it is not clear that reducing 
the remaining uncertainty would change the outcome of this rulemaking. 
The comments did not provide sufficient information to allow us to 
estimate the costs and benefits of delaying rulemaking until we 
generate additional information.
    (Comment 98) Other comments suggested that we should take some type 
of action other than issuing a regulation or generating additional 
information. A number of comments suggested that we address any 
problems with dietary supplements containing ephedrine alkaloids by 
using our existing authority to seize unsafe or adulterated dietary 
supplements. Other comments suggested that we address any problems by 
using our existing authority to investigate and prosecute unscrupulous 
multilevel marketing (MLM) distributors. Another comment suggested that 
we develop a level 1 guidance document rather than taking regulatory 
action.
    (Response) The comments did not provide sufficient information to 
establish that spending additional

[[Page 6847]]

resources on enforcement of existing regulations or on promulgating a 
level 1 guidance document would generate greater net benefits than 
issuing this final rule. Following guidance documents is strictly 
voluntary. The fact that some manufacturers continue to produce dietary 
supplements containing ephedrine alkaloids despite ongoing and well-
publicized concerns about the safety of such products suggests that 
voluntary guidance documents are unlikely to have a significant effect.
9. Benefit-Cost Analysis: Summary
    Removing dietary supplements containing ephedrine alkaloids from 
the market (i.e. taking this final action) will generate estimated 
benefits of between $43 million and $132 million per year. We used the 
following assumptions to calculate this range of benefits: A 10 percent 
reporting rate for adverse events, no potentially countervailing health 
effects from the use of substitute products and other weight loss 
alternatives, no countervailing health effects from potentially 
foregone weight loss, and the fact that consumers do not already 
understand and incorporate the risks posed by these products in their 
consumption decisions. Including the impact of substitute products and 
activities could reduce the rule's health benefit considerably, 
possibly to $0 per year, although that is unlikely. These 
countervailing effects may occur because this rule will not affect the 
underlying demand for products having functional characteristics 
similar to ephedrine alkaloids, and it is likely that products having 
similar functional characteristics may contain similar types of 
ingredients that may pose similar types of health risks. The range of 
benefits includes alternative assumptions about the value of a 
statistical life ($5 million and $6.5 million) and the value of a 
statistical life year ($0.1 million, $0.3 million, and $0.5 million). 
We also considered a reporting rate of 50 percent, which leads to 
estimated annual benefits of $9 million to $26 million, and 100 
percent, which leads to estimated annual benefits of $4 million to $13 
million. More precise estimates of the health benefits would depend on 
choosing a particular combination of assumptions.
    Removing these products from the market will generate one-time 
product reformulation costs of $10 million to $100 million, which 
amounts to a yearly cost of $1 million to $7 million when annualized 
over 20 years at an interest rate of 3 percent, and $1 million to $9 
million at an interest rate of seven percent. These costs could be 
partly offset by reductions in fees associated with legal actions 
involving these products. In addition to the social costs, removing 
dietary supplements containing ephedrine alkaloids from the market 
could also generate distributional effects under which some firms 
manufacturing or distributing dietary supplements containing ephedrine 
alkaloids may experience reduced profits, while firms manufacturing or 
distributing other dietary supplements or other weight loss 
alternatives may experience increased profits. In addition, removing 
dietary supplements containing ephedrine alkaloids from the market 
would also generate costs in the form of lost consumer utility or 
satisfaction because of the removal of a product from the market. We 
estimated lost utility to be $6 million to $81 million per year.
    Based on these estimates, the potential economic effects of this 
rule range from a net annual social cost of $90 million per year, if 
the rule's net health benefits are zero because of countervailing 
health effects or because consumers already understand and voluntarily 
accept the risks posed be these products, to an annual net social 
benefit of $125 million, if there are no countervailing health risks 
and consumers do not already understand and accept the known and 
potential risks.

           Table 8.--Summary of Options, Rounded to $ Millions
------------------------------------------------------------------------
      Option           Annual Cost      Annual Benefit          Net
------------------------------------------------------------------------
1. Take no new       $0                $0                 $0
 regulatory action
 (baseline)
------------------------------------------------------------------------
2a. Remove dietary   $7 to $90         $43 to $132        - $47 to $125
 supplements
 containing
 ephedrine
 alkaloids from
 the market (if
 consumer behavior
 does not already
 incorporate risk)
------------------------------------------------------------------------
2b. Remove dietary   $7 to $90         $0                 - $90 to - $7
 supplements
 containing
 ephedrine
 alkaloids from
 the market (if
 consumer behavior
 already
 incorporates
 risk)
------------------------------------------------------------------------
3. Require 2003      $0 to $1          $0 to $20          - $1 to $20
 warning atatement
------------------------------------------------------------------------
4. Require warning   NA                NA                 NA
 statement, but
 modify it or
 require only on
 certain products
------------------------------------------------------------------------
5. Generate          unknown           unknown            unknown
 additional
 information or
 take some action
 other than
 removal or
 warning
 statements
------------------------------------------------------------------------

B. Small Entity Analysis

    We have examined the economic implications of this final rule as 
required by the Regulatory Flexibility Act (5 U.S.C. 601-612) and in 
accordance with Executive Order 13272 (August 13, 2002). If a rule has 
a significant economic impact on a substantial number of small 
entities, the Regulatory Flexibility Act requires us to analyze 
regulatory options that would lessen the economic effect of the rule on 
small entities. We find that this final rule would have a significant 
economic impact on a substantial number of small entities.
    (Comment 99) Some comments addressed our estimate of the number of 
small firms in the analysis of the proposed rule. Some comments argued 
that we had ignored a large number of independent small distributors in 
the analysis of the proposed rule. One comment suggested we revisit our 
analysis of the impact of the rule on small businesses. One comment

[[Page 6848]]

suggested we obtain information on the impact of the rule on small 
entities by opening a dialogue with industry associations.
    (Response) We have revisited and revised our estimate of the number 
of firms based on a database of dietary supplement products that the 
Research Triangle Institute compiled under contract to FDA after 
publication of the proposed rule. This database listed 30 firms 
associated with 48 dietary supplement products containing ephedrine 
alkaloids (Ref. 159). To estimate the number of these firms that are 
small, we used a database of dietary supplement manufacturing practices 
that was also compiled by RTI under contract to FDA (Ref. 160). This 
database had size information for only a few of the 30 firms that we 
identified as relevant from the first database. Therefore, we estimated 
the number of small firms based on the percentage of all dietary 
supplement firms in the database that would qualify as small firms. The 
Small Business Administration (SBA) publishes definitions of small 
businesses by the North American Industry Classification System (NAICS) 
code. The firms in the database fell into the following NAICS codes: 
(1) 311222 Soybean Processing, (2) 311920 Coffee and Tea Manufacturing, 
(3) 325188 All Other Basic Inorganic Chemical Manufacturing, (4) 325199 
All Other Basic Organic Chemical Manufacturing, (5) 325411 Medicinal 
and Botanical Manufacturing, and (6) 325412 Pharmaceutical Preparation 
Manufacturing. SBA defines small businesses in these NAICS codes based 
on a maximum number of employees, as follows: 311222 and 311920--no 
more than 500 employees; 325411 and 325412--no more than 750 employees; 
and 325188 and 325199--no more than 1000 employees. The database of 
firms listed 1,566 individual plants and 146 parent companies. 
Essentially all individual plants qualified as small businesses (98 
percent under a maximum of 500 employees and 100 percent under a 
maximum of 1,000 employees). However, approximately 12 percent of the 
individual plants were associated with parent companies, and only about 
half of the parent companies qualified as small businesses (53 percent 
under a maximum of 500 employees and 58 percent under a maximum of 
1,000 employees). Based on this information, we estimated that about 94 
percent of the 30 firms associated with dietary supplement containing 
ephedrine alkaloids, or about 28 firms, would qualify as small 
businesses.
    There may also be a number of independent distributors that are not 
captured in our database of dietary supplement firms. All or most of 
these firms would probably qualify as small businesses. However, we do 
not have sufficient information to estimate the number of distributors 
or to compare their characteristics to the SBA definition of a small 
business for that industry. As we noted in the previous paragraphs, 
this final rule will generate shifts in demand that might adversely 
affect these firms. However, the most likely substitutes for dietary 
supplements containing ephedrine alkaloids are other dietary 
supplements, and the same distributors that handle dietary supplements 
containing ephedrine alkaloids might also handle these other dietary 
supplements. Therefore, the net distributive impact on small 
distributors may be small or nonexistent. Although demand shifts 
generated by this final rule might also increase business for other 
small businesses, we do not consider countervailing positive effects on 
other small entities when assessing the impact of our rules on small 
entities.
    In response to the request that we open a dialogue with industry 
associations, we note that small entities, and trade associations (with 
member small entities) submitted a number of comments regarding small 
business impact during the various comment periods for this rulemaking.
    In the preceding cost-benefit analysis, we estimated that removing 
dietary supplements containing ephedrine alkaloids from the market 
would generate annualized cost of $1 million to $9 million over 20 
years because of the need to reformulate products. This would 
correspond to a cost per firm across 30 firms of between $30,000 and 
$300,000 per year. In addition, we estimated that profits might be 
reduced by $0 to $13 million per year due to decreased sales. Profits 
may accrue to either manufacturers or distributors. If all profit 
losses affected manufacturers only, then the annual profit loss per 
firm across 30 firms would be between $0 and $ 430,000, which would 
give a total cost per firm of $30,000 to $730,000. Most of these firms 
are small, so even $30,000 per year (the lower bound) would be a 
significant additional burden. We previously estimated total sales to 
be $559 million to $806 million. If we assume that profits correspond 
to approximately 5 percent of sales, then annual profits would be $28 
million to $40 million. If we assume that all profits accrue to 
manufacturers, then profits would be $0.9 million to $1.3 million per 
year per firm across 30 firms. In that case, reformulation costs would 
represent 2 percent to 33 percent of total profits, while total costs 
would represent 2 percent to 81 percent of total profits. The 
Regulatory Flexibility Act does not specify a threshold for costs to 
have a significant economic impact, but the 2 ranges we have calculated 
reach a high fraction of total profit; for some individual small firms 
the fraction of profit would be higher. If some of the profit losses 
accrued to distributors rather than manufacturers, then the potential 
cost per firm across all firms would be lower. However, we have 
insufficient information to estimate the number of distributors or the 
sales or profits per distributor.
    (Comment 100) One comment argued that the PDP warning statement 
would have a significant economic impact on small businesses. This 
comment argued that the nonPDP warning statement would be adequate to 
protect consumers. This comment recommended that we eliminate the PDP 
warning statement.
    (Response) A PDP warning statement might have a significant impact 
on small businesses. We have analyzed the costs of the proposed warning 
statement as a whole (including both PDP and nonPDP components) in our 
analysis of impacts under Executive Order 12866. However, the comment 
did not provide sufficient information to differentiate the impact on 
small businesses from the impact on other regulated entities, or to 
differentiate the impact of the PDP warning from the impact of the 
nonPDP warning.
    (Comment 101) One comment recommended that we consider reasonable 
alternatives to the rule in order to reduce the burden on small 
businesses.
    (Response) The discussion of regulatory options in the preceding 
benefit-cost analysis pertains primarily to small businesses because 
nearly all affected firms are small businesses under SBA size 
definitions. We could develop a definition of a very small business 
(different from the SBA definition of a small business) and develop 
additional regulatory options to reduce the burden on those firms, but 
those options would also be similar to those in the benefit-cost 
analysis. As we stated elsewhere in this analysis, any option that 
would reduce the regulatory burden on very small firms would also 
reduce benefits by increasing the risk to public health. We do not have 
sufficient information to compare the value of the

[[Page 6849]]

regulatory relief for very small firms to the associated reduction in 
benefits.

IX. Environmental Impact

    Removing dietary supplements containing ephedrine alkaloids from 
the market will not have a significant impact on the human environment. 
Therefore, an environmental impact statement is not required.

X. Paperwork Reduction Act

    This final rule contains no collections of information. Therefore, 
clearance by OMB under the Paperwork Reduction Act of 1995 is not 
required.

XI. Federalism

    We have analyzed this final rule in accordance with the principles 
set forth in Executive Order 13132. FDA has determined that the rule 
has a preemptive effect on State law. Section 4(a) of the Executive 
order requires agencies to ``construe * * * a Federal statute to 
preempt State law only where the statute contains an express preemption 
provision or there is some other clear evidence that the Congress 
intended preemption of State law, or where the exercise of State 
authority conflicts with the exercise of Federal authority under the 
Federal statute.'' Section 402(f)(1)(A) of the act states that a 
dietary supplement or dietary ingredient shall be considered 
adulterated if it presents a significant or unreasonable risk of 
illness or injury under conditions of use recommended or suggested in 
the product's labeling. If no conditions of use are suggested or 
recommended in the product's labeling, the dietary supplement or 
dietary ingredient is considered to be adulterated if it presents a 
significant or unreasonable risk of illness or injury under ordinary 
conditions of use. We have concluded that dietary supplements 
containing ephedrine alkaloids present an unreasonable risk and are 
therefore adulterated under section 402(f)(1)(A) of the act.
    Section 402(f)(1)(A) of the act does not expressly preempt State or 
local laws. Therefore, under section 4(b) of Executive Order 13132, we 
are to construe our rulemaking authority as authorizing preemption of 
State law by rulemaking ``only when the exercise of State authority 
directly conflicts with the exercise of Federal authority under the 
Federal statute or there is clear evidence to conclude that Congress 
intended the agency to have the authority to preempt State law.''
    We are aware that several States have laws concerning dietary 
supplements containing ephedrine alkaloids, such as required label 
statements, which clearly contemplate the continued marketing of such 
products. Section 301(a) of the act (in relevant part) prohibits the 
introduction or delivery for introduction into interstate commerce of 
any adulterated food. In this rule, the agency has declared dietary 
supplements containing ephedrine alkaloids to be adulterated. As a 
result, State laws establishing label requirements or other 
requirements that contemplate the continued marketing of these products 
conflict with this final rule and, consequently, are preempted.
    Section 4(c) of Executive Order 13132 instructs us to restrict any 
Federal preemption of State law to the ``minimum level necessary to 
achieve the objectives of the statute pursuant to which the regulations 
are promulgated.'' This action meets the preceding requirement because 
it only applies to State laws that contemplate the continued marketing 
of this class of products.
    Section 4(d) of Executive Order 13132 states that when an agency 
foresees the possibility of a conflict between State law and federally 
protected interests within the agency's area of regulatory 
responsibility, the agency ``shall consult, to the extent practicable, 
with appropriate State and local officials in an effort to avoid such a 
conflict.'' Section 4(e) of Executive Order 13132 adds that, when an 
agency proposes to act through adjudication or rulemaking to preempt 
State law, the agency ``shall provide all affected State and local 
officials notice and an opportunity for appropriate participation in 
the proceedings.''
    In the present rulemaking, consultation with and notice to State 
officials under section 4(d) and (e) of Executive Order 13132 did not 
occur before we published the June 1997 proposal. Such consultation and 
notice was not possible because we published the proposed rule in the 
Federal Register of June 4, 1997, and Executive Order 13132 was not 
signed until August 4, 1999. OMB's guidance for implementing Executive 
Order 13132 states that, when a final rule may have been issued as a 
proposed rule before August 4, 1999, such that the intergovernmental 
consultation process had not occurred as called for by Executive Order 
13132, the agency's certification ``should so state'' (see Memorandum 
for Heads of Executive Departments and Agencies, and Independent 
Regulatory Agencies, dated October 28, 1999) (Ref. 161). Thus, we 
certify that the intergovernmental consultation process described in 
section 4(d) of Executive Order 13132 did not occur for the proposed 
rule, but we also believe that State and local governments had 
sufficient notice and an opportunity to participate in this rulemaking 
process. We note that the proposed rule was subject to a previous 
Executive Order, Executive Order 12612, which was also entitled, 
``Federalism,'' and had a similar consultation and notification 
obligation for federal agencies. When we issued the proposed rule, we 
notified the States, and State and local health departments, among 
others, submitted comments to the proposal (65 FR 17474, April 3, 2000) 
(stating that State and local health departments and government 
agencies had commented on the proposed rule)). Furthermore, a 
subsequent notice, published on March 5, 2003, expressly asked whether 
we should determine that dietary supplements containing ephedrine 
alkaloids present a ``significant or unreasonable risk of illness or 
injury'' under section 402(f)(1)(A) of the act (68 FR at 10417, 10419, 
and 10420). Although the March 2003 notice did not contain a separate 
Federalism analysis, we believe that States were aware of the March 
2003 notice because at least five State or local governments or 
legislators submitted comments in response to the March 2003 notice, 
and most of these comments urged us to ban the sale of such products.

XII. References

    The following references have been placed on display in the 
Division of Dockets Management (see ADDRESSES) and may be seen by 
interested persons between 9 a.m. and 4 p.m., Monday through Friday. 
(FDA has verified the Web site addresses, but FDA is not responsible 
for any subsequent changes to the nonFDA Web sites after this document 
publishes in the Federal Register.)
    1. Ephedra Monograph, Review of Natural Products, Der 
Marderosian, A., ed., DrugFacts.com (http://
www.factsandcomparisons.com), 2003.
    2. Chen, K. K. and C. F. Schmidt, ``Ephedrine and Related 
Substances,'' Medicine vol. 9, pp. 1-117, 1930.
    3. Mahuang (Appendix: Mahuanggen)., Chapter in Chang, H-M. and 
P. P-H. But, eds., Pharmacology and Applications of Chinese Materia 
Medica, pp. 1119-1124, Singapore: World Scientific Publishing Co. 
Pte. Ltd., pp. 1119-1124, 1987.
    4. Karch, S. B., ``Other Naturally Occurring Stimulants,'' 
Chapter in Karch, S. B., The Pathology of Drug Abuse, pp. 177-198, 
Boca Raton: CRC Press, 1996.
    5. ``Phenethylamines,'' Chapter in Bruneton, J., ed., 
Pharmacognosy, Phytochemistry, Medicinal Plants, pp. 711-715, New 
York: Laviosier Publishing, 1995.
    6. Betz, J. M., Tab F, ``Review of Plant Chemistry: Alkaloids of 
Ma Huang (ephedra

[[Page 6850]]

spp.),'' Food and Drug Administration, Briefing Materials for Food 
Advisory Committee Special Working Group on Foods Containing 
Ephedrine Alkaloids, Center for Food Safety and Applied Nutrition, 
pp. 1-14, 1995.
    7. World Health Organization (WHO) Monographs on Selected 
Medicinal Plants, Herba Ephedrae, pp. 145-153, 1999.
    8. Oshio, H., M. Tsukui, and T. Matsuoka, ``Isolation of l-
Ephedrine From `Pinelliae Tuber','' Chemical and Pharmaceutical 
Bulletin (Tokyo), vol. 26, pp. 2096-2097, 1978.
    9. Ghosal, S., R. B. Chauhan, and R. Mehta, ``Alkaloids of Sida 
Cordifolia,'' Phytochemical Reports, vol. 14, pp. 830-832, 1975.
    10. Cui, J., T. Zhou, J. Zhang, and Z. Lou, ``Analysis of 
Alkaloids in Chinese Ephedra Species by Gas Chromatographic 
Methods,'' Phytochemical Analysis, vol. 2, pp. 116-119, 1991.
    11. Food and Drug Administration, Tab E: Additional Market 
Review Information. Briefing Materials for Food Advisory Committee 
on Dietary Supplements Containing Ephedrine Alkaloids, Center for 
Food Safety and Applied Nutrition, pp. 1-35, 1996.
    12. Brevoort, P., ``The U.S. Botanical Market--An Overview,'' 
HerbalGram, vol. 36, pp. 49-57, 1996.
    13. Food and Drug Administration, Hardy, C., ``FDA Market Review 
Update Ephedrine Alkaloid-Containing Dietary Supplements,'' 2000.
    14. Food and Drug Administration, Food Advisory Committee on 
Dietary Supplements Containing Ephedrine Alkaloids, Meeting 
Transcript, Docket No. 1995N-0304, TR2, vol. 72-73, August 1996.
    15. Food and Drug Administration, Food Advisory Committee, 
Special Working Group on Foods Containing Ephedrine Alkaloids, 
Meeting Transcript, Docket No. 1995N-0304, TR1, vol. 40-41, October 
1995.
    16. Commission on Dietary Supplement Labels: Report of the 
Commission on Dietary Supplement Labels, 1997.
    17. U.S. General Accounting Office (GAO), ``Dietary Supplements, 
Uncertainties in Analyses Underlying FDA's Proposed Rule on 
Ephedrine Alkaloids,'' GAO/Health, Education, and Human Services 
Division (HEHS)/General Government Division (GGD), pp. 99-90, 1999.
    18. Food and Drug Administration, ``Assessment of Public Health 
Risks Associated with the Use of Ephedrine Alkaloid-containing 
Dietary Supplements,'' 2000, Docket No. 2000N-1200, vol. 29-30, pp. 
39-41, 2000.
    19. Kernan, W. N., C. M. Viscoli, L. M. Brass, et al., 
``Phenylpropanolamine and the Risk of Hemorrhagic Stroke,'' New 
England Journal of Medicine, vol. 343, pp. 1826-1832, 2000.
    20. Department of Health and Human Services, Office of Inspector 
General, ``Adverse Event Reporting for Dietary Supplements, An 
Inadequate Safety Valve,'' OEI-01-00-00180, 2001.
    21. Shekelle P., S. Morton, M. Maglione, et al., ``Ephedra and 
Ephedrine for Weight Loss and Athletic Performance Enhancement: 
Clinical Efficacy and Side Effects. Evidence Report/Technology 
Assessment,'' No. 76 (Prepared by Southern California Evidence-based 
Practice Center, RAND, under Contract No. 290-97-0001, Task Order 
No. 9), AHRQ Publication No. 03-E022, Rockville, MD, Agency for 
Healthcare Research and Quality, Docket No. 1995N-0304, BKG3, vol. 
300 (http://www.fda.gov/OHRMS/DOCKETS/98fr/95n-0304-bkg0003-ref-07-
01-index.htm), 2003.
    22. Shekelle, P. G., M. L. Hardy, S. C. Morton, et al., 
``Efficacy and Safety of Ephedra and Ephedrine for Weight Loss and 
Athletic Performance: A Meta-Analysis,'' Journal of the American 
Medical Association, vol. 289, pp. 1537-1545, 2003.
    23. U.S. General Accounting Office (GAO), ``Dietary Supplements 
Containing Ephedra, Health Risks and FDA's Oversight,'' Testimony: 
Before the Subcommittee on Oversight and Investigations, Committee 
on Energy and Commerce, U.S. House of Representatives, GAO-03-1042T, 
2003.
    24. U.S. General Accounting Office (GAO): ``Dietary Supplements: 
Review of Health-Related Call Records for Users of Metabolife 356,'' 
GAO-03-494, 2003.
    25. Caveney, S. and A. Starratt, ``Glutamatergic Signals in 
Ephedra (Scientific Correspondence),'' Lancet, vol. 372, p. 509, 
1994.
    26. Caveney, S., D. A. Charlet, H. Freitag, et al., ``New 
Observations on the Secondary Chemistry of World Ephedra 
(Ephedraceae),'' American Journal of Botany, vol. 88, pp. 1199-1208, 
2001.
    27. Food and Drug Administration, L. A. Love, Tab E, 
``Evaluation of the Safety of Food Products Containing Sources of 
Ephedrine Alkaloids,'' Briefing Materials for Food Advisory 
Committee Special Working Group on Foods Containing Ephedrine 
Alkaloids, Center for Food Safety and Applied Nutrition, pp. 1-51, 
1995.
    28. Food and Nutrition Board, Institute of Medicine, Dietary 
Reference Intakes: A Risk Assessment Model for Establishing Upper 
Intake Levels for Nutrients, Washington, DC, National Academy Press, 
pp. 1-71, 1999.
    29. Vasan, R. S., M. G. Larson, E. P. Leip, et al., ``Impact of 
High-Normal Blood Pressure on the Risk of Cardiovascular Disease,'' 
New England Journal of Medicine, vol. 345, pp. 1291-1297, 2001.
    29a. Vasan, R. S., J. M. Massaro, P. W. Wilson, et al., 
``Antecedent Blood Pressure and Risk of Cardiovascular Disease: The 
Framingham Heart Study,'' Circulation, vol. 105, pp. 48-53, 2002.
    30. Neaton, J. D., L. Kuller, J. Stamler, and D. N. Wentworth, 
Impact of Systolic and Diastolic Blood Pressure on Cardiovascular 
Mortality, Chapter in Laragh, J. H. and B. M. and Brenner, eds., 
Hypertension: Pathophysiology, Diagnosis, and Management, New York: 
Raven Press Ltd., pp. 127-144, 1995.
    31. National Institutes of Health (NIH), Report of Ephedra 
Working Group to the National Advisory Council for Complementary and 
Alternative Medicine (http://nccam.nih.gov/health/alerts/ephedra/
working-group.htm), 2003.
    32. Goldman, S. A., D. L. Kennedy, D. J. Graham, et al., ``The 
Clinical Impact of Adverse Event Reporting,'' MedWatch Continuing 
Education Article, pp. 1-11, 1996.
    33. Middleton, W. S. and K. K. Chen, ``Ephedrine,'' Archives of 
Internal Medicine, vol. 39, pp. 385-403, 1927.
    34. Haller, C. A. and N. L. Benowitz, ``Adverse Cardiovascular 
and Central Nervous System Events Associated With Dietary 
Supplements Containing Ephedra Alkaloids,'' New England Journal of 
Medicine, vol. 343, pp. 1833-1838, 2000.
    35. Hoffman, B. B. and R. J. Lefkowitz, ``Catecholamines, 
Sympathomimetic Drugs, and Adrenergic Receptor Antagonists,'' 
Chapter in Hardman, J. G., A. G. Gilman, and L. E. Limbird, eds., 
Goodman & Gilman's The Pharmacological Basis of Therapeutics, New 
York, St. Louis, McGraw-Hill Chapter, pp. 199-248, 1996.
    36. Astrup, A., S. Toubro, S. Cannon, et al., ``Thermogenic 
Synergism Between Ephedrine and Caffeine in Healthy Volunteers: A 
Double-Blind, Placebo-Controlled Study,'' Metabolism, vol. 40, pp. 
323-329, 1991.
    37. Taguchi, N.,T. Nishikawa, S. Inomata, et al., ``Hemodynamic 
Effects of Intravenous Ephedrine in Infants and Children 
Anesthetized With Halothane and Nitrous Oxide,'' Anesthesia and 
Analgesia, vol. 82, pp. 568-573, 1996.
    38. The Xamoterol in Severe Heart Failure Study Group, 
``Xamoterol in Severe Heart Failure,'' Lancet, vol. 336, pp. 1-6, 
1990.
    39. Packer, M., J. R. Carver, R. J. Rodeheffer, et al., ``Effect 
of Oral Milrinone on Mortality in Severe Chronic Heart Failure, The 
PROMISE Study Research Group,'' New England Journal of Medicine, 
vol. 325, pp. 1468-1475, 1991.
    40. Cowley, A. J. and A. M. Skene, ``Treatment of Severe Heart 
Failure: Quantity or Quality of Life? A Trial of Enoximone,'' 
Enoximone Investigators, British Heart Journal, vol. 72, pp. 226-
230, 1994.
    41. Brown, O. M., Adrenergic Drugs, Chapter in Smith, C. M. and 
A. M. Reynard, eds., Essentials of Pharmacology, Philadelphia, W. B. 
Saunders Co., pp. 75-85, 1995.
    42. Haller, C. A., P. Jacob, III and N. L. Benowitz, 
``Pharmacology of Ephedra Alkaloids and Caffeine After Single-Dose 
Dietary Supplement Use,'' Clinical Pharmacology and Therapeutics, 
vol. 71, pp. 421-432, 2002.
    43. White, L. M., S. F. Gardner, B. J. Gurley, et al., 
``Pharmacokinetics and Cardiovascular Effects of Ma-huang (Ephedra 
Sinica) in Normotensive Adults,'' Journal Clinical Pharmacology, 
vol. 37, pp. 116-122, 1997.
    44. Gurley, B. J., S. F. Gardner, L. M. White, et al., 
``Ephedrine Pharmacokinetics After the Ingestion of Nutritional 
Supplements Containing Ephedra Sinica (Ma Huang),'' Therapeutic Drug 
Monitoring, vol. 20, pp. 439-445, 1998.
    45. Eggertsen, R., A. Andreasson, T. Hedner, et al., ``Effect of 
Coffee on Ambulatory Blood Pressure in Patients With Treated 
Hypertension,'' Journal of Internal Medicine, vol. 233, pp. 351-355, 
1993.
    46. Myers, M. G. and R. A. Reeves, ``The Effect of Caffeine on 
Daytime Ambulatory Blood Pressure,'' American Journal of 
Hypertension, vol. 4, pp. 427-431, 1991.

[[Page 6851]]

    47. Jacobs, I., H. Pasternak, and D. G. Bell, ``Effects of 
Ephedrine, Caffeine, and Their Combination on Muscular Endurance,'' 
Medicine and Science in Sports and Exercise, vol. 35, pp. 987-994, 
2003.
    48. Berlin, I., D. Warot, G. Aymard, et al., ``Pharmacodynamics 
and Pharmacokinetics of Single Nasal (5 mg and 10 mg) and Oral (50 
mg) Doses of Ephedrine in Healthy Subjects,'' European Journal of 
Clinical Pharmacology, vol. 57, pp. 447-455, 2001.
    49. Boozer, C. N., P. A. Daly, P. Homel, et al., ``Herbal 
Ephedra/Caffeine for Weight Loss: A 6-Month Randomized Safety and 
Efficacy Trial,'' International Journal of Obesity and Related 
Metabolic Disorders, vol. 26, pp. 593-604, 2002.
    50. Kaplan, N., Expert Opinion: A Review of Boozer, CN, et al., 
Herbal Ephedra/Caffeine for Weight Loss: A 6-Month Randomized Safety 
and Efficacy Trial, Review Prepared for the Food and Drug 
Administration, 2003.
    51. Atkinson, R., Expert Opinion: A Review of Boozer, CN, et 
al., Herbal Ephedra/Caffeine for Weight Loss: A 6-Month Randomized 
Safety and Efficacy Trial, Review Prepared for the Food and Drug 
Administration, 2003.
    52. Espeland, M., Expert Opinion: A Review of Boozer, CN, et 
al., Herbal Ephedra/Caffeine for Weight Loss: A 6-Month Randomized 
Safety and Efficacy Trial, Review Prepared for the Food and Drug 
Administration, 2003.
    53. Hajjar, I. and T. A. Kotchen, ``Trends in Prevalence, 
Awareness, Treatment, and Control of Hypertension in the United 
States, 1988-2000,'' Journal of the American Medical Association, 
vol. 290, pp. 199-206, 2003.
    54. Lewington, S., R. Clarke, N. Qizilbash, et al., ``Age-
Specific Relevance of Usual Blood Pressure to Vascular Mortality: A 
Meta-Analysis of Individual Data for One Million Adults in 61 
Prospective Studies,'' Lancet, vol. 360, pp. 1903-1913, 2002.
    55. Collins, R., R. Peto, S. MacMahon, et al., ``Blood Pressure, 
Stroke, and Coronary Heart Disease, Part 2, Short-term Reductions in 
Blood Pressure: Overview of Randomised Drug Trials in Their 
Epidemiological Context,'' Lancet, vol. 335, pp. 827-838, 1990.
    56. Staessen, J. A., J. G. Wang, and L. Thijs, ``Cardiovascular 
Prevention and Blood Pressure Reduction: A Quantitative Overview 
Updated Until 1 March 2003,'' Journal of Hypertension, vol. 21, pp. 
1055-1076, 2003.
    57. Neal, B., S. MacMahon, and N. Chapman, ``Effects of ACE 
Inhibitors, Calcium Antagonists, and Other Blood-Pressure-Lowering 
Drugs: Results of Prospectively Designed Overviews of Randomised 
Trials,'' Blood Pressure Lowering Treatment Trialists' 
Collaboration, Lancet, vol. 356, pp. 1955-1964, 2000.
    58. MacMahon, S., R. Peto, J. Cutler, et al., ``Blood Pressure, 
Stroke, and Coronary Heart Disease. Part 1, Prolonged Differences in 
Blood Pressure: Prospective Observational Studies Corrected for the 
Regression Dilution Bias,'' Lancet, vol. 335, pp. 765-774, 1990.
    59. Cohn, J. N., S. O. Goldstein, B. H. Greenberg, et al., ``A 
Dose-Dependent Increase in Mortality With Vesnarinone Among Patients 
With Severe Heart Failure,'' Vesnarinone Trial Investigators, New 
England Journal of Medicine, vol. 339, pp. 1810-1816, 1998.
    60. Hampton, J. R., D. J. van Veldhuisen, F. X. Kleber, et al., 
``Randomised Study of Effect of Ibopamine on Survival in Patients 
With Advanced Severe Heart Failure,'' Second Prospective Randomised 
Study of Ibopamine on Mortality and Efficacy (PRIME II) 
Investigators, Lancet, vol. 349. pp. 971-977, 1997.
    61. Cuffe, M. S., R. M. Califf, K. F. Adams, Jr., et al., 
``Short-Term Intravenous Milrinone for Acute Exacerbation of Chronic 
Heart Failure: A Randomized Controlled Trial,'' Journal of the 
American Medical Association, vol. 287, pp. 1541-1547, 2002.
    62. Sears, M. R., ``Adverse Effects of Aeta-Agonists,'' Journal 
of Allergy and Clinical Immunology, vol. 110, pp. S322-S328, 2002.
    63. Wit, A. L., B. F. Hoffman, and M. R. Rosen, 
``Electrophysiology and Pharmacology of Cardiac Arrhythmias, IX. 
Cardiac Electrophysiologic Effects of Beta Adrenergic Receptor 
Stimulation and Blockade. Part C,'' American Heart Journal, vol. 90, 
pp. 795-803, 1975.
    64. Mahaffey, K. W., J. A. Puma, N. A. Barbagelata, et al., 
``Adenosine as an Adjunct to Thrombolytic Therapy for Acute 
Myocardial Infarction: Results of a Multicenter, Randomized, 
Placebo-Controlled Trial: The Acute Myocardial Infarction Study of 
Adenosine (AMISTAD) Trial,'' Journal of the American College of 
Cardiology, vol. 34, pp. 1711-1720, 1999.
    65. Kenchaiah, S., J. C. Evans, D. Levy, et al., ``Obesity and 
the Risk of Heart Failure,'' New England Journal of Medicine, vol. 
347, pp. 305-313, 2002.
    66. National Institutes of Health, National Heart, Lung, and 
Blood Institute, ``Clinical Guidelines on the Identification, 
Evaluation, and Treatment of Overweight and Obesity in Adults: The 
Evidence Report,'' NIH Publication No. 98-4083, pp. 1-262 (http://
www.nhlbi.nih.gov/guidelines/obesity/ob_gdlns.htm), 1998.
    67. Astrup, A., L.Breum, S. Toubro, et al., ``The Effect and 
Safety of an Ephedrine/Caffeine Compound Compared to Ephedrine, 
Caffeine and Placebo in Obese Subjects on an Energy Restricted Diet, 
A Double Blind Trial,'' International Journal of Obesity and Related 
Metabolic Disorders, vol. 16, pp. 269-277, 1992.
    68. Toubro, S., A. V. Astrup, L. Breu, F. Quaade, ``Safety and 
Efficacy of Long-Term Treatment With Ephedrine, Caffeine and an 
Ephedrine/Caffeine Mixture,'' International Journal of Obesity and 
Related Metabolic Disorders, vol. 17, suppl. 1, pp. S69-S72, 1993.
    69. Quaade, F., A. Astrup, L. Breum, et al., ``The effect of an 
Ephedrine/Caffeine Combination As a Supplement to a Weight-Reducing 
Diet. A Randomized, Placebo-Controlled, Double-Blind Trial,'' Ugesk 
Laeger, vol. 154, pp. 1258-1263, 1992.
    70. Toubro, S., A. Astrup, L. Breum, et al., ``The Acute and 
Chronic Effects of Ephedrine/Caffeine Mixtures on Energy Expenditure 
and Glucose Metabolism in Humans,'' International Journal of Obesity 
and Related Metabolic Disorders, vol. 17, suppl. 3, pp. S73-S77, 
1993.
    71. Astrup, A., C. Lundsgaard, J. Madsen, et al., ``Enhanced 
Thermogenic Responsiveness During Chronic Ephedrine Treatment in 
Man,'' American Journal Clinical Nutrition, vol. 42, pp. 83-94, 
1985.
    72. Astrup, A. and S. Toubro, ``Thermogenic, Metabolic, and 
Cardiovascular Responses to Ephedrine and Caffeine in Man,'' 
International Journal of Obesity and Related Metabolic Disorders, 
vol. 17, suppl. 1, pp. S41-S43, 1993.
    73. Astrup, A., S. Toubro, S. Cannon, P. Hein, et al., 
``Caffeine: A Double-Blind, Placebo-Controlled Study of its 
Thermogenic, Metabolic, and Cardiovascular Effects in Healthy 
Volunteers,'' American Journal Clinical Nutrition, vol. 51, pp. 759-
767, 1990.
    74. Breum, L., J. K. Pedersen, F. Ahlstrom, et al., ``Comparison 
of an Ephedrine/Caffeine Combination and Dexfenfluramine in the 
Treatment of Obesity. A Double-Blind Multi-Centre Trial in Ggeneral 
Practice,'' International Journal of Obesity and Related Metabolic 
Disorders, vol. 18, pp. 99-103, 1994.
    75. Pasquali, R., G. Baraldi, M. P. Cesari, et al., ``A 
Controlled Trial Using Ephedrine in the Treatment of Obesity,'' 
International Journal of Obesity and Related Metabolic Disorders, 
vol. 9, pp. 93-98, 1985.
    76. Pasquali, R., M. P. Cesari, N. Melchionda, et al., ``Does 
Ephedrine Promote Weight Loss in Low-Energy-Adapted Obese Women?,'' 
International Journal of Obesity and Related Metabolic Disorders, 
vol. 11, pp. 163-168, 1987.
    77. Pasquali, R., F. Casimirri, N. Melchionda, et al., ``Effects 
of Chronic Administration of Ephedrine During Very-Low-Calorie Diets 
on Energy Expenditure, Protein Metabolism and Hormone Levels in 
Obese Subjects,'' Clinical Science (London), vol. 82, pp. 85-92, 
1992.
    78. Krieger, D. R., Daly, P. A., Dulloo, A. G., et al., 
``Ephedrine, Caffeine and Aspirin Promote Weight Loss in Obese 
Subjects,'' Transactions of the Association of American Physicians, 
vol. 103, pp. 307-312, 1990.
    79. Daly, P. A., D. R. Krieger, A. G. Dulloo, et al., 
``Ephedrine, Caffeine and Aspirin: Safety and Efficacy for Treatment 
of Human Obesity,'' International Journal of Obesity and Related 
Metabolic Disorders, vol. 17, suppl. 1, pp. S73-S78, 1993.
    80. Horton, T. J. and C. A. Geissler, ``Aspirin Potentiates the 
Eeffect of Ephedrine on the Thermogenic Response to a Meal in Obese 
But Not Lean Women,'' International Journal of Obesity and Related 
Metabolic Disorders, vol. 15, pp. 359-366, 1991.
    80a. Food and Drug Administration, Memorandum, Beaston, P., 
M.D., Ph.D., ``Review of Boozer-Daly Article for Ephedrine and 
Caffeine,'' January 10, 2003, amended February 2, 2004.
    81. Food and Drug Administration, Memorandum: Evaluation of 
Scientific References Considered by RAND, February 2, 2004.
    81a. Friedman, L. M., C. D. Furberg, and D. L. DeMets, 
Fundamentals of Clinical Trials, 3rd ed., Springer, New York, 1998.
    81b. Ingelfinger, J.A., F. Mosteller, L. A. Thibodeau, et al., 
3rd ed., Biostatistics in Clinical Medicine, McGraw-Hill, Inc., New 
York, 1994.

[[Page 6852]]

    81c. Dawson, B. and R.G. Trapp, Basic and Clinical 
Biostatistics, 3rd ed., Lange Medical Books/McGraw-Hill, New York, 
2001.
    82. Niemann R.A. and M.L. Gay, ``Determination of Ephedrine 
Alkaloids and Synephrine in Dietary Supplements by Column-Switching 
Cation Exchange High-Performance Liquid Chromatography with 
Scanning-Wavelength Ultraviolet and Fluorescence Detection,'' 
Journal of Agricultural and Food Chemistry, vol. 51, pp. 5630-5638, 
2003.
    83. Chua, S. S. and S. I. Benrimoj, ``Non-Prescription 
Sympathomimetic Agents and Hypertension,'' Medical Toxicology and 
Adverse Drug Experience, vol. 3, pp. 387-417, 1988.
    84. Inchiosa, M. A., ``Dose-Related Cardiovascular Effects of 
Ephedrine in Humans,'' Report prepared for the Food and Drug 
Administration, July 6, 1999.
    85. Inchiosa, M. A., ``Bioavailability of Ephedrine Alkaloids,'' 
Report prepared for the Food and Drug Administration, June 14, 1999.
    86. Food and Drug Administration, Food Advisory Committee on 
Dietary Supplements Containing Ephedrine Alkaloids, Meeting 
Transcript, II, pp. 136, 137, 203, 205, 221, 225, 227, 229-230, 236-
237, 239-240, August 27-28, 1996.
    87. Food and Drug Administration, Food Advisory Committee on 
Dietary Supplements Containing Ephedrine Alkaloids, Meeting 
Transcript, II, pp. 137, 233, 237, 240, 272, and 277, August 27-28, 
1996.
    88. Kerin, K. J., Dietary Supplements Containing Ephedrine 
Alkaloids, Comments from Circadian Technologies, Inc., FDA Docket 
No. 1995N-0304, c3923:1-3, 2003.
    89. Lee, M. K., B. W. Cheng, C. T. Che, et al., ``Cytotoxicity 
Assessment of Ma-huang (Ephedra) Under Different Conditions of 
Preparation,'' Toxicological Sciences, vol. 56, pp. 424-430, 2000.
    90. Hikino, H., K. Ogata, C. Konno, et al., ``Hypotensive 
Actions of Ephedradines, Macrocyclic Spermine Alkaloids of Ephedra 
Roots,'' Planta Medica, vol. 48, pp. 290-293, 1983.
    91. Minematsu, S., Y. Kobayashi, N. Kobayashi, et al., ``Acute 
Ephedrae Herba and ephedrine poisoning in mice,'' Japanese Journal 
of Toxicology, vol. 4, pp. 143-149, 1991.
    92. Lee, M. K., Y. H. Wong, C. T. Che, et al., ``Adrenergic 
Agonistic Effects and Cyto-Toxicity of Chinese Ephedra (Ma-Huang) 
Used for Weight Reduction,'' Toxicological Sciences, The 
Toxicologist, vol. 48, p. 58, abstract no. 272, 1999.
    93. CANTOX Health Sciences International, ``Safety Assessment 
and Determination of a Tolerable Upper Limit for Ephedra,'' pp. 1-
273, Docket No. 2000N-1200, Sup2 (http://www.crnusa.org/
cantoxreportindex.html), 2000.
    94. Ad Hoc Committee Reports, On the Safety of Ma Huang (1995, 
1996), On the Safety of Dietary Supplements (1997), FDA Docket Nos. 
1995N-0304, RPT 1, vols. 36-38; 1995N-0304, C57, vol. 76; 1995N-
0304, C3391, vol. 280.
    95. Ephedra Education Council Expert Panel Report, Docket No. 
2000N-1200, C30, vol. 70, 2000.
    96. Boozer, C. N., P. A. Daly, D. Blanchard, et al., ``Herbal 
Ephedra/Caffeine for Weight Loss: A 6-Month Safety and Efficacy 
Trial,'' North American Association for the Study of Obesity, 2000.
    97. Dulfano, M. J. and P. Glass, ``Evaluation of a New 
B2 Adrenergic Receptor Stimulant, Terbutaline, in 
Bronchial Asthma. II. Oral Comparison With Ephedrine,'' Current 
Therapeutic Research, vol. 15, pp. 150-157, 1973.
    98. Tashkin, D. P., R. Meth, D. H. Simmons, et al., ``Double-
Blind Comparison of Acute Bronchial and Cardiovascular Effects of 
Oral Terbutaline and Ephedrine,'' Chest, vol. 68, pp. 155-161, 1975.
    99. Brown, S. D., Jr., and F. J. Landry, ``Recognizing, 
Reporting, and Reducing Adverse Drug Reactions,'' Southern Medical 
Journal, vol. 94, pp. 370-373, 2001.
    100. Lazarou, J., B. H. Pomeranz, and P. N. Corey, ``Incidence 
of Adverse Drug Reactions in Hospitalized Patients: A Meta-Analysis 
of Prospective Studies,'' Journal of the American Medical 
Association, vol. 279, pp. 1200-1205, 1998.
    101. American Herbal Products Association, Code of Ethics and 
Business Conduct, pp. 1-9, July 2002.
    102. Lake, C. R. and R. S. Quirk, ``CNS Stimulants and the Look-
Alike Drugs,'' Psychiatric Clinics of North America, vol. 7, pp. 
689-701, 1984.
    103. Furuya, I. and S. Watanabe, ``Discriminative Stimulus 
Properties of Ephedra Herb (Ephedra Sinica) in Rats,'' Yakubutsu 
Seishin Kodo (Japanese Journal of Psychopharmacology), vol. 13, pp. 
33-38, 1993.
    104. Tinsley, J. A. and D. D. Watkins, ``Over-the-Counter 
Stimulants: Abuse and Addiction,'' Mayo Clinic Proceedings, vol. 73, 
pp. 977-982, 1998.
    105. Ephedra Sinica, In Gruenwald, J., T. Brendler, and C. 
Jaenicke, eds., PDR for Herbal Medicines, 1st ed., Montvale: Medical 
Economics Co., Inc., pp. 826-827, 1998.
    106. The Complete German Commission E Monographs, Therapeutic 
Guide to Herbal Medicines (English Translation), Blumenthal, M., W. 
R. Busse, A. Goldberg, et al., pp. 125-126, 1998.
    107. Huang, K. C., Antiasthmatic Herbs, Chapter in The 
Pharmacology of Chinese Herbs, pp. 229-232, Boca Raton, CRC Press, 
1993.
    108. Hsu, H. Y., Y. P. Chen, S. J. Shen, et. al, Ephedrae Herba 
(Ephedra, Ma-Huang), Oriental Materica Medica: A Concise Guide, pp. 
52-53, Long Beach, CA, Oriental Healing Arts Institute, 1986.
    109. Food and Drug Administration, Clinical Research and Review 
Staff: Evaluation of Adverse Events Reported With the Use of 
Ephedrine Alkaloid-containing Dietary Supplements (June 1, 1997, 
through March 31, 1999), 2000.
    110. Food and Drug Administration, Chen, M. and C. B.Karwoski, 
Office of Post Market Drug Risk Assessment Postmarketing Safety 
Review, ``Evaluation of Adverse Event Reports Associated With 
Ephedrine Alkaloid-Containing Dietary Supplements,'' 2000.
    111. Woosley, R. L., ``Summary of Analysis of Adverse Event 
Reports for Dietary Supplements Containing Ephedrine Alkaloids,'' 
Report Prepared for the Food and Drug Administration, 1999.
    112. Benowitz, N. L., ``Review of Adverse Reaction Reports 
Involving Ephedrine-Containing Herbal Products,'' Report Prepared 
for the Food and Drug Administration, 2000.
    113. Ricaurte, G. A., ``Review of FDA Neuropsychiatric/central 
nervous system adverse events related to ephedrine-containing 
dietary supplements,'' Report Prepared for the Food and Drug 
Administration, 2000.
    114. Stoll, A. L., ``FDA Ephedrine Adverse Event Case Review: 
Summary,'' 1999.
    115. Woosley, R. L., ``FDA Addendum: Summary of Analysis of 
Adverse Event Reports for Dietary Supplements Containing Ephedrine 
Alkaloids,'' Report Prepared for the Food and Drug Administration, 
2000.
    116. Food and Drug Administration, Love, L. A., ``FDA Review of 
the Published Literature on the Physiological, Pharmacological and 
Toxic Effects of Ephedrine Alkaloids,'' 2000.
    117. U.S. House of Representatives, Committee on Government 
Reform, Minority Staff Report, ``Adverse Event Reports from 
Metabolife,'' 2002.
    118. Shekelle, P. G., E-Mail Response to Questions on the RAND 
Report on Ephedra, April 7, 2003.
    119. Shekelle, P. G., M. Maglione, and S. C. Morton, 
``Preponderance of Evidence. Judging What to Do About Ephedra,'' 
RAND Review, vol. 27, pp. 16-21, (http://www.rand.org/publications/
randreview/issues/spring2003/evidence.html), 2003, accessed in 
January 2004.
    120. ``Danish Medicines Agency: Letigen[reg] has been 
suspended,'' Danish Medicines Agency 2002 (http://
www.laegemiddelstyrelsen.dk/en/news/letigen--6--en.asp), accessed as 
of March 3, 2003. (To obtain a paper copy see the FOR FURTHER 
INFORMATION CONTACT section of this document.)
    121. ``Adverse Reactions Letigen,'' Danish Medicines Agency, 
2002 Company Report, 18, 2002.
    122. Walker, A. M., ``The Relation Between Voluntary 
Notification and Material Risk in Dietary Supplement Safety,'' 
Report prepared for the Food and Drug Administration, 2000.
    123. Food and Drug Administration, Letter from Melinda Plaisier 
to Congressman Henry Waxman (CA) dated July 11, 2001 and Tab C 
(preliminary evaluation of data from American Association of Poison 
Control Centers) enclosed with the letter, 2001.
    124. U.S. General Accounting Office (GAO), GAO Report, ``Dietary 
Supplements: Uncertainties in Analyses Underlying FDA's Proposed 
Rule on Ephedrine Alkaloids,'' GAO, 1999, submitted to Docket No. 
2000N-1200, BKG 1, vol. 32.
    125. Boozer, C. N., J. A. Nasser, S. B. Heymsfield, et al., ``An 
Herbal Supplement Containing Ma Huang-Guarana For Weight Loss: A 
Randomized, Double-Blind Ttrial,'' International Journal of Obesity 
and Related Metabolic Disorders, vol. 25, pp. 316-324, 2001.
    126. National Institutes of Health, National Heart, Lung, and 
Blood Institute, Third Report of the Expert Panel on Detection,

[[Page 6853]]

Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult 
Treatment Panel III) Final Report, NIH Publication No. 02-5215 
(http://www.nhlbi.nih.gov/guidelines/cholesterol), 2002.
    127. Bray, G. A. and F. L. Greenway, ``Current and Potential 
Drugs for Treatment of Obesity,'' Endocrine Reviews, vol. 20, pp. 
805-875, 1999.
    128. Yanovski, S. Z. and J. A. Yanovski, ``Obesity,'' New 
England Journal of Medicine, vol. 346, pp. 591-602, 2002.
    129. Food and Drug Administration, Division of Metabolic and 
Endocrine Drug Products, FDA Draft Guidance for the Clinical 
Evaluation of Weight-Control Drugs, 1996.
    130. International Committee on Harmonization (ICH): Guidance 
for Industry, S7A Safety Pharmacology Studies for Human 
Pharmaceuticals, ICH, pp. 1-11 (http://www.fda.gov/cder/guidance/
4461fnl.pdf), 2001.
    131. Filozof, C., C. Gonzalez, and E. Hofman, ``The Effect 
Ephedrine Plus Caffeine After a 4-week Portion-Controlled Diet,'' 
Ninth International Congress on Obesity, Abstract 591, Ninth 
International Congress on Obesity, vol. 26, suppl. 1, p. S156, 2002, 
August 2002.
    132. Food and Drug Administration, ``Evidence On The Safety And 
Effectiveness Of Ephedra: Implications For Regulation,'' (http://
www.fda.gov/bbs/topics/NEWS/ephedra/whitepaper.html), 2003.
    133. Boerth, J. M., C. F. Caley, ``Possible Case of Mania 
Associated With Ma-huang,'' Pharmacotherapy, vol. 23, pp. 380-383, 
2003.
    134. Samenuk, D., M. S. Link, M. K. Homoud, et al., ``Adverse 
Cardiovascular Events Temporally Associated With Ma-huang, an Herbal 
Source of Ephedrine,'' Mayo Clinic Proceedings, vol. 77, pp. 12-16, 
2002.
    135. Bent, S., T. N. Tiedt, M. C. Odden, et al., ``The Relative 
Safety of Ephedra Compared With Other Herbal Products,'' Annals of 
Internal Medicine, vol. 138, pp. 468-471, 2003.
    136. Morgenstern, L. B., C. M. Viscoli, W. N. Kernan, et al., 
``Use of Ephedra-Containing Products and Risk for Hemorrhagic 
Stroke,'' Neurology, vol. 60, pp. 132-135, 2003.
    137. Food and Drug Administration, Policy and Guidance Handbook 
For FDA Advisory Committees, pp. 113-114, 1994.
    138. Food and Drug Administration, Nightingale, S. L., FDA 
``Dear Colleague'' Letter Regarding the Research on Eosinophilia-
Myalgia Syndrome and Current Regulatory Status of L-Tryptophan, 1992 
(http://www.cfsan.fda.gov/dms/ds-ltr3.html).
    139. Chyka, P. A., McCommon, S. W., ``Reporting of adverse drug 
reactions by poison control centres in the U.S.,'' Drug Safety, vol. 
23, pp. 87-93, 2000.
    140. Office of Management and Budget: Regulatory Analysis, 
Circular A-4 (http://www.whitehouse.gov/omb/circulars/a004/a-
4.html), 2003.
    141. Shekelle P, S. Morton, M. Maglione, et al., ``Ephedra and 
Ephedrine for Weight Loss and Athletic Performance Enhancement: 
Clinical Efficacy and Side Effects. Evidence Report/Technology 
Assessment,'' No. 76 (Prepared by Southern California Evidence-Based 
Practice Center, RAND, under Contract No. 290-97-0001, Task Order 
No. 9), AHRQ Publication No. 03-E022, Rockville, MD, Agency for 
Healthcare Research and Quality, Table 23, 2003.
    142. Centers for Disease Control and Prevention, ``Life 
Expectancy'' (Data are for U.S. in 2001). National Center for Health 
Statistics, Centers for Disease Control and Prevention, U. S. 
Department of Health and Human Services (http://www.cdc.gov/nchs/
fastats/lifexpec.htm), accessed on December 9, 2003.
    143. Derdeyn, C. P., Powers, and W. J., ``Cost-Effectiveness of 
Screening for Asymptomatic Carotid Atherosclerotic Disease,'' 
Stroke, vol. 27, pp. 1944-1950, 1996.
    144. Agency for Healthcare Research and Quality (AHRQ), 
Healthcare Cost and Utilization Project, 2001 National Statistics, 
The ``National Bill'' for Conditions Treated in Hospitals (Top 100 
Clinical Classifications Software (CCS) Diagnoses), (http://
www.hcup.ahrq.gov/HCUPnet.asp), 2001 .
    145. GNC Plans to Drop Ephedra Private-Label Supplements, FDC 
Reports--Tan Sheet, pp. 11-12, 1996.
    146. Redfearn, S., ``Ephedra Products Thin Out: Braced for the 
Worst, Makers Sell New Herbs,'' The Washington Post, HE01, 2003 
(http://www.washingtonpost.com/wp-dyn/articles/A52270-
2003Jan13.html).
    147. Letter Correction: The Relative Safety of Ephedra Compared 
with Other Herbal Products, Annals of Internal Medicine, vol. 138, 
pp. 1012, 2003 (http://www.annals.org/cgi/content/full/138/12/1012), 
accessed on December 11, 2003.
    148. Dickinson, A., Letter, ``The Relative Safety of Ephedra 
Compared With Other Herbal Products,'' Annals of Internal Medicine, 
vol. 139, pp. 385-386 (http://www.annals.org/cgi/content/full/139/
5_Part_1/385-b) 2003, accessed on September 2, 2003.
    149. Nutrition Business Journal, 2002 Facts and Stats about 
Dietary Supplements (obtained from http://www.nnfa.org/facts), 2002.
    150. Kaufman, D. W., Kelly, J. P., Mitchell, A. A., ``Use of 
Ephedra-Containing Products in the U.S. Population,'' Data from the 
Slone Survey, FDA Docket No. 1995N-0304, emc126, vol. 297, 2003.
    151. Research Triangle Institute, ``Cost of Reformulating Foods 
and Cosmetics,'' Final Report, Report Prepared for the Food and Drug 
Administration, RTI Project Number 08184.003, 2002.
    152. Gugliotta, G., ``Ephedra Lawsuits Show Big Increase,'' The 
Washington Post, July 23, 2000, (http://www.washingtonpost.com/ac2/
wp-dyn?
pagename=article&node=&contentId=
A25415-2000Jul21¬Found=true), accessed on February 3, 2004.
    153. Gugliotta, G., ``Woman Wins $13.3 Million Against Dietary 
Company,'' The Washington Post, February 2, 2001 (http://
www.washingtonpost.com/ac2/wp-dyn?
pagename=article&node=&contentId=
A40923-2001Feb7¬Found=true), accessed on February 3, 2004.
    154. Prevention Magazine, ``Consumer Use of Dietary 
Supplements,'' 1999.
    155. National Consumers League, National Consumers League's 
Survey of Consumers Using Over-the-Counter (OTC) Medications, 2003.
    156. Geiger, C. J., ``Health Claims: History, Current Regulatory 
Status, and Consumer Research,'' Journal of the American Dietetic 
Association, vol. 98, pp. 1312-1322, 1998.
    157. MacKinnon, D. P., L. Nohre, J. Cheong, et al., 
``Longitudinal Relationship Between the Alcohol Warning Label and 
Alcohol Consumption,'' Journal of Studies on Alcohol, vol. 62, pp. 
221-227, 2001.
    158. Food and Drug Administration, Memorandum: Label Analysis 
for Dietary Supplements Containing Ephedrine Alkaloids, January 28, 
2004.
    159. Research Triangle Institute, Dietary Supplement Sales 
Information, Supplement to the Final Report, Report Prepared for the 
Food and Drug Administration, Contract No. 223-96-2290 Task Order 
4:1999 (http://www.foodriskclearinghouse.umd.edu/dspd.htm), accessed 
on October 1999.
    160. Research Triangle Institute, Survey of Manufacturing 
Practices in the Dietary Supplement Industry. Final Report, Report 
prepared for the Food and Drug Administration, RTI Project Number 
6673-6:2000 (http://www.foodriskclearinghouse.umd.edu/smpds.htm), 
accessed on May 17, 2000.
    161. Office of Management and Budget, Guidance for Implementing 
E.O. 13132, ``Federalism,'' October 28, 1999.

List of Subjects in 21 CFR Part 119

    Dietary ingredients, Dietary supplements, Foods.

0
Therefore, under the Federal Food, Drug, and Cosmetic Act and under 
authority delegated to the Commissioner of Food and Drugs, 21 CFR part 
119 is added as follows:
0
1. Part 119 consisting of Sec. 119.1 is added to read as follows:

PART 119--DIETARY SUPPLEMENTS THAT PRESENT A SIGNIFICANT OR 
UNREASONABLE RISK


Sec. 119.1   Dietary supplements containing ephedrine alkaloids.

    Dietary supplements containing ephedrine alkaloids present an 
unreasonable risk of illness or injury under conditions of use 
recommended or suggested in the labeling, or if no conditions of use 
are recommended or suggested in the labeling, under ordinary conditions 
of use. Therefore, dietary supplements containing ephedrine alkaloids 
are adulterated under section 402(f)(1)(A) of the Federal Food, Drug, 
and Cosmetic Act.

    Authority: 21 U.S.C. 321, 342, 343, 371.


[[Page 6854]]


    Dated: January 28, 2004.
Mark B. McClellan,
Commissioner of Food and Drugs.

    Dated: February 4, 2004.
Tommy G. Thompson,
Secretary of Health and Human Services.
[FR Doc. 04-2912 Filed 2-6-04; 2:00 pm]
BILLING CODE 4160-01-S