[Federal Register: October 23, 1998 (Volume 63, Number 205)]
[Rules and Regulations]
[Page 56789-56802]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr23oc98-7]

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

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Part 201

[Docket No. 77N-094W]


Over-the-Counter Drug Products Containing Analgesic/Antipyretic
Active Ingredients for Internal Use; Required Alcohol Warning

AGENCY: Food and Drug Administration, HHS.

ACTION:  Final rule.

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

SUMMARY: The Food and Drug Administration (FDA) is amending its
regulations to require an alcohol warning for all over-the-counter
(OTC) drug products, labeled for adult use, containing internal
analgesic/antipyretic active ingredients. The required warning
statements advise consumers with a history of heavy alcohol use to
consult a physician for advice about the use of OTC internal analgesic/
antipyretic drug products. FDA is issuing this final rule after
considering comments on the agency's proposed regulation for OTC
internal analgesic, antipyretic, and antirheumatic drug products; a
proposed regulation to establish an alcohol warning; recommendations of
its Nonprescription Drugs Advisory Committee (NDAC) and Arthritis Drugs
Advisory Committee (ADAC); and new data and information that have come
to the agency's attention. This final rule is part of the ongoing

[[Page 56790]]

review of OTC drug products conducted by FDA.

EFFECTIVE DATE:  April 23, 1999.

FOR FURTHER INFORMATION CONTACT:  Debbie L. Lumpkins, Center for Drug
Evaluation and Research (HFD-560), Food and Drug Administration, 5600
Fishers Lane, Rockville, MD 20857, 301-827-2241.

SUPPLEMENTARY INFORMATION:

I. Background

    In the Federal Register of November 16, 1988 (53 FR 46204), FDA
published a notice of proposed rulemaking, in the form of a tentative
final monograph (TFM), that would establish conditions in part 343 (21
CFR part 343) under which OTC internal analgesic, antipyretic, and
antirheumatic drug products are generally recognized as safe and
effective and not misbranded. In the preamble to the proposed rule of
this current rulemaking, the agency addressed concerns raised in the
1988 proceeding about the need for a warning on the increased risk of
liver toxicity when acetaminophen is taken with substances or drugs
that induce microsomal enzyme activity, i.e., alcohol, barbiturates, or
prescription drugs for epilepsy (53 FR 46204 at 46217). The agency
found that the available data did not provide a sufficient basis to
require such a warning at that time. Interested persons were invited to
submit new data or file written comments, objections, or requests for
oral hearing before the Commissioner of Food and Drugs regarding the
proposal.
    In response to the proposed rule, the agency received a number of
comments containing new data addressing the need for an alcohol warning
for acetaminophen. Copies of the comments received are on display in
the Dockets Management Branch (HFA-305), Food and Drug Administration,
5630 Fishers Lane, rm. 1061, Rockville, MD 20852.
    On June 29, 1993, NDAC met to consider the need for an alcohol
warning for acetaminophen. NDAC concluded that heavy drinkers are at
increased risk for developing liver toxicity when using acetaminophen
and recommended that the labeling of OTC analgesic/antipyretic drug
products containing this ingredient bear an alcohol warning. However,
NDAC recommended that the agency not implement an alcohol warning for
OTC analgesic/antipyretic drug products containing acetaminophen until
it had a chance to consider data on the risk of alcohol use with other
internal analgesic/antipyretic ingredients.
    On September 8, 1993, NDAC and ADAC (the Committees) met jointly to
evaluate the available data on the use of aspirin and other OTC
analgesics by heavy alcohol users or abusers. The Committees concluded
that the use of aspirin, ibuprofen, and naproxen sodium increases the
risk of upper gastrointestinal (UGI) bleeding in heavy alcohol users or
abusers. Concerning whether the data support an alcohol warning for OTC
drug products containing these ingredients, the Committees voted 12
yes, 2 no for aspirin; 12 yes, 2 no for ibuprofen; and 12 yes, 1 no,
and 1 abstention for naproxen sodium. The Committees further concluded
that a recommendation on the need for an alcohol warning for OTC drug
products containing other monograph salicylates (carbaspirin calcium,
choline salicylate, magnesium salicylate, or sodium salicylate) was
outside their advisory scope.
    In the Federal Register of November 14, 1997 (62 FR 61041), the
agency published a proposed amendment of part 201 (21 CFR part 201)
that would establish alcohol warnings for all OTC drug products labeled
for adult use containing internal analgesic/antipyretic active
ingredients. This warning would be required for all OTC internal
analgesic/antipyretic drug products whether marketed under an OTC drug
monograph or an approved new drug application (NDA).
    In the proposal to amend part 201, the agency advised that any
final rule based on the proposal will be effective 6 months after the
date of publication in the Federal Register. Therefore, on or after
April 23, 1999, any OTC drug product that is subject to this final
rule, that contains nonmonograph labeling may not be initially
introduced or initially delivered for introduction into interstate
commerce unless it is the subject of an approved application or
abbreviated application. Further, any OTC drug product subject to this
final rule that is repackaged or relabeled after the effective date of
the rule must be in compliance with the rule regardless of the date
that the product was initially introduced or initially delivered for
introduction into interstate commerce.

II. The Agency's Response to Comments

A. Comments on Specific Ingredients

    1. Two comments argued that the agency's proposed requirement for
an alcohol warning for OTC analgesic/antipyretic drug products
containing aspirin is not based on sound scientific evidence. One
comment asserted that it is necessary for FDA to demonstrate that a
significant risk of gastrointestinal (GI) bleeding would result if
heavy alcohol users were not specifically warned against the use of
aspirin. Both comments suggested that the proposed requirement is
contrary to agency statements in the TFM for OTC internal analgesic/
antipyretic drug products that warning statements should be ``limited
to those that are scientifically documented, clinically significant,
and important for the safe and effective use of products by consumers''
(53 FR 46204 at 46213).
    In support of this position, one comment included data that purport
to show that heavy alcohol use: (1) Does not increase the risk of
stomach bleeding (Refs. 1 through 4), (2) alcohol protects against GI
problems (Refs. 5 and 6), and (3) GI bleeding in patients who reported
prior aspirin and alcohol use is not more severe (Ref. 7). The comment
also asserted that its evaluation of the adverse drug reaction data
contained in FDA's Spontaneous Reporting System (SRS) failed to
demonstrate a correlation between GI bleeding and heavy alcohol use,
although the results of this evaluation were not included.
    Another comment supporting the need for an alcohol warning for OTC
analgesic/antipyretic drug products containing aspirin reviewed the
data evaluated by the agency during the development of its proposal. To
substantiate the need for an alcohol warning for aspirin, the comment
also included data from a recently published study of the relationship
between aspirin and nonsteroidal anti-inflammatory drug (NSAID) use and
GI perforation (Ref. 8).
    The agency continues to believe that warning statements should be
limited to those that are scientifically based, clinically relevant,
and important for the safe and effective use of these products by
consumers. The agency disagrees with the comments asserting that the
alcohol warning is not based on solid scientific evidence. An alcohol
warning is needed for OTC analgesic/antipyretic drug products
containing nonsteroidal anti-inflammatory ingredients, including
aspirin. This warning is based on the data and information on the
adverse GI effects of aspirin and other NSAID ingredients, the adverse
GI effects of alcohol use, and the documented risk of combining them.
    Although the previous comments pertain specifically to aspirin-
containing OTC analgesic/antipyretic products, the agency's response
will provide the scientific reasoning for applying the alcohol warning

[[Page 56791]]

requirement to the pharmacologic class of OTC analgesic/antipyretic
drug products containing nonsteroidal anti-inflammatory ingredients,
which include aspirin, nonaspirin salicylates, ibuprofen, ketoprofen,
and naproxen sodium.
    These OTC analgesic/antipyretic drug products contain NSAID
ingredients, which belong to the carboxylic acid class. Aspirin and
other salicylates are salicyclic acids; ibuprofen, ketoprofen, and
naproxen sodium are derivatives of propionic acid. All of these
ingredients share certain pharmacologic properties, including
inhibitory effects on prostaglandin synthesis and platelet function. As
with aspirin, propionic acid derivatives produce adverse GI side
effects, alter platelet function, and can affect bleeding time (Refs. 9
through 14). Adverse GI effects are caused by aspirin and nonaspirin
NSAID ingredients, which can irritate the mucosal epithelium (stomach
lining) directly and/or can suppress prostaglandin synthesis.
Prostaglandins normally help protect the stomach lining by promoting
secretion of mucus and bicarbonate, repair of epithelial (lining)
cells, immune cell function, and blood flow. Adverse bleeding effects
can occur because NSAID's inhibit platelet aggregation.
    Although there are data and information available concerning all of
these ingredients, the largest body of data relied upon by the agency
pertains to aspirin. Because these NSAID ingredients all share similar
pharmacologic properties and can all cause adverse GI effects,
including bleeding, it is reasonable for the agency to rely on the data
pertaining to individual ingredients and to reason and apply these data
to all of these NSAID ingredients. More specific comments concerning
other ingredients will be addressed elsewhere in section II of this
document.
    Drug-related adverse effects can be evaluated through clinical data
collected various ways, including randomized controlled trials, cohort
studies, case-control studies, surveys, and spontaneous case reports.
Prospective, randomized, blinded clinical trials require large patient
enrollments to demonstrate a difference between groups when adverse
events are infrequent, even if serious. Thus, most studies which
examine the adverse GI effects of NSAID's are observational rather than
experimental. Observational studies provide important information when
investigating an association between a risk and a predisposing event.
However, these studies may be subject to specific biases which should
be considered. For example, case-control studies examine the prevalence
of NSAID (and alcohol) exposure in patients who already have the
outcome (GI events or bleeding) with a control population, which is
matched for other factors. These studies may suffer from recall bias;
that is, individuals in cases may be more likely than controls to
remember that they took an NSAID (or alcohol). When reviewing these
data from various studies, the agency has taken into account the
limitations of each study method. Despite the limitations of individual
studies, the data generated by each of these methods collectively
provide a sound body of evidence from which it is scientifically
reasonable to assess risk. Therefore, the agency believes that the
collected body of scientific evidence supports the labeled warning.
    As previously discussed in the notice of proposed rulemaking (62 FR
61041 at 61049), the adverse GI effects of aspirin are well known.
Medical texts document adverse effects associated with the use of
aspirin. These effects include, but are not limited to, gastritis,
ulcerations, and colitis (Refs. 15 through 18). In addition, aspirin
irreversibly interferes with normal platelet function for the life of
the platelet, prolongs the bleeding time, and interferes with clotting
whenever bleeding occurs (Ref. 13). Nonsalicylate NSAID ingredients
reversibly inhibit platelet aggregation for as long as the drug is in
the blood (Refs. 13 and 14). GI mucosal damage caused by aspirin has
been widely acknowledged in the medical literature (Ref. 15 through
18), confirmed by endoscopic observational studies (Ref. 19), and
taught through medical texts to students of medicine (Ref. 20).
    In 1977, the Advisory Review Panel for OTC Analgesic and
Antipyretic Drug Products (the Panel) first reviewed relevant data and
concluded that aspirin causes adverse GI effects. The Panel concluded
that the adverse effects of aspirin on the GI system range from
relatively mild effects such as gastric distress (minor stomach pain,
heartburn, or nausea), mucosal irritation and occult (not easily seen)
bleeding, to less frequent but more serious effects such as mucosal
erosion, ulceration, and life-threatening massive bleeding. The Panel
further concluded that the acute use of aspirin may activate symptoms
of both gastric and duodenal ulcer (42 FR 35346 at 35386 through 35397,
July 8, 1977).
    In addition to the Panel's conclusions, FDA also evaluated
published literature, including studies which demonstrate adverse GI
effects even with low-dose aspirin use (Refs. 21 and 22). The agency
also reviewed data from controlled, prospective clinical trials on
aspirin for cardiovascular and cerebrovascular uses and established
that bleeding can occur with long-term aspirin use, even at low doses
(62 FR 61041 at 61050).
    Just as aspirin is well known to produce adverse GI effects,
including bleeding, it is also well known that alcohol is a gastric
toxin and that heavy alcohol use may cause a number of adverse GI
effects, including bleeding. Routinely heavy alcohol use is associated
with a number of medical conditions. These conditions include, but are
not limited to, esophagitis, varices, acute gastritis, hemorrhagic
lesions of the duodenal villi, and peptic ulcer disease (Refs. 23
through 28). Also, chronic heavy alcohol use can cause bleeding because
of increased prothrombin time, decreased circulating platelets, and
altered function of platelets (Ref. 13). Early (Ref. 23) and continuing
(Refs. 24 through 26) study of the effects of alcohol on the stomach
have been widely published in the scientific literature and alcoholic
gastritis is a well-recognized cause of acute hemorrhagic gastritis
(Ref. 29). These effects of heavy, chronic alcohol use on the GI system
and bleeding parameters are explained in many standard medical
textbooks (Refs. 25, 27 and 28).
    The Panel recognized alcohol as a major factor that may produce
acute gastric mucosal lesions, and thus increase the risk of bleeding
from the use of aspirin (42 FR 35346 at 35479). Given these
observations and the well established and recognized medical acceptance
of GI and bleeding problems associated with the use of either aspirin
or alcohol, the agency was concerned about the risks present for
consumers who routinely and heavily drink alcohol and also use aspirin.
This concern led to a review of relevant medical literature and studies
(Refs. 8, 30, and 31), which confirmed the increased risk of adverse GI
events, including bleeding, when alcohol use and aspirin use are
combined.
    Published studies which include randomized controlled clinical
trials (Refs. 32 through 35), case-control studies (Refs. 8, 36 through
39a), cohort studies (Ref. 40), meta-analyses (Refs. 41 and 42),
physician surveys (Ref. 31), and case reports (Ref. 43) have
established an association between NSAID's, including aspirin, and
adverse GI events, including bleeding. Because chronic alcohol use
causes GI disease and bleeding, some studies simply exclude these
patients from entry or analysis when assessing the risk of NSAID use on
adverse GI outcomes (Ref.

[[Page 56792]]

44). However, some studies have examined both NSAID and alcohol use
(Refs. 8, 30, 31, and 45) and assessed the risk of developing adverse
GI events, including bleeding.
    P. J. DeSchepper et al. (Ref. 45) measured fecal blood loss in 10
healthy males in a double-blind, parallel study and in 12 healthy
subjects in a double-blind crossover study. Fecal blood loss was
demonstrated with aspirin ingestion and concomitant ingestion of
alcohol significantly increased (by three times) this blood loss.
    D. Aarons et al. (Ref. 30) conducted a double blind prospective
study of 27 healthy volunteers with initial normal baseline endoscopies
who were given alcohol and either placebo, aspirin, or acetaminophen.
Repeat endoscopy showed that alcohol and aspirin together caused
significantly greater erythema (redness) due to irritation and
hemorrhage in the stomach than alcohol alone.
    The agency has reviewed adverse events reported to its SRS data
base (Ref. 43). From 1993 to 1995, 37 case reports were submitted for
serious UGI bleeding, 36 involving hospitalizations and 1 death. Most
bleeds were documented by endoscopy. In these reports, ibuprofen was
listed as the suspect drug in patients who reported chronic alcohol use
(nearly 80 percent reported alcoholism or more than two drinks/day). Of
important note, concomitant use of salicylates, primarily aspirin, was
reported in almost 50 percent of these cases, thus associating both
ibuprofen and/or salicylates with these reports of bleeding. From 1994
to 1996, five case reports were submitted for serious UGI bleeding with
naproxen sodium listed as the suspect drug in patients who reported
daily (or binge) alcohol ingestion. Two of these reports also listed
salicylate use and two reports listed concomitant ibuprofen use. From
1993 to 1996, 10 case reports were submitted for serious UGI bleeding
with aspirin listed as the suspect drug in patients who also reported
alcohol ingestion (more than 2 drinks/day or unspecified). All 10 cases
were hospitalized. Cases of concomitant NSAID ingredient use were
excluded. Thus, the agency's SRS data base provides additional serious
adverse events documenting the association between NSAID ingredient use
and UGI bleeding in persons with a history of chronic alcohol use.
    In a prospective community clinical case study, Lee et al. (Ref.
46) endoscoped 400 consecutive patients hospitalized for UGI hemorrhage
to identify factors which predispose patients who bleed from
hemorrhagic erosive gastritis. Of the 74 patients with stomach
bleeding, salicylate use (31 percent), alcohol use, usually chronic (27
percent), or both (16 percent) were reported. There was no case-matched
control and relative risk was not assessed. However, this study
demonstrates that patients who have experienced hemorrhagic erosive
gastritis (stomach bleeding) commonly report having used alcohol and/or
salicylates.
    Peura et al. (Ref. 31) surveyed American College of
Gastroenterology physicians to assess demographics, management
strategies, and outcomes for 1,235 patients who were diagnosed with GI
bleeding. OTC doses of NSAID's were associated with a three-fold
increased risk for developing GI bleeding and alcohol use increased
this risk to four-fold.
    Lanas et al. (Ref. 8) conducted a single-center, prospective, case-
controlled study, which examined the relationship between NSAID use,
including aspirin, and GI perforation. Detailed clinical histories and
laboratory tests were obtained in 76 hospital admitted patients with
surgically documented GI perforations and in 152 matched case controls.
Histories of NSAID use were confirmed by measuring platelet cyclo-
oxygenase activity. In the study cohort, 67 percent of the patients
used aspirin (90 percent of these were over-the-counter formulations).
The calculated odds ratio (OR) for GI perforation in patients who had
used an NSAID within a week prior to hospitalization was 6.64 (95
percent confidence interval: 3.6-12.2; p < 0.0001) as compared to those
who had not. Other independent risk factors for perforation included
smoking (OR: 3.88; 95 percent CI: 2.15-7.0; p<0.0001), alcohol
ingestion (OR: 3.25; 95 percent CI: 1.81-5.82; p<0.0001), and peptic
ulcer disease (OR: 3.29; 95 percent CI: 1.74-6.21; p<0.0005). The
combination of NSAID's, smoking, and alcohol increased the risk of GI
perforation (OR: 10.69; 95 percent CI: 3.60-29.87). Because the study
was conducted in Spain, a small number of patients in both cohorts
reported use of NSAID's which are not available in the United States.
However, the study conclusions remain valid for the NSAID class and,
importantly, for nonprescription aspirin.
    Although acute ingestion of aspirin and alcohol causes gastric
hemorrhage (Ref. 30) in previously normal gastric mucosa, the increased
bleeding risk from NSAID's in chronic heavy alcohol users can be
further compounded by coexisting problems such as prolonged prothrombin
time due to liver disease, decreased number of circulating platelets,
and pre-existing GI disease (e.g., esophageal varices, ulcers, or
alcoholic gastritis) (Ref. 13). Alcohol also potentiates the
prolongation of bleeding time produced by aspirin and nonaspirin
NSAID's, including ibuprofen (Ref. 14). A retrospective cohort study,
using a Medicaid data base, was designed to determine the risk and cost
of adverse GI effects associated with NSAID use (Ref. 47). Logistic
regression analysis showed NSAID use was significantly associated with
each defined GI side effect (i.e., ulcers, gastritis, bleeding)
(p<.001) and alcohol-related diagnoses were a significant independent
predictor of increased risk (p<.05) for GI bleeding and hemorrhagic
gastritis. Therefore, co-existing GI and bleeding problems in chronic
heavy alcohol users may pre-dispose to the increased bleeding risk from
NSAID ingredients.
    The data and studies presented provide sound and convincing
evidence to support the conclusion that consumers are at increased risk
of adverse GI effects when using OTC analgesic/antipyretic products,
including aspirin, in combination with routine heavy alcohol use (Refs.
8 and 31). While the data and studies show that there is an increased
risk to consumers who combine these drug products with routine heavy
alcohol use, the agency acknowledges that the data differ as to the
exact magnitude of this increased risk.
    The agency again convened expert advisors in 1993 (Refs. 48 to 50)
in three separate advisory committee meetings with NDAC and ADAC, to
discuss the question of whether OTC analgesic/ antipyretic products
containing aspirin should bear an alcohol warning. The advisory
committee experts concluded that aspirin increases the risk of UGI
bleeding in heavy alcohol users or abusers and overwhelmingly concluded
that the data support an alcohol warning for aspirin. A complete
discussion of this conclusion can be found in the proposed rulemaking
(62 FR 61043 through 61044).
    The agency has reviewed the data and information submitted with the
comments, which both oppose and support a requirement for an alcohol
warning on OTC analgesic/antipyretic drug products containing NSAID
ingredients, including aspirin. The agency's analysis of these data
follows.
    Holvoet et al. (Ref. 1) was reviewed by the Committees which
heavily criticized the study design and did not use it as a basis for
their recommendation (Ref. 48). Coggon, Langman, and Spiegelhelter
(Ref. 2) was a case-control

[[Page 56793]]

study in patients with GI bleeding which reported an increased risk (OR
of 3.7, 95 percent CI: 2.2-6.4) for patients who had recently used
aspirin; but this study did not detect an added risk associated with
alcohol use. However, the study groups were not balanced for alcohol-
use history (p<0.02), compromising the ability of the study to
determine the additional risk, if any, in heavy alcohol users. Bartle,
Gupta, and Lazor (Ref. 3) failed to detect an increased risk of acute
UGI bleeding with weekly alcohol ingestion of 280 milliliters. The
investigators noted, and the agency concurs, that more patients would
be required to assess whether or not an association exists. Although
Schubert et al. (Ref. 6) reported a decreased risk of duodenal ulcer
disease with alcohol use, the study lacked a matched case-control
comparator arm and failed to quantify alcohol ingestion and other co-
factors which may be associated with risks for developing ulcer
disease.
    Likewise, the Cohen et al. (Ref. 5) study submitted to demonstrate
that alcohol is protective against GI bleeding caused by aspirin is not
relevant because this study excluded patients without existing GI
disease and those who drank more than two alcoholic drinks per day.
Thus, the study excluded the very target population required to answer
the question addressed by the agency, namely, individuals who consume
three or more alcoholic drinks every day and/or have concomitant
alcohol associated GI disease. The investigators concluded, and the
agency concurs, that it is impossible to determine from this study that
alcohol protects patients who take aspirin.
     Jensen et al. (Ref. 7) reported that alcohol and aspirin use prior
to hospital admission for the treatment of UGI bleeding was not
associated with certain surrogate variables which were used to estimate
the severity of GI bleeding. All patients were selected because they
required medical treatment for severe UGI hemorrhage, and information
was collected regarding alcohol and aspirin use. However, the study was
not analyzed to evaluate whether reported concomitant aspirin and
alcohol use is associated with a higher risk for developing UGI
bleeding. Therefore, this study did not address the basic question
before the agency, namely, whether there is an increased risk of
stomach bleeding in patients who consumed both alcohol and aspirin.
    Soll (Ref. 4) is a review article on peptic ulcer disease presented
by an expert gastroenterologist. The article reviews the scientific
literature and concludes that NSAID's, including aspirin, produce
topical irritative effects on the mucosa as well as ulcerations as a
consequence of a systemic effect. Therefore, NSAID's, which are
rectally delivered or enteric coated may still cause adverse GI
effects. Similar reviews have been published elsewhere (Refs. 51 and
52). Thus, while the article was submitted in opposition to a warning,
the information in the article supports the scientific rationale for a
warning.
    A case-controlled study was also submitted which supports the need
for an alcohol warning on OTC analgesic/antipyretic drugs containing
NSAID ingredients (Ref. 8). This study has been previously summarized
earlier in this response to comment 1 of section II.A of this document.
    Given the data available at this time, the agency cannot precisely
quantify the increased risk of combining routine heavy alcohol use and
these OTC drug products. In order to require an alcohol warning,
however, it is not necessary that the agency be able to demonstrate
precisely how much the risk is increased. The available data
demonstrate clearly that the risk to consumers of combining heavy
routine alcohol use with these drug products is greater than the risk
of using either alcohol or these drug products alone. These data are
sufficient to establish the need for an alcohol warning on these OTC
products. In light of the clearly demonstrated increased risk to
consumers, the agency is requiring an alcohol warning about the risk of
stomach bleeding on aspirin and other NSAID-containing OTC drug
products.
    In summary, OTC analgesic/antipyretic drug products, including
aspirin, are known to cause adverse GI effects, including bleeding.
Chronic, heavy alcohol use is also associated with adverse GI effects,
including bleeding. Based on the agency's review of a large body of
scientific information and in concurrence with expert advisors, FDA has
determined that routine, heavy (three or more alcoholic drinks every
day) alcohol use in combination with use of OTC analgesic/antipyretic
drug products containing NSAID ingredients increases the risk of
adverse GI events, including stomach bleeding. The agency believes that
the most appropriate public health response to this information
concerning risk is to warn consumers who drink three or more alcoholic
drinks every day to consult their doctor about their use of these OTC
drug products. This conclusion is scientifically based, clinically
relevant, and important for the safe and effective use by consumers of
OTC analgesic/antipyretic drug products containing NSAID ingredients.
    2. One comment argued that FDA's conduct of this rulemaking
violates the Administrative Procedure Act (APA). The comment stated
that the APA requires that a notice of proposed rulemaking include
``either the terms or substance of the proposed rule or a description
of the subjects and issues involved'' (5 U.S.C. 553(b)). The comment
maintained that the agency's proposal fails to adequately describe the
basis for the requirement for an alcohol warning for OTC drug products
containing aspirin. The comment asserted that FDA denied interested
parties adequate notice of the action by failing to expressly state its
reliance on a ``switch rationale,'' i.e, the concern that an alcohol
warning on one analgesic would cause inappropriate ``switching'' to
other OTC analgesic/antipyretic drug products. The comment further
argued that the agency's failure to obtain the raw data from
unpublished epidemiological studies presented to the Committees that
made recommendations also effectively denied interested parties the
opportunity to comment fully.
    Another comment suggested that the ``switch rationale'' is flawed.
The comment asserted that there is no evidence that heavy alcohol users
would be persuaded to change their analgesic use based on an alcohol
warning. One comment noted that after several years of voluntary
alcohol warnings on products other than aspirin, market tracking data
for aspirin sales for the years of 1994 to 1997 have demonstrated that
``switching'' does not occur.
    The intent of the warning is to advise consumers with a history of
heavy alcohol use (three or more alcoholic drinks every day) to consult
a physician for advice about the use of all OTC analgesic/antipyretic
products and to advise that there is a specific risk associated with
use of these products. The agency agrees that it is important not to
encourage consumers who consume three or more alcoholic drinks every
day to begin to use another OTC analgesic/antipyretic drug product
before consulting their physician. In comment 1 of section II.A. of
this document, the agency describes the scientific basis for requiring
an alcohol warning for OTC analgesic/antipyretic drug products
containing NSAID's, including aspirin. This rationale is also present
in the agency's proposal (62 FR 61041 at 61049).
    As discussed in the proposed rule (62 FR 61041 at 61049), the
agency agreed with the assessment of the Advisory

[[Page 56794]]

Committees who made recommendations on the unpublished data presented
before the committees. Raw data were not evaluated by the agency, do
not serve as the agency's basis for this final rule, and are not
required to be placed in the administrative record. The agency
disagrees that interested parties were given insufficient opportunity
to comment fully on the data. Comments on the presentations to the
Committees as well as the Committees' recommendations (Ref. 53) were
included in the administrative record. Further, the comments'
criticisms of the unpublished data presented in September 1993 were
sent to the members of the Committees for their specific comment. Of
the responses received (Ref. 54), none stated that the comments'
criticisms changed their recommendation. The agency has included in the
administrative record the relevant data and information that were
considered and relied upon regarding the warning statement requirements
of the final rule. Therefore, the agency considers the requirements of
the APA to be fully satisfied.
    3. Three comments asserted that the imposition of an alcohol
warning on aspirin could result in a significant adverse impact on
public health. The comments said that placing an unnecessary ``stomach
bleeding'' warning on aspirin may cause consumers taking it for its
cardiovascular and cerebrovascular benefits to avoid using aspirin. The
comments suggested that poor compliance with cardiovascular and
cerebrovascular aspirin regimens could be detrimental to consumers at
risk for these events. One comment noted that consumers on a long-term
professional use regimen would be under a doctor's supervision and
would presumably be warned about the risks of aspirin use and would be
monitored for GI injury. Another comment maintained that the low doses
used in long-term professional use aspirin regimens have not been
associated with significant GI problems.
    In its proposal, the agency evaluated the published literature on
aspirin for cardiovascular and cerebrovascular uses and determined that
bleeding can occur with long-term aspirin use, even at low aspirin
doses. The proposal also discussed the use of alcohol in patients with
cardiovascular problems and noted the recommendations of the American
Heart Association (AHA) that consumers with these conditions should not
consume alcohol heavily (62 FR 61041 at 61050). The proposal further
reviewed the increased risk of cardiovascular diseases, such as heart
muscle disease, hypertension, disturbances in heart rhythm, and stroke
from heavy alcohol use. The intended purpose of this warning is to
promote a dialogue between physicians and individuals who consume three
or more drinks every day. The agency believes that this dialogue should
extend to consumers on long-term aspirin regimens who may be adding to
their risk of adverse vascular events by their alcohol consumption.
Therefore, the agency concludes that an alcohol warning on OTC
analgesic/antipyretic drug products containing aspirin will provide
important advice to consumers on long-term, low-dose vascular regimens.
    4. Two comments argued that to the limited extent that consumers
are at risk from aspirin use, they are already alerted to this risk by
warnings included in the TFM for OTC internal analgesic/antipyretic
drug products (53 FR 46204). Specifically, the comments asserted that
the proposed warning in Sec. 343.50(c)(1)(v)(B) that states: ``Do not
take this product if you have stomach problems (such as heartburn,
upset stomach, or stomach pain) that persist or recur, or if you have
ulcers or bleeding problems, unless directed by a doctor,'' is
sufficient to warn consumers with stomach problems, whether due to
heavy alcohol use or another condition, about the risk of aspirin.
    The warning in Sec. 343.50(c)(1)(v)(B) is intended to warn
consumers with diagnosed stomach ulcer or symptoms of stomach distress
to avoid the use of aspirin, unless directed to do so by a doctor.
However, as noted in the agency's proposal, acute hemorrhagic gastritis
accounts for 25 percent of major bleeding in heavy, chronic alcohol
users and this condition may be asymptomatic (62 FR 61041 at 61049).
For this reason, the agency finds that the currently proposed stomach
distress warning does not adequately inform individuals who consume
three or more alcoholic drinks every day of their risk.
    5. Two comments stated the belief that the agency's proposed
rulemaking did not evaluate the totality of the data for
nonprescription ibuprofen. One comment argued that ibuprofen, even at
prescription doses, has excellent GI tolerability. In support of its
position, the comment cited data from a variety of different studies
(Ref. 55) assessing the relative GI tolerability of prescription and
OTC ibuprofen. The comments continued that the proposed rule does not
acknowledge data demonstrating the excellent GI tolerabililty of
ibuprofen, even when taken by individuals who regularly consume
alcohol. Cited by the comment were: (1) The results of an endoscopic
study of the effects of alcohol administration on the GI tolerability
of 2,400 milligrams (mg) ibuprofen (twice the maximum daily OTC dose)/
day (d) in healthy subjects (Ref. 56), (2) epidemiological studies
previously evaluated by the agency (Refs. 57, 58, and 59) , and (3) an
assessment of adverse reaction reports for OTC analgesic/antipyretic
drug products containing ibuprofen (both prescription and OTC)
contained in the agency's SRS data base for 1974 to 1993.
    Another comment noted that while OTC drug products containing
ketoprofen and naproxen sodium have been required to include an alcohol
warning in their label, there are no clinical or meaningful
epidemiological data to support the need for a warning on these
products. Based on this lack of data, the comment maintained that an
alcohol warning should not be required for any of the currently
approved OTC NSAID's. To support its position, the comment cited the
lack of reports of injury from the use of these products with alcohol
and few reports of GI bleeding when these products are used as
directed.
    The agency concludes that an alcohol warning is needed for OTC
analgesic/antipyretic drug products containing ibuprofen. Endoscopic
data (Ref. 56) evaluating the GI tolerability in healthy subjects of
prescription doses of ibuprofen (2,400 mg/d for 1 day) with 100-proof
vodka are not adequate because the study did not assess the safety of
ibuprofen use in individuals who consume three or more alcoholic drinks
every day. Carson et al. (Ref. 59) reported that subjects with an
alcohol-related diagnosis who took prescription ibuprofen had no
material increase in bleeding. However, the Committees' evaluated the
study by Carson and concluded that the population studied may not be
generalizable (Ref. 48). The agency evaluated and discussed other
studies (Refs. 57 and 58), which were not convincing as discussed in
the proposed rule (62 FR 61041 at 61050).
    Data concerning the relative GI tolerability of OTC ibuprofen are
not sufficient to support the safety of ibuprofen in heavy alcohol
users. Data from case-control studies which looked at the association
between NSAID use and GI bleeding by Griffin et al. (Ref. 60), Savage
et al. (Ref. 39), and Garcia Rodriguez and Jick (Ref. 61) were
presented and publicly discussed at the October 11 and 12, 1995,
Arthritis Advisory Committee Meeting (Ref. 62). All three of these
studies found the use

[[Page 56795]]

of ibuprofen to be associated with a dose-dependent increase in risk
for GI bleeding. The study by Somerville et al. (Ref. 38), which also
looked at this issue, adds nothing to the discussion. Bradley et al.
(Ref. 63) compared the effectiveness of low-dose ibuprofen (1,200 mg/d)
to high-dose ibuprofen (2,400 mg/d) and high-dose acetaminophen (4,000
mg/d) in patients with osteoarthritis. This study confirmed the dose-
dependent increase in GI symptoms associated with ibuprofen use (1,200
mg/d: 7/62, 11.3 percent; versus 2,400 mg/d: 14/61, 23.0 percent). None
of these studies looked at the associated risks for gastrotoxicity and
ibuprofen in individuals who consume three or more alcoholic drinks
every day. DeArmond et al. (Ref. 64) is an abstract of safety data
generated from 48 clinical trials evaluating OTC naproxen sodium versus
ibuprofen and acetaminophen.
    As previously discussed, study results displaying comparative risks
among these analgesic products are difficult to interpret. However,
because adverse GI effects, including bleeding, occur with all NSAID
ingredients covered by this final rule, the warning is needed for all
of these ingredients.
    In conclusion, as previously discussed in comment 1 of section
II.A. of this document, based on the similar pharmacologic properties
of the nonaspirin NSAID ingredients available OTC as antipyretic/
analgesic drug products, the available scientific data for NSAID
ingredients, alcohol, and the combination of nonaspirin NSAID's and
alcohol, the agency concludes that an alcohol warning is needed for the
safe and effective use of OTC drug products containing ibuprofen,
ketoprofen, or naproxen sodium.
    6. Several comments objected to the agency's requirement for an
alcohol warning on OTC drug products containing carbaspirin calcium,
choline salicylate, magnesium salicylate, and sodium salicylate. These
objections were based on the lack of data supporting the risk of the
use of these products by individuals with a history of heavy alcohol
use. The comments did not include data.
    The agency notes that carbaspirin calcium, choline salicylate,
magnesium salicylate, and sodium salicylate were recognized by the
Panel as having similar adverse effects on the GI tract as aspirin (42
FR 35346 at 35417 through 35422). Similar to aspirin, these adverse
effects include gastric ulcer, exacerbation of peptic ulcer symptoms
(heartburn and dyspepsia), GI hemorrhage and erosive gastritis (Ref.
65). These adverse effects can occur even at low doses. Based on the
recognized individual GI toxicities of carbaspirin calcium, choline
salicylate, magnesium salicylate, sodium salicylate, and alcohol as
well as the Panel's recommendation that these OTC analgesic/antipyretic
drug products bear similar labeling, including a warning against use of
these OTC products in the presence of stomach distress, the agency
concludes that an alcohol warning is necessary for the safe and
effective use of OTC drug products containing these ingredients.

B. Comments on Labeling

    7. Several comments objected to the inclusion of trade names and
brand names in the proposed warning, because it would be confusing to
consumers and would use up valuable label space. Two comments suggested
using the name of the analgesic/antipyretic ingredient. Two comments
suggested using the term ``this product,'' ``the product,'' or
``product'' in place of the trade name or brand name so that the
warning would be generic for all OTC analgesic drug products. One
comment suggested that even these terms (``this product,'' etc.) are
superfluous and unnecessary. A comment contended that for cough/cold
and analgesic combination drug products, the trade name could confuse
consumers because only the analgesic ingredients pertains to the
alcohol warning. Thus, consumers may infer that the warning was
directed at each of the ingredients in a combination drug product.
    The agency agrees that clear labeling is necessary. Inclusion of
the name of the ingredient helps educate and alert the consumer by
making the warning more precise. The agency also believes that the name
of the specific analgesic/antipyretic active ingredient would generally
be more informative than the term ``this product'' or other similar
terms. Therefore, the agency is revising the warning to include the
analgesic/antipyretic ingredient name instead of the brand name.
    8. A number of comments were in disagreement as to the relative
importance of the warnings for acetaminophen, aspirin, and other
NSAID's. A number of comments said the established risks of
acetaminophen use by heavy alcohol users far outweigh the risks of
aspirin use by the same consumers. One comment submitted data from a
comparative risk analysis of aspirin and acetaminophen (Ref. 66). Based
on this analysis, the comment maintained that the number of expected
deaths from acetaminophen toxicity when used for the short-term
treatment of fever and pain is 12 times higher than that expected with
aspirin.
    Several comments complained that despite the much greater risk for
acetaminophen, the proposed alcohol warning conveys the impression that
for heavy alcohol users, the hazards of acetaminophen use and aspirin
(or NSAID) use is essentially the same. Thus, consumers may be led to
believe that they face a comparable risk with either analgesic. The
comments said the proposed warning minimizes the essential messages. In
support of this position, the comment included the results of a
labeling comprehension study (Ref. 67) that it maintained demonstrated
that consumers interpreted the warnings as conveying equivalent risks.
    The agency has reviewed the analysis submitted by one comment (Ref.
66). There were numerous flaws in the baseline assumptions, some of
which were noted by the analysis. The authors assumed that the maximum
recommended daily dose of aspirin is 2,600 mg, but the maximum daily
dose in OTC aspirin labeling is 4,000 mg. For comparative purposes,
alcohol consumption should have been defined in terms of absolute
alcohol. Deaths for GI bleeding and hepatotoxicity were based on
articles from the literature rather than actual death rates in the
United States attributed to either of these conditions. The authors
summarized the data from case reports of hepatotoxicity due to
``therapeutic misadventure'' with acetaminophen to estimate the rate of
hepatotoxicity associated with the drug. Cases of hepatotoxicity
requiring transplantation were discounted in the analysis. It was
assumed that the risk of GI bleeding with aspirin use starts at doses
of 1,500 mg/d and the risk of hepatotoxicity with acetaminophen starts
at about 4,000 mg/d. These data do not support an alcohol warning with
comparative rates of risk.
    The agency has also reviewed the labeling comprehension study (Ref.
67) and has determined that this study did not assess the risk
communication of either warning. In the study, the warnings were not
presented in context, as a consumer would be seeing them. Subjects were
not allowed to perform comparative assessments of the two labels. In
addition, the phrasing of three of the four agree/disagree statements
made ``agree'' responses more likely. Finally, the results were not
framed in terms of alcohol use, a key element in the relevant
population of consumers. However, the study did reveal how few
consumers were aware of these potential toxicities associated with
aspirin or acetaminophen.

[[Page 56796]]

    Although the risk of GI bleeding with aspirin is dose dependent, it
can occur at any dose, depending on other comorbidity factors (Ref.
68). In addition to dosage, hepatotoxicity due to acetaminophen use is
also dependent on factors such as liver glutathione stores, nutritional
state, age, and in some cases, chronicity of usage. Thus, the agency
concludes that the relative degree of risk between aspirin use and
acetaminophen use can not be drawn from this analysis.
    Finally, the agency believes there is some degree of risk for all
OTC analgesic/antipyretic drug products in subjects that are chronic,
heavy alcohol users. This risk is greater than for consumers of these
products who are not chronic, heavy alcohol users. However, the degree
of risk cannot be precisely calculated for the ``at risk'' population
because different risk assessments vary from study to study and may
increase with comorbid factors (Refs. 8 and 31) (62 FR 61041 at 61047).
Nevertheless, it is likely that the degree of risk is not exactly the
same for any two of these drug products or for any two individuals who
consume three or more alcoholic drinks every day. The purpose of the
alcohol warning in this final rule is to alert heavy alcohol users that
serious, specific adverse events can occur with concomitant use of OTC
drug products containing analgesic/ antipyretic ingredients and to seek
advice from their doctor in order to prevent serious adverse events
whenever possible.
    9. Several comments stated that the proposed alcohol warning for
acetaminophen does not describe the severity of potential liver damage.
One comment said the problem is not liver damage, but a significant
risk of dying. A second comment said the term ``liver damage'' is vague
and recommended that the warning include the phrase ``acute liver
failure'' or ``sudden liver failure,'' or the term ``severe liver
damage.''
    In the majority of case reports the agency evaluated,
acetaminophen-induced liver damage in heavy alcohol users did not
result in liver failure or death. Therefore, the agency concludes that
the statement ``Acetaminophen may increase your risk of liver damage''
provides an accurate description to the consumer.
    10. One comment argued that the proposed three-drink threshold is
not appropriate for the acetaminophen warning because it is far below
what is reported in the cases cited by the agency. Therefore, the
comment recommended that language be added to the warning to accurately
describe the chronic heavy alcohol user. However, suggested language
was not provided. One comment said that stating a specific number of
drinks (``3 or more alcoholic beverages daily'') would be better than
the general term ``excessive,'' because the later is very subjective
and each person could define it differently. Another comment suggested
that the warning does not adequately protect women. The comment based
its contention on the U.S. Departments of Agriculture (USDA) and the
Department of Health and Human Services (DHHS) guidelines that
recommend only one drink per day for women (two for men) and evidence
(Refs. 69 and 70) it believes demonstrates that women are more
susceptible to the hepatic effects of alcohol. The comment suggested
that the warning should be gender specific or should be changed to ``2
or more drinks a day'' in order to provide adequate protection for
women.
    The agency acknowledges that the level of alcohol consumption
included in the proposed warning was intended as a general guideline to
help consumers quantify their level of alcohol consumption (62 FR 61041
at 61052). This threshold is based on the recommendations from the
dietary guidelines set by the USDA and DHHS and the standard set by the
AHA. The agency notes that while the dietary guidelines for alcohol
consumption set by USDA and DHHS differentiate between men and women,
the standard set by AHA does not (62 FR 61041 at 61052).
    The agency agrees with the comment that suggested a specific number
of drinks is better than using the term ``excessive'' as a reference
point for consulting a physician because it is more meaningful to many
individuals as a specific number. The warning is intended to aid
consumers in characterizing heavy alcohol consumption, in view of the
inherent variability of individuals in their susceptibility to the
toxic effects of both alcohol and OTC analgesic/antipyretic drug
products.
    11. One comment suggested using the word ``drinks'' instead of
``beverages'' in the proposed warning which states: ``If you drink 3 or
more alcoholic beverages daily * * *.'' The comment said ``drinks'' is
better understood by consumers, and noted that the agency based its
analysis of alcohol consumption on the Dietary Guidelines for
Americans, which defines ``drink.'' The comment said number of
``beverages'' could be perceived as the number of different kinds of
drinks. For instance, a person could perceive four glasses of wine and
four beers as two beverages. Another comment suggested using the term
``every day'' rather than ``daily'' in the warning because ``daily'' is
often misunderstood to mean a single day, whereas ``every day'' is
clearer in communicating a repetitive pattern of drinking behavior.
    The agency agrees with the comments that the terms ``drinks'' and
``every day'' would better convey the intended message to consumers and
has revised the warning to state: ``If you consume 3 or more alcoholic
drinks every day * * *.''
    12. One comment suggested that organ-specific warnings may be more
appropriate for professionals than for consumers. The comment
questioned whether the proposed warning would leave consumers puzzled
as to which product to choose, one that causes liver damage or one that
causes stomach bleeding. Thus an organ-specific warning may discourage
consumers from consulting their physician, believing they can rely on
their ability to self-diagnose liver damage or stomach bleeding. The
comment also refuted the agency's evaluation of data relating to
consumers' perception of label warnings, cited in the proposed rule (62
FR 61041 at 61051), suggesting that a general alcohol warning is less
likely to prompt consumers into appropriate action than an explicit
warning. The comment said the study was not designed to determine
consumer understanding of the warnings tested and that flaws in that
study prevent meaningful conclusions. The comment submitted no data to
support its contention.
    The agency considers organ specific warnings to be more effective
than general warnings. Consumers are better equipped to make a decision
on whether to take a medicine or contact their doctor when they know
the specific risk involved. The agency believes that consumers with a
history of heavy alcohol use need to know the potential risk of OTC
analgesic/antipyretic use. If consumers are not advised of what may
happen (liver damage or stomach bleeding) or what to do (ask their
doctor), the agency believes they would be less likely to take the
warning seriously or to consult their doctor.
    13. Two comments recommended that the proposed warning be formatted
in a style that more closely follows the February 27, 1997 (62 FR
9024), proposed rule on OTC label format. One comment contended that
the use of specific headers for specific warnings are unnecessary and
redundant. Also, specific warnings take up additional space, disrupt
the logical flow of information, and distract from consumer

[[Page 56797]]

comprehension. No data were submitted by the comments.
    The issue of labeling format for specific warnings is broader than
this rulemaking which concerns a single alcohol warning. This issue
will be addressed in a future issue of the Federal Register when the
agency issues a final rule regarding labeling requirements for OTC drug
products.
    14. Several comments recommended reducing the maximum daily dose of
acetaminophen to 2 grams (g) for heavy alcohol users but submitted no
new data. Another comment supported the currently recommended maximum
daily dose of 4 g acetaminophen.
    The agency addressed this issue in the proposed rule (62 FR 61041
at 61044 to 61049) and evaluated a placebo-controlled, double-blind,
randomized study of various dosages of acetaminophen in alcoholics
(Ref. 71). The agency concludes that there is not sufficient evidence
to recommend a specific dosage of acetaminophen which is safe and
effective in subjects who use alcohol heavily.
    15. One comment suggested that the acetaminophen labeling should
warn against the use of more than one acetaminophen-containing product
at a time. The comment also recommended that, because of overdose risk
and risk of liver injury, acetaminophen preparations intended only for
adults should contain warnings against use in children, and pediatric
formulations should convey the need to follow instructions very
carefully. The comment also noted that the warning does not address the
effects of fasting on acetaminophen toxicity.
    The issues raised by the comment are outside of this rulemaking
which specifically addresses the need for an alcohol warning. However,
the issues raised by the comment will be addressed in the final rule
for OTC internal analgesic, antipyretic, and antirheumatic drug
products in a future issue of the Federal Register.
    16. One comment supported the agency's proposal and suggested that
the warning should be put on the leaflet inside the package.
    Information required to appear on the labeling by or under
authority of section 502(c) of the Federal Food, Drug, and Cosmetic Act
(21 U.S.C. 352(c)) must be placed conspicuously so as to be read and
understood by the consumer under customary conditions of purchase and
use. Manufacturers may also include package inserts containing the
required information, but such inserts are not required.

C. Comments on Product Exemptions

    17. One comment maintained that enteric-coated products provide
additional safety for aspirin users and urged FDA to recognize the
documented health and safety benefits of enteric coatings on aspirin.
The comment said that the enteric-coating minimizes gastric irritation
because the enteric-coating delays dissolution of aspirin in the acidic
environment of the gastric lumen. The comment further argued that this
delayed absorption reduces the intracellular accumulation of aspirin in
the gastric mucosa that can lead to cellular injury. In support of this
position, the comment included data from published clinical research
(Refs. 72, 73, and 74) and cited references (Refs. 75 and 76) to
demonstrate the safety of enteric-coated aspirin. Based on these
arguments, the comment suggested the agency take one of the following
actions: (1) Exempt enteric-coated aspirin from the proposed warning,
(2) defer action on a warning for this dosage form until the agency can
gather data that would challenge the documented benefits of enteric-
coated aspirin, or (3) require a separate warning for enteric-coated
products.
     The agency disagrees with the comment. The data provided by the
comments do not demonstrate the safety of enteric-coated dosage forms
of aspirin in consumers with a history of heavy alcohol use.
Furthermore, as previously discussed, aspirin's adverse GI effects are
due both to direct local irritation (the Davenport mechanism) and to
systemic effects which result in prostaglandin inhibition and platelet
dysfunction (Refs. 10 and 13).
    As discussed in comment 1 of section II.A of this document,
enteric-coated dosage forms may exert less direct local effect on the
gastric mucosa, but they are associated with the same risks (and
benefits) of other systemically absorbed aspirin products (Refs. 4 and
51). J. P. Kelly et al. (Ref. 77) examined 550 cases of UGI bleeding
confirmed by endoscopy and 1,202 controls in a multicenter case-control
study. Multiple logistic regression analysis demonstrated a similar
relative risk for plain, enteric-coated, and buffered aspirin at high
(RR: 5.8-7.0) and low (RR: 2.6-3.1) doses. C. A. Silagy et al. (Ref.
78) examined the adverse effects of low-dose enteric-coated aspirin
(100 mg/d) in 400 subjects 70 years or older for 12 months in a double-
blind, randomized, placebo-controlled trial. Clinically evident GI
bleeding occurred in the enteric-coated aspirin treated group but not
in the controls. Clinically evident bleeding from any site and
decreased hemoglobin levels were significantly greater (p<0.05) in the
aspirin-treated group than in the control group. In summary, clinical
trials demonstrate UGI bleeding in patients who also take enteric-
coated aspirin products. Therefore, the agency will require an alcohol
warning for these products.
    18. One comment requested that antacid and aspirin combination
products (highly buffered aspirin in solution) that produce sodium
acetylsalicylate, sodium citrate, and carbon dioxide when added to
water prior to ingestion, not bear an alcohol warning. In support of
this request, the comment submitted data documenting the chemical
characteristics and safety profile distinguishing these products from
plain aspirin. These data were previously reviewed by the Panel (42 FR
35346 at 35417) and are not resummerized in this document.
    The agency disagrees with the comment. The Panel believed there is
no valid clinical evidence to support the claim that highly buffered
aspirin for solution has significantly less potential to induce major
GI hemorrhage than other dosage forms of aspirin (42 FR 35346 at
35471). The agency concurred in comment 31 of the proposed rule for OTC
internal analgesics drug products that the direct toxic effects from
the Davenport mechanism may be reduced, but not eliminated, in highly
buffered aspirin-for-solution products (53 FR 46204 at 46220). In
addition, the indirect effects on systemic prostaglandin inhibition
still play an important role in the toxicity of such products.
Therefore, the agency will require an alcohol warning for these
products.
    19. One comment contended that OTC analgesic/antipyretic drug
products differ in their benefits and potential for injury, and that
any proposal to change the current labeling on such products should be
on a product-by-product basis. The comment argued that alcohol warnings
are not appropriate for products intended for relief of mild to
moderate symptoms associated with menstrual periods in teenagers, or
for OTC highly buffered aspirin solution products indicated for
overindulgence of food and drink.
    The agency disagrees that these products should be exempt from the
alcohol warnings. In comment 18 in section II.C of this document, the
agency discusses the need for an alcohol warning for OTC highly
buffered aspirin solution products. Concerning the need for warnings on
products intended for relief of mild to moderate symptoms associated
with menstrual periods in

[[Page 56798]]

teenagers, this population is not immune to heavy alcohol use as up to
32 percent of high school students have reported heavy drinking (Ref.
79).

D. Comments on Implementation

    20. A number of comments objected to the agency's proposed 6-month
implementation date for the final rule because of the potential
economic impact of the rule based on that timeframe. One comment
requested flexibility in considering the appropriate implementation
period for all OTC analgesic/antipyretic drug products or, at minimum,
for cough-cold products containing these ingredients. The comment
contended that the seasonal nature of cough-cold products requires
large inventory stockpiles and shipments prior to the cough-cold
season. Therefore, depending on the time of year that the rule becomes
final, significant inventory may need to be destroyed if products are
not shipped with required labeling by the effective date. The comment
stated that industry estimates indicate that the average time to
redesign and produce new labeling is 9.25 months. Therefore, it would
be impossible to comply with the proposed 6-month implementation
period. Trying to force these changes more quickly could lead to
labeling errors, resulting in consumer confusion, potential recalls,
and unavailability of some products in the marketplace.
    Although the agency has suggested stick-on labeling as a means to
comply with the 6-month implementation date, one comment believed that
this would not be practical or cost-effective for most combination
cough-cold products. This comment further argued that current warnings
dictated by monographs expend most of the available space on containers
and cartons, leaving insufficient room for placement of a sticker
containing the additional warnings.
    Several comments urged the agency to coordinate the implementation
of the alcohol warning with other labeling proposals impacting these
products. One comment requested that the agency make the rule effective
no sooner than the effective date of the final rule for a standardized
OTC labeling format (62 FR 9024). The comment noted that the agency
expects that the standardized labeling final rule will result in major
format and content changes to current OTC product labeling. If the
final rule for the alcohol warning is effective prior to the
standardized format final rule, manufacturers will incur significant
labeling costs for each of these rules separately. Another comment
requested that FDA extend the implementation date to 12 months.
    One comment stated that 8 months had already been expended to
complete the addition of the voluntary warning on its acetaminophen
products. The comment contended that 14 additional months would be
required to implement the alcohol warning for all products covered by
the final rule. The comment recommended that an effective date of 24
months be established for implementation of the final rule for affected
products that have not been updated to include the voluntary warning
suggested in the proposed rule, and 36 months for products that already
comply with the voluntary warning.
    Although the final rule will have an economic impact on some
manufacturers, the agency believes that the potential benefits of the
rule, including reduced risk of adverse effects, override any economic
concerns (see section III.C of this document). In an attempt to
minimize the economic impact, the agency has allowed for a 6-month
implementation period and the use of supplementary labeling (e.g.,
stick-on labels) to comply with the final rule. Further, manufacturers
that voluntarily included in their labeling the exact warning in the
agency's proposed rule will be permitted to exhaust their inventory of
labels. The agency believes that these measures will help reduce
labeling costs that manufacturers will incur to make the required
labeling changes. The agency concludes that a 6-month implementation
period for the required warning will ensure that consumers have the
most recent information for the safe and effective use of OTC
analgesic/antipyretic drug products.

III. Analysis of Impacts

    FDA has examined the impacts of this final rule under Executive
Order 12866 and the Regulatory Flexibility Act (5 U.S.C. 601-612).
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). Under the Regulatory
Flexibility Act, if a rule has a significant economic impact on a
substantial number of small entities, an agency must analyze regulatory
options that would minimize any significant impact of the rule on small
entities.
    Title II of the Unfunded Mandates Reform Act (2 U.S.C. 1501 et
seq.) requires that agencies prepare a written statement and economic
analysis before proposing any rule that may result in an expenditure by
State, local, and tribal governments, in the aggregate, or by the
private sector of $100 million (adjusted annually for inflation) in any
1 year.
    The agency believes that this rule is consistent with the
principles set out in the Executive Order and in these two statutes.
The purpose of this rule is to add warning statements to the labeling
of OTC drug products labeled for adult use that contain internal
analgesic/antipyretic active ingredients. The added statements warn of
the increased risk of adverse effects from the use of OTC analgesic/
antipyretic drug products by individuals who consume three or more
alcoholic drinks every day. This rule is intended to reduce the number
of specific adverse events associated with the use of these products by
such individuals.

A. Benefits

    As described earlier in this document, FDA finds that individuals
who routinely drink alcohol heavily (three or more drinks every day)
should be specifically warned of risk associated with their use of OTC
analgesic/antipyretic drug products. For example, both aspirin and
other NSAID's carry a dose-related risk of GI bleeding. Alcoholics are
also known to be at increased risk of liver damage and UGI bleeding.
However, because UGI bleeding and liver damage are not unexpected in
alcohol users, medical personnel may not routinely investigate the use
of OTC drug products by patients presenting with these problems.
Recently, in a number of cases, use of acetaminophen was found to be
associated with pathognomonic hepatotoxic changes among heavy alcohol
users and to be a contributing factor in their hospitalization. Many of
these patients required an extended hospital stay.
    FDA cannot quantify the expected benefits of this rule, because it
lacks the data to conduct a quantitative risk assessment. The agency
notes, however, that an estimated 11 million Americans, or about 5.5
percent of the U.S. population age 12 and older, are heavy drinkers
and, therefore, at risk (Ref. 80). Because alcohol warnings on OTC
analgesic/antipyretic drug products could reduce the number of
hospitalizations of heavy alcohol users for hepatic damage and UGI
bleeding, the potential benefits of the rule are substantial. For
example, the cost of a 7-day hospital stay (the average length of stay
in 1994 for an alcohol related discharge) is about $10,000 (Ref. 81).
(Length of stay was calculated as

[[Page 56799]]

weighted average of alcohol first-listed hospital discharges. Cost of
stay was estimated from the 1987 National Medical Expenditure Survey;
cost was converted to 1995 dollars using the CPI-U (consumer price
index--urban areas) for medical services.) If, among the 11 million
consumers potentially at risk, this rule prevented even 500 hospital
visits annually, the present value of the avoided costs would be about
$75 million. (This assumes a 7 percent discount rate and an infinite
time horizon.)

B. Costs

     OTC drug products containing internal analgesic/antipyretic active
ingredients, labeled for adult use, will require new labeling to
incorporate the warning statements. The agency's Drug Listing System
identifies 5,000 to 6,000 OTC analgesic/antipyretic drug products.
Assuming an average of 3 stock keeping units (SKU's)/product, up to
18,000 SKU's will require the alcohol warnings. In its analysis of the
proposed rule, FDA estimated the cost of redesigning a label at from
$2,000 to $3,000/SKU. No industry comment questioned this estimate.
Nevertheless, FDA now believes that the lower end of that range is more
likely, because the added warning requires only a straight-forward text
change without significant graphics redesign. Alternatively, a private-
label manufacturer estimated that the shorter implementation period
would add about $700/SKU. On the assumption that lost inventory cost
for branded SKU's will be twice as high, or $1,400, and that the market
share of branded and private label SKU's is 70 and 30 percent,
respectively, the added cost will amount to about $900. Thus, FDA
projects the total cost of the new warnings at about $3,000/SKU.
Consequently, the estimated one-time cost of this rule is about $54
million. The actual cost may be lower, because the agency is allowing
supplementary labeling (e.g., stick-on labeling), which could reduce
inventory losses.

C. Small Business Impacts

    The agency estimates that fewer than 75 OTC drug manufacturers will
incur costs. FDA does not have data on the size distribution of these
affected firms, but an analysis of an IMS America, Ltd. listing of OTC
drug manufacturers indicates that approximately 70 percent of all
identified OTC drug manufacturers employ fewer than 750 employees,
which is the Small Business Administration's definition of a small
pharmaceutical firm. Consequently, the agency finds that this rule may
have a significant impact on some OTC drug manufacturers, including
smaller firms and manufacturers of private label products. The effect
on individual firms will vary with the number of the firm's SKU's that
require relabeling and the size and cost of the firm's labeling
inventory. Most small firms will not incur significant regulatory costs
because they manufacture few affected SKU's and use less expensive
labeling stock. On the other hand, smaller firms tend to keep
relatively larger labeling inventories because of the volume price
discounts offered by printers. These firms could experience relatively
higher costs for lost inventories.
    This rule will not require any new reporting or recordkeeping
activities. Therefore, no additional professional skills are needed. No
small entities commented on the impact of the proposed rule or
suggested alternatives that would reduce the economic impact on their
establishments.

D. Alternatives

    The agency considered but rejected several less costly regulatory
alternatives, because they would not provide adequate health and safety
benefits. First, the agency considered extending the implementation
period from 6 months to 1 year. This alternative would have saved an
estimated $18 million due to smaller labeling inventory losses.
Nevertheless, as stated in section II.D of this document, in comment
20, the required warnings are necessary to alert consumers to the
potential for serious health outcomes. As the warnings provide
consumers with the critical information needed for making informed
decisions, the longer implementation phase-in would increase the period
over which consumers may make inappropriate choices. The agency
concluded that the reduced labeling cost associated with the longer
phase-in would not justify the increased risk to the public health that
would occur over the additional 6-month period.
    The agency then considered permitting a 1-year implementation
period for those products already labeled with less specific alcohol
warnings. This alternative also was rejected, based on the agency's
determination that most current warnings are inadequate, because they
fail to address the specific nature of the adverse consequence.

E. Conclusion

    The above cost estimates demonstrate that this rule is not
economically significant under Executive Order 12866. As discussed
previously, the agency concludes that this rule is the least burdensome
alternative that meets the agency objective of providing the public
with important health and safety information in a timely manner. As
this rule may have a significant impact on a substantial number of
small entities, this analysis, together with other relevant sections of
this document, serve as the agency's regulatory flexibility analysis,
as required under the Regulatory Flexibility Act. Finally, the Unfunded
Mandates Reform Act does not require a cost-benefit analysis of this
rule, because the rule will not result in an expenditure by State,
local, or tribal governments, in the aggregate, or by the private
sector of $100 million in any 1 year.

IV. Paperwork Reduction Act of 1995

    FDA concludes that the warning statement set forth in this document
is not subject to review by the Office of Management and Budget because
it does not constitute a ``collection of information'' under the
Paperwork Reduction Act of 1995 (44 U.S.C. 3501 et seq.) Rather, the
required warning statement is a ``public disclosure of information
originally supplied by the Federal government to the recipient for the
purpose of disclosure to the public'' (5 CFR 1320.3(c)(2)).

V. Environmental Impact

    The agency has determined under 21 CFR 25.30(h) that this action is
of a type that does not individually or cumulatively have a significant
effect on the human environment. Therefore, neither an environmental
assessment nor an environmental impact statement is required.

VI. References

    The following references have been placed on display in the Dockets
Management Branch (address above) and may be seen by interested persons
between 9 a.m. and 4 p.m., Monday through Friday.
    1. Holvoet, J. et al., ``Relation of Upper Gastrointestinal
Bleeding to Nonsteroidal Anti-inflammatory Drugs and Aspirin: a Case
Control Study,'' Gut, 32:730-734, 1991.
    2. Coggon, D., M. Langman, and D. Spiegelhalter, ``Aspirin,
Paracetamol and Haematemesis and Melaena,'' Gut, 23:340-344, 1982.
    3. Bartle, W., A. Gupta, and J. Lazor, ``Nonsteroidal Anti-
inflammatory Drugs and Gastrointestinal Bleeding: a Case-Control
Study,'' Archives of Internal Medicine, 146:2365-2367, 1986.
    4. Soll, A., ``Pathogenesis of Peptic Ulcer and Implications for
Therapy,'' New England Journal of Medicine, 322:909-916, 1990.
    5. Cohen, M. et al., ``Aspirin-Induced Human Antral Injury Is
Reduced by Vodka

[[Page 56800]]

Pretreatment,'' Digestive Diseases and Sciences, 33:513-517, 1988.
    6. Schubert, T. et al., ``Ulcer Risk Factors: Interactions
Between Helicobacter Pylori Infection, Nonsteroidal Use, and Age,''
American Journal of Medicine, 94:413-418, 1993.
    7. Jensen, D. et al., ``Effects of Aspirin and Alcohol Exposure
on Outcomes of Severe Upper Gastrointestinal Hemorrhage,'' draft of
an unpublished paper included in Comment No. C12, Docket No. 77N-
094W, Dockets Management Branch.
    8. Lanas, A. et al., ``Evidence of Aspirin Use in Both Upper and
Lower Gastrointestinal Perforation,'' Gastroenterology, 112:683-689,
1997.
    9. Gilman, A. G. et al., editors, The Pharmacological Basis of
Therapeutics, 8th ed., McGraw-Hill, New York, pp. 643 and 664-668,
1990.
    10. Dukes, M. N. G., editor, Meyler's Side Effects of Drugs,
12th ed., Elsevier, Amsterdam, pp. 201-206, 1992.
    11. Gennaro, A. R. et al., editors, Remington's Pharmaceutical
Sciences, 18th ed., Mack Publishing Co., Easton, PA, pp. 1112, 1117,
and 1118, 1990.
    12. Van Tyle, K. W., ``Internal Analgesic Products,'' in
Handbook of Nonprescription Drugs, 10th ed., American Pharmaceutical
Association, Washington, pp. 59 and 62, 1993.
    13. Schafer, A. I., ``Effects of Nonsteroidal Antiinflammatory
Drugs on Platelet Function and Systemic Hemostasis,'' Journal of
Clinical Pharmacology, 35:209-219, 1995.
    14. Deykin, D., et al., ``Ethanol Potentiation of Aspirin-
Induced Prolongation of the Bleeding Time,'' New England Journal of
Medicine, 306:852-54, 1982.
    15. Schukit, A., ``Alcoholism and Drug Dependency,'' in
Harrison's Principles of Internal Medicine, 13th ed., edited by K.
J. Isselbacher et al., McGraw-Hill Health Professions Division, New
York, pp. 2421-2422, 1994.
    16. Krome, R., ``Peptic Ulcer Disease,'' in Emergency Medicine A
Comprehensive Guide, 2nd ed., edited by J. Isselbacher, R. Krome,
and E. Ruiz, McGraw-Hill Book Co., New York, p. 310, 1988.
    17. Venho, V., ``Toxicants in the Gastrointestinal Tract,'' in
Gastrointestinal Toxicology, edited by K. Rozman and O. Hannien,
Elsevier, Amsterdam, p. 386, 1986.
    18. Lichtenstein, D., ``Approach to Patient with Acute
Gastrointestinal Hemorrhage,'' in Gastrointestinal Emergencies,
edited by M. Taylor, Williams and Wilkins, Baltimore, p. 105, 1992.
    19. Lanza, F., G. Royer, and R. Nelson, ``Endoscopic Evaluation
of the Effects of Aspirin, Buffered Aspirin, and Enteric-Coated
Aspirin on Gastric and Duodenal Mucosa,'' The New England Journal of
Medicine, 303:136-138, 1980.
    20. Cello, J. P., ``Figure 28 and 29, color plate IV,'' in
Gastrointestinal Disease Pathophysiology/Diagnosis/Management,
edited by M. H. Sleisenger and J. S. Fortran, W. B. Saunders,
Philadelphia, 1993.
    21. Naschitz, J. E. et al., ``Overt Gastrointestinal Bleeding in
the Course of Chronic Low-Dose Aspirin Administration for Secondary
Prevention of Arterial Occlusive Disease,'' American Journal of
Gastroenterology, 85:408-411, 1990.
    22. Weil, J. et al., ``Prophylactic Aspirin and Risk of Peptic
Ulcer Bleeding,'' British Medical Journal, 310:827-830, 1995.
    23. Beaumont, W., Experiments and Observations on the Gastric
Juice and the Physiology of Digestion, Reprinted for XIIIth
International Physiology Congress, Boston, 1929.
    24. Wolf, S. and H. G. Wolff, Human Gastric Function An
Experimental Study of a Man and His Stomach, Oxford University
Press, London.
    25. Palmer E. D., ``Gastritis: A Reevaluation,'' Medicine,
33:199, 1954.
    26. Tarnawski, A. et al., ``Alcohol Injury to the Normal Gastric
Mucosa: Endoscopic, Histologic, and Functional Assessment,''
Clinical Investigations in Medicine, 10:259-263, 1987.
    27. Bockus Gastroenterology, edited by W. S. Haubrich and F.
Schaffner, W. B. Saunders Co., Philadelphia, pp. 664-665, 1994.
    28. Weinstein, W. M. ``Gastritis and Gastropathies,'' in
Gastrointestinal Disease Pathophysiology/Diagnosis/Management,
edited by M. H. Sleisenger and J. S. Fortran, W. B. Saunders,
Philadelphia, p. 547, 1993.
    29. U.S. Department of Health and Human Services, PHS, National
Institutes of Health, NIH Publication No. 94:1447, p. 783, 1994.
    30. Aarons, D. et al., ``Comparative Effects of Alcohol Alone or
in Combination with Aspirin or Acetaminophen on Human Gastric
Mucosa,'' (abs.), Gastrointestinal Endoscopy, 28:124, 1982.
    31. Peura, D. A., et al., ``The American College of
Gastroenterology Bleeding Registry: Preliminary Findings,'' American
Journal of Gastroenterology, 92:924-928, 1997.
    32. UK-TIA Study Group, ``United Kingdom Transient Ischaemic
Attack (UK-TIA) Aspirin Trial: Interim Results,'' British Medical
Journal, 296:316-320, 1988.
    33. The SALT Collaborative Group, ``Swedish Aspirin Low-dose
Trial (SALT) of 75 mg Aspirin as Secondary Prophylaxis after
Cerebrovascular Ischaemic Events,'' Lancet, 338:1345-1349, 1991.
    34. Steering Committee of the Physicians' Health Study Research
Group, ``Final Report on the Aspirin Component of the Ongoing
Physicians' Health Study,'' New England Journal of Medicine,
321:129-135, 1989.
    35. Sievert, W. et al., ``Low-Dose Antacids and Nonsteroidal
Anti-inflammatory Drug-Induced Gastropathy in Humans,'' Journal of
Clinical Gastroenterology, 13:S145-48, 1991.
    36. Laporte, J. et al., ``Upper Gastrointestinal Bleeding in
Relation to Previous Use of Analgesics and Non-steroidal Anti-
inflammatory drugs,'' Lancet, 337:85-89, 1991.
    37. Griffin, M.R., et al., ``Nonsteroidal Anti-Inflammatory Drug
Use and Death from Peptic Ulcer in Elderly Persons,'' Annals of
Internal Medicine, 109:359-363, 1988.
    38. Somerville, K., G. Faulkner, and M. Langman, ``Nonsteroidal
Anti-inflammatory Drugs and Bleeding Peptic Ulcer,'' Lancet, 1:462-
464, 1986.
    39. Savage, R. et al., ``Variation in the Risk of Peptic Ulcer
Complications with Nonsteroidal Anti-inflammatory Drug Therapy,''
Arthritis Rheumatology, 36:84-90, 1993.
    39a. Nobili, A., et al., ``Non-steroidal Anti-inflammatory Drugs
and Upper Gastrointestinal Bleeding, a Post-marketing Surveillance
Case-control Study,'' Pharmacoepidemiology and Drug Safety, 1:65-72,
1992.
    40. Guess, H. A. et al., ``Fatal Upper Gastrointestinal
Hemorrhage or Perforation Among Users and Nonusers of Nonsteroidal
Anti-inflammatory Drugs in Saskatchewan, Canada 1983,'' Journal of
Clinical Epidemiology, 47:35-45, 1988.
    41. Bollini, P. et al., ``The Impact of Research Quality and
Study Design on Epidemiologic Estimates of the Effect of
Nonsteroidal Anti-inflammatory Drugs on Upper Gastrointestinal Tract
Disease,'' Archives of Internal Medicine, 152:1289-95, 1992.
    42. Gabriel, S. E. et al., ``Risk for Serious Gastrointestinal
Complications Related to Use of Nonsteroidal Anti-inflammatory
Drugs: A Meta-analysis,'' Annals of Internal Medicine, 115:787-96,
1991.
    43. Report from FDA Spontaneous Reporting System Database.
    44. Levy, M. et al., ``Major Upper Gastrointestinal Tract
Bleeding: Relation to the Use of Aspirin and Other Nonnarcotic
Analgesics,'' Archives of Internal Medicine, 148:281-285, 1988.
    45. DeSchepper, P. J. et al., ``Diflunisal Versus Aspirin: a
Comparative Study of Their Effect on Faecal Blood Loss in the
Presence and Absence of Alcohol,'' Current Medical Research and
Opinion, 5:520-524, 1978.
    46. Lee et al., ``Hemorrhagic Erosive Gastritis,'' American
Journal of Gastroenterology, 3:201-208, 1975.
    47. Bloom, B. S., ``Risk and Cost of Gastrointestinal Side
Effects Associated With Nonsteroidal Anti-inflammatory Drugs,''
Archives of Internal Medicine, 149:1019-1022, 1989.
    48. Summary of the minutes of the September 8, 1993, joint
meeting of the FDA Nonprescription Drugs and Arthritis Drugs
Advisory Committees, in OTC Vol. 03AWNPR, Docket No 77N-094W,
Dockets Management Branch, Food and Drug Administration.
    49. Summary minutes of the June 28 and 29, 1993 meeting of the
FDA Nonprescription Drugs Advisory Committee, in OTC Vol. 03AWNPR,
Docket No. 77N-094W, Dockets Management Branch.
    50. Summary minutes of the June 1, 1993, meeting of the FDA
Arthritis and Nonprescription Drugs Advisory Committees.
    51. Soll, A. H., ``Nonsteroidal Anti-inflammatory Drugs and
Peptic Ulcer Disease,'' Annals of Internal Medicine, 114:307-319,
1991.
    52. Wallace, J. L., ``Nonsteroidal Anti-inflammatory Drugs and
Gastroenteropathy: The Second Hundred Years,'' Gastroenterology,
112:1,000-1,006, 1997.
    53. Comments No. C193, C206, and MM19, Docket No. 77N-0094,
Dockets Management Branch.
    54. Comments No. C198, C199, C200, Docket No. 77N-0094, Dockets
Management Branch.
    55. Comment No. C11, Docket No. 77N-094A, Dockets Management
Branch.

[[Page 56801]]

    56. Lanza, F. et al., ``Ethanol, Aspirin, Ibuprofen, and the
Gastroduodenal Mucosa: An Endoscopic Assessment,'' American Journal
of Gastroenterology, 80:767-769, 1985.
    57. Kelly, J., D. Kaufman, and S. Shapiro, ``The Risk of Major
Upper Gastrointestinal Bleeding among Users of Aspirin, Ibuprofen,
Naproxen, at Various Levels of Alcohol Consumption,'' draft of an
unpublished paper included in Comment No. C188, Docket No. 77N-0094,
Dockets Management Branch.
    58. Henry, D., A. Dobson, and C. Turner, ``Variability in the
Risk of Major Gastrointestinal Complications from Nonaspirin
Nonsteroidal Anti-inflammatory Drugs,'' Gastroenterology, 105:1078-
1088, 1993.
    59. Carson, J. et al., ``The Relative Gastrointestinal Toxicity
of the Nonsteroidal Anti-inflammatory Drugs,'' Archives of Internal
Medicine, 147:1054-1059, 1987.
    60. Griffin, M. et al., ``Nonsteroidal Anti-inflammatory Drug
Use and Increased Risk for Peptic Ulcer Disease in Elderly
Persons,'' Annals of Internal Medicine, 114:257-263,1991.
    61. Garcia Rodriguez, L., and H. Jick, ``Risk of Upper
Gastrointestinal Bleeding and Perforation Associated with Individual
Nonsteroidal Anti-inflammatory Drugs,'' Lancet, 343:769-772, 1994.
    62. Transcripts of the October 11 and 12, 1995, meeting of the
FDA Arthritis Drugs Advisory Committee, in OTC Vol. 03AWNFR, Docket
No. 77N-094A, Dockets Management Branch.
    63. Bradley, J. et al, ``Comparison of an Anti-inflammatory Dose
of Ibuprofen, an Analgesic Dose of Ibuprofen, and Acetaminophen in
the Treatment of Patients with Osteoarthritis of the Knee,'' New
England Journal of Medicine, 325:87-91, 1991.
    64. DeArmond, B. et al, ``Safety Profile of Nonprescription
Naproxen Sodium,'' Clinical Pharmacology and Therapy, 57:136, 1995.
    65. Insel, P. A., ``Analgesic-Antipyretic and Anti-inflammatory
Agents and Drugs Employed in the Treatment of Gout: The Salicylates;
Gastrointestinal Effects,'' in ``Goodman and Gilman's The
Pharmacological Basis of Therapeutics,'' 9th ed., edited by J. G.
Hardman, A. G. Gilman, and L. E. Limbird, McGraw-Hill, New York, p.
626, 1996.
    66. Cher, D., R. Morgan, and S. Hyg, ``Comparison of the Risks
of Aspirin and Acetaminophen in Alcohol Users: Application of
Decision Analysis'' draft of an unpublished paper in Comment No.
C12, Docket No. 77N-094A, Dockets Management Branch.
    67. ``A Study of Two Pain Reliever Warning Statements'' by
Guideline Research Corp., draft of an unpublished paper in Comment
No. C12, Docket No. 77N-094A, Dockets Management Branch.
    68. Willett, L., J. Carson, and B. Strom, ``Epidemiology of
Gastrointestinal Damage Associated with Nonsteroidal Anti-
inflammatory Drugs,'' Drug Safety, 10:170-181, 1994.
    69. Norton, R., ``Alcohol Consumption and the Risk of Alcohol
Related Cirrhosis in Women,'' British Medical Journal, 295:80-82,
1987.
    70. Mezey, E. et al., ``Alcohol and Dietary Intake in the
Development of Chronic Pancreatitis and Liver Disease in
Alcoholism,'' American Journal of Clinical Nutrition, 48:148-151,
1988.
    71. Kuffner, E., G. Bogdan, and R. Dart, ``Evaluation of
Hepatotoxicity in Alcoholics from Therapeutic Dosing of
Acetaminophen,'' Journal of Toxicology, 35:561, 1997.
    72. Petroski, D., ``A Comparison of Enteric-Coated Aspirin
Granules with Plain and Buffered Aspirin: A Report of Two Studies,''
The American Journal of Gastroenterology, 81:26-28, 1986.
    73. Lanza, F., G. Royer, and R. Nelson, ``Endoscopic Evaluation
of the Effects of Aspirin, Buffered Aspirin, and Enteric-Coated
Aspirin on Gastric and Duodenal Mucosa,'' The New England Journal of
Medicine, 303:136-138, 1980.
    74. Savon, J. et al., ``Gastrointestinal Blood Loss with Low
Dose (325 mg) Plain and Enteric-Coated Aspirin Administration,'' The
American Journal of Gastroenterology, 90:581-585, 1995.
    75. Petroski, D., ``Endoscopic Comparison of Three Aspirin
Preparations and Placebo,'' Excerpta Medica, pp. 314-320, 1993.
    76. Petroski, D., ``Endoscopic Comparison of Various Aspirin
Preparations-Gastric Mucosal Adaptability to Aspirin Restudied,''
Current Therapeutic Research, 45:945-954, 1989.
    77. Kelly, J. P. et al., ``Risk of Aspirin-Associated Major
Upper-Gastrointestinal Bleeding with Enteric-Coated or Buffered
Product,'' Lancet, 348:1413-1416, 1996.
    78. Silagy, C. A. et al., ``Adverse Effects of Low-dose Aspirin
in a Healthy Elderly Population,'' Clinical Pharmacology and
Therapeutics, 54:84-89, 1993.
    79. National Institutes of Health, Eighth Special Report to the
U.S. Congress on Alcohol and Health from the Secretary of Health and
Human Services, U.S. Department of Health and Human Services,
National Institutes of Health, and National Institute on Alcohol
Abuse and Alcoholism, NIH Publication No. 94-3699, pp. 21-23, 1994.
    80. Substance Abuse and Mental Health Service Administration,
``Preliminary Results from the 1996 National Household Survey on
Drug Abuse;'' published August 1997; <http://www.health.org/pubs/
nhsda/96hhs/httoc.htm>
    81. National Institute on Alcohol Abuse and Alcoholism,
``Surveillance Report #40: Trends in Alcohol-Related Morbidity Among
Short-Stay Community Hospital Discharges, United States, 1979-94'',
published December 1996, ``http://silk/niaaa1/publication/
SR40.pdf''.

List of Subjects in 21 CFR Part 201

    Drugs, Labeling, Reporting and recordkeeping requirements.
    Therefore, under the Federal Food, Drug, and Cosmetic Act and under
authority delegated to the Commissioner of Food and Drugs, 21 CFR part
201 is amended as follows:

PART 201--LABELING

    1. The authority citation for 21 CFR part 201 continues to read as
follows:

    Authority: 21 U.S.C. 321, 331, 351, 352, 353, 355, 358, 360,
360b, 360gg-360ss, 371, 374, 379e; 42 U.S.C. 216, 241, 262, 264.

    2. Section 201.322 is added to subpart G to read as follows:

Sec. 201.322   Over-the-counter drug products containing internal
analgesic/antipyretic active ingredients; required alcohol warning.

    (a) People who regularly consume large quantities of alcohol (three
or more drinks every day) have an increased risk of adverse effects
(possible liver damage or gastrointestinal bleeding). OTC drug products
containing internal analgesic/antipyretic active ingredients may cause
similar adverse effects. FDA concludes that the labeling of OTC drug
products containing internal analgesic/antipyretic active ingredients
should advise consumers with a history of heavy alcohol use to consult
a physician. Accordingly, any OTC drug product, labeled for adult use,
containing any internal analgesic/antipyretic active ingredients
(including, but not limited to, acetaminophen, aspirin, carbaspirin
calcium, choline salicylate, ibuprofen, ketoprofen, magnesium
salicylate, naproxen sodium, and sodium salicylate) alone or in
combination shall bear an alcohol warning statement in its labeling as
follows:
    (1) Acetaminophen. ``Alcohol Warning'' [heading in boldface type]:
``If you consume 3 or more alcoholic drinks every day, ask your doctor
whether you should take acetaminophen or other pain relievers/fever
reducers. Acetaminophen may cause liver damage.''
    (2) Nonsteroidal anti-inflammatory analgesic/antipyretic active
ingredients--including but not limited to aspirin, carbaspirin calcium,
choline salicylate, ibuprofen, ketoprofen, magnesium salicylate,
naproxen sodium, and sodium salicylate. ``Alcohol Warning'' [heading in
boldface type]: ``If you consume 3 or more alcoholic drinks every day,
ask your doctor whether you should take [insert one nonsteroidal anti-
inflammatory analgesic/antipyretic active ingredient] or other pain
relievers/fever reducers. [Insert one nonsteroidal anti-inflammatory
analgesic/antipyretic active ingredient] may cause stomach bleeding.''
    (3) Combinations of acetaminophen with nonsteroidal anti-
inflammatory analgesic/antipyretic active ingredients--including but
not limited to aspirin, carbaspirin calcium, choline

[[Page 56802]]

salicylate, ibuprofen, ketoprofen, magnesium salicylate, naproxen
sodium, and sodium salicylate. ``Alcohol Warning'' [heading in boldface
type]: ``If you consume 3 or more alcoholic drinks every day, ask your
doctor whether you should take [insert acetaminophen and one
nonsteroidal anti-inflammatory analgesic/antipyretic active
ingredient--including, but not limited to aspirin, carbaspirin calcium,
choline salicylate, magnesium salicylate, or sodium salicylate] or
other pain relievers/fever reducers. [Acetaminophen and (insert one
nonsteroidal anti-inflammatory analgesic/antipyretic ingredient--
including, but not limited to aspirin, carbaspirin calcium, choline
salicylate, magnesium salicylate, or sodium salicylate] may cause liver
damage and stomach bleeding.''
    (b)  Requirements to supplement approved application. Holders of
approved applications for OTC drug products that contain internal
analgesic/antipyretic active ingredients that are subject to the
requirements of paragraph (a) of this section must submit supplements
under Sec. 314.70(c) of this chapter to include the required warning in
the product's labeling. Such labeling may be put into use without
advance approval of FDA provided it includes the exact information
included in paragraph (a) of this section.
    (c) Any drug product subject to this section that is not labeled as
required and that is initially introduced or initially delivered for
introduction into interstate commerce after April 23, 1999, is
misbranded under section 502 of the Federal Food, Drug, and Cosmetic
Act (21 U.S.C. 352) and is subject to regulatory action.

    Dated: July 22, 1998.
Michael A. Friedman,
Acting Commissioner of Food and Drugs.
Donna E. Shalala,
Secretary of Health and Human Services.
[FR Doc. 98-28520 Filed 10-21-98; 10:58 am]
BILLING CODE 4160-01-F