Annals of Epidemiology is a peer reviewed, international journal devoted to epidemiologic research and methodological development. The journal
emphasizes the application of epidemiologic methods to issues that affect the distribution and determinants of human illness in diverse contexts. Its primary
focus is on chronic and acute conditions of diverse etiologies and of major importance to clinical medicine, public health, and health care delivery. Annals
encourages the use of epidemiology in a multidisciplinary approach to understanding disease etiology. Review articles, reports from U.S. Federal and
International sources, Editorials, Commentaries, Brief Communications, Letters to the Editor, Book Reviews, and selected papers from major symposia are
also published.
EDITOR-IN-CHIEF
Richard Rothenberg, MD, MPH
ASSOCIATE EDITORS
STATISTICAL CONSULTANT
James G. Gurney, PhD
Irene Hall, PhD
Otto Wong, ScD, FACE
Richard F. Gillum, MD
Michael Goodman, MD, MPH
Michael H. Kutner, PhD
MANAGING EDITOR
Dorothy Fitzmaurice
BOOK REVIEW EDITOR
James G. Gurney, PhD
EDITORIAL BOARD
Elizabeth Barrett-Connor, MD
Felix Gutzwiller, MD, DrPH
Philip C. Nasca, PhD
University of California at San Diego
La Jolla, CA, USA
University of Zurich
Zurich, Switzerland
University of Massachusetts
Amherst, MA, USA
Catherine Hill, PhD
Roberta B. Ness, MD, MPH
Gustav Roussy Institute
Paris, France
University of Pittsburgh
Pittsburgh, PA, USA
Tomio Hirohata, MD, PhD, SDHyg
C. Ineke Neutel, PhD
Nakamura College
Fukuoka, Japan
University of Ottawa
Ottawa, Canada
George Kaplan, PhD
Trevor Orchard, MD
Theodore Colton, ScD
University of Michigan School of Public Health
Ann Arbor, MI, USA
University of Pittsburgh
Pittsburgh, PA, USA
Boston University School of Public Health
Boston, MA, USA
Richard A. Kaslow, MD, MPH
D. C. Rao, PhD
University of Alabama at Birmingham
Birmingham, AL, USA
Washington University School of Medicine
St. Louis, MO, USA
Ulrich Keil, MD, MPH, PhD
Elisheva Simchen, MD, MPH
University of Münster,
Münster, Germany
Hadassah University Hospital
Jerusalem, Israel
Tai Hing Lam, MD
Dimitrios Trichopoulos, MD
University of Hong Kong
Hong Kong, China
Harvard School of Public Health
Boston, MA
Richard F. Gillum, MD
Ruey S. Lin, MD, DrPH
Lars Wilhelmsen, MD
National Center of Health Statistics
Hyattsville, MD, USA
National Taiwan University
Taipei, China
Göteborg University
Gothenburg, Sweden
Eric Boerwinkle, PhD
University of Texas Health Sciences Center
Houston, TX, USA
Patricia A. Buffler, PhD, MPH
University of California School of Public Health
Berkeley, CA, USA
Jeffrey P. Davis, MD
Bureau of Public Health
Madison, WI, USA
Lawrence M. Friedman, MD
National Heart, Lung and Blood Institute
Bethesda, MD, USA
EDITORIAL OFFICE: Emory University School of Medicine, Dept. of Medicine, 69 Jesse Hill Jr. Drive, Atlanta, GA 30303-3219, Editorial office’s e-mail
address: [email protected], Dr. Rothenberg’s e-mail address: [email protected], 404-616-5602 (phone), 404-616-6847 (fax).
VOLUME 17, NUMBER 5S
MAY 2007
CONTENTS
S1 Introduction to Symposium
R. Curtis Ellison, MD
S3
Alcohol and Coronary Heart Disease: Drinking Patterns and Mediators of Effect
Eric B. Rimm, ScD, and Caroline Moats, MS
S8
Alcohol Intake and Noncoronary Cardiovascular Diseases
Kenneth Mukamal, MD
S13
Confounders of the Relation between Type of Alcohol and Cardiovascular Disease
Morten Grønbæk, MD, PhD
S16
Moderate Alcohol Use and Reduced Mortality Risk: Systematic Error in Prospective Studies
and New Hypotheses
Kaye Middleton Fillmore, PhD, Tim Stockwell, PhD, Tanya Chikritzhs, PhD, Alan Bostrom, PhD, and William Kerr, PhD
S24
Mechanism by which Alcohol and Wine Polyphenols Affect Coronary Heart Disease Risk
Francois M. Booyse, PhD, Wensheng Pan, MD, PhD, Hernan E. Grenett, PhD, Dale A. Parks, PhD, Victor M. Darley-Usmar, PhD,
Kelley M. Bradley, BS, and Edlue H. Tabengwa, PhD
S32
Changes in Cardiovascular Risk Factors Associated with Wine Consumption in Intervention
Studies in Humans
Federico Leighton, MD, and Inés Urquiaga, PhD
S37
Panel Discussion I: Does Alcohol Consumption Prevent Cardiovascular Disease?
S40
Moderate Alcohol Consumption and Insulin Sensitivity: Observations and Possible Mechanisms
Henk F.J. Hendriks, PhD
S43
Moderate Alcohol Consumption and Risk of Developing Dementia in the Elderly: The Contribution
of Prospective Studies
Luc Letenneur, PhD
S46
Moderate Alcohol Consumption and Cancer
Klim McPherson, PhD, FFPH, FMedSci
S49
Moderate Drinking, Inflammation, and Liver Disease
Gyongyi Szabo, MD, PhD
S55
Investigating the Association Between Moderate Drinking and Mental Health
Nady El-Guebaly, MD
S63
Alcohol Drinking and Total Mortality Risk
Arthur L. Klatsky, MD, and Natalia Udaltsova, PhD
S68
S71
Panel Discussion II: Net Effects of Moderate Alcohol Consumption on Health
Why We Don’t Know More about the Social Benefits of Moderate Drinking
Dwight B. Heath, PhD
S75
Patterns of Alcohol Consumption and Cardiovascular Risk in Northern Ireland and France
Alun Evans, MD, Pedro Marques-Vidal, MD, PhD, Pierre Ducimetière, PhD, Michele Montaye, MD, Dominique Arveiler, MD,
Annie Bingham, MA, Jean-Bernard Ruidavets, MD, Phillipe Amouyel, MD, PhD, Bernadette Haas, MD, John Yarnell, MD,
Jean Ferrières, MD, MSc, Frederic Fumeron, PhD, Gerald Luc, MD, Frank Kee, MD, and Francois Cambien, MD
S81
Harm, Benefits, and Net Effects on Mortality of Moderate Drinking of Alcohol Among Adults
in Canada in 2002
Jürgen Rehm, PhD, Jayadeep Patra, PhD, and Ben Taylor, MEpi
S87
Alcohol Use and Mental Health in Developing Countries
Vikram Patel, PhD
S93
Panel Discussion III: Moderation, Culture, and Risk
S95
Panel Discussion IV: Implications for Future Research
S98
Communicating Through Government Agencies
Richard Harding, PhD, and Creina S. Stockley, MSc, MBA
S103 Communicating with the Public: Dilemmas of a Chief Medical Officer
Richard Smallwood, MD
S108 The Influence of Print Media on their Readers’ Understanding of the Benefits of Moderate Drinking
Thomas Stuttaford, MD
S110 Panel Discussion V: The Message on Moderate Drinking
S112 Alcohol and Pleasure
Lionel Tiger, PhD
S114 Closing Remarks
R. Curtis Ellison, MD
Annals of Epidemiology is cited in EMBASE/Excerpta Medica, CABS/Current Advances in Ecological & Environmental Sciences, CAB
ABSTRACTS/Nutrition Abstracts and Reviews, Cambridge Scientific Abstracts, Index Medicus/MEDLINE, BIOSIS/Biological Abstracts,
CINAHL® and Cumulative Index to Nursing & Allied Health Literature™, Current Contents/Clinical Medicine, Research Alert, SciSearch.
For Customers in European countries: Personal rate: EUR 326, Institutional rate: EUR 673. For Customers in all countries except Europe and
Japan: Personal rate: USD 302, Institutional rate: USD 625. For customers in Japan: Personal rate: JPY 32,600, Institutional rate: JPY 64,800.
Prices include postage and are subject to change without notice. For additional information contact Elsevier Customer Support Department, P.O.
Box 945, New York, NY 10010, Tel: (212) 633-3730. Toll Free (for customers in North America): 1-888-4ES-INFO, Fax: (212) 633-3680,
E-mail: ([email protected]).
For orders, claims, product inquiries (no manuscript inquiries) please contact the Customer Support Department at the Regional Sales
Office nearest to you:
Orlando: Elsevier, Customer Service Department, 6277 Sea Harbor Drive, Orlando, FL 32887-4800, USA; phone: (+1) (877) 839-7126
[toll free number for US customers], or (+1) (407) 345-4020 [customers outside US]; Fax:(+1) (407) 363-1354; e-mail: [email protected]].
Amsterdam: Elsevier Inc., P.O. Box 211, 1000 AE Amsterdam, The Netherlands [Tel: (⫹31) 20-485-3757. Fax (⫹31) 20-485-3432.
E-mail: [email protected]].
Tokyo: Elsevier Inc., 9-15, Higashi-Azabu 1-chome, Minato-ku, Tokyo 106 Japan [Tel: (⫹81) 3-5561-5033. Fax: ( ⫹ 81) 3-5561-5047.
E-mail: [email protected]].
Singapore: Elsevier Inc., No. 1 Temasek Avenue, #17-01 Millenia Tower. Singapore 039192 [Tel: (⫹65) 434-3727. Fax: (⫹65) 3372230. E-mail: [email protected]].
Rio de Janeiro: Elsevier Inc., Rua Sete de Setembro 111/16 Andar, 20050-002 Centra, Rio de Janeiro - RJ, Brazil [Tel: (⫹55) (21) 509
5340. Fax: (⫹55) (21) 507 1991. E-mail: [email protected]] [Note (Latin America): for orders, claims and help desk information,
please contact the Regional Sales Office in New York as listed above.]
FOR ADVERTISING ORDERS & INQUIRIES: North & South America: Inez Herrero, Elsevier Inc., 360 Park Avenue South, New
York, NY 10010, Tel: 212-633-3122, Fax: 212-633-3820. International: Advertising Department, Elsevier Inc., The Boulevard, Langford
Lane, Kidlington, Oxford OX5 1GB, United Kingdom, Tel: (⫹44) (0) 1865 843565, Fax: (44) (0) 1865 843976 (E-mail:
[email protected]). Japan: Advertising Department, Elsevier Inc. K.K., Tel: (81) (3) 5561 5033, Fax: (81) (3) 5561 5047.
Claims for missing issues can be honored up to three months for domestic addresses and six months for foreign addresses. Duplicate
copies will not be sent to replace ones undelivered through failure to notify Elsevier Inc., of a change in address. Single copy, microform,
and back-volume information available upon request. POSTMASTER: Send address changes to AEP, Elsevier Science Inc., 360 Park Avenue
South, New York, NY 10010. Periodicals postage paid at New York, NY and at additional mailing offices.
© 2007 Elsevier Inc. All rights reserved.
This journal and the individual contributions contained in it are protected under copyright by Elsevier Science Inc., and the following
terms and conditions apply to their use:
Photocopying: Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of
the publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising
or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make
photocopies for non-profit educational classroom use.
Permissions may be sought directly from Elsevier’s Rights Department in Philadelphia, PA, USA: phone (215) 239–3804, fax: (215)
239-3805; email: [email protected]. Requests may also be completed online via the Elsevier homepage (http://www.
elsevier.com/locate/permissions).
In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive,
Danvers, MA 01923, USA; phone: (978) 7508400, fax: (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid
Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P OLP, UK; phone; (⫹44) 171 436 5931; fax: (⫹44) 171 436 3986.
Other countries may have a local reprographic rights agency for payments.
Author Support: Authors should contact Elsevier’s Author Support <[email protected]> for any concerns or questions.
Derivative works: Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation
within their institutions. Permission of the publisher is required for resale or distribution outside the institution.
Electronic Storage or Usage: Permission of the publisher is required to store or use electronically any material contained in this journal, including any article or part of an article. Contact the publisher at the address indicated.
Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by
any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher.
Address permissions requests to: Elsevier Rights & Permissions Department, at the mail, fax and e-mail addresses noted above.
Notice: No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.
Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.
Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality of such product or of the claims made of it by its manufacturer.
This paper meets the requirements of ANSI Standard Z39.48-1992 (Permanence of Paper).
Introduction to Symposium
R. CURTIS ELLISON, MD
For millennia, wine and other alcoholic beverages have
been associated with health and for centuries were a part
of the physician’s pharmacopoeia. During the late 19th
and early 20th centuries, the public focus switched primarily
to the harmful effects of excessive alcohol use and alcohol
abuse, and the role of alcohol in medical therapy ended.
In recent decades, however, there have begun to appear
reports on possible healthful aspects of moderate drinking,
with alcohol intake being especially associated with a reduced risk of coronary heart disease (CHD). There remain
questions about methodology used in observational studies
on alcohol and health, and of possible residual confounding
by lifestyle factors. However, the general consensus, based
on epidemiologic studies, short-term clinical trials of the
effects of alcoholic beverages on cardiovascular disease
risk factors, and basic science studies, is that moderate alcohol consumption is associated with less CHD and possibly
with a lower risk of other diseases of aging.
The American public became aware of potentially beneficial effects of moderate alcohol consumption on the risk of
heart disease in the early 1990s, primarily from a program on
the popular television show 60 Minutes describing the
‘‘French Paradox.’’ The impression from that program was
that the regular consumption of red wine by the French
might be an important factor in explaining the lower rates
of CHD in France than in the United States and Northern
Europe; this led not only to an increase in the sales of red
wine in the United States but has stimulated considerable
research on moderate drinking by scientists around the
world.
Some physicians and public health officials, only too
aware of the terrible personal and social problems associated
with alcohol abuse, have been reluctant to acknowledge
potential beneficial effects of moderate drinking, and others
have questioned the methodology of the epidemiologic
studies. These concerns will be important topics of discussion at this conference.
While positive effects of alcohol on high-density lipoprotein (HDL)-cholesterol have been known for years, an
immense amount of recent research has identified many
other potential mechanisms by which alcoholic beverages
may relate to CHD. Demonstrated have been effects of
From Boston University School of Medicine, Boston, MA.
Address correspondence to: R. Curtis Ellison, Boston University School
of Medicine, 715 Albany St, Boston, MA 02118. Tel.: (617) 638-8080; fax:
(617) 638-8076. E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
alcohol and wine polyphenolic compounds on coagulation,
fibrinolysis, endothelial function, lipid oxidation, glucose
metabolism, inflammation, and ventricular function, many
of which will be discussed at this conference. Also to be discussed will be the adverse effects of abusive drinking on the
liver, the brain, and other organs, as well as potentially
favorable effects of moderate drinking on other diseases of
aging, including stroke, congestive heart failure, diabetes,
metabolic syndrome, dementia, and even total mortality.
Further topics will include the social and cultural aspects
of alcohol consumption.
The goal of this symposium is to foster free communication among the participating scientists, many of whom
have made important contributions to this field over
many decades. It is a closed meeting, and the emphasis
will be on uninhibited dialogue, whether it be ‘‘politically
correct’’ or not. We are encouraging all of the participants
to speak freely, asking each to indicate clearly whether he
or she agrees, or does not agree, with what other speakers
may saydand the reasons why.
We are hoping that the end result of the meeting will be
up-to-date, scientifically sound, and balanced information
on moderate drinking, data that should be useful to other
scientists working in the field, to physicians and other
health-care providers, and to those who develop guidelines
for drinking and set policy related to alcohol use. We trust
that the published proceedings of this conference will
indeed be as close as possible to the ‘‘truth,’’ as we understand it now, about the health risk and benefits of the moderate consumption of alcoholic beverages.
STATEMENT ON CONFLICT OF INTEREST
The conference upon which the present publication is based
was sponsored by the Institute on Lifestyle & Health, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, and by the
International Center for Alcohol Policies (ICAP) in Washington, DC. ICAP provided funds for the conference,
including travel expenses and a modest honorarium for scientists to participate and to furnish manuscripts based on
their presentations.
While ICAP is an independent organization, it does
receive its major funding from companies in the beverage
alcohol business. Because of this, we were careful to abide
fully by the Dublin Principles (1), a set of guidelines for
research principles relating to alcohol. The Dublin Principles state that while ‘‘the beverage alcohol industry should
support independent scientific research which contributes
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.001
S2
Ellison
INTRODUCTION TO SYMPOSIUM
Selected Abbreviations
CHD Z coronary heart disease
HDL Z high-density lipoprotein
to a better understanding of the use, misuse, effects and
properties of alcohol and the relationships among alcohol,
health, and society,’’ the ‘‘academic and scientific communities should adhere to the highest professional, scientific, and
ethical standards in conduction and reporting on alcohol research, whatever the source of funding for such research.’’
Further, the Principles state specifically that while ‘‘scientific researchers should acknowledge the sources(s) of funding, (they) should be free to disseminate and publish the
results of their work.’’
AEP Vol. 17, No. 5S
May 2007: S1–S2
In the present publication, the opinions expressed in
each paper and discussion summary are those of the author(s) and are not necessarily those of the Guest Editor of
the supplement, the Annals of Epidemiology, the Institute
on Lifestyle & Health, or ICAP. Final decisions regarding
the editing of each manuscript rested with the Guest Editor
of this supplement and the Editor-in-Chief of Annals of Epidemiology and were uninfluenced by the source of funding.
REFERENCE
1. The Dublin Principles of Cooperation among the Beverage Alcohol Industry, Governments, Scientific Researchers, and the Public Health Community. ICAP, Dublin, Ireland, 1997. Available at http://www.icap.org/
portals/0/download/all_pdfs/Other_Publications/dp_english.pdf
Alcohol and Coronary Heart Disease: Drinking Patterns
and Mediators of Effect
ERIC B. RIMM, SCD, AND CAROLINE MOATS, MS
An inverse association between alcohol consumption and coronary heart disease (CHD) has been shown in
epidemiologic studies for more than 30 years; beneficial changes in high-density lipoprotein (HDL) cholesterol, clotting factors, insulin sensitivity, and inflammation provide biological plausibility. Recently, the
importance of including drinking patterns in defining ‘‘moderate drinking’’ has been appreciated. A recent
meta-analysis raised questions about systematic misclassification error in observational studies because of
inclusion among ‘‘nondrinkers’’ of ex-drinkers and/or occasional drinkers. However, misclassification
among a small percentage of nondrinkers cannot fully explain the inverse relation, and there is substantial
evidence to refute the ‘‘sick quitter’’ hypothesis. Furthermore, it has been shown that moderate alcohol consumption reduces CHD and mortality in individuals with hypertension, diabetes, and existing CHD. To
address the issue of residual confounding by healthy lifestyle in drinkers, in a large prospective study we restricted analysis to only ‘‘healthy’’ men (who did not smoke, exercised, ate a good diet, and were not obese).
Within this group, men who drank moderately had a relative risk for CHD of 0.38 (95% CI, 0.16–0.89)
compared with abstainers, providing further evidence to support the hypothesis that the inverse association
of alcohol to CHD is causal, and not confounded by healthy lifestyle behaviors.
Ann Epidemiol 2007;17:S3–S7. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol, Coronary Heart Disease, Drinking Patterns, Epidemiology, Ethanol, Myocardial
Infarction.
The study of the effects of alcohol on health in human populations has spanned a wide spectrum, ranging from controlled feeding studies to cross-cultural comparisons to
large scale prospective studies (1). Many detailed review papers have covered these areas of research d several are also
included in this issue of the Annals of Epidemiology. In this paper we focus on the inverse association between moderate alcohol consumption and CHD and highlight recent evidence
that contributes to the hypothesis that alcohol is causally related to lower risk through known biological mechanisms.
The first study to suggest an inverse association between
alcohol consumption and CHD was published more than 30
years ago (2). Since then, additional data confirming this association have been collected in more than 25 countries
from at least 100 different studies. Evidence from these studies suggests that beneficial changes in HDL cholesterol
levels, clotting factors, insulin sensitivity, and markers of inflammation provide biological plausibility to the observed
association. Recently, epidemiological studies have more
carefully addressed the importance of drinking patterns,
From the Departments of Epidemiology (E.B.R.) and Nutrition (E.B.R.,
C.M.), Harvard School of Public Health, Boston, MA; and Channing
Laboratory, Department of Medicine, Brigham and Women’s Hospital,
Harvard Medical School, Boston, MA (E.B.R.).
Address correspondence to: Eric B. Rimm, ScD, Department of Epidemiology, Harvard School of Public Health, 677 Huntington Ave., Boston,
MA 02215; Tel.: (617) 432-1843. E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
frequency, quantity, and beverage choice while also accounting for differences in physical activity, diet, and
changes in alcohol consumption over the adult lifespan.
The preponderance of evidence suggests that ‘‘moderate’’
drinkers of alcohol have the greatest cardiovascular benefit.
The definition of moderate drinking has a very broad range
(from 5 to 60 g of alcohol per day) depending on the study
population and the tool used for assessment. For purposes
of this summary, we will use the definition defined by the
2005 USDA Dietary Guidelines (3). Moderate drinking is
considered to be no more than one drink per day for women
and no more than two drinks per day for men. A drink is
more specifically defined as 12 ounces of regular beer, 5 ounces of wine, and 1.5 ounces of 80-proof distilled spirits. Thus,
at the extreme, two drinks a day would be up to approximately 30 g of alcohol.
Substantial epidemiological evidence from geographical
comparisons, large cohort studies, and several meta-analyses
links moderate alcohol consumption with a lower risk of cardiovascular disease. There are too many epidemiological
studies that have independently analyzed the association between alcohol and CHD risk to summarize in this paper,
though they have been summarized elsewhere (4–7). The
population, size, and length of follow-up all vary greatly
from study to study, which provides further support to the robustness and generalizability of the findings. For example, in
the first study in 1974, Klatsky et al (2) analyzed recent
health and lifestyle histories of 464 patients in the San
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.002
S4
Rimm and Moats
ALCOHOL AND CORONARY HEART DISEASE
Francisco Bay Area who had a multiphase health checkup
and subsequently a first myocardial infarction. The alcohol
histories of the cases suggested significantly lower alcohol
intake compared with that of 50,000 healthy examinees.
The Framingham Heart Study, a prospective epidemiologic
cohort study established in Massachusetts in 1948 to evaluate potential risk factors for CHD, is also important because
of the extremely long follow-up period (26 years in an early
report of alcohol and CHD). With a single measure of alcohol at baseline, the authors reported an inverse association
with subsequent CHD risk (8). Similar inverse associations
have been documented in such diverse populations as those
in France, Japan, Denmark, Germany, Finland, Korea,
Great Britain, the United States, and many others (9–15).
Despite the strength and consistency across studies, some
still argue that generalization of the results may not be possible because of the selected nature of several of these study
populations. However, general population surveys including
the National Health and Nutrition Examination Survey
(NHANES) in the United States have also found benefit
from moderate alcohol consumption (16).
More recent studies of alcohol and CHD have focused on
subgroups defined by age or health status. Although alcohol
in moderation will likely provide greater benefit for older
populations where rates of CHD are highest (10), the etiology of CHD is such that moderate consumption in middle
age also is beneficial. Several important risk factors for
CHD, such as obesity and the prevalence of diabetes, both
of which have been increasing in younger adults, are consistently reported to be inversely associated with alcohol consumption (17–22). In addition to subgroups across age strata,
benefit from moderate alcohol consumption has been documented in many other subpopulations, such as those persons
with diabetes and hypertension or patients followed up after
myocardial infarction (23–26). Thus, in summary, the epidemiological evidence for an inverse association between
moderate alcohol consumption and CHD is extensive and
generalizable to populations defined by age, ethnicity, geography, and prevalent health conditions.
Despite this evidence, some have raised doubt as to
whether the observed inverse association between alcohol
and CHD risk is causal. In 1988, Shaper et al (27) suggested
an alternative explanationdthe ‘‘sick quitter’’ hypothesisd
suggesting that the reference category of abstainers were at
higher risk of CHD not because they did not drink alcohol,
but rather because they stopped drinking because of existing
illness. Fillmore et al (28) recently published a subsequent
review and meta-analysis supporting the hypothesis by
Shaper and colleagues and adding further theories of systematic misclassification error due to changes in alcohol consumption with aging and poor health. Unfortunately, in
their review (28), the inclusion criteria were excessively
selective and thus the results were difficult to interpret.
AEP Vol. 17, No. 5S
May 2007: S3–S7
Although the hypothesis put forward by Shaper et al may
have a theoretical basis, a simple misclassification among
a small percentage of nondrinkers cannot fully explain
why men and women who consume one to three drinks
per day have a 10% to 40% lower risk of CHD when compared with the 10% to 60% of the population (depending
on the study) who abstain (29). There is substantial evidence to refute the ‘‘sick quitter’’ hypothesis. First, subsequent analyses from several large cohorts with data on
lifelong or long-term abstainers have successfully refuted
the hypothesis (30, 31). Second, moderate alcohol consumption not only reduces the risk of incident CHD in
healthy individuals, but also reduces CHD and mortality
in individuals with hypertension, diabetes, and existing
CHD (23, 25, 26, 32). Thus, if the inverse association persists in a ‘‘sick’’ population, this suggests that changes in
alcohol consumption after diagnosis do not appreciably
change the shape of the relationship between alcohol and
subsequent disease. Third, and most importantly (as we discuss in more detail below), there are more than 70 metabolic
studies which document important changes in the cardiovascular risk profile caused by daily drinking (29).
Some have also suggested that the inverse association between alcohol and CHD may not be causal because moderate
drinkers may also be more likely to eat better, exercise more,
and live a healthier life. Although most prospective studies
of alcohol and CHD control for potential confounders, uncontrolled or residual confounding could still explain the
beneficial effect attributed to alcohol. Ideally, a long-term
randomized trial of alcohol would help to rule out confounding. Such trials have been conducted to study changes in
markers of CHD such as HDL cholesterol, triglycerides, glycemic control, and clotting factors, but long-term trials
which follow hard clinical cardiovascular events may be difficult to conduct because of cost, ethical considerations
regarding the randomization of alcohol, and finally the
generalizability of results among what would need to be
a very highly selective population who would be willing to
be randomized to a specific drinking pattern for a long period.
We recently conducted a re-analysis of men in the
Health Professionals Follow-up Study (HPFS) to examine
the association between alcohol and CHD in the strata of
men already at low risk based on other health lifestyle
choices. This effectively eliminates the potential for residual
confounding by healthy behavior because only men with
a healthy lifestyle were included in the analysis.
Briefly, the HPFS is a prospective investigation of the dietary etiologies of heart disease and cancer among 51,529
men, aged 40 to 75 years at baseline in 1986 (9, 31). The
population consists of dentists, veterinarians, pharmacists,
optometrists, osteopathic physicians, and podiatrists. In
our initial analysis from this population with 4 years of
follow-up (9), we found a strong inverse association for
AEP Vol. 17, No. 5S
May 2007: S3–S7
Rimm and Moats
ALCOHOL AND CORONARY HEART DISEASE
Alcohol
HDL-C
Fibrinogen
HgA1c
Benefit due to:
3
ll
A
ib
rin
o
+F
Hg
A1
c
ge
n
L
D
+H
el
Beta coef (freq/wk)
0
-0.05
-0.1
-0.15
-0.2
-0.25
-0.3
-0.35
-0.4
Ba
se
M
od
A
ll3
ib
rin
o
Hg
A1
c
ge
n
L
D
+F
+H
Ba
se
M
od
Benefit due to:
a previous meta-analysis, we summarized the effects of alcohol on a variety of these biomarkers from experimental studies (29) and found strong and consistent evidence linking
moderate alcohol intake with higher concentrations of
HDL cholesterol and apolipoprotein A1 and lower concentrations of fibrinogen. Since the publication of this article,
several cross-sectional and experimental studies have also
reported strong positive associations between alcohol and
increased insulin sensitivity (35–37). The combined beneficial effect of alcohol on the underlying biological mechanisms, specifically HDL cholesterol, fibrinogen, and
hemoglobin A1c (as a marker of glycemic control and insulin
resistance), can almost entirely explain the reduced CHD
risk among moderate drinkers.
Historically, early studies by Langer et al. (38) estimated
that approximately 50% of the benefit of moderate alcohol
consumption could be explained by the direct effects of alcohol on HDL cholesterol. We extended this analysis and
included other biomarkers of CHD. In nested case-control
studies of CHD from the HPFS and the Nurses’ Health
Study, we modeled the inverse association between alcohol
and CHD among men and women with and without controlling for the potential biological mediators, HDL cholesterol, fibrinogen, and hemoglobin A1c (39) (Fig. 1). In the
figure, each subsequent model after the basic model controls
for an additional biomarker thought to be in the causal pathway between moderate alcohol consumption and reduced
risk of CHD. For both the men and the women, approximately half of the benefit (the magnitude of the regression
coefficient is halved) is explained away after including
HDL cholesterol in the model. Subsequent models illustrate
the change in the coefficient after controlling for fibrinogen,
hemoglobin A1c and all 3 biomarkers simultaneously. The
final model which controls for all 3 potential mediators suggests that all of the benefit of alcohol in men and 80% of the
benefit of alcohol in women can be explained by these biomarkers in the causal pathway between alcohol and CHD.
The importance of the role of drinking pattern and specifically beverage choice is still disputed. Similar to the
B
0
-0.02
-0.04
-0.06
-0.08
-0.1
-0.12
-0.14
-0.16
el
A
Beta coef (freq/wk)
moderate alcohol consumption that persisted even after exclusion of men with baseline health conditions which may
have led to a change in drinking pattern. In a subsequent
analysis of these men with 12 years of follow-up and 1,418
incident CHD cases, we investigated the role of drinking
pattern. Men with the lowest rates of CHD consumed alcohol 3 to 4 days per week, regardless of the type of beverage
(beer, wine, spirits) or whether or not alcohol was consumed
with meals (31).
Finally, in our most recent analysis with 16 years of
follow-up and 2,183 incident CHD cases, we identified 5
factors that define a low-risk individual: (1) nonsmoker,
(2) body mass index !25 kg/m2, (3) moderate-to-vigorous
activity 30 minutes or more per day, (4) the top 40% of
the distribution for a healthy diet score, and (5) moderate
alcohol consumption. Individually, these modifiable lifestyle factors were associated with lower risk of CHD, but
a combination of lifestyle factors had an even greater impact
(33). Men who adhered to all 5 low-risk factors had the lowest risk of CHD (relative risk Z 0.13; 95% confidence interval [CI] Z 0.09–0.19) compared with men with no low-risk
factors during the 16-year follow-up period. To address the
issue of residual confounding by healthy lifestyle among
drinkers, we restricted the analysis to the 8,867 healthy
men who adhered to the first 4 low-risk behaviors (not including alcohol). We then examined the association for alcohol consumption among this group of the healthiest men.
In this group we still had 106 incident CHD cases and found
that men who drank moderately (15.0–29.9 g/d) had a relative risk of 0.38 (95% CI, 0.16–0.89) compared with abstainers. This strong inverse association between moderate
alcohol consumption and CHD in predominantly healthy
individuals adds further evidence to support the hypothesis
that the inverse association is causal, and not confounded by
healthy lifestyle behaviors (34).
The strongest evidence to support a causal association between alcohol and CHD comes from metabolic studies
which demonstrate changes in lipids, hemostatic factors, inflammatory markers, and glycemic control parameters. In
S5
Alcohol
HDL-C
Fibrinogen
HgA1c
FIGURE 1. The inverse association between alcohol consumption and risk of CHD in the Health Professionals Follow-up Study (A) and
the Nurses’ Health Study (B). The b-coefficient (also called the regression coefficient) is the natural log of the odds ratio in frequency of
consumption per week. The basic model controls for age, smoking, and date of blood sample return (39).
S6
Rimm and Moats
ALCOHOL AND CORONARY HEART DISEASE
argument that moderate drinkers may have a healthier lifestyle, there are documented differences in healthy lifestyle
characteristics by predominant beverage choice and drinking pattern, which vary by country and cultural setting
(40). Thus it is more likely that in any given culture, the alcohol-containing beverage which is consumed in moderation, most frequently, and with meals will be of greatest
benefit. For example, in a large cohort from France (where
wine is the most frequent alcohol beverage of consumption),
Renaud et al. (11) evaluated the health risk of both wine and
beer and found that both beverage choices reduced the risk
of cardiovascular death, albeit wine was more beneficial. In
Germany, where much of the alcohol consumed is beer, the
risk of CHD events for drinkers was one half that of nondrinkers (12). In Japan, where spirits are the common beverage of choice, Kitamura et al. (13) found that alcohol
intake reduced premature incidence of CHD among urban
Japanese middle-aged men; within that study, less than
1% of alcohol consumed was from grape wine. When we
summarized the available evidence, the inverse association
between moderate alcohol intake and CHD was consistent,
regardless of beverage choice (41). This suggests that the direct metabolic effects of ethanol are largely responsible for
the cardiovascular benefits and that other constituents of
the beverages contribute little if anything to further benefit.
The evidence discussed above provides substantial support for the hypothesis that moderate drinking reduces the
risk of CHD. Beer, wine, and spirits all have demonstrated
significant benefits. These benefits are likely mediated
through strong and lasting effects of alcohol on HDL cholesterol, fibrinogen, and glycemic control. The ‘‘sick-quitter’’
hypothesis and the concern that moderate drinkers lead
a healthier lifestyle may explain a small proportion of the
benefit attributed to alcohol in some studies, but recent studies which have removed sick quitters, updated alcohol and
covariate information on diet and lifestyle factors, and separately documented benefits of alcohol among healthy and
unhealthy populations further add to the evidence that
moderate alcohol consumption is causally related to a lower
risk of CHD.
REFERENCES
1. Lucas DL, Brown RA, Wassef M, Giles TD. Alcohol and the cardiovascular system research challenges and opportunities. J Am Coll Cardiol.
2005;45:1916–1924.
2. Klatsky AL, Friedman GD, Siegelaub AB. Alcohol consumption before
myocardial infarction. Results from the Kaiser-Permanente epidemiologic
study of myocardial infarction. Ann Intern Med. 1974;81:294–301.
3. U.S. Department of Health and Human Services and U.S. Department of
Agriculture. Dietary Guidelines for Americans, 2005, 6th ed. Washington,
DC: U.S. Government Printing Office; 2005.
4. Maclure M. Demonstration of deductive meta-analysis: ethanol intake and
risk of myocardial infarction. Epidemiol Rev. 1993;15:328–351.
AEP Vol. 17, No. 5S
May 2007: S3–S7
5. Hill JA. In vino veritas: alcohol and heart disease. Am J Med Sci.
2005;329:124–135.
6. Rehm J, Gmel G, Sempos CT, Trevisan M. Alcohol-related morbidity and
mortality. Alcohol Res Health. 2003;27:39–51.
7. Klatsky AL. Drink to your health? Sci Am. 2003;288:74–81.
8. Friedman LA, Kimball AW. Coronary heart disease mortality and alcohol
consumption in Framingham. Am J Epidemiol. 1986;124:481–489.
9. Rimm EB, Giovannucci EL, Willett WC, Colditz GA, Ascherio A, Rosner
B, et al. Prospective study of alcohol consumption and risk of coronary
disease in men. Lancet. 1991;338:464–468.
10. Fuchs CS, Stampfer MJ, Colditz GA, Giovannucci EL, Manson JE, Kawachi I, et al. Alcohol consumption and mortality among women. N Engl J
Med. 1995;332:1245–1250.
11. Renaud SC, Guéguen R, Siest G, Salamon R. Wine, beer, and mortality in
middle-aged men from Eastern France. Arch Intern Med. 1999;159:1865–
1870.
12. Keil U, Chambless LE, Döring A, Filipiak B, Stieber J. The relation of alcohol intake to coronary heart disease and all-cause mortality in a beerdrinking population. Epidemiology. 1997;8:150–156.
13. Kitamura A, Iso H, Sankai T, Naito Y, Sato S, Kiyama M, et al. Alcohol
intake and premature coronary heart disease in urban Japanese men. Am J
Epidemiol. 1998;147:59–65.
14. Wannamethee SG, Shaper AG. Type of alcoholic drink and risk of major
coronary heart disease events and all-cause mortality. Am J Public Health.
1999;89:685–690.
15. Tolstrup J, Jensen MK, Tjonneland A, Overvad K, Mukamal KJ, Grønbæk
M. Prospective study of alcohol drinking patterns and coronary heart disease in women and men. BMJ. 2006;332:1244–1248.
16. Gartside PS, Wang P, Glueck CJ. Prospective assessment of coronary heart
disease risk factors: the NHANES I epidemiologic follow-up study
(NHEFS) 16-year follow-up. J Am Coll Nutr. 1998;17:263–269.
17. Wannamethee SG, Shaper AG, Perry IJ, Alberti KGMM. Alcohol consumption and the incidence of type II diabetes. J Epidemiol Comm Health.
2002;56:542–548.
18. Wannamethee SG, Camargo CA Jr, Manson JE, Willett WC, Rimm EB.
Alcohol drinking patterns and risk of type 2 diabetes among younger
women. Arch Intern Med. 2003;163:1329–1336.
19. Wannamethee SG, Field AE, Colditz GA, Rimm EB. Alcohol intake and
8-year weight gain in women: a prospective study. Obes Res.
2004;12:1386–1396.
20. Conigrave KM, Hu FB, Camargo CA Jr, Stampfer MJ, Willett WC, Rimm
EB. A prospective study of drinking patterns in relation to risk of type 2
diabetes among men. Diabetes. 2001;50:2390–2395.
21. Beulens JW, Stolk RP, van der Schouw YT, Grobbee DE, Hendriks HF,
Bots ML. Alcohol consumption and risk of type 2 diabetes among older
women. Diabetes Care. 2005;28:2933–2938.
22. Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Moderate
alcohol consumption lowers the risk of type 2 diabetes: a metaanalysis of prospective observational studies. Diabetes Care. 2005;28:
719–725.
23. Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Prior
alcohol consumption and mortality following acute myocardial infarction.
JAMA. 2001;285:1965–1970.
24. Beulens JW, Rimm EB, Stampfer MJ, Ascherio A, Hendriks HF, Mukamal
KJ. Alcohol consumption and risk of coronary heart disease in men with
hypertension. Ann Int Med. 2007;146(1):10–19.
25. Malinski MK, Sesso HD, Lopez-Jimenez F, Buring JE, Gaziano JM. Alcohol
consumption and cardiovascular disease mortality in hypertensive men.
Arch Intern Med. 2004;164:623–628.
26. Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Meta-analysis
of the relationship between alcohol consumption and coronary heart
disease and mortality in type 2 diabetic patients. Diabetologia.
2006;49:648–652.
AEP Vol. 17, No. 5S
May 2007: S3–S7
27. Shaper AG, Wannamethee G, Walker M. Alcohol and mortality in British
men: Explaining the U-shaped curve. Lancet. 1988;2:1267–1273.
28. Fillmore KM, Kerr WC, Stockwell T, Chikritzhs T, Bostrom A. Moderate
alcohol use and reduced mortality risk: systematic error in prospective
studies. Addiction Research & Theory. 2006;14:101–132.
29. Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects
on lipids and haemostatic factors. BMJ. 1999;319:1523–1528.
Rimm and Moats
ALCOHOL AND CORONARY HEART DISEASE
S7
35. Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, et al.
Insulin sensitivity and regular alcohol consumption: large, prospective,
cross sectional population study (Bruneck study). Br Med J. 1996;313:
1040–1044.
36. Vitelli LL, Folsom AR, Shahar E, Winkhart SP, Shimakawa T, Stevens J,
et al. Association of dietary composition with fasting serum insulin level d
the ARIC study. Nutr Metab Cardiovasc Dis. 1996;6:194–202.
30. Klatsky AL, Armstrong MA, Friedman GD. Risk of cardiovascular mortality in alcohol drinkers, ex-drinkers, and nondrinkers. Am J Cardiol.
1990;66:1237–1242.
37. Davies MJ, Baer DJ, Judd JT, Brown ED, Campbell WS, Taylor PR. Effects
of moderate alcohol intake on fasting insulin and glucose concentrations
and insulin sensitivity in postmenopausal women d a randomized
controlled trial. JAMA. 2002;287:2559–2562.
31. Mukamal KJ, Conigrave KM, Mittleman MA, Camargo CA Jr, Stampfer
MJ, Willett WC, et al. Roles of drinking pattern and type of alcohol consumed in coronary heart disease in men. N Engl J Med. 2003;348:109–118.
38. Langer RD, Criqui MH, Reed DM. Lipoproteins and blood pressure as biological pathways for effect of moderate alcohol consumption on coronary
heart disease. Circulation. 1992;85:910–915.
32. Tanasescu M, Hu FB, Willett WC, Stampfer MJ, Rimm EB. Alcohol consumption and risk of coronary heart disease among men with type 2 diabetes mellitus. Am J Coll Cardiol. 2001;38:1836–1842.
39. Mukamal KJ, Jensen MK, Grønbæk M, Stampfer MJ, Manson JE, Pischon
T, et al. Drinking frequency, mediating biomarkers, and risk of myocardial
infarction in women and men. Circulation. 2005;112:1379–1381.
33. Chiuve SE, McCullough ML, Sacks FM, Rimm EB. Healthy lifestyle
factors in the primary prevention of coronary heart disease among men.
Circulation. 2006;114:160–167.
40. Tjonneland A, Grønbæk M, Stripp C, Overvad K. Wine intake and diet in
a random sample of 48763 Danish men and women. Am J Clin Nutr.
1999;69:49–54.
34. Mukamal KJ, Chiuve SE, Rimm EB. Alcohol consumption and risk for coronary heart disease in men with healthy lifestyles. Arch Intern Med.
2006;166:2145–2150.
41. Rimm EB, Klatsky A, Grobbee D, Stampfer MJ. Review of moderate alcohol consumpton and reduced risk of coronary heart disease: is the effect
due to beer, wine, or spirits? Br Med J. 1996;312:731–736.
Alcohol Intake and Noncoronary Cardiovascular Diseases
KENNETH MUKAMAL, MD
Moderate drinking has complex associations with cardiovascular diseases other than coronary heart disease.
Recent cohort studies examining the relationship between alcohol use and ischemic stroke have shown
a modest association, with risk ratios approximating 0.8 and the lowest risk among those who drink less
than daily. In contrast, alcohol use is generally associated with an approximate dose-dependent risk for
hemorrhagic stroke throughout the full range of intake.
Several prospective studies of alcohol intake and congestive heart failure have found lower risk with
moderate drinking. This risk is also dose dependent through the moderate range, but its underlying mechanism remains uncertain. Accounting for the lower risk of myocardial infarction associated with moderate
intake does not eliminate the observed association. Cohort studies have found no association of long-term
alcohol intake with risk of atrial fibrillation below levels of at least 3 standard drinks per day. Finally, two
prospective studies have found lower risks of claudication or clinically more severe peripheral arterial
disease among moderate drinkers, an association also supported by cross-sectional studies of alcohol intake
and ankle-brachial index.
Ann Epidemiol 2007;17:S8–S12. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Stroke, Heart Failure, Peripheral Arterial Disease, Atrial Fibrillation.
INTRODUCTION
Although the relationship of alcohol consumption and coronary heart disease (CHD) has been extensively addressed
in dozens of studies (1–3), alcohol consumption has also
been associated with other cardiovascular diseases, including ischemic and hemorrhagic stroke, congestive heart failure (CHF), atrial fibrillation (AF), and peripheral arterial
disease (PAD) (4–7). These associations include both positive and inverse associations, even at moderate levels of
consumption, although the strength of evidence surrounding these relationships is not yet as strong as for CHD. Complicating this issue further, these cardiovascular diseases are
quite heterogeneous, both across diseases and even within
a single disease. In this review, I will examine and summarize
some of the evidence linking alcohol to these conditions
and identify new areas for future research.
NONCORONARY CARDIOVASCULAR DISEASES
Stroke
The clinical syndrome of stroke is a quintessential example
of the heterogeneity among cardiovascular diseases. Stroke
From the Division of General Medicine and Primary Care, Beth Israel
Deaconess Medical Center, Boston, MA.
Address correspondence to: Kenneth Mukamal, MD, Division of General Medicine & Primary Care, Beth Israel Deaconess Medical Center,
Rose-114, 330 Brookline Ave, Boston, MA 02215. Tel.: (617) 667-8975;
fax: (617) 667-2854. E-mail: [email protected].
Research related to this article has been funded entirely by the National
Institutes of Health and the American Heart Association.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
comprises two main types of cardiovascular disease d
ischemic and hemorrhagic d and several subtypes of each
(e.g., atherothrombotic, lacunar, embolic, subarachnoid
hemorrhage, intracerebral hemorrhage). Alcohol appears
to have decidedly different associations with these various
types.
Ischemic stroke, the most common type in the United
States and Europe, is the result of vascular occlusion of an
intracerebral artery, either through local thrombosis or distal embolism. It is noteworthy that two of the strongest
chronic risk factors for ischemic stroke are hypertension
and AF (8), both of which have been attributed to heavy
drinking (see below). Thus, even though ischemic stroke
shares an atherosclerotic and thrombotic etiology with
CHD, the particular factors that most strongly predict ischemic stroke risk appear to be in the subset of cardiovascular
risk factors that are exacerbated by heavy alcohol use.
As a consequence, it may not be surprising that the apparent
inverse association of moderate alcohol intake with ischemic stroke risk occurs at a lower dose of alcohol consumption and with a lower magnitude of risk reduction than
does the corresponding association with CHD risk (9, 10).
It is also not surprising that heavy drinking is consistently associated with higher stroke risk.
For many years, observational (and largely case-control)
studies of the relationship between alcohol consumption
and risk for ischemic stroke have suggested a relatively
strong inverse relationship (4, 11). For example, in the
Nurses’ Health Study cohort, Stampfer and colleagues documented 66 incident ischemic strokes and found relative risks
of 0.3 (95% confidence interval [CI], 0.1–0.7) for intake of 5
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.003
AEP Vol. 17, No. 5S
May 2007: S8–S12
Selected Abbreviations and Acronyms
AAA Z abdominal aortic aneurysm
AF Z atrial fibrillation
ANP Z atrial natriuretic peptide
CHD Z coronary heart disease
CHF Z congestive heart failure
CI Z confidence interval
PAD Z peripheral arterial disease
to 14 g of alcohol per day, and 0.5 (95% CI, 0.2–1.1) for intake levels of 15 or more g/d (12). However, more recent
studies have called the magnitude of lower risk into question. In a meta-analysis restricted to 19 cohort studies, Reynolds and colleagues found that light drinking (!12 g/d)
was associated with a relative risk of 0.82 (95% CI, 0.73–
0.92), but that consumption of 12 to 60 g/d was not associated with risk (4). In a representative study, Klatsky and
colleagues reported a U-shaped association between alcohol
intake and hospitalization for ischemic stroke among members of a single insurance plan. Relative risks were 0.8 (95%
CI, 0.7–1.0) among those who consumed between one drink
a month and one drink a day and 1.0 (95% CI, 0.8–1.2)
among those who consumed 3 or more drinks per day
(10). For comparison, these authors found a direct inverse
association between alcohol use and risk of CHD hospitalization, with relative risks of 0.6 among women and 0.7
among men who consumed 3 or more drinks per day (9).
In an analysis of the Cardiovascular Health Study, a lower
risk of ischemic stroke was confined to consumers of 1 to 6
drinks per week (hazard ratio 0.75; 95% CI, 0.53–1.06), although even this level of intake modestly but not significantly increased the risk among carriers of the APOE4
allele (13). In the same population (14), the corresponding
magnitude of lower risk for CHD was lower and occurred at
an intake of 14 or more drinks per week.
Dulli has emphasized how the heterogeneity of ischemic
stroke may influence its association with alcohol intake
(15). To address this, we examined ischemic stroke risk in
the Health Professionals Follow-up Study (16). Concordant
with previous work, lower risk was restricted to consumers of
an average of 0.1 to 9.9 g of ethanol per day, and particularly
to those whose consumption was spread out over 3 to 4 days
per week (hazard ratio 0.56; 95% CI, 0.31–1.02). However,
stroke subtype appeared to modify risk substantially. Compared with abstention, the hazard ratios associated with intake of 0.1 to 9.9 g/d were 0.76 (95% CI, 0.50–1.15) for
confirmed thrombotic stroke, but 2.20 (95% CI, 0.84–
5.76) for confirmed embolic stroke. Larger studies with effective classification of ischemic stroke subtypes are needed
to confirm these findings.
It should be repeated that little controversy exists regarding the relationship of heavier drinking with a higher risk of
ischemic stroke (4, 16). Moreover, even single episodes of
Mukamal
ALCOHOL INTAKE AND OTHER CARDIOVASCULAR DISEASES
S9
intake of 3 or more drinks or drinking to intoxication may
acutely and transiently increase the risk of ischemic stroke
for the ensuing 24 hours (17–21).
Hemorrhagic stroke is a common stroke type in Asia,
where most studies addressing it have been performed, and
includes both subarachnoid and intracerebral subtypes. In
a meta-analysis of cohort studies of subarachnoid hemorrhage, Feigin found increased risk restricted to heavier
intake (O150 g/wk), with a summary relative risk of 2.1
(95% CI, 1.5–2.8) (22). In a comparable meta-analysis of
case-control studies of intracerebral hemorrhage, Ariesen
found summary odds ratios of 2.05 (95% CI, 1.35–3.11)
for intake of 56 g or less per day, and 4.11 (95% CI, 2.54–
6.65) for intake of more than 56 g/d (23), although a large
cohort study published subsequently found an increase in
risk to be confined to consumers of 6 or more drinks per
day (24). These results are largely consistent with Renaud’s
hypothesis that the effects of moderate drinking are mediated, at least in part, through activity of platelets and prothrombotic factors (25). If this hypothesis is correct, one
would anticipate that moderate drinking would decrease
the risk of thrombotic events, such as myocardial infarction
and ischemic stroke, but increase the risk of bleeding in the
brain and possibly other sites (26), exactly as has been
observed.
Congestive heart failure
For decades (27), clinical investigators interested in the cardiac effects of ethanol focused their efforts on alcoholic cardiomyopathy, a dilated cardiomyopathy attributed primarily
to cardiotoxic effects of ethanol and secondarily also associated with malnutrition (28). Interestingly, although it has
long been recognized that only substantial ethanol intake
leads to cardiomyopathy (although less so in women) (29),
only recently have studies demonstrated that in subjects
with cardiomyopathy, controlled drinking could be as effective as abstinence in stabilizing or restoring myocardial
function (30).
More recent interest in CHF has been spurred by 3 observations. First, with the aging of Western populations, CHF
has rapidly become a serious public health concern. In
2003 alone, more than one million Americans were hospitalized with a primary diagnosis of CHF (31). Second, a substantial portion of the burden of CHF in the United States
and Europe is the consequence of CHD and ischemic cardiomyopathy. If moderate alcohol use is associated with a lower
risk of CHD, one might anticipate that it would also be associated with a lower risk of CHF. Third, physiological experiments have suggested that alcohol consumption may
induce effective atrial natriuretic peptide (ANP)–mediated
diuresis (32, 33) and lower pulmonary artery pressure (34),
both of which would be expected to reduce the incidence
S10
Mukamal
ALCOHOL INTAKE AND OTHER CARDIOVASCULAR DISEASES
of CHF; the association of alcohol intake with ANP has now
been confirmed in a large observational study (35).
The observed association of alcohol use with risk for
CHF has justified this interest but has also added new complexity. In at least 4 large prospective studies (7, 36–38), investigators have found that moderate drinking is associated
with approximately a 50% lower risk of CHF, although the
level of intake linked to lowest risk varies somewhat. Interestingly, in all 4 studies, accounting for incident coronary
disease did not eliminate the relationship, so that even
where previous myocardial infarction was not involved,
risk appeared to be, to some degree, inversely related to
moderate drinking. Some (39), but not all (40, 41), studies
of alcohol intake among patients with established left ventricular dysfunction have also suggested a benefit from light
or moderate drinking; one study has suggested an association with lower mortality in men but higher mortality in
women (42).
Atrial fibrillation
Atrial fibrillation (AF) is the most common chronic cardiac
arrhythmia and is closely linked with hypertension and
CHF. Clinicians have long recognized the temporal association of episodic heavy alcohol use with onset of AF, dubbed
the ‘‘holiday heart’’ syndrome (43, 44). However, the association of regular alcohol intake with risk for AF has only recently begun to receive attention. Given the association of
heavy drinking with risk for AF, one might hypothesize that
even moderate drinking would be associated with higher
risk, although animal models have suggested the opposite
(45, 46). In human populations, some studies have shown
positive relationships in men but not in women (47),
whereas some have even shown inverse relationships (48).
However, most studies have suggested a threshold effect
(6, 44, 49–51). In these studies, there has generally been
no discernible difference in risk between abstainers and
moderate drinkers, but a higher risk for the heaviest
drinkers. Although the absolute threshold of risk is uncertain, the largest study to date suggests that risk for AF does
not increase below a long-term intake level of at least 28
and probably 35 drinks per week (51).
An interesting area for future investigation is the effect of
alcohol consumption on individuals with chronic AF, most
of whom are likely to be prescribed warfarin or similar anticoagulants. Although guidelines have often explicitly recommended that individuals taking warfarin abstain
entirely from drinking (52), the limited available data suggest that the risk for excessive anticoagulation is not higher
in moderate drinkers than it is in abstainers, although it may
be elevated in binge drinkers (53, 54).
AEP Vol. 17, No. 5S
May 2007: S8–S12
Peripheral arterial disease
The final manifestation of cardiovascular disease to be discussed herein is peripheral arterial disease (PAD), most
commonly localized to the lower extremities; for practical
purposes, abdominal aortic aneurysm (AAA) is also considered in this category. Although PAD has generally received
the least attention from researchers among the major atherosclerotic vascular diseases, it is nonetheless widely prevalent and a major cause of disability, primarily related to
walking.
The body of evidence linking moderate drinking to PAD
is modest. The Strong Heart Study and Rotterdam Study
both found inverse cross-sectional associations of moderate
drinking with PAD (defined as a low ankle-brachial index)
that tended to be strongest in women (55, 56), although
other cross-sectional studies have not consistently confirmed
this finding (57, 58). Surprisingly, a subsequent analysis of
the progression of the ankle-brachial index in one of these
latter studies has suggested an inverse relationship between
alcohol use and a 5-year change in the ankle-brachial index
(59). To our knowledge, two prospective cohort studies have
assessed the risk of PAD in the general population. In an
analysis of symptomatic claudication from the Framingham
Heart Study (5), Djoussé and colleagues found hazard ratios
of 0.67 (95% CI, 0.42–0.99) associated with intake of 13–24
g/d for men and 0.44 (95% CI, 0.23–0.80) associated with intake levels of 7 to 12 g/d for women. In the Physicians’
Health Study (60), Camargo and colleagues found a hazard
ratio of 0.74 (95% CI, 0.57–0.97) associated with an intake
of 7 or more drinks per week, although the composite outcome included only 66 cases of lower-extremity arterial disease that required revascularization. In contrast, Ciccarone
and colleagues found no association of alcohol intake with
risk of PAD among diabetic adults (61).
There has been relatively little investigation into the relationship of moderate alcohol consumption with risk of
AAA. In an analysis of male smokers, Tornwall and colleagues found a U-shaped relationship, with relative risks
for AAA of 0.74 (95% CI, 0.49–1.13) among consumers
of up to 15 g of ethanol per day and 0.60 (95% CI, 0.36–
1.02) among consumers of 16 to 30 g of ethanol per day (62).
CONCLUSIONS
In summary, the relationship between moderate alcohol
consumption and noncoronary cardiovascular diseases is
complex, in part reflecting their heterogeneity. The relationship between alcohol use and the different subtypes of
stroke have received the most attention, but even these relationships are not without controversy. To date, evidence
generally suggests a J-shaped relationship between alcohol
use and ischemic stroke, with lower risk limited to
AEP Vol. 17, No. 5S
May 2007: S8–S12
consumers of approximately 1 drink every other day and
clearly higher risk among consumers of 3 or more drinks
per day. The risk of hemorrhagic stroke appears to rise in
a graded fashion with higher alcohol intake. The risks of
PAD and CHF both seem to be lower with moderate intake
in a range of up to 1 to 2 drinks per day, but larger prospective studies are needed to define these associations more precisely. Finally, AF seems to be unrelated to alcohol use
throughout the moderate range of consumption.
REFERENCES
Mukamal
ALCOHOL INTAKE AND OTHER CARDIOVASCULAR DISEASES
S11
18. Hillbom M, Kaste M. Ethanol intoxication: a risk factor for ischemic brain
infarction in adolescents and young adults. Stroke. 1981;12:422–425.
19. Hillbom M, Kaste M. Ethanol intoxication: a risk factor for ischemic brain
infarction. Stroke. 1983;14:694–699.
20. Hillbom M, Haapaniemi H, Juvela S, Palomaki H, Numminen H, Kaste M.
Recent alcohol consumption, cigarette smoking, and cerebral infarction
in young adults. Stroke. 1995;26:40–45.
21. Hillbom M, Numminen H, Juvela S. Recent heavy drinking of alcohol
and embolic stroke. Stroke. 1999;30:2307–2312.
22. Feigin VL, Rinkel GJ, Lawes CM, Algra A, Bennett DA, van Gijn J, et al.
Risk factors for subarachnoid hemorrhage: an updated systematic review of
epidemiological studies. Stroke. 2005;36:2773–2780.
23. Ariesen MJ, Claus SP, Rinkel GJ, Algra A. Risk factors for intracerebral
hemorrhage in the general population: a systematic review. Stroke.
2003;34:2060–2065.
1. Maclure M. Demonstration of deductive meta-analysis: ethanol intake and
risk of myocardial infarction. Epidemiol Rev. 1993;15:328–351.
24. Klatsky AL, Armstrong MA, Friedman GD, Sidney S. Alcohol drinking
and risk of hemorrhagic stroke. Neuroepidemiology. 2002;21:115–122.
2. Corrao G, Rubbiati L, Bagnardi V, Zambon A, Poikolainen K. Alcohol and
coronary heart disease: a meta-analysis. Addiction. 2000;95:1505–1523.
25. Renaud S, de Lorgeril M. Wine, alcohol, platelets, and the French paradox
for coronary heart disease. Lancet. 1992;339:1523–1526.
3. Di Castelnuovo A, Rotondo S, Iacoviello L, Donati MB, De Gaetano G.
Meta-analysis of wine and beer consumption in relation to vascular risk.
Circulation. 2002;105:2836–2844.
26. Kaufman DW, Kelly JP, Wiholm BE, Laszlo A, Sheehan JE, Koff RS, et al.
The risk of acute major upper gastrointestinal bleeding among users of aspirin and ibuprofen at various levels of alcohol consumption. Am J Gastroenterol. 1999;94:3189–3196.
4. Reynolds K, Lewis B, Nolen JD, Kinney GL, Sathya B, He J. Alcohol consumption and risk of stroke: a meta-analysis. JAMA. 2003;289:579–588.
5. Djousse L, Levy D, Murabito JM, Cupples LA, Ellison RC. Alcohol consumption and risk of intermittent claudication in the Framingham Heart
Study. Circulation. 2000;102:3092–3097.
27. Klatsky AL. Alcohol and cardiovascular diseases: a historical overview.
Novartis Found Symp. 1998;216:2–12. discussion 12–18, 152–158.
28. Piano MR. Alcoholic cardiomyopathy: incidence, clinical characteristics,
and pathophysiology. Chest. 2002;121:1638–1650.
6. Djousse L, Levy D, Benjamin EJ, Blease SJ, Russ A, Larson MG, et al.
Long-term alcohol consumption and the risk of atrial fibrillation in the
Framingham Study. Am J Cardiol. 2004;93:710–713.
29. Fernandez-Sola J, Estruch R, Nicolas JM, Pare JC, Sacanella E, Antunez E,
et al. Comparison of alcoholic cardiomyopathy in women versus men.
Am J Cardiol. 1997;80:481–485.
7. Klatsky AL, Chartier D, Udaltsova N, Gronningen S, Brar S, Friedman
GD, et al. Alcohol drinking and risk of hospitalization for heart failure
with and without associated coronary artery disease. Am J Cardiol.
2005;96:346–351.
30. Nicolas JM, Fernandez-Sola J, Estruch R, Pare JC, Sacanella E, UrbanoMarquez A, et al. The effect of controlled drinking in alcoholic cardiomyopathy. Ann Intern Med. 2002;136:192–200.
8. Manolio TA, Kronmal RA, Burke GL, O’Leary DH, Price TR. Short-term
predictors of incident stroke in older adults. The Cardiovascular Health
Study. Stroke. 1996;27:1479–1486.
9. Klatsky AL, Armstrong MA, Friedman GD. Red wine, white wine, liquor,
beer, and risk for coronary artery disease hospitalization. Am J Cardiol.
1997;80:416–420.
10. Klatsky AL, Armstrong MA, Friedman GD, Sidney S. Alcohol drinking
and risk of hospitalization for ischemic stroke. Am J Cardiol. 2001;
88:703–706.
31. American Heart Association. Heart Disease and Stroke Statistics d 2005
Update. Dallas (TX): American Heart Association; 2005.
32. Puddey IB, Vandongen R, Beilin LJ, Rouse IL. Alcohol stimulation of
renin release in man: its relation to the hemodynamic, electrolyte, and
sympatho-adrenal responses to drinking. J Clin Endocrinol Metab.
1985;61:37–42.
33. Colantonio D, Casale R, Desiati P, De Michele G, Mammarella M, Pasqualetti P. A possible role of atrial natriuretic peptide in ethanol-induced
acute diuresis. Life Sci. 1991;48:635–642.
11. Camargo CA Jr. Moderate alcohol consumption and stroke. The epidemiologic evidence. Stroke. 1989;20:1611–1626.
34. Greenberg BH, Schutz R, Grunkemeier GL, Griswold H. Acute effects
of alcohol in patients with congestive heart failure. Ann Intern Med.
1982;97:171–175.
12. Stampfer MJ, Colditz GA, Willett WC, Speizer FE, Hennekens CH. A
prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women. N Engl J Med. 1988;319:267–273.
35. Djousse L, Hunt SC, Eckfeldt JH, Arnett DK, Province MA, Ellison RC.
Alcohol consumption and plasma atrial natriuretic peptide. Am J Cardiol.
2006;98:628–632.
13. Mukamal KJ, Chung H, Jenny NS, Kuller LH, Longstreth WT Jr, Mittleman MA, et al. Alcohol use and risk of ischemic stroke among older adults:
the cardiovascular health study. Stroke. 2005;36:1830–1834.
36. Walsh CR, Larson MG, Evans JC, Djousse L, Ellison RC, Vasan RS, et al.
Alcohol consumption and risk for congestive heart failure in the Framingham Heart Study. Ann Intern Med. 2002;136:181–191.
14. Mukamal KJ, Chung H, Jenny NS, Kuller LH, Longstreth WT Jr, Mittleman MA, et al. Alcohol consumption and risk of coronary heart disease in
older adults: the cardiovascular health study. J Am Geriatr Soc.
2006;54:30–37.
37. Abramson JL, Williams SA, Krumholz HM, Vaccarino V. Moderate alcohol consumption and risk of heart failure among older persons. JAMA.
2001;285:1971–1977.
15. Dulli DA. Alcohol, ischemic stroke and lessons from a negative study.
Stroke. 2002;33:890–891.
16. Mukamal KJ, Ascherio A, Mittleman MA, Conigrave KM, Camargo CA
Jr, Kawachi I, et al. Alcohol and risk for ischemic stroke in men: the
role of drinking patterns and usual beverage. Ann Intern Med. 2005;
142:11–19.
17. Hillbom M, Kaste M. Does ethanol intoxication promote brain infarction
in young adults? Lancet. 1978;2:1181–1183.
38. Bryson CL, Mukamal KJ, Mittleman MA, Fried LP, Hirsch CH, Kitzman
DW, et al. The association of alcohol consumption and incident heart
failure: the Cardiovascular Health Study. J Am Coll Cardiol. 2006;
48:305–311.
39. Cooper HA, Exner DV, Domanski MJ. Light-to-moderate alcohol consumption and prognosis in patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000;35:1753–1759.
40. Aguilar D, Skali H, Moye LA, Lewis EF, Gaziano JM, Rutherford JD, et al.
Alcohol consumption and prognosis in patients with left ventricular
S12
Mukamal
ALCOHOL INTAKE AND OTHER CARDIOVASCULAR DISEASES
systolic dysfunction after a myocardial infarction. J Am Coll Cardiol.
2004;43:2015–2021.
41. Salisbury AC, House JA, Conard MW, Krumholz HM, Spertus JA.
Low-to-moderate alcohol intake and health status in heart failure patients.
J Card Fail. 2005;11:323–328.
42. Faris RF, Henein MY, Coats AJ. Influence of gender and reported alcohol
intake on mortality in nonischemic dilated cardiomyopathy. Heart Dis.
2003;5:89–94.
43. Greenspon AJ, Schaal SF. The ‘‘holiday heart’’: electrophysiologic studies
of alcohol effects in alcoholics. Ann Intern Med. 1983;98:135–139.
44. Rich EC, Siebold C, Campion B. Alcohol-related acute atrial fibrillation.
A case-control study and review of 40 patients. Arch Intern Med.
1985;145:830–833.
45. Nguyen TN, Friedman HS, Mokraoui AM. Effects of alcohol on experimental atrial fibrillation. Alcohol Clin Exp Res. 1987;11:474–476.
46. Kostis JB, Goodkind MJ, Skvaza H, Gerber NH Jr, Kuo PT. Effect of alcohol
on the atrial fibrillation threshold in dogs. Angiology. 1977;28:583–587.
AEP Vol. 17, No. 5S
May 2007: S8–S12
53. Hylek EM, Heiman H, Skates SJ, Sheehan MA, Singer DE. Acetaminophen and other risk factors for excessive warfarin anticoagulation.
JAMA. 1998;279:657–662.
54. Fihn SD, McDonell M, Martin D, Henikoff J, Vermes D, Kent D, et al.
Risk factors for complications of chronic anticoagulation. A multicenter
study. Warfarin Optimized Outpatient Follow-up Study Group. Ann
Intern Med. 1993;118:511–520.
55. Fabsitz RR, Sidawy AN, Go O, Lee ET, Welty TK, Devereux RB, et al.
Prevalence of peripheral arterial disease and associated risk factors in
American Indians: the Strong Heart Study. Am J Epidemiol. 1999;
149:330–338.
56. Vliegenthart R, Geleijnse JM, Hofman A, Meijer WT, van Rooij FJ, Grobbee DE, et al. Alcohol consumption and risk of peripheral arterial disease:
the Rotterdam study. Am J Epidemiol. 2002;155:332–338.
57. Jepson RG, Fowkes FG, Donnan PT, Housley E. Alcohol intake as a
risk factor for peripheral arterial disease in the general population in the
Edinburgh Artery Study. Eur J Epidemiol. 1995;11:9–14.
47. Frost L, Vestergaard P. Alcohol and risk of atrial fibrillation or flutter:
a cohort study. Arch Intern Med. 2004;164:1993–1998.
58. Iribarren C, Sidney S, Sternfeld B, Browner WS. Calcification of the aortic
arch: risk factors and association with coronary heart disease, stroke, and
peripheral vascular disease. JAMA. 2000;283:2810–2815.
48. Psaty BM, Manolio TA, Kuller LH, Kronmal RA, Cushman M, Fried LP,
et al. Incidence of and risk factors for atrial fibrillation in older adults.
Circulation. 1997;96:2455–2461.
59. Whiteman MC, Deary IJ, Fowkes FG. Personality and social predictors of
atherosclerotic progression: Edinburgh Artery Study. Psychosom Med.
2000;62:703–714.
49. Koskinen P, Kupari M, Leinonen H, Luomanmaki K. Alcohol and new onset atrial fibrillation: a case-control study of a current series. Br Heart J.
1987;57:468–473.
60. Camargo CA Jr, Stampfer MJ, Glynn RJ, Gaziano JM, Manson JE, Goldhaber SZ, et al. Prospective study of moderate alcohol consumption and
risk of peripheral arterial disease in US male physicians. Circulation.
1997;95:577–580.
50. Ruigomez A, Johansson S, Wallander MA, Rodriguez LA. Incidence of
chronic atrial fibrillation in general practice and its treatment pattern.
J Clin Epidemiol. 2002;55:358–363.
51. Mukamal KJ, Tolstrup JS, Friberg J, Jensen G, Grønbæk M. Alcohol consumption and risk of atrial fibrillation in men and women: the Copenhagen
City Heart Study. Circulation. 2005;112:1736–1742.
52. Weathermon R, Crabb DW. Alcohol and medication interactions. Alcohol Res Health. 1999;23:40–54.
61. Ciccarone E, Di Castelnuovo A, Salcuni M, Siani A, Giacco A, Donati
MB, et al. A high-score Mediterranean dietary pattern is associated with
a reduced risk of peripheral arterial disease in Italian patients with Type
2 diabetes. J Thromb Haemost. 2003;1:1744–1752.
62. Tornwall ME, Virtamo J, Haukka JK, Albanes D, Huttunen JK. Life-style
factors and risk for abdominal aortic aneurysm in a cohort of Finnish male
smokers. Epidemiology. 2001;12:94–100.
Confounders of the Relation between Type of Alcohol
and Cardiovascular Disease
MORTEN GRØNBÆK, MD, PHD
There have been numerous reports from epidemiologic studies showing that moderate drinkers have lower
rates of cardiovascular disease (CVD) than do those who drink heavily or not at all. A number of scientific
reports from scientists around the world suggest confounding may play a role in the reported beneficial
health effects associated with moderate drinking. Among potentially confounding variables for these reported associations are the frequency of alcohol consumption, drinking pattern (steady or binge drinking),
type of beverage, and differences in the pattern of drinking associated with different types of beverages. In
some papers, individuals who report primarily wine consumption have been shown to be at lower risk of
CVD and total mortality, and there is evidence for greater beneficial effects from more frequent, regular
drinking. However, other potential confounders include better cognitive function, higher socioeconomic
status, better subjective health, and a healthier diet, including food purchases, all of which are more common in regular drinkers and wine drinkers. Thus, the question of whether the beneficial effects of beverage
types differ, with additional benefits for wine, remains unresolved.
Ann Epidemiol 2007;17:S13–S15. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Wine, Cardiovascular Diseases, Confounding Factors (Epidemiology).
INTRODUCTION
A large number of simple prospective studies have quite
consistently shown that alcohol (ethanol) itself has cardioprotective properties and thereby contributes to the descending leg of the U-shaped relationship between alcohol
and all-cause mortality (1–4). A recent longitudinal study
has confirmed this by showing that in a randomly selected
healthy population, a change in intake over a 5-year period
implied an increased risk of cardiovascular disease (CVD)
among those who decreased their consumption, and a decreased risk among abstainers who increased their intake
of alcohol (5). The effect of alcohol seems relatively
short-term and acute when CVD is the outcome, with
much longer latency times when it comes to cancer (6).
During the past decade, some epidemiological studies
have suggested that drinkers of wine have a lower incidence
of CVD than drinkers of other types of alcohol (7–11). It has
been shown in some populations that beer, wine, and spirits
may also have a differential effect on other health outcomes
(8, 12, 13) (Fig. 1).
These differences in ‘‘effect’’ on health may be explained
by true differences; there are experimental studies to support
several plausible biological explanations. Wine contains
several components that may contribute to a possible
From the Centre for Alcohol Research, National Institute of Public
Health.
Address correspondence to: Morten Grønbæk, MD, Centre for Alcohol
Research, National Institute of Public Health, Copenhagen, Denmark.
E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
antithrombotic and anticarcinogenic effect, which may exert their action locally in parallel with the possible effect of
ethanol. Polyphenolic substances, for example, have been
shown to inhibit platelet aggregation. An alternative explanation is confounding. In many countries and cultures there
are differences between drinkers of different types of alcohol
and these differences may have confounded the results.
Drinking Patterns and Other Confounders
Among the potentially confounding variables for the reported associations between alcohol intake and health outcomes is drinking pattern, which is defined herein as the
frequency of alcohol consumption, divided into steady or
binge drinking. Both in large American and Danish cohorts
it has been shown that men who drink alcohol more frequently have a lower risk of developing CVD than those
who drink less frequently, independent of the amount of intake (14, 15). The findings are less convincing for women
(15) (Fig. 2).
Several explanations may account for a possible interaction between sex and drinking frequency. One explanation
is sex-specific drinking habits, such as drinking with meals.
It cannot be excluded that men who drink frequently are
more likely to drink with meals, which may contribute to
a greater risk reduction when compared with men with
less frequent alcohol intake.
Other potential confounders include cognitive function,
socioeconomic status, subjective health, and dietary habits,
since all of these factors have been shown to be associated
with intake of different types of beverages (16–19). In the
Copenhagen City Heart Study, conducted on a cohort of
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.004
S14
Grønbæk
CONFOUNDERS OF THE ALCOHOL-CVD RELATION
AEP Vol. 17, No. 5S
May 2007: S13–S15
wine. The strength of the association was maintained as
the total consumption of wine increased over time in that
cohort (18).
It is likely that dietary factors need more thorough attention. Some studies have found wine drinkers to have
a healthier diet than drinkers of beer or spirits, and variation
in diet associated with the preferred drink may explain why
wine seems to have an additional beneficial effect on health
outcomes. However, self-reported diet and other lifestyle
factors may themselves be underreported or overreported.
For example, intake of certain food items may be either exaggerated or understated depending on health recommendations. These misclassifications are likely to be associated
with educational level, income, and a preference for different alcoholic beverage types.
We tested this idea by studying the relationship between
the purchase of beer and wine and various other food items.
The purpose of the study was to investigate whether wine
purchasers buy healthier food items on the whole and
thereby have a healthier diet than do beer buyers. The
data were collected from sales tickets obtained from supermarkets in Denmark. The results showed that wine buyers
tended to buy more olives, fruits, vegetables, poultry, cooking oil, and low-fat products than beer buyers (20). Beer
buyers tended to buy precooked dishes, sugar, cold cuts,
chips, pork, butter, sausages, lamb, and soft drinks more often than wine buyers (Fig. 3). Generally, both beer and wine
buyers shop more frequently on weekends than on other
days, but wine buyers are more likely to purchase food items
1,8
Percent wine
of alcohol intake
1,6
0
1-30
>30
Mortality
1,4
1,2
1
0,8
0,6
0,4
0
1-7
8-21
22-35
>35
Alcohol Consumption, drinks/wk
FIGURE 1. Relative all-cause mortality in relation to intake of
total alcohol, estimated in a model which allows separate effect of
wine intake at two levels: drinkers of 1–30% wine and more than
30% wine of total alcohol intake. Relative risk is set at 1.00
among non-drinkers (less than one drink per week). The estimates were adjusted for age, sex, educational level, smoking
habits, physical activity, and body mass index. (From Grønbæk M,
Becker U, Johansen D, Gottschau A, Schnohr P, Hein HO.
Ann Intern Med 2000;133:411–419.)
Danish men, a strong dose-response relationship was recently found between IQ scores in young adulthood and
beverage preferences later in life. Intelligence seemed to
correlate strongly in a positive way to a preference for
Men
A
Frequency
Hazard ratio of CHD
1.5
p=0.02
p=0.001
p<0.001
1 2-4 5-7
1 2-4 5-7
1 2-4 5-7
7-13
14-20
p=0.57
p=0.12
p=0.01
0.5
1
2-4 5-7
Drinks/week
1-6
C
Trendtest:
Hazard ratio of CHD
p<0.001
Women
Trend test:
1.0
Days/week
Amount
B
Trendtest:
1.5
p=0.25
1
21+
1-6
D
p=0.22
p<0.001
2-4 5-7
1 2-4
7-13
5-7
1
2-4 5-7
14+
Trendtest:
p=0.002
p<0.001
p<0.001
1.0
0.5
Drinks/week 1-6 7-13 14-20 21+ 1-6 7-13 14-20 21+ 1-6 7-13 14-20 21+
Days/week
1
2-4
5-7
1-6 7-13 14+ 1-6 7-13 14+ 1-6 7-13 14+
1
2-4
5-7
FIGURE 2. Hazard ratios of coronary heart disease according to drinking frequency and amount of alcohol intake. (From Tolstrup J,
Jensen MJ, Tjonneland A, Overvad K, Mukamal KJ, Grønbæk M. BMJ 2006;332:1244–1248.)
AEP Vol. 17, No. 5S
May 2007: S13–S15
Grønbæk
CONFOUNDERS OF THE ALCOHOL-CVD RELATION
S15
2. Corrao G, Bagnardi V, Zambon A, Arico S. Exploring the dose-response
relationship between alcohol consumption and the risk of several
alcohol-related conditions: a meta-analysis. Addiction. 1999;94:1551–
1573.
3. Fuchs CS, Stampfer MJ, Colditz GA, Giovannucci EL, Manson JE, Kawachi I, et al. Alcohol consumption and mortality among women. N Engl J
Med. 1995;332:1245–1250.
4. Klatsky AL, Armstrong MA, Friedman GD. Alcohol and mortality. Ann
Intern Med. 1992;117:646–654.
5. Grønbæk M, Johansen D, Becker U, Hein HO, Schnohr P, Jensen G, et al.
Changes in alcohol intake and mortality: a longitudinal populationbased study. Epidemiology. 2004;15:222–228.
6. Leinoe EB, Hoffmann MH, Kjaersgaard E, Nielsen JD, Bergmann OJ,
Klausen TW, et al. Prediction of haemorrhage in the early stage of acute
myeloid leukaemia by flow cytometric analysis of platelet function. Br J
Haematol. 2005;128:526–532.
7. Grønbæk M, Becker U, Johansen D, Gottschau A, Schnohr P, Hein HO.
Type of alcohol consumed and mortality from all causes, coronary heart
disease, and cancer. Ann Intern Med. 2000;133:411–419.
8. Grønbæk M, Deis A, Sorensen TI, Becker U, Schnohr P, Jensen G. Mortality associated with moderate intakes of wine, beer, or spirits. Br Med J.
1995;310:1165–1169.
9. Klatsky AL, Armstrong MA. Alcoholic beverage choice and risk of coronary artery disease mortality: do red wine drinkers fare best? Am J Cardiol.
1993;71:467–469.
10. Klatsky AL, Friedman GD, Armstrong MA, Kipp H. Wine, liquor, beer
and mortality. Am J Epidemiol. 2003;158:585–595.
11. Wannamethee SG, Sharper A. Type of alcohol drink and risk of major coronary heart disease events and all-cause mortality. Am J Public Health.
1999;89:685–690.
12. Grønbæk M, Becker U, Johansen C, Tonnesen H, Jensen G, Sorensen
TIA. Population based cohort study of the association between alcohol
intake and caner of the upper digestive tract. Br Med J. 1998;317:844–
848.
13. Pedersen A, Johansen C, Grønbæk M. Relations between amount and type
of alcohol and colon and rectal cancer in a Danish population-based cohort study. Gut. 2003;52:861–867.
FIGURE 3. Purchases of other foods associated with sales of beer
or wine (based on 3.5 million receipts from Danish supermarkets).
Logistic regression: beer versus wine. (From Johansen D, Friis K,
Skovenborg E, Grønbæk M. BMJ 2006;332:519–522.; reprinted
with permission from the British Medical Journal.)
consistent with the Mediterranean diet, whereas beer buyers
tend to buy ‘‘traditional’’ food items.
In conclusion, the question of whether the beneficial effects of beverage types differ, with additional benefits for
wine, remains unresolved. It may still be the drink, but it
may also be the drinker.
REFERENCES
1. Boffetta P, Garfinkel L. Alcohol drinking and mortality among men enrolled in a American Cancer Society prospective study. Epidemiology.
1990;1:342–348.
14. Mukamal K, Conigrave M, Mittleman M, Camargo C, Stampfer MJ, Willett WC. Roles of drinking pattern and type of alcohol consumed in
coronary heart disease in men. N Engl J Med. 2003;348:109–118.
15. Tolstrup J, Jensen MJ, Tjonneland A, Overvad K, Mukamal KJ, Grønbæk
M. Prospective study of alcohol drinking patterns and coronary heart
disease in women and men. Br Med J. 2006;332:1244–1248.
16. Barefoot JC, Grønbæk M, Feaganes JR, McPherson RS, Williams RB, Siegler IC. Alcohol beverage preference, diet, and health habits in the UNC
Alumni Heart Study. Am J Clin Nutr. 2002;76:466–472.
17. Grønbæk M, Mortensen E, Mygind K, Andersen A, Becker U, Gluud C.
Beer, wine, spirits and subjective health. J Epidemiol Community Health.
1999;53:721–724.
18. Frost L, Vestergaard P, Mosekilde L, Mortensen LS. Trends in incidence
and mortality in the hospital diagnosis of atrial fibrillation or flutter in
Denmark, 1980–1999. Int J Cardiol. 2005;103:78–84.
19. Tjonneland A, Grønbæk M, Stripp C, Overvad K. Wine intake and diet in
a random sample of 48763 Danish men and women. Am J Clin Nutr.
1999;69:49–54.
20. Johansen D, Friis K, Skovenborg E, Grønbæk M. Food buying habits of
people who buy wine or beer: cross sectional study. BMJ. 2006;332:519–
522.
Moderate Alcohol Use and Reduced Mortality Risk: Systematic
Error in Prospective Studies and New Hypotheses
KAYE MIDDLETON FILLMORE, PHD, TIM STOCKWELL, PHD,
TANYA CHIKRITZHS, PHD, ALAN BOSTROM, PHD,
AND WILLIAM KERR, PHD
We have provided recent evidence suggesting that a systematic error may be operating in prospective
epidemiological mortality studies that have reported ‘‘light’’ or ‘‘moderate’’ regular use of alcohol to be
‘‘protective’’ against coronary heart disease. Using meta-analysis as a research tool, a hypothesis first suggested by Shaper and colleagues was tested. Shaper et al suggested that people decrease their alcohol consumption as they age and become ill or frail or increase use of medications, some people abstaining from
alcohol altogether. If these people are included in the abstainer category in prospective studies, it is reasoned that it is not the absence of alcohol elevating their risk for coronary heart disease (CHD) but, rather,
their ill health. Our meta-analytic results indicate that the few studies without this error (i.e., those that did
not contaminate the abstainer category with occasional or former drinkers) show abstainers and ‘‘light’’ or
‘‘moderate’’ drinkers to be at equal risk for all-cause and CHD mortality. We explore the history of this
hypothesis, examine challenges to our meta-analysis, and discuss options for future research.
Ann Epidemiol 2007;17:S16–S23. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Coronary Heart Disease, Mortality, Meta-analysis, Prospective Studies, Systematic Bias.
INTRODUCTION
From the University of California, San Francisco (K.M.F., A.B.); Centre
for Addictions Research of British Columbia, University of Victoria, Victoria, Canada (T.S.); National Drug Research Institute, Curtin University,
Perth, Australia (T.C.); and Alcohol Research Group, Berkeley, CA (W.K.).
Address correspondence to: Kaye Middleton Fillmore, 1310 Brewster
Dr, El Cerrito, CA 94530. Tel.: (510) 233-8368. E-mail: [email protected].
edu.
Disclosure: The majority of funding for research performed by Dr. Fillmore has been derived from the National Institutes of Health (NIAAA).
She has received a minor amount of seed money from NordAN, a collection
of Scandinavian groups interested in the control of the accessibility of alcohol, and a minor amount of money from the International Center for Alcohol Policies to support an in-house paper on the nature of contemporary
alcohol-related research and has received travel expenses from the same
group at an earlier time. She has consulted for NIAAA and for the World
Health Organization (WHO, Geneva and Europe). Dr. Stockwell periodically conducts consulting work for WHO and Health Canada on alcohol
and other drug research issues. He is in receipt of funding from the Centre
for Addictions Research of British Columbia, the British Columbia Ministry of Health, WHO, and the Canadian Institutes for Health Research. He
has previously received travel expenses from the International Center for
Alcohol Policies but has not received personal fees or research funds
from alcohol or tobacco manufacturers or from pharmaceutical companies.
Dr. Chikritzhs has received all of her research funding from the National
Drug Strategy Commonwealth, Department of Health and Aging, Australia, through competitive grants with no links to the alcohol beverage industry. She has performed minor consulting to the health department in
Western Australia. Dr. Bostrom has received the majority of his funding
from the NIAAA and has performed statistical consulting for pharmaceutical companies and other industries, none connected with the alcohol beverage industry. The majority of work performed by Dr. Kerr has been
derived from the NIAAA.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
The objectives of our paper consist of, first, providing a brief
summary of the history of research efforts surrounding
a hypothesis originally articulated by Shaper, Wannamethee, and Walker in 1988 (1), who hypothesized that
a systematic misclassification error was present in most prospective studies assessing associations between alcohol use
and coronary heart disease (CHD). This history laid the
groundwork for testing the hypothesis using a meta-analysis
of prospective studies that reported the associations between
alcohol use and mortality risk from all causes and CHD (2).
Second, challenges to the findings from this meta-analysis
are evaluated. Third, options for future research are
discussed.
BRIEF HISTORY OF THE SHAPER,
WANNAMETHEE, AND WALKER HYPOTHESIS
Shaper and colleagues suggested that the error of including
persons terminating or decreasing their alcohol consumption to very occasional drinking in the abstainer category
biased the findings toward making drinkers appear to be
less vulnerable to CHD and abstainers more vulnerable in
prospective studies. As people age and become ill or frail
or increase use of medications, their alcohol consumption
decreases, some abstaining altogether. If these people are
included in the abstainer category, then it is not the absence
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.005
AEP Vol. 17, No. 5S
May 2007: S16–S23
Selected Abbreviations and Acronyms
CHD Z coronary heart disease
HDL Z high-density lipoprotein
RCT Z randomized clinical trial
of alcohol that is elevating their risk for CHD but, rather,
their compromised health.
Although meta-analyses appraising alcohol use and CHD
morbidity and/or mortality have been performed in the past
(3–8), our meta-analysis (2) is the second to explicitly
address the hypothesis (the first meta-analysis [7] utilized
inclusion criteria for studies without error and differed
radically from our own). Fifty-four prospective studies evaluating alcohol’s association with all-cause mortality (35 prospective studies evaluating CHD mortality) were included.
A systematic misclassification error was committed by
including as ‘‘abstainers’’ many people who had reduced or
stopped drinking, a phenomenon associated with aging
and ill health. The studies judged to be error free found no
significant all-cause or cardiac protection, suggesting that
the cardiac protection afforded by alcohol may have been
overestimated. Our results do not prove the error but suggest
that the protective effect of alcohol for CHD may have been
exaggerated in most epidemiological studies to date and the
analysis may reopen the debate in this domain.
The history of the Shaper et al hypothesis stretches back
for a quarter of a century. In 1981, two prospective studies
(9, 10) suggested that ‘‘moderate’’ drinking was associated
with the reduction of incidence of CHD. These two publications drew considerable scientific and public attention,
although earlier studies had reported similar findings.
Many prospective and case-control studies followed, confirming what became known as the ‘‘protective effect.’’ Biological studies demonstrated plausible mechanisms whereby
‘‘moderate’’ doses of alcohol will affect the clotting processes, will increase the protective fraction of cholesterol
(high-density lipoprotein [HDL] cholesterol), and may
increase elasticity of blood vessels (e.g., 11, 12) in contrast
to the positive linear association with heavier alcohol use
and blood pressure. However, none of the studies assessing
plausible biological mechanisms tell us whether alcohol
necessarily is beneficial, even if possible. This directs scientific attention to the importance of strong evidence from
epidemiological studies to positively determine if the lower
risk of ‘‘light’’ to ‘‘moderate’’ drinkers is actually caused by
the theoretical benefits derived from laboratory research
versus the possibility that the lower risk is caused primarily
by confounding variables or error. It should also direct
researchers to more precisely measure levels and patterns
of consumption consistent with different proposed biological mechanisms.
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
S17
There have been at least two major critiques of the
epidemiological studies contributing to the protective effect
finding. First, considerable evidence suggests that healthy
behaviors tend to cluster in the group of ‘‘protected’’
drinkers (11, 13–18) and no amount of statistical controls
in prospective studies can eliminate it. We acknowledge
this problem but do not pursue it herein.
Second is the Shaper et al hypothesis. The hypothesis
was taken seriously in the relevant epidemiological community. Some analysts strove to correct for the error. A deductive meta-analysis (7) utilized seven studies to suggest
that the hypothesis had been eliminated. All but one
was with the abstainer misclassification error according
to our definition of it. By 1996, it could be claimed that
the hypothesis had been eliminated (16), citing five prospective studies and one case-control study. Each of these
prospective studies contains one or both of the abstainer
errors that we evaluate in our meta-analysis. In light of
this history, it appeared to us that conclusions from both
of these publications did not utilize the rigor we would
eventually employ in our efforts to define the abstainer
error.
Although the Shaper et al hypothesis was seemingly
laid to rest by the mid-1990s, the medical epidemiological
literature had all but ignored two major scientific literatures existing somewhat outside its domain, both pertinent
to it.
The first supports the Shaper et al hypothesis. Numerous cross-sectional and longitudinal studies (some of the
latter with multiple measurement points and in populations characterized with greater longevity) attest that as
people progress into late middle and old age, their consumption of alcohol declines in tandem with ill health,
frailty, dementia, and/or use of medications (19–33). Longitudinal studies (including those with more than two
measurements) found that many people shift over time between complete abstinence and occasional drinking (34–
36), alcohol use declines with advancing age (evidence
even from the medical epidemiological literature itself
[37]), and the alleged protective effect of alcohol is reduced when accounting for variation in drinking patterns
over time (38).
The second consists of efforts made in alcohol-related science reflecting more than 50 years of struggle to accurately
describe the complex and sometimes elusive patterns of
drinking, including variations in these patterns over time
(39, 40). These efforts emanated from knowledge that drinking patterns vary across and within populations and that, in
any one individual, drinking patterns may be complex, fluctuating over the life course. In addition, it was early understood that specific types of drinking patterns are closely
related to specific disease outcomes. Although some recent
studies in the medical epidemiological literature have
S18
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
sought to incorporate variations in drinking patterns over
time into their studies (41),* most deal with absolute volume of drinking and some are highly limited in measurement (e.g., measuring drinking solely with respect to its
frequency or its quantity per occasion or limiting the time
frame of assessment to several days or weeks), thereby containing extremely imprecise measurement and possibly
masking potential error. The cause of this, in part, is owed
to the fact that many of these studies were not originally
or specifically designed to investigate the issue at hand.
Hence the alcohol-related findings were published as an afterthought, containing not only imprecise measurement of
a difficult-to-measure behavior but also insufficient information on probable confounders (42).
It is in these contexts that our research team felt it
worthwhile to evaluate the hypothesis of Shaper et al in
a cross-study approach because (a) it had not been adequately tested; (b) other literatures, often ignored by medical epidemiologists, supported that hypothesis; and (c) the
great majority of medical epidemiological studies had not
integrated a body of knowledge pointing to the necessity
of utilizing crisp operational definitions and measurement
instruments that would accurately reflect the changing nature of drinking patterns over time among individual
respondents.
ARGUMENTS CHALLENGING OUR
META-ANALYSIS FINDINGS
We turn attention to what we perceive to be arguments
challenging our analyses.
Classification of the Relevant Studies
Two errors were evaluated: the inclusion of former drinkers
and occasional drinkers in the abstainer category. Studies
without either error did not show abstainers to be at higher
risk of all-cause mortality (n Z 7 studies) and CHD mortality (n Z 2 studies) mortality than were ‘‘light’’ or ‘‘moderate’’ drinkers. Studies with both errors (26 [all-cause
mortality studies]; 25 [CHD mortality studies]) showed
what had been repeatedly reported in this literature:
a J-shape for all-cause mortality and a negative linear shape
for CHD. Studies with only the former drinker misclassification error (21 [all-cause mortality studies] and 8 [CHD mortality studies]) showed a J-shape for all-cause mortality and
a nonsignificant negative linear association for CHD
mortality. Where evaluated, former drinkers in studies
only partially or fully without error tended to be at higher
*However, in this particular case the baseline data consisted of a food frequency questionnaire, and the categorization of drinking committed the
error of including occasional drinkers with abstainers.
AEP Vol. 17, No. 5S
May 2007: S16–S23
mortality risk than long-term abstainers. Occasional
drinkers appeared to be more like ‘‘light’’ drinkers, something for which there is no plausible biological mechanism.
Our operational definitions were rigorous in an effort to
isolate studies that were error-free. Inattention to designating or excluding former drinkers clearly indicated error
(Table 1 [43–56]). Some studies discriminated former
drinkers and occasional drinkers from long-term abstainers
in their questionnaires but combined these groups with
abstainers in their models. Other analysts made efforts to
designate former or occasional drinkers as separate groups
or exclude them altogether from their models. However,
because of question wording or sequence, former drinkers
were judged to have only been partially dealt with in
such studies. For example, a difficult study to categorize, illustrating the difficulties and ambiguities in measurement
of drinking behavior, was performed by Klatsky (16).
Although considerable effort was made to achieve
a ‘‘pure’’ abstainer category, abstainers were defined, in
part, as never or almost never drinking. ‘‘Almost never’’
was regarded by our research team as highly subjective,
suggesting that very infrequent drinkers might have been
included in the abstainer group.
The studies were coded to the presence or absence of
these errors by two of the investigators. When there was disagreement in coding, the entire research team evaluated the
studies and consensus was reached. Additionally, studies in
question were assigned to independent analyses. The cases
TABLE 1. Examples of factors alerting us to potential
abstainer error
Factors alerting us to error
Inattention to classify or exclude
former drinkers
Abstainers were coded to include
‘‘lifetime abstainers, occasional
drinkers, former drinkers and
possibly men not admitting to
drinking alcohol’’
Limited time frame for assessing
abstinence (e.g. 24- to 48-hour
recall, 3-day recall, 1-week recall, 1month recall, number of drinks in
last 3 days, number of drinks in last
month)
Daily drinking assessment (e.g., daily
drinking in a week or number of
drinks a day or daily drinking)
Use of terms ‘‘almost never drink’’ or
‘‘rarely/never drink’’ or ‘‘never or
less than once a month’’
Former drinkers assigned a category
that resulted only from men who
volunteered the information
Probable error
Former drinkers misclassified
(e.g., 9, 12, 43–51)
Infrequent and former drinkers
misclassified (e.g., 52)
Infrequent and former drinkers
misclassified (e.g., 10, 53)
Infrequent and former drinkers
misclassified (e.g., 54)
Infrequent and former drinkers
misclassified (e.g., 12, 55, 56)
Former drinkers misclassified
(e.g., 45)
AEP Vol. 17, No. 5S
May 2007: S16–S23
in which this occurred, more often than not, were those that
made an effort to be error free but, because of the questionnaire content, most probably did not completely exclude either former drinkers and/or occasional drinkers from the
abstainer category.
Only 7 studies for all-cause and 2 studies for CHD mortality were judged error free. Despite these small numbers, the
fact that the analysis was solely hypothesis driven attests to
the strong possibility that the nonsignificance of findings
among the reduced number of error-free studies was due to
the absence of error, rather than to random data dredging.
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
S19
cross-study heterogeneity but without complete success.
This suggests to us that there are other errors or confounding
in these studies that should be explored and that new
hypotheses are warranted.
WHERE DO WE GO FROM HERE?
An argument might be advanced that a number of valid
competing hypotheses could eliminate our finding, given
that error was operating in the majority of these studies. It
took us, literally, a number of years to collect data from
the analysts who had performed the studies because quite often these data were simply not reported.* We tested for multiple variables. In some cases, these variables altered the
curves in the various strata designating error or lack of it,
but our hypothesis was robust to these tests.
Noteworthy is that all meta-analyses to date (our own included) find heterogeneity in cross-study results. Although
our testing of competing hypotheses sometimes reduced
the amount of heterogeneity, it was not eliminated. Utilizing good hypotheses, some analysts (5, 57)dourselves
includeddhave gone to extraordinary lengths to reduce
Our analyses have a number of weaknesses. However, we
believe the major weakness is reflected in the nature of
the studies themselves, primarily the dominance of inadequate and imprecise measurement of the major explanatory
variable. Quite obviously, better study designs with wellarticulated questions describing drinking patterns over
long periods are in order. As well, the morbidity and casecontrol literatures in this domain require careful scrutiny
for measurement error, and attention should be paid to other
disease entities that have reported abstainers to be at higher
risk than drinkers.
Still other issues should be addressed. Klatsky (58)
states that ‘‘.the approximately 10% lower total mortality
risk is not large enough to completely preclude the possibility of indirect explanation.’’ He suggests that potential
genetic or environmental traits, yet unknown, may operate
to account for the increased risk among nondrinkers. We
are in agreement that there may be other explanations
for the increased risk of abstainers, including the methodological issues we have addressed in our published paper as
well as serious problems of confounding. It should not be
forgotten that epidemiology deals with crude approximations, beset by confoundingdoften unmeasureddespecially when single estimates or limited measurements are
used.
The observation that most of the protection afforded
by alcohol operates almost solely for CHD, compared
with other diseases, is a critical issue in alcohol epidemiology that should be systematically addressed.y It may be
stated in the form of questions: Why is it that statistically
significant J-shape associations are obtained for some diseases, whereas significant linear positive associations are
found for others? Is this due solely to disease specificity
(i.e., the expected nature of the disease in association
with alcohol use) or is it due to measurement error or confounding? Most scientific exercises opt for the first explanation, and, in the tradition of medicine, the juncture of
epidemiological observation (attempting to best resolve
confounding and error) and laboratory affirmation of plausible mechanisms for a given disease is where the story
ends. We posit a competing hypothesisdexplicated in
*In several cases, the analysts were deceased or had left the field and no
longer had the data set in their possession. In two cases, the analysts refused
to answer our questions.
yNoteworthy is that alcohol has also been reported to be ‘‘protective’’ for
other conditions (e.g., cognitive function, ischemic stroke, and even some
cancers).
Use of General Population Studies, Including Younger
Adults
Our published study reported results from all studies utilized,
regardless of age of measurement. It should be understood
that the studies in this research domain have more differences than similarities (as seen in Table A2 of our published
paper [2]). In fact, these dissimilarities have often been
utilized as an argument to advance the notion that consensus in study results has been found despite cross-study
differences.
Studies ranged in ages of measurement. Because the protective effect is thought to operate among middle-aged and
older adults, we controlled for respondents’ ages in the represented studies, among other pertinent variables. Additionally, we conducted separate analyses of those studies
with samples of respondents 35 years of age and older at
initial measurement point. The latter did not yield different
results compared with the larger sample of studies.
Competing Hypotheses
S20
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
AEP Vol. 17, No. 5S
May 2007: S16–S23
TABLE 2. Illustrations of a competing hypothesis: Logic and preliminary evidence
The hypothesis: The difference in shape of associations of alcohol with different disease outcomes is a function of measurement error mediated through study
design. The hypothesis is based on the following:
1. J-shaped associations are more often observed for alcohol-related outcomes in diseases dominated by prospective studies, whereas positive linear associations are more often
observed for outcomes dominated by case-control studies. Utilizing a meta-analysis of multiple disease outcomes (59), Table 2 and Figs 1 and 2 illustrate that
significant linear associations are typically found for diseases dominated by case-control designs, whereas J-shaped associations are typically found for diseases
dominated by prospective study designs.
2. Prospective studies evaluating alcohol-related disease outcomes are more likely to contain the abstainer error than case-control studies. Abstainer error is less likely to occur
in case control studies because greater attention is paid to accurate assessment of exposure, that is, greater attention is paid to past exposure and the duration of exposure,
particularly for disease outcomes taking many years to develop. Corrao et al. (59) reported characteristics of the studies for a number of disease outcomes utilized in
their meta-analysis. The clear majority (67%) of case-control studies assessed drinking before diagnosis or assessed lifetime consumption. The clear majority
(68%) of prospective studies assessed current consumption (as opposed to consumption in the past or all consumption during follow-up). Assuming that the
systematic abstainer error is valid, these cross-design differences suggest that it is more likely to be operating in prospective than in case-control studies (i.e.,
because case-control studies are more attentive to duration of exposure, they tend to self-correct for this error). (It is noteworthy that Corrao, Bagnardi,
Zambon, and Arico (57) found design effects [case-control vs. prospective study] in their meta-analysis for some disease outcomes but did not interpret them
via the mechanism of abstainer error.)
The hypothesis has promise because it is supported in at least two additional domains. First, our preliminary observations in a sample of prospective studies
evaluating both forms of cancer risk and CHD risk in the same study suggest that when abstainers are found to be at higher CHD risk compared with varieties
of drinkers, J-shaped or negative linear curves are also likely to be found for some forms of cancer (e.g., Boffetta and Garfinkel [54] for all cancers combined;
Semenciw et al [61] for prostate cancer and all cancers combined; Goldberg et al [60] for all cancers combined; Thun et al [51] for all other cancers, excluding
alcohol-related cancers and colorectal cancer, among men and colorectal cancer and all other cancers, excluding alcohol-related cancers, among women;
Doll et al [45] for other cancers, excluding lung and large bowel). Second, another meta-analysis supports the proposition. The meta-analysis of alcohol and
breast cancer by Ellison et al in 2001 (61) finds increased risk at 12 g of alcohol per day for hospital-based case-control studies (compared with cohort/
prospective studies and community case-control samples). Two large cohort studies were analyzed independently in this study for breast cancer mortality.
Both find a J-shape curve in contrast to the linear associations found by the entire pool of studies dominated by case-control design studies.
Should this hypothesis be supported, resolution of the abstainer error (most likely to occur in prospective study designs) should bring the conflicting findings
from the two designs and from multiple disease outcomes into agreement.
Table 2done that may resolve some of the contradictions
in the alcohol-related epidemiological literature: The
difference in shape of associations of alcohol with different disease outcomes is a function of measurement error
mediated through study design (see Tables 2 and 3; Figs.
1 and 2).
A further note regarding design is warranted. It has
sometimes been suggested that the epidemiological association between alcohol use and CHD can best be resolved
through randomized clinical trial (RCT)dcertainly the alcohol effect may be more precisely ascertained, particularly
in exploring whether low alcohol intake actually conveys
a health benefit. RCTs are, of course, the ‘‘gold’’ standard
of medical and clinical research and, indeed, may be an asset to the literature regarding alcohol and CHD incidence
should the period of measurement be long term and the
sample sufficiently large to properly assess both morbidity
and mortality. But use of this design faces serious ethical,
human subject, logistic, and design obstacles and challenges, and, consequently, any findings may be limited.
Additionally, observational and RCTs have been found
to differ, with nonrandomized studies showing larger effects than RCTs in some cases (56). Furthermore, results
from RCTs would probably not change recommendations
currently in existence for the general population. The considerable scientific effort and the massive financial
resources devoted to initiating an RCT may not be
warranted in view of the need to better explore ‘‘healthier’’
alternatives to cardiac protection. In any event, any RCT
should be performed free of any vested commercial
interests.
DISCUSSION
The often used term ‘‘moderate drinking’’ may be inappropriate for at least two reasons. First, its meaning varies across
culture and time. Second, the operational definitions of
‘‘moderate’’ drinking vary enormously (62) Such imprecision should be unacceptable to the scientific community.
It has been suggested that a more useful term may be
‘‘nonharmful’’* drinking and that some effort be made to
precisely qualify the boundaries of potentially beneficial
use (written communication with A. G. Shaper, April
2006).
History attests to exaggeration of both the beneficial
and the adverse effects of the use of alcohol (63, 64). In
the context of the wide variation of the perceptions and
*Some have suggested ‘‘low-risk drinking’’ as an alternative term (46),
although one author of this paper finds it to be inappropriate due to the
loaded nature of the term ‘‘risk.’’
AEP Vol. 17, No. 5S
May 2007: S16–S23
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
TABLE 3. Proportion case-control studies and prospective
studies in a meta-analysis by 12 disease types: Significant
associations finding a positive linear association versus
a J-shape association
Positive linear
association with
alcohol
Neoplastic conditions
Larynx
Esophagus
Oral cavity and
pharynx
Breast
Liver
Colon
Rectum
Nonneoplastic conditions
Chronic pancreatitis
Hemorrhagic stroke*
Cirrhosis of the liver
J-shaped association
with alcohol
Ischemic stroke*
Coronary heart
disease
Total No. of
studies
Proportion
case-control
studies
Proportion
prospective
designs
20
14
15
100
96
93
0
4
7
29
10
16
6
82
80
75
66
18
20
25
34
2
9
9
100
66
66
0
34
34
6
28
50
50
100
Adapted from Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of
alcohol consumption and the risk of 15 diseases. Prev Med. 2004;38:613–619.
*Continuous models indicate a very slight J shape with the nadir for ischemic stroke
at 15 g/d and for hemorrhagic stroke at 3 g/d, but ‘‘moderate’’ drinkers (25 g/d) were
not found to be significantly different from abstainers in the categorical models.
uses of alcohol and the rhetoric surrounding both, it
should not be surprising that ideological, political, and
economic forces and vested interests play important
roles in how alcohol is studied with respect to both
S21
benefits and harms associated with its use. In contemporary
times, these strong and sometimes exceedingly powerful
opposing forces meet over the assertion that ‘‘moderate
drinking’’ is beneficial to health because, of course, such
an assertion has massive political and economic ramifications. It is therefore incumbent for scientists in this field
to devote themselves not to proving an assertion but to
questioning it and to remain free, to the extent possible,
of identification with any of these strong influences.
Such caution is warranted among scientists when alcohol
is the topic of inquiry, regardless of whether the study is
on the biological, behavioral, population, or institutional
level.
Our own group’s position on the health benefits of alcohol results from extensive discussion among ourselves and
what seemed to us to be a herculean effort to eliminated
rather than provedthe Shaper, Wannamethee, and
Walker hypothesis (1). On the basis of the fine contributions of laboratory science demonstrating plausible and
real mechanisms for cardiac protection, our conclusion is
that alcohol (among other substances, lifestyles, and behaviors) conveys benefit to the heart. But the lot falls to
epidemiology to establish whether, in fact, human populations will benefit greatly from the use of alcohol and if they
should be advised to use the substance for medicinal purposes. With others (15), we conclude that the actual outcomes in human populations for cardiac benefit have been
exaggerated and we rely on Feinstein (65) for succinctly
stating some of the problems facing contemporary epidemiology, alcohol-related epidemiology included: ‘‘The people who struggle to understand those results can be helped
by recalling the old adage that statistics are like a bikini
bathing suit: what is revealed is interesting; what is
concealed is crucial.’’
(26.52)
(7.13)
5
Liver cirrhosis
4.5
Oral cavity & pharynx cancer
4
3.5
RR
3
2.5
2
(6.45)
Hemorrhagic
stroke
Larynx cancer
Esoph.cancer
Chronic
pancreatitis
Breast cancer
Liver cancer
Rectum cancer
1.5
Colon
cancer
1
0.5
0
Abstainers
25 g/day
50 g/day
100 g/day
Alcohol Consumption
FIGURE 1. Significant associations of alcohol use with disease outcome. All diseases dominated by case-control studies (>66%) of the
studies evaluated. (From Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15
diseases. Prev Med. 2004;38:613–619.)
S22
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
AEP Vol. 17, No. 5S
May 2007: S16–S23
4.5
4
3.5
RR
3
2.5
Ischemic stroke
2
1.5
1
Conorary heart disease
0.5
0
Abstainers
25 g/day
50 g/day
100 g/day
Alcohol Consumption
FIGURE 2. Significant associations of alcohol use with disease outcome. All diseases dominated by prospective studies (>50%) of the
studies evaluated. (From Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15
diseases. Prev Med. 2004;38:613–619 [essential hypertension also evaluated but, in essence, only one study was included in the evaluation,
insufficient for comparison in our study].)
We are indebted to A.G. Shaper for his crisp thinking and comments regarding the matters discussed in this paper and for his persistence in reminding those engaged in alcohol and disease outcome research that
competing hypotheses should be pursued with diligence.
REFERENCES
12. Rimm EB, Giovannucci EL, Willett WC, Colditz GA, Ascherio A, Rosner B,
et al. Prospective study of alcohol consumption and risk of coronary
disease in men. Lancet. 1991;338:464–468.
13. Andreasson S. Alcohol and J-shaped curves. Alcohol Clin Exp Res.
1998;22(Suppl):359S.
14. Fillmore KM, Golding JM, Graves KL, Kniep S, Leino EV, Romelsjo A,
et al. Alcohol consumption and mortality. I. Characteristics of drinking
groups. Addiction. 1998;93:183–203.
1. Shaper AG, Wannamethee G, Walker M. Alcohol and mortality in British men: Explaining the U-shaped curve. Lancet. 1988;2:1267–1273.
15. Jackson R, Broad J, Connor J, Wells S. Alcohol and ischaemic heart
disease: probably no free lunch. Lancet. 2005;366:1911–1912.
2. Fillmore KM, Kerr WC, Stockwell T, Chikritzhs T, Bostrom A. Moderate
alcohol use and reduced mortality risk: Systematic error in prospective
studies. Addiction Res Theory. 2006;14:101–132.
16. Klatsky A. Alcohol, coronary disease, and hypertension. Annu Rev Med.
1996;47:149–160.
3. Corrao G, Rubbiati L, Bagnardi V, Zambon A, Poikolainen K. Alcohol
and coronary heart disease: A meta-analysis. Addiction. 2000;95:1505–
1513.
4. English DR, Holman CDJ, Milne E, Winter MG, Hulse GK, Codde JP,
et al. The Quantification of Drug Caused Morbidity and Mortality in Australia. Canberra: Commonwealth Department of Human Services and
Health; 1995.
5. Gmel G, Gutjahr E, Rehm J. How stable is the risk curve between alcohol
and all-cause mortality and what factors influence the shape? A precisionweighted hierarchical meta-analysis. Eur J Epidemiol. 2003;18:631–642.
6. Holman CDJ, English DR, Milne E, Winter MG. Meta-analysis of alcohol
and all-cause mortality: A validation of NHMRC recommendations. Med J
Aust. 1996;164:141–145.
7. Maclure M. Demonstration of deductive meta-analysis: Ethanol intake and
risk of myocardial infarction. Epidemiol Rev. 1993;15:328–335.
8. Rehm J, Gutjahr E, Gmel G. Alcohol and all-cause mortality: A pooled
analysis. Contemp Drug Probl. 2001;28:337–361.
9. Klatsky A, Friedman GD, Siegelaub AB. Alcohol and mortality: A tenyear Kaiser-Permanente experiment. Ann Intern Med. 1981;95:139–145.
17. Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK,
et al. Cardiovascular risk factors and confounders among nondrinking
and moderate-drinking U.S. adults. Am J Prev Med. 2005;28:369–
373.
18. Shaper AG, Wannamethee SG. Epidemiological confounder in the
relationship between alcohol and cardiovascular disease. In: Paoletti R,
Klatsky A, Poli A, et al., eds. Moderate alcohol consumption and cardiovascular disease. Dordrecht (Netherlands): Kluwer Academic Publishers;
2000:105–112.
19. Adams WL, Garry PJ, Rhyne R, Hunt WC, Goodwin JS. Alcohol intake in
the healthy elderly. Changes with age in a cross-sectional and longitudinal
study. J Am Geriatr Soc. 1990;38:211–216.
20. Cahalan D. Problem drinking and short-term change: a community [dissertation]. Washington (DC): The George Washington University;
1968.
21. Cryer PC, Jenkins LM, Cook AC, Ditchburn JS, Harris CK, Davis AR,
et al. The use of acute and preventative medical services by a general
population: relationship to alcohol consumption. Addiction. 1999;94:
1523–1532.
10. Marmot MG, Rose G, Shipley MJ, Thomas BJ. Alcohol and mortality: A
U-shaped curve. Lancet. 1981;14:580–583.
22. Cryer PC, Saunders J, Jenkins LM, Neale H, Cook AC, Peters TJ. Clusters
within a general population of alcohol abstainers. Int J Epidemiol. 2001;
30:756–765.
11. Fagrell B, De Faire U, Bondy S, Criqui M, Gaziano M, Gronbaek M, et al.
The effects of light to moderate drinking on cardiovascular diseases. J Intern Med. 1999;246:331–340.
23. Dent O, Grayson D, Waite L, Cullen JS, Creasey H, Broe GA. A longitudinal study of alcohol consumption and functional disability in a community sample of older people. Australas J Ageing. 2000;19:185–189.
AEP Vol. 17, No. 5S
May 2007: S16–S23
Fillmore et al.
ERROR IN STUDIES OF ALCOHOL AND MORTALITY
S23
24. Dent OF, Sulway MR, Broe GA, Creasey H, Kos SC, Jorm AF, et al. Alcohol
consumption and cognitive performance in a random sample of Australian
soldiers who served in the Second World War. BMJ. 1997;314:1655.
46. Friedman LA, Kimball AW. Coronary heart disease mortality and alcohol consumption in Framingham. Am J Epidemiol. 1986;124:481–
489.
25. Graham K, Schmidt G. The effects of drinking on health of older adults.
Am J Drug Alcohol Abuse. 1988;24:465–481.
47. Fuchs CD, Stampfer MJ, Colditz GA, Giovannucci EL, Manson JE, Kawachi I, et al. Alcohol consumption and mortality among women. N Engl J
Med. 1995;332:1245–1250.
26. Gronbaek M, Becker U, Johansen D, Gottschau A, Schnohr P, Jensen G,
et al. Type of alcohol consumed and mortality from all causes, coronary
heart disease and cancer. Ann Intern Med. 2000;133:411–419.
27. Knupfer G, Room R. Abstainers in a metropolitan community. Q J Stud
Alcohol. 1970;31:108–131.
28. Kunz JL. Alcohol use and reported visits to health professionals: An
exploratory study. J Stud Alcohol. 1997;58:474–479.
29. Ruchlin HS. Prevalence and correlates of alcohol use among older adults.
Prev Med. 1997;26:651–657.
30. Seppa K, Pitkajarvi T, Sillanaukee P. Alcohol consumption profile by time
in middle-aged men: a longitudinal study based on three different diagnostic instruments. Alcohol Alcohol. 1999;34:65–70.
31. Shaper AG, Wannamethee G. The J-shaped curve and changes in drinking habit. In: Chadwick DJ, Goode JA, eds. Alcohol and cardiovascular
diseases. London: Wiley Chichester; 1998:173.
32. Stall R. Change and stability in quantity and frequency of alcohol use
among aging males: A 19-year follow up study. Br J Addict.
1986;81:537–544.
33. Zarkin GA, Bray JW, Babor TF, Higgins-Biddle JC. Alcohol drinking patterns and health care utilization in a managed care organization. Health
Serv Res. 2004;39:553–570.
34. Fillmore KM, Hartka E, Johnstone BM, Leino EV, Motoyoshi M, Temple
MT. A meta-analysis of life course variation in drinking. Br J Addict.
1991;86:1221–1267.
35. Kerr WC, Fillmore KM, Bostrom A. Stability of alcohol consumption over
time: Evidence from three longitudinal surveys from the United States.
J Stud Alcohol. 2002;63:325–333.
36. Moore AA, Gould R, Reuben DB, Greendale GA, Carter MK, Zhou K,
et al. Longitudinal patterns and predictors of alcohol consumption in
the United States. Am J Public Health. 2005;95:458–464.
37. Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: Meta-analysis of effects
on lipids and haemostatic factors. BMJ. 1999;19:1523–1528.
38. Emberson JR, Shaper AG, Wannamethee SG, Morris RW, Whincup PH.
Alcohol intake in middle age and risk of cardiovascular disease and
mortality: Accounting for intake variation over time. Am J Epidemiol.
2005;161:856–863.
39. Special issue. Proceedings: International workshop on consumption measures and models for use in policy development and evaluation. Alcohol
Clin Exp Res. 1998;22(2 Suppl):1S–81S.
40. Room R. Drinking patterns and alcohol-related social problems: Frameworks for analysis in developing societies. Drug Alcohol Rev. 1998;
17:389–398.
41. Mukamal KJ, Conigrave KM, Mittleman MA, Camargo CA, Stampfer MJ,
Willett WC, et al. Roles of drinking pattern and type of alcohol consumed
in coronary heart disease in men. N Engl J Med. 2003;348:109–118.
42. Smith GD, Phillips AN. Confounding in epidemiological studies: Why
‘‘independent’’ effects may not be all they seem. BMJ. 1992;305:757–759.
43. Britton A, Marmot M. Different measures of alcohol consumption and risk
of coronary heart disease and all-cause mortality: 11-year follow-up of the
Whitehall II Cohort Study. Addiction. 2004;99:109–116.
44. DeLabry LO, Glynn RJ, Levenson MR, Hermos JA, LoCastro JS, Vokonas
PS. Alcohol consumption and mortality in an American male population:
Recovering the U-shaped curve d Findings from the Normative Aging
Study. J Stud Alcohol. 1992;53:25–32.
45. Doll R, Peto R, Hall E, Wheatley K, Gray R. Mortality in relation to
consumption of alcohol: 13 years’ observations on male British doctors.
BMJ. 1994;309:911–918.
48. Gronbaek M, Johansen D, Becker U, Hein HO, Schnohr P, Jensen G,
et al. Changes in alcohol intake and mortality: a longitudinal populationbased study. Epidemiology. 2004;15:222–228.
49. Semenciw RM, Morrison HI, Mao Y, Johansen H, Davies JW, Wigle DT.
Major risk factors for cardiovascular disease mortality in adults: Results
from the Nutrition Canada Survey Cohort. Int J Epidemiol.
1988;12:317–324.
50. Simons LA, McCallum J, Friedlander Y, Ortiz M, Simons J. Moderate alcohol intake is associated with survival in the elderly: The Dubbo Study.
Med J Aust. 2000;173:121–124.
51. Thun MJ, Peto R, Lopez AD, Monaco JH, Henley SJ, Heath CWJ, et al.
Alcohol consumption and mortality among middle-aged and elderly U.S.
adults. N Engl J Med. 1997;337:1705–1714.
52. Hart CL, Smith GD, Hole DJ, Hawthorne VM. Alcohol consumption and
mortality from all-causes, coronary heart disease and stroke: Results from
a prospective cohort study of Scottish men with 21 years of follow up.
BMJ. 1999;318:1725–1729.
53. Keil U, Chambless LE, Doring A, Filipiak B, Stieber J. The relation of alcohol intake to coronary heart disease and all-cause mortality in a beerdrinking population. Epidemiology. 1997;8:150–156.
54. Boffetta P, Garfinkel L. Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study. Epidemiology.
1990;1:342–348.
55. Gaziano JM, Gaziano TA, Glynn RJ. Light-to-moderate alcohol consumption and mortality in the physicians’ health study enrollment cohort. J Am
Coll Cardiol. 2000;35:96–105.
56. Hoffmeister H, Schelp F-P, Mensink GBM, Deitz E, Bohning D. The relationship between alcohol consumption, health indicators and mortality in
the German population. Int J Epidemiol. 1999;28:1066–1072.
57. Corrao G, Bagnardi V, Zambon A, Arico S. Exploring the dose response
relationship between alcohol consumption and the risk of several alcoholrelated conditions: A meta-analysis. Addiction. 1999;94:1551–1571.
58. Klatsky A, Udaltsova N. Alcohol drinking and total mortality risk. Ann
Epidemiol. 2007;17:S63–S67.
59. Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of
alcohol consumption and the risk of 15 diseases. Prev Med. 2004;38:
613–619.
60. Goldberg RJ, Burchfiel CM, Reed DM, Wergowske G, Chiu D. A prospective study of the health effects of alcohol consumption in middleaged and elderly men: The Honolulu Heart Program. Circulation. 1994;
89:651–659.
61. Ellison RC, Zhang Y, McLennan CE, Rothman KJ. Exploring the relation
of alcohol consumption to risk of breast cancer. Am J Epidemiol.
2001;154:740–747.
62. Letenneur L. Moderate alcohol consumption and risk of developing dementia in the elderly: The contribution of prospective studies. Ann Epidemiol.
2007;17:S43–S45.
63. Babor TF, Rosenkrantz B. Public health, public morals and public order:
Social science and liquor control in Massachusetts, 1880–1916. In: Barrows S, Room R, Verhey J, eds. The social history of alcohol: Drinking
and culture in modern society. Berkeley (CA): Alcohol Research Group,
Medical Research Institute of San Francisco; 1987.
64. Grant M. Health claims in early British alcohol beverage advertising. In:
Barrows S, Room R, Verhey J, eds. The social history of alcohol: Drinking
and culture in modern society. Berkeley (CA): Alcohol Research Group,
Medical Research Institute of San Francisco; 1987.
65. Feinstein AR. Scientific standards in epidemiological studies of the
menace of daily life. Science. 1988;242:1257–1263.
Mechanism by which Alcohol and Wine Polyphenols Affect
Coronary Heart Disease Risk
FRANCOIS M. BOOYSE, PHD, WENSHENG PAN, MD, PHD, HERNAN E. GRENETT, PHD,
DALE A. PARKS, PHD, VICTOR M. DARLEY-USMAR, PHD, KELLEY M. BRADLEY, BS,
AND EDLUE M. TABENGWA, PHD
The reduction in coronary heart disease (CHD) from moderate alcohol intake may be mediated, in part, by
increased fibrinolysis; endothelial cell (EC)–mediated fibrinolysis should decrease acute atherothrombotic
consequences (eg, plaque rupture) of myocardial infarction (MI). We have shown that alcohol and individual polyphenols modulate EC fibrinolytic protein (t-PA, u-PA, PAI-1, u-PAR and Annexin-II) expression
at the cellular, molecular, and gene levels to sustain increased fibrinolytic activity. Herein we describe the
sequence of molecular events by which EC t-PA expression is increased through common activation of p38
MAPK signaling. Up-regulation of t-PA gene transcription, through specific alcohol and polyphenol transcription factor binding sites in the t-PA promoter, results in increased in vitro fibrinolysis and in vivo clot
lytic activity (using real-time fluorescence [Fl] imaging of Cy5.5-labeled fibrin clot lysis in a mouse model).
Fl-labeled fibrin clots injected into untreated C56Bl/6 wild-type control mice are lysed in approximately 2
hours and clot lytic rates significantly increased in mice treated with either alcohol, catechins, or quercetin
(4–6 weeks). Fl-labeled clot lysis in ApoE knock-out mice (atherosclerosis model) showed impaired in vivo
clot lysis that was ‘‘normalized’’ to wild-type control levels by treatment with alcohol, catechin, or quercetin
for 6 to 8 weeks.
Ann Epidemiol 2007;17:S24–S31. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Coronary Heart Disease Prevention, Fibrinolysis, Endothelial Cell.
INTRODUCTION
Cardiovascular disease, in particular coronary heart disease
(CHD), and the ensuing acute atherothrombotic complications resulting in myocardial infarction (MI), is the leading
cause of death among adults in the Western world (1–5). Epidemiological studies have shown that light-to-moderate
drinkers of alcoholic beverages or wine, particularly red
wine (1 to 4 drinks per day), have significantly lower mortality rates (20%–40%) than nondrinkers or heavier drinkers
(6–10). These reduced rates appear to be due largely to a reduced risk for CHD-related mortality attributed, at least in
part, to an attenuation in acute CHD-related vascular
thrombotic and atherothrombotic complications (11–14).
CHD is a complex multifactorial disease, yet significant insights have been gained in our understanding of the etiology
underlying the initiation, progression, and acute consequences of CHD. The pathogenesis of CHD and acute atherothrombotic complications resulting in MI involves the
From the University of Alabama at Birmingham.
Address correspondence to: Edlue M. Tabengwa, PhD, University of
Alabama at Birmingham, 845 19th St South, BBRB 809, Birmingham,
AL 35294-2170. Tel.: (205) 934-4296; fax: (205) 975-7121. E-mail: tabengwa@
uab.edu.
This work has been supported, in part, by Grant HL070610 (National
Heart, Lung and Blood Institute) and Grant AA 11674 (National Institute
on Alcohol Abuse and Alcoholism).
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
complex interplay between multiple altered cellular and
molecular functions. Functions are typically affected at the
level of the heart muscle (cardiomyocytes), blood vessels
(endothelial cells [ECs] and smooth muscle cells [SMC]),
blood cells (platelets and monocytes), and plasma components (lipoproteins, clotting factors, fibrinogen, etc.). Consequently, any systemic factors (such as alcohol or wine
components) that will reduce, minimize, or restore these
altered CHD-related functions will reduce the overall risk
for CHD-related mortality.
Alcohol affects a diverse array of vascular and biochemical functions that have potential cardioprotective benefits.
Induced changes in lipid profile, including increases in
high-density lipoprotein (HDL) cholesterol and its subtypes
(HDL2, HDL3), are thought to represent a major mechanism to reduce the risk for CHD-related mortality (15–
17). However, other changes in vascular, myocardial,
hemostatic, and EC functions may be equally important
in collectively contributing to reducing this overall risk.
Identified functions include decreased platelet aggregation/function (18, 19); decreased myocardial ischemiareperfusion injury (20, 21); increased EC-dependent
vasorelaxation (22); simultaneous activation of EC antiapoptotic and proapoptotic pathways (23); decreased plasma
levels of factor VII (24) and fibrinogen (17, 25); increased
fibrinolysis (26–31); increased levels of atrial natriuretic
peptide (32); and inhibition of initiation/progression
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.006
AEP Vol. 17, No. 5S
May 2007: S24–S31
Selected Abbreviations and Acronyms
CHD Z coronary heart disease
EC Z endothelial cell
EMSA Z electrophoretic mobility shift assay
FDP Z fibrin degradation products
Fl Z fluorescence imaging
HDL Z high-density lipoprotein
LDL Z low-density lipoprotein
MI Z myocardial infarction
PMG Z plasminogen
SMC Z smooth muscle cell
TF Z tissue factor
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
S25
proliferation (50, 51); and increased fibrinolysis and up-regulation of fibrinolytic protein gene transcription in cultured
human ECs (52, 53).
Fig. 1 schematically summarizes described mechanisms
by which moderate alcohol and/or wine polyphenols can affect different vascular, cellular, and hemostatic functions
that likely act in combination or perhaps synergistically to
provide potential protective benefits that may reduce overall risk for CHD.
Normal hemostasis
of atherosclerotic lesion formation in hyperlipidemic mice
(33).
Similarly, unfractionated red wine, dealcoholized red
wine, and individual red wine components (ie, polyphenols)
have been associated with reduced CHD mortality (7, 8, 10,
34). Functions affected include increased vasorelaxation of
human and rat aortic rings (35–37); antiatherogenic effects
in apoE-deficient mice (reduced LDL oxidation/aggregation, foam cell formation, lesion progression) (38, 39); inhibition of endothelin-1 synthesis in bovine aortic ECs (40,
41); down-regulation of tissue factor (TF) gene transcription
in cultured human ECs and monocytes (42–44); reduced
thrombosis (45); reduced inflammation (46, 47); inhibition
of platelet aggregation/function (48, 49); inhibition of SMC
ECs play a major role in maintaining normal hemostasis and
vascular patency by regulating the balance between the synthesis and interaction of coagulation proteins to generate
thrombin and promote blood clotting (fibrin/thrombosis)
and fibrinolytic proteins to generate plasmin and facilitate
blood clot lysis. Under normal physiological conditions,
the amount of blood clot formed is readily degraded or lysed
by the fibrinolytic system, thereby limiting any thrombotic
complications. However, if the fibrinolytic system is overwhelmed by an acute event, such as the rupture of an unstable plaque, causing sudden exposure and release of factors
(ie, TF) leading to increased activation of the coagulation
system, this would result in the rapid formation of an atherothrombus and the ensuing potential of an MI.
FIGURE 1. Mechanisms by which moderate alcohol and/or wine polyphenols may provide cardiovascular disease protection. The central
theme of this presentation is focused on endothelial cell (EC) fibrinolysis (blood clot lysis) (highlighted by bolded arrows).
S26
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
AEP Vol. 17, No. 5S
May 2007: S24–S31
FIGURE 2. Regulation of EC fibrinolysis and maintenance of hemostasis.
Described studies will focus specifically on the sequence
of molecular mechanisms by which alcohol and principal
red wine polyphenols (catechin and quercetin) can individually modulate the expression of the EC fibrinolytic proteins
at the cellular, molecular and gene levels to sustain increased fibrinolysis (blood clot lytic or ‘‘busting’’ activity).
Components of the EC fibrinolytic system include tissuetype plasminogen activator (PA), t-PA; urokinase-type
PA, u-PA; type-1 PA inhibitor, PAI-1; receptor/binding
proteins for t-PA/plasminogen (Pmg), Annexin-II (AnnII); and u-PA receptor, u-PAR. Increased fibrinolytic
activity will be expected to decrease the risk for both early
thrombotic as well as later acute CHD-related (plaque
rupture-induced) atherothrombotic consequences of MI
and hence the overall risk for CHD-related mortality.
Regulation of fibrinolysis
Fibrinolysis requires the conversion of the inactive circulating proenzyme (Pmg) to its active two-chain plasmin form
by t-PA and u-PA (54). Plasmin will degrade fibrin/blood
clots to maintain normal hemostasis by forming fibrin degradation products (FDPs), which are subsequently cleared by
the liver. ECs are a major site of synthesis of the fibrinolytic
proteins, t-PA, u-PA and the major physiological regulator
of fibrinolysis, PAI-1 (54–56). EC-mediated fibrinolysis is
further regulated by and localized to the EC surface via specific receptors or binding proteins for t-PA and Pmg (Ann
II) (57) and u-PA (u-PAR) (58). To maintain the required
normal level of EC-mediated fibrinolytic activity
necessitates the regulated synthesis and complex multicomponent interactions of all of these different fibrinolytic components. Consequently, systemic factors that alter the
expression/interaction of one or more of these EC fibrinolytic components to decrease EC fibrinolytic activity will
increase the risk for thrombotic/atherothrombotic complications. Conversely, circulating systemic factors, such as
alcohol or wine components (polyphenols) that alter the expression/interaction of one or more of these EC fibrinolytic
components to increase fibrinolytic activity, will be expected to significantly reduce the overall risk for acute
thrombotic/atherothrombotic events and complications associated with CHD, including MI. The individual fibrinolytic components and their interactions, resulting in the
regulated expression of surface-localized EC fibrinolysis are
depicted schematically in Fig. 2.
The demonstrated ability of alcohol and polyphenols to
increase the expression and interaction of EC fibrinolytic
proteins, to increase fibrinolysis and provide protection in
vitro and in vivo (as detailed below), is remarkably consistent with epidemiological studies that have led to the suggestion that moderate alcohol consumption may mediate
additional cardioprotection by promoting fibrinolysis
through changes in the activity, level, or interaction of
one or more of the components of the fibrinolytic system
(26, 30).
Alcohol increases t-PA in bovine aortic ECs (59) and
simultaneously increases t-PA and decreases PAI-1 in cultured human ECs (60, 61). We have previously shown
that low concentrations of ethanol (0.01%–0.1%, vol/vol,
AEP Vol. 17, No. 5S
May 2007: S24–S31
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
S27
concentrations present in 2 glasses of red wine) would affect
the expression of t-PA and/or u-PA mRNA, in vivo, in
thoracic aortic endothelium. In situ hybridization analysis
demonstrated that ethanol and each of the individual
polyphenols rapidly (w 6 hours) increased both t-PA and
u-PA mRNA expression in the thoracic aortic endothelium,
as shown in Fig. 3. In situ t-PA/u-PA mRNA results were
further confirmed by reverse-transcriptase polymerase chain
reaction (RT-PCR) measurement of t-PA/u-PA mRNA
levels in adjacent tissue sections to those simultaneously analyzed by in situ hybridization. These results clearly demonstrated that ingestion of moderate levels of alcohol or
catechin or quercetin rapidly achieved sufficiently high
blood concentrations for each individual component to induce demonstrable gene activation to confirm our previous
in vitro results with cultured human ECs. Similar results
were obtained with alcohol-/polyphenol-induced increased
expression of u-PA mRNA, in vivo, in rat thoracic aortic
endothelium (data not shown).
FIGURE 3. Alcohol- and polyphenol-induced in vivo expression of t-PA mRNA in rat aortic endothelium. Top, In situ hybridization; note increased mRNA expression at endothelium
surface (dark line), compared with control; bottom, real-time
PCR analysis of t-PA, u-PA, and PAI-1 mRNA levels in adjacent
sections.
<25 mM) induces sustained (w24 hours) increased (w2- to
3-fold) surface-localized fibrinolytic activity in cultured human ECs (28, 29, 62). Induced, increased EC fibrinolytic activity is concomitant with the up-regulation of t-PA,
u-PA, u-PAR, and Ann-II antigen/activity, mRNA and
gene transcription (63–65) and simultaneous downregulation of PAI-1 antigen/mRNA and gene transcription
(66, 67). We have also shown that individual polyphenols,
catechin and quercetin, similarly induce increased surfacelocalized fibrinolytic activity in cultured human ECs and is
associated with both increased t-PA and u-PA mRNA and
gene transcription (53).
Expression of t-PA mRNA, in vivo, in rat thoracic
aorta (in situ hybridization)
These studies were carried out to verify our in vitro results on
the effects of alcohol and polyphenols on fibrinolysis and tPA/u-PA mRNA expression in cultured ECs. Initial animal
studies were carried out, using a rat model, to determine
whether direct delivery (gavage, by stomach tube) of a moderate amount of ethanol (present in 2 drinks) or individual
polyphenols (equivalent to catechin or quercetin
MAPK signaling pathways involved in alcohol- and
polyphenol-induced t-PA expression
To address the question whether alcohol and polyphenols
may act through activation of similar cellular kinases we determined whether the MAPK cascades (ie, extracellular signal regulated kinases, ERK1 and ERK2; c-Jun N-terminal
kinase [JNK]; and, p38 MAPK) were involved in mediating
alcohol- and polyphenol-induced effects on fibrinolytic protein gene transcription in cultured human ECs. Low alcohol
levels, quercetin or catechin each transiently activated
(phosphorylated) all 3 of the MAPKs, p38, JNK and
ERK1/2, in a time-dependent manner (Western blot). Preincubation of cultured human ECs with the physiologic inhibitors of MAPK pathways prior to induction with ethanol,
catechins, or quercetin indicated that inhibition of only p38
MAPK attenuated the alcohol-, catechin- or quercetininduced expression of both t-PA mRNA (RT-PCR) and
antigen levels (Western blot). Inhibition of the JNK pathway had minimal effect on quercetin-induced t-PA expression, and no effect on the alcohol-and catechin-induced
t-PA gene expression. ERK inhibition had no effect on alcohol- or polyphenol-induced t-PA expression. Therefore,
these findings suggest that the p38 MAPK is the signaling
pathway upstream of alcohol and polyphenol transcription
factors (68). These combined results provide a novel mechanism by which alcohol or individual wine components (i.e.,
alcohol, catechin, and quercetin) can act through a common
shared signaling pathway in the t-PA promoter to increase tPA gene transcription, concomitant with t-PA–mediated
increased EC fibrinolysis.
S28
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
AEP Vol. 17, No. 5S
May 2007: S24–S31
FIGURE 4. Real-time fluorescence imaging analysis of spontaneous Cy5.5-labeled fibrin clot lysis, in vivo, in a C57Bl/6 mouse model.
Regulation of t-PA gene transcription
We have previously demonstrated that the PCR amplified
3.63-kb promoter and 50 flanking region(s) of the t-PA
gene (ligated to a pGL-3 luciferase [luc] reporter gene) contained regulatory sequences important in the transcriptional regulation by alcohol (63) and polyphenols,
including catechin and quercetin (53). Further studies focusing on the identification of regulatory elements and associated transcription factors in this 3.63-kb t-PA
promoter fragment, responsive to alcohol and individual
polyphenols, are currently being finalized. These studies involve generating overlapping 50 -30 promoter deletion fragments, each ligated into the reporter gene to generate 50
promoter deletion constructs from the 3.63-kb t-PA/luc
construct. A combination of transient transfection studies
and electrophoretic mobility assays are being used to identify the possible involved cis regulatory elements. This region contains putative binding sites for a number of
transcription factors including, Sp1 and NFkB. Both these
transcription factors have been shown to regulate basal
and induced t-PA transcription. Therefore, we hypothesize
that alcohol and polyphenols regulate t-PA gene transcription via the activation of one or both these factors. A
combination of oligonucleotide mutational analysis, electrophoretic mobility shift assays (EMSAs), and super EMSAs were used to identify a distinct combination of NFkB
and Sp1 binding sites for alcohol, catechin and quercetin,
activation of t-PA gene transcription (69, 70).
Real-time fluorescence imaging analysis of Cy5.5labeled fibrin clot lyses, in vivo, using a mouse model
We have modified an established pulmonary microembolism mouse model, using 125I-labeled fibrin clots to evaluate
in vivo clot lysis (71), and developed a novel real-time in
vivo clot lyses assay, using Cy5.5 fluorescence (Fl)–labeled
fibrin clots to examine the effects of alcohol and individual
polyphenols on in vivo fibrinolysis (72). Briefly, Fl-labeled
microclots (3.2 mg of protein) are injected into the tail
vein of a mouse and immediately become lodged in the
lung. This allows for the quantitative measurement of in
vivo clot lysis at 10-minute scan intervals (usually up to
120 minutes) in each individual mouse, using the GE Explore OptixÔ acquisition imaging system. The sequence
of actual real-time Fl-imaging analyses (10-minute scans,
10–90 minutes) of Cy5.5-labeled fibrin clot lysis in the
WT mouse lung, captured with the GE Explore OptixÔ acquisition imaging system, is shown in Fig. 4 (left). The quantitative measurement of a representative time-dependent
change in spontaneous Fl-labeled clot lysis, in vivo, in the
lung of a WT mouse, under these specific experimental conditions described above, is shown graphically in Fig. 4 (right).
At the end of the scan period, the mice are sacrificed and the
thoracic aorta and plasma harvested for measurement of
changes in fibrinolytic protein mRNA and protein levels.
Analyses of the time-dependent change in quantitative
measurements of spontaneous clot lysis in vivo demonstrated
that untreated C56Bl/6 WT control mice rapidly lysed their
clots in approximately 2 hours. The rate of in vivo clot lysis
was significantly increased in mice treated with either
alcohol, catechin or quercetin for 4 to 6 weeks (72). This
alcohol- and polyphenol-induced increase in clot lyses was
concomitant with an increase in expression of thoracic
aortic t-PA mRNA levels, in vivo (real-time PCR).
Real-time analysis of Fl-labeled clot lysis in an apoEdeficient (k/o) mouse model of atherosclerosis demonstrated
that in vivo clot lytic rates were significantly impaired in
these mice, compared with WT control mice. However,
we have recently shown that this impaired clot lytic activity
in apoE k/o mice can be ‘‘restored’’ or ‘‘normalized’’ to WT
control levels, in vivo, by treatment with alcohol, catechins,
or quercetin for 6 weeks.
These results demonstrate that alcohol and different
individual polyphenols can independently stimulate and
AEP Vol. 17, No. 5S
May 2007: S24–S31
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
S29
FIGURE 5. Proposed molecular mechanisms by which moderate alcohol and wine polyphenols induce increased EC t-PA-mediated
fibrinolysis to promote increased blood clot lysis.
promote increased blood clot lysis, in vivo, in WT control
mice. Furthermore, alcohol or polyphenols can also induce
the ‘‘restoration’’ or ‘‘normalization’’ of impaired blood
clot lysis, in vivo, in an atherosclerosis setting. These combined results strongly suggest that moderate consumption of
alcoholic beverages, containing only an alcohol component
or alcohol plus polyphenol components (i.e., red wine) will
be expected to similarly promote and sustain increased
blood clot lysis, in vivo, to substantially reduce the risk for
acute CHD-related atherothrombotic consequences of MI
and hence the overall risk for CHD-related mortality.
Molecular mechanisms by which alcohol and
polyphenols induce increased t-PA mediated fibrinolysis
Described studies have identified and defined a molecular
mechanism by which moderate levels of alcohol or individual wine components (i.e., principal red wine polyphenols, catechins, and quercetin) induce increased EC t-PA
expression, resulting in increased EC fibrinolysis. Increased
EC fibrinolysis will promote increased blood clot lysis and
will be expected to contribute, in part, to reducing the risk
for early thrombotic events and later acute atherothrombotic (following plaque rupture) consequences of MI.
The possible molecular events that interact sequentially
to initiate and promote increased t-PA–mediated blood
clot lysis are shown schematically in Fig. 5. The proposed
molecular activation sequences include: alcohol- and/or
catechin- and/or quercetin-induced activation of p38
MAPK signaling which can lead to; activation and nuclear
translocation of specific transcription factors, NFkB and
Sp1; binding of transcription factors to their inducerspecific, separate individual binding sites in the t-PA
promoter to activate gene transcription; translation of increased t-PA mRNA to increased t-PA protein expression
and transport to the EC surface; localization of increased
t-PA at the EC surface by binding to the increased, simultaneously expressed t-PA binding protein, Ann-II; increased blood clot lysis and formation of fibrin
degradation products (FDPs). In addition to the increased
expression of t-PA and Ann-II, the concurrent decreased
expression of inhibitor PAI-1 and its regulatory action is
also shown.
This work was supported, in part, by Grant HL070610 (National Heart
Lung Blood Institute) and Grant AA11674 (National Institute on
Alcohol Abuse and Alcoholism).
REFERENCES
1. Barnett E, Halverson J. Local increases in coronary heart disease mortality
among blacks and whites in the United States, 1985–1995. Am J Public
Health. 2001;91:1499–1506.
2. Jousilahti P, Salomaa V, Kuulasmaa K, Niemela M, Vartiainen E. Total and
cause specific mortality among participants and non-participants of
S30
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
population based health surveys: a comprehensive follow up of 54 372 Finnish men and women. J Epidemiol Community Health. 2005;59:310–315.
3. Bakx C, Schwarte J, van den Hoogen H, Bor H, van Weel C. First
myocardial infarction in a Dutch general practice population: trends in
incidence from 1975–2003. Br J Gen Pract. 2005;55:860–863.
4. Machado DB, Crow RS, Boland LL, Hannan PJ, Taylor HA Jr, Folsom
AR. Electrocardiographic findings and incident coronary heart disease
among participants in the atherosclerosis risk in communities (ARIC)
Study. Am J Cardiol. 2006;97:1176–1181.
AEP Vol. 17, No. 5S
May 2007: S24–S31
22. Rendig SV, Symons JD, Longhurst JC, Amsterdam EA. Effects of red wine,
alcohol, and quercetin on coronary resistance and conductance arteries.
J Cardiovasc Pharmacol. 2001;38:219–227.
23. Liu J, Tian Z, Gao B, Kunos G. Dose-dependent activation of antiapoptotic and proapoptotic pathways by ethanol treatment in human vascular
endothelial cells. differential involvement of adenosine. J Biol Chem.
2002;277:20927–20933.
24. Salem RO, Laposata M. Effects of alcohol on hemostasis. Am J Clin Pathol. 2005;123(Suppl):S96–105.
5. Goldberg RJ, Glatfelter K, Burbank-Schmidt E, Lessard D, Gore JM.
Trends in community mortality due to coronary heart disease. Am Heart
J. 2006;151:501–507.
25. Wang Z, Barker TH, Fuller GM. Alcohol at moderate levels decrease
fibrinogen expression in vivo and in vitro. Alcohol Clin Exp Res.
1999;23:1927–1932.
6. Doll R, Peto R, Hall E, Wheatley K, Gray R. Mortality in relation to
consumption of alcohol: 13 years’ observations on male British doctors.
[see comment] BMJ. 1994;309:911–918.
26. Ridker PM, Vaughan DE, Stampfer MJ, Glynn RJ, Hennekens CH. Association of moderate alcohol consumption and plasma concentration of endogenous tissue-type plasminogen activator. JAMA. 1994;272:929–933.
7. Theobald H, Bygren LO, Carstensen J, Engfeldt P. A moderate intake of
wine is associated with reduced total mortality and reduced mortality
from cardiovascular disease. J Stud Alcohol. 2000;61:652–656.
27. Veenstra J, te Wierik E, Kluft C. Alcohol and fibrinolysis. Fibrinolysis.
1995;4:64–67.
8. Ruf JC. Overview of epidemiological studies on wine, health and mortality. Drugs Exp Clin Res. 2003;29:173–179.
28. Aikens ML, Benza RL, Grenett HE, Tabengwa EM, Davis GC, Demissie S,
et al. Ethanol increases surface-localized fibrinolytic activity in cultured
endothelial cells. Alcohol Clin Exp Res. 1997;21:1471–1478.
9. Wellmann J, Heidrich J, Berger K, Doring A, Heuschmann PU, Keil U.
Changes in alcohol intake and risk of coronary heart disease and all-cause
mortality in the MONICA/KORA-Augsburg cohort 1987–97. Eur J Cardiovasc Prev Rehabil. 2004;11:48–55.
29. Abou-Agag LH, Tabengwa EM, Tresnak JA, Wheeler CG, Taylor KB,
Booyse FM. Ethanol-induced increased surface-localized fibrinolytic activity in cultured human endothelial cells: kinetic analysis. Alcohol Clin Exp
Res. 2001;25:351–361.
10. Renaud SC, Gueguen R, Conard P, Lanzmann-Petithory D, Orgogozo JM,
Henry O. Moderate wine drinkers have lower hypertension-related mortality: a prospective cohort study in French men. Am J Clin Nutr. 2004;
80:621–625.
30. Lee KW, Lip GY. Effects of lifestyle on hemostasis, fibrinolysis, and platelet
reactivity: a systematic review. Arch Intern Med. 2003;163:2368–2392.
11. Mansvelt EP, van Velden DP, Fourie E, Rossouw M, van Rensburg SJ,
Smuts CM. The in vivo antithrombotic effect of wine consumption on
human blood platelets and hemostatic factors. Ann N Y Acad Sci.
2002;957:329–332.
12. de Lorgeril M, Salen P, Martin JL, Boucher F, Paillard F, de Leiris J. Wine
drinking and risks of cardiovascular complications after recent acute myocardial infarction [see comment] Circulation. 2002;106:1465–1469.
13. Fernandez-Jarne E, Martinez-Losa E, Serrano-Martinez M, Prado-Santamaria M, Brugarolas-Brufau C, Martinez-Gonzalez MA. Type of alcoholic
beverage and first acute myocardial infarction: a case-control study in
a Mediterranean country. Clin Cardiol. 2003;26:313–318.
31. de Lange DW, van de Wiel A. Drink to prevent: review on the cardioprotective mechanisms of alcohol and red wine polyphenols. Semin Vasc
Med. 2004;4:173–186.
32. Guillaume P, Jankowski M, Gutkowska J, Gianoulakis C. Effect of chronic
moderate ethanol consumption on heart brain natriuretic peptide. Eur J
Pharmacol. 1996;316:49–58.
33. Emeson EE, Manaves V, Emeson BS, Chen L, Jovanovic I. Alcohol inhibits the progression as well as the initiation of atherosclerotic lesions
in C57Bl/6 hyperlipidemic mice. Alcohol Clin Exp Res. 2000;24(9):
1456–1466.
34. Rimm EB, Katan MB, Ascherio A, Stampfer MJ, Willett WC. Relation between intake of flavonoids and risk for coronary heart disease in male
health professionals. Ann Intern Med. 1996;125:384–389.
14. Augustin LS, Gallus S, Tavani A, Bosetti C, Negri E, La VC. Alcohol consumption and acute myocardial infarction: a benefit of alcohol consumed
with meals? Epidemiology. 2004;15:767–769.
35. Flesch M, Schwarz A, Bohm M. Effects of red and white wine on endothelium-dependent vasorelaxation of rat aorta and human coronary arteries.
Am J Physiol. 1998;275:H1183–H1190.
15. Suh I, Shaten J, Cutler JA, Kuller LH. The Multiple Risk Factor Intervention Trial Research Group. Alcohol use and mortality form coronary heart
disease: the role of high-density lipoprotein cholesterol. Ann Intern Med.
1992;116:881–887.
36. Andriambeloson E, Stoclet JC, Andriantsitohaina R. Mechanism of endothelial nitric oxide-dependent vasorelaxation induced by wine polyphenols
in rat thoracic aorta. J Cardiovasc Pharmacol. 1999;33:248–254.
16. Renaud S, Lanzmann-Petithory D, Gueguen R, Conard P. Alcohol and
mortality from all causes. Biol Res. 2004;37:183–187.
17. Mukamal KJ, Jensen MK, Grønbæk M, Stampfer MJ, Manson JE, Pischon
T, et al. Drinking frequency, mediating biomarkers, and risk of myocardial
infarction in women and men. Circulation. 2005;112:1406–1413.
18. Zhang QH, Das K, Siddiqui S, Myers AK. Effects of acute, moderate ethanol consumption on human platelet aggregation in platelet-rich plasma
and whole blood. Alcohol Clin Exp Res. 2000;24:528–534.
19. Ruf JC. Alcohol, wine and platelet function. Biol Res. 2004;37:209–215.
20. Miyamae M, Diamond I, Weiner MW, Camacho SA, Figueredo VM. Regular alcohol consumption mimics cardiac preconditioning by protecting
against ischemia-reperfusion injury. Proc Natl Acad Sci U S A.
1997;94:3235–3239.
21. Guiraud A, de Lorgeril M, Boucher F, Berthonneche C, Rakotovao A, de
Leiris J. Cardioprotective effect of chronic low dose ethanol drinking: insights into the concept of ethanol preconditioning. J Mol Cell Cardiol.
2004;36:561–566.
37. Rakici O, Kiziltepe U, Coskun B, Aslamaci S, Akar F. Effects of resveratrol
on vascular tone and endothelial function of human saphenous vein and
internal mammary artery. Int J Cardiol. 2005;105:209–215.
38. Hayek T, Fuhrman B, Vaya J, Rosenblat M, Belinky P, Coleman R, et al.
Reduced progression of atherosclerosis in apolipoprotein E-deficient mice
following consumption of red wine, or its polyphenols quercetin or catechin, is associated with reduced susceptibility of LDL to oxidation and
aggregation. Arterioscler Thromb Vasc Biol. 1997;17:2744–2752.
39. Aviram M, Fuhrman B. Wine flavonoids protect against LDL oxidation
and atherosclerosis 3. Ann N Y Acad Sci. 2002;957:146–161.
40. Corder R, Douthwaite JA, Lees DM, Khan NQ, Viseu Dos Santos AC,
Wood EG, et al. Endothelin-1 synthesis reduced by red wine. Nature.
2001;414:863–864.
41. Stoclet JC, Chataigneau T, Ndiaye M, Oak MH, El Bedoui J, Chataigneau
M, et al. Vascular protection by dietary polyphenols. Eur J Pharmacol.
2004;500:299–313.
42. Di Santo A, Mezzetti A, Napoleone E, Di Tommaso R, Donati MB, de
Gaetano G, et al. Resveratrol and quercetin down-regulate tissue factor
AEP Vol. 17, No. 5S
May 2007: S24–S31
Booyse et al.
ALCOHOL AND WINE POLYPHENOLS EFFECT ON CHD
S31
expression by human stimulated vascular cells. J Thromb Haemost.
2003;1:1089–1095.
59. Laug WE. Ethyl alcohol enhances plasminogen activator secretion by
endothelial cells. JAMA. 1983;250:772–776.
43. Casani L, Segales E, Vilahur G, Bayes de Luna A, Badimon L. Moderate
daily intake of red wine inhibits mural thrombosis and monocyte tissue
factor expression in an experimental porcine model. Circulation.
2004;110:460–465.
60. Venkov CD, Su M, Shry Y, Vaughan DE. Ethanol-induced alterations in
the expression of endothelial-derived fibrinolytic components. Fibrinol
Proteol. 1997;11(2):115–118.
44. Di Santo A, Mezzetti A, Napoleone E, Di Tommaso R, Donati MB, de
Gaetano G, Lorenzet R. Resveratrol and quercetin down-regulate tissue
factor expression by human stimulated vascular cells. J Thromb Haemost.
2003;1:1089–1095.
45. De Curtis A, Murzilli S, Di Castelnuovo A, Rotilio D, Donati MB, de Gaetano G, et al. Alcohol-free red wine prevents arterial thrombosis in
dietary-induced hypercholesterolemic rats: experimental support for the
‘French paradox’. J Thromb Haemost. 2005;3:346–350.
46. Zern TL, Fernandez ML. Cardioprotective effects of dietary polyphenols.
J Nutr. 2005;135:2291–2294.
47. Avellone G, Di Garbo V, Campisi D, De Simone R, Raneli G, Scaglione
R, et al. Effects of moderate Sicilian red wine consumption on inflammatory biomarkers of atherosclerosis. Eur J Clin Nutr. 2006;60:41–47.
48. Bucki R, Pastore JJ, Giraud F, Sulpice JC, Janmey PA. Flavonoid inhibition of platelet procoagulant activity and phosphoinositide synthesis.
J Thromb Haemost. 2003;1:1820–1828.
49. Hubbard GP, Wolffram S, Lovegrove JA, Gibbins JM. Ingestion of quercetin inhibits platelet aggregation and essential components of the collagenstimulated platelet activation pathway in humans. J Thromb Haemost.
2004;2:2138–2145.
50. Iijima K, Yoshizumi M, Hashimoto M, Kim S, Eto M, Ako J, et al. Red wine
polyphenols inhibit proliferation of vascular smooth muscle cells and downregulate expression of cyclin A gene. Circulation. 2000;101:805–811.
51. Alcocer F, Whitley D, Salazar-Gonzalez JF, Jordan WD, Sellers MT, Eckhoff DE, et al. Quercetin inhibits human vascular smooth muscle cell
proliferation and migration. Surgery. 2002;131:198–204.
52. Zhao X, Gu Z, Attele AS, Yuan CS. Effects of quercetin on the release
of endothelin, prostacyclin and tissue plasminogen activator from human
endothelial cells in culture. J Ethnopharmacol. 1999;67:279–285.
53. Abou-Agag LH, Aikens ML, Tabengwa EM, Benza RL, Shows SR, Grenett
HE, et al. Polyphyenolics increase t-PA and u-PA gene transcription in
cultured human endothelial cells. Alcohol Clin Exp Res. 2001;25:155–162.
54. Lijnen HR. Elements of the fibrinolytic system. Ann N Y Acad Sci.
2001;936:226–236.
55. Vassalli JD, Sappino AP, Belin D. The plasminogen activator/plasmin
system. J Clin Invest. 1991;88:1067–1072.
56. Lijnen HR, Collen D. Mechanisms of physiological fibrinolysis. Baillieres
Clin Haematol. 1995;8:277–289.
57. Cesarman GM, Guevara CA, Hajjar KA. An endothelial cell receptor for
plasminogen/tissue plasminogen activator (t-PA). II. Annexin II-mediated
enhancement of t-PA-dependent plasminogen activator. J Biol Chem.
1994;269:21198–21203.
58. Barnathan ES. Characterization and regulation of the urokinase receptor
of human endothelial cells. Fibrinolysis. 1992;6:1–9.
61. Miyamoto A, Yang SX, Laufs U, Ruan XL, Liao JK. Activation of guanine
nucleotide-binding proteins and induction of endothelial tissue-type
plasminogen activator gene transcription by alcohol. J Biol Chem.
1999;274:12055–12060.
62. Aikens ML, Grenett HE, Benza RL, Tabengwa EM, Davis GC, Booyse FM.
Alcohol-induced up-regulation of plasminogen activators and fibrinolytic
activity in cultured human endothelial cells. Alcohol Clin Exp Res.
1998;22:375–381.
63. Grenett HE, Aikens ML, Torres JA, Demissie S, Tabengwa EM, Davis GC,
et al. Ethanol transcriptionally upregulates t-PA and u-PA gene expression
in cultured human endothelial cells. Alcohol Clin Exp Res. 1998;22:849–
853.
64. Tabengwa EM, Abou-Agag LH, Benza RL, Torres JA, Aikens ML, Booyse
FM. Ethanol-induced up-regulation of candidate plasminogen receptor
annexin II in cultured human endothelial cells. Alcohol Clin Exp Res.
2000;24:754–761.
65. Tabengwa EM, Grenett HE, Benza RL, Abou-Agag LH, Tresnak JK,
Wheeler CG, et al. Ethanol-induced up-regulation of the urokinase receptor in cultured human endothelial cells. Alcohol Clin Exp Res.
2001;25:163–170.
66. Grenett HE, Aikens ML, Tabengwa EM, Davis GC, Booyse FM. Ethanol
downregulates transcription of the PAI-1 gene in cultured human endothelial cells. Thromb Res. 2000;97:247–255.
67. Grenett HE, Wolkowicz PE, Benza RL, Tresnak JK, Wheeler And CG,
Booyse FM. Identification of a 251-bp fragment of the PAI-1 gene promoter that mediates the ethanol-induced suppression of PAI-1 expression.
Alcohol Clin Exp Res. 2001;25:629–636.
68. Tabengwa EM, Drawz SL, Pan W, Grenett HE, Booyse FM. Wine components activate t-PA gene transcription through the JNK and p38 mitogen
activated protein kinase pathways. J Am Coll Cardiol. 2005;45(3 Suppl
A):367A.
69. Pan W, Booyse FM, Grenett HE, Pasten MC, Drawz SL, Tabengwa EM.
Alcohol induction of tissue plasminogen activator expression in human
endothelial cells through Sp1 and NFkappaB activation: involvement
of p38 MAPK pathway. Arterioscler Thromb Vasc Biol. In press.
70. Pan W, Booyse FM, Bradley KM, Tabengwa EM. Transcriptional activation of tissue plasminogen activator by dietary polyphenols catechin
and quercetin is mediated through Sp1 and NFkappaB: involvement of
p38 MAPK pathway. Arterioscler Thromb Vasc Biol. In press.
71. Bdeir K, Murciano JC, Tomaszewski J, Koniaris L, Martinez J, Cines DB,
et al. Urokinase mediates fibrinolysis in the pulmonary microvasculature.
Blood. 2000;96:1820–1826.
72. Pan W, Booyse FM, Grenett HE, Parks DA, Zinn K, Tabengwa EM. Development of a real-time fluorescence imaging in vivo clot lysis method:
alcohol- and polyphenol (catechin and quercetin)-induced effects.
Thromb Haemost. In press.
Changes in Cardiovascular Risk Factors Associated with Wine
Consumption in Intervention Studies in Humans
FEDERICO LEIGHTON, MD, AND INÉS URQUIAGA, PHD
Evidence that links moderate wine consumption to cardiovascular health corresponds mostly to ecological
observations. Intervention studies using moderate wine consumption with ischemic heart disease as the end
point will probably not be available soon because they require long-term follow-up and adequately randomized experimental groups. In contrast, short-term studies focused on risk factors are feasible and should
provide evidence suitable for a critical assessment of the apparent beneficial role of moderate drinking,
as well as other lifestyle measures, on cardiovascular health.
Our intervention studies suggest an increase in HDL-cholesterol, decrease in the omega-6/omega-3 ratio, and in some cases a slight increase in triglyceride levels from moderate drinking. Observed changes in
hemostasis include reduced coagulation and increased fibrinolysis; effects on blood pressure have been inconsistent. There is a reduction in inflammatory markers and an increase in endothelial function. Effects of
wine are greater for subjects on a Mediterranean diet than those on an occidental diet. Several key biochemical or physiological processes related to atherogenesis are positively modified by wine consumption. From
our observations, we conclude that wine in moderation and as part of the diet is directly responsible for
changes that may help decrease the risk of cardiovascular disease.
Ann Epidemiol 2007;17:S32–S36. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS: Wine, Alcohol Consumption, Risk Factors, Mediterranean Diet, Intervention Studies,
Hemostasis, Inflammation, Endothelium, Vascular, Nitric Oxide, Oxidative Stress.
INTRODUCTION
Epidemiological studies have shown an inverse association
between moderate alcohol consumption and cardiovascular
disease. The consistency of the findings strongly suggests
a causal relationship. Intervention studies could help define
the role of alcoholic beverages (wine, beer, and distilled
spirits) in coronary heart disease (CHD). However, adequately randomized cohorts of volunteers and the time required for prospective intervention studies make these
almost impossible. A reasonable approximation for prospective intervention studies is offered by primary or even
secondary prevention trials on high-risk subjects, yet extrapolation of results to the entire population is difficult. This is
also true for the effect of moderate drinking on all-cause
death or longevity. An alternative approach is to perform
short-term prospective intervention studies with wine, measuring the changes observed in biochemical or physiological
cardiovascular risk factors. This strategy, generally used to
define dietary and lifestyle habits for the prevention of
CHD, represents a sophisticated approach that avoids misleading results, such as those that have existed in studies
on dietary and total blood cholesterol (1).
Age, sex, family history, smoking, a sedentary lifestyle,
obesity, and diabetes are among the best-established cardiovascular risk factors and are considered in algorithms employed for risk calculation (e.g., Framingham Heart Study,
Multiple Risk Factor Intervention Trial). For several reasons, however, these factors are not adequate for evaluating
the possible changes in cardiovascular risk associated with
intervention studies of moderate alcohol consumption.
Other established and novel risk factors should be considered in wine or dietary intervention studies, including (a)
plasma lipids, (b) hemostasis components, (c) markers of inflammation, (d) endothelial function, (e) blood pressure, (f)
anthropometric parameters, (g) oxidative stress, and (h) homocysteine. Information exists for these factors and will be
discussed herein.
Hypothesis
From the Molecular Nutrition Laboratory, Universidad Católica de
Chile, Santiago.
Address correspondence to: Dr. Federico Leighton, Facultad de Ciencias
Biológicas, Universidad Católica de Chile, Casilla 114-D, Santiago, Chile.
E-mail: [email protected].
Supported by the Molecular Basis of Chronic Disease Fund (PUCPBMEC) of the Catholic University, Chile, 2003–2005.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
When improvements are observed in individual risk factors
following intervention studies, they probably represent a reduction in the overall risk of CHD. A more accurate prediction of the change in risk would require an evaluation of the
predictive capacity of a specific cluster of factors, beyond
calculations made for independent risk factors.
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.007
AEP Vol. 17, No. 5S
May 2007: S32–S36
Selected Abbreviations and Acronyms
CHD Z coronary heart disease
HDL Z high-density lipoprotein
eNOS Z endothelial nitric oxide synthase
8-OHdG Z 8-hydroxy-20 -deoxyguanosine
LDL Z low-density lipoprotein
Moderate Wine Consumption Intervention Studies
to Monitor Cardiovascular Risk
The search for the biochemical mechanisms to explain
the apparent healthy effects of moderate wine and other alcoholic beverage consumption has led to numerous studies
with different experimental designs, in which a wide range
of biochemical, physiological, and anthropometric parameters have been evaluated. Some of these studies are reviewed
and their main conclusions summarized. The analysis of individual parameters selected considers both the frequency
with which they have been included in intervention studies
and the evidence that renders them valid as cardiovascular
risk factors.
Lipid Factors. Ethanol-mediated high-density lipoprotein (HDL) cholesterol elevation is generally considered
the main mechanism explaining the benefits of moderate alcohol consumption, with Rimm et al (2) estimating that
HDL explains one half of the beneficial effects. In a metaanalysis conducted by the authors (3), calculations suggested that 30 g of alcohol a day (approximately 320 mL
of wine) would cause an average increase of 8.3% in HDL
levels. We conducted an intervention study on 44 healthy
young male volunteers who received a high-fat diet, rich
in monounsaturated or polyunsaturated fatty acids and supplemented isocalorically with either a soft drink, 250 mL of
white wine, 250 mL of red wine, or fruits and vegetables for
3-week periods each. We observed increases in HDL levels
ranging between 4% and 18% among those who received
wine, which was greater than that seen with fruits and vegetables; red wine appeared to be slightly better than white
wine for this purpose.
In a recent Internet-based intervention to promote the
Mediterranean diet in Scotland, increases in HDL levels
were observed among those who followed it; these increases
were attributed to increased consumption of vegetables,
fruits, and legumes, as well as the monounsaturated fatty
acid/saturated fatty acid ratio of the diets (4). Increased
levels of apolipoprotein A-1 have also been observed after
moderate alcohol consumption, particularly during the transient postprandial phase (5). The mechanism through which
HDL exerts its beneficial effects has not been fully elucidated; experimental evidence supports reverse cholesterol
transport mediated by HDL, but the results are not consistent (5). An additional explanation for the antiatherogenic
Leighton and Urquiaga
WINE AND CARDIOVASCULAR RISK FACTORS
S33
effects of HDL is its potent endothelial nitric oxide synthase
(eNOS)–inducing activity, which is apparently mediated by
scavenger receptors class B type I (SR-BI) and would explain
its premenopausal estrogen-dependent protective effects
(6).
A negative consequence of alcohol consumption on
plasma lipids is the increase observed in triglycerides. Although fasting triglyceride levels have been shown to increase in some experimental observations, the change is
minor compared with increases in HDL, particularly when
alcohol consumption is kept below 30 g/d.
Hemostasis. The beneficial effects of wine are reflected
at many levels of the hemostatic process, including primary
hemostasis, fibrinolysis, coagulation, and thrombosis. Almost all the changes described are protective and, after
changes in HDL, constitute the second main explanation
for the lower risk of CHD associated with moderate alcohol
consumption (2). The effects of alcohol on hemostasis have
been reviewed recently (7). In addition to changes induced
after 1 month of daily consumption of 240 mL of red wine
(8), studies revealed that subjects on a high-fat diet showed
increases in procoagulant fibrinogen (22%), factor VIIc
(9%) and factor VIIIc (4%). Decreases in the natural anticoagulants antithrombin III (3%), protein C (11%), protein S
(6%), and plasminogen activator inhibitor-1 (20%) were
also observed. In contrast, subjects on a Mediterranean
diet showed less of an increase in fibrinogen (4%), antithrombin III (5%), protein C (3%), and protein S (2.7%),
and decreases in factor VIIIc (9%) and plasminogen activator inhibitor-1 (21%). In both diets, red wine supplementation resulted in decreased plasma fibrinogen and factor VIIc,
and increased tissue plasminogen activator antigen and plasminogen activator inhibitor-1 antigen. Divergent effects
were observed with wine consumption on antithrombin
III. Whereas the high-fat diet groups showed a 10% decrease, there was a slight increase in the Mediterranean
diet group. These results confirm those described in other
studies and emphasize the important role of the diet in the
modulation of the biological effects of alcohol. Clearly,
there is synergy between the Mediterranean diet and wine
consumption for the prevention of cardiovascular disease.
With regard to thrombosis, observations in animal
models suggest that in addition to the antithrombotic effect
of alcohol, wine phenolics, per se, also play an important
role, which is mediated by the enzyme eNOS synthase. In
human studies the available evidence supports a principal
role for ethanol in the changes observed with wine administration (9). In contrast, wine phenol antioxidants apparently play a major role in the protective effects of nitric
oxide, perhaps an explanation for the modulation by the
Mediterranean diet of the changes in hemostasis associated
with moderate alcohol consumption.
S34
Leighton and Urquiaga
WINE AND CARDIOVASCULAR RISK FACTORS
Inflammation. There is growing evidence that inflammation plays a central role at various stages of the atherosclerotic
process (10). Phenolic compounds in wine have several biological activities that make them potential therapeutic agents
for cardiovascular disease, as recently reviewed by Curin
and Andriantsitohaina (11). Cyclooxygenase, lipoxygenase,
prostaglandin biosynthesis, monocyte adhesion and nuclear
factor kB activation are among the inflammatory responses
modified by phenolics. Notwithstanding this evidence, intervention studies with wine and other alcoholic beverages
point to a central role of ethanol in the anti-inflammatory responses observed. A prospective randomized crossover trial
compared the effects of red wine and gin (12) and concluded
that both wine and gin exert anti-inflammatory effects by reducing plasma fibrinogen and interleukin 1a levels; wine had
the additional effect of decreasing hs-C-reactive protein, as
well as monocyte and endothelial adhesion molecules. The
effects of red and white wine were compared by Williams et
al (13), who found that plasma interleukin 6 was elevated
after the consumption of either, a result the authors correlated with blood alcohol levels.
Beer also exerts anti-inflammatory effects, and intervention studies show a decrease in plasma C-reactive protein
and fibrinogen levels (14). These results clearly suggest that
alcohol is responsible for several anti-inflammatory changes
observed after moderate drinking and that wine phenolics
may have additional anti-inflammatory activities. Among
the effects attributed to wine phenolics are the increased level
of u-3 fatty acids after wine consumption, the anti-inflammatory properties of which have been well described (15).
Endothelial Function. Endothelial dysfunction is an
early event in the development of atherosclerosis. It accompanies most of the risk factors that result in atherosclerosis
development and is present along the various stages of the
disease. Indeed, endothelial dysfunction seems to be an efficient marker of the composite effect of the various factors involved in predisposition to atherosclerosis, as well as a valid
marker for monitoring changes in these risk factors. Nevertheless, its value as a surrogate marker for atherosclerotic
cardiovascular risk has not been established firmly, probably
because the methods employed for its clinical evaluation are
still too cumbersome for it to be included in the extensive
prospective studies needed (16, 17).
We have shown that a high-fat diet induces endothelial
dysfunction and that red wine counteracts this effect (18),
a phenomenon apparently specific to red wine (19). In a
randomized controlled intervention study performed in
healthy, normotensive, moderate-drinking men, Zilkens
et al (20) did not find changes in endothelial function measured by flow-mediated brachial artery dilatation. However,
we have shown that clear-cut results are observed after 4
weeks of red wine consumption in volunteers who received
AEP Vol. 17, No. 5S
May 2007: S32–S36
a high-fat diet, known to depress endothelial function, but
not in those who received a Mediterranean diet (18). The
flow-mediated dilatation procedure to measure endothelial
function is very sensitive. In acute experiments it has been
shown that smoking a single cigarette will decrease endothelial function for approximately 60 minutes, a change prevented when red wine or dealcoholized red wine is
consumed while smoking (21).
In contrast to other risk factors, the correction of endothelial dysfunction after wine administration seems dependent on wine phenolics, an effect attributed to their
antioxidative properties and their direct effects on eNOS.
In nonacute experiments, ethanol may have a direct effect
on endothelial function. In fact, studies to explore the
mechanism of chronic ethanol-induced acute respiratory
distress syndrome have shown that under experimental conditions long-term ethanol stimulation modulates vascular
endothelial function and increases nitric oxide production
by increasing eNOS protein (22).
Wine, u-3 fatty acids, and the Mediterranean diet in general protect endothelial function. Furthermore, the pathogenesis of metabolic syndrome has been attributed to
endothelial dysfunction, an attractive hypothesis in the
search for the biochemical explanation of this syndrome,
as well as in defining strategies to reduce it and to monitor
the efficacy of corrective measures (23).
Blood Pressure. There is substantial evidence that demonstrates a positive relationship between alcohol consumption and blood pressure. Alcohol consumption of 3 or more
standard drinks per day is associated with and predictive of
hypertension. Routine screening for excessive alcohol consumption should therefore have a major impact on reducing
the prevalence of hypertension among the general population (24). The response to a high dose of alcohol is biphasic:
first vasodilation occurs, followed by a pressor effect, a hemodynamic pattern that should be considered when blood pressure changes are not monitored continuously (25). Zilkens
et al. found that both beer and red wine (in doses of approximately 40 g of alcohol per day) increase systolic blood pressure by 1.9 and 2.9 mm Hg, respectively, suggesting that red
wine does not decrease blood pressure (20).
In contrast, Thadhani et al. (26) examined the association between alcohol consumption and the subsequent
risk of hypertension in 70,891 women ranging from 25 to
42 years of age and found a J-shaped curve, with light
drinkers demonstrating a modest decrease in risk. In addition to dose, meals may also affect the hypertensive effect
of alcoholic beverages; an elevated risk for hypertension
was found by Stranges et al. (27) to occur in subjects consuming alcohol without meals. In an intervention study
with red wine (including 53 women drinking 11.4 to
14.2 g of ethanol per day and 42 men drinking 22.8 to
AEP Vol. 17, No. 5S
May 2007: S32–S36
28.4 g/d), we found that after 6 weeks, systolic and diastolic
pressure values did not change significantly as compared
with those in a control group that did not consume wine.
In another intervention study, involving elderly people, administered the same doses, wine had a slight hypotensive effect. Finally, in a third intervention study done with young
male volunteers, both white wine (at a dosage of 23.3 g of
alcohol per day) and fruits and vegetables showed a mild hypotensive effect, not apparent with red wine.
Overall, it appears that moderate consumption of wine,
defined herein as less than 30 g of alcohol per day for males
and half that dose for females, is associated either with
a slight hypotensive response or with no change in blood
pressure. The results from a prospective study of moderate
alcohol consumption in young women suggests that the
same relationship observed for wine is also present for other
alcoholic beverages (26).
Anthropometric Parameters. Waist circumference,
more than body mass index, is being recognized as a marker
of intra-abdominal fat content and the metabolic syndrome.
Vadstrup et al (28) examined the long-term association between the amount and type of alcohol consumed and subsequent waist circumference among a sample of men and
women from the Copenhagen City Heart Study (28). Their
conclusions were that consumption of beer and spirits was
associated with the development of high waist circumference, whereas moderate-to-high wine consumption apparently had the opposite effect. These results are in
agreement with our proposition that red wine, through
eNOS enhancement, should help in the control and prevention of the metabolic syndrome (23).
In a recent intervention involving 53 women drinking
red wine at a dose equivalent to 11.4 to 14.2 g of ethanol
per day and 42 men drinking 22.8 to 28.4 g/d, we found
that compared with basal values, waist circumference decreased 1.0 cm (p ! 0.0001) after 6 weeks of wine consumption; a similar decrease of 1.3 cm (p ! 0.0001) was observed
for volunteers who did not drink wine. The conclusion is
that wine consumption, in this short-term 6-week intervention, did not increase waist circumference.
Oxidative Stress. Intervention studies using red wine have
shown that wine phenolics are, to a certain degree, absorbed and
that either the original compounds or their metabolites exert antioxidant effects. Several studies have shown protection of lowdensity lipoprotein (LDL) cholesterol against oxidation in vitro.
Tsang et al. (29) gave red wine to volunteers for 2 weeks and
found a decrease in conjugated dienes and in TBARS in copper-oxidized LDL (29); they were able to detect metabolites of
wine phenolics in plasma. Their conclusion, similar to that of
others doing similar interventions, is that moderate
Leighton and Urquiaga
WINE AND CARDIOVASCULAR RISK FACTORS
S35
consumption of red wine exerts potential protective effects related to phenolic antioxidants.
In an intervention study in which young male volunteers
were given Mediterranean or occidental (high-fat) diets accompanied by 240 mL of red wine daily over a 4-week period, total phenols in plasma were elevated after wine
consumption. An increase in total plasma antioxidant capacity was observed, both in those on the Mediterranean
diet and in those on the occidental diet, and for those
who received wine with either diet (30). A concurrent decrease in leukocyte DNA oxidation was also observed,
both with the Mediterranean diet and with wine.
In another similar intervention study, we used white
wine, red wine, or a supplement of fruits and vegetables
that provided the same caloric count as the wine. We found
that red wine exerts marked protection on DNA oxidation,
a 56.5% decrease in 8-OHdG. A significant effect was also
observed with diet supplementation by fruits and vegetables,
a 36.4% decrease, and a smaller (12%) decrease was observed with white wine (3).
In another intervention, performed with elderly male
and female volunteers, we found the same degree of protection of DNA by red wine consumption in both groups. This
result is particularly interesting, as the women received one
half of the dosage of wine of men (12g ethanol per day vs. 24
g ethanol per day). In an intervention study, Natella et al.
(31) measured the wave of plasma lipoperoxides associated
with postprandial hyperlipemia; they found that 250 mL of
wine with a meal almost completely prevented the increase
in plasma lipoperoxides, an effect attributed by the authors
to antioxidant protection in the lumen of the intestine (31).
Homocysteine. Plasma homocysteine levels are strong
independent biomarkers for CHD-related death (32). It
has been reported that 40g daily of alcohol in the form of
red wine or spirits, but not beer, increases homocysteine
levels (33). These observations should take into consideration that there is a U-shaped relation between alcohol consumption and homocysteine concentration, with an average
of less than 14 g daily being associated with lower homocysteine concentrations (34).
CONCLUSIONS
The examination of cardiovascular risk factors for which
changes have been observed in intervention studies confirm
the possibility that randomized prospective studies are feasible to evaluate the effect of moderate wine consumption on
the prevention of CHD. The relative contribution of individual biomarkers remains to be systematically evaluated.
An example of this is the analysis that led Lee et al. (32)
to consider homocysteine and interleukin 6 as strong independent markers of subsequent CHD-related death.
S36
Leighton and Urquiaga
WINE AND CARDIOVASCULAR RISK FACTORS
AEP Vol. 17, No. 5S
May 2007: S32–S36
The close correlation of endothelial function with risk
factors and its consistent response to procedures that
decrease cardiovascular risk underscores the need of developing procedures that would permit fast, simple, and reliable
measurements in large-scale observations.
The interventions already conducted raise 4 important issues that should be considered in interpreting results and establishing health-related recommendations: (1) wine and
the Mediterranean diet exert synergistic protective effects
on cardiovascular risk factors; (2) some of the protective
changes observed after moderate wine consumption are
seen only in the presence of high-fat diets; (3) wine or other
alcoholic beverages with meals lead to maximal protection;
(4) limitation of consumption to below 30 g of alcohol daily
for men and half that dose for women is necessary to prevent
negative changes in cardiovascular risk factors such as hypertension, hypertriglyceridemia, and hyperhomocysteinemia.
14. Sierksma A, van der Gaag MS, Kluft C, Hendriks HF. Moderate alcohol consumption reduces plasma C-reactive protein and fibrinogen levels; a randomized, diet-controlled intervention study. Eur J Clin Nutr. 2002;56:1130–1136.
REFERENCES
22. Polikandriotis JA, Rupnow HL, Hart CM. Chronic ethanol exposure stimulates endothelial cell nitric oxide production through PI-3 kinase-and hsp90dependent mechanisms. Alcohol Clin Exp Res.. 2005;29:1932–1938.
1. Kromhout D, Menotti A, Kesteloot H, Sans S. Prevention of coronary
heart disease by diet and lifestyle: evidence from prospective cross-cultural,
cohort, and intervention studies. Circulation. 2002;105:893–898.
2. Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects
on lipids and haemostatic factors. Br Med J. 1999;319:1523–1528.
3. Leighton F, San Martı́n A, Castillo O, Pollak F, Perez DD, Strobel P, et al.
Red wine, white wine and diet, intervention study. Effect on cardiovascular risk factors. Proc XXV Congrès Mondial de la Vigne et du Vin, Parı́s,
June 19–23, 2000. pp. 63–70.
4. Papadaki A, Scott JA. The Mediterranean eating in Scotland experience
project: evaluation of an Internet-based intervention promoting the Mediterranean diet. Br J Nutr. 2005;94:290–298.
5. Hannuksela ML, Ramet ME, Nissinen AE, Liisanantti MK, Savolainen
MJ. Effects of ethanol on lipids and atherosclerosis. Pathophysiology.
2004;10:93–103.
6. Gong M, Wilson M, Kelly T, Su W, Dressman J, Kincer J, et al. HDLassociated estradiol stimulates endothelial NO synthase and vasodilation
in an SR-BI-dependent manner. J Clin Investig. 2003;111:1579–1587.
7. Salem RO, Laposata M. Effects of alcohol on hemostasis. Am J Clin
Pathol. 2005;123(Suppl):S96–105.
8. Mezzano D, Leighton F, Martinez C, Marshall G, Cuevas A, Castillo O,
et al. Complementary effects of Mediterranean diet and moderate red
wine intake on haemostatic cardiovascular risk factors. Eur J Clin Nutr.
2001;55:444–451.
9. de Gaetano G, Di Castelnuovo A, Donati MB, Iacoviello L. The mediterranean lecture: wine and thrombosisdfrom epidemiology to physiology
and back. Pathophysiol Haemost Thromb. 2003;33:466–471.
10. Libby P. Inflammation and cardiovascular disease mechanisms. Am J Clin
Nutr. 2006;83(Suppl):456S–460S.
11. Curin Y, Andriantsitohaina R. Polyphenols as potential therapeutical agents
against cardiovascular diseases. Pharmacol Rep. 2005;57(Suppl):97–107.
12. Estruch R, Sacanella E, Badia E, Antunez E, Nicolas JM, Fernandez-Sola J,
et al. Different effects of red wine and gin consumption on inflammatory biomarkers of atherosclerosis: a prospective randomized crossover trial. Effects
of wine on inflammatory markers. Atherosclerosis. 2004;175:117–123.
13. Williams MJ, Sutherland WH, Whelan AP, McCormick MP, de Jong SA.
Acute effect of drinking red and white wines on circulating levels of
inflammation-sensitive molecules in men with coronary artery disease.
Metabolism. 2004;53:318–323.
15. Urquiaga I, Guasch V, Marshall G, San Martin A, Castillo O, Rozowski J,
et al. Effect of Mediterranean and Occidental diets, and red wine, on plasma
fatty acids in humans. An intervention study. Biol Res.. 2004;37:253–261.
16. Wierzbicki AS, Chowienczyk PJ, Cockcroft JR, Brett SE, Watts GF, Jenkins BS, et al. Cardiovascular risk factors and endothelial dysfunction.
Clin Sci (Lond). 2004;107:609–615.
17. Dupuis J. Noninvasive evaluation of endothelial vascular reactivity: should
the quest continue? Can J Cardiol. 2005;21:1047–1051.
18. Cuevas AM, Guasch V, Castillo O, Irribarra V, Mizon C, San Martin A,
et al. A high-fat diet induces and red wine counteracts endothelial dysfunction in human volunteers. Lipids. 2000;35:143–148.
19. Shimada K, Watanabe H, Hosoda K, Takeuchi K, Yoshikawa J. Effect of
red wine on coronary flow-velocity reserve. Lancet. 1999;354:1002.
20. Zilkens RR, Burke V, Hodgson JM, Barden A, Beilin LJ, Puddey IB. Red
wine and beer elevate blood pressure in normotensive men. Hypertension.
2005;45:874–879.
21. Papamichael C, Karatzis E, Karatzi K, Aznaouridis K, Papaioannou T, Protogerou A, et al. Red wine’s antioxidants counteract acute endothelial dysfunction
caused by cigarette smoking in healthy nonsmokers. Am Heart J. 2004;147:E5.
23. Leighton F, Miranda-Rottmann S, Urquiaga I. A central role of eNOS in
the protective effect of wine against metabolic syndrome. Cell Biochem
Funct. 2006;24:291–298.
24. Miller PM, Anton RF, Egan BM, Basile J, Nguyen SA. Excessive alcohol
consumption and hypertension: clinical implications of current research.
J Clin Hypertens. 2005;7:346–351.
25. Rosito GA, Fuchs FD, Duncan BB. Dose-dependent biphasic effect of ethanol on 24-h blood pressure in normotensive subjects. Am J Hypertens.
1999;12:236–240.
26. Thadhani R, Camargo CA Jr, Stampfer MJ, Curhan GC, Willett WC,
Rimm EB. Prospective study of moderate alcohol consumption and risk
of hypertension in young women. Arch Intern Med. 2002;162:569–574.
27. Stranges S, Wu T, Dorn JM, Freudenheim JL, Muti P, Farinaro E, et al.
Relationship of alcohol drinking pattern to risk of hypertension: a population-based study. Hypertension. 2004;44:813–819.
28. Vadstrup ES, Petersen L, Sorensen TI, Grønbæk M. Waist circumference in
relation to history of amount and type of alcohol: results from the Copenhagen City Heart Study. Int J Obes Relat Metab Disord. 2003;27:238–246.
29. Tsang C, Higgins S, Duthie GG, Duthie SJ, Howie M, Mullen W, et al.
The influence of moderate red wine consumption on antioxidant status
and indices of oxidative stress associated with CHD in healthy volunteers.
Br J Nutr. 2005;93:233–240.
30. Leighton F, Cuevas A, Guasch V, Perez DD, Strobel P, San Martin A,
et al. Plasma polyphenols and antioxidants, oxidative DNA damage and
endothelial function in a diet and wine intervention study in humans.
Drugs Exp Clin Res.. 1999;25:133–141.
31. Natella F, Ghiselli A, Guidi A, Ursini F, Scaccini C. Red wine mitigates
the postprandial increase of LDL susceptibility to oxidation. Free Radic
Biol Med. 2001;30:1036–1044.
32. Lee KW, Hill JS, Walley KR, Frohlich JJ. Relative value of multiple
plasma biomarkers as risk factors for coronary artery disease and death in
an angiography cohort. Can Med Assoc J. 2006;174:461–466.
33. van der Gaag MS, Ubbink JB, Sillanaukee P, Nikkari S, Hendriks HF. Effect of consumption of red wine, spirits, and beer on serum homocysteine.
Lancet. 2000;355:1522.
34. Dixon JB, Dixon ME, O’Brien PE. Reduced plasma homocysteine in obese
red wine consumers: a potential contributor to reduced cardiovascular risk
status. Eur J Clin Nutr. 2002;56:608–614.
Panel Discussion I: Does Alcohol Consumption Prevent
Cardiovascular Disease?
The panel on the relationship between alcohol consumption and cardiovascular disease was chaired by R.
Curtis Ellison and the panelists were Goya Wannamethee, Eric B. Rimm, Kenneth J. Mukamal, Morten
Grønbæk, Kaye M. Fillmore, Francois M. Booyse, Federico Leighton, and Murray A. Mittleman. This discussion was centered initially on whether the putative protection of moderate alcohol consumption against
coronary heart disease (CHD) reported in epidemiological studies is due to methodologic ‘‘errors.’’
Ann Epidemiol 2007;17:S37–S39 Ó 2007 Elsevier Inc. All rights reserved.
‘‘ERRORS’’ IN OBSERVATIONAL STUDIES AND
THE ASSOCIATION BETWEEN ALCOHOL AND
CORONARY HEART DISEASE
Wannamethee began the discussion by stating that from the
evidence currently available, there may well be protection
against coronary heart disease (CHD) from moderate drinking, and many mechanisms for such an action have been
shown. However, she believed that the magnitude of the
protection remains a question, and there is still a problem
with epidemiological studies that use ‘‘nondrinkers’’ as controls. Even if the problem with ‘‘sick quitters’’ being included
in this group is being addressed in newer studies, other factors, many of which we may not yet understand, may represent confounding when it comes to nondrinkers.
Mittleman stated that epidemiologists address known
biases, but cannot completely eliminate all sources of bias,
especially from lifestyle factors. However, the central question one should consider is: To what extent does bias affect
the results of our studies? He asserted that there is good evidence that potential biases and residual confounding probably do not overcome the protective effects observed in most
epidemiological studies. He further commented that future
studies should also evaluate the relation between alcohol
drinking pattern and the risk of CHD. While carrying out
randomized trials for cardiovascular end points is daunting,
he suggested that intermediate-term trials may be feasible
and results from such trials would help us understand the biological mechanisms.
To respond to Fillmore’s criticisms of methods in observational studies, and ‘‘errors’’ in the assessment of the health
effects of light-to-moderate alcohol consumption, Mittleman said that most of the harmful effects from moderate
drinking that she demonstrated were related to cancer and
that, in fact, prospective studies also show such adverse effects. Fillmore insisted that better measures of the duration
of drinking and exposure over longer periods of observation
are still needed to help resolve the issue. Ellison countered
by pointing out that in analyses among the generally light
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
drinkers in the Framingham Heart Study (where there are
repeated assessments of alcohol intake), similar results are
usually found regardless of whether one uses alcohol data assessed at baseline, average data over time, or the most recent
updated data prior to the occurrence of CHD as the exposure
variable. Wannamethee stated that heavier drinkers are
more likely than others to stop drinking completely at
some point in their drinking career; as people age, many of
them become light drinkers. She added that taking change
into account may have little effect for light drinkers, but
does among heavy drinkers.
Arthur Klatsky commented: ‘‘We all agree that ‘sick quitters’ are inappropriate as controls; but I am not concerned
about including occasional drinkers in the referent group.
Our own studies find that about 20% of the drinkers consume alcohol infrequently, perhaps once a month, and we
have consistently found that these infrequent drinkers
have essentially the same risk as lifetime abstainersdnot
at higher or lower risk.’’ He then asked Fillmore if she could
cite any study that finds infrequent drinkers to be at higher
risk, or if there was any scientific evidence suggesting that
including occasional drinkers in the reference group with
abstainers is really a methodologic ‘‘error,’’ as she had
suggested.
Fillmore replied that ‘‘occasional drinkers’’ may include
mainly individuals who have always been occasional
drinkers, but also some who have decreased their consumption because of illness or infirmity. In her view, there is
sufficient evidence from numerous longitudinal studies
showing that the decline in consumption with increasing
age and illness is a process. The meta-analytic efforts conducted by her group were a crude effort to take this process
into account. Rimm said that his own studies had found that
using light drinkers (rather than abstainers) as the referent
group still yielded a dose-response curve of increasing protection against CHD with increasing alcohol consumption.
To him, this suggested that there is a biological basis for the
protective effects. Wannamethee added that, similar to the
results of Klatsky, the occasional drinkers included in her
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.008
S38
DOES ALCOHOL CONSUMPTION PREVENT CARDIOVASCULAR DISEASE?
own studies were not sick people. However, she stated that
they differ from ‘‘abstainers’’ who, in some studies, may include former heavy drinkers.
JUDGING THE EVIDENCE THAT ALCOHOL
CONSUMPTION DECREASES THE RISK OF CHD
Klim McPherson stated that he had come to the conference
with the assumption that moderate alcohol consumption
does decrease the risk of CHD, by about 10% to 15%, and
what he had heard confirmed this belief. But he pointed to
lessons learned from the presumed protective effects against
CHD of hormone replacement therapy (HRT), which had
not been supported by a large clinical trial, the Women’s
Health Initiative (WHI). He said that we should bear these
and similar results in mind as we interpret findings on alcohol and CHD. Rimm responded that the WHI trial should
remind us that a single clinical trial on alcohol and CHD
will not answer all the questions. He added that the WHI
was done among women at an average age of 65 years, and
in reality very few women start taking HRT 15 years after
menopause. When women in WHI were stratified by age,
there was actually less harm and even some benefit among
the youngest women. ‘‘This points out the danger of having
only one clinical trial of alcohol and CHD and using its results as the ‘final answer.’ Any results from such a trial may
well apply only to a subset of the populationdthose with
a certain lifestyle, certain diet, certain genetics, certain
age group, etc.dso we should not expect that a single trial
will be applicable to everyone.’’
Mukamal added that for HRT, or vitamin E, or almost any
other lifestyle factor, overall health may be more favorable
among the exposed group, making comparisons much harder.
However, this is not the case for alcohol consumption. Smoking, for example, is much more common among drinkers than
abstainers, yet the beneficial effects of alcohol on CHD still
remain even though we realize that confounding from smoking may be incompletely controlled for in many studies.
Thomas Stuttaford added that proof that alcohol reduces
the risk of CHD depends on what ‘‘standard of evidence’’
one uses. There are two different standards of acceptance:
if it is a criminal case, the standard is ‘‘beyond reasonable
doubt’’; in a commercial case, it is ‘‘the balance of probability.’’ What matters to clinicians is very much the balance of
probability. On the other hand, epidemiologists are like the
criminal lawyers and want the evidence to be beyond reasonable doubtdoften an impossibility. Lionel Tiger pointed
out that there is almost a ‘‘disconnect’’ between research evidence and advice given to patients, as Stuttaford had suggested. Richard Smallwood added that those setting
government policy are motivated not by the balance of
AEP Vol. 17, No. 5S
May 2007: S37–S39
probabilities or beyond reasonable doubt, but by ‘‘beyond
reasonable criticism.’’
Smallwood repeated the key question: ‘‘For alcohol and
heart disease, are we now ‘beyond reasonable doubt’ that
there is a protective effect?’’ Rimm responded that the answer to that question depends on whom one asks. Among
cardiovascular epidemiologists, most believe it is indeed beyond a reasonable doubt based on the evidence from both
epidemiological and basic science findings. He continued:
‘‘We call ourselves experts and at some point we have to
conclude that there is enough evidence to say that alcohol
reduces the risk of CHD. Even with 15 clinical trials over
15 or more years, we will never have enough evidence to
convince some people. Besides, we all come with our own
biases: people working in addiction clinics will have different biases from those working with heart disease prevention.
But recall,’’ he added, ‘‘that stating that the science is strong
enough to show such protection does not necessarily relate
directly to policy decisions, as we will hear later.’’
AGE TO BEGIN DRINKING
Ellison asked if there was any evidence which suggested advantages from starting alcohol intake before one reached
middle age. Leighton responded that the short-term effects
are only one part of the story, and there may be long-term
effects of alcohol that we do not yet understand. He pointed
out that atherosclerosis begins at younger ages and there is
a need to control for that as early as possible. Measures
such as endothelial function, although cumbersome to
undertake, would greatly enhance our understanding of
the effects of moderate drinking and the age at which they
are seen.
DIFFERENCES AMONG POPULATIONS
Rehm brought up three points to be considered. First, the
majority of people in the world do not drink alcohol. In
‘‘drinking countries,’’ there may not be enough differences
between people for good comparisons. Hence more studies
should look at the developing world where drinking is
not the norm. Second, the decline in drinking with age
is a very tricky feature. In North America, the heaviest
drinking is among the young; in most European countries,
however, heavy drinking continues even into the 50s and
60s, and there may be relatively less heavy drinking in
early adulthood. Thus it is important to study specific cohorts in different countries. Third, the relative risk for
CHD from alcohol intake goes down as people age. Therefore a 90-year-old person would not have the same relative
protection against death as a younger person. Mittleman
disagreed with the last statement, stating that as the risk
AEP Vol. 17, No. 5S
May 2007: S37–S39
DOES ALCOHOL CONSUMPTION PREVENT CARDIOVASCULAR DISEASE?
for mortality approaches 100%, the estimated effects on
the relative risk of any single factor become smaller. However, alcohol protects against CHD at all ages and, in fact,
Rehm’s own studies show that the beneficial effect of
alcohol on the risk of CHD is not reduced among older
people.
Wannamethee noted that most of the studies showing
a protective effect of alcohol against CHD come from individuals who are of Caucasian origin, and such a protective effect may not apply to other groups, such as African
Americans. Others stated that there are limited data available on minorities, and much more research is needed.
Dwight Heath pointed out a similar problem in many
parts of the developing world, in that very sophisticated
and rigorous studies exist only from a few countries. Alcohol data from much of the rest of the world are often
based on social surveys or hospital records, making it difficult to evaluate the relation of alcohol to disease among
such populations.
Ricardo Brown said that this discussion reminded him of
diabetes research. ‘‘In diabetes, we have come up with
hemoglobin A1C as a biomarker of severity. In alcohol
research, there is great variability in what is defined as
‘moderate drinking.’ There is also variability in what a standard drink is. None of us seems to know of a stable biomarker
that would be more reliable as an indicator of moderate
drinking than data from questionnaires.’’
GENETIC EFFECTS
Henk Hendriks mentioned that genetic studies have shown
that polymorphisms of the alcohol dehydrogenase gene,
which regulates alcohol metabolism, modify the association
between alcohol and CHD, which adds to the evidence that
alcohol reduces the risk of CHD. Rimm stated that although
genetic studies were very important, it would be simplistic to
say that one gene determines the effect of alcohol, as the
body tends to compensate, and many other genetic factors
will undoubtedly be found. Booyse cautioned that it is
important to keep in mind that thousands of genes are activated by alcohol or polyphenols, working together to produce a final effect. The significance of individual functional
changes still needs to be understood. At present, it would be
an error to pick and choose among genesdit is a matter of
accumulating data.
S39
TRANSLATING RESEARCH ON ALCOHOL
AND HEALTH
Nady el-Guebaly offered that there is a considerable difference between advice given by a doctor to a patient and that
given to a population. Any discussion of moderate drinking,
he observed, may be more productive at the individual level.
Ellison asked if, as a basis for giving advice on drinking, there
was any need for further observational epidemiological studies on alcohol and CHD. Richard Harding responded that
one of the purposes of epidemiological studies is to inform
mechanistic studies, such as those described by Booyse and
Leighton. These findings, in turn, should be used to inform
the design of future epidemiological studies, so both types of
study are needed. Advice to the public should be based on
the totality of evidence.
For determining what advice to give regarding drinking,
Rimm pointed out difficulties in observational epidemiological studies. ‘‘Among the ‘healthiest people’ in our cohorts
(who are lean, do not smoke, who exercise, and who eat
a healthy diet.), the risk of CHD is very low to begin with.
Among such individuals, it might be difficult to show an effect of moderate drinking. Drinking should not be viewed as
a substitute for addressing risk factors such as obesity or lack
of exercise.’’ He concluded that if a doctor does not mention
to a patient to stop smoking or to lose weight but gives only
one recommendation, to drink moderately, the physician
would be making a mistake. Ellison challenged his conclusion: ‘‘If a person cannot stop smoking or cannot lose weight,
he or she might well benefit from having a drink a day.’’
Rimm replied that that was true, but that other practices,
such as not smoking and not being obese, impact many
health conditions. ‘‘For example, smoking has dramatically
strong effects on mortality, much greater than drinking.’’
He added that there may be the potential problem of the patient focusing only on the drinking advice and not worrying
about other lifestyle factors. But he concluded that his own
data have shown that even individuals who meet all other
criteria for a ‘‘healthy lifestyle’’ have better health outcomes
if they drink moderately.
Mittleman added that even though all of the genetic factors involved in the protective effect of alcohol on CHD
may not be fully understood, current evidence supports an
association, and it cannot all be explained by bias. ‘‘This relationship underscores the importance of reconciling basic
and epidemiological research and the need for translational
research that helps inform the clinician who treats patients
as to the proper advice regarding alcohol consumption.’’
Moderate Alcohol Consumption and Insulin Sensitivity:
Observations and Possible Mechanisms
HENK F.J. HENDRIKS, PHD
Light to moderate alcohol consumption is associated with a reduced risk for cardiovascular diseases. Epidemiologic studies, like our analysis of the European Prospective Investigation into Cancer and Nutrition
study, suggest that moderate alcohol consumption is also associated with a reduced risk of type 2 diabetes,
reported for various populations. This risk reduction may be explained by an increase in insulin sensitivity
after moderate alcohol consumption. Indeed, a positive association between alcohol consumption and insulin sensitivity is consistently reported in cross-sectional studies. Mechanisms for the effect of alcohol on
insulin sensitivity may include modulation of changes in the endocrine functioning of fat tissue, modulation
of the inflammatory status of several organs, and/or modulation of glucose and fatty acid metabolism.
Ann Epidemiol 2007;17:S40–S42. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS: Alcohol Drinking, Diabetes Mellitus, Insulin, Adiponectin, Epidemiology, Mechanism,
Nutrition Intervention, Human.
ALCOHOL AND DIABETES
The prevalence of type 2 diabetes is rising. Between 2000
and 2030, a 37% increase in the worldwide prevalence of diabetes is expected (1). Next to aging of the population, the
lack of physical activity and high energy intake leading to
overweight and obesity have been shown to be largely responsible for the so-called obesity epidemic. Moderate alcohol consumption has been suggested to be a lifestyle factor
that may modify the risk of developing type 2 diabetes.
A meta-analysis was performed based on all epidemiologic studies available describing the association between
moderate alcohol consumption and type 2 diabetes risk
(2). The 15 original cohort studies that were included comprise 11,959 incident cases of type 2 diabetes in 369,682 individuals who, on average, were followed up for 12 years. A
U-shaped association was obtained after pooling all data,
suggesting a 30% reduced risk of type 2 diabetes in moderate
alcohol consumers, whereas no risk reduction was observed
in those consuming more than 48 g of alcohol per day. The
lower type 2 diabetes risk was consistent over most included
studies, but risk estimates differed more across studies than
expected. This appeared not to be caused by multiple adjustments, the definition of type 2 diabetes used, or by differences between men and women and low and high body
mass index. In the absence of long-term randomized intervention studies, a meta-analysis can be considered the best
available evidence.
However, relatively few data were available in the metaanalysis on the association between moderate drinking and
the incidence of type 2 diabetes in women. Therefore this
association was studied in one of the Dutch Prospect-EPIC
(European Prospective Investigation into Cancer and Nutrition) cohorts consisting of 16,330 women aged 49 to 70
years, who were free from diabetes and were followed up
for 6.2 years (3). During follow-up, 760 new cases of type
2 diabetes were documented. A linear inverse association
between alcohol consumption and type 2 diabetes risk was
observed. The hazard ratio for type 2 diabetes risk varied between 0.9 and 0.6 up to consumption levels of 210 g of alcohol per week. Beverage type seemed not to affect the
association. Lifetime alcohol consumption was associated
with type 2 diabetes in a U-shaped fashion. These findings
support the evidence of a decreased risk of type 2 diabetes
with moderate alcohol consumption and expand the finding
to a population of older women.
In an additional meta-analysis, the relationship between
alcohol consumption and coronary heart disease and mortality in type 2 diabetic patients was investigated. Statistical
pooling showed lower risks in those who consumed alcohol
than in those who did not for total mortality, fatal and total
coronary heart disease (4).
POSSIBLE MECHANISMS IN TYPE 2
DIABETES REDUCTION
From TNO Quality of Life, Zeist, Netherlands.
Address correspondence to: Henk Hendriks, PhD, TNO Quality of Life,
PO Box 360, Zeist 3700 AJ, Netherlands. E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
A reduced type 2 diabetes risk in moderate alcohol consumers may be explained by an increase in insulin sensitivity. Indeed, an association between alcohol consumption
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.009
AEP Vol. 17, No. 5S
May 2007: S40–S42
Selected Abbreviations and Acronyms
CRP Z C-reactive protein
EPIC Z European Prospective Investigation into Cancer and Nutrition
HOMA Z homeostatic model assessment
and insulin sensitivity is consistently reported in cross-sectional studies (5, 6).
Randomized controlled trials, however, report contradictory results. An alcohol intake reduction study (7) showed
that high-density lipoprotein cholesterol and liver enzymes
were reduced in healthy young men, but the intervention
did not affect the insulin sensitivity index (assessed by clamp
technique) or homeostatic model (HOMA) assessment, calculated on the basis of fasting insulin and glucose. A borderline increase in an insulin-resistant subgroup was observed
in a study using 23 healthy middle-aged men consuming
40 g of whisky with dinner daily for 17 days (vs. tap water
for 17 days) (8). Insulin sensitivity index improvement correlated positively with the relative alcohol-induced increase
in plasma adiponectin level in that study. Another study
showed that consumption of 30 g of alcohol per day for an
8-week period had beneficial effects on insulin and triglyceride concentrations and insulin sensitivity in a group of 51
nondiabetic postmenopausal women (9).
Possible reasons for the variations in study outcome may
be related to the outcome measurement analyzed, the duration of the intervention, the dose of alcohol used, and the
group of subjects chosen. Additional randomized controlled
trials are needed to further substantiate improved insulin
sensitivity in moderate alcohol consumers.
Insulin sensitivity may be affected by moderate alcohol
consumption through several mechanisms. Possible mechanisms include modulation of changes in the endocrine functioning of fat tissue, modulation of the inflammatory status
of several organs, or modulation of metabolism.
Fat tissue has been appreciated more over the last several
years as an organ that is actively involved in communication
with other organs through endocrine and other mechanisms, for example, through the production of adipocytokines, such as adiponectin. Plasma adiponectin
concentrations are significantly lower in subjects with obesity and type 2 diabetes. Also, low fasting plasma adiponectin concentrations are associated with high basal and low
insulin-stimulated skeletal muscle insulin receptor tyrosine
phosphorylation, which is one possible cause of decreased
insulin sensitivity.
Epidemiologic studies have shown that moderate alcohol
consumption is associated with increased adiponectin concentrations (10). Randomized control trials have shown
that moderate alcohol consumption will increase plasma
adiponectin levels by about 11% (8, 11). In these trials,
increases in plasma adiponectin concentrations correlated
Hendriks
ALCOHOL INTAKE AND INSULIN SENSITIVITY
S41
with increased insulin sensitivity. Adiponectin is thought
to improve insulin sensitivity by suppression of glucose production, or increased glucose uptake and fatty acid oxidation
in muscle tissue. Possibly, other adipokines like leptin, resistin, and acylation-stimulating protein are also affected.
Alternatively, inflammation of adipose tissue may cause
insulin resistance. Interestingly, there appears to be an association between moderate alcohol consumption and inflammatory status. C-reactive protein (CRP) is one of the
inflammatory biomarkers, being a nonspecific acute-phase
protein produced by the liver in response to tissue injury, infection, and inflammation. A mild CRP elevation within
the normal, nonacute-phase range has recently emerged as
a valuable marker of cardiovascular risk. Biomarkers for inflammation have been shown to be affected by moderate alcohol consumption both in epidemiologic studies as well as
in randomized controlled trials. These include CRP (12, 13)
and several others.
A final possible mechanism by which moderate alcohol
consumption may affect insulin sensitivity is through
a change in intermediary metabolism. Studies are now under
way to investigate the effects of moderate alcohol consumption on glucose and fatty acid metabolism in plasma and
muscle. These studies may use several genotypes of alcohol
metabolizing enzymes generating differing levels of alcohol
metabolites which may modulate insulin sensitivity.
The contributions of Martijn van der Gaag, Aafje Sierksma, and Joline
Beulens to this research are greatly acknowledged. We thank Eric
Rimm, Ken Mukamal, and Lando Koppes for collaboration.
REFERENCES
1. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care.
2004;27:1047–1053.
2. Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Moderate
alcohol consumption lowers the risk of type 2 diabetes: a meta-analysis
of prospective observational studies. Diabetes Care. 2005;28:719–725.
3. Beulens JW, Stolk RP, van der Schouw YT, Grobbee DE, Hendriks HF,
Bots ML. Alcohol consumption and risk of type 2 diabetes among older
women. Diabetes Care. 2005;28:2933–2938.
4. Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Meta-analysis
of the relationship between alcohol consumption and coronary heart
disease and mortality in type 2 diabetic patients. Diabetologia. 2006;49:
648–652.
5. Bell RA, Mayer-Davis EJ, Martin MA, D’Agostino RB Jr, Haffner SM.
Associations between alcohol consumption and insulin sensitivity and
cardiovascular disease risk factors: the Insulin Resistance and Atherosclerosis Study. Diabetes Care. 2000;23:1630–1636.
6. Kiechl S, Willeit J, Poewe W, Egger G, Oberhollenzer F, Muggeo M, et al.
Insulin sensitivity and regular alcohol consumption: large, prospective,
cross sectional population study (Bruneck study). BMJ. 1996;313:1040–
1044.
7. Zilkens RR, Burke V, Watts G, Beilin LJ, Puddey IB. The effect of alcohol
intake on insulin sensitivity in men: a randomized controlled trial. Diabetes Care. 2003;26:608–612.
S42
Hendriks
ALCOHOL INTAKE AND INSULIN SENSITIVITY
AEP Vol. 17, No. 5S
May 2007: S40–S42
8. Sierksma A, Patel H, Ouchi N, Kihara S, Funahashi T, Heine RJ, et al.
Effect of moderate alcohol consumption on adiponectin, tumor necrosis
factor-alpha, and insulin sensitivity. Diabetes Care. 2004;27:184–189.
11. Beulens JW, van Beers RM, Stolk RP, Schaafsma G, Hendriks HF. The effect of moderate alcohol consumption on fat distribution and adipocytokines. Obesity (Silver Spring). 2006;14:60–66.
9. Davies MJ, Baer DJ, Judd JT, Brown ED, Campbell WS, Taylor PR. Effects
of moderate alcohol intake on fasting insulin and glucose concentrations
and insulin sensitivity in postmenopausal women: a randomized controlled
trial. JAMA. 2002;287:2559–2562.
12. Imhof A, Froehlich M, Brenner H, Boeing H, Pepys MB, Koenig W. Effect
of alcohol consumption on systemic markers of inflammation. Lancet.
2001;357:763–767.
10. Shai I, Rimm EB, Schulze MB, Rifai N, Stampfer MJ, Hu FB. Moderate
alcohol intake and markers of inflammation and endothelial dysfunction
among diabetic men. Diabetologia. 2004;47:1760–1767.
13. Sierksma A, van der Gaag MS, Kluft C, Hendriks HF. Moderate alcohol
consumption reduces plasma C-reactive protein and fibrinogen levels;
a randomized, diet-controlled intervention study. Eur J Clin Nutr.
2002;56:1130–1136.
Moderate Alcohol Consumption and Risk of Developing Dementia
in the Elderly: The Contribution of Prospective Studies
LUC LETENNEUR, PHD
Moderate alcohol consumption, after controlling for potential confounding factors, has been found to be
associated with a lower risk of developing dementia in several prospective epidemiological studies from Europe, the United States, and China. When the type of alcoholic beverage consumed is analyzed, moderate
wine intake has been systematically associated with lower risk. However, moderate consumption has very
different definitions across studies, ranging from monthly or weekly drinking to 3 to 4 drinks per day. In
addition, different results have been observed according to sex; some studies found the same effect in
men and women, while others found either no association or a stronger association in women. All of these
results lead to the conclusion that the observed association is fragile and needs further confirmation.
Ann Epidemiol 2007;17:S43–S45. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Dementia, Alzheimer Disease.
Dementia is the most common disorder affecting the brain
in older people. Epidemiological studies have reported several risk factors and a consensus has emerged that sex, education, and dietary and vascular factors are likely to be
important. Among dietary factors, alcohol consumption
was found to be associated with a lower risk for dementia
in some studies. We shall review several prospective studies
on this topic.
The first prospective study to explore alcohol intake was
the PAQUID (Personnes Agees QUID) program (1). A
sample of 3777 subjects aged 65 years or older was followed
up for 3 years and 99 incident cases of dementia were diagnosed. Alcohol consumption data were collected at baseline, with wine the main type of alcohol consumed,
usually on a daily basis. Four categories of individuals were
defined: nondrinkers, mild drinkers (consuming 1 or 2 drinks
per day), moderate drinkers (consuming 3 or 4 drinks per
day) and heavy drinkers (consuming more than 4 drinks
per day). Lower risks of developing dementia were found
among drinkers compared with nondrinkers, but the relationship was significant only for moderate drinkers (mild
drinkers: risk ratio [RR] Z 0.81; moderate drinkers: RR Z
0.19; heavy drinkers: RR Z 0.31). No modification effect
was found according to sex and the association did not
change after adjusting for age, sex, education, occupation,
and baseline cognition.
From INSERM, U593; Université Victor Segalen. Bordeaux, France.
Address correspondence to: Luc Letenneur, PhD, INSERM U593, Case
11, Université Victor Segalen, 146 rue Léo Saignat, 33076 Bordeaux
Cedex, France. E-mail: [email protected].
Disclosure: Dr. Letenneur received a grant from the Office National
Interprofessionnel des Vins (ONIVINS) to analyze the association between
alcohol consumption and cardiovascular diseases. It was not linked to the
study of dementia.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
The Rotterdam Study followed 5395 subjects aged 55
years or older over a period of 6 years; 197 incidents cases
of dementia were diagnosed (2). The number of drinks of
alcohol (beer, wine, fortified wine, or spirits) was collected
at baseline and 5 categories of intake were studied: no
drinks consumed, less than 1 drink per week, more than
1 drink per week but less than 1 per day, 1 to 3 drinks
per day, more than 3 drinks per day. The risk of developing
dementia was lower among drinkers compared with nondrinkers (Table 1) and was significant in the 1 to 3 drinks
per day category. The pattern was different in men and
women. No association was found in women, whereas
a lower risk was found for men drinking 1 to 3 drinks
per day. A modification effect was found when the Apolipoprotein E4 allele (ApoE4) was taken into consideration;
the risk was lower among drinkers with an ApoE4 allele,
whereas it was less clear for drinkers without the ApoE4
allele (Table 1). No difference was found according to
beverage type, although beer tended to give marginally
lower risk than wine.
In a prospective study of elderly people living in North
Manhattan (3), 2126 subjects were followed up for 4 years
and 260 incident cases of dementia were diagnosed. The
number of drinks per week was collected at baseline and subjects were classified as nondrinkers, light drinkers (less than
1 drink per month to 6 drinks a week), moderate drinkers
(1–3 drinks a day), and heavy drinkers (more than 3 drinks
a day). Light and moderate categories were aggregated because of a low number of moderate drinkers. Indeed, in
this sample, 70% of the subjects were nondrinkers. When
analyzing the association between each alcoholic beverage
type and dementia, it was found that wine was significantly
associated with a lower risk among light to moderate
drinkers (hazard ratio [HR] Z 0.64, p Z 0.018). When
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.010
S44
Letenneur
ALCOHOL CONSUMPTION AND DEMENTIA
Selected Abbreviations and Acronyms
Apo E4 Z apolipoprotein E4 allele
HR Z hazard ratio
OR Z odds ratio
PAQUID Z Personnes Agees QUID
RR Z risk ratio
analyzing the risk of Alzheimer’s disease adjusted for age and
sex, a decreased risk was observed in wine drinkers (HR Z
0.59, p Z 0.018), but the association became insignificant
when education and the ApoE4 genotype (HR Z 0.69,
p Z 0.11) were included. The HRs were greater than 1 for
light to moderate beer or spirits drinkers (beer: HR Z
1.39, p Z 0.094, spirits: HR Z 1.34, p Z 0.152). When
wine, beer, and spirits were analyzed simultaneously with
full adjustment, the risk for Alzheimer’s disease was lower
in wine drinkers (HR Z 0.55, p Z 0.015), but higher for
beer (HR Z 1.47, p Z 0.065) or spirits (HR Z 1.51, p Z
0.062) drinkers. A modification effect was found with the
ApoE4 genotype. A significantly lower risk of dementia
was found in light to moderate wine drinkers without an
ApoE4 allele (HR Z 0.44, p Z 0.004) compared with nondrinkers, whereas the association disappeared for ApoE4 allele bearers (HR Z 1.10, p Z 0.093). No modification effect
by sex was found.
The association between alcohol intake and risk for dementia was also examined in studies originally designed to
explore cardiovascular events. During follow-up, cognitive
functioning was explored and nested case-control studies
were performed. In the Copenhagen City Heart study (4),
a nested case-control included 83 cases of dementia and
1626 controls. Alcohol intake was collected in two ways:
the number of drinks per week (less than 1, 1–7, 8–14,
15–21, 22 or more) and the frequency of intake (never/
hardly ever, monthly, weekly, daily). No association was
found between the number of drinks of alcohol consumed
per week and the risk of dementia. When beer, wine, and
spirits intake were analyzed simultaneously, a reduced risk
was observed only for wine drinkers (monthly: HR Z 0.43
[0.23–0.82]; weekly: HR Z 0.33 [0.13–0.86]; daily: HR Z
0.57 [0.15–2.11]). Beer drinkers tended to have a higher
risk (monthly: HR Z 2.28 [1.13–4.60]; weekly: HR Z
2.15 [0.98–4.78]; daily: HR Z 1.73 [0.75–3.99]) and no
AEP Vol. 17, No. 5S
May 2007: S43–S45
clear association was found in spirits drinkers (monthly:
HR Z 0.81 [0.42–1.57]; weekly: HR Z 1.65 [0.74–3.69];
daily: HR Z 1.12 [0.43–2.92]). No difference was found
between men and women.
Another nested case-control study was performed within
the Cardiovascular Health Study, which included 373 cases
of dementia and 373 controls (5). Levels of alcohol intake
were defined as: 0 drinks per week; less than 1 drink per
week; 1–6 drinks per week; 7–13 drinks per week; 14 or
more drinks per week. The association between alcohol intake and the risk of dementia followed a J-shaped curve, with
a nadir for the category of 1–6 drinks per week (Table 2).
The pattern was different for men and women: all drinker
categories were associated with a lower risk in women,
whereas a J-shaped curve was found for men (Table 2). A
modification effect was observed by ApoE4; when the
ApoE4 allele was absent, the risk was significantly lower
among subjects who consumed 1 to 6 drinks per week.
When the ApoE4 allele was present, the HR was below
1.00 only for light drinkers and above 1.00 for heavier
drinkers (Table 2). The odds of dementia were lower (although not significantly) for wine drinkers (less than 1 drink
per week: HR Z 0.72 [0.46–1.11]; 1–6 drinks per week:
HR Z 0.72 [0.39–1.33]; more than 6 drinks per week:
HR Z 0.62 [0.25–1.50]). However, the trend was not the
same for beer (less than 1 drink per week: HR Z 0.84
[0.48–1.47]; 1–6 drinks per week: HR Z 0.74 [0.36–1.54];
more than 6 drinks per week: HR Z 1.96 [0.71–5.47]) or
spirits drinkers (less than 1 drink per week: HR Z 0.84
[0.48–1.45]; 1–6 drinks per week: HR Z 1.17 [0.59–2.30];
more than 6 drinks per week: HR Z 1.08 [0.55–2.13]).
Several other prospective studies have reported an association between alcohol consumption and dementia. A Canadian study (6) reported that at least weekly consumption
of alcohol was associated with a decreased risk of Alzheimer’s disease (odds ratio [OR] Z 0.68, [0.47–1.00]. In
Sweden (7), the risk of dementia was estimated to be 0.5
(0.3–0.7) among light to moderate drinkers (1–21 drinks
per week in men, 1–14 drinks per week in women). In China
(8), light to moderate drinkers (1–21 drinks per week
in men, 1–14 drinks per week in women) had a lower
risk (RR Z 0.52 [0.32–0.85]) than nondrinkers, but a nonsignificant increased risk was observed in heavy drinkers
TABLE 1. Hazard ratios (and 95% confidence intervals) of dementia according to alcohol consumption in the Rotterdam Study (2)
Total
Men
Women
ApoE4 absent
ApoE4 present
No alcohol
!1 drink/wk
>1 drink/wk but !1/d
1–3 drinks/d
>4 drinks/d
1.00
1.00
1.00
1.00
1.00
0.82 (0.56–1.22)
0.60 (0.27–1.34)
0.91 (0.58–1.44)
1.26 (0.67–2.37)
0.69 (0.35–1.34)
0.75 (0.51–1.11)
0.53 (0.28–1.0)
0.91 (0.55–1.49)
1.39 (0.73–2.64)
0.46 (0.23–0.94)
0.58 (0.38–0.90)
0.40 (0.21–0.74)
0.85 (0.47–1.57)
0.67 (0.31–1.46)
0.60 (0.30–1.21)
1.0 (0.39–2.59)
0.88 (0.32–2.44)
–
–
–
AEP Vol. 17, No. 5S
May 2007: S43–S45
Letenneur
ALCOHOL CONSUMPTION AND DEMENTIA
S45
TABLE 2. Odds ratios (and 95% confidence intervals) of incident dementia according to alcohol consumption (drinks per week) in the
Cardiovascular Health Study (5)
Total
Men
Women
ApoE4 absent
ApoE4 present
None
!1
1–6
7–13
>14
1.00
1.00
1.00
1.00
1.00
0.65 (0.41–1.02)
0.82 (0.38–1.78)
0.52 (0.30–0.90)
0.56 (0.33–0.97)
0.60 (0.24–1.52)
0.46 (0.27–0.77)
0.36 (0.17–0.77)
0.57 (0.28–1.17)
0.37 (0.20–0.67)
0.62 (0.21–1.81)
0.69 (0.37–1.31)
1.42 (0.58–3.48)
0.23 (0.09–0.61)
0.64 (0.30–1.38)
1.49 (0.33–6.65)
1.22 (0.60–2.49)
2.40 (0.86–6.64)
0.39 (0.14–1.10)
0.60 (0.24–1.51)
3.37 (0.67–17.1)
(RR Z 1.45 [0.43–4.89]). A greater reduction of risk was observed for men (RR Z 0.37) than for women (RR Z 0.76).
All these studies tend to show the same result: light to
moderate alcohol consumption is associated with a lower
risk of developing dementia. Which mechanisms may be involved in the risk reduction of dementia? One possibility is
that alcohol might act by reducing cardiovascular risk factors, either through an inhibitory effect of ethanol on platelet aggregation, or through the alteration of the serum lipid
profile. A second possibility is that alcohol might have a direct effect on cognition through the release of acetylcholine
in the hippocampus. Finally, another possible mechanism is
through antioxidant activity of alcohol, particularly wine,
which has been found to have important antioxidant effects.
However, the definition of light to moderate alcohol intake varies considerably across the studies reviewed. The
classification of drinking as moderate ranges from monthly
or weekly drinking to 3 to 4 drinks per day, and many studies
reported an association for an intake of less than 1 drink per
day. As alcohol intake is self reported, it is also expected to
be underreported. However, neither a linear dose-response
nor a J-shaped curve was systematically found over all studies, and the association sometimes differed according to sex.
The type of alcohol does not appear to be consistent across
studies, yet wine intake is systematically associated with
lower risk. If alcohol per se were associated with a decreased
risk of developing dementia, the same pattern would be expected for beer and wine drinkers, yet beer has been found to
be associated with higher risk in several studies.
When analyzing these results and discrepancies, one can
wonder about the nature of the association between alcohol
consumption and the risk of dementia. It can be hypothesized that alcohol intake (especially light to moderate intake) is only a marker of a broader psychosocial behavior
that is associated with a decreased risk of developing
dementia. However, the analyses were controlled for many
other risk factors and the association with alcohol was still
significant. It is possible that important confounders (not
yet identified) were not considered, which might explain
some of the discrepancies between optimal intake, sex, or
type of alcohol. Light to moderate wine drinkers may prove
to be moderate with regard to other risk factors of dementia,
and alcohol intake would only be an indicator of such behavior. Until such factors have been identified, we must
be careful in how we interpret results relating to alcohol
consumption. People should not be encouraged to drink
more in the belief that this will protect them against
dementia.
REFERENCES
1. Orgogozo JM, Dartigues JF, Lafont S, Letenneur L, Commenges D, Salamon
R, et al. Wine consumption and dementia in the elderly: a prospective community study in the Bordeaux area. Rev Neurol. 1997;153:185–192.
2. Ruitenberg A, van Swieten JC, Witteman JC, Mehta KM, van Duijn CM,
Hofman A, et al. Alcohol consumption and risk of dementia: the Rotterdam
Study. Lancet. 2002;359:281–286.
3. Luchsinger J, Tang MX, Siddiqui M, Shea S, Mayeux R. Alcohol intake and
risk of dementia. J Am Geriatr Soc. 2004;52:540–546.
4. Truelsen T, Thudium D, Grønbæk M. Amount and type of alcohol and risk
of dementia. The Copenhagen Heart Study. Neurology. 2002;59:1313–
1319.
5. Mukamal K, Kuller L, Fitzpatrick A, Longstreth W, Mittleman M, Siscovick
D. Prospective study of alcohol consumption and risk of dementia in older
adults. JAMA. 2003;289:1405–1413.
6. Lindsay J, Laurin D, Verreault R, Hébert R, Helliwell B, Hill G, et al. Risk
factors for Alzheimer’s disease: A prospective analysis of the Canadian
Study of Health and Aging. Am J Epidemiol. 2002;156:445–453.
7. Huang W, Qiu C, Winblad B, Fratiglioni L. Alcohol consumption and incidence of dementia in a community sample aged 75 years and older. J Clin
Epidemiol. 2002;55:959–964.
8. Deng J, Zhou DH, Li J, Wang YJ, Gao C, Chen M. A 2-year follow-up study
of alcohol consumption and risk of dementia. Clin Neurol Neurosurg.
2006;108:378–383.
Moderate Alcohol Consumption and Cancer
KLIM MCPHERSON, PHD, FFPH, FMEDSCI
Alcohol consumption may increase the risk for at least six types of cancer, although the evidence is not
uniformly strong. Alcohol is not a carcinogen in standard tests but is likely a cocarcinogen, at least in
the digestive tract. The strongest association is with cancer of the oral cavity and pharynx (seven-fold increase for heavy consumption). For moderate consumption (2–3 drinks/day for men and 1–2 drinks/day for
women), the risks never increase above twofold and are mostly less than 25% above baseline. The evidence
from the epidemiological literature suggests that 25g/d of alcohol is associated with a relative risk of 1.9 for
cancers of the oral cavity and pharynx, 1.4 for cancers of the esophagus and larynx, about 10% for colorectal
cancer (mechanisms unclear), and 20% for liver cancer (mechanisms well described). The association
between alcohol and breast cancer is not strong and not necessarily causative, at least for moderate
consumption.
In the United Kingdom, cancer deaths attributable to alcohol consumption above the recommended
limits (caused mostly by colorectal cancer) are maximal for men aged 65 to 74 years and for women at
ages 75 to 84 years. It is postulated that if everyone in the United Kingdom drank ‘‘sensibly,’’ there would
be a marked increase in cancer deaths.
Ann Epidemiol 2007;17:S46–S48. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Cancer, Epidemiology.
Total alcohol consumption worldwide, including heavy
drinking, is responsible for only 5.2% of cancers in men
and 1.7% among women (1). Since the risk relationships between alcohol consumption and cancer are dose dependent,
this suggests that moderate drinking increases the risk for
only a small proportion of all cancers. In public health terms,
this relationship is of minor significance.
A recent meta-analysis of the relationships between
alcohol consumption and disease has reviewed 156 studies
involving 117,000 subjects (2). The findings on the incidence of cancer suggest a strong relationship with cancers
of the oral cavity, pharynx, esophagus, and larynx. The
observed magnitude of the association with cancers of the
colon, rectum, liver, and breast is weaker. The summary
relative risks are shown in Table 1.
The strongest relative risk is mostly seen for the rarest
cancers. Of the more common cancers (Fig. 1), namely, cancers of the colon, breast, liver, and esophagus, the relationship with moderate alcohol consumption is weaker overall
and hence less likely to be causal.
Ethanol is not a carcinogen by standard laboratory tests.
However, it may act as a cocarcinogen in the esophagus and
nonglandular parts of the stomach, as determined by animal
experiments. The mechanisms underlying the association
with colon cancer and breast cancer remain unclear and
From the University of Oxford, United Kingdom.
Address correspondence to: Klim McPherson, The John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. Tel: þ44 1865 221 021. E-mail:
[email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
thus for such moderate associations need not be causal.
Risk for both colon and rectal cancer is similar for men
and women. Whether these associations are a manifestation
of confounding with some dietary factor, smoking, other
substance misuse, or publication bias remains unresolved.
However, all are clearly plausible explanations. The evidence on the relationship of alcohol with cancer of stomach
or pancreas suggests no link (3).
The relationship between alcohol and the risk for breast
cancer remains ambiguous. Epidemiological studies suggest
an association between increased risk and heavy drinking,
although results are not consistent. In these cases, the association is more likely to be accounted for through unknown
confounders and publication bias than to be causative. However, some have argued strongly for a causal link. To date, no
evidence can usefully distinguish between the two explanations (4, 5). Assuming a causal relationship, Boffetta et al.
(1) estimate that 60% of alcohol-attributable cancers in females are due to breast cancer. However, this may be an artefact. Calculations of the effect of alcohol consumption on
cancer generally assume the relative risks to be causal. Thus
these estimates are likely to exaggerate the true effect of
alcohol on cancer among populations.
Clearly some of the epidemiological evidence indicates
a causal relationship between alcohol intake and the risk
of cancer, particularly of the upper digestive tract for which
the risk relationship with consumption is strong. It is therefore interesting to ask questions about not only how much
cancer may be caused by alcohol, but also how much may
be preventable by changes in alcohol consumption.
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.011
AEP Vol. 17, No. 5S
May 2007: S46–S48
McPherson
ALCOHOL CONSUMPTION AND CANCER
S47
TABLE 1. Relative risks (and 95% confidence intervals) of various cancer sites associated with alcohol consumption*
Alcohol consumption
Cancer site
Studies (No.)
Cases (No.)
25 g/d
50 g/d
100 g/d
15
20
14
24
8
4
12
507
3789
3233
32175
1321
1420
5360
1.86 (1.76–1.96)
1.43 (1.38–1.48)
1.39 (1.36–1.42)
1.25 (1.2–1.29)
1.19 (1.12–1.27)
1.09 (1.08–1.12)
1.05 (1.01–1.09)
3.11 (2.85–3.39)
2.02 (1.89–2.16)
1.93 (1.85–2.00)
1.55 (1.44–1.67)
1.40 (1.25–1.56)
1.19 (1.14–1.24)
1.10 (1.03–1.18)
6.54 (5.76–7.24)
3.86 (3.42–4.35)
3.59 (3.34–3.85)
2.41 (2.07–2.80)
1.81 (1.5–2.19)
1.42 (1.3–1.55)
1.21 (1.05–1.39)
Oral cavity and pharynx
Larynx
Esophagus
Breast
Liver
Rectum
Colon
*From Corrao et al. (2).
Calculations of attributable fractions have examined by how
much cancer incidence (or mortality) would be reduced if
alcohol were not consumed at alldan unlikely prospect. It
is more interesting to examine what might happen if heavy
drinkers reduced their consumption to sensible levelsd
probably an equally unlikely prospect. In the United Kingdom, sensible drinking is defined as 3 to 4 drinks a day for
men or 2 to 3 for women, where a drink is assumed to contain
8 g of ethanol. This will enable some estimate of the contribution of sensible drinking to cancer, assuming that the
estimated risks represent a causative relationship. Indeed,
one might speculate what might happen if everyone, even
current abstainers, drank so-called ‘‘sensible’’ amounts.
Fig. 2 shows the effect these strategies could have on cancer deaths. This analysis shows that, since women drink less
than men, the effect of heavy-drinking women changing to
moderate drinking is minimaldeven for breast cancer. For
men, the effect is also small, reflecting the proportion of
the population who drink more than the recommended
limits. On the other hand, were both women and men to
drink ‘‘sensibly,’’ then cancer deaths would increase dramatically.
CONCLUSIONS
Moderate alcohol consumption as currently consumed has
a small effect on the incidence of cancer. However, its importance for increased incidence were populations to drink
more overall should not be ignored. The mechanisms by
which alcohol might induce common cancers require
more rigorous research. Merely repeating findings from
Effects on cancer deaths on changing drinking patterns
% Cancer incidence by site and sex
70
1800
Males
Females
Adult Deaths in E&W
60
50
%
1600
40
30
20
1400
1200
1000
800
600
400
ec
Female
Male
Female
Male
Female
on
Male
Current
Heavy - Sensible
All - Sensible
R
C
ol
r
ve
Li
st
ea
O
es
op
Br
ou
ha
g
ry
n
La
tu
m
0
s
0
x
200
an ra
d lc
ph a
ar vit
yn y
x
10
O
Colon
Liver
Oesophagous
Breast
Site
FIGURE 1. Proportion of cancer incidence by site and by sex.
FIGURE 2. Estimated cancer deaths in the UK by sex attributable to current drinking patterns (left columns); all heavy drinkers
drinking moderately (middle columns); and everyone, including
abstainers, drinking moderately (right columns).
S48
McPherson
ALCOHOL CONSUMPTION AND CANCER
observational epidemiological studies is unlikely to enhance
our understanding, particularly as positive findings have
a greater chance of publication. The disparity of the scientific effort dedicated to investigating the causative nature
of the apparent protection of moderate alcohol against coronary heart disease compared with investigating the apparent raised risk of breast cancer seems completely misplaced.
The latter is assumed to be causal, while the former is contested vigorously. On the whole, the breast cancer association seems to be accepted without serious criticism and yet
the public health implications, where a lack of causality
has been demonstrated, are quite as profound.
AEP Vol. 17, No. 5S
May 2007: S46–S48
REFERENCES
1. Boffetta P, Hashiba M, La Vecchia C, Zatonski W, Rehm J. The burden of
cancer attributable to alcohol drinking. Int J Cancer. 2006;119:884–887.
2. Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15 diseases. Prev Med. 2004;38:613–619.
3. Doll R, Forman D, La Vecchia C, Woutersen R., Alcoholic beverages and
cancer of the digestive tract and larynx. In: Macdonald I, ed. Health issues
related to alcohol consumption: ILSI, 1999.
4. McPherson K. Alcohol and breast cancer. Eur J Cancer. 1998;34:1307–
1308.
5. McPherson K, Cavallo F, Rubin E. Health issues related to alcohol consumption. In: Macdonald I, ed. Alcohol and breast cancer. Malden (MA):
Blackwell Science; 1999:217–242.
Moderate Drinking, Inflammation, and Liver Disease
GYONGYI SZABO, MD, PHD
It is well known that heavy drinking increases the risk of alcohol-related liver disease (ALD). Female gender, hepatitis C or B, obesity, and other cofactors increase susceptibility to ALD, so ‘‘safe’’ levels of alcohol
consumption in regard to ALD vary among individuals. Inflammation is one mechanism by which alcohol
causes liver damage. Increasing evidence suggests that in contrast to the proinflammatory activation by
chronic excessive alcohol consumption, acute moderate alcohol administration has anti-inflammatory effects. Long-term alcohol administration results in increased baseline nuclear regulatory factor kB (NF-kB)
activation in the livers of mice; in contrast, acute alcohol administration in mice attenuates lipopolysaccharide (LPS)-induced NF-kB activation in the liver and serum tumor necrosis factor alpha (TNFa) induction. Consistent with this notion, peripheral blood monocytes from patients with alcoholic hepatitis
spontaneously produce increased amounts of TNFa and respond to ex vivo LPS stimulation with increased
TNFa levels, while acute moderate alcohol consumption in normal volunteers results in the attenuation of
TNFa production by various stimulants and attenuates monocyte production of other proinflammatory cytokines. To date, no evidence for a beneficial role of the anti-inflammatory effect of acute moderate alcohol
consumption on the liver has been demonstrated, but this may contribute to the effect of alcohol on other
organ systems.
Ann Epidemiol 2007;17:S49–S54. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Alcoholic Liver Diseases, Inflammation.
INTRODUCTION
Alcoholic liver disease (ALD) remains a major cause of morbidity and mortality worldwide. Alcohol-induced reactive
oxygen and nitrogen species, as well as inflammatory and genetic factors, contribute to the development and progression
of ALD. In contrast to the effects on coronary heart disease,
moderate alcohol consumption has no benefit for liver disease and may even cause liver disease in susceptible individuals. Activation of inflammatory pathways by long-term
heavy alcohol intake, as opposed to attenuation of inflammation by acute moderate alcohol consumption, likely contributes to some of the organ-specific effects of alcohol.
ALCOHOL CONSUMPTION AND LIVER DISEASE
While multiple studies suggest a beneficial effect of moderate alcohol consumption on coronary heart disease, a linear
correlation exists between the amount and duration of alcohol use and liver disease. In the Western world, up to 50% of
cases of end-stage liver disease are related to alcohol use (1).
The ‘‘threshold’’ of alcohol consumption associated with
ALD depends on the daily amount and the duration of
From the Hepatology and Liver Center, University of Massachusetts
Medical School.
Address correspondence to: Gyongyi Szabo, MD, PhD, Hepatology and
Liver Center, University of Massachusetts Medical School, 364 Plantation
St, Worcester, MA 01605. E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
alcohol intake. Furthermore, gender difference is a major
factor in ALD. In males, a daily intake of 40 to 80 g of alcohol over a period of 10 to 12 years can lead to ALD, while in
females only 10 to 30 g of daily alcohol can lead to liver disease (2, 3). A large, population-based prospective study in
Denmark found that the nadir of the estimated relative
risk of developing liver disease was at an alcohol intake of
1 to 6 beverages per week, and above this level a steep increase in relative risk was observed (2). The level of alcohol
intake above which the relative risk was significantly greater
than one was observed at 7 to 13 beverages per week for
women and 14 to 27 beverages per week for men. While
the association with the amounts and duration of alcohol
abuse is well established, at this time, there are insufficient
data to evaluate whether different patterns of drinking
would affect development of alcoholic liver disease.
Although the determination of a safe ‘‘threshold’’ of alcohol intake, with respect to dose and length of time, would be
important for evaluating the global benefits and potential
harms of moderate alcohol use, the amount of ‘‘safe’’ alcohol
intake that would not cause ALD is difficult to determine.
Long-term heavy alcohol use predictably results in the development of fatty liver in 90% to 100% of patients, while
only 10% to 35% develop alcoholic hepatitis and 8% to
20% develop alcoholic cirrhosis (4). The severity of steatosis
predicts the risk of progression to fibrosis or cirrhosis in alcoholic patients (5). Most individuals who habitually drink
more than 60 g/d of alcohol will develop fatty liver and
subjects with severe microvesicular/macrovesicular fat deposition are more likely to develop fibrosis and evolve to
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.012
S50
Szabo
ALCOHOL, INFLAMMATION, AND LIVER DISEASE
Selected Abbreviations and Acronyms
ADH Z alcohol dehydrogenase
ALD Z alcohol-related liver disease
ALDH Z aldehyde dehydrogenase
ALT Z alanine aminotransferase
Cyp2E Z cytochrome P450 2E1
HBV Z hepatitis B virus
HCV Z hepatitis C virus
LPS Z lipopolysaccharide
MCP-1 Z monocyte chemotactic protein 1
NF-kB Z nuclear regulatory factor kB
ROS Z reactive oxygen species
TNFa Z tumor necrosis factor a
cirrhosis (5, 6). Some studies suggest that even moderate
drinking can increase the risk of cirrhosis (7). A metaanalysis of 8 studies found a pooled relative risk estimate
between 1.6 and 3.0 for moderate drinkers (!25 g/d) compared with nondrinkers in populations that included both
men and women (8). Moderate drinking does not appear
to have any beneficial effects on liver cancer. While liver
cancer is associated with excessive and prolonged alcohol
use, moderate doses of alcohol (25g/d) resulted only in a minimally elevated risk of liver cancer in both men and women
(8).
It remains to be understood why only a subset of people
with prolonged heavy alcohol use will develop ALD, and
why cirrhosis occasionally develops even in ‘‘social
drinkers’’ (9–11). Currently, there is a limited amount of
knowledge regarding specific coexisting host and external
factors that may interact with alcohol to cause and accelerate liver disease. Of those factors, female gender is a wellestablished risk factor for ALD, as lower doses of alcohol
over a briefer period result in liver damage among females.
Women have a significantly higher relative risk of developing alcohol-related liver disease than men for any given
level of alcohol intake (2). The underlying mechanisms
are not fully understood, though differences in the alcohol
dehydrogenase enzyme and hormonal factors have been suggested. Importantly, ALD progresses more frequently in
women even after cessation of alcohol use (4). The presence
of other liver diseases, particularly chronic infection with
hepatitis C (HCV) or B viruses (HBV), significantly increases the risk of progressive liver damage in combination
with both moderate and excessive alcohol use (12). The
prevalence of chronic HCV infection is increased in patients with long-term alcohol consumption, and alcohol
use has been identified as an independent risk factor for progression of liver disease in chronic HCV infection (13). It
remains to be determined whether the additive effects of
alcohol and HCV infection are related to the effects of
alcohol on HCV viral replication, host immune factors, or
hepatocytes (14). Genetic factors, such as hereditary hemochromatosis, are also associated with an increased risk of
AEP Vol. 17, No. 5S
May 2007: S49–S54
alcohol-induced liver injury, as alcohol consumption greatly
increases the prevalence of cirrhosis in these patients (15).
Obesity and exposure to other substances/drugs present additional risks for alcoholic liver disease (16, 17). On the basis of these multiple factors, ‘‘safe’’ levels of alcohol
consumption with regard to liver disease may significantly
vary among individuals depending on the presence or absence of cofactors that may modulate the effects of alcohol
on the liver.
MECHANISMS OF LIVER INJURY
AND THE ROLE OF INFLAMMATION
IN ALCOHOLIC LIVER DISEASE
The mechanisms by which alcohol consumption causes liver
damage involves a number of physiological and biochemical
changes that lead to liver pathology. Alcohol dehydrogenase (ADH) is the dominant enzyme in alcohol oxidation
producing acetaldehyde, which is further oxidized to acetate
by aldehyde dehydrogenase (ALDH). When tissue alcohol
levels exceed 10 mmol/L concentration, the microsomal enzyme cytochrome P450 2E1 (Cyp2E) is induced and participates in alcohol metabolism. It has been proposed that
Cyp2E1 induction plays a role in ALD and that this enzyme
is a source of oxidative stress that results in direct damage to
hepatocytes, contributing to inflammation (18).
The natural history of ALD is characterized by chronic inflammation in the liver (Fig. 1). There are multiple lines of
evidence for inflammatory cascade activation in ALD, including neutrophil and other inflammatory cell recruitment
to the liver and increased circulating levels of proinflammatory cytokines such as tumor necrosis factor a (TNFa), interleukin 6 (IL-6), IL-8, and monocyte chemotactic protein 1
(MCP-1) (18, 19). This inflammation is most likely triggered
by reactive oxygen species (ROS) and endotoxin (lipopolysaccharide, LPS). Alcohol consumption is associated with
impaired function of the intestinal mucosa in patients with
various stages of alcoholic liver injury, as well as in rodents
after short-term alcohol administration (20, 21). Several
studies have shown malabsorption of xylose in alcoholics
without liver disease (22). In human alcoholics, LPS levels
are increased in the blood and in the portal circulation
(23). As it enters the liver, LPS is recognized by the Tolllike receptor 4 complex expressed on the resident macrophages, Kupffer cells, in the liver and induces downstream
activation of intracellular pathways that result in activation
of various proinflammatory genes including TNFa, IL-8, and
MCP-1 (24). Kupffer cells isolated from mice after alcohol
administration demonstrate increased spontaneous and
stimulation-induced TNFa production (24). The increased
production of TNFa by Kupffer cells is thought to be of particular importance in the pathogenesis of ALD, as TNFa is
AEP Vol. 17, No. 5S
May 2007: S49–S54
FIGURE 1. Currently accepted model of the pathomechanisms
of alcoholic hepatitis. Chronic alcohol use is associated with increased endotoxin levels in the portal blood presumably from increased gut permeability. Endotoxin in the liver activates Kupffer
cells, the resident macrophages, to produce chemokines (IL-8,
MCP-1), proinflammatory cytokines (TNFa, IL-1), reactive oxygen species (ROS), and transforming growth factor b (TGFß).
The chemokines recruit other inflammatory cells, particularly
polymorphonuclear leukocytes (PMNL) into the liver, which amplify the inflammation. ROS and TNFa can damage hepatocytes
by inducing apoptosis. In addition, alcohol and its metabolites
have direct negative effects on hepatocyte function and survival.
Moreover, Kupffer cell–derived inflammatory mediators, TGFb,
and LPS activate stellate cells to proliferate and produce collagen,
leading to fibrosis and progression of liver damage.
one of the principal mediators of the inflammatory response
transducing differential signals that regulate cellular activation, proliferation, cytotoxicity, and apoptosis (25). LPS recognition by Toll like receptor 4 expressed on hepatic stellate
cells and sinusoidal epithelial cells may also contribute to the
progression of ALD (26, 27). The continued presence of increased inflammatory mediators and activation of Kupffer
cells in the liver contribute to the activation of stellate cells
by inflammatory mediators and induce collagen deposition,
fibrosis, and progression to cirrhosis.
DIFFERENT REGULATION OF INFLAMMATORY
PATHWAYS BY ACUTE AND CHRONIC
ALCOHOL USE
Increasing evidence suggests that in contrast to the proinflammatory activation in alcoholic hepatitis by chronic excessive alcohol consumption, acute alcohol administration
has anti-inflammatory effects. Consistent with this notion,
peripheral blood monocytes obtained from patients with
alcoholic hepatitis spontaneously produced increased
amounts of TNFa and responded to ex vivo LPS stimulation
with increased TNFa levels (19, 28), while acute alcohol
consumption in normal volunteers results in the attenuation
of TNFa production by various stimulants (29). Acute alcohol intake in humans also attenuated monocyte production
Szabo
ALCOHOL, INFLAMMATION, AND LIVER DISEASE
S51
of other proinflammatory cytokines, including IL-1b and the
chemokines, IL-8 and MCP-1 (29, 30), suggesting that common pathways in the regulation of these cytokines may be
affected by alcohol. Indeed, activation of the nuclear regulatory factor kB (NF-kB), a common element in the promoter region of these proinflammatory genes, is inhibited
by acute alcohol intake. NF-kB nuclear translocation,
which is associated with proinflammatory cytokine induction, is decreased in monocytes of individuals after shortterm alcohol intake (29). Similar inhibition of NF-kB
activation has been reported in other studies using in vitro
alcohol administration and other cell types (31). In contrast
to the studies above, increased NF-kB activation has been
reported in monocytes of chronic alcoholics with alcoholic
hepatitis (28).
In animals, administration of an alcohol-containing liquid diet to mice for 4 weeks results in fat deposition in the
liver, similar to changes seen in humans after excessive alcohol use (Fig. 2). Furthermore, alcohol feeding causes liver
damage, indicated by increased serum levels of alanine aminotransferase in alcohol-fed animals as compared with those
fed control liquid or chow diet. Activation of the proinflammatory cascade is demonstrated by the increased baseline
NF-kB activation in the livers of alcohol-fed mice and not
in the livers of control mice (Fig. 2). In contrast, acute alcohol administration in mice attenuates LPS-induced NF-kB
activation in the liver and serum TNFa induction (Fig. 3).
Together, these results indicate that acute ingestion of alcohol and chronic heavy use have opposite effects on inflammatory pathways, inflammatory cell activation in the
liver, and in isolated monocytes (Fig. 4).
CONCLUSIONS
Moderate alcohol use in the amount of 20 to 30 g/d for men
and 10 to 15 g for women is likely to be safe for most individuals. Nevertheless, an individual’s threshold for alcoholinduced liver disease depends on multiple susceptibility
factors and coexisting liver conditions. The presumed
‘‘safe’’ limits of alcohol use, therefore, cannot apply to people with cofactors such as chronic viral hepatitis, obesity,
hemochromatosis, and other factors that increase the susceptibility of the liver to alcohol-induced injury. Given
the frequency of chronic HCV infection and the increasing
epidemic of obesity in Western countries, universal limits of
‘‘safe’’ alcohol use are difficult to determine for the liver.
There is an obvious anti-inflammatory effect of acute,
moderate alcohol intake on inflammatory cell functions
that may be a critical component of cardioprotective effects
of alcohol use. To date, there is no evidence for a beneficial
role of the anti-inflammatory effect of acute moderate
alcohol consumption on the liver, but the effects on
S52
Szabo
ALCOHOL, INFLAMMATION, AND LIVER DISEASE
AEP Vol. 17, No. 5S
May 2007: S49–S54
FIGURE 2. Chronic alcohol feeding results in liver injury and inflammatory activation in mice. Mice (3–4 mice per group) were fed with
a Lieber–de-Carli liquid diet containing 4.5% vol/vol alcohol, pair-fed liquid or chow diet. Top panels: Liver sections were stained with
hematoxylin-eosin to demonstrate steatohepatitis. Bottom right: Serum ALT was determined using a commercially available kit demonstrating an increase in alcohol-fed mice. Bottom left: Nuclear extracts from livers were tested in an electromobility gel shift assay for
NF-kB binding to a consensus NF-kB oligonucleotide as described (31), demonstrating an increase in NF-kB in alcohol-fed animals.
FIGURE 3. Acute alcohol administration inhibits NF-kB activation and LPS-induced TNFa production in mice. Mice (4–5 per group)
were given acute alcohol (2 mL in intraperitoneal injection), followed by an LPS challenge. Serum and livers were collected 2 hours after
the LPS administration. Right panel: Serum TNFa levels were determined by enzyme-linked immunosorbent assay demonstrating decreased
TNFa in acute alcohol-treated mice. Left panel: Nuclear extracts from livers were evaluated for NF-kB activation in an electrophoretic
mobility shift as described (31), demonstrating the decreased NF-kB in acute alcohol-treated mice.
AEP Vol. 17, No. 5S
May 2007: S49–S54
Szabo
ALCOHOL, INFLAMMATION, AND LIVER DISEASE
S53
FIGURE 4. Opposite effects of acute and chronic alcohol consumption on inflammation. Acute alcohol consumption has an antiinflammatory effect through its effects on macrophages by inhibiting stimulation-induced proinflammatory pathways, while chronic
alcohol use is associated with up-regulation of the same pathways in inflammatory cells.
inflammation may contribute to the effects of alcohol on
other organ systems.
REFERENCES
9. Lelbach WK. Cirrhosis in the alcoholic and its relation to the volume of
alcohol abuse. Ann N Y Acad Sci. 1975;252:85–105.
10. Savolainen V, Liesto K, Mannikko A, Penttila A, Karhunen PJ. Alcohol
consumption and alcoholic liver disease: evidence of a threshold level of
effects of ethanol. Alcohol Clin Exp Res. 1999;17:1112–1117.
11. Savolainen V, Perola M, Lalu K, Penttila A, Vertanen I, Karhunen PJ.
Early perivenular fibrogenesis d precirrhotic lesions among moderate alcohol consumers and chronic alcoholics. J Hepatol. 1995;23:524–531.
1. Orholm M, Sorensen TI, Bentsen K, Hoybye G, Eghoje K, Christoffersen
P. Mortality of alcohol abusing men prospectively assessed in relation to
history of abuse and degree of liver injury. Liver. 1985;5:253–260.
12. Westin J, Lagging LM, Spak F, Aires N, Svensson E, Lindh M, et al. Moderate alcohol intake increases fibrosis progression in untreated patients
with hepatitis C virus infection. J Viral Hepat. 2002;9:235–241.
2. Becker U, Deis A, Thorkild IA, Gronbaek M, Borch-Johnson K, Muller
CF, et al. Prediction of risk of liver disease by alcohol intake, sex and
age: A prospective populations study. Hepatology. 1996;23:1025–1029.
13. Jamal MM, Saadi Z, Morgan T. Alcohol and hepatitis C. Dig Dis.
2005;23:285–296.
3. Fuchs CS, Deis A, Sorensen TI. Prediction of risk of liver disease by alcohol intake, sex, and age: a prospective population study. Hepatology.
1995;332:1245–1250.
4. McCullough AJ. Alcoholic liver disease. In: Schiff ER, Sorrell FF, Maddrey WC, eds. Schiff’s Diseases of the liver. Philadelphia: Lippincott,
Raven; 1999:941–971.
5. Teli MR, Day CP, Burt AD, Bennett MK, James OF. Determination of
progression to cirrhosis or fibrosis in pure alcoholic fatty liver. Lancet.
1995;14:987–990.
6. Fromenty B, Grimbert S, Mansouri A, Beaugrand M, Erlinger S, Rotig A,
et al. Hepatic mitochondrial DNA deletion in alcoholics: association with
microvesicular steatosis. Gastroenterology. 1995;108:193–200.
7. Kondili LA, Taliani G, Cerga G, Tosti ME, Babameto A, Resuli B. Correlation of alcohol consumption with liver histological features in noncirrhotic patients. Eur J Gastroenterol Hepatol. 2005;17:155–159.
8. Corrao G, Bagnardi V, Zambon A, Arico S. Exploring the dose-response
relationship between alcohol consumption and the risk of several
alcohol-related conditions: a meta analysis. Addiction. 1999;94:1551–
1573.
14. Szabo G, Dolganiuc A. Subversion of plasmacytoid and myeloid dendritic
cell functions in chronic HCV infection. Immunobiology. 2005;210:237–
247.
15. Fletcher LM, Dixon JL, Purdie DM, Powell LW, Crawford DH. Excess
alcohol greatly increases the prevalence of cirrhosis in hereditary hemochromatosis. Gastroenterology. 2002;122:281–289.
16. Naveau S, Giraud V, Borotto E, Aubert A, Capron F, Chaput JC. Excess
weight risk factor for alcoholic risk factor. Hepatology. 1997;25:108–111.
17. Raynard B, Balian A, Falik D, Capron F, Bedossa P, Chaput JC, et al. Risk
factors of fibrosis in alcohol-induced liver disease. Hepatology.
2002;35:635–638.
18. Arteel G, Marsano L, Mendez C, Bentley F, McClain C. Advances in alcohol liver disease. Best Prac Res Clin Gastroenterol. 2003;17:625–647.
19. MacClain CJ, Cohen DA. Increased tumor necrosis factor production by
monocytes in alcoholic hepatitis. Hepatology. 1989;9:349–351.
20. Parlesak A, Schafer C, Schutz T, Bode JC, Bode C. Increased intestinal
permeability to macromolecules and endotoxemia in patients with chronic
alcohol abuse in different states of alcohol-induced liver disease. J Hepatol.
2000;32:742–747.
S54
Szabo
ALCOHOL, INFLAMMATION, AND LIVER DISEASE
21. Stern M, Carter EA, Walker WA. Food proteins gut mucosal barrier. IV.
Effects of acute and chronic ethanol administration on handling and uptake of bovine serum albumin by rat small intestine. Dig Dis Sci.
1986;31:1242–1248.
22. Thompson AD, Heap LC, Ward RJ. Alcohol induced malabsorption in the
gastrointestinal tract. In: Preedy VR, Watson RR, eds. Alcohol and the
gastrointestinal tract. Boca Raton (FL): CRC Press; 1996:203–218.
23. Urbaschek R, McCuskey RS, Rudi V, Becker KP, Stickel F, Urbaschek B,
et al. Endotoxin, endotoxin-neutralizing-capacity, sCD14, sICAM-1, and
cytokines in patients with various degrees of alcoholic liver disease. Alcohol Clin Exp Res. 2001;25:261–268.
24. Nagy LE. Recent insights into the role of the innate immune system in the
development of alcoholic liver disease. Exp Biol Med. 2003;228:882–890.
25. Beutler B. TNF, immunity and inflammatory disease: lessons of the past
decade. J Investig Med. 1995;43:227–235.
26. Paik YH, Schwabe TF, Bataller R, Russo MP, Jobin C, Brenner DA. Tolllike receptor 4 mediates inflammatory signaling by bacterial lipopolysaccharide in human hepatic stellate cells. Hepatology. 2003;37:1043–1055.
AEP Vol. 17, No. 5S
May 2007: S49–S54
27. Uetsugi T, Froh M, Arteel G, Bradford BU, Thurman RG. Toll-like receptor 4 is involved in the mechanism of early alcohol-induced liver injury in
mice. Hepatology. 2001;34:101–108.
28. Hill DB, Barve S, McClain C. Increased monocyte nuclear factor-kB activation and tumor necrosis factor production in alcoholic patients. J Lab
Clin Med. 2000;135:387–395.
29. Mandrekar P, Catalano D, White B, Szabo G. Moderate intake in humans
attenuates monocyte inflammatory responses: inhibition of nuclear regulatory factor kappa B and induction of interleukin 10. Alcohol Clin Exp Res.
2006;30:135–139.
30. Szabo G, Chavan S, Mandrekar P, Catalano D. Acute alcohol consumption attenuates IL-8 and MCP-1 induction in response to ex vivo stimulation. J Clin Immunol. 1999;19:67–76.
31. Mandrekar P, Dolganiuc A, Bellerose G, Kodys K, Romics R, Nizamani R,
et al. Acute alcohol inhibits the induction of regulatory factor kB activation through CD14/toll-like receptor 4, interleukin-1 and tumor necrosis
factor receptors: a common mechanism independent of inhibitory kappa
B alpha degradation? Alcohol Clin Exp Res. 2002;26:1609–1614.
Investigating the Association Between Moderate Drinking
and Mental Health
NADY EL-GUEBALY, MD
In an attempt to relate ‘‘moderate drinking’’ to ‘‘mental health,’’ inadequacies of definition for both terms
become apparent. Moderate drinking can be variously defined by a certain number of drinks to ‘‘nonintoxicating’’ to ‘‘noninjurious’’ to ‘‘optimal,’’ whereas mental health definitions range from ‘‘the absence of psychopathology’’ to ‘‘positive psychology’’ to ‘‘subjective well-being.’’ Nevertheless, we evaluated the relation
by conducting an electronic search of the literature from 1980 onwards using the terms ‘‘moderate drinking,’’ ‘‘moderate alcohol consumption,’’ ‘‘mental health,’’ and ‘‘quality of life.’’
Most studies report a ‘‘J-shaped curve,’’ with positive self-reports of subjective mental health associated
with moderate drinking but not with heavier drinking. The relevance of expectancies has been unevenly
acknowledged, and studies on the cultural differences among expectancies are largely lacking. The potential role of moderate drinking in stress reduction and studies of social integration have yielded inconsistent
results as previous levels of drinking, age, social isolation, and other factors have often not been adequately
controlled. Future anthropological, epidemiological, and pharmacological interactions preferably must be
studied through a prospective design and with better definitions of moderate drinking and mental health.
Ann Epidemiol 2007;17:S55–S62. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Drinking Behavior, Mental Health, Stress.
INTRODUCTION
Difficulties in capturing the concept of ‘‘moderate drinking’’
(1, 2) are matched by the challenges in identifying valid indicators of mental health. While mental health can be simply defined as ‘‘the absence of psychopathology,’’ Vaillant
and Vaillant (3) have reviewed 6 empirical approaches to
mental health, which are outlined.
Among these approaches, studies of ‘‘moderate drinking’’
have investigated psychosocial benefits, including positive
subjective health and positive mood effects, both anticipated and experienced. Another line of investigation has
been the reduction of psychopathology. The current body
of research integrates large epidemiological studies with anthropological data.
Most studies use the definitions of ‘‘moderate drinking’’ as
set out in available guidelines, expressed as a number of
drinks. Main methodological caveats reviewed include the
array of definitions of ‘‘moderate drinking’’ and ‘‘mental
health,’’ the need to differentiate between ‘‘lifestyle’’ abstinence and those abstainers recovering from drinking problems, and the assessment of psychological benefits in the
context of expectancies, social context and cultural norms.
From the Addiction Centre, Foothills Hospital.
Address correspondence to: Nady el-Guebaly, MD, Foothills Hospital,
Addiction Centre, 1403 29th St NW, Calgary, Alberta T2N 2T9, Canada.
E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
Further study is required of the risks-benefits of moderate
drinking in non-Western countries, as well as the directions
of causality and their policy implications.
METHOD
An electronic search of the available literature from 1980
onwards was conducted using the terms ‘‘moderate drinking,’’ ‘‘moderate alcohol consumption,’’ ‘‘mental health,’’
and ‘‘quality of life.’’ The search engines involved were
MEDLINE, PyschInfo, and CINAHL. The search was complemented by a retrieval of the references cited in previously
published literature reviews.
CHALLENGES OF DEFINITION
What Is ‘‘Moderate’’ Drinking?
The conceptual evolution and meanings of ‘‘moderate’’ consumption has been summarized by Eckardt et al. (1). Moderate drinking can be defined as any drinking that is
‘‘nonintoxicating;’’ in other words, consumption that is
controlled or restrained, for example, as evident in current
campaigns against drinking-driving (Social motivation).
Another meaning includes ‘‘noninjurious’’ drinking, or consumption, the cumulative effect of which does not result in
health deterioration or harm. Herein, consumption is below
an upper limit beyond which some health malfunction occurs (Medical motivation). In another meaning, moderate
drinking is defined as ‘‘statistically normal.’’ In this case, it
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.013
S56
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
Selected Acronym
NIAAA Z National Institutes on Alcohol Abuse and Alcoholism
is defined by the norm or mean value of consumption for
a particular age or other population group and within a standard deviation of the group mean. Finally, moderate drinking levels may be viewed by some as those that are
‘‘optimal,’’ consumption levels that have some beneficial effect specific to particular diseases identified in comparisons
with both nondrinkers and heavier drinkers. This level is
also represented by the nadir of the J-curve for health
outcomes.
More recently, the National Institutes on Alcohol Abuse
and Alcoholism (NIAAA) in the United States endorsed
the above typology and listed a number of caveats related
to current national definitions of ‘‘moderation’’ (2). In the
United States, for example, official dietary guidelines define
moderate drinking as no more than one drink per day for
women and no more than two drinks per day for men (4).
Difficulties with this definition arise from a number of
factors. There is a range of differences among individuals
in drinking experience and tolerance. For example, twoto four-fold differences have been described among individuals in the pharmacokinetics and pharmacodynamics of
alcohol metabolism. Differences also result from the time
period over which alcohol is consumed, or its interaction
with genetic vulnerability for a particular medical condition. Additional confounders may be introduced by lifestyle
variables between drinkers and nondrinkers, demographics,
including age or gender, as well as variation in drinking patterns. Furthermore, methodological differences arise from
variation in patterns of consumption, drink sizes, in vivo
and in vitro reactivity and extrapolation from animal
models. Finally, cultural dimensions are another source of
variability and include interaction of food, traditions and taboos, as well as drinking settings. Bearing in mind the above
caveats, the NIAAA report, however, does not suggest
a need to modify the existing guidelines in the United
States.
It is interesting to note that this review highlights the relative consistency of the drinking levels suggested as ‘‘moderate,’’ at least in Western societies. In 1862, Sir Francis
Anstie set the upper limit of safe drinking at 1.5 ounces of
absolute alcohol, that is, approximately three 5-oz glasses
of wine, three 12-oz bottles of beer and three 1-oz shots of
80% proof whiskey (5). A century and a half of extensive research has reduced these levels by only one third. While past
studies focused on comparisons between moderate and
heavy drinkers, current epidemiological surveys have also
added the comparison with abstainers.
AEP Vol. 17, No. 5S
May 2007: S55–S62
What Is Mental Health?
Difficulties in capturing the concept of ‘‘moderate drinking’’
are matched by the challenges in identifying valid indicators
of mental health. Vaillant and Vaillant (3) contrast 6 empirical approaches to mental health. While mental health was
initially merely defined as ‘‘the absence of psychopathology,’’ empirical approaches have further conceptualized
mental health in a number of ways. One definition looks
at mental health as ‘‘above normal,’’ a mental state that is
objectively desirable. This is in keeping with Sigmund
Freud’s definition of mental health as the capacity to work
and to love.
A second view looks at mental health as ‘‘positive psychology,’’ epitomized by the presence of multiple human
strengths. ‘‘Maturity,’’ part of healthy adult development,
is another dimension of mental health, as is ‘‘socioemotional
intelligence,’’ accurate and conscious perception that monitors the social adaptation of one’s own emotions. Additional approaches define mental health as ‘‘subjective
well-being,’’ the subjective experience of happiness, contentment and being desired, and ‘‘resilience,’’ the capacity
for successful adaptation and homeostasis.
In addition to searching for ‘‘absence of psychopathology,’’ the studies reviewed have mostly investigated psychosocial benefits. These also include a range of dimensions.
Psychosocial benefits include, for example, positive subjective health, a self-perception of good health, whether or
not moderate drinking enhances it. Positive psychology,
such as mood effects, both anticipated and experienced, is
another important aspect. Expectancy research is of particular relevance in investigations of psychosocial benefits.
Brown et al (6) have identified six independent expectations
by drinkers of their drinking. These include:
1.
2.
3.
4.
5.
6.
Positive transformation of experience
Enhanced social and physical pleasure
Enhanced sexual performance and experience
Increased power and aggression
Increased social assertiveness
Reduced tension
According to Brown and colleagues, nonproblematic
drinkers emphasize expectations of social and physical pleasure (6), while problem drinkers anticipate reduced tension
reduction. Gustafson (7) reported that consumers of high
levels of alcohol had higher expectations along all dimensions than those consuming low levels of alcohol. Expectations and experiences are both subject to substantial
cultural differences.
MODERATE DRINKING AND MENTAL HEALTH
Within the constraint of the above challenges in definition,
the association of moderate drinking and mental health has
AEP Vol. 17, No. 5S
May 2007: S55–S62
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
S57
TABLE 1. Subjective health and alcohol*
Study and sample
Method and controls
Poikolainen et al (14), random study of 6040 Finns,
age 25–64 y
Power et al (15), 1958 cohort of all UK births
(3/3–3/9), 9605 of whom reported alcohol
consumption
Poikolainen and Vartiainen (16), FINRISK Study,
random sample of 6040 Finns age 25–64 y
Grønbæk et al (17), WHO Copenhagen survey,
random sample of 12,039 Danes age 18–100 y
Guallar-Castillon et al (11), 1993 Spanish
National Health Survey of 19,573, age >16 y
Regression; controls: age, sex, education, marital
status, isolation, disability pension, smoking, exdrinker, decrease in drinking for health reasons
Within prospective study, cross-sectional selfreports at 33 y of age; controls: heavy/problem
drinkers at age 23 y
Regression; controls: age, sex, education, marital
status, no friends, sickness pension, smoking,
lifelong abstinence/ex-drinker, decrease in
drinking last 12 mo
Regression; controls: age, sex, intake of other
alcoholic beverages, physical activity, body
mass, education, social networks, chronic
disease, smoking
Regression; controls: age, sex, education, job
status, social support, location, drinking, leisure
time and chronic disease; ex-drinkers included
Results
Adjusted OR, subjective health
40–99 g/wk Z 0.7 (3.3–9 drinks)
Poor-fair self-rated health (units/week;
men/women) Moderate (11–35/6–20) Z
9%/11%
Adjusted OR, suboptimal health (drinks/d), wine
only, men: lowest at 5–9 Z 0.67 Suboptimal
health (drinks/d), wine only, women: lowest at
0–4 Z 0.81
Adjusted OR, suboptimal health (drinks/d, wine
only): 3–5 Z 0.65
With Mediterranean lifestyle and wine with meals,
the higher the consumption of wine and beer
(not spirit), the lower the prevalence of
suboptimal health; no ‘‘J’’ shape
OR Z odds ratio; UK Z United Kingdom; WHO Z World Health Organization.
*Adapted from Peele and Brodsky (10) and updated by author.
been the topic of a number of research lines. The initial studies were mostly of a social-anthropological nature and were
reviewed by Baum-Baicker (8). Overall small samples were
generally involved, interestingly, involving medical students in several cases. The review is considered to be mainly
of historical interest.
Fifteen years later, comprehensive reviews by Chick (9), as
well as by Peele and Brodsky (10) present a more sophisticated
body of research that integrate larger epidemiological studies
with anthropological data. The following tables are largely
based on the review by Peele and Brodsky, modified to fit
the focus of this paper and updated as required (Tables 1–7).
TABLE 2. Positive mood effects: Anticipation and experience
Study and sample
Positive mood effects in surveys
US (18, 19)
Finland (20, 21)
Australia (22, 23)
Quebec, Canada (12)
Methods and controls
National survey data 1970 and 1983
Cross sectional ‘‘interviews’’
1988: N Z 3624, age 15–69 y
1992: N Z 2370 university students
Population surveys
‘‘Social representation’’ of drinking, i.e., objective
assessment vs. personal experience
UK (24)
Mass-observation archives
Cultural differences in expectations (poor; North America–based immigrant minorities)
Teahan (25)
Irish vs. Canadian male alcoholics
Marin (26)
Mexican Americans vs. non-Hispanic whites
Experimental/pharmacological research and social settings
Doty and De Wit (27)
Subjects’ choice of alcohol vs. placebo in social
setting vs. alone
Lowe and Taylor (24); Lowe et al (28)
Young drinkers given alcohol
Adapted from Peele and Brodsky (10) and updated by author.
Results
About 50% of current drinkers ‘‘feel happy and
cheerful’’ or ‘‘friendly’’ compared with 8%
‘‘aggressive’’ or 2% ‘‘sad’’
Positive effects reported more often than negative
54% identified relaxation, stress reduction, and
improved psychological well-being
‘‘Objective’’ assessments of drinking effects
frequently cite harm but not from ‘‘personal
experience’’
Drinking is usually enjoyable
‘‘Irish’’ drink more for tranquilization and
detachment while ‘‘Canadians’’ drink more for
social and sexual enhancement
Mexican Americans expect both more positive and
negative effects
Subjects in social settings have more positive
reactions
Young drinkers laugh more viewing a comedy film;
in pubs and in daily life. Expectancies?
S58
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
AEP Vol. 17, No. 5S
May 2007: S55–S62
TABLE 3. Stress reduction: Response dampening, social learning, attention allocation
Study and sample
Methods and controls
Results
Naturalistic research
Culbert (29)
Drinking and life stressors
De Castro (30)
Daily activities report of moderate drinking
Cooper et al (31)
Moderate vs. problem drinkers
Moderate drinkers do not increase and even reduce drinking in response to
major life stressors
Moderate alcohol use has a calming effect even if use does not aim at reducing
anxiety
Moderate drinkers report that alcohol creates a general calming sensation,
while problem drinkers use alcohol to cope with major life issues. Is it
pharmacological effect, social learning or expectancy?
Adapted from Peele and Brodsky (10) and updated by author.
Positive self-reports of subjective mental health associated with moderate drinking (Table 1) are complemented
by lower reports of suboptimal health. Among the studies
examined, the only one that found no J-shaped association
also did not control for ex-drinkers (11). Overall, an association of positive mood with moderate drinking is consistent
in a number of surveys from different countries (Table 2). Of
note, Demers et al (12) reported an interesting difference
between descriptions of ‘‘personal experience’’ of drinking
as usually enjoyable, as compared to an ‘‘objective’’ assessment of drinking effects, where harm was more frequently
cited. The relevance of expectancies in this line of research
has been unevenly acknowledged. Studies on the cultural
differences among expectancies are largely lacking and
those that are available are often based on immigrant minorities in different countries.
Drinking plays a role in the potential for stress reduction
(Table 3) and a popular way to conceptualize the benefits of
drinking has been with regard to people drinking alcohol to
‘‘relax.’’ Several theoretical psychological models have been
developed around this concept that include stress response
dampening, social learning, expectancy, and attention allocation. Studies in this area often show inconsistent results
(10, 13).
TABLE 4. Social integration and reduction of ‘‘social anxiety’’
Study and sample
Methods and controls
Camacho et al (32), probability sample of
4590 residents of Alameda County, Calif,
>35 y
Covariates with alcohol consumption;
inclusion of people in recovery?
Leifman et al (33), N Z 45,746, all Swedish
military conscripts 18–19 yr old males over
2-y period
Percentage in each drinking category
indicating various social integration/
sociability indicators; controls: alcoholic
father, rural vs. urban
Elderly
Alexander and Duff (34)
Retirement communities
Hanson (35)
Elderly men born in 1914
Graham (36)
Elderly in small Ontario community.
Lifetime abstainers and former drinkers
removed from study
Beck Depression and Anxiety Inventories
(BDI & BAI), SF/36, Cambridge
Cognitive Exam for Mental Disorders of
Elderly
Cassidy et al (37); cross section of 278
community dwelling women, age >70,
recruited through media and clubs
Youth
Murphy et al (38); sample of 353 college
students receiving credits
Regression; Daily Drinking Questionnaire;
Rutgers Alcohol Problem Index; two Life
Satisfaction Scales (LS)
Results
Not married (drinks/mo)
0 Z 35%; 1–30 Z 23% (1–2/d)
No group membership (drinks/mo);
0 Z 46%; 1–30 Z 32%
Often insecure with others (U shape)
Abstain Z 7%; 26–100 g alcohol/wk Z
3%; O250 g/wk Z 7%
No intimate conversations
Abstain Z 8%; 26–100 g/wk Z 3%;
O250 g/wk Z 9%
Greater social interaction associated with
heavier drinking
Isolation associated with greater
consumption and problem drinking
Current drinkers engaged in more social
activities
Physically active women half as likely to be
depressed; more likely if ever smoking >20
cigarettes/d; no association of alcohol with
mood or quality of life but better cognition
Drinking among young men but not women
associated with positive curvilinear
relation with social satisfaction but no
other LS scale, i.e., school, family, dating,
and future
AEP Vol. 17, No. 5S
May 2007: S55–S62
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
S59
TABLE 5. Mental healthdDepression and anxiety
Study and sample
Bell et al (39); random sample,
2029 adults in 3 southern US counties
Neff and Husaini (40); random sample,
713 adults in Tennessee
Neff (41); random sample, 1784 residents of San
Antonio, Tex
Lipton (42); random sample,
928 non-Hispanic white, Epidemiological
Catchment Area Study (Los Angeles)
Green et al (43); random HMO Oregon in
1990: 3074 males, 3947 females
Caldwell et al (44); random sample,
2404 adults age 20–24, Canberra region,
Australia
Paschall et al (45); from US longitudinal study of
adolescent health, i.e., in 1995 adolescents from
middle and high schools and then 2002
Alati et al (46); 812 patients attending emergency
department in Queensland over 14 days,
age 16–84 y; Australia
Methods and controls
Group mean comparisons; controls: race,
sex, age, marital status, social class;
separate analyses for experience of
different stressful life events
Regression on depression scores (CES-D
scale) by drinking categories; controls:
race, sex, education
Analysis of variance by drinking; controls:
fatalism, religiosity, social desirability
Regression on time 2 scores by drinking;
controls: sex, age, physical health,
depression at time 1; life stressors
Predictability of alcohol consumption by
MHI-5 and BSI-8 Depression screens
Results
Anxiety scale: moderate!light!heavy!abstain
Depression scale: moderate!light!abstain!heavy
but no definition of drinking category
Depressive symptoms: moderate!abstain!heavy
Depressive scale: abstain!light!heavy Abstainers:
‘‘recovery’’ excluded
Depression scale: moderate!light/
moderate!heavy!abstain!light
MHI-5 more predictive than depression alone specifically
among women; women with prior diagnosis of
depression drink less; ‘‘very heavy drinkers excluded’’
Goldberg Depression and Anxiety Scales; For young men, nondrinker/occasional and hazardous/
Positive and Negative Affect Schedule
harm consumption both associated with higher levels
and AUDIT
of anxiety and depression. Male abstainers less
extroverted and healthy, unrelated to past alcohol use
or current tobacco or marijuana use; female abstainers
did not show more distress
9 items from CES-D scale for general
Adjusted for health and socioeconomic factors; frequency
population; 7 categories of alcohol use
of depressive symptoms similar for moderate drinkers
and long term abstainers but higher for heavy drinkers
AUDIT and HADS
U-shaped relation for men between alcohol consumption
and anxiety/depression; more linear relation for women
AUDIT Z alcohol use disorders identification test; BSI-8 Z Brief Symptoms Inventory-8; CES-D Z Center of Epidemiological Studies Depression (scale); HADS Z Hospital
Anxiety and Depression Scale; HMO Z health maintenance organization; MHI Z Mental Health Index.
Adapted from Peele and Brodsky (10) and updated by the author.
Studies of social integration have been complemented by
studies into drinking and the reduction of social anxiety
(Table 4), supporting a positive influence of moderate drinking. However, there are some caveats around these data.
Some data sets do not explicitly control for ex-drinkers
who may be in recovery. Age-specific studies have been carried out, but evidence among the elderly is mixed. For some,
moderate drinking may be associated with reduced isolation,
perhaps linked with the general consumption of alcohol in
social venues. Among young people, one study of students
reports that the social impact of drinking appears to be
more apparent in men than in women (38).
Investigations into the ‘‘reduction of psychopathology’’
have targeted the prevalence of symptoms of depression
and anxiety with positive associations reported from moderate drinking (Table 5). The same positive association is reported with broader studies of psychological distress and/or
psychiatric diagnoses (Table 6). Of importance is that abstainers in cultures or family settings where abstinence is
normative are not likely to show mental health deficiencies.
Lastly, another indicator of mental health may be work performance associated with higher income or absence of disability and/or stress (Table 7). The moderate drinking
category appears to be associated with higher income, less
sickness, absence, disability, or distress. However, a data
analysis of cause or consequence has not been conducted.
CONCLUSIONS
In this review of studies around the association of moderate
drinking and mental health, several methodological caveats
are evident. These relate, first, to the existing array of definitions of ‘‘moderate drinking’’ and ‘‘mental health.’’ Most
studies appear to use definitions of ‘‘moderate drinking’’ as
laid out in dietary guidelines, favoring those that refer to
the number of drinks. Studies into mental health mostly
highlight subjective well-being, positive psychological attributes and the absence or reduction of psychopathology. Another caveat relates to the complexity in the assessment of
psychological benefits within the context of expectancies,
and social and cultural norms. Furthermore, there is
a need to differentiate between ‘‘lifestyle’’ abstainers and
those abstainers who are recovering from drinking problems.
Several challenges and questions also remain. Crossfertilization is required between investigations of mental
S60
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
AEP Vol. 17, No. 5S
May 2007: S55–S62
TABLE 6. Mental health: General
Study and sample
Methods and controls
Power et al (15), 1958 cohort of all UK births
(3/3–3/9), 9605 of whom reported alcohol
consumption
Within prospective study, cross-section of
psychological distress at 33 y; controls: heavy/
problem drinkers at age 23 y
Roberts et al (47), Whitehall II study of
10314 UK civil servants
Cross-section of psychosocial measures at
screening in longitudinal study; light drinkers:
men 1–10; women 1–6 units/wk
Vaillant (48), 50-y longitudinal study of 443
inner city Boston men
Percentage of each lifetime drinking group with
each mental health indicator
Abstainers and culture
Leifman et al (33), Swedish conscripts who abstain
Young conscript abstainers whose fathers abstained
vs. those whose fathers drank
Results
Psychological distress (units/wk; men/women)
Abstain Z 10%/15% (U-shaped relation)
Moderate (11–35/6–20) Z 5%/9%
Heavy (35þ/20þ) Z 11%/23%; U shape
Psychological measures: light
drinkersOabstainers; MenOwomen, overall
and for positive affect, level of upset; men and
women Z for hostility
Psychiatric diagnosis: abstain Z 35%;
Moderate ! 21 drinks/wk Z 17%
Abuse Z 39%
Lower psychological adjustment: HSRS ! 60
Abstain Z 19%; moderate Z 1%; abuse Z 19%
No psychiatric/anxiety/emotional disadvantage
among abstainers whose fathers abstained
HSRS Z Health-Sickness Rating Scale.
Note: Abstainers in cultures where abstinence is normative will not show mental health deficiencies.
Adapted from Peele and Brodsky (10) and updated by author.
health outcomes associated with moderate social drinking
(such as subjective health or stress reduction) with investigations into the prevention and/or reduction of clinical syndromes such as depression or social phobia. One unanswered
question is: What are the upper limits of ‘‘moderate drinking’’ in psychiatric comorbidities?
In addition, the risk-benefits of moderate drinking in
non-Western countries are ‘‘terra incognita.’’ They cannot
be extrapolated from studies of immigrants to Western
countries. Future anthropological, epidemiological, and
pharmacological interactions preferably must be studied
through a prospective design. Current studies of the association of psychological benefits with moderate drinking in
major demographic and socio-cultural groups are mostly
cross-sectional and may also require more sophisticated prospective designs.
Directions of causality remain to be further explored, as
do their policy implications. Is moderate drinking improving
mental health or are mentally healthy individuals more
likely to drink moderately? Moderate drinking may be the
result of a host of risk and protective factors derived from genetics, developmental phases, or environmental influences,
TABLE 7. Work performance
Study and sample
Camacho et al (32), panel study of probability
sample of 4590 residents of Alameda County,
Calif, >35 y of age
Vasse et al (49) 471 participants at 3 sites
in worksite health project, Netherlands
Mansson et al (50), 5 complete birth-year
cohorts of 3751 men aged 47–48 y in Malmö,
Sweden
Marchand et al (51), sample constituted 10,387
employees from 422 occupations from 1998
Quebec Health & Social Survey (QHSS)
Riise et al (52), cross-section of 23,312
individuals aged 40–47 y, Hordaland, Norway
Methods and controls
Covariates with alcohol consumption
Regression on sickness absence by stress alcohol
consumption; controls: age, gender, education,
marital status, blue-collar vs. white-collar
worksite, smoking
11-y prospective study; controls: smoking,
hypertension, cholesterol and body mass;
MAST scale; ‘‘Sick Abstainer’’ ruled out
Regression; Ilfeld Scale for psychological distress
over past week; work strains from Karasek’s skill
utilization; decision authority and psychological
demands, etc
Covariance analysis; SF12 Health Survey,
with physical and mental components
MAST Z Michigan Alcohol Screening Test; OR Z odds ratio; RR Z relative risk; SF12 Z 12-item short form.
Adapted from Peele and Brodsky (10) and updated by author.
Results
Disabled (drinks/mo)
0 Z 27%; 61–90 Z 10%
(2–3 drinks/d); O91 Z 17%
OR, sickness absence
Abstainers Z 4.6; Moderate Z 1.0
(!3 drinks/d for men and 2 for
women); Excessive Z 2.0
Adjusted RR, disability pension
Abstainers Z 1.8; low Z 1.0; high 1.3
6.1% variance of psychological distress
between occupations; alcohol use U-shape
relation with distress, i.e., moderate intake
not associated with distress
Difference between occupations with
managers in top health category; U-shape
relation with alcohol
AEP Vol. 17, No. 5S
May 2007: S55–S62
for example, involving social control, behavioral choice,
social learning or stress and coping theories.
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
S61
21. Nystrom M. Positive and negative consequences of alcohol drinking
among young university students in Finland. Br J Addict. 1992;87:715–
722.
22. Hall W. Changes in the public perceptions of the health benefits of
alcohol use., 1989–1994. Aust N Z J Public Health. 1996;20:93–95.
REFERENCES
1. Eckardt MJ, File SE, Gessa GL, Grant KA, Guerri C, Hoffman PL, et al.
Effects of moderate alcohol consumption on the central nervous system.
Alcohol Clin Exp Res. 1998;22:998–1040.
2. Gunzerath L, Faden V, Zakhari S, Warren K. National Institute on Alcohol Abuse and Alcoholism report on moderate drinking. Alcohol Clin Exp
Res. 2004;28:829–847.
23. Hall W, Flaherty B, Homel P. The public perception of the risks and
benefits of alcohol consumption. Aust N Z J Public Health. 1992;16:
38–42.
24. Lowe G, Taylor SB. Effects of alcohol on responsive laughter and amusement. Psychol Rep. 1997;80:1149–1150.
25. Teahan JE. Alcohol expectancies of Irish and Canadian alcoholics. Int J
Addict. 1988;23:1057–1070.
3. Vaillant GE, Vaillant CO. Normality and mental health. In: Sadock BJ,
Sadock VA, eds. Kaplan’s & Sadock’s Comprehensive textbook of psychiatry. Philadelphia: Lippincott, Williams & Wilkins; 2005:583–597.
26. Marin G. Expectancies for drinking and excessive drinking among
Mexican-Americans and non-Hispanic whites. Addict Behav. 1996;21:
491–507.
4. US Departments of Agriculture and Health and Human Services. Nutrition and Your Health: Dietary Guidelines for Americans. Bethesda
(MD): USDA/DHHS; 1995.
27. Doty P, De Wit H. Effect of setting on the reinforcing and subjective
effects of ethanol in social drinkers. Psychopharmacology (Berl).
1995;118:19–27.
5. O’Brien R, Chafetz M. The encyclopedia of alcoholism. New York: Facts
on File Publications; 1982.
28. Lowe G, Britton R, Carpenter E, Castle H, Clayton C, Hulme C, et al.
Social drinking and laughter. Psychol Rep. 1997;81:684.
6. Brown SA, Goldman MS, Inn A, Anderson LR. Expectations of reinforcement from alcohol: Their domain and relation to drinking patterns. J Consult Clin Psychol. 1980;48:429–436.
29. Culbert DS. Drinking, expectancies, life stress and social support as predictors of alcohol use. Dissert Abst Int. 1989;51:424-B.
7. Gustafson R. Does moderate dose of alcohol reinforce feelings of pleasure,
well-being, happiness and joy? Psychol Rep. 1991;69:220–222.
30. De Castro JM. Social, circadian, nutritional and subjective correlates of
the spontaneous pattern of moderate alcohol intake of normal humans.
Pharmacol Biochem Behav. 1990;35:923–931.
8. Baum-Baicker C. The psychological benefits of moderate alcohol consumption: A review of the literature. Drug Alcohol Depend. 1985;
15:305–322.
31. Cooper ML, Frone MR, Russell M, Mudar P. Drinking to regulate positive
and negative emotions: A motivational model of alcohol use. J Pers Soc
Psychol. 1995;69:990–1005.
9. Chick J. Can light or moderate drinking benefit mental health? Eur Addict
Res. 1999;5:74–81.
32. Camacho TC, Kaplan GA, Cohen RD. Alcohol consumption and mortality in Alameda County. J Chronic Dis. 1987;40:229–236.
10. Peele S, Brodsky A. Exploring psychological benefits associated with
moderate alcohol use: a necessary corrective to assessments of drinking
outcomes? Drug Alcohol Depend. 2000;60:221–247.
33. Leifman H, Kuhlhorn E, Allebeck P, Andreasson S, Romelsjo A. Abstinence in late adolescence d Antecedents and covariates to a sober lifestyle and its consequences. Soc Sci Med. 1995;41:335–348.
11. Guallar-Castillon P, Rodriguez-Artalejo F, Ganan LD, Banegas JRB, Urdinguio PL, Cabrera RH. Consumption of alcohol beverages and subjective
health in Spain. J Epidemiol Community Health. 2001;55:648–652.
34. Alexander F, Duff RW. Social interaction and alcohol use in retirement
communities. Gerontologist. 1988;28:632–636.
12. Demers A, Kishchuk N, Bourgault C, Bisson J. When anthropology meets
epidemiology: Using social representation to predict drinking patterns.
Subst Use Misuse. 1996;31:847–871.
13. Stritzke WGK, Lang AR, Patrick CJ. Beyond stress and arousal: A reconceptualization of alcohol-emotion relation with reference to psychophysiological methods. Psychol Bull. 1996;120:376–395.
14. Poikolainen K, Vartiainen E, Korhonen HJ. Alcohol intake and subjective
health. Am J Epidemiol. 1996;144:346–350.
15. Power C, Rodgers B, Hope S. U-shaped relation for alcohol consumption
and health in early adulthood and implication for mortality. Lancet.
1998;352:877.
16. Poikolainen K, Vartiainen E. Wine and good subjective health. Am J
Epidemiol. 1999;150:47–50.
17. Grønbæk M, Mortensen EL, Mygind K, Andesen AT, Becker U, Gluud C,
et al. Beer, wine, spirits and subjective health. J Epidemiol Community
Health. 1999;53:721–724.
18. Cahalan D. Problem drinkers: A national survey. San Francisco: Jossey
Bass; 1970.
19. Roizen R. Loosening up: General population views of the effects of alcohol. In: Room R, Collins G, eds. Alcohol and disinhibition: Nature and
meaning of the link (Research Monograph 12; DHHS Publication No
(ADM) 83-1246). Rockville (MD): National Institute on Alcohol Abuse
and Alcoholism; 1983:236–257.
20. Makela K, Mustonen H. Positive and negative consequences related to
drinking as a function of annual alcohol intake. Br J Addict. 1988;
83:403–408.
35. Hanson BS. Social network, social support and heavy drinking in elderly
men d a population study of men born in 1914, Malmo, Sweden. Addiction. 1994;89:725–732.
36. Graham K. Alcohol abstention among older adults: Reasons for abstaining and characteristics of abstainers. Eur Addict Res. 1998;6:473–
487.
37. Cassidy K, Kotynia-English R, Acres J, Flicker L, Lautenschlager NT. Association between lifestyle factors and mental health measures among
community-dwelling older women. Aust N Z J Psychiatry. 2004;38:940–
947.
38. Murphy JG, McDevitt-Murphy ME, Barnett NP. Drink and be merry?
Gender, life satisfaction and alcohol consumption among college students.
Psychol Addict Behav. 2005;19:184–191.
39. Bell RA, Keeley KA, Buhl JM. Psychopathology and life events among
alcohol users and nonusers. In: Seixas F, ed. Currents in alcoholism.
New York: Grune & Stratton; 1977:103–123.
40. Neff JA, Husaini BA. Life events, drinking patterns and depressive symptomatology. J Stud Alcohol. 1982;43:301–318.
41. Neff JA. Life stressors, drinking patterns and depressive symptomatology:
Ethnicity and stress-buffer effects of alcohol. Addict Behav. 1993;
18:373–387.
42. Lipton RI. The relationship between alcohol, stress and depression in
Mexican Americans and non-Hispanic whites. Behav Med. 1997;23:
101–111.
43. Green CA, Freeborn DK, Polen MR. Gender and alcohol use: The roles of
social support, chronic illness and psychological well-being. J Behav Med.
2001;24:383–399.
S62
El-Guebaly
MODERATE DRINKING AND MENTAL HEALTH
44. Caldwell TM, Rodgers B, Jorm AF, Christensen H, Jacomb PA, Korten AE,
et al. Patterns of association between alcohol consumption and symptoms of
depression and anxiety in young adults. Addiction. 2002;97:583–594.
45. Paschall MJ, Freisthler B, Lipton RI. Moderate alcohol use and depression
in young adults: Findings from a national longitudinal study. Am J Public
Health. 2005;95:453–457.
46. Alati R, Kinner S, Najman JM, Fowler G, Watt K, Green D. Gender differences in the relationship between alcohol, tobacco and mental health
in patients attending an emergency department. Alcohol Alcohol.
2004;39:463–469.
47. Roberts R, Brunner E, Marmot M. Psychological factors in the relationship
between alcohol and cardiovascular morbidity. Soc Sci Med. 1995;
41:1513–1516.
AEP Vol. 17, No. 5S
May 2007: S55–S62
48. Vaillant GE. The natural history of alcoholism revisited. Cambridge
(MA): Harvard University Press; 1995.
49. Vasse RM, Nijhuis FJN, Kok F. Associations between work stress, alcohol
consumption and sickness absence. Addiction. 1998;93:231–241.
50. Mansson N-O, Rastam L, Eriksson K-F, Israelsson B. Alcohol consumption
and disability pension among middle-aged men. Ann Epidemiol.
1999;9:341–348.
51. Marchand A, Demers A, Durand P, Simard M. The moderating effect of
alcohol intake on the relationship between work strains and psychological
distress. J Stud Alcohol. 2003;64:419–427.
52. Riise T, Moen BE, Nortvedt MW. Occupation, lifestyle factors and healthrelated quality of life: The Hordaland Health Study. J Occup Environ Med.
2003;45:324–332.
Alcohol Drinking and Total Mortality Risk
ARTHUR L. KLATSKY, MD, AND NATALIA UDALTSOVA, PHD
To evaluate further the relation between alcohol consumption and total mortality, we have carried out new
Cox proportional hazards model analyses of 21,535 deaths through 2002 in the Kaiser Permanente study.
This follow-up includes 2,618,523 person-years of observation, with a mean follow-up of 20.6 years. We adjusted for age, sex, ethnicity, body mass index, marital status, education, and smoking. New methodology
was used to stratify light-moderate drinkers into groups felt more or less likely to include under-reporters.
The analysis reconfirms that the relation of alcohol drinking to total mortality is J-shaped, with reduced risk
(mainly because of less cardiovascular disease) for lighter drinkers and increased risk for persons reporting
more than 3 drinks per day. Infrequent (occasional) drinkers have risk similar to that of lifelong abstainers,
while former drinkers are at increased risk, especially for noncardiac death. The general shape of the relation of alcohol to mortality is similar for men and women. Age differences are substantial, with the apparent
benefit from light-moderate drinking not seen before middle life. Our data indicate further that the apparent magnitude of benefit of lighter drinking is probably reduced by systematic underreporting.
Ann Epidemiol 2007;17:S63–S67. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Cardiovascular Disease, Total Mortality.
INTRODUCTION
Total mortality has been studied as one global measure of
the effects of alcohol drinking. In 1926, Pearl (1) reported
that heavy drinkers had the highest risk of death but that
light-moderate drinkers were at lower risk than alcohol
abstainers; this was probably the first report of the J-shaped
alcohol-mortality relation. Pearl had no biological explanation for the possible benefit of lighter drinking and concluded that it was ‘‘not harmful.’’ In recent decades,
a number of studies and several meta-analyses (2–5) have
confirmed this J-shaped alcohol-mortality relation and
shown that it is a composite of disparate relations of alcohol
drinking to various conditions. Heavy drinkingdherein
operationally defined as usual daily intake of 3 or more
standard-sized drinksdis associated with excess mortality
from a number of cardiovascular and noncardiovascular
causes. The lower mortality risk associated with lighter
drinking is largely attributable to lower risk of coronary
heart disease (CHD) and other atherothrombotic vascular
conditions. Substantial evidence for several plausible biological mechanisms for protection by alcohol against CHD
has contributed to widespread acceptance of a causal hypothesis. These include effects via high-density-lipoprotein
(HDL) cholesterol, antithrombotic actions, better endothelial function, and other benefits (6–10).
In the absence of randomized controlled trial data, unresolved confounding might be involved in the apparent protection by light-moderate alcohol intake. Such confounding
would represent an alternative explanation for lower risk of
light drinking or higher risk of abstainers, but has not been
demonstrated despite many attempts. One hypothesis has
been that a spurious apparent relation might be the result
of inclusion of persons who become abstainers because of illness (‘‘sick quitters’’) among the abstainer referent group.
Originally propounded in the 1980s by Sharper, Wannamethee, and Walker (11), the ‘‘sick quitter’’ argument was recently reopened by a meta-analysis (12) that suggested
that this potential problem may also involve inclusion of
a second high-risk group among abstainers, one composed
of persons reporting drinking less than monthly. It was suggested that these ‘‘occasional’’ drinkers might include highrisk persons who reduced their intake because of illness (12).
PRIOR KAISER-PERMANENTE STUDIES
From the Division of Cardiology, Oakland Medical Center (A.L.K.) and
the Division of Research, Kaiser Permanente Medical Care Program
(N.U.), Oakland, CA.
Address correspondence to: Arthur L. Klatsky, MD, Senior Consultant in
Cardiology, Kaiser Permanente Medical Center, 280 W MacArthur Blvd.
Oakland, CA 94611. E-mail: [email protected].
Supported by the Robert Wood Johnson Foundation’s Program of Research Integrating Substance Use in Mainstream Healthcare and by a grant
from the Kaiser Foundation Research Institute. Data collection during the
period 1978–1985 was supported by the Alcoholic Beverage Medical
Research Foundation, Inc., Baltimore, MD.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
Our first report about alcohol and total mortality risk (13)
utilized a nested case-control design with four matched
groups of persons with different alcohol drinking categories.
Persons in each group were individually matched to the
2015 persons who reported 6 or more drinks per day. The
J-curve results were clear, as was the fact that lower risk of
light drinkers was due to fewer CHD deaths. However, the
data source was 1964–1968 health examinations, which
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.014
S64
Klatsky and Udaltsova
ALCOHOL DRINKING AND TOTAL MORTALITY RISK
AEP Vol. 17, No. 5S
May 2007: S63–S67
UPDATED MORTALITY ANALYSIS
Selected Abbreviation
Methods
CHD Z coronary heart disease
failed to separate ex-drinkers from lifelong abstainers and
lumped all light-moderate drinkers into the ‘‘2 or fewer
per day’’ category.
To correct these potential sources of bias for future work,
from 1978–1985, we collected more detailed alcohol intake
information from 127,212 persons who voluntarily took
health examinations at Northern California Kaiser Permanente facilities and who provided self-classified ethnicity
(14). The alcohol queries created clear separation of lifelong
abstainers from former drinkers and subdivided light-moderate drinkers into three categories. One of these was a large
category of persons who reported less than one drink per
month. The drinking category distributions of these subjects
were as follows: lifelong abstainers Z 15,272; ex-drinkers Z
4,124; ‘‘occasional’’ (!1 drink per month) Z 27 ,007; more
than1 drink per month but less than 1 drink per day Z
4,303; 1–2 drinks per day Z 23,110; 3–5 drinks per day Z
8,419; and more than 6 drinks per day Z 1,977.
Two published reports of total mortality are based on
these data (15, 16). Both used Cox proportional hazards
models with 7 covariates and included separate ex-drinker
and occasional drinker categories. With 4501 deaths
through 1988 (15) and 16,431 deaths through 1998 (16),
both analyses confirmed the J-curve demonstrated increased
risk of death among ex-drinkers and showed neither increased nor decreased risk among occasional drinkers.
ALL DEATHS
We present herein updated Cox proportional hazards model
analyses of 21,535 deaths through 2002. This follow up includes 2,618,523 person-years of observation, with a mean
follow-up of 20.6 years. Models described as ‘‘fully adjusted’’
included age ( 10 years), sex, ethnicity (white Z referent,
black, Asian, Hispanic, other), body mass index (weight
[kg]/height [M]2) (3 categories), marital status (3 categories), education (3 categories), smoking (4 categories),
and 7 alcohol intake categories. Detailed descriptions of
data and of analytic methods have been published (13–
17). New methodology used to stratify light-moderate
drinkers into groups thought more or less likely to include
under-reporters will be described in the following
paragraphs.
Results and Specific Comments
Covariate effects and sex differences. Comparison of
age-sex adjusted and fully adjusted models (Fig. 1, left panel)
shows an increased risk for ex-drinkers in both, no increased
risk for occasional drinkers in either, and a J-curve for lightmoderate and heavier drinking. The fully adjusted model exhibits less increase in risk for ex-drinkers and heavy drinkers
and more apparent benefit at light-moderate drinking. Stepwise models (not shown) indicated that the major factor
involved was control for smoking.
Data stratified by sex (Fig. 1, right panel) show more apparent benefit of light drinking in women. Explanations are not
evident, but could include differences in biological effects of
alcohol or in drinking pattern.
Ethnic differences. Data for whites, African Americans, and
Asian Americans (Fig. 2) indicate that the apparent benefit
2.2
2
2
1.8
1.8
1.6
1.6
MEN AND WOMEN
1.4
1.4
1.2
1.2
1
1
0.8
0.8
0.6
0.6
Never
Ex
Occas
<1 d
Age/Sex Adj
1-2 d
3-5 d
Fully Adj
6+ d
Never
Ex
Occas < 1 d
Men
1-2 d
3-5 d
6+ d
Women
FIGURE 1. Comparison of total mortality, age-sex adjusted, and fully adjusted model (left panel); fully adjusted data stratified by sex (right
panel).
AEP Vol. 17, No. 5S
May 2007: S63–S67
Klatsky and Udaltsova
ALCOHOL DRINKING AND TOTAL MORTALITY RISK
BLACKS AND WHITES
S65
ASIANS
1.6
3
1.5
2.5
1.4
1.3
2
1.2
1.5
1.1
1
1
0.9
0.5
0.8
0.7
0
Never
Ex
Occas < 1 d
1-2 d
3-5 d
Never
6+ d
Ex
White
Black
Occas < 1 d
1-2 d
3-5 d
6+ d
Asians
FIGURE 2. Fully adjusted models for blacks and whites (left panel) and Asians (right panel).
at lighter drinking is greatest in white persons, but a statistically significant lower risk at light drinking is also present in
African Americans. For Asian Americans, who have a much
larger proportion of abstainers and a smaller proportion of
deaths attributed to CHD, there is borderline reduced mortality risk at light drinking and a greater increased risk
among ex-drinkers and heavy drinkers.
Interval between baseline data and death. All relations appear
to become attenuated with passage of time (Fig. 3, left panel).
This probably is due to a general reduction of alcohol intake
in this population (18), resulting in less harm from heavy
drinking and less benefit from light-moderate drinking.
Cardiovascular and noncardiovascular deaths. The previously
observed marked disparity (15, 16) is again evident (Fig. 3,
right panel). The cardiovascular curve is substantially due
to CHD deaths. For cardiovascular mortality, even past
drinkers and heavy drinkers do not have much increase in
risk. This relative specificity argues against a confounded explanation for lower CHD risk among light-moderate
drinkers.
Differences according to baseline age. With few cardiovascular deaths, persons younger than 40 years old at baseline
show no benefit at light intake and have significantly increased risks starting at a reported intake of 1 to 2 drinks
per day (Fig. 4, left panel). In middle life (baseline age between 40 and 60 years) the J-curve appears. Among older
persons (Fig. 4, right panel), the benefits are more evident
and the risks of heavy drinking are not great, but they are
less favorable than those of light-moderate intake.
Differences according to likelihood of under-reporting.
Under-reporting of alcohol intake is widely believed to
be present (19), but it is virtually impossible to measure
and control for. The presumed effect is to systematically
reclassify some true ‘‘heavy’’ drinkers as apparent ‘‘lightmoderate’’ drinkers. Several likely effects on outcomes
DEATH WITHIN 10 & ≥ 10 YEARS
2
CV AND NON CV DEATHS
1.8
1.8
1.6
1.6
1.4
1.4
1.2
1.2
1
1
0.8
0.8
0.6
0.6
Never
Ex
Occas < 1 d
< 10 YRS
1-2 d
3-5 d
10+ YRS
6+ d
Never
Ex
Occas < 1 d
NON CV
1-2 d
3-5 d
6+ d
CV
FIGURE 3. Fully adjusted model for interval between baseline data and death (left panel); fully adjusted model for cardiovascular (CV)
and noncardiovascular (non-CV) deaths (right panel).
S66
Klatsky and Udaltsova
ALCOHOL DRINKING AND TOTAL MORTALITY RISK
AEP Vol. 17, No. 5S
May 2007: S63–S67
BASELINE AGE < 60
2.4
< 40
40-49
50-59
2.2
2
BASELINE AGE ≥ 60
1.8
60-69
70+
1.6
1.4
1.8
1.6
1.2
1.4
1
1.2
1
0.8
0.8
0.6
0.6
Never
Ex
Occas
<1d
1-2 d
3-5 d
6+ d
Never
Ex
Occas < 1 d
1-2 d
3-5 d
6+ d
FIGURE 4. Fully adjusted model for differences according to baseline age: Deaths before age 60 (left panel); deaths at or after age 60 years
(right panel).
would be a lowered apparent threshold for harm, creation of
an apparent continuous relation in place of a true threshold
relation, and lessening of the apparent benefit of lightmoderate drinking.
We tried to identify a group of persons among moderate
drinkers that was more likely to be under-reporters, inferred
from computer-stored information about lifetime responses
to queries about alcohol and about alcohol-related diagnoses. This subset included persons who, on another occasion,
indicated intake of three or more drinks per day or who ever
had a diagnosis of an alcohol-related condition; a detailed
description has been published (19). Twenty-seven percent
of persons reporting 1 to 2 drinks per day fell into this more
likely under-reporter group. These persons were about twice
as likely to have high aspartate aminotransferase levels (19),
a fact indirectly pointing to probable heavy drinking. The
LIKELY/ NOT LIKELY UNDER-REPORTER
1.4
1.3
1.2
1.1
1
0.9
likely under-reporting group accounted for much of the
relation of moderate drinking to increased prevalence of
systemic hypertension (19).
Fig. 5 shows mortality among non–heavy drinkers separately for those in the suspected under-reporter group and
those not in that group. Reduced mortality risk is concentrated in those unlikely to be under-reporters. The suspected
under-reporters show little if any apparent mortality benefit.
CONCLUSIONS
This new analysis reconfirms that the relation of alcohol
drinking to total mortality is J-shaped, with increased risk
for persons reporting more than 3 drinks per day and reduced
risk for lighter drinkers. The reduced risk of lighter drinkers
is due almost entirely to lower risk of death from cardiovascular disease. Former drinkers are at increased risk, especially for noncardiac death. Infrequent (occasional)
drinkers have risk similar to that of lifelong abstainers.
There are sex and ethnic differences in the magnitude of
the alcohol-associated risk, although the general shape of
the relation is similar.
Age differences are substantial, with greater increased
alcohol-associated risks among younger persons. Apparent
benefit from light-moderate drinking is not seen before middle life. Among confounders, control for associated smoking
is of special importance. The apparent magnitude of benefit
at lighter drinking is probably reduced by systematic underreporting.
0.8
Never
Ex
Occas
Likely
<1d
1-2 d
3-5 d
6+ d
Not Likely
REFERENCES
1. Pearl R. Alcohol and longevity. New York: Knopf; 1926.
FIGURE 5. Fully adjusted model for mortality among non-heavy
drinkers, separately for those in the suspected under-reporter
group and those not in that group.
2. Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15 diseases. Prev Med. 2004;38:613–
619.
AEP Vol. 17, No. 5S
May 2007: S63–S67
3. Corrao G, Rubbiati L, Bagnardi V, Zambon A, Poikolainen K. Alcohol
and coronary heart disease: a meta-analysis. Addiction. 2000;95:1505–
1523.
4. Maclure M. Demonstration of deductive meta-analysis: Ethanol intake
and risk of myocardial infraction. Epidemiol Rev. 1993;15:328–351.
5. Rehm J, Greenfield TK, Rogers JD. Average volume of alcohol consumption, patterns of drinking, and all-cause mortality: results from the US
National Alcohol Survey. Am J Epidemiol. 2001;153:64–71.
6. Booyse FM, Parks DA. Moderate wine and alcohol consumption: Beneficial effects on cardiovascular disease. Thromb Haemost. 2001;86:517–
528.
7. Dreon DM, Krauss RM. Chapter 20: Alcohol, lipids and lipoproteins. In:
Zakhari S, Wassef M, eds. Alcohol and the cardiovascular system. Rockville (MD): National Institutes of Health; 1997.
8. Greenberg SS, David Giles MD, Lancaseter J. Chapter 25: Alcohol, nitric
oxide, and vascular smooth muscle. In: Zakhari S, Wassef M, eds. Alcohol
and the cardiovascular system. Rockville (MD): National Institutes of
Health; 1997.
9. Hines LM, Rimm EB. Moderate alcohol consumption and coronary heart
disease: A review. Postgrad Med J. 2001;77:747–752.
10. Zakhari S. Molecular mechanisms underlying alcohol-induced cardioprotection: Contribution of hemostatic components. Alcohol Clin Exp Res.
1999;23:1108–1110.
Klatsky and Udaltsova
ALCOHOL DRINKING AND TOTAL MORTALITY RISK
S67
11. Sharper AG, Wannamethee G, Walker M. Alcohol and mortality in British men: Explaining the U-shaped curve. Lancet. 1988;3:1267–1273.
12. Fillmore KM, Kerr WC, Stockwell T, Chikritzhs T, Bostrom A. Moderate
alcohol use and reduce mortality risk: Systematic error in prospective
studies. Addict Res Theory. 2006;14:101–132.
13. Klatsky AL, Armstrong MA, Friedman GD. Alcohol and mortality. A ten
year Kaiser-Permanente experience. Ann Intern Med. 1981;95:139–145.
14. Collen MF, Davis LF. The multitest laboratory in health care. J Occup
Med. 1969;11:355–360.
15. Klatsky AL, Armstrong MA, Friedman GD. Alcohol and mortality. Ann
Intern Med. 1992;117:646–654.
16. Klatsky AL, Friedman GD, Armstrong MA, Kipp H. Wine, liquor, beer,
and mortality. Am J Epidemiol. 2003;158:585–595.
17. Arellano MG, Peterson GR, Petitti DB, Smith RE. The California Automated Mortality Linkage System (CAMLIS). Am J Public Health.
1984;74:1324–1330.
18. Klatsky AL, Armstrong MA, Landy C, Udaltsova N. The effect of coronary disease on changes in drinking in an older population. Alcohol
Res. 2003;8:211–213.
19. Klatsky AL, Gunderson E, Kipp H, Udaltsova N, Friedman GD. Higher
prevalence of systemic hypertension among moderate alcohol drinkers:
an exploration of the role of underreporting. J Stud Alcohol.
2006;67:421–428.
Panel Discussion II: Net Effects of Moderate Alcohol Consumption
on Health
This discussion focused on the effects of moderate alcohol consumption on conditions other than cardiovascular diseases and on total mortality. The panel was chaired by Richard Smallwood, and the panelists
were Henk F.J. Hendriks, Luc Letenneur, Klim McPherson, Gyongyi Szabo, Nady el-Guebaly, and Arthur
Klatsky.
Ann Epidemiol 2007;17:S68–S70. Ó 2007 Elsevier Inc. All rights reserved.
MODERATE ALCOHOL CONSUMPTION
AND TYPE 2 DIABETES
Smallwood initiated the discussion by asking if there is
enough evidence to support a causal relationship between
alcohol consumption and risk of type 2 diabetes. Hendriks
replied that while findings from epidemiologic studies
seem to be promising, he did not think that one could
draw a definite conclusion and that more research is needed
to understand the biological mechanisms. Eric Rimm stated
that ‘‘sick quitters’’ may be a concern for investigators when
they assess the effect of alcohol intake on the risk of coronary heart disease (CHD) because so-called ‘‘sick quitters’’
may have other diseases, such as hypertension, a risk factor
for CHD. Studies that evaluate the relationship between alcohol and type 2 diabetes, however, are unlikely to be influenced by ‘‘sick quitter’’ bias because no data suggest that
conditions such as hypertension or gout associated with excessive drinking would necessarily put a subject at higher
risk for type 2 diabetes.
ALCOHOL CONSUMPTION
AND COGNITIVE FUNCTION
Letenneur stated that while he and his colleagues found that
moderate alcohol drinkers tend to have lower risk of
Alzheimer’s disease, because of marked variability among
studies he is not convinced that such a reduced risk is attributable to alcohol consumption per se. He suspected that factors other than alcohol may play a role. Kenneth Mukamal
commented that there were disparities in findings with regard to a possible interaction between genes and alcohol
consumption on the risk of dementia. For example, the Rotterdam Study and the Cardiovascular Health Study (CHS)
had different results on the relation of apolipoprotein E to
Alzheimer’s disease; no one was able to explain why the results differed.
When asked about experimental studies on alcohol and
indices of cognition in animals, Letenneur responded that
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
several animal studies have shown that the administration
of alcohol improves memory. Some have shown that it
works through the retinoic acid pathway, and others found
that alcohol increases the survival of neurons of the hippocampus, which is a primary target of Alzheimer’s disease.
Jürgen Rehm asked about competing risk factors and
death, and how these factors were controlled for when studying alcohol consumption and Alzheimer’s disease. Letenneur commented that most studies, in fact, had a short
follow-up period; thus attrition should not be a major threat
to the validity of such studies. Furthermore, since drinkers
tended to have longer survival times, if competing or survival bias exists, it would only reduce the protective effects
of alcohol on cognitive function or risk of Alzheimer’s
disease.
Several participants commented that if light-to-moderate alcohol consumption does have a protective effect on
dementia, it should have a stronger effect on vascular dementia via its demonstrated effects on blood vessels. Mukamal stated that their studies had not shown significant
differences by type of dementia, and the results from the
CHS found a similar U-shaped relationship for vascular dementia and for Alzheimer’s disease between alcohol consumption and white matter assessed by magnetic
resonance imaging. Since many cerebral pathological
changes probably have a vascular component, he speculated
that there may be an overlap between these two diseases that
could be difficult to tease out perfectly.
ALCOHOL AND CANCER
Rimm stated that, to date, at least five studies have reported
that the relation of alcohol consumption to the risk of breast
cancer was modified by folate intake. Since animal studies
suggest that alcohol does not have direct carcinogenic effect, he was wondering if alcohol has an effect only on selected people in the population who are on a certain diet
or exposed to other risk factors.
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.015
AEP Vol. 17, No. 5S
May 2007: S68–S70
McPherson commented that the underlying biological
mechanisms linking alcohol consumption to the risk of cancer vary depending on the type of cancer. He suspected that
the further away the cancer site is from the gastrointestinal
tract, the more complicated the mechanisms are. He believed that the current findings of increased risk of breast
and colon cancer only suggest an association, rather than
prove causation. Mukamal pointed out there was strong evidence to suggest that moderate alcohol consumption is associated with higher levels of sex steroid hormones in
women, which explains an association with both higher
bone mineral density and with increased risk of breast cancer among moderate alcohol drinkers. Hendriks disagreed,
saying that data from his trials did not show any evidence
that alcohol intake increases the sex steroid hormones and
he suspected that other biological mechanisms rather than
the estrogenic effect of alcohol may play a role. Some investigators hypothesized that the risk of certain types of cancers,
such as head and neck or esophageal cancer, may be related
to the type of beverage consumed. However, McPherson
responded that, in his opinion, there is no clear evidence
supporting differences in risk of cancer according to
beverage type.
MODERATE ALCOHOL CONSUMPTION,
INFLAMMATION, AND LIVER DISEASE
Szabo said that data suggest that there is a linear relationship
between amount of fat deposition in the liver and progression of liver disease, as well as cirrhosis; the risk of progression to liver cirrhosis, even after stopping alcohol use, is
greater in women than in men. Szabo further commented
that distinguishing nonalcoholic steatosis hepatitis
(NASH) from alcoholic steatosis hepatitis (ASH) is difficult based on the pathologic findings. She agreed with others
that in many early studies, NASH may have been misdiagnosed as ASH. Cofactors, such as hepatitis B or hepatitis C,
were not known at the time of many early studies where
most cases were attributed only to alcohol consumption.
Rehm commented that both incidence and mortality of
liver disease vary greatly among countries. In European
countries, there are 5-fold differences in certain age groups
and 10-fold lower rates in women in some countries. Alcohol intake alone seemed unable to explain such big differences. He speculated that some other cofactors may
interact with alcohol to put certain people at much higher
risk. Szabo suggested that genetic factors that may be unique
to certain populations could play a role. For example, in certain parts of Europe there is a high frequency of genes leading to hemochromatosis. Subjects with these genes are at
higher risk of developing alcohol-induced liver disease and
have a low threshold for the amount of alcohol intake
S69
PANEL DISCUSSION
that will lead to its clinical manifestations. Likewise, the frequency of hepatitis C infection varies greatly in different
part of Europe and tends to be higher than in the United
States. Szabo speculated that genetic factors and hepatitis
infection may also play roles in the high frequency of
NASH in certain parts of Mexico and in the Hispanic population in the United States. Marjana Martinic asked if bacterial contamination of beverages in parts of Mexico may
contribute to high rates of cirrhosis. Rehm indicated that
the prevalence of liver cirrhosis is still high in those areas,
even after the local government tried to eliminate various
risk factors, including contaminated home brews.
Klatsky asked why all long-term heavy alcohol drinkers
do not develop alcohol-related liver disease. Szabo agreed
that was the case, in that only about 20% of heavy drinkers
develop liver disease. She added that a few studies have
looked at genetic factors that may affect susceptibility to cirrhosis but, as of now, we do not have very good markers
or genetic profiles that can serve as biomarkers to test the
population for such effects.
Ellison said that heavy drinkers have often been advised
to stop drinking completely for 1 or 2 days per week to let the
‘‘liver recover,’’ but most data currently available have
shown that the more frequently one drinks, the stronger
the protective effects on obesity, CHD, and total mortality.
He wondered if there is any evidence suggesting that people
should avoid alcohol consumption for a day or two per week
to protect their liver. Szabo stated that she was not aware of
any clinical data to support this claim. Rehm added that
their results from several different data sets all suggest that
the more spread out the same amount of alcohol consumption, the lower the risk of morbidity and mortality. He did
not believe there is sound empirical evidence to suggest
a practice of setting aside one day per week for not drinking,
as long as overall drinking is moderate.
DOES DRINKING WITH MEALS MODIFY
THE EFFECTS OF ALCOHOL?
It was stated that if one consumes alcohol with food, it reduces the blood alcohol levels by about one half; does this
suggest that drinking alcohol with meals could reduce the
toxic effects on the liver and reduce the risk of cirrhosis
and cancer by one half? Szabo did not think that studies
have carefully studied this issue, but admitted that drinking
patterns can certainly influence blood alcohol levels.
Morten Grønbæk added that data seem to suggest that the
protective effect of alcohol on CHD does not vary by
whether one drinks with meals or at other times. Since determination of blood alcohol levels requires blood samples,
it is not easy for investigators to distinguish between the first
pass metabolism and overall alcohol stimulation. Grønbæk
S70
PANEL DISCUSSION
AEP Vol. 17, No. 5S
May 2007: S68–S70
There was a comment about the difficulty of quantifying
some aspects of mental health that may result from alcohol,
such as relief of anxiety, feeling good, or happiness. ElGuebaly stated that it is a great challenge for studies of alcohol and mental health to get detailed information on mental
health status. While some instruments are available and
have been used to assess these dimensions, ‘‘the devil is in
the details.’’ Investigators should not put a lot of trust into
a simple question; one may have to ask numerous questions,
say 15 or 20, to obtain the necessary detailed information on
mental health.
responded that her question was pertinent, as the way that
investigators ask a question can certainly affect the results
of a study. However, in his studies almost 90% of subjects
could answer the questions, and questionnaires were completed within the clinic; thus literacy could not be a major
issue that would affect the findings he presented.
Rehm pointed out that some studies have shown that the
protective effects of alcohol consumption on CHD are less
apparent among certain ethnic groups, for example, African
Americans. He wondered if such a difference could be attributed to different drinking patterns among different racial
groups. Klatsky commented that differences in drinking pattern is certainly one explanation, and he asserted that African Americans, in general, tend to drink less than white
persons. Alun Evans raised the issue of whether the lack of
protective effects on mortality among the Asian population
in Klatsky’s study could be due to cultural prohibitions
against alcohol among Asians, possibly leading to underreporting bias. Klatsky responded that indices of heavy
drinking were also infrequent among Asians, suggesting
that under-reporting was not a major problem in that group.
He added that bias, if a factor, should be present across all
categories of alcohol consumption.
MODERATE ALCOHOL CONSUMPTION
AND TOTAL MORTALITY
AGE TO BEGIN ALCOHOL CONSUMPTION
Smallwood commented on the view among some scientists
that the protective effects of alcohol on CHD have been exaggerated and that such protection occurs only at high levels
of drinking. This issue has recently been considered by officials in Australia and may prompt changes in policy in that
country to make alcohol less accessible. In response to this,
Klatsky said that he was not aware of any new evidence
suggesting that only heavy drinking is associated with protection against CHD. In fact, almost all studies have consistently shown a U-shaped curve between amount of alcohol
intake and CHD mortality, as well as total mortality, with
the lowest mortality rate among moderate alcohol drinkers.
He added that there are certainly errors in assigning the specific cause of death, but misclassification of total mortality
is unlikely to occur. Thus the statement that any protective
effect of alcohol on mortality requires high levels of intake is
not justified by recent scientific data.
Kaye Fillmore said that since alcohol consumption data
are usually based on questionnaires, she was concerned
whether investigators were able to get reliable data in studies
that included subjects with illiteracy or semiliteracy. Klatsky
The question was raised as to the age at which drinking may
have potentially beneficial health effects. Rimm believed
that there is no beneficial effect among people younger
than 50 years of age. He stated, however, that he was not
aware of any study that allowed the investigators to test
whether people who started drinking alcohol before age 50
years or later would have different risks of heart disease or
mortality. Furthermore, the latency period of any effect of
alcohol on heart disease is likely to be long, especially if
we believe that biological mechanisms of such protective effects are mediated through alcohol’s impact on high-density
lipoprotein cholesterol and on inflammation. Thus consumption in the 30s or 40s might lead to benefits later in
life. Ellison mentioned that he was aware of one study
among relatively young people, by Rosenberg and colleagues
from 1981, at a time when myocardial infarction (MI)
among women was high because of the combination of cigarette smoking and birth control pills containing high levels
of sex steroids. That study reported that the risk of MI among
women younger than 50 years was 30% lower for drinkers
than for abstainers.
believed that a body exposed to the same amount of alcohol
will probably generate the same amount of reactive oxygen
radicals. McPherson pointed out that most studies on alcohol and cancer basically looked at consumption, irrespective
of when and how alcohol is consumed, and he was not aware
of any study that examined alcohol consumption and cancer
according to whether alcohol is consumed with or without
food.
MODERATE ALCOHOL CONSUMPTION
AND MENTAL HEALTH
Why We Don’t Know More about the Social Benefits
of Moderate Drinking
DWIGHT B. HEATH, PHD
The scientific literature about drinking shows that a preoccupation with the harms that occasionally befall
those who drink too much or too fast is coupled with a virtual ignorance (in both senses of that term) about
the benefits that more often accrue to the vast majority who are moderate drinkers. It is ironic that reviews
of the subject that appear under the World Health Organization (WHO) imprint have paid almost no attention to two thirds of WHO’s definition of health: ‘‘.mental and social well-being.,’’ while focusing
overwhelmingly on the physical aspects. This imbalance is so striking that in a recent WHO-affiliated publication the authors suggested that social consequences may be ‘‘the forgotten dimension.’’ This, despite the
fact that drinking plays very important roles in many societies with respect to such widely recognized and
appreciated benefits as celebration, relaxation, sociability, enhancement of food, concretization of the
social order, and time out, in addition to medical and therapeutic benefits. The time is ripe for increasing
collaboration to investigate such social benefits in ways that would be of greater interest and of use to those
who heretofore have neglected them.
Ann Epidemiol 2007;17:S71–S74. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Health Behavior, Social Behavior, Social Values, Health Benefits.
When first invited to write for this conference, it seemed appropriate to do a comprehensive review of the literatured
but I had pretty much done that a few years earlier and there
has been very little additional work done in the interim. To
be sure, that book (1) included discussions of ‘‘heavy’’ or
‘‘excessive’’ drinking as well as moderate drinking, but the
emphasis on social benefits was more sharply delineated
than in most of what is written about alcohol; the scope
was worldwide and with considerable historical depth; and
the ‘‘types’’ of drinking were clearly specified. Instead of
briefly recapitulating some of the major points that I have
already made elsewhere, it seems appropriate, in this context, that I come to grips with the simple but serious questions of why there is still so little reported in the scientific
literature about social benefits of drinking. This is especially
pertinent in view of the fact that drinking has long been so
widespread among societies of the world. Then too it is often
viewed as a quintessentially social activity. And it is so much
more often enjoyed and enjoyable than it is harmful, risky,
or problematic.
The major points of my paper are elementary: it would
appear logical and relevant that the familiar panoply of
so-called ‘‘alcohol-related problems’’ be somehow arrayed
From the Department of Anthropology, Brown University, Providence,
RI.
Address correspondence to: Dr. Dwight B. Heath, Professor of Anthropology, Brown University, 47 Barnes St., Providence, RI 02906. Tel.:
(401) 863 3251.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
and weighed against a yet-to-be-articulated similar list of
‘‘alcohol-related solutions’’ or ‘‘.satisfactions.’’ Also, the
heretofore largely neglected social component of health
should be more seriously addressed.
Those who founded the World Health Organization
(WHO) were careful to be both inclusive and exclusive in
spelling out their simple definition of health: ‘‘.a state of
complete physical, mental, and social well-being, and not
merely the absence of disease or infirmity’’ (2, p. 100). During WHO’s 60 years there has been a great deal of attention
paid to the links between alcohol and physical well-being,
but very little about such links with either mental or social
well-being. All I ask is that there be some attempt to redress
that imbalance.
It is a pleasure to let someone else who can speak with
more authority champion the mental aspect (3), but I feel
both qualified and obliged to address the general ignorance
(or perhaps it is willful ignorance) of what I consider to be
the equally impressive linkage between alcohol use and
social well-being.*
*For close to fifty years, I have pretty well tried to keep my finger on the
pulse of social and cultural studies about uses of alcohol around the world
and throughout history. (Don’t just take my word for it: a colleague
whom I’ve never met or heard from wrote last year that ‘‘There is no
more important figure in the international anthropological study of alcohol.(4, p. 4). Such an effort may not qualify for the lofty designation of
‘‘meta-analysis,’’ but it is a substantial review and as close as we get in terms
of qualitative studies.
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.016
S72
Heath
SOCIAL BENEFITS OF MODERATE DRINKING
Selected Abbreviations and Acronyms
ICAP Z International Center for Alcohol Policies
WHO Z World Health Organization
It seems at first paradoxical that the social benefits of
drinking have been virtually ignored. But the intellectual
history of what little work has been done on the subject
pretty clearly reveals not only the gaps but some clear reasons why they persist even after half a century of burgeoning
international and multidisciplinary studies of alcohol.
The problem is not, as some suggest, that too few alcohol
investigators are clearly communicating their findings to
public health and other authorities, or that they are neglecting to make the policy implications apparent. On the contrary, there is a coterie of alcohol investigators who have
gone to great lengths to communicate their findings, clearly
and forcefully, to public health and other officials, and who
carefully spell out in great detail what they consider to be the
policy implications. Unfortunately, those investigators are
not involved in the kinds of meticulous, long-term, biological, medical, and other scientific research such as has been
discussed in this conference. And the public implications
that they recommend are diametrically opposed to what
might be made if they were truly concerned to promote public health and social welfare and to minimize the risks that
are associated with alcohol.
I refer to a small but vocal international group of sociologists, statisticians, and others who like to characterize their
work as ‘‘epidemiological’’ and yet whose methods differ
markedly from those that have been discussed here, and
whose findings consistently emphasize the risks and harms
of drinking, with only rare allusion to (and no systematic
investigation of) the benefits. Not surprisingly, the recommendations are consistently a variety of restrictive measures
by which they suggest that an administration could and
should diminish alcohol consumption throughout the
population.
They insist that they do not embrace the idea of prohibitiondexcept for ‘‘young’’ people and females ‘‘who might
become pregnant’’dbut they unabashedly favor less drinking by everyone and claim that many of the risks of harm
are very similar for moderate drinkers as for heavy or excessive drinkers. Their policy recommendations have been
consistent for decades and supposedly apply to various countries and cultures: increased taxation of beverage alcohol,
limited hours and outlets for the sale thereof, an elevated
‘‘minimum drinking age,’’ increased restrictions on advertising, and tighter controls on commerce, among others.
It is in that context that I feel it is important simply to
mention a number of the social benefits of moderate drinking that are often cited. With few exceptions, those who
AEP Vol. 17, No. 5S
May 2007: S71–S74
have written in any detail about social aspects of drinking*
found in their careful analyses of populations of various sizes,
ethnic compositions, political and economic systems, or
other differences, that moderatey drinking is consistently
linked with such positive associations as celebration, feasting, sociability, in-group identification and social support,
relaxation, the enjoyment of food, and other aspects of
social well-being (1). Please note that these are not just
anthropologists referring to quaint and curious customs of
small, exotic, and relatively isolated tribes; the same holds
almost equally for urban industrial populations that are in
the mainstream of Western cultural traditions, intimately
engaged in promoting globalization.
A dirty little quasi-secret is that most of what has been
written under the rubric of social and epidemiological studies of alcohol does not, like the studies I mentioned in the
previous paragraph, deal with behavior or even with spontaneous comments by drinkers and their associates, nor with
close observation and careful analysis by trained and experienced observers.
Using sampling, survey, and statistical techniques that
had been developed for marketing research in the United
States, some sociologists and others claim to count and measure drinking practices, associated attitudes, and outcomes
among various populations. Such studies are actually based
on responses to questionnaires (the preferred term is ‘‘social
surveys’’), in which the questions have been carefully
worded and ordered in sequence by specialists who need
have had no prior familiarity with the populations being
studied, and respondents are usually required to choose
among a small number of precoded standard answers (rather
than in their own words). Often the data are gathered by
part-time low-paid employees who know little about alcohol, research, or the population, and usually by telephone.
This is not news to this readership, but most others would
be surprised and concerned to learn that it is social research
with no social component; data-gathering about behavior
that in only a very remote way is associated with the behavior in question.
In-house publications and a flurry of professional journals
that proliferated in the last quarter of the 20th century
spread the word about such new findings on the newly
important subject of alcohol. Then, as the self-designated
official-sounding Alcohol and Policy Project, a small
*What I refer to as the relevant literature is by social scientists of different disciplines, various national and educational backgrounds, and contrasting theoretical and methodological orientations, most of whom have
no special interest in drinking or its outcomes, but who reported on them
in connection with projects on very different topics.
y
ICAP (5) has already documented how varied are official views of what
constitutes ‘‘moderate’’ or ‘‘sensible’’ drinking; I see no need to cavil beyond
the definitions that are discussed there. A simple approximation may be
‘‘drinking without drunkenness.’’
AEP Vol. 17, No. 5S
May 2007: S71–S74
international cadre of investigators, periodically reviewed
and summarized their work in a series of multiauthored
books (6–8).
It pretty much started a generation ago with an overview
of the sparse epidemiological and social literature on drinking that was then available (9). That book ended with the
appropriately tentative conclusions that ‘‘.changes in the
overall consumption of alcoholic beverages have a bearing
on the health of a people in any society. Alcohol control
measures can be used to limit consumption; thus control
of availability becomes a public health issue’’ (9, p. 90).
Although the exact nature of that ‘‘bearing’’ was never
made clear, those two sentences quickly became the cornerstone of the powerful and lucrative ‘‘consumption theory of
prevention,’’ whereby less consumption was said to result in
fewer problems and more consumption in more problems.
To his credit, it was one of their own number (10) who
meticulously traced the steps by which the original cautious
sociological statement was rapidly transformed by a few bureaucrats into a starkly axiomatic assertion of constant and
universal correlation.
Their simplistic rule, that more consumption correlates
with more problems, and that less consumption would result in fewer problems, was contradicted by abundant empirical evidence (often including their own data), but it
persists nevertheless. No one now should be comfortable
with the recent remark that ‘‘.from the point of view
of minimizing the social harm from drinking, the general
conclusion is that the lower the consumption the better’’
(6, p. 85); it is as if a decade of intensive scrutiny of drinking patterns had never taken place, during which even
some of the most dyed-in-the-wool proponents of the consumption theory have belatedly conceded that how one
drinks must be considered as well as how much one drinks
(11), especially as relates to the occurrence of alcoholrelated problems.
All of my carping about methods would be of little relevance if the survey-researchers actually addressed health issues in the spirit of the WHO definition. But they have
focused almost solely on pathology rather than well-being,
and the few data that they offer concerning mental and social harms are generally garnered from police-blotters, hospital records, and other sources that were never collected
or intended for such use.
The contrast with descriptive analyses is dramatic. The
fact that a small number of dissidents wrote about benefits
as well as harms associated with drinking prompted Room
(12) to be dismissive of qualitative studies about drinking
and its outcomes. Zeroing in on ethnographers, his interpretation was that their naı̈ve indiscretion of what he called
‘‘problem deflation’’ was not accidental but deliberate,
fueled by reaction against a parental ‘‘dry generation’’ (ignoring the fact that he was describing age-mates), combined
Heath
SOCIAL BENEFITS OF MODERATE DRINKING
S73
with blind allegiance to an obsolete and simplistic functionalist world view, and a romantic inclination to side with any
remote and noble ‘‘Other’’ in a Rousseauan rejection of
Western values.
But if the scientific method has anything to do with social
reality, it must take some account of what the majority of
people believe and say most of the time about their own
experience and how it relates to the process of living in
the complex relationships that comprise our social universe.
The fact is that most people who drink do so deliberately,
and they say that they enjoy it and benefit from it.
Social well-being, already recognized as a major component of the very definition of health, is also a major contributor to that other intangible (but crucial), quality of life. If
the prevailing list of ‘‘alcohol-related problems’’ is deplored
for its erosion of quality of life, then certainly it is time to
recognize some list of ‘‘alcohol-related benefits’’ that are
generally said to enhance quality of life.
It is not up to me to propose a list of questions (knowing
how critical are wording, sequencing, alternative choices,
and other technical details of survey-research), but surely
our ingenious and experienced colleagues who specialize
in such work can devise reliable and significant ways to elicit
measurable data about the predominantly pleasant outcomes that most people associate with alcohol, just as they
have managed to elicit what they consider useful data about
the unpleasant experiences and expectations (which, in the
real world, are far less prevalent.)
Probably the nearest approximation to date is a book in
which the editors rhetorically ask whether the social consequences of alcohol consumption are ‘‘the forgotten dimension (13, p. 1).’’ Two other contributors to that volume
allude to benefits as well as harms of drinking: Pernanen
(14, p. 55) specifically mentions meeting new people and
making new friends, and Rehm (15, p. 12) lists relaxation,
stress-reduction, improved social events, general subjective
well-being, group cohesion, cultural identification, benefits
from taxation, and (unspecified) ‘‘mortality, morbidity and
disability reduction if used regularly and moderately.’’ On
the right track, Rehm (15, p. 15) then asks, ‘‘For what outcomes do we have solid evidence about the relation between
alcohol and outcome?’’ to which I hasten to reply that there
are many. A few examples, with substantive references,
include sociability (16); celebration (17); adjunct to ritual
(18); relaxation (19); enhancement of food (20); social
integration (21); in-group identity (4); feasting and rewarding political supporters (22); facilitating expressions of
affect, rewarding reciprocal labor, building social capital,
settling disputes (23); as well as negotiating business, offering homage to spirits, deities and/or ancestors. And there are
many others too numerous to mention.
Although a number of social harms, some of them quite
serious, are commonly reported in connection with heavy
S74
Heath
SOCIAL BENEFITS OF MODERATE DRINKING
or excessive drinking, no such harms tend to be reported in
connection with moderate drinking.
A major hurdle that remains is how to ‘‘operationalize’’
the ‘‘translation’’ of what most social scientists consider
‘‘solid evidence’’ (often in expository qualitative format
that others dismiss as ‘‘anecdotal’’) to the kind of evidence
that alcohol-epidemiologists consider solid (numerical,
quantitative, subject to easy tabulation, graphing, and statistical manipulation).
In the interest of expanding and improving our knowledge of both social relations and alcohol, this would appear
to be long overdue. In the interest of rounding out our
understanding of the human experience, it poses an exciting
opportunity for mutually fruitful collaboration among investigators whose concerns are similar but whose paths too
rarely converge. In the interest of scientific accuracy, I propose this as an important challenge as well as an important
opportunity.
REFERENCES
AEP Vol. 17, No. 5S
May 2007: S71–S74
8. Holder HD, Edwards G, eds. Alcohol and public policy: Evidence and
issues. Oxford (UK): Oxford University Press; 1995.
9. Bruun K, Edwards G, Lumio M, Mäkelä K, Pan L, Popham ER, et al.
Alcohol control policies in public health perspective. Helsinki: Finnish
Foundation for Alcohol Studies; 1975.
10. Room R. Alcohol control and public health. Annu Rev Public Health.
1984;5:293–317.
11. Heath DB. Anthropology, alcohol, and a parallel career: Serendipity compounded. Social History Alcohol Rev. 2003;17:9–30.
12. Room R. Alcohol and ethnography: A case of ‘‘problem deflation’’? Curr
Anthropol. 1984;25:169–191.
13. Klingemann H, Gmel G, eds. Mapping the social consequences of alcohol
consumption. Dordrecht: Kluwer Academic; 2001.
14. Pernanen K. Consequences of drinking to friends and the close social
environment. In: Klingemann H, Gmel G, eds. Mapping the social consequence of alcohol consumption. Dordrecht: Kluwer Academic; 2001:
53–65.
15. Rehm J. Concepts, dimensions, and measures of alcohol-related social consequencesdA basic framework for alcohol-related benefits and harm. In:
Klingemann H, Gmel G, eds. Mapping the social consequence of alcohol
consumption. Dordrecht: Kluwer Academic; 2001:11–19.
16. Oldenberg R. The great good place: Cafes, coffee shops, community
centers, beauty parlors, general stores, bars, hangouts, and how they get
you through the day. New York: Paragon House; 1989.
1. Heath DB. Drinking occasions: Comparative perspectives on alcohol and
culture. Philadelphia: Brunner/Mazell; 2000.
17. Peele S, Grant M, eds. Alcohol and pleasure: A health perspective. Philadelphia: Brunner/Mazell; 1999.
2. World Health Organization. Preamble to the Constitution of the World
Health Organization as adopted by the International Conference. WHO;
1948.
18. McAllister PA. Xhosa beer drinking rituals: Power, practice and performances in the South African rural periphery. Durham (NC): Carolina
Academic Press; 2006.
3. El-Guebaly N. Investigating the association between moderate drinking
and mental health. Ann Epidemiol. 2007;17:S55–S62.
19. Coulombe CA, ed. The muse in the bottle: Great writers on the joys of
drinking. New York: Kensington Publishing; 2002.
4. Wilson TM, ed. Drinking cultures: Alcohol and identity. Oxford (UK):
Berg; 2005.
20. de Garine I, de Garine V, eds. Drinking: Anthropological approaches. New
York: Berghahn; 2001.
5. International Center for Alcohol Policies. ICAP Reports 14: International
drinking guidelines. Washington: International Center for Alcohol Policies; 2004.
21. Douglas M, ed. Constructive drinking: Perspectives on drink from anthropology. Cambridge (UK): Cambridge University Press; 1987.
6. Babor T, Caetano R, Casswell S, Edwards G, Giesbrecht N, Graham K,
et al. Alcohol: No ordinary commodity: Research and public policy. Oxford: Oxford University Press; 2003.
7. Edwards G, Anderson P, Babor TF, Casswell S, Ferrence R, Giesbrecht N,
et al. Alcohol policy and the public good. Oxford (UK): Oxford University
Press; 1994.
22. Dietler M, Hayden B. Feasts: Archaeological and ethnographic perspectives on food, politics, and power. Washington (DC): Smithsonian Institution Press; 2001.
23. Butler BY. Holy intoxication to drunken dissipation: Alcohol among
Quichua-speakers in Otavalo, Ecuador. Albuquerque: University of New
Mexico Press; 2006.
Patterns of Alcohol Consumption and Cardiovascular
Risk in Northern Ireland and France
ALUN EVANS, MD, PEDRO MARQUES-VIDAL, MD, PHD, PIERRE DUCIMETIÈRE, PHD,
MICHELE MONTAYE, MD, DOMINIQUE ARVEILER, MD, ANNIE BINGHAM, MA,
JEAN-BERNARD RUIDAVETS, MD, PHILLIPE AMOUYEL, MD, PHD,
BERNADETTE HAAS, MD, JOHN YARNELL, MD, JEAN FERRIÈRES, MD, MSC,
FREDERIC FUMERON, PHD, GERALD LUC, MD, FRANK KEE, MD,
AND FRANCOIS CAMBIEN, MD
The PRIME Study was begun in 1991 and recruited 10,600 men aged 50 to 59 years in the WHO MONICA
Project centers of Belfast, Lille, Strasbourg and Toulouse. Although drinkers in France and Northern Ireland consumed almost identical amounts, the pattern of consumption was different. In Northern Ireland
beer and spirits were the staple beverages, whereas in France it was predominantly red wine; in France,
90% of men drank at least one unit per week versus 61% in Northern Ireland. Frenchmen drank evenly
throughout the week, whereas in Northern Ireland two thirds of the consumption took place on Friday
and Saturday nights. In the 5-year follow up of PRIME in France, the usual cardiovascular protective effect
of increasing consumption (up to 45 units per week) was shown and the level of significance for trend in
consumption was highly significant (p Z 0.006); in Northern Ireland, this pattern was less consistent
and did not attain significance. It remains a matter of conjecture whether in Northern Ireland the beneficial
effects of alcohol consumption were annulled by a pattern of drinking that increases blood pressure, a wellestablished risk factor for heart disease, or whether the protection in France resulted from the consumption
of wine along with food.
Ann Epidemiol 2007;17:S75–S80. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Cardiovascular Disease, Drinking Behavior.
BACKGROUND
In 1980, Ducimetière and his colleagues (1) drew comparisons between the Paris Prospective Study, the Seven Countries Study, and the Pooling Project. In France, coronary
heart disease (CHD) death rates were distinctly lower
than those in Northern and Eastern Europe and the United
States. The authors reported that ‘‘numerous hypotheses
have been proposed to explain the anomalous epidemiology
of CHD in Francedconsumption of olive oil, of alcohol, especially wine, garlic, and so on.’’ The authors quite properly
continued: ‘‘The answers can only come from careful prospective study,’’ and that ‘‘age, blood pressure and cholesterol levels are responsible for comparable risk gradients in
France just as in other countries.’’ They added that ‘‘not
From the PRIME Study Group (a complete listing of members and affiliations may be found at the end of the article).
The PRIME Study was funded by the Research and Development Office,
a directorate of the Northern Ireland Health and Social Services Central
Services Agency (RRG project 9.8), and the Merck, Sharpe and DohmeChibret Laboratory under an agreement between INSERM and participating laboratories (see end of article for the list of laboratories).
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
measured or not yet identified variables contribute to the
risk of CHD among American men.’’
The following year, three of these authors coined the
term ‘‘The French Paradox’’ after comparing levels of heart
disease with fat intake data provided by the Organisation for
Economic Cooperation and Development (OECD) (2).
Consumption of animal fat in France was high despite the
fact that CHD mortality data showed low age-standardized
rates for males (35–64 years of age), lying just above Greece
and Japan.
Different diagnostic practices in France had been invoked as a cause for the paradox. The World Health Organization’s (WHO) MONICA Project was set up to
evaluate the contribution of major risk factor change to diverse trends in CHD and stroke incidence around the world
(3). MONICA rankings, based on MONICA categories of
fatal events, placed some middle- and low-mortality populations, such as the French, higher than they would be based
on official CHD mortality rates. However, rates for non-fatal
myocardial infarction correlated quite well with the official
mortality for the same populations. Moreover, when ranked
by event rates for certain MONICA definitions, French
MONICA populations had low rates that were comparable
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.017
S76
Evans et al.
ALCOHOL INTAKE IN NORTHERN IRELAND AND FRANCE
Selected Abbreviations and Acronyms
CETP Z cholesterol ester transfer protein
CHD Z coronary heart disease
ECTIM Z Etude Cas-Temoins de l’Infarctus du Myocarde
HDL Z high-density lipoprotein
MONICA Z Multinational monitoring of trends and determinants in cardiovascular disease
MORGAM Z MONICA, Risk, Genetics, Archiving, and Monograph
OECD Z Organisation for Economic Cooperation and Development
PRIME Z Prospective Epidemiological Study of Myocardial Infarction
WHO Z World Health Organization
AEP Vol. 17, No. 5S
May 2007: S75–S80
per day but increased commensurately with greater intakes.
The odds ratios for myocardial infarction in B2 homozygotes
decreased from 1.0 in nondrinkers to 0.34 in those drinking
75 g/d or more. Similar but weaker effects were also present
for the B1 allele.
On the strength of the results from the ECTIM Study, it
was decided to mount a prospective study, the Prospective
Epidemiological Study of Myocardial Infarction (PRIME)
(10), whose methods are outlined below.
Methods
with those in other European populations in Italy, southern
Germany, Switzerland. and Spain.
In a dietary comparison of Belfast and Toulouse mounted
within the MONICA Project (4), alcohol consumption in
Toulouse was significantly higher than in Belfast: the Toulousains drink red wine, rather than beer and spirits, and
drink daily; in Belfast fewer people drink but those who do
make up for it, particularly in binge drinking at weekends.
Large studies have now shown a reduced relative risk for
CHD in moderate drinkers. It is suggested that a steady consumption of wine in France reduces platelet aggregation (5)
and that binge drinking induces rebound aggregation.
The Étude Cas-Tèmoins de l’Infarctus du Myocarde (ECTIM) Study is a large case-control study of myocardial
infarction mounted between the MONICA centers in
Belfast, Lille, Strasbourg, and Toulouse, for which male patients 25 to 64 years of age were recruited (6). Despite an almost four-fold higher level of heart disease in Northern
Ireland than in France, classical risk factor levels were
very similar. After adjusting for classic risk factors and country of recruitment by logistic regression, alcohol displayed
a protective effect against myocardial infarction (7). The
magnitude of the effect, which was comparable in the two
countries, was related to wine consumption in France and
beer and spirits consumption in Northern Ireland. This effect largely disappeared when high-density lipoprotein
(HDL) cholesterol was included in the model. Among 605
controls in ECTIM, of a host of lipid and hemostatic factors
investigated, only HDL parameters and plasminogen activator inhibitor 1 activity were increased by alcohol consumption (8).
In a later work from the ECTIM Study, an important
gene-environment interaction between alcohol consumption and a polymorphism of the cholesterol ester transfer
protein (CETP) gene (CETP/Taq1B) was observed (9).
CETP cycles HDL cholesterol to the low-density lipoprotein
cholesterol pool, and reduced levels of enzyme will lead to
higher levels of HDL cholesterol. The gene has two alleles,
B1 (60%) and B2 (40%), and B2 carriers had highly significantly lower levels of CETP and higher levels of HDL cholesterol. The effect on plasma HDL cholesterol was not
observed in B2 carriers drinking less than 25 g of alcohol
The PRIME study recruited 10,600 men aged 50 to 59 years
from the populations of the WHO MONICA Centres of
Belfast, Northern Ireland, and Lille, Strasbourg and Toulouse in France from 1991 through 1993, assembling roughly
equal cohorts in each center. The basic procedures have
been described (11) but, briefly, the standard questionnaire
also gathered information on medical history, chest pain,
physical activity, and tobacco and alcohol consumption.
Participants were informed of the aims of the study, and
those who agreed to take part signed a consent form. Ethical
clearance was obtained from the appropriate local Research
Ethics Committees.
Alcohol consumption was assessed by means of a validated questionnaire that recorded the subjects’ mean consumption (in units) of wine, beer, cider, and spirits for
each day of the week. The intake of alcohol (expressed in
milliliters of pure ethanol/week) was estimated from the average number of milliliters of ethanol in one unit of each
type of alcoholic beverage (11). Subjects were considered
former drinkers if they reported no current alcohol consumption but indicated that they had consumed alcohol in
the past (for any period of time).
Blood pressure was measured once at the end of the examination after a 5-minute rest in the sitting position. Measurements were performed with an automatic device (Spengler
SP9), which also recorded heart rate (12).
Follow up for CHD events over 5 years
Subjects were contacted annually by postal questionnaire,
by telephone, or through their family doctors. Possible
events were followed up through the hospital or the family
doctor. All death certificates were recorded and the circumstances ascertained; loss at follow-up amounted to 0.8% in
Belfast and 1.4% in France. A medical committee independently validated the coronary events with information gathered from all medical sources. Myocardial infarction and
definite coronary deaths were defined according to MONICA criteria. ‘‘Hard’’ CHD cases were defined as coronary
deaths or nonfatal myocardial infarction (13). The criteria
also allowed for the classification of cases of angina, differentiating stable and unstable angina.
AEP Vol. 17, No. 5S
May 2007: S75–S80
Evans et al.
ALCOHOL INTAKE IN NORTHERN IRELAND AND FRANCE
body mass index, heart rate, tobacco smoking, educational
level, marital status and professional activity, it was found
that blood pressure was higher in Northern Irish drinkers
screened on Mondays and Tuesdays and decreased until
Thursday (14). In French drinkers, blood pressure levels
were constant throughout the week (day by country interactions, p ! 0.05). In both countries, drinkers had higher
blood pressure levels than non-drinkers. Data from the Belfast MONICA Register for 1991–1993 for fatal definite and
possible myocardial infarction indicated a higher prevalence
of events on Mondays and Tuesdays and a lower incidence
on Fridays (p Z 0.05) in both sexes.
A total of 9750 subjects, free of CHD at baseline, were included in the analyses of the 5-year follow up: 7532 in France
and 2398 in Northern Ireland. There were 106 coronary
deaths or myocardial infarctions and 94 angina pectoris
events in France; the corresponding events in Northern Ireland were 61 and 60. A combined classification into ‘‘CHD
events’’ (angina pectoris, myocardial infarction, or coronary
death) (12) yielded 197 events (2.7%) in France and 121
(5.0%) in Northern Ireland. In France, after adjusting for
other CHD risk factors, subjects in the highest quartile of alcohol consumption had a significantly lower risk of developing angina pectoris in comparison with nondrinkers. For
myocardial infarction and all CHD events, the risk also decreased from the first to the fourth quartile (p for trend Z
0.02). Conversely, in Northern Ireland, no significant relationship was found between alcohol consumption and the
development of angina pectoris or all CHD events, although
alcohol consumption appeared to reduce the risk of myocardial infarction. There was a suggestion that wine intake had
a protective effect in Northern Ireland (Fig. 1). Excluding
subjects on sick leave or with possible alcohol-related
Results of new analyses
Although drinkers in France and Northern Ireland consumed almost identical amounts, the pattern of consumption was entirely different (13). Mean intake, expressed in
milliliters of alcohol per week, was slightly higher in Northern Ireland than in France (mean G standard deviation:
326 G 333 vs. 318 G 249 mL/wk; Kruskal-Wallis Z 8.20,
p Z 0.01). In Northern Ireland, beer and spirits were the staple beverages, whereas in France it was predominantly red
wine. In France, 90% of men drank at least one unit of alcohol per week; the proportion in Northern Ireland was 61%.
In addition, while the French cohort drank more or less
evenly throughout the week, in Northern Ireland two thirds
of the consumption took place on Friday and Saturday
nights. In both countries, current smokers had higher consumption of all types of alcoholic beverages than nonsmokers. In Northern Ireland, subjects who reported some
physical activity consumed significantly less alcohol than
sedentary subjects, whereas higher consumption was associated with dyslipidemia in France. In France total alcoholic
intake, irrespective of type, was inversely associated with socioeconomic status and educational level. The same was
true for total intake in Northern Ireland, although red
wine was positively associated with higher socioeconomic
and educational status.
The weekly patterns of alcohol consumption were investigated. In this analysis, those drinking at least one unit of
alcohol per week amounted to 5663 and 1367 in France
and Northern Ireland, respectively. (Some subjects were excluded from the study: 593 because of preexisting CHD,
1825 because of antihypertensive or cholesterol-lowering
treatment, and an additional 207 subjects in France who
were screened on a Saturday.) After adjustment for age,
Relative risk of all coronary heart disease (CHD) events
according to level of alcohol intake, France and
Northern Ireland
1
Relative Risk
S77
0.5
>4
14
4
.<
41
7
7<
4<
11
22
.<
22
4
.<
11
no
ne
0<
26
5<
.<
.<
26
8<
12
>4
41
44
1
5
8
.<
12
0<
no
ne
0
Alcohol consumption
France
P Value for trend = 0.005
(ml ethanol/week)
N.Ireland
P Value for trend = 0.72
FIGURE 1. Relative risk of all coronary heart disease (CHD) events accordingto level of alcohol intake, France and Northern Ireland.
S78
Evans et al.
ALCOHOL INTAKE IN NORTHERN IRELAND AND FRANCE
AEP Vol. 17, No. 5S
May 2007: S75–S80
Relative risk of all coronary heart disease (CHD) events
according to alcohol intake, excluding subjects on sick
leave or with alcohol-related disease
Relative Risk
1
0.5
France
Alcohol consumption
(ml ethanol/week)
P Value for trend = 0.006
4
>4
1
4
7
7<
.<
41
22
4
11
4<
.<
22
0<
.<
11
ne
no
1
>4
4
5<
.<
44
1
5
26
8<
.<
26
8
12
0<
.<
12
no
ne
0
N.Ireland
P Value for trend = 0.84
FIGURE 2. Relative risk of all coronary heart disease (CHD) events according to alcohol intake, excluding subjects on sick leave or with
alcohol-related disease.
diseases from the non-drinker group, removing the bias of
former drinking (‘‘sick quitters’’), made no difference to
the results (Fig. 2).
DISCUSSION
In the PRIME study we have observed profound differences
in the pattern of consumption, the type of consumption and
the amount consumed at one time, and the degree of cardioprotection afforded. The PRIME finding of higher blood
pressure in men screened on Mondays and Tuesdays (14)
confirms a similar finding from the British Regional Heart
Study in 1991 (15), which did not extend to Northern Ireland. The crucial point is that these effects were absent in
France where the pattern of drinking is different. Could
the effect observed in Northern Ireland have been due to
bias? There is a famous example of how disease detection
can vary as a result of selective participation. In the Medical
Research Council’s study of miners in South Wales in the
early 1950s, cases of pneumoconiosis were picked up on xray early in the survey, whereas cases of tuberculosis were detected at the end (16). The miners had been left to select
when they were screened, and those with pneumoconiosis
came early as they had fair insight into their condition
and were eligible for compensation, whereas those with tuberculosis came late; they suspected they had a disease
which bore stigma and which was ineligible for compensation. Could such factors have been operating in Northern
Ireland? Were subjects with raised blood pressure appearing
early in the week because they wanted it checked? In PRIME
the men were randomly allocated their day of screening and
no differences in several characteristics of the men screened
on different days were detected. Therefore, we have to accept the effect as real and it helps to explain the observation
that heart attacks are most common in Northern Ireland on
Mondays and least common on Fridays.
There are a number of plausible mechanisms for these effects: acute alcohol consumption modulates intracellular sodium and decreases adrenoreceptor-mediated cardiovascular
reactivity; smoking cigarettes while drinking may increase fibrinogen; salted foods consumed while drinking may raise
blood pressure; or the large water load from beer consumed
may lead to a secondary hyperaldosteronism. The apparent
lack of a cardioprotective effect from alcohol consumption
in Northern Ireland in comparison to France is an important
finding (11). Again, could bias have been operating? How
much of the lack of protection could have been caused by
under-reporting, or even over-reporting, because of guilt surrounding alcohol use (see below), or is it that those who
drink a lot on two or three occasions a week simply cannot
accurately recollect how much they have consumed? It is important to remember that in Northern Ireland only 61% of
men drank (as opposed to 90% in France); two thirds of their
intake took place at weekends, yet they drank rather more
per occasion than in France (13). French subjects consuming
alcohol moderately would have a very shrewd idea of how
much they consumed. Klatsky and his colleagues have estimated that the increased odds ratio for hypertension in persons reporting one to two drinks per day appears to be due, in
part, to under-reporting (17).
Could the apparent lack of protection in Northern Ireland have been due to relatively small numbers of events?
This question will have to await the imminent analyses of
the 10-year follow-up that will approximately double the
AEP Vol. 17, No. 5S
May 2007: S75–S80
number of events. This will permit a fuller investigation of
the effects of different patterns of alcohol consumption. Another question relates to the type of alcohol consumed. Naturally enough, men in France, where wine is produced,
consume more wine than men in Northern Ireland. There
was a suggestion in Northern Ireland that wine was protective, but the group deriving the apparent benefit was small
and skewed toward higher socioeconomic status; this group
enjoys better health status and, even after adjustment, residual confounding remains an issue. Wine may exert a protective effect because of the bioflavonoids (18) and estrogen
mimickers (e.g., phytoestrogens) it contains (19).
Whether the beneficial effects of alcohol consumption in
Northern Ireland were annulled by a pattern of drinking
that increases blood pressure, a well-established risk factor
for heart disease, or whether consumption of wine along
with food gives further protection in France by increasing
the absorption of bioflavonoids, is a matter for conjecture.
However, to add a word of caution in interpreting these
data, Ingster and Feinleib (20) have postulated that the increase in salicylates in food could account for the decline in
cardiovascular mortality observed in the United States. Alcohol users notoriously dose themselves with salicylates and
some of the ‘‘protection’’ could be due to this.
DRINKING PATTERNS IN NORTHERN IRELAND
These very different patterns of alcohol consumption are
deep seated. How did the pattern in Northern Ireland become established? In his book The Classic Slum, Roberts
(21) gives an account of social conditions in Salford, in
Northern England around 1900. Roberts records that, ‘‘To
the great mass of manuals the local public house spelled paradise.’’ The caption to a photograph of men about to enter
a bar reads, ‘‘The shortest way out of Manchester.’’ He also
states that many small business employers of labor paid out
their weekly wages there, so the weekend would have got
off to a bad start. The same system operated in Belfast
with stevedores paying the dockers in the bar at the end of
the week (Boyd A, Cassells J, letter/telephone communication, June 2006). This practice was common among Irish
construction workers in London until recently; in addition,
in the 19th century, Irish Navvies were often paid in beer
(22). The payment of a weekly wage also helped to establish
these patterns of drinking. Among the professional classes,
being paid monthly may have induced a different pattern.
The tight grip of religion in Northern Ireland is undoubtedly also exerting an effect on these patterns of alcohol consumption. The Irish Nobel Laureate William Butler Yeats,
a member of the Protestant Ascendancy, wrote of ‘‘the fascination of what’s difficult,’’ in relation to the Protestant
Ethic, which may be translated as ‘‘hard working and
Evans et al.
ALCOHOL INTAKE IN NORTHERN IRELAND AND FRANCE
S79
abstemious, with a tendency to dourness.’’ Protestant fundamentalists go to the intemperate lengths of ‘‘good-living,’’
which means total abstinence, probably denying themselves
some health benefits from drinking temperately. It should be
appreciated, however, that these attitudes and habits exist
on both sides of the religious divide, but there may be
a lesser tolerance to alcohol abuse in Protestant families
(23). Even in less extreme religious groups, the desire to
obey the Commandment, ‘‘to keep the Sabbath day
holy,’’ combined with the strong work ethic that demands
turning up sober and early for work on weekdays, serves to
punctuate the drinking pattern and builds up a thirst for
the weekend. Added to this, the guilt which surrounds alcohol consumption, no doubt fueled by religious beliefs,
could be a cause of underreporting. It is an odd coincidence that the town with the biggest problem with hard
drugs in Northern Ireland is situated in Ian Paisley’s constituency. It seems that fundamentalist doctrines and religious certainties are comforting for some, but create
a need for escapism in others. It may be the case that alcohol is commonly used as a vehicle for escape in Northern Ireland, whereas in France it is more une aide
nourriture. Added to this may be the fact that in France
a small intake of alcohol with food begins at an early
age, whereas in Northern Ireland, in the past, at least, alcohol consumption began relatively late in life, almost exclusively without food, and then tended to be overdone.
The focus on abstention prevents young people from
learning to drink safely, as they are in more permissive alcohol cultures in southern Europe (24). It seems that the
best advice on alcohol intake should be the same as that
on how to feed a horse: ‘‘little and often.’’ These crosscultural differences in alcohol intake will be assessed
further in MORGAM (MONICA, Risk, Genetics,
Archiving, and Monograph), a large European pooling
project (25).
A protective effect of alcohol consumption against CHD
in France was apparent despite the higher smoking and
blood pressure levels associated with increasing levels of
drinking, suggesting that the effect is strong. Moreover,
the fact that a large majority of men are deriving a possible
benefit from alcohol and that in Northern Ireland this effect
is virtually absent, could go some way to explaining the
‘‘French Paradox.’’ By 1992, Ducimetière and Richard
(26) were asking, ‘‘Is there a French Paradox?’’ The authors
believed that the different mortality experienced in France
and in its neighboring countries could be much more readily
explained by the specific alcohol drinking habits of its population than by any other dietary peculiarity. They also concluded, ‘‘The idea that the dietary lipid concept might
represent the spine of CHD etiology, but not its whole
body, has gained ground during the past 10 years and recognition of this proposition means that there is no place for
S80
Evans et al.
ALCOHOL INTAKE IN NORTHERN IRELAND AND FRANCE
AEP Vol. 17, No. 5S
May 2007: S75–S80
any French Paradox.’’ However, perhaps, like the ‘‘Legal
Paradox,’’ the ‘‘French Paradox’’ is one of those libels which
‘‘.may be the more a libel for being true (27).’’
7. Marques-Vidal P, Ducimetiere P, Evans A, Cambou J-P, Arveiler D. Alcohol consumption and myocardial infarction: a case-control study in France
and Northern Ireland. Am J Epidemiol. 1996;143:1089–1093.
The PRIME Study Group and affiliated projects/institutions: Coordinating
Center, INSERM U258, Paris-Villejuif, France (P.D., A.B.), Strasbourg
MONICA Project, Department of Epidemiology and Public Health, Faculty of Medicine, Strasbourg, France (D.A., B.H.); Toulouse MONICA
Project, INSERM U558, Department of Epidemiology, Paul SabatierToulouse Purpan University, Toulouse, France (J.F., J.B.R.); Lille MONICA Project, INSERM U508, Pasteur Institute, Lille, France (P.A.,
M.M.); Department of Epidemiology and Public Health, Queen’s University, Belfast, Northern Ireland (A.E., J.Y., F.K.); Department of Atherosclerosis, INSERM UR545, Lille, France (G.L., J. M. Bard); Laboratory
of Haematology, La Timone Hospital, Marseilles, France (I. JuhanVague); Laboratory of Endocrinology, INSERM U326, Toulouse, France
(B. Perret); Vitamin Research Unit, University of Bern, Bern, Switzerland
(K. F. Gey); Trace Element Laboratory, Department of Medicine, Queen’s
University Belfast, Northern Ireland (Jayne Woodside, Ian Young),
Faculté de Médecine Pitie-Salpetriere, INSERM 525, France (Francois
Cambien); Faculté de Médecine Xavier Bichat, INSERM U695, France
(Frédéric Fumeron); Instituto de Medicina Molecular, Faculdade de
Medicina, Universidade de Lisboa, Portugal (Pedro Marques-Vidal).
We thank the organizations which allowed the recruitment of the
PRIME subjects : the health screening centers organized by the Social Security of Lille (Institut Pasteur), Strasbourg, Toulouse and Tourcoing;
Occupational Medicine Services of Haute-Garonne, of the Urban Community of Strasbourg; the Association Inter-entreprises des Services Médicaux du Travail de Lille et environs; the Comité pour le Développement
de la Médecine du Travail; the Mutuelle Générale des PTT du Bas-Rhin;
the Laboratoire d’Analyses de l’Institut de Chimie Biologique de la Faculté de Médecine de Strasbourg; the Department of Health (NI) and
the Northern Ireland Chest Heart and Stroke Association.
9. Fumeron F, Arveiler D, Evans AE. Alcohol intake modulates the effect of
a polymorphism of the cholesterol ester transfer protein gene on high density lipoprotein and the risk of myocardial infarction. J Clin Invest.
1995;3:1664–1671.
REFERENCES
1. Ducimetiere P, Richard JL, Cambien F, Rakotovao R, Claude JR. Coronary
heart disease in middle-aged Frenchmen: Comparisons between Paris Prospective Study, Seven Countries Study and Pooling Project. Lancet.
1980;1:1346–1350.
2. Richard JL, Cambien F, Ducimetiere P. Particularités epidemiologiques
de la maladie coronarienne en France. Nouv Presse Med. 1981;10:1111–
1114.
3. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas AM,
Pajak A. Myocardial infraction and coronary deaths in the World Health
Organization MONICA Project: registration procedures, event rates, case
fatality rates in 38 populations from 21 countries in 4 continents. Circulation. 1994;90:583–612.
4. Evans AE, Ruidavets JB, McCrum EE, Cambou JP, McClean R, DousteBlazy P, et al. Autres pays, autres coeurs? Dietary patterns, risk factors
and ischaemic heart disease in Belfast and Toulouse. Q J Med.
1995;88:469–477.
5. Renaud S, de Logeril M. Wine, alcohol, platelets and the French Paradox
for coronary heart disease. Lancet. 1992;339:1523–1526.
6. Parra HJ, Arveiler D, Evans AE, Cambou JP, Amouyel P, Bingham A,
et al. A case-control of lipoprotein particles in two populations at contrasting risk for coronary heart disease: the ECTIM Study. Arterioscler
Thromb. 1992;344:1315–1318.
8. Marques-Vidal P, Cambou JP, Nicaud V, Luc G, Evans A, Arveiler D,
et al. Cardiovascular risk factors and alcohol consumption in France and
Northern Ireland. Atherosclerosis. 1995;115:225–232.
10. Yarnell JWG. The PRIME Study: Classical risk factors do now explain the
severalfold differences in risk of coronary heart disease between France and
Northern Ireland. Q J Med. 1998;91:667–676.
11. Marques-Vidal P, Montaye M, Arveiler D, Evans A, Bingham A, Ruidavets J-B, et al. Alcohol consumption and cardiovascular disease: differential effects in France and Northern Ireland. The Prime Study. Eur J
Cardiovasc Prev Rehabil. 2004;11:336–343.
12. Ducimetiere P, Ruidavets J-B, Montaye M, Haas B, Yarnell J. Five year incidence of angina pectoris and other forms of coronary heart disease in
healthy men age 50–59 in France and Northern Ireland: the Prospective
Epidemiological Study of Myocardial Infarction (PRIME) study. Int J Epidemiol. 2001;3:1057–1062.
13. Marques-Vidal P, Arveiler D, Evans A, Montaye M, Bingham A, Ruidavets JB, et al. Patterns of alcohol consumption in middle-aged men from
France and Northern Ireland. The PRIME Study. Eur J Clin Nutr.
2000;54:321–328.
14. Marques-Vidal P, Arveiler D, Evans A, Amouyel P, Ferriere J, Ducimetiere
P. Different alcohol drinking and blood pressure relationships in France
and Northern Ireland: The PRIME Study. Hypertension. 2001;38:1361–
1366.
15. Wannamethee G, Sharper AG. Alcohol intake and variations in blood
pressure by day of examination. J Hum Hypertens. 1991;5:59–67.
16. Cochrane AL. The detection of pulmonary tuberculosis in a community.
Br Med Bull. 1954;10:91–95.
17. Klatsky AL, Gunderson EP, Harald K, Udaltsova N, Friedman GD. Higher
prevalence of systematic hypertension among moderate alcohol drinkers:
an exploration of the role of underreporting. J Stud Alcohol.
2006;67:421–428.
18. Yarnell JWG, Evans AE. The Mediterranean diet revisited - towards
resolving the (French) paradox. Q J Med. 2000;93:783–785.
19. Sans S, Evans AE. Are cardiovascular disease trends driven by gadflies?
[letter] Int J Epidemiol. 2001;3:624–625.
20. Ingster LM, Feinleib M. Could salicylates in food have contributed to the
decline in cardiovascular disease mortality? A new hypothesis. Am J Public
Health. 1997;87:1554–1557.
21. Roberts R. The classic slum. Food, drink and physic. Manchester: Manchester University Press; 1971.
22. Cowley U. The men who built Britain: A history of the Irish navy. Dublin:
Wolfhound Press; 2004.
23. Walls P. The health of Irish descended Catholics in Glasgow: A qualitative study of the links between health risk and religious and ethnic identities. PhD thesis. Glasgow: University of Glasgow, 1996.
24. Tilki M. The social contexts of drinking among Irish men in London. Drug
Educ Prev Pol. 2006;13:247–261.
25. Evans A, Salomaa V, Kulathinal S, Asplund K, Cambien F, Ferrario M,
et al. MORGAM (an international pooling of cardiovascular cohorts).
Int J Epidemiol. 2005;34:21–27.
26. Ducimetiere P, Richard JL. Dietary lipids and coronary heart disease: is
there a French Paradox? Nutr Metab Cardiovasc Dis. 1992;2:195–201.
27. The Shorter Oxford English Dictionary. Oxford (UK): Oxford University
Press; 1964:1428.
Harm, Benefits, and Net Effects on Mortality of Moderate Drinking
of Alcohol Among Adults in Canada in 2002
JÜRGEN REHM, PHD, JAYADEEP PATRA, PHD, AND BEN TAYLOR, MEPI
Alcohol is an important risk factor contributing to the burden of chronic diseases and injuries, but is also
associated with some health benefits. This study estimates risks and benefits associated with moderate consumption of alcohol in terms of mortality for Canada in 2002 by age and sex. Distribution of exposure was
taken from a Canadian survey and corrected for per capita consumption from production and sales data; risk
relationships were taken from published literature to calculate alcohol-attributable fractions for moderate
consumption. If moderate consumption is based on average volume alone, 866 net deaths in 2002 among
those younger than 70 years of age were due to moderate consumption of alcohol (1.3% of all the deaths in
this age group, consisting of 1653 deaths caused and 787 deaths prevented). When heavy drinking episodes
were excluded, the net effect was beneficial (55 prevented deaths, 0.09% of all deaths); the net burden was
higher for younger ages and the net benefits for older ages.
The net impact of average moderate alcohol consumption on mortality depends on patterns of drinking.
Beneficial net effects are seen only when heavy drinking occasions are excluded. Policies should strive to
reduce the burden of moderate alcohol consumption while preserving the beneficial impacts.
Ann Epidemiol 2007;17:S81–S86. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Mortality, Health Policy, Canada.
INTRODUCTION
Alcohol is accountable for high levels of mortality, morbidity, and social problems (1, 2), with more than 60 causes of
death attributed to alcohol consumption, either positively
or negatively (3–6). Among the detrimental consequences
are different categories of cancer, such as esophageal, liver,
breast, and oropharyngeal cancer, certain gastrointestinal
diseases (7), hypertension, cardiac arrhythmias, cirrhosis
of the liver, pancreatitis, alcohol dependence, motor vehicle
accidents, suicide, and violence (4, 5, 8–10). On the other
hand, when alcohol is consumed in moderation, there is evidence of decreased risk for ischemic heart disease and ischemic stroke (4, 5, 8, 11). In addition, according to current
epidemiological standards, there is some evidence of a protective effect of moderate consumption on diabetes and cholelithiasis or gallstones (12). For other conditions, such as
peripheral vascular disease, cognitive functioning (13, 14),
stress reduction (15), mood elevation (16–18), and other
From the Centre for Addiction and Mental Health (CAMH), Toronto,
Canada (J.R., J.P., B.T.); Public Health Sciences, University of Toronto
(J.R., B.T.); Addiction Research Institute, Zurich, Switzerland (J.R.);
Technische Universitaet, Dresden, Germany (J.R.); and Human Development and Applied Psychology, University of Toronto (J.P.).
Address correspondence to: Jürgen Rehm Professor and Chair, Addiction
Policy, Public Health Sciences, Faculty of Medicine University of Toronto,
Canada Senior Scientist and Co-Head, Section Public Health and Regulatory Policies Centre for Addiction and Mental Health 33 Russell Street,
Room 2035 Toronto, Ontario Canada M5S 2S1. Tel.: þ1 416 535 8501
ext. 6907; fax: þ1 416 260 4156. E-mail: [email protected]
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
subjective psychosocial effects, the evidence is neither consistent nor conclusive (12).
Since there are clearly both beneficial and deleterious effects associated with moderate levels of alcohol consumption, this paper attempts to provide a balance sheet. Thus,
the impact of moderate drinking on chronic diseases and injuries is given by estimating the proportion of premature
chronic disease deaths ‘‘caused’’ or ‘‘prevented’’ by alcohol
in Canada for the year 2002.
METHODS
Identification of Diseases and Meta-analyses
To identify the chronic and nonchronic disease conditions
attributable to alcohol, a comprehensive review of metaanalyses, as well as of the biological literature (4, 5), was carried out. We took a conservative approach and included
only categories where there was an established biological
pathway in addition to an established temporal order, consistent effect, and dose-response relationship (19). The
same standards were used for both deleterious and beneficial
conditions.
Measurement of Exposure
To measure alcohol consumption, we followed the approach
of English et al (8) and used four gender-specific drinking
categories based on average volume of alcohol consumed
(see Table 1 for definition). Moderate drinking was
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.018
S82
Rehm et al.
EFFECTS OF MODERATE DRINKING IN CANADA
AEP Vol. 17, No. 5S
May 2007: S81–S86
Selected Abbreviations and Acronyms
AAF Z alcohol-attributable fraction
CAS Z Canadian Addiction Survey
ICD-10 Z International Classification of Diseases and Health-related Problems, 10th revision
categorized as consumption up to 39 g (inclusive) of pure alcohol per day on average for men, and up to 19 g (inclusive)
of pure alcohol on average per day for women. These specific
drinking categories were used because most of the reviewed
meta-analyses provide results based on these categories.
These drinking categories were entirely based on average
volume of drinking. As a result, it should be noted that in
an extreme case a person can be a moderate consumer on average, even if all the drinking comes from irregular heavy
drinking occasions.
The prevalence data of different levels of current alcohol
consumption between 2003 and 2004 were collected
through the Canadian Addiction Survey (CAS) (20).
This representative survey included a sample size of
13,909 men and women and a 47% individual response
rate. It was decided to use this survey despite the relatively
low response rate because it included the necessary exposure
measures for the intended calculations, a large sample size,
and the closest temporal proximity to available mortality
data. In addition, it has been shown that higher response
rates in surveys do not essentially change the results with respect to the distribution of alcohol consumption (21). Aside
from these reasons, as alcohol consumption has been more
or less stable in Canada over the past years, the difference
of 2 to 3 years between exposure and mortality/morbidity
data is negligible. However, to ensure that the characteristics of the CAS sample are similar to those in the Canadian
population, this sample was weighted to correspond to the
age and sex distribution of the Canadian population.
As patterns of drinking have been shown to influence the
risk of injuries (22) and other causes of death (4, 5), we carried out two separate analyses. The main scenario will give
a balance sheet for on-average moderate drinking, as it
was usually defined (see also above) and without taking patterns of drinking into consideration. The second analysis
was based on moderate drinking under the assumption
that there were no heavy drinking occasions.
Data on Mortality
Mortality data for Canada for the year 2002, with underlying
cause of death coded according to the International Classification of Diseases version 10 (ICD-10), were obtained from
Statistics Canada (23).
Considerations of age. Comparisons of causes of death
and illness in the elderly have some particular methodological problems. First, with increasing age, death certificates
tend to be less valid than for younger populations because
of the multiple causes involved (24, 25). Second, relative
risks for substance-attributable mortality and morbidity indicators tend to be applicable for middle-age deaths only
TABLE 1. Prevalence in percentages of different drinking categories in Canada 2003/2004*
Drinking categories
Overall (all ages)
15–29 y
30–44 y
45–59 y
60–69 y
70–79 y
80þ y
y
Abstention and very light drinking
Female
66.9
Male
40.4
Drinking category I:y moderate drinking
Female
24.9
Male
46.8
Drinking category IIy
Female
6.3
Male
6.5
Drinking category IIIy
Female
1.9
Male
6.3
Total
100
Drinking categories
Abstainer or very light drinker
Drinking category I: moderate drinking
Drinking category II
Drinking category III
59.0
30.2
62.1
35.1
65.3
40.0
68.4
45.0
70.5
48.3
72.6
51.5
34.8
51.6
31.0
48.6
27.1
45.5
23.2
42.4
20.7
40.4
18.1
38.3
3.2
8.7
4.3
8.2
5.5
7.6
6.6
7.1
7.4
6.7
8.2
6.4
3.0
9.4
100
2.6
8.1
100
2.2
6.8
100
1.8
5.5
100
1.5
4.6
100
1.1
3.8
100
Females
Males
z
0–!0.25 g/d
0.25–!20 g/d
20–!40g/d
40þ g/d
0–!0.25 g/d
0.25–!40 g/d
40–!60 g/d
60þ g/d
*Average volume of alcohol consumption was based on a smoothed quantity frequency measure derived from the Canadian Addiction Survey (CAS) and corrected for per capita
consumption (9).
y
The drinking categories were based on the following definitions in grams of pure alcohol per day.
z
One drink (either a can of beer or a small glass of wine or one shot of spirits) contains, on average, 13.6 g of pure alcohol in Canada.
AEP Vol. 17, No. 5S
May 2007: S81–S86
Rehm et al.
EFFECTS OF MODERATE DRINKING IN CANADA
fZ0
Where:
i: exposure category with baseline exposure or no alcohol
iZ0
RR(i): relative risk at exposure level i compared to no
consumption
P(i): prevalence of the ith category of exposure
We used the most comprehensive meta-analysis for each
chronic or nonchronic condition to derive the information
of relative risk for the diverse categories. This was usually
based on Gutjahr et al (28), a series of meta-analyses based
on English et al (8), but the meta-analysis also incorporated
all the literature published since English et al (8). For injuries, which are usually defined directly, we used the relative risk approach (3).
600
Number of deaths
400
200
0
-200
-400
-600
-800
Men
Women
Overall
-1000
15-29 yrs 30-44 yrs 45-59 yrs 60-69 yrs 70-79 yrs 80+ yrs
Age groups
FIGURE 1. Number of net deaths attributable to ‘‘on-average’’
moderate drinking for all age groups.
200
Number of deaths
and there is some indication that relative risks tend to converge to 1 with increasing age (10), thus leading to an overestimation of alcohol-attributable fractions in older age
groups. Therefore, in this analysis, most of the calculations
and the main conclusions are based on data for those 15 to
69 years of age (inclusive). However, the results for the
age group 70 years of age and older are included in the overview (Figs. 1 and 2).
Computing Alcohol-Attributable Fractions due to
Moderate Drinking. The alcohol-attributable fraction
(AAF) is defined as that fraction of the death or disease in
a population that would not have occurred if the effect associated with drinking were absent (26, 27). Since alcohol
may ‘‘cause’’ or ‘‘prevent’’ deaths, the AAF can be positive
or negative.
The formula AAFmod for moderate drinking was derived
from the basic formula of AAF shown below. AAFmod was
then directly calculated from alcohol exposure prevalence
proportions in Canada and the disease- and sex-specific risks
pooled relative risks from meta-analyses.
k
h
i.h X
i
AAFmodZ Pmod RRmod 1
Pi RRi 1 þ 1
S83
0
-200
-400
-600
Men
Women
Overall
-800
-1,000
-1,200
15-29 yrs 30-44 yrs 45-59 yrs 60-69 yrs 70-79 yrs 80+ yrs
Age groups
FIGURE 2. Number of net deaths attributable to moderate
drinking without any occasional heavy drinking occasions of all
age groups: Sensitivity analysis.
For all those conditions with 100% AAF by definition
(e.g., alcoholic cardiomyopathy) and all neuropsychiatric
conditions with the exception of epilepsy, we took a conservative approach by attributing only a fraction (10%) to
moderate drinkers.
AAFmod was calculated separately by sex and age (age
groups: 15–29 years, 30–44, 45–59, 60–69, 70–79, >80).
AAFmod was then applied to the mortality data in order to
estimate the number of alcohol-attributed deaths by age
and sex due to moderate drinking.
RESULTS
Alcohol-Attributable Mortality due to Moderate
Drinking
Table 2 provides estimates of alcohol-attributable deaths,
that is, deaths caused or prevented by moderate alcohol consumption for those younger than 70 years of age. Overall, in
Canada for the year 2002, a net total of 866 alcohol-attributable premature deaths were estimated, accounting for 628
deaths among men and 238 among women. These numbers
were derived by multiplying AAFmod with the number of
deaths for each category, thereby producing numbers with
decimals. As a result, there may be minor rounding errors
after collapsing numbers across different categories.
The 866 alcohol-attributable deaths constituted 1.3%
(866/64,221) of the deaths in Canada among people under
70 years of age. These were net figures, that is, the estimates
of deaths prevented by alcohol have been taken into account. Deaths caused by moderate drinking outweighed
deaths prevented by more than twofold. Overall, there
were 1653 deaths (1233 male and 420 female deaths) attributable to alcohol and 787 deaths (605 male and 182 female
deaths) prevented by alcohol. Both were calculated using
the same epidemiological procedure described above.
Out of all alcohol-attributable deaths in people between
the ages of 15 and 69 years in Canada in 2002 (9), 33.2%
(1653/4985) of deaths caused and 83.1% (787/947) of
S84
Rehm et al.
EFFECTS OF MODERATE DRINKING IN CANADA
AEP Vol. 17, No. 5S
May 2007: S81–S86
TABLE 2. Number of deaths attributable to moderate alcohol drinking (compared with abstainers) by sex, age, and disease category in
Canada, 2002 (9)
15–29 y
Condition
Malignant neoplasms
Oropharyngeal cancer
Esophageal cancer
Liver cancer
Laryngeal cancer
Breast cancer
Other neoplasms
Total
Neuropsychiatric conditions
Alcoholic psychoses
Alcohol dependence syndrome
Alcohol abuse
Unipolar major depression
Degeneration of nervous system due to alcohol
Epilepsy
Alcoholic polyneuropathy
Total
Cardiovascular diseases
Hypertensive disease
Alcoholic cardiomyopathy
Cardiac arrhythmias
Esophageal varices
Total
Digestive diseases
Alcoholic gastritis
Cirrhosis of the liver
Acute and chronic pancreatitis
Chronic pancreatitis (alcohol induced)
Total
Skin diseases
Psoriasis
Total (psoriasis)
Unintentional injuries
Motor vehicle accidents
Poisonings
Falls
Fires
Drownings
Other unintentional injuries
Total
Intentional injuries
Suicide, self-inflicted injuries
Homicide
Other intentional injuries
Total
Ethanol and methanol toxicity, unintentional intent
Finding of alcohol in blood
TOTAL DETRIMENTAL EFFECTS
Beneficial effects
Diabetes mellitus
Ischemic heart disease
Cerebrovascular disease
Cholethiasis
TOTAL BENEFICIAL EFFECTS
TOTAL NET EFFECTS
30–44 y
45–59 y
Male Female Male Female
0
0
0
0
1
1
0
0
1
0
0
0
1
0
0
0
0
0
1
0
2
4
5
3
1
1
14
1
1
1
0
13
1
17
0
0
0
0
0
1
0
1
2
3
2
0
0
2
0
9
0
0
2
0
2
0
0
1
0
1
0
0
0
0
0
Male
25
47
25
13
60–69 y
Female
3
113
5
5
6
2
45
1
65
0
1
1
0
0
1
0
3
4
9
4
0
0
2
0
19
2
1
5
0
8
0
0
1
0
1
0
0
0
0
0
0
5
1
0
7
0
0
0
0
130
16
5
4
9
26
189
Male
29
58
28
21
Overall (all ages)
Female
Male
Female
Total
5
140
6
9
8
3
29
1
57
58
111
56
35
0
9
269
13
16
16
6
86
4
141
70
127
72
41
86
13
409
1
3
1
0
0
2
0
7
3
8
2
0
0
1
0
15
1
2
1
0
0
1
0
4
10
20
8
0
1
6
0
44
2
6
3
0
0
4
0
15
12
26
10
0
1
10
0
59
9
2
15
0
26
1
0
4
0
5
10
1
16
0
27
2
0
4
0
6
20
4
38
1
62
3
1
9
0
13
23
4
47
1
75
0
4
0
0
4
0
27
4
1
32
0
9
1
0
11
0
23
4
0
28
0
7
1
0
8
0
56
9
1
66
0
20
3
0
23
0
76
12
1
89
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
1
1
25
9
1
1
2
5
44
94
29
8
8
9
39
186
20
13
2
2
3
7
47
44
20
18
5
12
37
135
18
10
5
1
2
11
48
9
2
8
1
2
17
39
4
2
2
0
0
5
13
276
67
38
19
31
118
550
68
34
10
5
8
28
151
344
100
48
24
40
146
701
51
18
1
70
0
0
264
13
10
0
23
0
0
70
74
14
0
89
2
0
313
19
7
0
26
1
0
99
58
10
0
68
0
0
394
19
4
0
23
0
0
159
11
2
0
13
0
0
262
3
1
0
3
0
0
92
195
44
1
240
2
0
1233
54
21
0
75
1
0
420
249
66
1
315
3
0
1653
0
2
0
0
L2
262
0
0
1
0
L1
68
1
37
1
0
L39
274
1
6
10
0
L17
81
2
241
5
0
L249
145
5
32
27
0
L64
95
3
303
8
1
L315
53
8
55
36
0
L98
6
6
583
15
1
L605
628
15
93
73
0
L182
238
21
676
88
1
L787
866
AEP Vol. 17, No. 5S
May 2007: S81–S86
Rehm et al.
EFFECTS OF MODERATE DRINKING IN CANADA
S85
TABLE 3. Sensitivity analysis on number of deaths attributable to moderate alcohol drinking excluding occasional heavy drinking
occasions (compared with abstainers) by sex and age in Canada, 2002 (9)
15–29 y
Injuries
Total detrimental effects*
Total beneficial effects
Total net effectsy
30–44 y
45–59 y
60–69 y
Overall (all ages)
Male
Female
Male
Female
Male
Female
Male
Female
Male
Female
Total
39
44
L2
42
10
13
L1
12
41
71
L39
31
11
35
L17
17
31
200
L249
L49
11
92
L64
27
8
203
L315
L112
2
75
L98
L24
118
517
L605
L88
34
214
L182
32
152
731
L787
L55
*Total detrimental effects of all conditions, including injuries.
y
Effects on 100% AAF categories were taken out.
deaths prevented by alcohol were found among people in
the moderate drinking category.
Among premature deaths caused by moderate consumption of alcohol, injuries accounted for 1016 deaths (male:
790; female: 226), malignant neoplasms accounted for 409
deaths (male: 269; female: 141), and digestive diseases accounted for 87 (male: 64; female: 23). There were 701 unintentional injuries (male: 550; female: 151) and 315
intentional injuries (male: 240; female: 75). Among deaths
prevented by alcohol among those under the age of 70, 765
deaths were attributed to cardiovascular disease (male: 598;
female: 167). Ischemic heart disease constituted almost 86%
of the total deaths saved.
Fig. 1 gives an overview of net effects for the entire life
span. A linear trend can be clearly made out: the older the
age group, the higher the net beneficial effects of alcohol.
However, as noted above, there is reason to believe that
in the oldest age groups, the beneficial effect of alcohol is
overestimated.
Fig. 1 helps to reconcile the results of our analysis with
meta-analyses on average volume of alcohol consumption
and all-cause mortality, such as those by English et al (8)
and Rehm et al (29). As many epidemiological studies use
samples with relatively elderly populations (ages 50 and
above at baseline), an overall beneficial net effect would
be expected and occurs in practice.
Sensitivity Analyses
Table 3 provides total numbers of deaths due to injuries
when we separate out the effect of occasional heavy drinking
occasions among moderate drinkers. This would correspond
to a definition of moderate drinking which is not only based
on average consumption, but also on zero heavy drinking occasions. In this sensitivity analysis, we also took out the minimal effects of moderate drinking as calculated in the main
scenario (10%) on 100% AAF conditions by assuming an
attributable fraction of zero.
Moderate drinking caused 152 (male: 118; female: 34) injury deaths without considering heavy drinking occasions,
which account for 15% of the main scenario. Overall, deaths
due to detrimental effects, after parsing out the occasional
heavy drinking, were reduced by 56% (with occasional
heavy drinking: 1653 attributable deaths; without occasional heavy drinking: 731 attributable deaths), which had
an overall beneficial net effect of 55 deaths (0.09% of all
deaths). Fig. 2 shows a precipitous pattern of deaths, that
is, net deaths by age group over the entire age span. (Please
note that the warning on potential problems with reliability
with older ages also applies here.)
DISCUSSION
Overall, the net impact of on-average moderate alcohol
consumption on mortality depends on patterns of drinking.
Only where occasional heavy drinking was not taken into
account did on-average moderate alcohol consumption
have beneficial net effects. One important aspect here is
the injury burden, which in most cases is linked to singleoccasion heavy drinking. Policies should strive to reduce
the burden of moderate alcohol consumption while preserving the beneficial impacts.
The preparation of this paper was facilitated, in part, by
work on two projects: the Second Canadian Study on Social
Costs of Substance Abuse receiving funding from various
Canadian national agencies, under the umbrella of the Canadian Centre on Substance Abuse, and a contract from
Heath Canada to produce an overview of detrimental and
beneficial effects of alcohol.
REFERENCES
1. Room R, Babor T, Rehm J. Alcohol and public health: a review. Lancet.
2005;365:519–530.
2. World Health Organization. The World Health Report 2002: Reducing
risks, promoting healthy life. Geneva: WHO; 2002.
3. Corrao G, Bagnardi V, Zambon A, Arico S. Exploring the dose-response
relationship between alcohol consumption and the risk of several alcohol-related conditions: a meta-analysis. Addiction. 1999;94:1551–1573.
4. Rehm J, Room R, Graham K, Monteiro M, Gmel G, Sempos CT. The
relationship of average volume of alcohol consumption and patterns of
drinking to burden of disease d an overview. Addiction. 2003;98:1209–
1228.
S86
Rehm et al.
EFFECTS OF MODERATE DRINKING IN CANADA
5. Rehm J, Room R, Monteiro M, Gmel G, Graham K, Rehn N, et al. Alcohol use. In: Ezzati M, Lopez AD, Rodgers A, et al., eds. Comparative quantification of health risks. Global and regional burden of disease attributable
to selected major risk factors. Geneva: World Health Organization;
2004:959–1108.
6. Schultz JM, Rice DP, Parker DL, Goodman RA, Stroh G, Chalmers N.
Quantifying the disease impact of alcohol with ARDI software. Public
Health Rep. 1991;106:443–450.
7. Taylor B, Rehm J, Gmel G. Moderate alcohol consumption and gastrointestinal tract. Digestive Dis. 2005;23:170–176.
8. English D, Holman C, Milne E, Winter M, Hulse G, Codde G. The
quantification of drug caused morbidity and mortality in Australia 1995.
Canberra, Australia: Department of Human Services and Health; 1995.
9. Rehm J, Baliunas D, Brochu S, Fischer B, Gnam W, Patra J, et al. The
costs of substance abuse in Canada 2002. Ottawa, Canada: Canadian Centre on Substance Abuse; 2006.
10. Rehm J, Patra J, Popova S. Alcohol-attributable mortality and potential
years of life lost in Canada 2001: implication for prevention and policy.
Addiction. 2006;101:373–384.
AEP Vol. 17, No. 5S
May 2007: S81–S86
16. Castaneda R, Sussman N, Levy O’Malley M, Westreich L. A review of the
effects of moderate alcohol intake on psychiatric and sleep disorders. In:
Galanter M, ed. Recent development in alcoholism: The consequences
of alcoholism. New York: Plenum Press; 1998:197–226.
17. Castaneda R, Sussman N, Westreich L, Levy O’Malley M. A review of the
effects of moderate alcohol intake on the treatment of anxiety and mood
disorders. J Clin Psychiatry. 1996;57:207–212.
18. Pohorecky LA. Stress and alcohol interaction: An update of human
research. Alcohol Clin Exp Res. 1991;15:438–459.
19. Hill AB. The environment and disease: association or causation? Proc R
Soc Med. 1965;58:295–300.
20. Canadian Centre on Substance Abuse. Canadian Addiction Survey 2004:
Microdata eGuide. Ottawa: Canadian Centre on Substance Abuse; 2004.
21. Gmel G, Rehm J. Measuring alcohol consumption. Contemp Drug Probl.
2004:31.
22. Gmel G, Klingemann S, Muller R, Brenner D. Revising preventive paradox: the Swiss case. Addiction. 2001;96:273–284.
23. Statistics Canada. Table 051-0001. CANSIM.
11. Puddey IB, Rakic V, Dimmitt SB, Beilin LJ. Influence of pattern of drinking on cardiovascular disease and cardiovascular risk factors d a review.
Addiction. 1999;94:649–663.
24. Kestefoot HE, Verbeke G. On the relationship between all-cause, cardiovascular, cancer, and residual mortality rates with age. Eur J Cardiovasc
Prev Rehabil. 2005;12:175–181.
12. Ashley MJ, Rehm J, Bondy S, Single E, Rankin J. Beyond ischaemic heart
disease: Are there other health benefits from drinking alcohol? Contemp
Drug Probl. 2000;27:735–777.
25. Minder CE. [Quality of coding of causes of death in Switzerland.] German.
Soz Praventivmed. 1984;29:248–250.
13. Chick J. Can light or moderate drinking benefit mental health? Eur Addict
Res. 1999;5:74–81.
14. Eckardt MJ, File SE, Gessa GL, Grant KA, Guerri C, Hoffman PL, et al.
Effects of moderate alcohol consumption on the central nervous system.
Alcohol Clin Exp Res. 1998;22:998–1040.
15. Baum-Baicker C. The health benefits of moderate alcohol consumption:
A review of the literature. Drug Alcohol Depend. 1985;15:207–227.
26. Walter SD. The estimation and interpretation of attributable risk in health
research. Biometrics. 1976;32:829–849.
27. Walter SD. Prevention of multifactorial disease. Am J Epidemiol.
1980;112:409–416.
28. Gutjahr E, Gmel G, Rehm J. Relation between average alcohol consumption and disease: an overview. Eur Addict Res. 2001;7:117–127.
29. Rehm J, Gutjahr E, Gmel G. Alcohol and all-cause mortality: a pooled
analysis. Contemp Drug Probl. 2001;28:337–361.
Alcohol Use and Mental Health in Developing Countries
VIKRAM PATEL, PHD
This paper provides an overview of mental health and alcohol use in developing countries. The review
shows that mental disorders are common and pose a significant burden on the health of developing nations.
There are close associations between poor mental health and other public health and social development
priorities. Although the overall use of alcohol at the population level is relatively low, with high abstention
rate, drinking patterns among those who do drink are often hazardous. The consumption of alcohol is
heavily gendered and is characterized by a high proportion of hazardous drinking among men. Hazardous
drinkers do not only consume large amounts of alcohol, but also do so in high-risk patterns, such as drinking
alone and bingeing. Hazardous drinking is associated with depressive and anxiety disorders as well as suicide
and domestic violence. The limited evidence base suggests that moderate or casual drinking is not associated with social or health hazards; any likely benefits of moderate drinking for mental health have not been
studied in developing countries. The implications of this evidence base for future research and policy are
discussed.
Ann Epidemiol 2007;17:S87–S92. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Mental Health, Developing Countries, Alcohol Abuse, Comorbidity.
INTRODUCTION
MENTAL HEALTH IN DEVELOPING COUNTRIES
This paper assesses evidence of mental health harm and benefits associated with drinking alcohol in the context of developing countries. Developing countries, categorized as
low or middle income in the World Bank’s classification, account for more than 80% of the world’s population and, unsurprisingly, are home to the majority of individuals living
with mental disorders.
This paper has 4 parts: it provides an overview of mental
health in developing countries, reviews what we know about
drinking patterns and their correlates in these countries,
considers the evidence on the relationship between drinking
and mental health, and suggests implications for policy and
future research.
Research for this article has been derived from two major
sources:
At any given time, about 10% of the adult population globally and about one in three adults attending a primary health
center suffers from a mental disorder. Depression and anxiety (the ‘‘common mental disorders’’) and alcohol and drug
abuse (the ‘‘substance abuse disorders’’) are the most frequent of all mental disorders. Psychotic disorders, such as
schizophrenia and bipolar disorder, although relatively less
common, are profoundly disabling. It is no surprise that
mental disorders figure prominently in the list of leading
global causes of disability (see Table 1) (3, 4). The burden
is the greatest during the most productive years of lifed
young adulthooddwhen about 75% of all mental disorders
seen in adults begin (5). Among people aged 10 to 59 years
in developing countries, four conditions linked to mental
health and alcohol abuse can be found in the 10 leading
causes of death (road traffic accidents, self-inflicted injuries,
violence, and cirrhosis of the liver) (4). If disease burden is
measured through the number of years lived with disability
(YLD), then unipolar depressive disorders is the leading
contributor to disease burden in developing countries;
schizophrenia and alcohol abuse disorders also figure in
the leading 10 causes of YLD (4) (Table 1). Altogether, neuropsychiatric disorders account for 9.1% of disabilityadjusted life years (DALYs) in low-income countries and
17.7% of DALYs in middle-income countries.
The enormous gap between mental health needs and services in developing countries has been addressed in a series
of high-profile international documents, culminating in the
2001 World Health Report (3) and the WHO Mental
Health Atlas (6). Of the 400 million people with mental
A search of MEDLINE and PsycInfo databases, using the
following search terms: alcohol* AND (psycholo* OR psychiatr* OR mental* health OR mental* disorder*) AND
develop* countr*; and alcohol* AND (psycholo* OR psychiatr* OR mental* health OR mental* disorder*) AND
(Africa* OR Asia* OR South America*); for the period
1996 to date
A hand search of books and chapters pertaining to alcohol
use in developing countries, in particular two publications
from the World Health Organization (WHO) (1, 2)
From London School of Hygiene and Tropical Medicine, London, UK.
Address correspondence to: Vikram Patel, London School of Hygiene
and Tropical Medicine, Keppel Street, London WC1E 7HT, UK. E-mail:
[email protected].
*Manuscript provided but not presented at symposium.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.019
S88
Patel
MENTAL HEALTH IN DEVELOPING COUNTRIES
AEP Vol. 17, No. 5S
May 2007: S87–S92
Selected Abbreviations and Acronyms
DALY Z disability-adjusted life years
HIV/AIDS Z Human Immunodeficiency Virus/ Acquired Immunodeficiency Syndrome
ICD-10 Z International Statistical Classification of Diseases and Related
Health Problems, 10th Revision
MDG(s) Z Millennium Development Goal(s)
WHO Z World Health Organization
YLD Z years lived with disability
disorders, most live in developing countries. Meanwhile,
more than 90% of global mental health resources are concentrated in rich countries. In many developing countries,
for example, there is about one psychiatrist for every million
people (6). As a result, the majority of individuals suffering
from mental disorders do not seek professional help, and
families bear the brunt of the untreated morbidity and
disability.
Formal health care in developing countries often takes
the form of primary and traditional medical care. In primary
care, mental disorders typically go undetected, with patients
receiving a cocktail of treatments targeting the various
symptoms of mental disordersdfor example, sleeping pills
for sleep problems and vitamins for fatigue (7). Psychosocial
treatments are rarely provided. Often, only persons with psychotic disorders with disturbed behavior are brought to specialist mental health services (if these are available). In such
establishments, care is heavily biased toward drug therapies.
Mental illness is strongly associated with stigma (8); human
rights violations and institutionalization characterize services for severe mental disorders (9).
At the same time, the profile of mental health in developing countries is increasing. More governments are designing and implementing mental health policies (4); more
donors are supporting mental health–related work; more
TABLE 1. Leading causes of years lived with disability in low
and middle income countries, 2001 (4)
Low- and middle-income countries
1
2
3
4
5
6
7
8
9
10
Cause
YLD (millions
of years)
Percentage of
total YLD
Unipolar depressive disorders
Cataracts
Hearing loss, adult onset
Vision disorders, age-related
Osteoarthritis
Perinatal conditions
Cerebrovascular disease
Schizophrenia
Alcohol use disorders
Protein-energy malnutrition
43.22
28.15
24.61
15.36
13.65
13.52
11.10
10.15
9.81
9.34
9.1
5.9
5.2
3.2
2.9
2.8
2.3
2.1
2.1
2.0
Note: Neuropsychiatric conditions and self-inflicted injuries are presented in bold
font.
Source: Global Burden of Disease (4).
public health professionals and policy-makers are taking
an interest in mental health issues. The pace of reform is
slow, however. With every new challenge to the public
health sector, mental health is relegated to the shadows.
Thus mental health is absent from the United Nations
Millennium Development Goals (MDGs), which set out
a vision for development with a focus on health and education (see Table 2) (10, 11), despite the close link of mental
health with many of the individual MDGs (12). For instance, a major reason why children are not able to either
enroll in schools or complete primary education (MDG 2)
is related to their developmental and mental health (e.g.,
due to various types of learning disabilities). Depression is
one of the most common health problems affecting women
during pregnancy (MDG 5) and after childbirth; depression
during motherhood is associated with low birth weight and
infant failure to thrive, both of which are linked to infant
mortality (MDG 4) (13, 14). There are several areas of confluence between HIV/AIDS (MDG 6) and mental health:
people with mental health problems, particularly alcohol
use disorders, are at greater risk for HIV/AIDS; individuals
with HIV/AIDS are more likely to suffer mental health
problems, and these problems, in turn, can affect their
overall health outcomes (15). In general, virtually all
population-based studies of the risk factors for mental
disordersdparticularly, for depressive and anxiety disorders
and substance abuse disordersdshow higher prevalence
among the poor and marginalized. Mental disorders impoverish people through the costs of health care and as a result
of lost employment opportunities (16). Treatment, thus,
may help people rise out of poverty (MDG 1).
For many years, we lacked evidence that anything could
be done for mental disorders in poor countries. However,
a number of clinical trials have been published recently
from across the developing world, demonstrating the efficacy and cost-effectiveness of locally-feasible treatments
for depression, schizophrenia, and substance abuse. Studies
have demonstrated that community care for schizophrenia
is feasible and leads to superior clinical and disability outcomes (17). Both antidepressant and psychosocial treatments are efficacious for depression (18). Community
initiatives reduce the rates of substance abuse disorders
TABLE 2. The UN Millennium Development Goals
(MGD) (11)
Goal 1
Goal 2
Goal 3
Goal 4
Goal 5
Goal 6
Goal 7
Goal 8
Eradicate extreme poverty and hunger
Achieve universal primary education
Promote gender equality and empower women
Reduce child mortality
Improve maternal health
Combat HIV/AIDS, malaria, and other diseases
Ensure environmental sustainability
Develop a global partnership for development
AEP Vol. 17, No. 5S
May 2007: S87–S92
(19). Perhaps the best examples that management of suffering is possible derive not from trials, but from the remarkable
work of grassroots organizations implementing mental
health interventions (20).
ALCOHOL USE IN DEVELOPING COUNTRIES
The pattern of alcohol use in many developing countries is
closely linked to three broad historical phases. The preindustrial, precolonial phase was characterized by traditional
alcohol consumption, often within the context of specific
culturally sanctioned events (such as religious ceremonies).
Records of the use of traditional alcohols, brewed from locally grown grains, vegetables, or fruits, can be found in
the majority of developing countries. The second phase
was associated with the advent of colonial rule that affected
vast regions of the developing world. The colonial rulers
from Europe brought with them distilled alcohols and
beer, native to their own cultures. The introduction of these
foreign beverages led to new legislations in some countries
regarding taxation and the consumption of alcohol. Traditional alcohols continued to be widely used but remained
outside the tax net and concern of the colonists; European
alcohols were more expensive than local drinks and were
mainly consumed by the colonists and the more affluent sections of the local community. The third historical phase coincides with the postcolonial period. In many developing
countries, it has been characterized by globalization of economic markets and the increasing availability of international brands of distilled alcohols.
Today, distilled alcohols are freely available in many developing countries and include both local and international
brands. Significant variations exist in patterns of drinking
betweendand withindcountries in terms of the proportion
of total drinking accounted for by traditional and distilled
alcohols. As a general rule, however, distilled alcohols and
wine are generally consumed by more urban and affluent
groups, while traditional alcohols (such as sorghum-based
beer and palm wine in Africa) are consumed by more rural
and poorer populations; beer seems to be popular across
the social classes. In addition, in many societies, traditional
beverages are more popular among older drinkers, while distilled malt and barley-based beers are preferred by younger
generations and those with ‘‘more European or American
cultural orientation’’ (1, p. 22). There is considerable variation in the age of initiation of drinking. In many tribal societies, this may take place relatively early, but typically in the
context of a traditional ritual (as opposed to drinking for
pleasure, as in developed countries).
A number of studies address the prevalence of alcohol use
and alcohol use disorders in developing countries (e.g., 1, 2).
However, there is wide variation in the study methods used,
Patel
MENTAL HEALTH IN DEVELOPING COUNTRIES
S89
including how alcohol use and various disorders are defined,
as well as in the sampling strategies. These variations make
comparisons difficult. Typically, surveys target populations
considered to be at high risk (e.g., young people), individuals in primary health care, or those in psychiatric facilities.
The recently completed World Health Surveys will provide
a global picture of alcohol use and its association with socioeconomic and health factors. Meanwhile, a number of findings can be discerned from the existing literature.
First, per capita alcohol consumption is relatively low
in developing countries (see Table 3), with the lowest
rates reported in Africa, Southeast Asia, and the Middle
East. This, in part, attests to the heavy influence of social,
cultural, and economic factors on alcohol use. In comparison with developed countries, a much smaller proportion
of populationdin particular, among womendconsumes
alcohol in developing countries.
Second, alcohol use and alcohol use disorders are more
prevalent in men than in women across regions, with gender
gaps particularly wide in developing countries. It is likely
that both psychosocial and biological etiologies converge
to lead to the greater risk of alcohol use disorders in men.
In many countries, drinking and intoxication among men
are more socially acceptable than among women and may
have important social meanings, such as maintaining friendships or coping with stressful situations. In many cultures,
the role of machismo (as in Latin American cultures), the
importance of male sexuality, is recognized as a key factor
in shaping drinking patterns (e.g., 21). Thus excessive
drinking celebrates male courage, sexual prowess, maturity,
and the ability to take risks, including sexual risks. The association of masculinity and drinking and the use of alcohol as
a means of coping with stress by men are key factors behind
the rising toll of alcohol-related premature mortality in East
European men (22).
Third, alcohol consumption among men often takes the
form of binge drinking, typically outside of the home, with
other men (e.g., 23, 24). A study from Papua New Guinea,
for example, reported that men tend to drink in groups,
TABLE 3. Median per capital consumption of alcohol per adult
15 years of age and over, by who geographical region
Consumption (L)
Region
Mean
AFRO
AMRO
EMRO
EURO
SEARO
WPRO
1.37
6.98
0.30
8.60
1.15
5.54
Countries with
survey data/total number
Percent of
Median
of countries
population covered
0.95
5.74
0.53
8.26
0.99
1.95
28/46
32/35
12/21
49.52
7/11
20/27
76.72
99.96
90.33
99.99
98.38
99.94
Modified from World Health Organization (35), World Health Organization (1).
L Z liters. AFRO-African; AMRO-Americas; EMRO-Eastern Mediterranean;
EURO-Europe; SEARO-South-east Asia; WPRO-Western Pacific.
S90
Patel
MENTAL HEALTH IN DEVELOPING COUNTRIES
usually with a goal to get drunk (2). Drink-diary studies of
male hazardous drinkers in India reported that binge drinking was significantly associated with the use of traditional
and illicit alcohols (25). Beverage choice was related to socioeconomic status, with cost and ease of access being key
determinants. In Mexico and some other Latin American
countries, the legacy of fiesta has been identified as an important influence on male binge drinking (2).
THE RELATIONSHIP BETWEEN ALCOHOL USE
AND MENTAL HEALTH
It must be acknowledged that the evidence base for the association of alcohol use and mental health in developing
countries is weak, particularly from a population perspective. Alcohol dependence and harmful alcohol use are mental disorders in their own right in WHO’s International
Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) (26). For the purpose of this
section, however, this paper will concentrate on their association with other mental disorders.
A number of important recent studies of high-risk populations in developing countries indicate moderate levels of
comorbidity between alcohol abuse and mental illness. A
study in São Paulo, Brazil revealed that the prevalence of
substance misuse among Brazilians with severe mental illness was lower than in developed countries (27). Nevertheless, the very presence of comorbidity worsens the prognosis
and impact of mental disorders. Another Brazilian study reported an association between severity of alcohol dependence and psychiatric symptoms, stressing the importance
of early detection (28). In a study from India, a linear relationship was found between comorbidity of mental disorders
and alcohol and poorer quality of life (29).
Population-based studies reveal a strong association
among hazardous drinking, poor mental health (especially,
depressive and anxiety disorders), and suicide. For example,
two studies on the association of alcohol use and mental
health were carried out on a sample of male industrial
workers in Goa, India (23, 24). A survey of 1013 workers
found that one fifth of all respondents were hazardous
drinkers (24). In general, these men had begun drinking at
an earlier age and had lower educational levels than nonhazardous drinkers. While hazardous drinkers often recognized
that they had a drinking problem, only a small proportion
(14%) had sought help. Hazardous drinkers were significantly more likely to suffer from a common mental disorder
(depressive or anxiety disorders) or to have experienced an
adverse health outcome, such as hospital admission. This
study demonstrated a significant degree of comorbidity between common mental disorders and hazardous drinking,
AEP Vol. 17, No. 5S
May 2007: S87–S92
similar to that reported by researchers in developed
countries (30, 31).
In a subsequent case-control investigation of the impact
of alcohol consumption, two groups of drinkers (hazardous
and nonhazardous or moderate drinkers) were compared
with a group of abstinent men (23). Hazardous drinkers reported a higher number of sick leave days, increased rates
of tobacco use, more frequent injury in the form of fractures,
higher disability scores, more money spent on health, and
poorer mental health than their moderate-drinking or abstinent counterparts. Whereas hazardous drinkers did not report financial difficulties, their spouses were more likely to
attribute financial difficulties to their husbands’ drinking.
The study did not find any trends suggesting adverse impact
of moderate drinking on any of these indicators. As compared with moderate drinkers, hazardous drinkers tended
to drink alone, in bars, and preferred noncommercial alcoholic beverages, which are cheaper and have relatively
high alcohol concentration. These findings suggest that
the adverse association between male alcohol use and mental health in India is concentrated among men who drink
hazardously.
Similar relationships have been reported elsewhere in the
world. In some East European countries, a strong association
between per capita alcohol consumption and suicide rates
has been reported (1). In Chile, in the early 1980s, 38.6%
of suicides were identified as ‘‘alcohol-related’’; a more recent study from Ethiopia revealed a linear relationship
between adolescent suicide attempts and alcohol consumption (1). Several causal explanations have been cited. For
example, alcohol may disinhibit suicidal impulses (and aggression in general), whereas chronic and heavy alcohol
use may lead to a gradual disintegration of the person’s social
life, depression, and, thus, an elevated risk of suicide (1).
Another important consequence of alcohol consumption
in developing countries is related to mental health of individuals living with problem drinkers. Several studies from
developing countries have shown higher levels of family dysfunction and family violence among alcohol-dependent
people and alcohol abusers (23). In studies conducted either
with samples of alcohol-dependent subjects or in clinical situations, spouses (usually, female) of alcoholics were reported to suffer from significant stress levels and various
physical and mental health problems. Studies of individuals
in primary care in India showed significantly higher rates of
depressive and anxiety disorders among women (32); concerns about spousal drinking behavior and the related experience of domestic violence were key risk factors. Higher
rates of common mental disorders in women have been
found in virtually all studies from developing countries
(and, indeed, in developed countries); gender disadvantage,
intimate partner violence, and alcoholism were cited as
major factors to explain this increased risk (33). A recent
AEP Vol. 17, No. 5S
May 2007: S87–S92
community survey of women in India has confirmed
these hypotheses; depressive and anxiety disorders were
stronglydand independentlydassociated with intimate
partner violence and concerns about spouses’ drinking
habits (34).
Patel
MENTAL HEALTH IN DEVELOPING COUNTRIES
S91
relatively few hazardous drinkers seek help because of
stigma, lack of services, and lack of awareness, a concerted
campaign is needed to educate the community about the
health dimensions of hazardous drinking and definitions of
moderate consumption, combined with community-based
interventions. This strategy provides the potential for primary prevention.
IMPLICATIONS
The evidence base on the prevalence of alcohol use and alcohol use disorders in developing countries presents a mixed
picture. Although the overall use of alcohol at the population level is relatively low, with high abstention rate, drinking patterns among those who do drink are often hazardous.
Poor people are more likely to consume cheaper, traditional
alcohols, linked to some adverse health and social consequences. The consumption of alcohol is heavily gendered
and is characterized by a high proportion of hazardous drinking among men. Hazardous drinkers do not only consume
large amounts of alcohol, but also do so in high-risk patterns,
such as drinking alone and bingeing. Hazardous drinking is
strongly associated with common mental disorders and suicide and domestic violence, especially among women. Rates
of alcohol dependence are relatively low in many developing countries, but this condition is associated with high
levels of disability and mortality. The recent report ‘‘Global
Burden of Disease and Risk Factors’’ (5) has highlighted the
contribution of alcohol, as a risk factor, to the global burden
of disease; alcohol use disorders account for nearly 4% of the
attributable-disease burden. The disease outcomes assessed
include depression and suicide. This burden was concentrated in men under the age of 60 years.
On the other hand, there is little evidence on the harmsd
or benefitsdof moderate drinking. Moderate consumption
may be beneficial to individuals, although the ‘‘less tangible
benefits of conviviality, sociability, and in some cases social
solidarity are difficult to quantify’’ (1, p. 46). Little attention
has been paid to these nonquantifiable aspects of moderate
consumption as they relate to mental health and the general
quality of life. Important research questions remain on
health and social impacts of different patterns of alcohol
consumption in developing countries. Such research should
not only focus on the negative impact of hazardous drinking,
but also address the potential benefits of moderate consumption. The use of qualitative research methods would be
highly relevant in such inquiries.
The major challenge in terms of alcohol use and mental
health in developing countries is to reduce the rates of hazardous drinking and alcohol dependence in the population.
This may be achieved, for example, by linking alcohol taxation to the level of alcohol content in a given beverage
and by strengthening the enforcement of licensing to sell
drinks with high alcohol concentration. In addition, since
REFERENCES
1. World Health Organization. Global Status Report on Alcohol. Geneva:
WHO; 1999.
2. Riley L, Marshall M. Alcohol and Public Health in 8 Developing Countries. Geneva: World Health Organization; 1999.
3. World Health Organization. The World Health Report 2001: Mental
health: New Understanding, New Hope. Geneva: WHO; 2001.
4. Lopez A, Mathers C, Ezzati M, Jamison D, Murray C. Global Burden of
Disease and Risk Factors. Washington (DC): Oxford University Press
and the World Bank; 2006.
5. Kessler R, Bergland P, Demler O, Jin R, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:
593–602.
6. World Health Organization. Mental Health Atlas. Geneva: WHO; 2005.
7. Linden M, Lecrubier Y, Bellantuono C, Benkert O, Kisely S, Simon G.
The prescribing of psychotropic drugs by primary care physicians: An international collaborative study. J Clin Psychopharmacol. 1999;19:132–
140.
8. Jamison KR. The many stigmas of mental illness. Lancet. 2006;367:533–
534.
9. National Human Rights Commission. Quality Assurance in Mental
Health. New Delhi: NHRC; 1999.
10. Sachs JD, McArthur JW. The Millennium Project: A plan for meeting the
Millennium Development Goals. Lancet. 2005;365:347–353.
11. World Health Organization. WHO and the Millennium Development
Goals. Available at: http://www.who.int/mdg/en/. Accessed March 24,
2005.
12. Miranda JJ, Patel V. Achieving the Millennium Development Goals: Does
mental health play a role? PLoS Med. 2005;2:e291.
13. Patel V, Prince M. Maternal psychological morbidity and low birth weight
in India. Br J Psychiatry. 2006;188:284–285.
14. Patel V, Rahman M, Jacob K, Hughes M. Effect of maternal mental health
on infant growth in low income countries: New evidence from South Asia.
BMJ. 2004;328:820–823.
15. Collins PY, Holman AR, Freeman MC, Patel V. What is the relevance of
mental health to HIV/AIDS care and treatment programs in developing
countries? A systematic review. AIDS. 2006;20:1571–1582.
16. Patel V, Kleinman A. Poverty and common mental disorders in developing
countries. Bull World Health Organ. 2003;81:609–615.
17. Chatterjee S, Patel V, Chatterjee A, Weiss H. Evaluation of a community
based rehabilitation model for chronic schizophrenia in a rural region of
India. Br J Psychiatry. 2003;182:57–62.
18. Patel V, Araya R, Bolton P. Treating depression in the developing world.
Trop Med Int Health. 2004;9:539–541.
19. Wu Z, Detels R, Zhang J, Li V, Li J. Community-based trial to prevent drug
use among youths in Yunnan, China. Am J Public Health. 2002;92:1952–
1957.
20. Patel V, Thara R. Meeting the Mental Health Needs of Developing Countries: NGO Innovations in India. New Delhi: Sage; 2003.
S92
Patel
MENTAL HEALTH IN DEVELOPING COUNTRIES
21. Pyne HH, Claeson M, Correia M. Gender dimensions of alcohol consumption and alcohol-related problems in Latin America and the Caribbean:
World Bank Discussion Paper No. 433. Washington (DC): The World
Bank; 2002.
22. Bromet EJ, Gluzman SF, Paniotto VI, Webb CP, Tintle NL, Zakhozha V,
et al. Epidemiology of psychiatric and alcohol disorders in Ukraine: Findings from the Ukraine World Mental Health survey. Soc Psychiatry
Psychiatr Epidemiol. 2005;40:681–690.
23. Gaunekar G, Patel V, Rane A. The impact and patterns of hazardous
drinking amongst male industrial workers in Goa, India. Soc Psychiatry
Psychiatr Epidemiol. 2005;40:267–275.
24. Chagas Silva M, Gaunekar G, Patel V, Kukalekar DS, Fernandes J. The
prevalence and correlates of hazardous drinking in industrial workers:
A study from Goa, India. Alcohol Alcohol. 2003;38:79–83.
25. Gaunekar G, Patel V, Jacob KS, Vankar G, Mohan D, Rane A, et al.
Drinking patterns of hazardous drinkers: A multicenter study in India.
In: Haworth A, Simpson R, eds. Moonshine Markets: Issues in Unrecorded
Alcohol Beverage Production and Consumption. New York: BrunnerRoutledge; 2004:125–144.
AEP Vol. 17, No. 5S
May 2007: S87–S92
28. Lima AF, Pechansky F, Fleck MP, De Boni R. Association between psychiatric symptoms and severity of alcohol dependence in a sample of Brazilian
men. J Nerv Ment Dis. 2005;193:126–130.
29. Singh J, Mattoo SK, Sharan P, Basu D. Quality of life and its correlates in
patients with dual diagnosis of bipolar affective disorder and substance
dependence. Bipolar Disord. 2005;7:187–191.
30. Hickie IB, Koschera A, Davenport TA, Naismith SL, Scott EM. Comorbidity of common mental disorders and alcohol or other substance misuse
in Australian general practice. Med J Aust. 2001;175(Suppl):S31–36.
31. Lynskey MT. The comorbidity of alcohol dependence and affective disorders: Treatment implications. Drug Alcohol Depend. 1998;52:201–
209.
32. Patel V, Pereira J, Coutinho L, Fernandes R, Fernandes J, Mann A. Poverty, psychological disorder and disability in primary care attenders in
Goa, India. Br J Psychiatry. 1998;172:533–536.
33. Patel V, Araya R, de Lima M, Ludermir A, Todd C. Women, poverty and
common mental disorders in four restructuring societies. Soc Sci Med.
1999;49:1461–1471.
26. World Health Organization. International Statistical Classification of Diseases and Related Health Problems, 10th Revision. 2nd ed. Geneva: World
Health Organization; 2005.
34. Patel V, Kirkwood BR, Pednekar S, Pereira B, Barros P, Fernandes J, et al.
Gender disadvantage and reproductive health risk factors for common
mental disorder in women: A community survey in India. Arch Gen
Psychiatry. 2006;63:404–413.
27. Rossi Menezes P, Ratto LC. Prevalence of substance misuse among individuals with severe mental illness in Sao Paulo. Soc Psychiatry Psychiatr
Epidemiol. 2004;39:212–217.
35. World Health Organization. Global status report on alcohol 2004.
Geneva: World Health Organization.
Panel Discussion III: Moderation, Culture, and Risk
This discussion centered around the role of cultural influences on drinking and on the benefits and harms
associated with moderate drinking. The panel was chaired by Morten Grønbæk and the panelists were
Dwight Heath, Nady el-Guebaly, Alun Evans, Jürgen Rehm, Solomon Rataemane, and Murray A. Mittleman.
Ann Epidemiol 2007;17:S93–S94. Ó 2007 Elsevier Inc. All rights reserved.
DEFINING MODERATE DRINKING
Mittleman opened the session with comments on definitions, stating that ‘‘moderate’’ alcohol consumption can
no longer be based solely on average consumption in grams
per week. Citing the work presented by Evans, he emphasized the importance of drinking patterns in balancing the
health benefits and risks. Evans discussed the high levels
of weekend drinking in Northern Ireland that subsided to almost none on other days, contrasting it with French patterns
where consumption throughout the week is fairly constant.
He emphasized that although the average consumption per
week for these two groups was similar, the health effects
were quite different. The participants agreed that assessments of the benefits and the harms of different drinking
patterns could help provide a definition for ‘‘moderate’’ consumption and would help advance the fields of alcohol
research and policy.
NET EFFECTS OF MODERATE CONSUMPTION
Mittleman and Rimm pointed out that the gravity of outcomes associated with alcohol abuse makes it difficult to balance the potential health benefits of moderate consumption.
Heath added that, indeed, the world view of alcohol is dominated by people advocating alcohol control because of the
risks of abuse.
Ellison asked what the results would be if binge drinkers
were not included in the ‘‘moderate’’ drinking group in the
Global Burden of Disease (GBD) studies. Rehm replied
that the net balance of drinking is positive as long as it
reflects truly ‘‘moderate’’ alcohol consumption, when even
occasional binge drinkers are excluded. Much more needs
to be understood about the many factors that contribute
to the net effect of moderate alcohol consumption.
CULTURAL INFLUENCES
El-Guebaly suggested that while policy must be implemented at the local level, research on alcohol consumption
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
should pay more attention to the global level. Cultural influences affect the perceived relative risks and benefits of
moderate alcohol consumption; furthermore, changes are
occurring in drinking patterns around the world. For example, the presentation by Rataemane described cultural
changes in South Africa that are resulting in increased alcohol abuse, particularly among the country’s youth. ElGuebaly reiterated that these rapidly changing cultures,
which include countries in Africa and the Middle East,
need to establish some cultural protective factors against risky drinking. Heath noted that anthropological studies have
shown that culture is constantly changing and that education is a useful tool for making changes in culture that could
result in beneficial alcohol consumption.
Lionel Tiger stated that in the United Sates binge drinking follows the stages of the life cycle, a phenomenon he
described as ‘‘chrono-drinking.’’ Heavy alcohol consumption is most likely to occur during adolescence and young
adulthood. Tiger suggested that such trends should be considered when defining moderate alcohol consumption.
Rehm was quick to point out that the observed age differences in drinking behavior in the United States is not
seen in other parts of the world (e.g., in Central Europe),
where the heaviest drinking occurs among middle-aged or
older adults. Therefore implementing policies and alcohol
education under the assumption of heavy drinking only
among young people is not necessarily appropriate in all
cases.
Ernesto Tempesta said that in Italy alcohol consumption
is a normative behavior and that this part of Mediterranean
culture results in lower rates of binge drinking. He stated
that the protective social factor in Italy was self-regulation
of drinking by the community itself. He argued that policies
often attempt to standardize a complex cultural issue,
homogenizing the solutions. In his opinion, this approach
is an attempt to provide a simple solution to a complex
problem and has proven unsuccessful.
Tiger stated that even though alcohol is considered
a ‘‘drug’’ by some cultures, it has survived in most societies,
a ‘‘consumer-tested product’’ that continues to be used
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.020
S94
MODERATE, CULTURE, AND RISK
moderately for the benefits it provides. He added that drugs
generally do not last in society if they are dangerous; governments and public health officials are quick to place marked
restrictions on their use. He believed that alcohol, on the
other hand, can be well managed and integrated into
societies.
Evans pointed out that throughout history alcohol has often been a replacement for unclean drinking water, which
has helped establish its place in a number of cultures as a result. In other societies, the choice to consume alcohol with
or without food is also a result of cultural influences. To
define moderate consumption and to find the proper balance
for its net effects of risks and benefits, we must first understand the role of cultural influences.
POPULATION LEVEL STUDIES
Kaye Fillmore questioned some of the methodological steps
used to produce the results in the GBD study, as presented by
Rehm. The use of survey data and the omission of other data,
in her view, cast doubt onto some of the results of that study.
For example, for countries where data were unavailable, the
GBD study used estimates from neighboring countries’ data,
and definitions for cardiovascular diseases varied from country to country.
In response to Fillmore’s comments, Rehm emphasized
that comparability is important for identifying and understanding relationships between risk factors and diseases
across countries. He stated that survey data can often be inaccurate and need to be adjusted to reflect the actual known
levels of alcohol consumption; underestimation can affect
the validity of survey data collected. The problem of adjusting data to achieve comparability is an issue for all country-
AEP Vol. 17, No. 5S
May 2007: S93–S94
level studies. Rehm added that although drinking patterns
are important, data obtained on the pattern of drinking
are not as useful as volume data when performing countrylevel analyses. To make results meaningful, better methods
are needed to ensure comparability even, for example,
among definitions of coronary heart disease. Further development is also needed for estimating unrecorded levels of
consumption, which can be attributable either to noncommercial alcoholic beverages consumed in a country or to
poor data collection. In these instances, comparability
adjustments provide a means for producing more accurate
per capita consumption data.
Rimm suggested that the problem with using data from
country-level studies relates to their use in policy formulation; he believed that policy is often based on misanalyses
of data at the national level. Rehm agreed, but pointed
out that epidemiology and GBD studies are only tools to
be used for determining policy. Rimm remained concerned,
however, that poor advice can be given to the public as a
result of poor data. For instance, the general public is rarely
informed that the frequency of alcohol intake may be just as
important, if not more important, than average consumption.
Mittleman suggested that researchers reassess the assumptions used for modeling current epidemiologic studies;
in his view, this may be more important than the quality and
quantity of the data collected. Fillmore remained concerned
that alcohol research tends to maximize the protective
effects of alcohol, while policy tends to minimize them.
The session concluded with agreement that while research
has provided some large-scale overviews of the risks and
benefits of alcohol consumption, models have not been
developed to understand these phenomena adequately at
the population level.
Panel Discussion IV: Implications for Future Research
This discussion focused on the implications of the available evidence on the benefits and risks of moderate
drinking for future research. The panel was chaired by Eric Rimm and panelists were Francois Booyse, Kaye
Fillmore, Morten Grønbæk, Kenneth Mukamal, Jürgen Rehm, and Goya Wannamethee.
Ann Epidemiol 2007;17:S95–S97. Ó 2007 Elsevier Inc. All rights reserved.
DRINKING PATTERNS
There was overwhelming agreement among the panelists
and discussants that drinking patterns represent a key area
in which additional research with improved methodology
is needed. Grønbæk suggested that distinguishing steady
from nonsteady drinking and an understanding of the impact of binge drinking would be important contributions
to the field of alcohol research. Wannamethee echoed this
point of view and emphasized that additional attention
should be paid to those at the high end of the drinking spectrum and to the potential harmful effects of such drinking.
Given the different health outcomes of different patterns
of drinking, improvements in methodology would have important implications for any recommendations that might
be offered.
Wannamethee stressed the importance of understanding
the underlying biological mechanisms linking drinking patterns to various outcomes. Biomarkers (e.g., lipids) could
provide important insights into these relations. In the
United Kingdom, where binge drinking is on the rise, there
is evidence of increases in hemorrhagic stroke and liver disease. Understanding the underlying mechanisms, in her
view, would provide us with useful guides for conducting
intervention studies. Rimm pointed out that current
approaches for assessing alcohol consumption vary widely:
some emphasizing frequency, while others recording the
habitual number of drinks on the usual number of days, volumes and types of beverage consumed, or consumption on
the weekend. Most participants agreed that improvement
in methodology in assessing drinking pattern is urgently
needed.
Richard Harding concurred with the importance of focusing on pattern of drinking, saying that most recommendations on drinking currently issued to the public are based
only on quantity of alcohol consumed. Teasing out the relative importance of quantity and pattern would have important implications for drinking guidelines. Rehm added that
assessment of drinking patterns should aim not only at the
individual level, but also at the aggregate level. Fillmore
said that without studying the relationship between patterns
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
and outcomes, there could be no consensus on the magnitude of the protective effect of moderate drinking.
CROSS-CULTURAL ASPECTS
OF ALCOHOL CONSUMPTION
The cross-cultural dimensions of drinking patterns were
highlighted as another important research area that requires
attention. Fillmore emphasized the need to conduct more
and better research in developing countries where the characteristics of drinkers might be different from those in the
developed countries of Europe and North America and
where longevity and other health issues were quite different.
A closer look at various populations and the risks for coronary disease was also necessary, including better measures
of abstinence over time.
Yet, as Wannamethee and Fillmore both pointed out,
confounding remains an important issue in alcohol research
and needs to be addressed in relation to drinking patterns.
Cross-problem approaches in prospective studies, Fillmore
suggested, might offer additional information by focusing
on different variables and interactions. Mukamal added
that the effect of alcohol consumption on various outcomes
may also be influenced by other factors; for example, folate
fortification of foods may modify the relationship of alcohol
intake to the risk of breast cancer. Similarly, various external factors may influence different drinking patterns. For example, the impact of social stress on the development of
problematic drinking patterns is poorly understood. All of
these issues may vary among cultures and merit further
investigation.
SOCIAL DIMENSIONS
The participants realized that while drinking patterns are
linked with disease outcomes, they are also associated with
a range of social outcomes. However, few epidemiologic
studies, such as the Global Burden of Disease, have assessed
the effect of alcohol on social outcomes at the aggregate
level. Rehm pointed to the need to quantify social outcomes
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.021
S96
AEP Vol. 17, No. 5S
May 2007: S95–S97
PANEL DISCUSSION
with standardized and comparable instruments and measures. Both social and cultural determinants play important
roles, but they are hard to measure. While a range of social
instruments exists, they are currently not combined into
useful summary measures which, like disability-adjusted
life years (DALYs), for example, can be used to bring together information on various disease states. Insofar as social
measures exist, they have not been cross-culturally validated
and are not easily comparable.
Echoing Rehm’s comments, Rimm suggested that available data and measures should be modified so that they
might be used to add information on the social dimensions,
particularly of benefits. While there is a willingness to use
substitute measures and settle for what is available with regard to negative health outcomes, he pointed out that social
outcomes are generally discounted because of the lack of
instruments that are considered adequate.
ALCOHOL ACROSS THE AGE SPECTRUM
Drinking problems among young people is of particular concern. Grønbæk believed this is an important topic which
needs further research, particularly in light of the dearth
of long-term studies of drinking patterns. However, as Marjana Martinic pointed out, some long-term data do exist, notably the European School Survey Project on Alcohol and
Drugs (ESPAD), which has helped elucidate drinking patterns and trends across a number of European countries.
Wannamethee agreed with the need to pay greater attention to drinking among young people. At the same time, it
should be noted that variations in drinking continue across
the life span. Therefore greater attention is also needed to
the effects of alcohol in the elderly on, for example, the
development of markers of atherosclerosis.
EDUCATION
Another research area, pointed out by Grønbæk, relates to
the role that education can play in addressing drinking problems among young people; a better understanding of educational approaches and their impact is important. Similarly,
more research is also needed on how to target interventions
at high-risk drinkers and to influence them to reduce their
alcohol consumption. Picking up on the theme of education,
Rehm observed that, currently, there is insufficient evidence
to support whether or not educational programs on alcohol
work. Of the projects that are under way, only a few are adequately designed and long-term follow-up of the behavioral
outcomes is generally lacking, making meta-analyses of
long-term and randomized trials impossible.
However, as Martinic pointed out, what might actually
be needed is greater clarity around what measures and
outcomes are being examined when it comes to education.
Is the goal of education the provision of information, raising
awareness, or changing behavior? Should these three outcomes be given equal weight when it comes to assessing effectiveness? Echoing these comments, Dwight Heath stated:
‘‘If we don’t know enough about the impact of education,
can we really say it does not work?’’
From Richard Smallwood’s point of view, the skills of epidemiologists would be usefully directed at assessing educational interventions. More work is needed to determine
which populations should be targeted with various interventions, what these interventions should be, and whether they
work.
BIOLOGICAL MECHANISMS
Summing up the developments that have occurred over the
past decade, Booyse noted that significant strides have been
made in understanding the mechanisms that underlie the
cardioprotective effects of moderate drinking, including
changes in measures of vasorelaxation and in lipoprotein
profile. This is in large part due to a convergence among various studies and improvement of methodologies. However,
he also issued a note of caution. Much of the research has
been conducted on animal models in paradigms that may
not be fully applicable to human adults; laboratory models
may not fully reproduce the effects of drinking in humans.
For example, many of the protective effects of moderate
drinking in humans are the result of long-term exposure. Efforts to understand the underlying molecular mechanisms,
on the other hand, such as the functional expression of
genes, rely on short time-span experiments.
He commented further that gene expression following alcohol administration is studied in animal models. Typically,
a single activation is needed to switch on expression over
a 24- to 36-hour period before basal levels of expression
are once again reached. On the basis of this evidence, one
might conclude that to achieve a steady level of gene expression that confers protection, a drink every day would be
needed. This can be achieved with micromolar levels in
the laboratory setting, but it is difficult to extrapolate to
drinking behavior in humans.
Better animal models are needed, advised Booyse, and
critical evaluation of data to assess whether the systems
used are realistic. Small-scale studies that measure specific
biological end points in humans may therefore at times be
a more reliable approach than animal experiments. Consensus is needed, however, on what these measurable functions
should be.
Mukamal added that much more needs to be learned
about the biological mechanisms underlying the acute effects of alcohol consumption, even at low blood alcohol
concentrations. Much of the impact of alcohol occurs over
AEP Vol. 17, No. 5S
May 2007: S95–S97
the course of hours, not years; this relationship is still poorly
understood.
Klim McPherson added that while the effects of moderate
drinking on coronary heart disease are important, the implications for the potential risks for other conditions, such as
S97
PANEL DISCUSSION
breast cancer, are at least equally so. Models are therefore
also critical in this area of research that would allow the effects of alcohol on various systems to be reconciled. In fact,
there is probably enough evidence already to do so, he
added. What is needed is integration.
Communicating Through Government Agencies
RICHARD HARDING, PHD, AND CREINA S. STOCKLEY, MSC, MBA
A comparison of worldwide recommendations on alcohol consumption reveals wide disparity among countries. This could imply that many of the recommendations do not adequately accommodate the science,
given that the science is equally valid worldwide. Such a view, however, would be an oversimplification
of the problem that those who formulate such guidelines face. The objective of guidelines is to influence
and change behavior among target populations. It follows, therefore, that several factors then become relevant: behavior that is thought to be in need of change, the culture and mindset of the target populations,
and the kind of message that is likely to be effective. There are some tensions between advice intended only
to reduce the prevalence of misuse and that which also seeks to reflect the evidence on the beneficial health
effects of moderate consumption.
Ann Epidemiol 2007;17:S98–S102. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Public Policy, Health Policy, Health Planning Guidelines.
INTRODUCTION
Do governments take into consideration scientific evidence
when establishing recommended maximum levels of alcohol
consumption? If so, to what extent? If the levels are not
based on science, then what does influence them?
A review of guidelines produced by governments worldwide reveals widely differing views of what are regarded as
appropriate, ‘‘safe,’’ or low-risk maximum levels of alcohol
consumption and associated consumption patterns. ‘‘Safe’’
or low-risk consumption is considered to be the amount of
alcohol that an individual can safely consume without significantly increasing the risk of negative health and social
effects (1). This definition can be extended to include the
amount of alcohol that an individual can safely consume
to potentially experience or gain positive health effects in
the longer term.
From a review of the literature, moderate alcohol consumption is generally defined as approximately 20 to 30 g alcohol per day (2–4), a level at which the moderate
consumption of alcohol can reduce the risk of cardiovascular
disease by 20% to 50% (5). The recommended maximum
levels of alcohol consumption for men and women are listed
in Table 1 (daily levels) and Table 2 (weekly levels). Of the
reports, reviews, and analyses that preceded the setting of
these recommended levels of intake, only those undertaken
Consultant, London, UK (R.H.) and from The Australian Wine Research Institute, Glen Osmond, South Australia, Australia (C.S.S.).
Address correspondence to: Creina Stockley, The Australian Wine Research Institute, PO Box 197, Glen Osmond 5064, South Australia, Australia. Tel.: þ61-8-8303-6600; fax: þ61-8-8303-6601. E-mail: Creina.
[email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
by Australia, Canada, the United States and the United
Kingdom (UK) took into account the potential beneficial
health effects of alcohol consumption.
It is not difficult to identify inconsistencies among these
recommendations. In some countries, and even within different regions of the same country, recommended levels
vary, sometimes up to 2- or 3-fold. Some governments
make recommendations for daily intakes, some weekly,
and some both daily and weekly. For example, as shown in
Tables 1 and 2, safe consumption for men lies between 27
and 50 g of alcohol per day, and between 47 and 280 g per
week. Most, but not all, governments make different recommendations for men and women, where a safe level of consumption for women is generally approximately one-half
that considered to be safe for men. Only some governments,
however, also take an individual’s age and body weight into
account, specify levels for individuals with certain medical
conditions, or provide special recommendations for pregnant women. In addition, there is no consistency on the sizes
of standard drinks across countries, which generally reflects
differences in cultures and customs.
Based on scientific evidence, a consistent message could
be expected worldwide. However, such differences are less
surprising when one also considers other factors.
DIFFERENCES IN GOVERNMENT
RECOMMENDATIONS
There are a number of possible reasons why government recommendations for safe alcohol consumption differ, and the
lack of a single international recommendation that is satisfactory for all.
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.022
AEP Vol. 17, No. 5S
May 2007: S98–S102
Harding and Stockley
ALCOHOL AND GOVERNMENT AGENCIES
S99
TABLE 1. Summary of recommended maximum daily levels of intake of alcohol (Reprinted from 38)
Country
Australia
Austria
Canada
Czech Republic
France
Italy
Japan
Netherlands
New Zealand
Poland
Portugal
Romania
Slovenia
Spain
Basque country
Catalonia
Sweden
Switzerland
United Kingdom
USA
Men, g/d
40
24
27.2
24
29
24–36
39.5
29.7
30
20
28–42
20.7–32.5
20
30
70
32–50
20
24
24–32
28
Women, g/d
20
Comment
Four standard drinks a day (two for women) with two alcohol-free days per week
regarded as low risk
16
27.2
16
20
12–24
19.8
20
10
14–28
20.7–32.5
10
30
28
32–50
20
24
16–24
14
Advise not to drink at least 2 days within a week
Should not exceed 60 g/d on special one-time drinking occasion
Up to 5 days a week
Based only on wine consumption
Up to 32.5 g as beer or 20.7 g as wine
Not to exceed 50 g per drinking occasion (30 g for women)
Wine officially considered an integral part of Mediterranean diet
1 g/kg body weight (0.5 g/kg body weight for women)
Recognize that moderate alcohol intake may have certain positive medical benefits
Recognize that moderate drinking for men over 40 and postmenopausal women
confers health benefits
Recognize that moderate drinking may lower risk of coronary heart disease among
men over 45 and women over 55
Science
The scientific evidence relating to both abusive and moderate alcohol consumption is itself not sufficiently consistent
to produce precise recommendations for safe drinking.
There is no clear scientific evidence that uniformly applies
to all population groups. Indeed, the many factors influencing the definition of safe alcohol consumption for a specific
population group include age, body mass index, ethnicity,
family history, mental and physical health, and the use of
medications. Consequently, definitions are likely to vary
among population groups, as well as across countries and
within them individually. Thus most governments use
simple messages and recommendations that apply to the
majority of the general population rather than complex
ones that include recommendations for a number of specific
population groups. However, the recommendations of the
Australian and UK governments have evolved to also include definitions of safe alcohol consumption for specific
population groups in addition to a basic generalized
recommendation.
Culture
The science base for the health consequences of both alcohol abuse and moderate consumption is not the only criterion or factor that is considered by governments when
producing guidelines. Indeed, the purpose of recommendations is not to facilitate debate and discourse about the
science but to facilitate a change in behavior. Thus it is important to take into account the prevailing drinking culture
of a population, because only in that way is it possible to produce a public health message that is likely to be respected
and regarded.
Interestingly, as evidenced by Tables 1 and 2, those countries with a Mediterranean-style diet, lifestyle, and consumption patterns, such as drinking wine with daily meals,
appear to have higher recommended maximum levels for
TABLE 2. Summary of recommended maximum weekly levels
of intake of alcohol (Reprinted from 38)
Country
Men, Women,
g/wk
g/wk
Australia
Canada
Denmark
Finland
Ireland
New Zealand
Poland
South Africa
United Kingdom
280
190
252
165
210
210
100
252
168
140
122
168
110
140
140
50
168
112
USA
196
98
Comment
Recognize that moderate drinking
for men over 40 and postmenopausal
women confers health benefits
Recognize that moderating drinking
may lower risk of coronary heart
disease among men over 45 and
women over 55
S100 Harding and Stockley
AEP Vol. 17, No. 5S
May 2007: S98–S102
ALCOHOL AND GOVERNMENT AGENCIES
alcohol consumption than do other countries, especially
those with a culture of binge beer and spirits drinking (6).
Distinctions among countries’ consumption patterns are disappearing, however, as beverage preferences have begun to
converge globally (6). This can be seen in a trend toward
binge drinking and intoxication among young adults irrespective of country (7).
Economics and Terms of Reference
Although the specific population groups being targeted by
recommendations may differ, generally most governments
are principally concerned with reducing the economic,
health, and social consequences of alcohol misuse per se.
Their recommendations are aimed at the population groups
that are misusing alcohol or drinking cultures that are likely
to lead to misuse. This is also the approach advocated by the
World Health Organization (8, 9). Indeed, of the 2 billion
people that consume alcoholic beverages worldwide, approximately 76.3 million or 3.8 % have alcohol-related
problems due to alcohol abuse, and 1.8 million (0.09%)
are estimated to be likely to die from alcohol-related harms
(10). Alcohol is the fifth highest cause of the global burden
of disease behind childhood and maternal underweight, unsafe sex, hypertension, and tobacco use and is estimated to
cause 20% to 30% of esophageal and liver cancers, cirrhosis
of the liver, and epileptic seizures worldwide (11).
In this context, the positive consequences of moderate
consumption appear much less relevant, and it is not
surprising that they are not usually taken into account in
government recommendations. Indeed, while only four
governments have publicly recognized the potential health
benefits of moderate alcohol consumption in the preamble
to their recommendations, only the Australian and UK
governments have incorporated them into guidelines,
such as the Australian Drinking Guidelines: Health Risks and
Health Benefits (12) and the UK’s Sensible Drinking Guidelines (13), respectively.
The public health consequences of moderate drinking are
much more likely to be taken into account where recommendations are aimed at the whole population and are set
in the context of the diet as a whole, as is the case with
the US Dietary Guidelines for Americans (14). Although
such guidelines may acknowledge that there may be benefits
to health, these benefits are primarily conferred on middleaged and older individuals and need to be adequately
weighed against other risks to health (15–18).
Target Audience
An additional layer of complexity is the intended audience
for particular guidelines. For example, in Australia, recommendations are written for the general population, as well
as for the use of health professionals and policymakers.
The Canadian guidelines, on the other hand, were written
primarily for health promotion within the general population and are specifically intended to assist physicians in providing appropriate advice to patients. Furthermore, the UK
guidelines have offered a review of medical evidence on
alcohol consumption that has helped to develop policies
fostering responsible, nonabusive drinking. In contrast, the
US guidelines serve as authoritative advice for the general
population about how good dietary habits can promote
health and reduce risk for major chronic diseases and also
serve as the basis for federal food and nutrition education
programs.
Ministerial Approval
Finally, in many countries, any government guidelines may
have to be approved or endorsed by government ministers
before they can be implemented. This step adds a further
source of variability as, inevitably, political judgment is involved. In light of these factors, therefore, it is likely, if not
inevitable, that governments produce recommendations
that differ markedly from one another.
POTENTIAL PROBLEMS WITH
RECOMMENDATIONS
It may be argued that in some instances recommendations
intended to change the behavior of those misusing alcohol
or who are at risk of doing so may conflict with those intended to maximize the potential beneficial effects in a population. For example, advice to abstain from drinking on
some days of the week is at odds with the potential beneficial
effects of regular, daily moderate drinking on the cardiovascular system. This consumption pattern has been observed
to prolong any acute and short-term beneficial effects of alcohol and phenolic components on hemostasis (19, 20) and
it also maintains or promotes some long-term beneficial effects, including that on blood pressure (21, 22). The importance of this observation may be eroded, however, by the
observation that in many cultures, people who drink regularly tend to drink too much, and light drinkers tend to
not drink regularly (23).
In addition, advice on the beneficial effects of moderate
consumption is not useful to populations at high risk for alcohol abuse. For example, although moderate consumption
may confer some cardioprotection in the young (24, 25),
this benefit is generally not considered relevant for young
adults, for whom the risk of mortality and morbidity from accidents is far greater (26). This is succinctly stated in the US
Dietary Guidelines for Americans (14) and in the Canadian
Low-Risk Drinking Guidelines (27).
Conversely, recommendations that seek to reduce overall alcohol consumption in a population may also reduce
AEP Vol. 17, No. 5S
May 2007: S98–S102
that of moderate consumers. This effect may eventually be
reflected in an increased incidence of cardiovascular disease
within a population and is likely to have an economic impact and an effect on general health. Indeed, in the developed world, cardiovascular disease is the leading cause of
mortality, accounting for 25% to 50% of all deaths. Its incidence is increasing in developing countries.
It is important, therefore, when formulating recommendations on maximum levels of alcohol consumption to recognize these potential problems and to seek ways of
resolving them.
FUTURE TRENDS
It is likely that nutrition research will generate findings that
will continue to change alcohol’s role as part of a healthy
diet and lifestyle. It is already established that the extent
of the beneficial health effects of moderate alcohol consumption is related to the background diet (20, 28), with
the largest effects of drinking found when alcohol is consumed as part of a diet that is high in saturated fat (29–
32). Another important finding is that beneficial health
effects of moderate drinking are enhanced in diets that are
already high in plant-derived phenolic compounds, such
as a Mediterranean-style diet. The more recent findings on
the potential beneficial effects of moderate alcohol consumption on diabetes mellitus and obesity (33–37) are especially important in light of present trends in body weight of
individuals worldwide. Similar future findings would make
a strong case for the importance of taking the positive outcomes of moderate consumption of alcohol into account
when formulating both recommended maximum levels of
drinking and population dietary guidelines.
REFERENCES
1. Dufour MC. What is moderate drinking? Defining ‘‘drinks’’ and drinking
levels. Alcohol Res Health. 1999;23:5–14.
2. Beilin LJ, Puddey IB, Burke V. Alcohol and hypertension d kill or cure?
J Hum Hypertens. 1991;10:S1–5.
3. Donahue RP, Abbott RD, Reed DM, Yano K. Alcohol and hemorrhagic
stroke: the Honolulu Heart Study. JAMA. 1986;255:2311–2314.
4. Russell M, Cooper ML, Frone MR, Welte JW. Alcohol drinking patterns
and blood pressure. Am J Public Health. 1991;81:452–457.
5. Marmot MG. Evidence based policy or policy based evidence? BMJ.
2004;328:906–907.
6. Bloomfield K, Stockwell T, Gmel G, Rehn N. International comparisons
of alcohol consumption. Alcohol Res Health. 2003;27:95–109.
7. Britton N, Nolte E, White IR, Grønbæk M, Powles J, Cavallo F, et al.
A comparison of the alcohol-attributable mortality in four European countries. Eur J Epidemiol. 2003;18:643–651.
8. World Health Organization. Global Status Report on alcohol 1999. Geneva: WHO; 1999.
Harding and Stockley
ALCOHOL AND GOVERNMENT AGENCIES
S101
9. World Health Organization. Public health problems caused by harmful use
of alcohol. Geneva: WHO; 2005.
10. World Health Organization. Global Status ReportdAlcohol policy. Geneva: WHO; 2004.
11. World Health Organization. The World Health Report 2002dReducing
risks, promoting healthy life. Geneva: WHO; 2002.
12. Australian Alcohol Guidelines: Health risks and benefits. National Health
and Medical Research Council; 2001.
13. UK Department of Health. Sensible drinking. London: Department of
Health; 1995.
14. US Department of Agriculture and Health and Human Services. Nutrition
and your health: Dietary guidelines for Americans. 2000.
15. Klatsky AL. Alcohol, coronary disease, and hypertension. Annu Rev Med.
1996;47:149–160.
16. Klatsky AL, Armstrong MA, Friedman GD. Alcohol and mortality. Ann
Intern Med. 1992;117:646–654.
17. Rehm J, Sempos CT. Alcohol consumption and all-cause mortality. Addiction. 1995;90:471–480.
18. Stampfer MJ, Colditz GA, Willett WC, Speizer FE, Hennekens CH.
A prospective study of moderate alcohol consumption and the risk of
coronary disease and stroke in women. N Engl J Med. 1988;319:267–
273.
19. Hendriks HF, Veenstra J, Velthuis-te Wierik EJ, Scaafsma G, Kluft C.
Effect of moderate dose of alcohol with evening meal on fibrinolytic factors. Br Med J. 1994;308:1003–1006.
20. Renaud SC, Beswick AD, Fehily M, Sharp DS, Elwood PC. Alcohol and
platelet aggregation: the Caerphilly Prospective Heart Disease Study.
Am J Clin Nutr. 1992;55:1012–1017.
21. Klatsky AL. Blood pressure and alcohol intake. In: Laragh JH, Brenner
BM, eds. Hypertension: pathophysiology, diagnosis and management.
New York: Raven Press; 1995:2649–2667.
22. Puddey IB, Beilin LJ, Vandogen R, Rouse IL, Rogers P. Evidence for a
direct effect of alcohol consumption on blood pressure in normotensive
subjects: a randomized controlled trial. Hypertension. 1985;7:707–
713.
23. Rossow I. Successful drinking d or drinking among the successful? Addiction. 2000;95:1584.
24. Power C, Rodgers B, Hope S. U-shaped relation for alcohol consumption
and health in early adulthood and implications for mortality. Lancet.
1998;352:877.
25. Thadhani R, Camargo CA Jr, Stampfer MJ, Curhan GC, Willett WC,
Rimm EB. Prospective study of moderate alcohol consumption and
risk of hypertension in young women. Arch Intern Med. 2002;162:569–
574.
26. Single E, Robson L, Xie X, Rehm J. Morbidity and mortality attributable to
alcohol, tobacco and illicit drug use in Canada. Am J Public Health.
1999;89:385–390.
27. Health Canada. Canada’s Guidelines for Healthy Eating. Ottawa, Canada:
Minister of Public Works and Government Services in Canada; 2002.
28. de Lorgeril M, Salen P. Wine ethanol, platelets, and Mediterranean diet.
Lancet. 1999;353:1067.
29. Cuevas AM, Guasch V, Castillo O, Irribarra V, Mizon C, San Martin A,
et al. A high-fat diet induces and red wine counteracts endothelial dysfunction in human volunteers. Lipids. 2000;35:143–148.
30. Leighton F, Cuevas A, Guasch V, Perez DD, Strobel P, San Martin A,
et al. Plasma polyphenols and antioxidants, oxidative DNA damage and
endothelial function in a diet and wine intervention study in humans.
Drugs Exp Clin Res. 1999;25:133–141.
31. Mansvelt EPG, van Velden DP, Fourie E, Rossouw M, van Rensburg SJ,
Smuts CM. The in vivo antithrombotic effects of moderate and regular
wine consumption on human blood platelets and haemostatic factors.
Bull O I V. 2002;75:660–676.
S102 Harding and Stockley
ALCOHOL AND GOVERNMENT AGENCIES
AEP Vol. 17, No. 5S
May 2007: S98–S102
32. Mezzano D, Leighton F, Strobel P, Martinez C, Marshall G, Cuevas A,
et al. Mediterranean diet, but not red wine, is associated with beneficial
changes in primary haemostasis. Eur J Clin Nutr. 2003;57:439–446.
35. Dixon JB, Dixon ME, O’Brien PE. Reduced plasma homocysteine in obese
red wine consumers: a potential contributor to reduced cardiovascular risk
status. Eur J Clin Nutr. 2002;56:608–614.
33. Avogaro A, Watanabe RB, Gottardo L, de Kreutzenber S, Tiengo A, Pacin
G. Glucose tolerance during moderate alcohol intake: insights on insulin
action from glucose/lactate dynamics. J Clin Endocrinol Metab. 2002;87:
1233–1238.
36. Wannamethee SG, Camargo CAJ, Manson JE, Willett WC, Rimm EB.
Alcohol drinking patterns and risk of type 2 diabetes mellitus among younger women. Arch Intern Med. 2003;163:1329–1336.
34. de Vegt F, Dekker JM, Groeneveld WJ, Nijpels G, Stehouwer CD, Bouter
LM, et al. Moderate alcohol consumption is associated with lower risk for
incident diabetes and mortality: the Hoorn Study. Diabetes Res Clin Pract.
2002;57:53–56.
37. Wannamethee SG, Sharper AG, Perry IJ, Alberti KG. Alcohol consumption and the incidence of type II diabetes. J Epidemiol Community Health.
2002;56:542–548.
38. International Center for Alcohol Policies (ICAP). Report 14: International drinking guidelines. Washington (DC): ICAP; 2003.
Communicating with the Public: Dilemmas of a Chief Medical Officer
RICHARD SMALLWOOD, MD
Chief Medical Officers, or others directly responsible for the health of a population, have difficulty in dealing with alcohol. They must decide where to draw the line to minimize harm, as there is no doubt that alcohol abuse has serious adverse personal and social effects. The prohibition of alcohol, however, would
likely be both unacceptable and unenforceable. In certain countries, unhealthy drinking habits, especially
among the young (as in the United Kingdom) or among older adults (as in Russia), may call for more stringent harm-reduction strategies than in other countries. In Australia, drink driving rates have dropped markedly in recent decades, but excessive drinking by the young is increasing. Recommendations by health
agencies have included increased taxation, training of health professionals, dealing with advertising, and
improved regulation of licensed premises. In a setting where the public continually hears of the adverse effects of alcohol, there is little chance that potentially beneficial effects of moderate drinking will be considered in policy decisions. No society has yet solved the riddle of achieving the ideal balance where the
majority can enjoy the social and health benefits of moderate drinking, while the harm that alcohol causes
is minimized.
Ann Epidemiol 2007;17:S103–S107. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS:
Alcohol Drinking, Health Policy, Risk Reduction.
INTRODUCTION
In most Western societies, alcohol is deeply embedded in
the culture and has been for many centuries. In most Western cultures, the majority of the population consumes alcohol. Most use it sensibly and in ways that they perceive as
enhancing the quality of their daily lives. Some, however,
use alcohol in a way that carries considerable risk, to both
themselves and others, so that the cost to society of the
harm that is caused becomes a major social and political issue (1). Set against this harm, it has become apparent in recent years that alcohol in moderation may confer
considerable health benefit (2), although the evidence for
this is not universally accepted (3). How, then, does government or the community at large decide where to draw the
line in order to minimize harm, given that prohibition of alcohol would in all likelihood be both unacceptable and unenforceable? How should government try to inform the
public about potential benefit and harm as it seeks understanding and support for its policy position?
THE UNITED KINGDOM-A CASE STUDY
Britain ‘‘has reached a point where it is necessary to call time
on runaway alcohol consumption,’’ according to the
Address correspondence to: Richard Smallwood, MD, Faculty of Medicine, Dentistry and Health Sciences; University of Melbourne, Level 6,
Medical Building, Melbourne, Victoria 3010, Australia. E-mail: r.small
[email protected].
Until December 2005, Richard Smallwood was a member of the Medical
Advisory Group to the Australian Association of Brewers.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
Academy of Medical Sciences in its 2004 report entitled
‘‘Calling Time, The Nation’s Drinking as a Health Issue’’
(4). The Academy’s report sought a policy initiative from
government that would reduce alcohol consumption in Britain by one third, which would return it to the level in 1970
(5). The measures advocated included reducing the blood
alcohol limit for drivers from 0.08 to 0.05 g/100 mL, raising
taxes on alcohol, reducing the alcohol allowances of travelers within the European Union, and reviewing the beverage alcohol industry’s advertising practices.
The Academy of Medical Sciences took the view that
measures to control overall levels of alcohol consumption
were needed if any significant reduction in harm was to be
achieved. They had little doubt that there was increasing
harm due to increasing alcohol misuse in the UK. For example, in 2001, a report from the Chief Medical Officer had
stated that deaths from cirrhosis in young men (aged 25–
44 years) had risen by almost an order of magnitude from
49 per 100,000 in 1970 to 470 per 100,000 in 2000 (5).
This coincided with a 50% rise in alcohol consumption
across the country over that time, occasioned at least in
part by a reduction in the price of alcohol (4).
A recent examination of mortality rates from cirrhosis in
Britain between 1960 and 2002 showed steep increases during the 1990s (6). Among men in Scotland, mortality from
cirrhosis doubled, and in England and Wales it increased by
two thirds. Among women throughout the UK, deaths from
cirrhosis rose by nearly 50%. These striking increases were
proportionally much greater than those found in other
Western European countries, and the authors concluded
that they largely reflected increased alcohol consumption,
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.023
S104
Smallwood
DILEMMAS OF ALCOHOL POLICY
Selected Abbreviations and Acronyms
BAC Z blood alcohol concentration
OECD Z Organization for Economic Cooperation and Development
UK Z United Kingdom
although the sharper rise in cirrhosis mortality in Scotland
was not entirely explained by alcohol intake alone.
With these and many other pieces of evidence that harm
was on the rise, with alcohol misuse in England alone costing an estimated £6.4 billion per annum, with those under
16 years of age drinking twice as much as they did 10 years
earlier, what was the government’s response?
The Alcohol Harm Reduction Strategy for England (7)
based its approach on the following:
Education and communication;
Identification and treatment;
Alcohol-related crime and disorder;
Supply and industry responsibility.
The actions proposed were viewed by some as a considerable ‘‘watering down’’ of what had been proposed in the interim report. It could be argued that government had
eschewed the interventions for which there is best evidence
for efficacy, for example, those that directly affect access to
alcohol, such as raising alcohol taxes or restricting hours of
sale and numbers of outlets. The Royal College of Physicians
expressed ‘‘real concerns as to whether [the plans] will be
translated into sufficient action on what is the second commonest cause of preventable death in this country’’ (8). The
Editor of British Medical Journal called the strategy ‘‘the
dampest of squibs’’ (9). The Academy of Medical Sciences’
view that the strong correlation between the overall population consumption of alcohol, levels of heavy drinking, and
levels of harm meant that the critical intervention had to
be controlling the population’s consumption. In fact, the
UK government proceeded to further deregulate the alcohol
sector.
It is of interest to note that the Academy had taken into
account the health benefits of mild to moderate alcohol consumption, but this did not sway them from their principal
goal of controlling overall consumption.
Governments, and in matters of health the Chief Medical Officer, have to consider the scientific evidence as well as
various experts’ interpretations of the evidence. However,
this is just one input to the formulation of policy. Evidence
showing a dose-response relationship between exposure of
the population to alcohol and the degree of resulting harm
might be convincing, but it is not the only factor to be taken
into account. For example, what might seem to be a straightforward public health measure may be contrary to policy for
industry. It might run counter to community views of rights
and responsibilities and could lead to the rejection of
AEP Vol. 17, No. 5S
May 2007: S103–S107
a ‘‘nanny state’’ approach. Costs and benefits need to be
carefully weighed. For government, this includes political
cost. On this occasion, the UK government would appear
to have opted for a series of measures that are reasonable
in themselves, and perhaps necessary, but in the absence
of more direct control of population-wide consumption,
are of uncertain efficacy in reducing overall harm.
THE AUSTRALIAN SCENE
Over 80% of men and nearly 70% of women in Australia regard regular consumption of alcohol by adults as acceptable
(10). Although Australians are sometimes assumed to be
a heavy-drinking people, their average alcohol consumption
(7.3 liters of alcohol per person in 2002) is substantially
lower than that in Britain (9.4 liters per person) (11). Moreover, this represents a decrease in consumption from a peak
of 9.8 liters per person 20 years ago. Therefore there might
not be the same sense of urgency as in the UK among health
authorities to stem a ‘‘rising tide’’ of harm from alcohol
misuse.
Nonetheless, alcohol is accepted as an important contributor to the total burden of disease, second only to tobacco as a preventable cause of ill health and mortality.
The three major categories of harm from alcohol are problems arising from long-term, heavy alcohol use, alcohol
dependence, and acute intoxication (12). It has been
estimated that 4.9% of the total burden of disease is alcohol
related, but after the health benefits of moderate consumption are taken into account, this figure falls to 2.2% (13).
For both men and women, the burden of disease is highest
in the 15- to 24-year-old age group, due principally to road
and other injury. Preventing alcohol-related harm among
young people is an increasing focus, as there has been
a marked rise in the levels of harmful drinking among the
young in the last 20 years. For example, the numbers of
12- to15-year-old boys who drink to a level where they are
at risk for acute harm has risen from 11% to 18% between
1984 and 2002 (14). Among 12- to15-year-old girls, the
numbers have risen from 10% to 24%. Over the same period,
the numbers for16- to17-year-olds rose from 29% to 43% for
boys, and from 30% to 47% for girls. This trend is in the opposite direction from alcohol consumption of the population as a whole, whose levels have fallen, as well as in the
opposite direction from smoking. Although there have
been numerous programs designed to reduce risky drinking
behaviors among the young, some of which seem to have
been effective at least in the short term (13, 15, 16), their
overall impact on youth drinking patterns is not readily
discernible.
It is estimated that the alcohol industry in Australia
spends more than $100 million in promoting its products
AEP Vol. 17, No. 5S
May 2007: S103–S107
each year and $1.2 billion in sports sponsorship (17). Health
authorities remain particularly concerned about the cumulative effect of sophisticated advertising in shaping young
peoples’ ideas about drinking norms. In addition, there is
concern about the increasing availability of ‘‘ready-todrink’’ products, which are seen as being targeted toward
the young. An Alcoholic Beverages Advertising Code is
in place nationally, but the view of many health bodies is
that ongoing scrutiny of alcohol advertising and promotion
will be required to ensure industry compliance with the code
(11).
In 1995, the Royal Australasian College of Physicians
and the Royal Australian and New Zealand College of Psychiatrists produced a policy document entitled Alcohol Policy: Using Evidence for Better Outcomes (18). As might be
inferred from the title, the principal focus of the document
was on lessening harm at a population level. There were
nine recommendations, which covered the following:
Taxation reform to help achieve public health goals
The training of health professionals
Research
Drink driving
Community education and engagement
Alcohol advertising
Enhanced treatment services
Data collection
Improved regulation of licensed premises
Some of the interventions required action from the Colleges themselves, but many recommendations required government involvement and leadership. Given the public
health focus of the documentdpreventing and treating
alcohol-related problemsdit is perhaps not surprising that
no mention was made of the health benefits to individuals
who drink in moderation.
Australia can reasonably claim an excellent record in reducing the harm related to drink driving. In the past 30
years, road deaths per 100,000 among the population have
declined by a factor of 3 (from 45% above the Organization
for Economic Cooperation and Development [OECD] median to 12% below) (18). This is due in part to a reduction
in alcohol-related crashes. Community tolerance of drink
driving and community behaviors have changed over that
time, with numerous public campaigns related to drink driving as part of a broader road safety strategy. Regulatory measures, such as random breath testing, blood alcohol
concentration (BAC) limits, the loss of licenses if limits
are exceeded, and ‘‘zero tolerance’’ for drivers on provisional
licenses (for which there is a mandatory level of 0.00 g/100
mL BAC) has had a discernible impact on crashes related to
alcohol.
Nonetheless, drink driving remains a major cause of morbidity and mortality, particularly among the young. It is still
implicated in one third of all road deaths (18). Drivers with
a BAC of 0.05 g/100 mL, the present legal limit throughout
Smallwood
DILEMMAS OF ALCOHOL POLICY
S105
Australia, are twice as likely to crash as those with a level of
zero; and those at 0.08 g/100 mL are 7 times as likely to crash
(18). In 2003, police in the State of Victoria breath-tested
over 1.4 million drivers and 5600 were found to have
a BAC over 0.05 g/100 mL (11). These statistics have led
some to propose a zero BAC limit for all drivers, an example
of a severe regulatory impost with perhaps only a modest
gain.
It is worth reflecting briefly on the somewhat differing approaches taken by health bodies in Australia to alcohol and
tobacco, the two major preventable causes of morbidity and
mortality. The objective with tobacco, pursued over decades, has been increasingly explicit: to reduce tobacco
use to as close to zero as possible (19). Smoking rates among
adults have now been reduced to well under 20% (12), and
this has been achieved by a series of increasingly restrictive
regulatory measures, which have kept pace with changing
community attitudes. Government knows that there is no
health benefit from smoking and that the benefit for social
cost of reducing tobacco use is substantial. It is therefore prepared to override the protestations of a disgraced industry.
With alcohol, the position is more complex. There are
widely, if not universally, accepted health benefits from alcohol consumed in moderation. The industry is seen as an
important employer and contributor to the economy and
the export market, and its bona fides in promoting responsible drinking are, to a degree, accepted. The majority of the
community accepts alcohol use by adults as a long-standing
part of the Australian culture. All this has to be set against
the harm that alcohol can cause, and it will undoubtedly affect the way in which the government will seek to minimize
harm in its policy formulations.
THE ‘‘PUBLIC HEALTH APPROACH’’
Some public health professionals are unequivocal in their
pursuit of minimizing harm by advocating tighter population-wide control by government of overall alcohol consumption. This ‘‘public health approach’’ is embodied in
the publication, Alcohol: no ordinary commodity. Research
and public policy, (2003) by Babor et al (20) The authors examined various harm-prevention strategies and rated their
effectiveness on the basis of research evidence. In brief,
the interventions that they concluded were most effective
included regulation of alcohol promotion, restriction of
hours or days of sale, restricted numbers of outlets, government monopoly of retail sales, alcohol taxes, various regulatory measures to control drink driving, and brief
interventions for hazardous drinkers. Among the least effective were alcohol education in schools and colleges, public
service messages, warning labels, and voluntary codes of
practice. No weight is given to the evidence for health
S106
Smallwood
DILEMMAS OF ALCOHOL POLICY
benefits of mild-to-moderate drinking or to the utility of
some of the ‘‘softer’’ policy options.
The approach of this group has its adherents and its detractors, the latter presumably seeing the need for policy
to take into account both the benefits and the harms of alcohol to a given society. Scientific evidence on any topic often gives rise to differing interpretations and conclusions,
and alcohol policy is no exception. The science does not
come into a ‘‘green field site’’ where there are no preconceived ideas or strong convictions about what, if anything,
needs to be done. Evidence-based policy not infrequently
gives way to ‘‘policy-based evidence’’ (21), where the view
of the evidence is influenced by preconceived convictions
about actions to be taken, rather than the evidence changing the convictions.
STRIKING THE RIGHT BALANCE
How, then, should Chief Medical Officers strike the right
balance in helping formulate public policy, in assessing
the science, and in fostering a community which is informed
and understands the good and the bad about alcohol, and
can thereby adopt a rational view and rational behaviors?
In all probability, the formula will need to differ in different circumstances and in different countries. For example,
the approach in 21st century Russia would need to be very
different from that in Australia. It may be that the Chief
Medical Officer’s stance would be different in the UK
from that of the Australian Chief Medical Officer. The
UK Chief Medical Officer is confronted by rapidly rising alcohol consumption and harm, which arguably require more
rigorous regulatory control, as advocated by the Academy of
Medical Sciences. In Australia, the responsibility for alcohol policy rests with State and Territory Governments as
well as the Federal Government, which makes problematic
a consistent approach to policy development. There is, however, a National Alcohol Strategy, and the National Health
and Medical Research Council has promulgated national
guidelines that define the levels of alcohol intake that are associated with low, moderate, or high risk of harm in the
short or longer term. These guidelines appear to be widely
known and accepted, although their impact on drinking behaviors is unclear. There appears also to be some understanding of alcohol’s health benefits, as well as its harms,
despite the fact that health authorities, in particular health
promotion bodies, are unlikely to emphasize the benefits
(22). Certainly, it would be difficult to find any suggestion
from health organizations to abstainers that their health
might benefit from consuming alcohol in moderation.
In practice, in Australia, senior health officials in all governments have tended toward the ‘‘public health approach’’
to alcohol, focusing on minimizing harm and being prepared
AEP Vol. 17, No. 5S
May 2007: S103–S107
to advocate known effective measures. Most would recognize an increasing concern with the rising levels of harm
from drinking among the young. They would support more
rigorous regulatory restriction for access to alcohol by the
young, along with effective enforcement, in addition to education and health promotion initiatives. There is also ongoing concern about drink driving. Australia still has work
to do to lessen the potential harms in these areas, and this
is very likely to preoccupy the thinking of health authorities,
rather than any discussion of potential health benefits. In
fact, the health benefits are now being questioned, even in
the lay press (23), reinforcing the emphasis on the prime
need to tackle harm.
I do not believe that any society has yet solved the riddle
of achieving the ideal balance where the social and health
benefits of moderate drinking can be enjoyed by the majority, while the harm that alcohol causes is minimized. All we
can do is continue to work toward that goal.
REFERENCES
1. Collins DJ, Lapsley HM. Counting the costs: estimates of the social costs of
drug abuse in Australia 1998-9. Canberra: Commonwealth Department of
Health and Ageing; 2002.
2. Doll R, Peto R, Hall E. Alcohol and coronary heart disease reduction among
British doctors: confound or causality. Eur Heart J. 1997;18:23–25.
3. Jackson R, Broad J, Connor J, Wells S. Alcohol and ischaemic heart
disease: probably no free lunch. Lancet. 2005;366:1911–1912.
4. Academy of Medical Sciences. Calling Time: the nation’s drinking as
a health issue. London: Academy of Medical Sciences; 2004.
5. Dyer O. Government must take unpopular decisions to reduce alcohol
consumption. BMJ. 2004;328:542.
6. Leon DA, McCambridge J. Liver cirrhosis mortality rates in Britain from
1950 to 2002: an analysis of routine data. Lancet. 2006;367:52–56.
7. Cabinet Office PsSU. Alcohol harm reduction strategy for England. London: Cabinet Office; 2004.
8. Eaton L. UK government announces plan to tackle alcohol. BMJ.
2004;328:659.
9. Smith R. Why are we complacent about alcohol? BMJ. 2004;328,
doi:10.1136/bmj.328.7441.0-g. Accessed March 12, 2007.
10. Australian Institute of Health & Welfare (AIHW). National drug household survey: first results. Canberra: AIHW; 2002.
11. Parliament of Victoria. Inquiry into strategies to reduce harmful alcohol consumption. In: Drugs and Crime Prevention Committee. Melbourne; 2004.
12. Australian Institute of Health & Welfare (AIHW). Australia’s Health
2002. Canberra: AIHW; 2002.
13. Foxcroft D, Ireland D, Lister-Sharp DJ. Long-term primary prevention for
alcohol misuse in young people: a systematic review. Addiction.
2003;98:397–411.
14. Centre for Behavioural Research in Cancer. Victorian Secondary School students’ use of licit and illicit substances in 2002. Cancer Council Victoria; 2004.
15. McBride N, Farrington F, Midford R. Harm minimization in school drug
education: final results of school health and alcohol harm reduction
programme (SHAHRP). Addiction. 2004;99:278–291.
16. Midford R, McBride N. Alcohol education in schools. In: Heather N,
Stockwell T, eds. The essential handbook of treatment and prevention
of alcohol problems. Chichester: John Wiley; 2004:299–319.
AEP Vol. 17, No. 5S
May 2007: S103–S107
Smallwood
DILEMMAS OF ALCOHOL POLICY
S107
17. Community Alcohol Action Network. Alcohol-The Facts; 2004. Available at: http://www.adf.org.au/inside/caan.html. Accessed April 15,
2006.
20. Babor TF, Caetano R, Casswell S, Edwards G, Giesbrecht N, Graham K,
et al. Alcohol: no ordinary commodity-research and public policy. Oxford:
Oxford University Press; 2003.
18. The Royal Australasian College of Physician (RACP) TRAaNZCoPR. Alcohol Policy: using evidence for better outcomes. Sydney: RACP; 2005
Available at: http://www.racp.edu.au/hpu/policy/index.htm Accessed April
15, 2006.
21. Marmot MG. Evidence based policy or policy based evidence? BMJ.
2004;328:906–907.
19. Wakefield M, Hill D. Tobacco control. In: Moodie R, Hulme A, eds.
Hands on health promotion. Melbourne: I.P Communications; 2004:
149–160.
22. Stronach B, Munro G. Minimising harm from alcohol. In: Moodie R,
Hulme A, eds. Hands on health promotion. Melbourne: I.P. Communications; 2004:195–206.
23. Margo J. Singing alcohol’s praises out of tune with reality. The Australian
Financial Review. 2005:67.
The Influence of Print Media on their Readers’ Understanding
of the Benefits of Moderate Drinking*
THOMAS STUTTAFORD, MD
Journalism reflects and reinforces popular opinion. It is from the circulation that the owner’s dividends and
the journalists’ salaries are derived, and journalists are constantly reminded of this. Altering public opinion
on health-related issues occurs through numerous interactions. Survey data suggest that the most effective
mechanism for changing individual perceptions on health is through communication between patients and
their primary care physician; the second most successful means for influencing individuals is through the
print media. Moreover, the media provide the public with information on available solutions for minimizing
the risks of health hazards. It is, therefore, a major responsibility of print media to keep people informed on
various health issues.
There are obstacles in producing the correct message to provide the public regarding alcohol, as there are
both health risks and benefits. Although a quantifiable measure cannot be used to define moderate drinking
for alldas recommendations vary by sex, age, race, and other factorsdprint media should contribute to the
education of the public on the benefits and harms of moderate consumption. This will require nonhomogeneous messages for the public and greater efforts by editors to increase the availability of health-related
stories in the press.
Ann Epidemiol 2007;17:S108–S109. Ó 2007 Elsevier Inc. All rights reserved.
KEY WORDS: Public Health Messages, Public Opinion, Moderate Alcohol Consumption, Alcohol Abuse,
Medical Journalism.
ALCOHOL AND SOCIETY
For many millennia alcoholic drinks have had different roles
in society. They have been regarded as nourishment, as a social lubricant, and for their medicinal powers. At different
times over the last few thousand years one of these roles,
or another, has received greatest emphasis. In classical society both Greeks and Romans prized the medicinal qualities
of alcohol and these were accepted in biblical times; later
both Orthodox Jews and Muslims rejected it (there could
be genetic reasons and variations in the way alcohol is metabolized that may account for this). In Europe, both in
Christian monastic life and the Renaissance, the medical
role was paramount. In the eighteenth, nineteenth and early
twentieth centuries, both medicinal and social uses played
a prominent role in public and personal life. However, its
medical uses were also important.
Very few people would disagree that there is a significant
social and cultural component to alcohol consumption. In
fact, the French wine merchant Joseph Dargent famously
stated, ‘‘No government could survive without champagne
.in the throats of our diplomatic people it is like oil in
Address correspondence to: Thomas Stuttaford, MD, Medical Editor,
The Times, London, UK. E-mail: [email protected].
*Based on an abstract prepared by the author and transcribed notes from
his presentation entitled ‘‘The Influence of Print Media on their Reader’s
Understanding of the Benefits of Moderate Drinking.’’
Dr Stuttaford is Medical Editor, The Times, London, UK.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
the wheels of an engine’’ (1). Moreover, alcohol has other
capacities as it serves a role in celebration, relaxation, and
other activities.
FOCUS ON ALCOHOL ABUSE
In the twentieth century, the adverse pharmacological effects of an excessive alcohol intake became the dominant
aspect of drinking, and hence the one that received maximum press cover. Journalism reflects and reinforces popular
opinion. It is from the circulation that the owner’s dividends
and the journalists’ salaries are derived, and journalists are
constantly reminded of this. The anti-alcohol movement
seems to have owed more to movements from across the
Atlantic than from Europe; this aversion to alcohol may
have had its origins in the Puritan movement, or possibly
from orthodox Jewry.
However the anti-alcohol movement arose, and became
as powerful as it is, it can not have initially been based on
good scientific epidemiological evidence that had analyzed
the statistics on moderate drinking. Whatever the origins
of the movement, it was undoubtedly nurtured by the press.
PROVIDING INFORMATION TO THE PUBLIC
Altering public opinion on health-related issues occurs
through numerous interactions. Thus far, survey data suggest
that the most effective mechanism for changing individuals’
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.024
AEP Vol. 17, No. 5S
May 2007: S108–S109
perceptions on their health is through communication
between patients and their primary care physician or general
practitioner. The second most successful means for influencing individuals is through the print media. Print media,
ahead of radio, television, specialists, and the government,
provides people with the proper information to expand their
knowledge on health issues. Moreover, it provides the public
with information on available solutions for minimizing the
risks of health hazards. It is, therefore, a major responsibility
of print media to keep people informed on various health
issues.
There are, however, topics of discussion in health that are
complex. For instance, alcohol consumption has been a delicate issue of discussion for quite some time. The adverse
risks of alcohol misuse are often difficult to measure against
the benefits of moderate alcohol consumption. Consequently, governments and media outlets, both print and
broadcast, have produced and disseminated messages surrounding the dangers of alcohol consumption. In fact, about
50 years ago medical students were taught that even a single
drop of alcohol was harmful to one’s health.
The public have an insatiable appetite for information
that applies to their own, or their family’s, health. Alcohol
not only does this but it also crosses over into the readers’
social and frequently professional life. Newspaper editors
are always pleased to take stories about alcohol. The readers
don’t expect scientists to agree and realize that however
good the scientists, the weight they place on the different
aspects of a piece of research will inevitably vary according
to their training, interests, and even early upbringing.
JOURNALISTS AND THE MESSAGE ON
ALCOHOL DRINKING
The medical journalists derive their knowledge, if qualified,
from colleagues, and all of us rely on medical journals, the
Internet, medical conferences, lectures, and press conferences. We are greatly influenced by personal contact with
academic specialists working in those disciplines that treat
systems of the body that are affected either beneficially or
adversely by alcohol. The sources are so numerous that we
are influenced by research agencies that follow relevant
Stuttaford
INFLUENCE OF PRINT MEDIA ON DRINKING
S109
advances in knowledge. Inevitably, reporting reflects the
journalist’s, and the newspaper’s, prejudices and the need
for any story to have the potential to boost the circulation.
Presently, scientific data illustrate that there are some
benefits to moderate alcohol consumption; however, there
are some obstacles in producing the correct message to provide the public. One concern is the difficulty of determining
a quantifiable definition of moderate alcohol consumption.
For example, two UK units of alcohol a day is considered
moderate alcohol consumption, yet these two units could
be detrimental to the health of some individuals. From the
alcohol research available, studies show that varying levels
of alcohol consumption can elicit different outcomes
depending on an individual’s age, gender, race, and physical
health. For instance, studies show that people with obesity
are more likely to contract liver disease from alcohol consumption than nonobese individuals. Similarly, individuals
of Indian descent are at a higher risk of contracting liver
disease due to binge drinking.
Regardless of the fact that a quantifiable measure cannot
be used to define moderate drinking for all, print media
should contribute to the education of the public on the benefits and harms of moderate consumption. This will require
nonhomogeneous messages for the public and greater efforts
by editors to increase the availability of less glamorous
health-related stories in the press, especially in the United
Kingdom.
It is difficult to argue against the scientific data showing
that there are cardiovascular health benefits to moderate
alcohol consumption. In addition, new research suggests
that there could be advantages to moderate drinking with
respect to type 2 diabetes. There is a message regarding
the benefits of and harms of moderate drinking, and the
media need to inform the public of it.
REFERENCE
1. Joseph Dargent, French vintner. Quoted in New York Herald Tribune, July
21, 1955. Available at: www.bartleby.com/63/13/6413.html. Accessed
March 13, 2007.
Panel Discussion V: The Message on Moderate Drinking
This discussion focused on the benefits and risks of moderate drinking in terms of communications to the
public, through governmental regulations, official guidelines, the press, and advice to individuals. The
panel was chaired by R. Curtis Ellison, with panelists Richard Harding, Arthur Klatsky, Richard Smallwood, Thomas Stuttaford, and Lionel Tiger.
Ann Epidemiol 2007;17:S110–S111. Ó 2007 Elsevier Inc. All rights reserved.
MESSAGES ON ALCOHOL FOR THE PUBLIC
Stuttaford was asked how you can gauge whether information on alcohol intended for the public was reaching its intended audience. He explained that media outlets, like
other businesses, conduct extensive focus groups and market
research to help them identify what interests the public, as
well as which journalists are influential in providing such information. Medical market research firms, he added, also
have extensive demographic information on the influential
journalists in medicine and their impact on the medical field
and beyond.
Ellison asked if there is any evidence that the messages
being disseminated to the public on potential health benefits of moderate drinking were increasing, or decreasing, alcohol abuse. Kaye Fillmore responded that she did not think
the data are available yet, but she was worried about alcohol
being introduced into traditionally nondrinking cultures
and being advertised as a ‘‘healthy beverage.’’ Leighton reported on their experiences in Chile. Between 1997 and
2003, he explained, a program called Science, Wine and
Health was carried out with the participation of The Catholic University of Chile. This program provided information
on moderate drinking to the public. Leighton and his group
conducted surveys of drinking patterns and attitudes before
and after the program was implemented, so as to be able to
gauge impact. Their findings indicated that providing such
information did not increase consumption in Chile. Rather,
people were drinking less on any given occasion, but doing
so more frequently and with food. In addition, there was
a shift from weekend drinking to daily drinking and, overall,
there was greater awareness of drinking and its effects.
Goya Wannamethee stated that there was evidence that
encouraging moderate drinking raises average alcohol consumption in a country. To provide clarification, McPherson
responded that the evidence showed a strong correlation between average consumption and heavy consumption across
a given population. However, he noted, it cannot be inferred
from these data whose drinking is leading to an increase. An
increase in the average consumption will not necessarily
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
mean increases at both the lower and the upper ends of
the drinking spectrum, and the health implications would
be quite different for these two extremes.
CULTURAL INFLUENCES
The need to tailor recommendations and information to the
specific needs of different groups was also echoed in discussions around cultural differences. As Fillmore noted, there is
still insufficient evidence about drinking patterns and
changes that are occurring in parts of the world where alcohol consumption is traditionally low. How does one provide
messages in countries where only few people drink? Agreeing with this view, Rehm added that just as the message cannot be the same for all people, it also cannot be the same for
all countriesda single message about drinking or moderate
drinking is not a viable conclusion. Although it might be appropriate to offer a moderation message for countries where
the majority of people drink, such recommendations may
not be appropriate where the majority are abstainers.
Enrico Tempesta described results from Italy, where percapita wine consumption over the past 15 years has decreased by 40%. However, the percentage of subjects with
alcohol-related problems has remained stable at about 7%.
He also pointed out the importance of cultural influences,
even within Italy. In areas such as Tuscany and Umbria,
80% of the population consume alcohol, but there are low
rates of alcohol abuse. Certain other areas, where only
60% or less of the population consume alcohol, rates of
excessive drinking are higher. It is therefore difficult to
generalize about the impact on abuse of alcohol-related
messages to the public because of marked differences by
regions and cultures.
Ricardo Brown asked what would be the public health
message if data showed that part of the population would
not benefit at all from drinking? He pointed out many studies, especially from the United States, where subjects from
certain segments of the population (especially ethnic minorities) are underrepresented. Ellison responded that no one
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.025
AEP Vol. 17, No. 5S
May 2007: S110–S111
has made a general recommendation that everyone should
drink For example, the United Kingdom’s Sensible Drinking
guidelines state specifically that those who might consider
moderate drinking for their health are ‘‘middle-aged and
older men and post-menopausal women.’’ Klatsky agreed,
noting that there are certainly many people who should
not consider ‘‘drinking for health,’’ and these include those
with previous alcohol problems or liver disease. Young people, at least for the intermediate term, will probably not benefit from drinking alcohol, he said. Therefore, the message
cannot be the same for all people.
Rehm added that the message also cannot be the same for
each country. The majority of the people in the world are abstainers, he noted; what would be the appropriate message
for those people? Klatsky replied that in countries in which
there is little or no coronary heart disease or other chronic
diseases, there would be little merit in telling people that
they would be better off if they consumed alcohol. Furthermore, he added, there are many behaviors that can be advised for preventing heart disease; alcohol is not at the top
of the list. Stuttaford added that, at least in the United Kingdom, advice about drinking is tailored to the different ethnic
groups that make up what is now a multiracial country.
However, he added, practitioners may not always be aware
of research that points to various ethnic differences. In addition, advice that differs by race can at times be a sensitive
issue for those receiving it and may be taken as criticism
where that was not the intent.
Murray Mittleman stated that one of the challenges for
preventive medicine, and not just for alcohol consumption,
is that the benefits for individuals can rarely be observed. For
example, the heart attack that does not happen, or a case of
diabetes that does not occur, cannot be followed individually. The adverse effects on people, however, are obvious,
as with a car crash after someone has drunk excessively.
Thus, it is always a challenge in presenting risks/benefits
for any type of behavior.
S111
THE MESSAGE ON MODERATE DRINKING
GIVING ADVICE TO INDIVIDUALS
Klatsky described the approach that he has used as a medical
practitioner in giving advice about alcohol consumption.
Such advice depends on the individual’s age, sex, past alcohol use experience, cardiovascular risk, and other factors.
We can apply similar approaches in giving population advice, as well. He added that most people know very well
what the difference is between light to moderate drinking
and binge or excessive drinking. While some patients may
rationalize their heavy drinking because of its purported
health effects, he has yet to find someone who had developed alcohol abuse because of messages about the health
effects of moderate drinking. Medical practitioners, in his
view, have a ‘‘solemn duty’’ to tell the truth about alcohol
consumption, as they understand it, to all of their patients.
GENERAL CONCERNS ABOUT HEALTH ADVICE
Tiger pointed out that the ‘‘medical model’’ has become the
‘‘guilt-piety’’ modeldthe notion that diseases occur because
of a moral failing on the part of the individual. He agreed
with Klatsky that people know when they drink alcohol appropriately or excessively. As for what the government
should tell people, Tiger continued that we have seen the
rise of what he calls ‘‘concernocrats,’’ people who are motivated by well-meaning concern for the welfare of others
and are prepared, therefore, to exert their power over the behavior and disposition of others. ‘‘It is a very tempting thing
for well-meaning people to make statements about other
people’s health and welfare because the population is primed
for quasi-religious advice from the druids, the priests who advise us. Those of us in the comment business need to be extremely careful about what we say, not because people may
drink too much, or too little, but because we may nominate
ourselves as ‘gurus’ of everyone’s destiny and inspect their
behaviors too carefullydand it is none of our business!’’
Alcohol and Pleasure*
LIONEL TIGER, PHD
Curt Ellison asked me to speak on the subject of pleasure because he heard me discuss my book, The Pursuit of Pleasure,
which appeared in l992 (1). I wrote the book in part to emphasize that an evolutionary approach to human behavior
did not necessarily have to focus on the usual array of subjects
associated with the approach in popular and some scientific
circles: aggression, hypersexuality, scarcity, hierarchy, secularism, man’s inhumanity to woman and to man, and the
like. It seemed clear that a species as complex and inventive
as ours had to reveal as many positive (or ‘‘pull’’) factors in its
make-up, as ‘‘push’’ or (deprivation) factors.
There were many books on pain but relatively few on
pleasure as a serious subject. Hence, in my book I paid great
attention to food and drink, which are not only critical to
our survival but are also a rather dazzling exposition of our
skill (where possible) in converting necessity into elegance.
If substances affect behavior, so also does behavior affect
the body, and in part this conference is about the links between the pleasures and abuses of alcohol and the inner
bodily operations associated with health or morbidity.
Some of these internal mechanisms are brilliantly and convincingly described by participants in the meeting.
My concern is with the broad epidemiological connection between a drugdalcoholdwhich has been employed
by people for thousands of years with reasonable balance
and controlled inspection. Furthermore, it is not only a relatively safe substance (especially in the case of wine), but it
clearly produces pleasure and an enormous array of personal,
social, ritual, religious, and similar results, nearly all of
which are subject to informal but effective social control.
That the substance, when used in moderation, also appears
to confer some health benefit is an unexpected boond
although not quite so unexpected if one assumes that humans are careful managers of their lives and environments
and that alcohol has been consumer-tested for centuries.
When we were all asked earlier for a definition of ‘‘the
moderate use of alcohol,’’ I replied that it was when alcohol
was employed to have more fun with one’s friends and family. Clearly, if it is used to mask or deny serious personal and
social issues, then alcohol consumption is not ‘‘moderate,’’
regardless of the amounts used. However, as a factor in dining, celebrating, ritualizing, easing the day’s end, it is certainly widely useddwhere alcohol is useddand, by and
Address correspondence to Lionel Tiger, PhD, Charles Darwin Professor
of Anthropology, Rutgers University, 131 George St, New Brunswick, NJ
08901. E-mail: [email protected].
*Delivered as after-dinner remarks at the symposium.
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
large, with far less ruckus than the dry drugs such as cocaine,
heroin, and the like. Even the common antianxiety and antidepressant drugs that were the largest-selling drugs in the
world for years were withdrawn from ready access because
they, in fact, caused serious problems for long-time users.
But the essential point made at this meeting is that, given
the conditions described in the papers, moderate longterm use of alcohol is, in fact, desirable, even preferable,
especially in the later years of the life cycle.
When I lived in Ghana, it was customary at a party to
pour some gin into the soil for the ancestors, in order to expand the circle of those affected by the festivity to include
the dead. In the demanding environments of modern life,
when it is often necessary to encounter strangers, a unit of
alcohol is very likely to reduce awkwardness and tension
and generate sociability. Enter a party with 100 potentially
threatening strangers; an alcoholic drink will ease that entrance and facilitate relatively agreeable interactions, which
might otherwise require much time and laborious effort to
enjoy.
Humans are cooperative hunter-gatherers, broadly
equipped and eager to share and contribute. In college football movies (which are invariably terrible) there’s always
a young player on the sidelines during a crisis saying,
‘‘Send me in, Coach!’’ An offer of a drinkdtea, coffee,
beer, whiskey, whateverdis a universal statement about
inclusion, at least for a few moments of time.
And usually consumption is moderated by the very group
in which it takes place. In Japan, I pour your sake and you
pour mine. Buying rounds of drinks in a fair manner is another control. A study of the Canadian Army showed that
‘‘drinking level’’ was a robust feature of after-hour activity
dit was assumed everyone would more or less drink the
same amount, both for financial fairness and to enforce informal moderation. Neighborhood bartenders and publicans
may perform the same function; it is possible that ‘‘bingedrinking’’ may be facilitated by the fact that it may often
occur away from the controls of the home environment.
Certainly this is the case with the migrant football fans,
for whom binge drinking is both an expression of solidarity
with their own group and preparation for possible aggression
against strangers.
There are also other occasions that can produce destructive alcohol use, such as initiation ceremonies for young
men into warrior or similar groups. Yet a nontrivial consequence of such ceremonies, which occur from Spokane to
South Africa, is the formation of what is usually a lifetime
bond, with immense consequences for the life cycles of
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.026
AEP Vol. 17, No. 5S
May 2007: S112–S113
individuals. The binge may need to happen only once. And
so long as it occurs under reasonably prudent circumstances,
it may be socially and organizationally cheap at the price.
However, of course, such events can go dangerously awry
and I have on a number of occasions provided legal testimony about the failure of initiating groups such as fraternities to install necessary safeguards during ceremonies that
are part of otherwise normal function. At the same time,
there may be national pathologies with immensely destructive consequence, such as male drunkenness in Russia and
other countries in the region. The result includes a startling
decline in birth rates in Russia, if only because women find
men unreliable and unavailable, in part because of their degraded potential fertility. President Putin has made enhanced reproduction a national goal but has nonetheless
not addressed the role of alcohol in creating conditions for
the problem.
The burden of my argument in The Pursuit of Pleasure is
that in postpuritanical cultures there is bias against pleasure,
which has to be justified, whereas work and duress find odd
favor in public discourse. Even in Japan, the government
embarked on a campaign to convince citizens to work less
and play more. It failed. Perhaps one predictor might have
been that no Japanese publisher would take my book The
Pursuit of Pleasure, although they had published others of
mine. In the book I suggested redefining GDPdgross domestic productdinto a very different GDPdgross domestic
pleasuredas a way of evaluating communities, not by the
things they made but rather by the enjoyment of life they
had.
But back to alcohol. Obviously, it has countless uses in
social activity ranging from the religious, as in Communion,
Hebrew prayers over wine, or the purported derivation of the
word ‘‘wine’’ (de vin) from divine. In societies that use it
widely, it is clearly a sexual facilitatordperhaps too much
so as on college campuses, where cases of assault, rape, degrading behavior, and the like have been linked with alcohol. Since women are smaller and metabolize alcohol
more slowly, they may be ill advised to ‘‘drink level,’’ a point
brilliantly made in the quatrain by the American wit Dorothy Parker (2) who wrote:
I like to have a martini,
Two at the very most At three I’m under the table,
At four, I’m under the host!
Tiger
ALCOHOL AND PLEASURE
S113
Other issues about alcohol abuse may have to do with
time. First of all, persons who consume alcohol in the morning are likely to do so because of how the body works and the
impact is greater than for those who wait for the traditional
‘‘sundowner.’’ The sad drunk drinks early, if not also often.
And there may be a life-cycle featuredwhat might be
called ‘‘chrono-bingeing’’ may refer to the fact that people
may binge excessively in youth but not as they mature.
When epidemiologists parse the link between drink and
health, clearly they have to account for potentially substantial changes in usage during the course of a whole life. The
methodological issues are demanding, if only because the
exemplary result of moderate alcohol use can be principally
assessed accurately only at a one-time event, death.
There is also the confounding factor that when consumption occurs during social occasions, perhaps it is the impact
of socializing that has a medical impact. A recent study of
drinking by monkeys showed that monkeys who were isolated from other monkeys in their cages drank more than
those in groups. It is also clear that drinking alone among
humans poses different potential risks than that mediated
by the group process. I have, by the way, thought that, apart
from pricing, one reason for overeating in fast-food restaurants is precisely that the setting and experience are inadequately satisfying socially so that caloric bulk replaces what
has been traditional ‘‘slow food’’ social interaction for century after century.
Finally, these proceedings record a fine conference and
another one is surely unnecessary because the data are so
strong that the benign connection between moderate alcohol use and health seems beyond sensible critique. Of
course, grimness, such as in the case of Russian males, remains a central medical concern. However, the resolve of
persons with a supernaturally based or temperamental hostility to alcohol has to be countered, as well as the currently
more ominous (and more self-righteous and secular) efforts
of concernocrats to announce what is good for the rest of
us; though who asked them?
REFERENCES
1. Tiger L. The pursuit of pleasure. New York: Little Brown; 1992.
2. Parker D. Quotation #32049. From: Contributed quotations. Available at:
http://www.quotationspage.com/quote/32049.html
Closing Remarks
R. CURTIS ELLISON, MD
What do we know today about the risks and benefits of
moderate alcohol consumption? While all may not agree, I
believe that, for most people for whom alcohol is not contraindicated by health, ethical, or religious reasons, there are
‘‘intelligent’’ ways of consuming alcohol that minimize the
risks while providing health benefits. If consumed in ‘‘moderation’’ (when that term is defined to exclude binge or irresponsible drinking), alcohol can be considered as
a component of a healthy lifestyle (along with not smoking,
avoiding obesity, eating a healthy diet, and getting regular
exercise). As our colleagues at Harvard have demonstrated,
the risk of many of the diseases associated with aging can be
reduced markedly by people following such a lifestyle (1–3).
The goal of this conference has been to get closer to the
truth about moderate drinking. We have learned many ways
in which we can improve our research techniques, especially
ways of focusing more on the pattern of drinking among subjects in our studies. We should no longer define ‘‘moderate
drinking’’ as a daily, weekly, or monthly average amount below a certain number of drinks or grams of alcohol. We must
include in such a definition the drinking pattern, so that
studies across cultures are more comparable. In addition,
while drinking patterns are linked with health outcomes,
they are also linked with a range of social outcomes. From
the perspective of epidemiologic research, social outcomes
are rarely taken into consideration in aggregate level estimates, such as the burden of disease attributable to alcohol.
There is a real need to quantify social outcomes with standardized and comparable instruments and measures.
More research is needed in judging the applicability of
short-term results from animal studies to the human condition, where the potential benefits generally relate to longterm use. Small-scale studies that measure specific biological
end points in humans may therefore at times be a more reliable approach; consensus is needed, however, on what these
measurable functions should be. Assessing the effects of alcohol on endothelial function has been suggested as one
such measure that would provide important data in epidemiological studies.
We were not focusing on policy at this conference. And
we realize that national guidelines depend on much more
than the scientific data, and vary markedly according to
From Boston University School of Medicine.
Address correspondence to: R. Curtis Ellison, Boston University School
of Medicine, 715 Albany St, Boston, MA 02118. E-mail: [email protected].
Ó 2007 Elsevier Inc. All rights reserved.
360 Park Avenue South, New York, NY 10010
the culture, the usual drinking patterns of the people in a particular country, and the behaviors that they are intended to
change. However, we do have an abundant amount of epidemiologic and biological data that we can provide for physicians to use when dealing with individual patients, and such
information should be both scientifically sound and balanced. As has been pointed out, advice for an individual
can be quite different from general advice to a population.
But the data are very clear: individuals without contraindications who decide that they want to consume alcohol can
be told how to do so in the healthiest and least risky manner
dby drinking small amounts on a regular basis.
As for our colleagues who are setting policy and guidelines regarding alcohol consumption, recall that Cole has
pointed out that policy should be based not on paternalism
(‘‘We professionals know what is best’’) but on education
and the assumption that ‘‘informed people make healthful
choices (4).’’ But, Cole adds, ‘‘health education should be
education and not propaganda; it becomes the latter when
only one side of an issue is presented to the public.’’ Some
continue to be overly concerned that we must not share
all the facts with the public, that if they hear that a little alcohol is good for their health, they will immediately think
that a lot must be better! Specifying exactly how many
drinks or grams or units of alcohol can be considered ‘‘moderate’’ can never work for all people. And many at this conference have emphasized that people are not stupid, that
they know the difference between moderate and excessive
drinking. The Greek playwright Euboulos, more than 2000
years ago, provided a good definition. ‘‘For sensible men I
prepare only three kraters: one for health, the second for
love and pleasure, and the third for sleep. After the third
one is drained, wise men go home (5)’’.
We are all concerned about the problems with alcohol
abuse, especially among young people. The skills of epidemiologists would be usefully directed at assessing interventions
intended to decrease such misuse of alcohol. More work is
needed to determine which populations should be targeted
with various interventions, what these interventions should
be, and whether they work.
We have heard the strong moderating effects that culture
can have on problems associated with alcohol abuse, on
making excessive drinking socially unacceptable. Regardless
of what some say, cultures can be changed, as evidenced by
the example of the United States where smoking cigarettes
in most indoor places has become unacceptableda dramatic
cultural change that took place over only a couple of
1047-2797/07/$–see front matter
doi:10.1016/j.annepidem.2007.01.027
AEP Vol. 17, No. 5S
May 2007: S114–S115
decades. Similarly, with regard to drinking, in recent years
many countries have been successful in markedly reducing
the frequency of drunk driving. However, we have not
been successful at all in reducing the heavy drinking and
drunkenness that are becoming common in many of our
cultures, particularly among young people. Many at this
conference have emphasized the need for further, and
better, research on how to deal with such problems of
abusive drinking.
No one is advising that everyone should drink for their
health, and there is little support for encouraging alcohol
consumption for the prevention of coronary heart disease
in areas of the developing world where risk of the disease
is relatively low. However, I trust that the material presented and discussed at this conference will provide others
with up-to-date and balanced information that will allow
us all to judge better the risks and benefits associated with
moderate alcohol consumption.
The Chairman of the Symposium and Guest Editor of this supplement
wishes to acknowledge the tremendous intellectual and logistical support
for the conference and for the publication of this supplement by Marjana
Martinic, PhD, Vice President for Public Health, and her research assistant Yaw Nyame, BSE, of the International Center for Alcohol Policies,
Washington, DC. Further, he acknowledges the valuable input into the
conference and the editing of the supplement by Yuqing Zhang, MD,
ScD, Professor of Medicine, Boston University School of Medicine, Boston, MA.
Ellison
CLOSING REMARKS
S115
We are grateful for the time, effort, and intellectual input of all of the
participants at the symposium. In addition to the first authors of the papers
included in this supplement, we wish to acknowledge other scientists,
listed below, who were present and made important contributions to the
symposium:
Ricardo Brown, PhD, National Institute on Alcohol Abuse & Alcoholism, Bethesda, MD
Monica Gourovitch, PhD, Distilled Spirits Council of the United
States, Washington, DC
Murray A. Mittleman, DrPH, Harvard School of Public Health, Boston,
MA
Solomon Rataemane, MD, University of Limpopo/MEDUNSA, Pretoria, South Africa
Enrico Tempesta, MD, Osservatorio Permanente sui Giovani e l’Alcool,
Roma, Italy
S. Goya Wannamethee, PhD, Royal Free & University Hospital Medical School, London, UK
REFERENCES
1. Rimm EB, Willett WC, Hu FB, Sampson L, Colditz GA, Manson JE, et al.
Folate and vitamin B6 from diet and supplements in relation to risk of
coronary heart disease among women. JAMA. 1998;279:359–364.
2. Stampfer MJ, Hu FB, Manson JE, Rimm EB, Willett WC. Primary prevention of coronary heart disease in women through diet and lifestyle. N Engl J
Med. 2000;343:16–22.
3. Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, et al. Diet,
lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med.
2001;345:790–797.
4. Cole P. The moral bases for public health interventions. Epidemiology.
1995;6:78–83.
5. Available at: www.users.globalnet.co.uk/wloxias/symposium.htm.