Vol. 32, No.

11
November 2008

Alcoholism: Clinical and Experimental Research

Incidence of Cardiovascular and Cerebrovascular Disease

in Danish Men and Women With a Prolonged Heavy
Alcohol Intake
Ulla Arthur Hvidtfeldt, Marie Engholm Frederiksen, Lau Caspar Thygesen,
Mads Kamper-Jrgensen, Ulrik Becker, and Morten Grnbk

Background: Several epidemiological studies have found U- or J-shaped relationships between

alcohol intake and cardiovascular conditions. The inuence of heavy drinking is, however, sparsely studied. The objective of the present study was to examine whether alcohol addicts have
higher incidence rates of cardio- and cerebrovascular diseases than the population in general.
Methods: The cohort comprised 19,185 subjects (15,368 men and 3,817 women) who attended
outpatient clinics for alcohol abusers within the Copenhagen Hospital Corporation (1954 to
1992). Incidence rates were standardized (SIR) according to sex, age and calendar time to compare subjects cardio- and cerebrovascular incidence with that of the general population of Copenhagen.
Results: During the period 1977 to 2001 a total of 9,397 events of cardio- and cerebrovascular
disease were observed. In both men and women, statistically signicant higher incidence rates
than would be expected in a standard population were observed for cardiovascular diseases (e.g.,
ischemic heart diseases, men: SIR = 1.76; 95% CI 1.691.83; women: SIR = 2.44; 95% CI 2.19
2.73) and cerebrovascular diseases (e.g., hemorrhagic stroke, men: SIR = 2.71; 95% CI 2.45
2.99; women: SIR = 2.77; 95% CI 2.183.48).
Conclusions: The study indicates increased risks of cardio- and cerebrovascular diseases in subjects with an excessive alcohol intake.
Key Words: Alcohol, Cardiovascular Diseases, Cerebrovascular Diseases, Incidence.

VAST NUMBER of epidemiological studies among

subsets of normal, light- to moderately drinking populations have found the relationship between alcohol and cardio- and cerebrovascular disease to be U- or J-shaped, i.e., a
moderate intake of alcohol appears to be protective of cardioand cerebrovascular disease compared to abstinence and
heavy drinking (Camacho et al., 1987; Colditz et al., 1985;
Gronbaek et al., 2000; Mukamal et al., 2003; Reynolds et al.,
2003; Rimm et al., 1991; Thun et al., 1997; Truelsen et al.,
1998; Wannamethee and Shaper, 1996). This protective effect
of alcohol has especially been observed for ischemic heart disease, where the lowest risk is seen at an intake of 1 to 2 drinks
per day (Klatsky, 1994; Klatsky et al., 1990; McElduff and
Dobson, 1997; Rimm et al., 1991). However, the inuence of
heavy drinking on cardiovascular conditions is sparsely stud-

From the Centre for Alcohol Research (UAH, MEF, LCT, MG),
National Institute of Public Health, Copenhagen, Denmark; Department of Epidemiology Research (MKJ), Statens Serum Institut,
Copenhagen, Denmark; and Hvidovre University Hospital (UB),
Alcohol Unit, Hvidovre, Denmark.
Received for publication January 3, 2008; accepted June 23, 2008.
Reprint requests: Ulla Arthur Hvidtfeldt, Centre for Alcohol
Research, National Institute of Public Health, ster Farimagsgade
5A, DK-1399 Copenhagen K, Denmark; Fax: +45-39464002
+45-39208010; E-mail: ula@niph.dk
Copyright  2008 by the Research Society on Alcoholism.
DOI: 10.1111/j.1530-0277.2008.00776.x
1920

ied (Nicholls et al., 1974; Pell and DAlonzo, 1973; Robinette

et al., 1979). Often large population studies include only a
small percentage of heavy drinking subjects resulting in limited power to study the health effects of heavy drinking
(Rimm et al., 1991). Also, cohorts comprising heavy drinkers
may be biased towards the inclusion of heavy drinkers who
tolerate alcohol well or who have a relatively more favorable
drinking pattern or lifestyle.
Studies within this eld of research are often limited by the
lack of differentiation between various cardiovascular conditions, though some have found the association between alcohol and different cardio- and cerebrovascular diseases to be
disparate. For instance, while alcohol appears to protect
against atherosclerosis (Iso et al., 2004; Klatsky, 1994;
Klatsky et al., 1990; Malarcher et al., 2001; McElduff and
Dobson, 1997; Mukamal et al., 2003, 2005a; Reynolds et al.,
2003; Rimm et al., 1991; Sacco et al., 1999; Stampfer et al.,
1988), adverse effects have been documented in hypertension
(Beilin and Puddey, 2006; Corrao et al., 1999; Klatsky et al.,
1990; MacMahon, 1987), hemorrhagic stroke (Corrao et al.,
1999; Klatsky, 1994; Klatsky et al., 1990, 2002; Reynolds
et al., 2003; Stampfer et al., 1988), and cardiomyopathy
(Gavazzi et al., 2000; Komajda et al., 1986; Lazarevic et al.,
2000). Cardiomyopathy in particular has been suspected of
being associated with a heavy prolonged alcohol intake, and
several clinical studies have indicated a causal relationship
(Gavazzi et al., 2000; Komajda et al., 1986; Lazarevic et al.,
Alcohol Clin Exp Res, Vol 32, No 11, 2008: pp 19201924

CARDIO- AND CEREBROVASCULAR RISKS IN ALCOHOL ADDICTS

2000). Consequently, it may be advantageous to analyse the

different subgroups of cardio- and cerebrovascular conditions
separately (Klatsky, 1994).
The objective of the present study was to examine whether
excessive alcohol intake is associated with higher cardio- and
cerebrovascular incidence rates compared to the general
population.
MATERIALS AND METHODS
The present study was based on Danish data from The Copenhagen Alcohol Cohort comprising patients who have attended Copenhagen outpatient clinics for alcoholics. This alcohol unit consists of 5
outpatient clinics located on each of the somatic hospitals within the
Copenhagen hospital Corporation. Since 1954, all patients have been
registered consecutively. Through interviews with the trained staff of
nurses and social workers, information on sex, age, date of birth as
well as social and alcohol related issues were obtained for each subject at rst contact with the outpatient clinic. Information on average
number of drinks per day and duration of alcohol misuse was
recorded in medical records independently of the interview, but at
the same time as the interview was carried out. By review of these
medical records data on average number of drinks per day and duration of alcohol misuse was attached to the rest of the dataset.
In total, the cohort comprises 15,368 men and 3,817 women
between the age of 14 and 83 years at time of rst admission (mean
age was 38.2 years in men and 38.7 years in women). The structured
registration of patients was changed in 1992, which is why patients
registered after this time were not included in the present analysis.
The Danish Civil Registration System was established in 1968,
ensuring that all live-born infants as well as current and new residents
are assigned a unique Personal Identication Number (PIN). Using
the PIN the cohort data were linked with the Danish National
Patient Registry, which contains information on all somatic hospital
admissions since 1st of January 1977. For each admission a discharge
diagnosis classied according to the International Classication of
Diseases is recorded. In the period from 1977 to 1993 diagnoses were
classied according to the eighth revision, and from 1994 onwards
diagnoses were classied according to the 10th revision. Cohort data
were furthermore linked with the Central Danish Death Register. In
this way complete follow-up on the health status of the subjects was
ensured and we were able to identify subjects cardio- and cerebrovascular hospitalizations and deaths from 1977 to 2001.
In the present study, we excluded subjects who died before 1977
(n = 1,458), and subjects who had been admitted to the hospital due
to a cardio- or cerebrovascular disease prior to their rst attendance
at the outpatient clinic. However, this was only possible in subjects
entering the cohort after 1st of January 1977, as information before
this period was not available.
The primary end point of the present study was cardio- and cerebrovascular morbidity and mortality. We grouped the various cardio- and cerebrovascular conditions into the following groups:
Ischemic heart diseases (ICD-8 codes 410-14 and ICD-10 codes
I20-I25), acute myocardial infarction (AMI) (ICD-8, 410; ICD-10,
I21-22), other heart diseases (ICD-8, 420-29; ICD-10, I30-I52),
cerebrovascular diseases (ICD-8, 430-38; ICD-10, I60-I67 and I69),
ischemic stroke (ICD-8, 433; ICD-10, I63), haemorrhagic stroke
(ICD-8, 430-31; ICD-10, I60-I62), arterial, arteriolar and capillary
diseases (ICD-8, 440-8; ICD-10, I70-9), atherosclerosis and aorticand arterial aneurysms (ICD-8, 440-42; ICD-10, I70-72), and cardiomyopathy (ICD-8, 425; ICD-10, I42-I43).
The outcome variable of the present study was based on either the
rst cardio- or cerebrovascular hospital admission or a cardio- or
cerebrovascular death diagnosis. The death diagnosis was merely
considered if the subject had not previously been admitted to the
hospital with the respective cardio- or cerebrovascular diagnosis.

1921

Subjects who died on the same day as their rst cardio- or

cerebrovascular hospital admission were coded based on their admission diagnosis, irrespective of whether they died of the same or a different cardio- or cerebrovascular condition.
We coded the cardio- and cerebrovascular hospital admissions and
deaths in the following manner: Subjects admitted with one or more
of the above mentioned diagnoses were coded in accordance with the
main group (e.g., ischemic heart diseases). Subjects diagnosed with
one of the above mentioned specic diseases were coded according to
this group as well (e.g., acute myocardial infarction). If a subject had
more than one of the relevant diagnoses, he or she was coded as
belonging to these corresponding main groups.
The data were analysed by standardized incidence ratios (SIR),
comparing the observed number of cases in the cohort with an
expected number obtained from incidences in the general population
(Breslow and Day, 1987). This method allowed us to examine to
which degree the incidence of cardio- and cerebrovascular disease in
the cohort differed from that of the general population. The cohort
was compared to the citizens of the County of Copenhagen and
Municipalities of Copenhagen and Frederiksberg. This was preferred
over a comparison with the general population of Denmark, as the
study population merely comprised citizens of Copenhagen, and the
life expectancy, and hence the health status, in Copenhagen is lower
than that of the general Danish population (Juel, 2004).
We calculated person-years at risk for each subject from the date
of entering the outpatient clinic, or from the 1st of January 1977
(if the subjects entry date was before this date), until time of cardioor cerebrovascular hospital admission, immigration, date of death,
or until the end of the study period (31st of December 2001), whichever occurred rst.
Expected numbers of cardio- and cerebrovascular cases were
derived by multiplying age- (5-year groups) and sex-specic incident
cases and corresponding person-years at risk in 5-year intervals of
the follow-up period taking account of possible time trends in disease
occurrence. These were summarized in order to obtain the overall
expected numbers of cases indirectly standardized for age and time
trends.
The SIR was estimated as the ratio of the observed number of
cases to the expected number. In addition, tests of signicance and
95% condence intervals were calculated according to formulas
given by Breslow & Day assuming that the observed numbers followed a Poisson distribution (Breslow and Day, 1987).

RESULTS
Baseline characteristics are given in Table 1. More than
90% of the patients fullled the ICD-10 criteria for alcohol
addiction. Average alcohol intake in men was 21.8 drinks per
day, ranging from 0 to 95 (median = 20.0) while women had
Table 1. Characteristics of the Study Population and the General Danish
Population
Danish
population
Men
Alcohol intake, drinks day
2.2a
Duration of abuse, years

Mean age at entry, years

Married, %
52.4
Educated, %
75.1
Working, %
69.2
a

Based on sales statistics.

Copenhagen alcohol
cohort

Men
Women
Women (n = 13,814) (n = 3,563)
2.2a

50.2
68.2
58.4

21.8
13.8
38.2
22.3
52.7
26.7

16.5
8.8
38.7
23.2
44.5
22.4

1922

HVIDTFELDT ET AL.

an average alcohol intake of 16.5 drinks per day, ranging

from 1 to 72 (median = 15.0). In comparison, the general
Danish adult population consumes on average 11.5 l of pure
alcohol per year; corresponding to an average daily alcohol
intake of 2.2 drinks per day. The average duration of abuse at
entry to the outpatient clinic was 13.8 years in men and
8.8 years in women and 77% of the men and 74% of the
women were drinking daily as opposed to periodic consumption. The percentage of working subjects was considerably
lower in the cohort compared to the general Danish population since only 26.7% of the men (vs. 69.2% in the general
population) and 22.4% of the women (vs. 58.4%) were working. The proportion of the cohort without education was
48% in men and 56% in women, while this percentage in the
general Danish population is 25 and 32, respectively. In addition, the percentage of married subjects was also considerably
lower in the cohort compared to the Danish population
in both men (22% vs. 52%) and women (23% vs. 50%)
(Kjoeller and Rasmussen, 2002) (Table 1).
The calculated SIRs and 95% condence intervals for both
men and women are shown in Table 2. During the period 1st
of January 1977 to 31st of December 2001, a total of 7,403
events of cardiovascular diseases and 1,994 events of cerebrovascular diseases were observed in the cohort.
Statistically signicant higher incidence rates than expected,
were observed for all the studied main- and subgroups. In
both men and women, strong excess risks were observed for
cardiomyopathy, arterial, arteriolar and capillary diseases,
and atherosclerosis and aneurysms. In women, this applies
for ischemic heart diseases and other heart diseases as well.
Moderately elevated risks are noted for acute myocardial
infarction in both men and women and for ischemic heart diseases and other heart diseases in men.
In both men and women, a strong excess risk was observed
in the main group of cerebrovascular diseases, as well as in
the 2 subgroups ischemic stroke, and haemorrhagic stroke.
DISCUSSION
Findings from this study indicate that alcohol addicts are
at increased risk of both cardio- and cerebrovascular diseases.
Several plausible explanations for a lowered risk of cardioand cerebrovascular disease among moderate drinkers com-

pared to abstainers have been discussed. The most important

of these mechanisms involves an increased blood concentration of high-density lipoprotein cholesterol, decreased plasma
brinogen levels, increased insulin sensitivity, and a reduction
in platelet aggregation and thereby protection against coronary heart disease (Davies et al., 2002; Klatsky, 1994;
MacMahon, 1987; Marmot and Brunner, 1991; Mukamal
et al., 2005b; Renaud and de Lorgeril, 1992; Rimm et al.,
1999). The apparent unfavourable effect of a heavy alcohol
intake on the pathophysiology of cardio- and cerebrovascular
diseases compared to a lower intake is less well described.
However, it has been hypothesized, that a heavy alcohol
intake could induce cardio- and cerebrovascular diseases by
increasing blood pressure and triglyceride levels (Puddey and
Beilin, 2006).
Previous population studies have suggested some adverse
effects of heavy alcohol consumption on cardio- and cerebrovascular diseases (cf. the U- or J-shaped curve) (Camacho
et al., 1987; Colditz et al., 1985; Corrao et al., 1999; Gronbaek
et al., 2000; Mukamal et al., 2003; Reynolds et al., 2003;
Rimm et al., 1991; Thun et al., 1997; Truelsen et al., 1998;
Wannamethee and Shaper, 1996). In this regard, our results
correspond with ndings of previous studies. However, the
risks of ischemic heart diseases and other heart diseases in
men, and the risk of AMI in both men and women were only
moderately elevated compared to the standard population.
As subjects in the present study have been exposed to excessive alcohol consumption long before the beginning of our
follow up, our ndings could reect a healthy drinker selection, i.e., heavy drinkers sensitive to alcohol may have
become sick or died before attending the outpatient clinic and
thus excluded from the present study. Consequently, the present study population may consist of particularly healthy alcoholics compared to the excluded alcoholics. However,
previous studies comparing the Copenhagen Alcohol Cohort
to the general population with regard to other health outcomes are not indicative of a particularly healthy cohort. One
study reported a 27-fold excess mortality from alcoholic liver
cirrhosis in men and a 35-fold excess mortality in women in
the Copenhagen Alcohol Cohort compared to the general
Danish population (Kamper-Jorgensen et al., 2004). In addition, another study reported increased risks of different types
of cancer, among them pancreatic cancer (RR = 1.3; 95%

Table 2. Observed (O) and Expected (E) Number of Cases and Standardised Incidence Ratios (SIR) With 95% Confidence Intervals (CI)
Men

Ischemic heart diseases

AMI
Other heart diseases
Cardiomyopathy
Arterial, arteriolar, etc.
Atherosclerosis, aneurysms
Cerebrovascular diseases
Ischemic stroke
Hemorrhagic stroke

Women

SIR

95% CI

SIR

95% CI

2433
1306
2017
174
1053
921
1668
447
398

1383.5
908.4
1044.2
61.0
472.1
376.5
728.6
180.8
146.7

1.76
1.44
1.93
2.85
2.23
2.45
2.30
2.47
2.71

1.691.83
1.361.52
1.852.02
2.443.31
2.102.37
2.292.61
2.182.40
2.252.71
2.452.99

325
131
400
13
158
128
326
62
75

133.0
69.8
130.5
4.3
55.8
39.2
105.6
24.4
27.0

2.44
1.88
3.07
3.02
2.83
3.27
3.09
2.54
2.77

2.192.73
1.572.23
2.773.38
1.615.18
2.413.31
2.723.88
2.763.44
1.953.26
2.183.48

CARDIO- AND CEREBROVASCULAR RISKS IN ALCOHOL ADDICTS

CI 1.01.8) and liver cancer (RR = 3.9; 95% CI 2.85.4)

(Tonnesen et al., 1994). Subjects entering a treatment program at baseline increase their chance of alcohol abstinence
and if a large proportion of the subjects became abstainers
during follow up, the observed SIRs could be blurred, as
abstinence too is considered a risk factor of cardio- and cerebrovascular diseases. In addition, a change from heavy consumption to more moderate levels could also affect the risks
towards lower SIR estimates.
It is recognized that alcohol intake and smoking habits are
strongly correlated (Jensen et al., 2003), and as smoking is a
well-established risk factor for cardio- and cerebrovascular
diseases, the results of the present study could be confounded
by smoking. Information on the smoking habits of the cohort
was not available. Methodological studies have concluded,
however, that only substantially different distributions of
smoking habits in the 2 groups of comparison would cause a
considerable confounding effect (Axelson, 1989). Furthermore, a previous study investigating the incidence of cancers
in the cohort found that the SIR of lung cancer was only
modestly increased in this cohort compared to the general
population (SIR = 2.6; 95% CI 2.32.8) (Tonnesen et al.,
1994). This indicates that smoking to some extend may
account for some of the higher incidence of cardio- and cerebrovascular diseases in this cohort.
Information on the morbidity of the subjects was obtained
through linkage with the Danish National Patient Registry
(LPR), which only holds information on hospitalizations.
Therefore, regarding information on diseases frequently treated in primary health care, such as angina pectoris, the Register will only provide information on the most serious
proportion of the disease spectrum, while diseases such as
acute myocardial infarction, almost inevitably leading to hospitalization or perhaps sudden death, are more likely to be
registered. As a result, underreporting of cases regarding the
less serious conditions in question may have occurred. Furthermore, subjects with undiagnosed cardio- or cerebrovascular diseases must be expected to appear in the cohort, causing
underreporting of morbidity. With regard to certain diseases
of which alcohol is considered a risk factor, it is possible that
a doctor will provide a more thorough examination, and thus
detect more cases, if she is aware of the patients excessive
alcohol abuse. This type of differential misclassication could
have exaggerated the effect of alcohol on cardio- and cerebrovascular diseases.
Information on hospitalizations was only available in the
Danish National Patient Registry from January 1, 1977. This
hindered exclusion of subjects who had been admitted to the
hospital with cardio- or cerebrovascular diseases prior to this
date. This makes a causal inference questionable, as the temporal separation of cause and effect may be lacking. Additional analyses excluding the rst 5 years of follow up showed
similar incidence ratios (data not shown), indicating that preexisting disease was not a crucial problem in this study.
It is important to note that the groups in question presumably differ with respect to lifestyle characteristics inuencing

1923

the risk of cardio- and cerebrovascular diseases other than the

amount of alcohol consumed, e.g., smoking, diet, physical
activity or body mass index. Therefore, we are not able to
conclude that the observed higher incidence rates of the
cohort compared to the general population are caused by
alcohol, i.e., a causal inference may be questionable. Finally,
although the applied type of analysis accounts for age and
calendar time, no element of duration of abuse before attendance at the outpatient clinic is included. In this regard,
subjects are treated equally irrespective of whether they are
life-long or short-term abusers. This could lead to differential
misclassication if these 2 groups vary according to treatment
response.
The present study was based on a large cohort of heavily
drinking men and women who were followed prospectively.
Both study design and the size of the study population distinguishes this study from previous studies on alcohol abusers
(Nicholls et al., 1974; Pell and DAlonzo, 1973; Robinette
et al., 1979). Selection bias due to loss of follow up was not
relevant in the present study because of the access to population covering register data. The considerable size of the cohort
made it possible to examine effects on both men and women.
Moreover, we were able to distinguish between various types
of cardio- and cerebrovascular conditions.
ACKNOWLEDGMENTS
We thank Statistician Sren Rasmussen, National Institute
of Public Health, Denmark for valuable comments.
CONFLICTS OF INTEREST
None.
FUNDING SOURCES
Helsefonden and the Danish Ministry of the Interior and
Health.
REFERENCES
Axelson O (1989) Confounding from smoking in occupational epidemiology.
Br J Ind Med 46:505507.
Beilin LJ, Puddey IB (2006) Alcohol and hypertension: an update.
Hypertension 47:10351038.
Breslow NE, Day NE (1987) Statistical methods in cancer research. Volume
IIThe design and analysis of cohort studies. IARC Sci Publ 82:1406.
Camacho TC, Kaplan GA, Cohen RD (1987) Alcohol consumption and mortality in Alameda County. J Chronic Dis 40:229236.
Colditz GA, Branch LG, Lipnick RJ, Willett WC, Rosner B, Posner B,
Hennekens CH (1985) Moderate alcohol and decreased cardiovascular mortality in an elderly cohort. Am.Heart J 109:886889.
Corrao G, Bagnardi V, Zambon A, Arico S (1999) Exploring the doseresponse relationship between alcohol consumption and the risk of several
alcohol-related conditions: a meta-analysis. Addiction 94:15511573.
Davies MJ, Baer DJ, Judd JT, Brown ED, Campbell WS, Taylor PR (2002)
Effects of moderate alcohol intake on fasting insulin and glucose concentrations and insulin sensitivity in postmenopausal women: a randomized controlled trial. JAMA 287:25592562.

1924

Gavazzi A, De Maria R, Parolini M, Porcu M (2000) Alcohol abuse and

dilated cardiomyopathy in men. Am J Cardiol 85:11141118.
Gronbaek M, Becker U, Johansen D, Gottschau A, Schnohr P, Hein HO,
Jensen G, Sorensen TI (2000) Type of alcohol consumed and mortality from
all causes, coronary heart disease, and cancer. Ann Intern Med 133:411
419.
Iso H, Baba S, Mannami T, Sasaki S, Okada K, Konishi M, Tsugane S (2004)
Alcohol consumption and risk of stroke among middle-aged men: the JPHC
Study Cohort I. Stroke 35:11241129.
Jensen MK, Sorensen TI, Andersen AT, Thorsen T, Tolstrup JS, Godtfredsen
NS, Gronbaek M (2003) A prospective study of the association between
smoking and later alcohol drinking in the general population. Addiction
98:355363.
Juel K (2004) Mortality in Denmark Throughout 100 years: Danes Live
Longer, But Why 34 years Shorter Than Swedish Men and French
Women? [in danish]. National Institute of Public Health, Copenhagen.
Kamper-Jorgensen M, Gronbaek M, Tolstrup J, Becker U (2004) Alcohol
and cirrhosis: doseresponse or threshold effect? . J Hepatol 41:2530.
Kjoeller M, Rasmussen NK (2002) Health and Morbidity in Denmark 2000
and the Development Since 1987 [in danish] . National Institute of Public
Health, Copenhagen.
Klatsky AL (1994) Epidemiology of coronary heart disease inuence of alcohol. Alcohol Clin Exp Res 18:8896.
Klatsky AL, Armstrong MA, Friedman GD (1990) Risk of cardiovascular
mortality in alcohol drinkers, ex-drinkers and nondrinkers. Am J Cardiol
66:12371242.
Klatsky AL, Armstrong MA, Friedman GD, Sidney S (2002) Alcohol drinking and risk of hemorrhagic stroke. Neuroepidemiology 21:115122.
Komajda M, Richard JL, Bouhour JB, Sacrez A, Bourdonnec C, Gerbaux A,
Rozensztajn L, Lablanche JM, Matinat D, Morand P (1986) Dilated cardiomyopathy and the level of alcohol consumption: a planned multicentre
case-control study. Eur Heart J 7:512519.
Lazarevic AM, Nakatani S, Neskovic AN, Marinkovic J, Yasumura Y,
Stojicic D, Miyatake K, Bojic M, Popovic AD (2000) Early changes in left
ventricular function in chronic asymptomatic alcoholics: relation to the
duration of heavy drinking. J Am Coll Cardiol 35:15991606.
MacMahon S (1987) Alcohol consumption and hypertension. Hypertension
9:111121.
Malarcher AM, Giles WH, Croft JB, Wozniak MA, Wityk RJ, Stolley PD,
Stern BJ, Sloan MA, Sherwin R, Price TR, Macko RF, Johnson CJ, Earley
CJ, Buchholz DW, Kittner SJ (2001) Alcohol intake, type of beverage, and
the risk of cerebral infarction in young women. Stroke 32:7783.
Marmot M, Brunner E (1991) Alcohol and cardiovascular disease: the status
of the U shaped curve. BMJ 303:565568.
McElduff P, Dobson AJ (1997) How much alcohol and how often? Population based case-control study of alcohol consumption and risk of a major
coronary event. BMJ 314:11591164.

HVIDTFELDT ET AL.

Mukamal KJ, Ascherio A, Mittleman MA, Conigrave KM, Camargo CA Jr,

Kawachi I, Stampfer MJ, Willett WC, Rimm EB (2005a) Alcohol and risk
for ischemic stroke in men: the role of drinking patterns and usual beverage.
Ann Intern Med 142:1119.
Mukamal KJ, Conigrave KM, Mittleman MA, Camargo CA Jr, Stampfer
MJ, Willett WC, Rimm EB (2003) Roles of drinking pattern and type of
alcohol consumed in coronary heart disease in men. N Engl J Med 348:109
118.
Mukamal KJ, Jensen MK, Gronbaek M, Stampfer MJ, Manson JE,
Pischon T, Rimm EB (2005b) Drinking frequency, mediating biomarkers, and risk of myocardial infarction in women and men. Circulation
112:14061413.
Nicholls P, Edwards G, Kyle E (1974) Alcoholics admitted to four hospitals in
England. II. General and cause-specic mortality. Q J Stud Alcohol 35:841
855.
Pell S, DAlonzo CA (1973) A ve-year mortality study of alcoholics. J Occup
Med 15:120125.
Puddey IB, Beilin LJ (2006) Alcohol is bad for blood pressure. Clin Exp Pharmacol Physiol 33:847852.
Renaud S, de Lorgeril M (1992) Wine, alcohol, platelets, and the French paradox for coronary heart disease. Lancet 339:15231526.
Reynolds K, Lewis B, Nolen JD, Kinney GL, Sathya B, He J (2003) Alcohol
consumption and risk of stroke: a meta-analysis. JAMA 289:579588.
Rimm EB, Giovannucci EL, Willett WC, Colditz GA, Ascherio A, Rosner B,
Stampfer MJ (1991) Prospective study of alcohol consumption and risk of
coronary disease in men. Lancet 338:464468.
Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ (1999) Moderate
alcohol intake and lower risk of coronary heart disease: meta-analysis of
effects on lipids and haemostatic factors. BMJ 319:15231528.
Robinette CD, Hrubec Z, Fraumeni JF Jr (1979) Chronic alcoholism and subsequent mortality in World War II veterans. Am.J Epidemiol 109:687700.
Sacco RL, Elkind M, Boden-Albala B, Lin IF, Kargman DE, Hauser WA,
Shea S, Paik MC (1999) The protective effect of moderate alcohol consumption on ischemic stroke. JAMA 281:5360.
Stampfer MJ, Colditz GA, Willett WC, Speizer FE, Hennekens CH (1988) A
prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women. N Engl J Med 319:267273.
Thun MJ, Peto R, Lopez AD, Monaco JH, Henley SJ, Heath CW Jr, Doll R
(1997) Alcohol consumption and mortality among middle-aged and elderly
U.S. adults. N Engl J Med 337:17051714.
Tonnesen H, Moller H, Andersen JR, Jensen E, Juel K (1994) Cancer morbidity in alcohol abusers. Br J Cancer 69:327332.
Truelsen T, Gronbaek M, Schnohr P, Boysen G (1998) Intake of beer, wine,
and spirits and risk of stroke : the copenhagen city heart study. Stroke
29:24672472.
Wannamethee SG, Shaper AG (1996) Patterns of alcohol intake and risk of
stroke in middle-aged British men. Stroke 27:10331039.