Table 3

Cohort studies of consumption of specific types of alcoholic drink and relative risk of coronary heart disease

WineBeerSpirits
StudyConsumptionRelative risk (95% confidence interval)ConsumptionRelative risk (95% confidence interval)ConsumptionRelative risk (95% confidence interval)Comment
Yano et al 1977,19 Kagan et al 198120
7591 Men aged 46-68 (190 deaths from myocardial infarction or coronary heart disease in 6 years follow up) in Honolulu heart studyml/day*:ml/day*:ml/day*:Adjusted for age, drink types not simultaneously adjusted for
0
1
>/=2
1.0+
0.67+
0.71 (NS)+
0
1-299
>/=300
1.0+
0.74+
0.57 (P<0.001)+
0
1-2
>/=3
1.0+
0.93+
0.71 (NS)+
Kozarevic et al 198021
11 121 Men aged 35-62No of drinks/day:No of drinks/day:No of drinks/day (weak rakija, 30% alcohol):Adjusted for age, drink types not simultaneously adjusted for
  (estimated 166 with non-fatal myocardial infarction or fatal coronary heart disease in 7 years follow up) in Yugoslavia<1
>/=1
1.0+
0.7 (NS)+
<1
>/=1
1.0+
0.6 (NS)+
<1
>/=1
1.0+
0.8 (NS)+
No of drinks/day (strong rakija, 40% alcohol):
<11.0+
>/=10.6 (NS)+
Salonen et al 198322
4063 Men aged 30-59 (209 with acute myocardial infarction in 7 years follow up) in eastern FinlandNegligible consumptionNo of bottles/week:No of drinks/week:Adjusted for age, smoking, serum cholesterol and triglyceride concentrations, blood pressure
<5
>/=5
1.0
0.8 (0.5 to 1.4)
<1
>/=1
1.0
0.3 (0.1 to 0.7)
Kittner et al 198323
9150 Men aged 35-79 (164 died from coronary heart disease in 8 years follow up, 306 died in 12 years follow up) in Puerto Rico heart health programNegligible consumptionBeer and spirits combined (drinkers v non-drinkers):
Sudden coronary heart disease: Non-sudden coronary heart disease:
1.3 (0.8 to 2.0)
0.7 (0.4 to 1.0)
Alcohol intake assessed by single 24 hour recall; adjusted for age, smoking, exercise, location of residence
Non-fatal myocardial infarction:0.7 (0.5 to 1.0)
Friedman et al 198624
2106 Men aged 30-59 (24 year follow up) in Framingham heart studyNo of drinks/day in non-smokers++:No of drinks/day in non-smokers++:No of drinks/day in non-smokers++:Adjusted for systolic blood pressure, serum cholesterol concentration, haemoglobin concentration, left ventricular hypertrophy, relative weight
0
2
1.0)
0.25 (P=0.07)
0
2
1.0
0.49 (P=0.15)
0
2
1.0
0.66 (P=0.21)
Stampfer et al 198825
87 526 Women aged 34-59 (200 with fatal or non-fatal myocardial infarction in 4 years follow up) in nurses' health studyTotal alcohol (g/day):Total alcohol (g/day):Total alcohol (g/day):Adjusted for family history of coronary heart disease, menopause, hormone replacement therapy, age, smoking, body mass index, hypertension, high cholesterol concentration, exercise, intake of dietary fat and cholesterol
01.001.001.0
<50.9 (0.7 to 1.2)<50.3 (0.2 to 0.8)<51.1 (0.7 to 1.8)
>/=50.4 (0.2 to 0.8)>/=51.0 (0.6 to 1.6)>/=50.7 (0.4 to 1.3)
Rimm et al 199126
44 059 Men aged 40-75 (350 with fatal or non-fatal heart disease or revascularisation in 2 years follow up) in health professionals follow up studyNo of drinks/day&:No of drinks/day&:No of drinks/day&:Adjusted for family history of coronary heart disease; smoking; age; body mass index; profession; diabetes; hypertension; high cholesterol concentration; intake of dietary fat, fibre, and cholesterol
01.001.001.0
20.98 (0.64 to 1.50)20.80 (0.54 to 1.18)20.55 (0.39 to 0.77)
Farchi et al 199227
1563 Men aged 45-64 (166 died Quintiles of mean alcohol intake Alcohol consumption mainly or from cardiovascular disease in 15 years follow up) in Italian rural cohorts study(g/day):exclusively wineAdjusted for age, smoking, and occupation. Men with prevalent cardiovascular disease at baseline were excluded
22.71.0
56.40.77 (0.34 to 1.76)
77.80.67 (0.29 to 1.58)
108.21.31 (0.64 to 2.66)
164.71.61 (0.79 to 3.31)
Klatsky et al 1992,3 1993,28 1990,29 198630
129 170 Men and women (600 died from coronary heart disease in 7 years follow up) in Kaiser Permanente studyNo of drinks/week@:No of drinks/week@:No of drinks/week@:Adjusted for age, race, sex, smoking, body mass index, marital status, education
<21.0<21.0<21.0
>/=20.5 (0.4 to 0.7)>/=20.7 (0.5 to 0.9)>/=20.6 (0.5 to 0.8)
Gronbaek et al 199531
7217 Women and 5633 men aged 30-70 (1119 died from coronary heart disease in 12 year follow up) in Copenhagen city heart studyDrinks#:Drinks#:Drinks#:Age, smoking, sex
Never1.0Never1.0Never1.0
Monthly0.69 (0.62 to 0.77)Monthly0.79 (0.69 to 0.91)Monthly0.95 (0.85 to 1.06)
Weekly0.53 (0.45 to 0.63)Weekly0.87 (0.75 to 0.99)Weekly1.08 (0.93 to 1.26)
1-2 Daily0.47 (0.35 to 0.62)1-2 Daily0.79 (0.68 to 0.91)1-2 Daily1.16 (0.98 to 1.39)
3-5 Daily0.44 (0.24 to 0.80)3-5 Daily0.72 (0.61 to 0.88)3-5 Daily1.35 (1.00 to 1.83)
  • * Consumption of alcohol calculated as consumption of each drink type without conversion to alcohol. In this population consumption of wine and spirits was small compared with beer: average monthly consumption was 1-2 glasses of wine, 1-2 glasses of spirits, and 22 12 oz glasses of beer.20

  • +Relative risks (age adjusted) derived from rates of coronary heart disease for each category of alcohol consumption divided by rate for abstainers.

  • ++Alcohol content per drink was defined as 4 oz for wine (16.75% alcohol), 8 oz for beer (5% alcohol), and 2 oz for spirits (50% alcohol). We calculated relative risks from published

  • ß coefficients from a Cox proportional hazards model which accounted for all three drink types simultaneously.24

  • &Drinks of each type defined as a 4 oz glass of wine (10.8 g alcohol), a 12 oz glass of beer (13.2 g alcohol), and a shot of spirits (15.2 g alcohol).

  • @The amount of alcohol per drink was not defined.

  • #Drinks defined to contain 12 g alcohol on average.

  • 1 oz alcohol=28.3 g.

  • NS=Non-significant.