This page has been archived.

Back to Part D. Science Base, Section 8. Ethanol

 

PART D: SCIENCE BASE

Table D8-1. The Relationship Between Moderate Alcohol Consumption and Mortality (age 65+)

Inclusion Criteria: Prospective, Case-Control, Cross-Sectional Studies; Human Subjects; Publication Dates 1997 and After

Citation Design Population Exposure Outcome Duration Results Statistics
Camargo et al., 1997 Prospective cohort 22,071 men in Physicians' Health Study, aged 40-84 years with no history of MI, stroke, transient ischemic attack, or cancer <1 drink/wk;
1 drink/wk;
2–4 drinks/wk;
5–6 drinks/wk;
7–13 drinks/wk;
>14 drinks/wk
all-cause mortality 10.7 years Multivariate RR (age >52 y) <1 drink/wk 1.00; 1 drink/wk 0.81(0.63–1.03);
2–4 drinks/wk 0.71 (0.57–0.89);
5–6 drinks/wk 0.88 (0.69–1.12);
7–13 drinks/wk 1.02(0.86–1.22);
>14 drinks/wk 1.63 (1.23–2.14)
95 percent confidence interval; P-value for association - linear p=0.04; non-linear p<0.001 RR of cause-specific mortality also provided. Cohort had exceptionally low mortality rate, only 34 percent of that expected in a general population of white men with same age distribution during a similar period.
Chyou et al., 1997 Prospective cohort 8,006 Japanese-American men living in Hawaii, between 45–68 years at initial examination in 1965–1968 occasionally; lightly; moderately; heavily overall mortality 22 years J-shaped pattern in risk for intake of alcohol; synergistic interaction between BMI and alcohol—Men with intermediate BMI (21.21–26.30 kg/m2) and drank occasionally to lightly (0.01–24.99 oz/mo) RR 1.00 (reference group); Men with lowest BMI (<21.21 kg/m2) and drank moderately to heavily (>25 oz/mo) RR 1.63 (1.33–1.99) Synergistic interaction between BMI and alcohol, p=0.0017; RR - 95 percent confidence interval Increase in risk due to the interactive effect of low BMI and high alcohol intake was stronger (and statistically significant) than when each of these risk factors was considered separately
Dawson, 2001 Prospective cohort 42,910 adults 18 years and older; data from 1988 National Health Interview Study linked with the National Death Index for 1988 through 1985 abstainers; infrequent drinkers; light; moderate mortality 7.5 year followup Relative to lifetime abstainers and infrequent drinkers, the risk of death from external causes increased directly with volume of intake. No evidence for reduced risk of death among light or moderate drinkers.  
Dawson, 2000   37,682 U.S. adults age 25 years and older; data from 1988 National Health Interview Study linked with the National Death Index for 1988 through 1985 lifetime abstainers; past-year abstainers; light; moderate; heavy; very heavy all-cause mortality   OR past-year abstainers 1.00; light 0.76 (0.68–0.84); moderate 0.84 (0.74–0.96); very heavy 1.17 (0.93–1.47) 95 percent confidence interval When lifetime abstainers used as reference, the protective effect of moderate drinking fell short of significance. When dependence was considered, light and moderate drinkers without dependence had a reduced mortality risk regardless of reference group.
Farchi et al., 2000 Prospective cohort 1536 males aged 45–65 in 1965 in Northern and Central Italy <12 g/d; 13–48 g/d; 49–84 g/d; 85–120 g/d; over 120 g/d age-adjusted life expectancy; total mortality 30 years Age-adjusted life expectancy (years+/-SE) <12 g/d–19.6+/-0.9; 13–48 g/d–20.9+/-0.5; 49–84 g/d–21.6+/-0.4; 85–120 g/d–19.4+/-0.6; over 120 g/d–20.6+/-0.2 Years+/-SE Taking smoking habit into account, longest survival was observed in non-smokers drinking 4-7 drinks daily. Stratifying for physical activity, the longest survival was in men engaged in heavy physical activity at work drinking 1-4 drinks per day.
Gaziano et al., 2000 prospective cohort 89,299 U.S. men from the Physicians' Health Study who were age 40–84 years in 1982 and free of known MI, stroke, cancer or liver disease rarely/never drinkers; 1 drink/wk; 2–4 drinks/ wk; 5–6 drinks/wk; 1 drink/d; > 2 drinks/d total mortality 5.5 years of followup RR of total mortality rarely/never drinkers 1.00; 1 drink/wk 0.74; 2–4 drinks/ wk 0.77; 5–6 drinks/wk 0.78; 1 drink/d 0.82; > or = 2 drinks/d 0.95 Total mortality significant at 95 percent CI, except >or= 2 per day (0.79-1.14) CVD mortality L-shaped with apparent risk reductions even at highest category of > or = 2 drinks per day; no clear harm or benefit for total or common site-specific cancers
Hoffmeister et al., 1999 Prospective cohort 15,400 representative sample of German population and 2,370 regional sample of the Berlin-Spandau, age 25–69 years 0 g/d; 1–20 g/d; 21–40 g/d; 41–80 g/d; >80 g/d all-cause mortality 7 years for Berlin-Spandau population All-cause mortality hazard ratio (HR) for men - 0 g/day 1.00; 1–20 g/d 0.51 (0.29–0.90); 21–40 g/d 0.90 (0.51–1.56); 41–80 g/d 0.93 (0.49–1.76); >80 g/d 0.44 (0.10–1.86); All-cause mortality hazard ratio (HR) for women - 0 g/day 1.00; 1–20 g/d 0.83 (0.47–1.47); 21–40 g/d 1.29 (0.61–2.72); 41–80 g/d 0.81 (0.25–2.65); >80 g/d 4.20 (1.23–4.30) 95 percent confidence interval 65 percent of men and 85 percent of women were light or moderate drinkers.
Jackson et al., 2003 Prospective cohort 112,528 U.S. men from the Physicians' Health Study, 1320 of whom reported a baseline history of stroke rarely or never drink; very light (<1 drink/d); light (1–6 drinks/wk); moderate (> 1 drink/d) total mortality 4.5 years RR rarely or never drink 1.00; <1 drink/d 0.88 (0.60–1.28); 1–6 drinks/wk 0.64 (0.48–0.85); > 1 drink/d 0.71(0.54–0.94) 95 percent confidence interval; p=0.03 for trend RR for cardiovascular mortality - very light 0.89 (0.58-1.36); light 0.56 (0.40-0.79); moderate 0.64 (0.46-0.99); p=-0.008 for trend
Keil et al., 1997 Prospective cohort 1071 and 1,013 women, age 45–65 years, from the Ausburg region of Germany nondrinkers; drinkers (further divided by grams of alcohol/d) total mortality 8 years Hazard rate ratio nondrinkers 1.00; drinkers 0.59 (0.36–0.97); For different alcohol groups - 20–39.9 g/d 0.46 (0.20–0.80); > 80 g/day 1.04 (0.54–2.00) 95 percent confidence interval Total mortality HRR showed U-shaped curve.
Maskarinec et al., 1998 prospective cohort 40,000 persons with Caucasian, Chinese, Filipino, Japanese, and native Hawaiian ethnicity none, low alcohol intake (1–7 drinks/wk); higher levels of intake (>7 drinks/wk) all-cause mortality 20 years Men and women with low alcohol intake (1–7 drinks/wk) had 20 percent reduction in total mortality. At higher levels of intake, women and Asian men experienced no mortality benefit.
Mukamal et al., 2001 prospective cohort 1,913 adults hospitalized with AMI between 1989 and 1994 in 45 U.S. community and tertiary care hospitals none; less than 7 drinks/wk; 7 or more drinks/wk; (1 drink = 15 g alcohol) all-cause mortality 3.8 years Hazard ratio (full model) abstainers 1.00; <7 drinks 0.79 (0.60–1.03); > 7 drinks 0.68 (0.45–1.05) 95 percent confidence interval, p=0.01 for trend
Muntwyler et al., 1998 Prospective cohort 5,358 men from Physicians' Health Study who reported a history of MI and provided information on alcohol intake rarely/never drinkers; 1–4 drinks/ month; 2–6 drinks/wk; 1 drink/d; > 2 drinks/ d total mortality 5 years Multivariate RR-age 65–84y rarely/never drinkers 1.00; 1–4 drinks/mo 0.84 (0.65-1.07); 2–6 drinks/wk 0.70 (0.54–0.91); 1 drink/day 0.81 (0.64–1.02); > 2 drinks/d 0.89 (0.55–1.47) 95 percent confidence interval Total mortality and alcohol association did not differ significantly by age classification (40-64 y vs. 65-84 y)
San Jose et al., 1999 Prospective cohort 18,973 residents in Eindhoven, Netherlands abstainers; light (1–14 units/wk); moderate (15–28 units/wk); excessive (> or =29 units/wk) mortality   light or moderate drinkers had lower mortality than either abstainers or heavier drinkers  
Simons et al., 2000 Prospective cohort 1,235 men and 1,570 women age 60 years and over living in Dubbo, New South Wales, first examined in 1988-89 zero consumption; 1–7 drinks/week; 8–14 drinks/wk; 15–28 drinks/wk, >28 drinks/wk (1 drink = 10 g alcohol) mortality 116 months Hazard ratio: Men (60–74 y.o.) - no consumption 1.00; 1–7 drinks/wk 0.68 (.49–.94); 8–14 drinks/wk 0.58 (.39–.85); 15–28 drinks/wk 0.62 (.40–.95) >28 drinks/wk 0.56 (.33–.96); Women (60+ y.o.) - no consumption 1.00; 1–7 drinks/wk 0.78 (.61–.99); 8–14 drinks/wk 0.66 (.45–.97); 15–28 drinks/wk 0.67 (.29–1.55) 95 percent confidence interval Any intake of alcohol was associated with reduced mortality in men up to 75 years and in women over 64 years. After almost 10 years follow-up, men taking any alcohol lived on average 7.6 years longer and women on average 2.7 months longer, than non-drinkers. Study also provides HR for pattern of alcohol intake.
Theobald et al., 2000 Prospective cohort 1,828 individuals age 18-65 years lifelong abstainers; ex-drinkers; <50 g/wk; <140 g/wk total mortality 22 years RR compared with intake of wine less than once a week or not at all – Intake of wine once a week or more 0.58 (0.40–0.84); RR compared to lifelong abstainers and <50 g - ex-drinkers 2.64 (1.56-4.49) 95 percent confidence interval
Thun et al., 1997 Prospective cohort 490,000 people (251,420 women and 238,206 men) age 30–104 in 1982 that were part of the Cancer Prevention Study II nondrinkers; less than daily (but at least 3/wk); remaining reported in units per day (i.e. 1/day, 2/day, etc); (1 drink = 12 g alcohol) all-cause mortality 9 years RR for 60-79 y.o. with low cardiovascular risk - nondrinkers 1.00; less than daily 0.8 (0.8–0.9); 1 drink/d 0.8 (0.8–0.9); 2 drinks/d 0.8 (0.8–0.9), 3 drinks/d 0.9 (0.9–1.0); > 4 drinks/d 1.0 (0.9–1.1); RR for 60–79 y.o. with high cardiovascular risk - nondrinkers 1.00; less than daily 0.8 (0.8–0.9); 1 drink/d 0.8 (0.8–0.8); 2 drinks/d 0.8 (0.8–0.8), 3 drinks/d 0.8 (0.7–0.9); > 4 drinks/d 0.8 (0.7–0.8) 95 percent confidence interval
Woo et al., 2002 Prospective cohort 2,032 Chinese subjects aged 70 years and older (mean age 80 years) abstinence; occasional (less than once to up to twice per week); regular (three of more times weekly) mortality 3 years OR abstinence 1.00; occasional 0.625 (0.41,0.95); regular 0.684 (0.44,1.07) 95 percent confidence interval - However not statistically significant after adjusting for age and baseline self-perceived health

Back to Top

Continue to Part D. Science Base, Section 9. Food Safety