Table 5.
Summary of main mortality studies.
| Study Funding/COI a |
Design (Mean/Median Years of Follow-Up) | n (Women) | Categories of Alcohol Consumption/Type of Drink | Variable/s | Reference Group (HR = 1) | Outcomes/Conclusions b |
|---|---|---|---|---|---|---|
| de Gaetano et al., 2016 [34] Assobirra, the Italian Association of the Beer and Malt Industries/ Some authors were consultants for the Web Newsletter of Assobirra, or were on the board/received lecture fees from Fundación Cerveza y Salud, FIVIN, the Beer and Health Foundation, ERAB, or Cerveceros de España. |
Systematic review | Wine, beer, and spirits | All-cause mortality | Evidence suggests a J-shaped relationship between alcohol consumption and total mortality, with lower risk for moderate alcohol consumers than for abstainers or heavy drinkers. Specific data on beer are not conclusive, although some results indicate a positive role of drinking beer in moderation (1 drink/day, about 12 g of ethanol) against mortality for any cause |
||
| 1 meta-analysis of 34 prospective studies [65] | Over 1 million adults | Low to moderate Women: 1 drink/day. Men: 2 drinks/day / Wine, beer, and spirits |
All-cause mortality | Low to moderate consumption of alcohol significantly reduces total mortality, while higher doses increase it | ||
| 1 Prospective cohort [66] (12–18 y) |
36,250 men | Wine and beer | CV death All-cause mortality |
Non-drinkers | Moderate wine or beer drinking reduced the risk of CV death. Only moderate wine drinking was associated with lower all-cause mortality: RR: 0.67 (0.58 to 0.77; p < 0.001) |
|
| 1 Prospective cohort [67] (16.8 y) |
7735 British men 40–59 y old | 1 SDU: Half pint beer (8–10 g alcohol). Frequency: Non-drinkers; Occasional (1–2 SDU/month); Weekend drinkers; Daily or on most days. Quantity: 1–2, 3–6, >6 / Wine, beer, and spirits |
All-cause mortality | Occasional drinkers | Regular beer drinking [HR: 0.84 (0.71 to 1.01)] showed no significant difference vs. occasional drinking | |
| 1 Prospective cohort [68] Copenhagen City Heart Study (25 y) |
14,223 adults | 1 SDU: 1 bottle beer (12 g alcohol). Never, Hardly ever, Monthly, Weekly Daily: 1–2 SDUs Daily: >2 SDUs / Wine, beer, and spirits |
All-cause mortality | Never beer drinkers | In men, monthly beer intake (RR: 0.86 (0.77 to 0.97)) was associated with lower mortality, and daily intake >2 beers (RR: 1.14 (1.02 to 1.27)) to increased risk. In women the associations were not statistically significant: Monthly beer intake (RR: 0.98 (0.88 to 1.08)), and daily intake >2 beers (RR: 1.31 (0.92 to 1.88)) At a medium education level, monthly beer intake was associated with lower risk (RR: 0.87 (0.77 to 0.97)), and at low [RR:1.20 (1.07 to 1.34) and medium education level (RR:1.18 (1.02 to 1.37)), >2 beers daily was associated with increased risk. |
|
| 1 Prospective cohort [69] (12.6 y) |
380,395 adults (247,795 women) | For beer: Never. Light: 0.1–2.9 g/day, 3–9.9 g/day, 10–19.9 g/day, 20–39.9 g/day (only for men). ≥20 g/day (extreme for women) ≥40 g/day (extreme for men) / Wine and beer |
All-cause mortality | Light consumers (0.1–2.9 g/day) | In women: Compared to low-level consumers, lifetime non-drinkers (HR: 1.06; 1.02 to 1.12), and consumers of beer at amounts ≥3 g/day displayed significantly higher overall mortality risk. In men: Lifetime non-drinkers (HR: 1.07; 0.98 to 1.16) and consumers of 3–9.9 g/day (HR: 1.04; 0.98 to 1.10) showed no significant differences compared to light consumers. Consumers of beer amounts ≥10 g/day displayed a significantly higher overall mortality risk. |
|
| Stockwell et al., 2016 [53] None declared |
Systematic review/meta-analysis of 87 studies (13.4 y) |
3998,626 adults | Abstainer. Former drinker. Occasional: <1.30 g/day. Low: 1.30 to <25 g /day. Medium: 25 to <45 g/day High: 45 to <65 g/day. Higher: ≥65 g/day / Alcohol in general |
All-cause mortality | Abstainer OR occasional drinker |
Standard adjustment: Significant protective effect for low-volume (RR: 0.86 (0.83 to 0.90); p < 0.0001) and occasional drinkers (RR: 0.84 (0.79 to 0.89); p < 0.0001) as compared with abstainers. Abstainers were at significantly higher risk (RR: 1.19 (1.12 to 1.27); p < 0.0001) as compared to occasional drinkers. Full adjustment: No significant protection was estimated for occasional (RR: 0.95 (0.85 to 1.05)), low-volume (RR: 0.97 (0.88 to 1.07)), or medium-volume drinkers (RR: 1.07 (0.97 to 1.18)) as compared with abstainers. |
| Xi et al., 2017 [51] None declared |
Population survey data linked to mortality data (8.2 y) |
333,247 adults | 1 SDU: 14 g alcohol. Lifetime abstainers. Lifetime infrequent drinkers. Former drinkers. Current light drinkers. Moderate: >3 to ≤14 drinks/w for men or >3 to ≤7 drinks/w for women. Heavy drinkers. Binge drinking / Alcohol in general |
All-cause, cancer, or CVD mortality. | Lifetime abstainers | All cause-mortality: Decreased for Light (HR 0.79 (0.76 to 0.82)) and Moderate (HR 0.78 (0.74 to 0.82)) drinkers. Increased in Heavy: HR: 1.11 (1.04 to 1.19) and binge (HR: 1.13 (1.04 to 1.23)) drinkers. CVD-specific mortality: Light: HR 0.74 (0.69 to 0.80); Moderate: HR 0.71 (0.64 to 0.78) |
| Bell et al., 2017 [49] National Institute for Health Research, Welcome Trust, the Medical Research Council prognosis research strategy Partnership and other government health-related agencies. |
Prospective cohort (6 y) |
1937,360 (51% women) | 1 SDU c: 8 g Non-drinkers. Former drinkers Occasional drinkers: drinks rarely or occasionally. Moderate: Men: 21 SDU/w or 3 SDU/day. Women: 14 SDU/w or 2 SDU/day Heavy drinkers / Alcohol in general |
CV death and all-cause mortality | Moderate drinkers | Non-drinkers (former and occasional drinkers removed) had an increased risk of CV death (HR: 1.32 (1.27 to 1.38)) and all-cause mortality (HR: 1.24 (1.20 to 1.28)). |
| Suadicani, 2008 [61] The King Christian X’s Foundation, The Danish Medical Research Council, The Danish Heart Foundation, and The Else & Mogens Wedell Wedellsborg Foundation. |
Prospective cohort (16 y) |
3022 Caucasian males 53–74 y old |
1 SDU: 10–12 g ethanol / Wine, beer, and spirits |
All-cause and IHD-related death within the different blood phenotypes | Alcohol abstainers (comparison only for wine drinkers) | For beer, the median (P20, P80) number of drinks/week among those with the non-O phenotype was significantly higher in those who died (overall mortality): 10.5 (0, 15.5) vs 7.5 (0, 10.5); p ≤ 0.001. The effect of wine intake on all-cause mortality among middle-aged and elderly men may depend on ABO phenotypes. Among non-O phenotype, drinking 1–8 drinks/w: HR: 0.8 (0.7 to 1.8) and drinking >8 drinks/w: HR: 0.7 (0.6 to 0.98) |
CV. Cardiovascular; CVD: Cardiovascular disease; ERAB: European Foundation for Alcohol Research; FIVIN: the Foundation for Wine and Nutrition Research; HR: Hazard Ratio; IHD: Ischemic heart disease; OR: Odds Ratio; SDU: Standard drinking unit; (to): 95%CI a When funding is provided by industries and/or foundations that might represent a conflict of interest, it is written in bold. b Outcomes for prospective studies and meta-analyses, and Conclusions for reviews. c Since Bell et al. follow UK guidelines, 1 SDU was assumed to be 8 g alcohol. Adjustments: de Gaetano et al., 2016: (A) 1 Prospective cohort [67]; For age, social class, smoking, physical activity, body mass index, lung function, evidence of CHD on questionnaire, diabetes, and regular medication. (B) 1 Prospective cohort [68]; For other types of alcohol, sex, smoking, body mass index, physical activity in leisure time, cohabitation, and education. (C) 1 Prospective cohort [69]; For age at recruitment, BMI and height, former drinking, time since alcohol quit-ting, smoking status, duration of smoking, age at start smoking, educational attainment, and energy intake. In women also for menopausal status, ever use of replacement hormones and number of full-term pregnancies. Stockwell et al., 2016: Standard adjustment for between-study variation in covariates: Former drinker, Occasional (<1.30 g/day), Low volume (1.30 to <25 g /day), Medium volume (25 to <45 g/day), High volume (45 to <65 g/day), Higher volume (65 g/day), All drinkers combined. Full adjustment for study characteristics: median age at intake, sex, Caucasian/non-Caucasian, drinking measure adequacy, former drinker bias, and occasional drinker bias. Xi et al., 2017: Model 1: Adjusted for age, sex, and race or ethnicity. Model 2: Additional adjustments for education, marital status, body mass index, physical activity, smoking, and diabetes) Bell et al., 2017: HRs adjusted for age (and age 2), sex, socioeconomic deprivation, and smoking status. Suadicani, 2008: Age adjusted (only for wine drinking).