Abstract P227: Association of Alcohol Intake with Cardiovascular and Total Mortality

Abstract

Introduction: Moderate alcohol intake is associated with reduced risk of ischemic heart disease and total mortality but might have a detrimental role in several other diseases (in particular some cancers). The association of moderate alcohol intake with reduced total mortality has therefore been questioned. Hypothesis: Alcohol intake have a non-linear association with cardiovascular (CV) and total mortality. Methods: Using harmonised data from the cohorts in the MOnica Risk, Genetics, Archiving and Monograph (MORGAM) Project, the association of alcohol intake with risk of cardiovascular disease and total mortality was assessed using multivariable Cox regression. Multiple imputation (n=10) was used to accommodate missing data for covariates. Results: 67.4% of the populations studied were drinkers (median intake among drinkers 11 g/day of ethanol, no missing data). Their mortality was observed over median follow-up of 13.8 yrs and covariates (age, sex, smoking, hypertension, diabetes, history of CVD, BMI and level of education) were available for 19 cohorts (1 from Australia and 18 from Europe), including 197,113 individuals (mean age 51±12 y, 60% men, no missing data) and 36,730 deaths (12,644 coronary or cerebrovascular). Former drinkers (2.8%) were excluded from the reference group, which thus comprised habitual teetotallers only (17.7%). Because of missing data, we failed to distinguish between teetotallers and former drinkers in n=24,036 (12.2%) non-drinkers; these individuals were analysed separately but not included in the reference group. In comparison with the reference group, intake of alcohol up to 5 g/day was associated with a 10.5% (95%CI: 6.7% to 14.2%) reduction in the risk of total mortality and 11.8% (5.2% to 17.9%) in the risk of CV mortality; intake between 5 and 10 g/day with a 6.5% (2.0% to 11.0%) and a 3.4% (0.3% to 15.5%) reduction, respectively, while intake over 20 g/day was associated with a 14.5% (9.6% to 19.7%) increase in risk of total mortality and a 6.3% (-1.5% to 14.7%) rise in the risk of CV mortality. Findings were similar in men and women and according to level of education, whereas they were heterogeneous across Countries, with greater protection by alcohol observed in Italy and France and lowest in Australia. The reduced risk of mortality associated with alcohol in moderation was more marked when the beverage of preference was wine rather than beer or spirits; the increased risk associated with heavy intake appeared in individuals whose beverage of preference was beer or spirits. Conclusions: Using a large multi Country cohort, we confirmed that intake of more than 2 alcoholic units per day is associated with a detrimental health effect, while intake of alcohol in moderation (up to 1 unit per day) is associated with reduced risk of CV and total mortality, with small variations according to Country and type of alcoholic beverage.