Revisión sistemática y metaanálisis

Full protocol here

​

Review title
Effect of low-to-moderate alcohol consumption on all-cause mortality: systematic review and meta-analysis.

Review question
What is the long-term association between low-to-moderate alcohol intake (>0 & <50 g/d in males and >0 & <25 g/d in females) and all-cause mortality, as compared to never drinkers and also after excluding potential biases due to former drinkers (i.e., the "sick quitter effect")? What is the likelihood of bias to explain an apparent potential inverse association?

Intervention, exposure
Low-to-moderate alcohol consumption.

Comparator/control
The comparator will be never drinkers.

​

Context
Contrary to >100 conventional epidemiological studies showing benefits of low-to-moderate alcohol intake on cardiovascular risk or all-cause mortality, a small number of Mendelian Randomization (MR) studies have found a null or even a harmful linear relationship across all the range of alcohol intake, with some harms also associated with low-to-moderate intake, meaning that no level of consumption is healthy (1,2). MR studies have methodological advantages over conventional epidemiologic studies. Particularly, even small-to-moderate intakes of alcohol were associated in MR studies with a higher risk of stroke or heart disease and these findings have challenged the usual tenet that small-to-moderate alcohol consumption may reduce all-cause mortality. MR analyses avoid the problems of self-selection, thus reducing confounding, because genes for the enzymes related to the metabolism of alcohol (and, consequently, to its consumption) are distributed at random in the population. They may also provide a proxy of lifetime consumption. Also, MR studies remove potential reverse causation. In addition, several meta-analyses on alcohol and CVD published in the last 2 years have also attributed the protection by low-to-moderate alcohol intake to some potential flaws, particularly the "sick quitter" effect, of the observational studies, that they included in their systematic reviews (3,4). However, in their assessments (3,4) they
omitted a number of prospective studies, also excluding large and important cohorts (5-8). Furthermore, they did not show any detailed table disclosing the main characteristics and individual estimates from
each study. Some recent recommendations are supporting complete abstention as the healthiest option, not only for younger subjects but for the totality of the population (9). These recommendations on abstention started to be promoted after the results published in 2018 by the Global Burden of Diseases, Injuries, and Risk Factors Study 2016 (10). Notwithstanding, the updated estimates by the Global Burden of Disease using data collected up to 2020 concluded that small amounts of alcohol consumption (and not 0 intake) are likely to be associated with the
best indicators of health, depending on age, sex and the underlying distribution of causes of disease and death in each population of the world (11). In this context, sufficiently comprehensive meta-anaysis including the totality of the evidence and assessing potential sources of bias with full disclosing of the original estimates is needed.
(1) Carr S, Bryazka D, McLaughlin SA, et al. A burden of proof study on alcohol consumption and ischemic heart disease. Nat Commun. 2024;15:4082.
(2) Kassaw NA, Zhou A, Mulugeta A, Lee SH, Burgess S, Hyppönen E. Alcohol consumption and the risk of all-cause and cause-specific mortality-a linear and nonlinear Mendelian randomization study. Int J Epidemiol. 2024;53(2):dyae046.
(3) Stockwell T, Zhao J, Clay J, et al. Why Do Only Some Cohort Studies Find Health Benefits From Low-Volume Alcohol Use? A Systematic Review and Meta-Analysis of Study Characteristics That May Bias Mortality Risk Estimates. J Stud Alcohol Drugs. 2024;85:441-452.
(4) Zhao J, Stockwell T, Naimi T, Churchill S, Clay J, Sherk A. Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Netw Open. 2023;6:e236185. Erratum in JAMA Netw Open. 2023;6:e2315283.
(5) Bell S, Daskalopoulou M, Rapsomaniki E, et al. Association between clinically recorded alcohol consumption and initial presentation of 12 cardiovascular diseases: population based cohort study using linked health records. BMJ. 2017;356:j909.
(6) Ma H, Li X, Zhou T, Sun D, Shai I, Heianza Y, Rimm EB, Manson JE, Qi L. Alcohol Consumption Levels as Compared With Drinking Habits in Predicting All-Cause Mortality and Cause-Specific Mortality in Current Drinkers. Mayo Clin Proc. 2021;96:1758-1769.
(7) Li Y, Pan A, Wang DD, et al. Impact of Healthy Lifestyle Factors on Life Expectancies in the US Population. Circulation. 2018;138:345-355.
(8) Tian Y, Liu J, Zhao Y, et al. Alcohol consumption and all-cause and cause-specific mortality among US adults: prospective cohort study. BMC Med. 2023;21:208.
(9) Burton R, Sheron N. No level of alcohol consumption improves health. Lancet. 2018;392:987-8.
(10) GBD 2016 Alcohol Collaborators. Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2018;392:1015-1035. Erratum in: Lancet. 2018;392:1116. Erratum in: Lancet. 2019;393:e44.
(11) GBD 2020 Alcohol Collaborators. Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020. Lancet. 2022;400:185-235. Erratum in: Lancet. 2022;400:358.

Main outcome
All-cause mortality as relative risk from low-to-moderate alcohol intake.
 

Strategy for data synthesis

A random-effects meta-analysis will be conducted. Relative Risk (RR) with 95% Confidence Intervals (CI) will be used to combine all studies, treating Odds Ratios (OR) and Hazard Ratios (HR) as equivalent to RR. Results stratified by sex will be considered as separate reports. If there are differences in the cut-off points for alcohol intake in grams per day across various articles, we will standardize the RR with 95% CI to >0 & <50 g/day for males and >0 & <25 g/day for females. The mean or median alcohol consumption for each category will be used as the corresponding dose consumption. Consistency between articles will be assessed using Cochrane’s Q test and quantified by the I² statistic. Combined associations will be analyzed using random-effects models according to the Der Simonian and Laird method. The overall effect will be depicted in forest plots. Potential publication bias will be evaluated using a funnel plot and the Egger statistical test. Based on the characteristics of individual studies that may lead to biased results, a meta-regression may be performed.