Figure 2.

Linear Mendelian randomization analyses of alcohol consumption with all-cause and cause-specific mortality. The point estimates are represented by squares and the 95% confidence intervals by horizontal bars. Adjustments were made for age, sex, assessment center, birth location, single nucleotide polymorphism array, and the top 40 genetic principal components. IVW—inverse-variance weighted Mendelian randomization; MR-PRESSO—Mendelian Randomization Pleiotropy RESidual Sum and Outlier; W-Median—weighted median Mendelian randomization; W-Mode—weighted mode Mendelian randomization; CVD—cardiovascular disease. g/day—grams per day