**Background:** Alcohol consumption is a major risk factor for poor health, accounting for about 3 million deaths and over 130 million disability-adjusted life years worldwide in 2016. In China, alcohol consumption has increased markedly since the 1990s, especially among men. Previous studies, mainly in Western populations, have linked alcohol to several cancers, cardiovascular diseases, liver cirrhosis, infectious diseases, and injuries, but evidence for many less-common or non-fatal outcomes is limited or contradictory. Moreover, the generalizability of Western findings to Chinese populations is uncertain due to differences in drinking patterns, disease prevalence, and alcohol metabolism genetics. This study aimed to comprehensively assess the associations between alcohol consumption and a wide range of disease outcomes in Chinese men and women using both observational and genetic approaches.
**Methods:** The study included 512,724 participants (210,205 men, 302,519 women) aged 30–79 years from the China Kadoorie Biobank, recruited from 10 geographically diverse areas during 2004–2008. Baseline data on alcohol consumption, lifestyle, and medical history were collected via questionnaires. Participants were classified as non-drinkers, ex-drinkers, occasional drinkers, or current drinkers (weekly). Among current drinkers, alcohol intake was estimated in grams per week. Follow-up for morbidity and mortality was conducted through linkage to death registries, disease registries, and the national health insurance system, with a median follow-up of 12.1 years. Cox proportional hazard models were used to estimate hazard ratios (HRs) for 207 disease outcomes in men, comparing ever-regular drinkers with occasional drinkers and assessing dose–response per 280 g/week higher usual alcohol intake among current drinkers. Analyses were stratified by age-at-risk and study area and adjusted for education and smoking. Genetic analyses used two East Asian-specific variants (ALDH2-rs671 and ADH1B-rs1229984) to create a genetic instrument predicting >60-fold difference in mean alcohol intake in men. Genotype-predicted mean alcohol intake was related to disease risks using Cox models stratified by age-at-risk and study area and adjusted for genomic principal components.
**Key Results:** Among men, 33% reported regular drinking at baseline, with a mean intake of 286 g/week. During follow-up, 134,641 men experienced at least one hospitalization or death. Alcohol consumption was significantly associated with higher risks of 61 disease outcomes from 15 ICD-10 chapters. Of these, 28 were previously considered alcohol-related by the WHO, including tuberculosis, six site-specific cancers (larynx, esophagus, liver, colon, rectum, lips/oral cavity/pharynx), diabetes, epilepsy, hypertensive diseases, cerebrovascular diseases, chronic ischemic heart disease, cardiomyopathy, pneumonia, alcoholic liver disease, liver cirrhosis, pancreatitis, and external causes. The HR per 280 g/week higher intake for aggregated WHO alcohol-related diseases was 1.22 (95% CI 1.19–1.25), ranging from 1.12 (1.05–1.20) for pneumonia to 1.97 (1.80–2.15) for esophageal cancer. Additionally, 33 diseases not previously classified as alcohol-related showed significant positive associations, including lung cancer (HR 1.16, 95% CI 1.04–1.30), stomach cancer, cataract, gastroesophageal reflux disease, gastric ulcer, gout (HR 1.94, 1.43–2.63 for purpura and other hemorrhagic conditions), and several fracture types. Only three diseases showed inverse associations (other nontoxic goiter, hyperplasia of prostate, inguinal hernia). Genetic analyses confirmed positive associations for aggregated WHO alcohol-related diseases (HR per 280 g/week higher genotype-predicted intake: 1.14, 95% CI 1.09–1.20) and new alcohol-associated diseases (1.06, 1.01–1.12). For specific diseases, genetic HRs were 1.38 (1.27–1.49) for stroke, 2.30 (1.58–3.35) for liver cirrhosis, and 2.33 (1.49–3.62) for gout. No significant genetic association was found for ischemic heart disease. Among women, due to low drinking prevalence (2% current drinkers), associations could not be reliably assessed, but genetic analyses in women showed no adverse effects, supporting that the findings in men are due to alcohol itself rather than pleiotropy.
**Clinical Implications:** This study provides robust evidence that alcohol consumption increases the risk of a much broader range of diseases than previously recognized, including 33 conditions not classified as alcohol-related by the WHO. The genetic analyses support causal interpretations for many of these associations, particularly for stroke, liver cirrhosis, and gout. The findings challenge the notion of protective effects of moderate drinking on ischemic heart disease or stroke, as no such benefits were observed in genetic analyses. The results underscore the need for public health strategies to reduce alcohol consumption in China, where drinking is prevalent among men and increasing. The alcohol-attributable disease burden is likely substantially higher than current estimates, which have been based on a narrower set of diseases and may have been biased by residual confounding in observational studies. Future global burden of disease estimates should incorporate these new findings.