TL;DR
This study reveals that alcohol consumption in Chinese men is linked to a significantly higher risk of 61 diseases, including 33 not previously recognized as alcohol-related by the WHO. Using genetic data, the researchers confirmed these associations, highlighting the broad impact of alcohol beyond well-known risks and underscoring the need for stronger public health interventions in China.
What Problem Does This Paper Address?
Globally, alcohol consumption is a major contributor to disease and death, but understanding its full spectrum of health impacts, especially in diverse populations, remains a challenge. While the WHO has identified certain alcohol-related diseases, there's a persistent uncertainty about the causal links for many others, and whether observed associations in Western populations generalize to other ethnicities. Furthermore, observational studies are prone to biases like confounding and reverse causation (e.g., sick individuals quitting alcohol). This paper aims to comprehensively investigate the associations between alcohol consumption and a wide range of diseases in Chinese men, a population with distinct drinking patterns and disease profiles, and to leverage genetic data to strengthen causal inference.
Methods
The researchers employed a two-pronged approach:
- Observational Cohort Study: They analyzed data from the China Kadoorie Biobank (CKB), a large prospective cohort of over 512,000 adults. Participants provided detailed information on their alcohol consumption habits at baseline. Hospitalization and death records were followed for over 12 years, with over 1.1 million ICD-10 coded events analyzed. Cox proportional hazard models were used to estimate Hazard Ratios (HRs) for various disease outcomes, comparing different drinking categories and assessing dose-response relationships.
- Genetic Epidemiology (Mendelian Randomization - MR): A subset of participants (168,050) had their genotypes analyzed for two common East Asian variants, ALDH2-rs671 and ADH1B-rs1229984. These variants significantly influence alcohol metabolism and intake. By using these genetic variants as instrumental variables, the researchers could infer causal relationships between genetically predicted alcohol consumption and disease risks, mitigating biases inherent in observational studies. They constructed a genetic instrument that predicted a wide range of alcohol intake in men.
Datasets
The primary dataset used was the China Kadoorie Biobank (CKB) cohort, comprising over 512,000 adults recruited from diverse urban and rural areas across China between 2004 and 2008. Follow-up data extended to January 2019, with over 1.1 million ICD-10 coded hospitalization events and deaths. A subset of 168,050 participants had their genotypes analyzed.
Key Insights and Ideas
- Broadening the Scope of Alcohol-Related Harm: The study significantly expands the list of diseases linked to alcohol, identifying 33 conditions not previously recognized by the WHO as alcohol-related. This suggests that the health burden of alcohol is likely underestimated.
- Genetic Confirmation of Causal Links: The use of Mendelian Randomization provided robust evidence that the observed associations between alcohol consumption and many diseases are likely causal, not just correlational. This is particularly important for diseases where observational studies have shown conflicting or weak associations.
- Sex-Specific Differences: The study highlights that while alcohol consumption is widespread among Chinese men, it's very low in women. This allowed researchers to use women's genetic data to investigate whether the observed risks in men were due to the genetic variants themselves (pleiotropy) or the alcohol consumption they induced. The findings suggest the risks are primarily driven by alcohol intake.
- Beyond "Alcohol-Related" Diseases: The identification of new alcohol-associated diseases like cataract and gout underscores that alcohol's impact extends far beyond the traditionally recognized categories.
Key Results
- Observational Findings (Men): Alcohol intake was positively associated with 61 diseases, including 33 not defined as alcohol-related by the WHO. Examples include cataract (HR 1.21 per 280g/week) and gout (HR 1.57 per 280g/week). Overall morbidity risk increased by 1.12 per 280g/week higher intake.
- Genetic Findings (Men): Genotype-predicted alcohol intake was positively associated with established (HR 1.14 per 280g/week) and new alcohol-associated diseases (HR 1.06 per 280g/week). Specific diseases showed stronger genetic associations, including liver cirrhosis (HR 2.30 per 280g/week), stroke (HR 1.38 per 280g/week), and gout (HR 2.33 per 280g/week). Notably, ischemic heart disease showed no significant association.
- Women: Due to very low alcohol consumption (2%), statistical power to assess self-reported intake was limited. However, genetic findings in women suggested that the excess risks observed in men were not due to pleiotropic effects of the genetic variants.
Limitations and Future Work
- Under-reporting of Heavy Drinking: The authors acknowledge that heavy drinking might have been under-reported, potentially leading to an underestimation of risks.
- Longitudinal Trajectories: The study primarily used baseline and resurvey data to estimate usual intake but couldn't fully capture the effects of complex longitudinal drinking trajectories.
- Limited Power for Rare Diseases/Outcomes: For less common diseases or specific outcomes (e.g., certain cancers in women), statistical power was limited.
- Beverage Type: The study couldn't differentiate risks based on beverage type (e.g., spirits vs. wine) as spirits are dominant in China, and the genetic instrument didn't distinguish.
- Hypothesis-Generating for Some Findings: For certain individual diseases, the genetic analyses had limited power, making those findings hypothesis-generating.
Future work should focus on larger genetic studies to confirm associations for less common diseases, investigate gene-environment interactions (especially concerning acetaldehyde metabolism), and explore beverage-specific effects. The findings strongly advocate for public health initiatives to reduce alcohol consumption in China.
Summary updated