Alcohol Consumption and Health Outcomes: Evidence from Mendelian Randomization

Alcohol consumption has been shown to be causally associated with more than 60 diseases, and recent debates have challenged the conventional view that low-to-moderate alcohol drinking has a protective effect on health conditions such as coronary heart disease and diabetes. In this study, we investigate the causal influence of alcohol consumption on various health outcomes by utilizing a Mendelian randomization (MR) on the genetic variation of aldehyde dehydrogenase 2 gene (ALDH2). The genetic basis of the MR approach relies upon the random allocation of genes at meiosis in humans, which likely resembles the random assignment into treatment groups in randomized controlled trials (RCT). Through collaboration with WeGene, a leading private genetic testing company in China, we conduct an online survey of alcohol drinking behavior based on its customer database in November 2018. The final sample contains about 3,000 respondents from mainland China with linked alcohol consumption and individual genotyping data. A special variant of the ALDH2 genotype that is common in Chinese (affecting about 35-45% of East Asians, but less than 1% of Caucasians) is known to cause the Alcohol Flushing Response (also known as Asian Flush) and adverse alcohol reaction (e.g. nausea, headache after drinking) due to decreased acetaldehyde metabolism, thereby plays a strong protective role against heavy drinking behavior and can be used as an instrumental variable (IV) to disentangle environmental confounders, which are otherwise very difficult to fully account for in observational studies. Our preliminary results confirm an inverse relationship between the presence of ALDH2*2 allele (rs671) and alcohol consumption frequency, as well as the maximum number of drinks consumed in a single occasion, especially in males. By using ALDH2*2 allele as an IV for alcohol consumption to account for potential confounders that may relate to individual socioeconomic background, we find that higher alcohol consumption in general is associated with an increased risk of chronic diseases. On the other hand, the IV point estimate of alcohol consumption on cancer is not significant. Our results highlight the practicability of using genetic information to gain new insights into the long-term health outcomes of alcohol drinking behavior.