Search for a command to run...
ABSTRACT Aims As part of a larger study to estimate the global burden of disease attributable to alcohol: to quantify the relationships between average volume of alcohol consumption, patterns of drinking and disease and injury outcomes, and to combine exposure and risk estimates to determine regional and global alcohol‐attributable fractions (AAFs) for major disease and injury categories. Design, methods, setting Systematic literature reviews were used to select diseases related to alcohol consumption. Meta‐analyses of the relationship between alcohol consumption and disease and multi‐level analyses of aggregate data to fill alcohol–disease relationships not currently covered by individual‐level data were used to determine the risk relationships between alcohol and disease. AAFs were estimated as a function of prevalence of exposure and relative risk, or from combining the aggregate multi‐level analyses with prevalence data. Findings Average volume of alcohol consumption was found to increase risk for the following major chronic diseases: mouth and oropharyngeal cancer; oesophageal cancer; liver cancer; breast cancer; unipolar major depression; epilepsy; alcohol use disorders; hypertensive disease; hemorrhagic stroke; and cirrhosis of the liver. Coronary heart disease (CHD), unintentional and intentional injuries were found to depend on patterns of drinking in addition to average volume of alcohol consumption. Most effects of alcohol on disease were detrimental, but for certain patterns of drinking, a beneficial influence on CHD, stroke and diabetes mellitus was observed. Conclusions Alcohol is related to many major disease outcomes, mainly in a detrimental fashion. While average volume of consumption was related to all disease and injury categories under consideration, pattern of drinking was found to be an additional influencing factor for CHD and injury. The influence of patterns of drinking may be underestimated because pattern measures have not been included in many epidemiologic studies. Generalizability of the results is limited by methodological problems of the underlying studies used in the present analyses. Future studies need to address these methodological issues in order to obtain more accurate risk estimates.