In the Nov. 8, 2011, issue of CMAJ, Latino-Martel and colleagues1 suggested that new evidence connecting alcohol consumption and increased risk of cancer means that drinking guideline limits are too high, at least as far as the prevention of cancer is concerned.
As scientific advisors for the forthcoming Canadian guidelines,2 we acknowledge the clear evidence that having even one drink per day may significantly increase the risk of cancer if this pattern is maintained over several years. We also acknowledge the importance of communicating this information to consumers. This was considered in our efforts to find a basis for a national consensus to replace the diverse sets of guidelines previously offered by different Canadian provinces.3,4
Has alcohol just now approached the status reached by tobacco over 60 years ago, when the connection between smoking and lung cancer was first established? Consuming alcohol is more complex because low levels of alcohol consumption may increase the risk of many conditions and reduce the risk of others — notably heart disease and diabetes. We relied on evidence from meta-analyses of alcohol consumption and all-cause mortality5 to identify a level of consumption at which potential risks and benefits are, for the average person, balanced in comparison with abstainers (i.e., at the ascending portion of the J-shaped curve where risk approached 1.0). The science underlying these studies is not perfect but it does provide a simple and intuitive basis for advice on upper limits for average daily consumption of alcohol (the level at which lifetime risk of premature mortality from all causes does not exceed that of an abstainer). The best available evidence was judged to suggest weekly upper limits of 10 standard drinks for women and 15 for men, so to limit the risk of serious illnesses (note: one Canadian standard drink contains 17.05 mL ethanol). Recommendations for upper daily limits, strategies to reduce short-term risks associated with drinking, and other recommendations are also provided in an independent scientific report to be published later in 2011.2 We strongly agree with Latino-Martel and colleagues1 that these are low-risk, not zero risk, guidelines, and that people deserve complete information about risks and possible benefits of alcohol upon which to make informed decisions.