The article by Kunzmann and coworkers (1), evaluating the association of lifetime alcohol use with mortality and cancer risk in older adults, provides important information concerning potential benefits/risks of alcohol consumption. According to the 2015 National Survey on Drug Use and Health (NSDUH), 87.6 percent of people ages 18 or older reported that they drank alcohol at some point in their lifetime and 56.9 percent reported that they drank in the past month (2). Subjects who drink alcohol frequently ask their physicians/healthcare providers about the potential risks and benefits of alcohol consumption and seek advice on limits for consumption.

Most studies in the literature suggest that alcohol effects form a J-shaped curve on cardiovascular events and overall mortality, with modest alcohol consumption actually reducing overall risk. The U.S. Dietary Guidelines support this concept and indicate that safe, moderate drinking for women is one standard drink or less per day and for men is two standard drinks or less per day. However, concerns have been raised by some investigators about alcohol consumption at any level in certain disease processes, such as cancer, and there are vocal advocates, both for and against alcohol consumption, who impact the lay press with biased information.

This particular article is scientifically very helpful because it evaluates a cohort of almost 100,000 adults with over 800,000 person-years of follow-up (median 8.9 years), and there were a large number of deaths and primary cancers. Moreover, drinking consumption was well-characterized and divided into multiple gradations of intake. The investigators also adjusted for potential confounders. The authors’ findings generally support the U.S. Dietary Guidelines (3). They indicate a J-shaped curve related to overall mortality and cardiovascular mortality. Similar to the Dietary Guidelines, their data suggest differences between males and females, with females being more susceptible to harmful effects of alcohol at equivalent amounts of alcohol consumption. The overall cancer risk related to alcohol appears to be minimal at modest amounts of alcohol consumption and is higher in specific alcohol-related cancers.

Although statistical approaches are quite good, some additional considerations can help us better understand the data. For example, it is known that different amounts of drinking can have significant effects on the combined risk of cancer or death. A simple approach could be utilized with groups such as “never” + “infrequent drinkers” versus “heavy” + “very heavy drinkers” leading to (hopefully) non-overlapping 95% confidence intervals.

Importantly, a ‘J’ shape association is discussed at length, which relates to the hormesis effect (see for example, https://en.wikipedia.org/...). Given that this is a large cohort, we should be able to establish this statistically rather just comparing confidence intervals. The model could include a parameter for hormesis (4).

Supplemental Figure 4 gives interesting information, but not enough data to interpret the figure. Without more information (not included in the article), it is impossible to reach clear conclusions about the risks of consuming different forms of alcohol.

Our greatest concern relates to BMI stratification which is divided into only two groups (18.5-25 versus >25) in this paper. We suggest that BMI as a co-variate, should be grouped as 18.5-25, 25-30, 30-35, 35-40, >40 to adequately account for its influence on morbidity and mortality. In addition, BMI could be used as a continuous variable in the analysis. This is important because several “alcohol-related” cancers such as liver, breast, and colon cancer, are also associated with obesity, and approximately 2/3 of Americans are overweight or obese (BMI > 25). As an example, non-alcoholic fatty liver disease (NAFLD) is associated with obesity and can lead to cirrhosis and cancer (5-7). Moreover, liver cancer is increasingly recognized in patients with NAFLD but without cirrhosis (7). An elevated BMI increases the cirrhosis/liver cancer risk in the >25 BMI range (5, 6). Thus, stratification by more detailed BMI groups would have been useful to better distinguish the effects of alcohol from those of obesity.

In summary, this is an interesting article that gives us new clues to the effects of differing levels of drinking on the risk of death from all causes as well as from cancer. More information is needed to fully interpret these data, however.

Craig J. McClain, MD, FACN, FACG, AGAF, FAASLD
Professor, Medicine and Pharmacology & Toxicology
Distinguished University Scholar
Associate Vice President for Health Affairs/Research
Associate Vice President for Translational Research
University of Louisville
Louisville, KY 40202

Shesh N. Rai, PhD, FASA
Professor of Bioinformatics and Biostatistics
Wendell Cherry Chair in Clinical Trial Research
University of Louisville
Louisville, KY 40202




1. Kunzmann AT, Coleman HG, Huang W-Y, Berndt SI. The association of lifetime alcohol use with mortality and cancer risk in older adults: A cohort study. PLoS Med. 2018 Jun 19;15(6):e1002585. doi: 10.1371/journal.pmed.1002585. eCollection 2018 Jun. PMID: 29920516; PMCID: PMC6007830
DOI: 10.1371/journal.pmed.1002585

2. Substance Abuse and Mental Health Services Administration (SAMHSA). 2015 National Survey on Drug Use and Health (NSDUH). Table 2.41B—Alcohol Use in Lifetime, Past Year, and Past Month among Persons Aged 12 or Older, by Demographic Characteristics: Percentages, 2014 and 2015. Available at: https://www.samhsa.gov/da.... Accessed July 6, 2018.

3. US Dietary Guidelines. https://health.gov/dietar.... Accessed July 6, 2018

4. Hunt D and Rai SN. Testing threshold and hormesis in a random effects dose-response model applied to developmental toxicity data. Biometrical Journal, 47(3):319-328, 2005. PMID: 16053256.

5. Campbell PT, Newton CC, Freedman ND, et al., Body Mass Index, Waist Circumference, Diabetes and risk of Liver Cancer for U.S. Adults. Cancer Res. 2016 Oct 15;76(20):6076-6083. PMID: 27742674; PMCID: PMC5141518; DOI: 10.1158/0008-5472.CAN-16-0787

6. Chen Y, Want X, Wang J, Yan Z, Luo J. Excess body weight and the risk of primar liver cancer: An updated meta-analysis of prospective studies. Eur J Cancer. 2012 Sep;48(14):2137-45. doi: 10.1016/j.ejca.2012.02.063. Epub 2012 Mar 23. PMID: 22446023; DOI: 10.1016/j.ejca.2012.02.063

7. Karagozian R, Derdak Z, Baffy G. Obesity-associated mechanisms of hepatocarcinogenesis. Metabolism. 2014 May;63(5):607-17. doi: 10.1016/j.metabol.2014.01.011. Epub 2014 Feb 5. PMID:
24629562; DOI: 10.1016/j.metabol.2014.01.011