Critique 170: An update on the association of alcohol consumption with the risk of cancer — 1 September 2015
Cao Y, Willett WC, Rimm EB, Stampfer MJ, Giovannucci EL. Light to moderate intake of alcohol, drinking patterns, and risk of cancer: results from two prospective US cohort studies. BMJ 2015;351:h4238; doi: 10.1136/bmj.h4238
Objectives: To quantify risk of overall cancer across all levels of alcohol consumption among women and men separately, with a focus on light to moderate drinking and never smokers; and assess the influence of drinking patterns on overall cancer risk.
Design: Two prospective cohort studies.
Setting: Health professionals in the United States.
Participants: 88 084 women and 47 881 men participating in the Nurses’ Health Study (from 1980) and Health Professionals Follow-up Study (from 1986), followed until 2010.
Main outcomes and measures: Relative risks of cancer.
Results: 19 269 and 7571 (excluding non-advanced prostate cancers) incident cancers were documented among women and men, respectively, over 3 144 853 person years. Compared with non-drinkers, light to moderate drinkers had relative risks of total cancer of 1.02 (95% confidence interval 0.98 to 1.06) and 1.04 (1.00 to 1.09; Ptrend=0.12) for alcohol intake of 0.1-4.9 and 5-14.9 g/day among women, respectively. Corresponding values for men were 1.03 (0.96 to 1.11), 1.05 (0.97 to 1.12), and 1.06 (0.98 to 1.15; Ptrend=0.31) for alcohol intake of 0.1-4.9, 5-14.9, and 15-29.9 g/day, respectively. Associations for light to moderate drinking and total cancer were similar among ever or never smokers, although alcohol consumption above moderate levels (in particular ≥30 g/day) was more strongly associated with risk of total cancer among ever smokers than never smokers. For a priori defined alcohol related cancers in men, risk was not appreciably increased for light and moderate drinkers who never smoked (Ptrend=0.18). However, for women, even an alcohol consumption of 5-14.9 g/day was associated with increased risk of alcohol related cancer (relative risk 1.13 (95% confidence interval 1.06 to 1.20)), driven by breast cancer. More frequent and heavy episodic drinking was not further associated with risk of total cancer after adjusting for total alcohol intake.
Conclusion: Light to moderate drinking is associated with minimally increased risk of overall cancer. For men who have never smoked, risk of alcohol related cancers is not appreciably increased for light and moderate drinking (up to two drinks per day). However, for women who have never smoked, risk of alcohol related cancers (mainly breast cancer) increases even within the range of up to one alcoholic drink a day.
This is an extremely thorough and well-done analysis of the association of alcohol consumption over many years with the risk of cancer. The authors conclude that men reporting an average intake of more than 15 grams of alcohol per day, but not less, have a significant increase in risk of alcohol-related cancers. In women, however, even the consumption of an average of 5.0 – 14.9 grams was associated with a slight increase in total cancer risk, stated to be primarily from an increase in the risk of breast cancer. For both genders, there seemed to be a dose-response increase in risk with larger amounts of alcohol.
Specific comments from Forum reviewers: Reviewer Finkel stated: “This paper concerns a relationship of vital importance, on several fronts. It offers those of our group expert in statistical analysis or in the nuances of epidemiology a field day of opportunities to comment. As a non-expert, I came away with the impression of a very competently executed study with adequate power. As one clinically expert in cancer and intimately familiar with one aspect of drinking, wine with meals, I regret further that information on the type of beverage consumed and on the pattern of drinking were not presented. I also wonder how significant in the life of the study population are the demonstrated statistically significant differences.”
The results of this study also indicate the importance of smoking as a risk factor for these cancers. The risk of breast cancer in women was increased even among non-smokers. However, for other alcohol-related cancers in women and all cancers in men, the authors state: “Joint analysis of alcohol intake and smoking suggested that, compared with alcohol, smoking was a much stronger risk factor for cancer, and independent influence of alcohol, especially in the range of light to moderate drinking, could be minimal.”
Missing information in the present paper: A primary concern of the Forum about the conclusions of this paper relate to the absence of consideration of a number of other lifestyle factors and drinking patterns: dietary factors (especially folate intake), whether or not the alcohol was consumed with food, whether the “moderate” drinkers were regular drinkers or binge drinker, etc. Forum member Van Velden was worried that such data (which are known to be available for the cohorts providing information for the present paper) were not included in the paper: “It is difficult to evaluate this data if we do not know the intake of other dietary factors, especially folate intake. Cancer incidence is related to multifactorial influences of lifestyle and genetic factors; alcohol does not work in isolation.”
The frequency of drinking (number of days per week on which alcohol was consumed) and the highest number of drinks consumed in one day were related to the risk of total cancer, but neither showed a significant trend when adjusted for total alcohol intake. From the data presented, it is not possible to determine if subjects whose intake was in the “moderate” range consumed their alcohol on a regular basis or drank only on weekends; this could make a big difference in their risk of cancer. Reviewer Waterhouse commented: “It’s a pity that with all the brain power on this paper, they did not figure out how to better address regular vs binge drinking, and they totally avoided type of beverage (spirits/wine/beer).”
Along the same lines, Forum member Stockley noted: “After looking at all of the additional data available from these studies, including much contained in supplemental data in the journal, I wonder why more information on the subjects had not been discussed in the actual paper.” She noted, for example, that the supplemental information provided suggests a lower risk of cancer for wine drinkers. For example, women reporting 1 – < 2 drinks per day who consumed wine had a multivariable risk ratio, versus non-drinkers, of 1.0 (95% CI 0.94-1.07). For the same level of alcohol intake, consumers of beer had a RR 1.07 (CI 0.96-1.19) and of liquor a RR of 1.12 (CI 1.04-1.20).
Further, for women who reported consumption of an average of ≥ 3 drinks/day, the fully adjusted RR for wine was 1.03. while for beer it was 1.34 and for liquor it was 1.26. A similar pattern (lower risk for wine drinkers) was seen for men as well. Concerning the total cancers in men, the multivariable RR has no significant p-value for trend for wine (0.95), while it was 0.01 for liquor and 0.09 for beer. Forum member Lanzmann-Petithory noted the same pattern: “I see to the naked eye some differences between wine and other alcoholic beverages. Those differences were not enough flagrant to arouse the interest of the authors, but they exist.” The paper states only that “All alcohol beverage types showed an association suggesting that the ethanol but not other components of the beverages was the primary factor.” This statement is inconsistent with much of their data.
Failure to comment on the net health effects of light-to-moderate drinking: Forum reviewers were somewhat surprised that the authors did not comment on the risk of other health outcomes or on mortality in this paper. These investigators have previously reported decreases in the risk of cardiovascular disease and total mortality among moderate drinkers in these two cohorts, but these were not mentioned in this paper. Reviewer Ursini noted: “I fully endorse the review by other Forum members. Although I’m not an expert in epidemiology I feel disappointed by the absence of data on mortality. In other words we still don’t know whether moderate drinking is good or not. My previous conclusion was not affected at all by this study.”
Forum members were reminded of the recent timely commentary on scientific publications that give outcomes for only one condition (cancer, in this case) by Vinay Prasad of the National Cancer Institute at NIH. Prasad’s article, “But how many people died? Health outcomes in perspective,” pointed out the importance of reporting all health outcomes when judging exposures in clinical trials or other prospective studies, and especially the need to not ignore total mortality. Reporting only on one outcome may give a false message; the commentary states: “We need to know an intervention has a net benefit before we recommend it to a healthy person. Overall mortality should be reported routinely in this population, particularly in settings where the cost to do so is trivia (i.e., in observational studies).”
Forum member Barrett-Connor disagreed with the last sentence of the article by Prasad. “The cost of following up in most studies is definitely not trivial; cost is one of several reasons it is so difficult to get funding agencies to pay for follow up of very large observational studies, although such follow up is extremely valuable. Nevertheless, I do agree this paper is worth commenting on and agree with citing the Prasad paper with this caveat.” Reviewer Ellison felt that given that the authors of the Cao et al paper already have available data on mortality, it would not have been much additional work to at least comment on it. Forum member Finkel stated: “I echo heartily other members’ complaint that the net effects of drinking are not explored or even discussed. Might information on beverage choice, drinking patterns, folate status, common apparent confounders, for example, have been mined from the huge data bank (or are these being saved for separate papers)?”
Reviewer Skovenborg commented: “I agree with the concerns of others about the lack of data on other lifestyle factors, especially issues such as folate intake. The cancer-alcohol association is an important issue for study and the results have important public health interest. It is very interesting that the authors found minimal differences among women with, versus without, a family history of breast cancer. Given this finding it is disappointing to read the following conclusion in the accompanying editorial by Rehm: ‘Finally, people with a family history of cancer, especially women with a family history of breast cancer, should consider reducing their alcohol intake to below recommended limits, or even abstaining altogether, given the now well established link between moderate drinking and alcohol related cancers.’” Reviewer de Gaetano added: “I’m wondering how many women may stop drinking alcohol in moderation and may die of cardiovascular disease; this is especially true after menopause, when the cardiovascular risk becomes exceedingly high. In every report we present on alcohol and health, we always include data on different alcoholic beverages and on total mortality.”
There is little question that heavy drinking markedly increases the risk of many of the “alcohol-related cancers” (especially the upper aero-digestive cancers). And no physician or agency advises people to consume in excess of the typical upper limits of no more than one or two drinks/day. Hence the finding in this study of an increase in breast cancer risk among women from light to moderate drinking has implications for giving advice to the public. The Forum was disappointed that in the present analyses not all dietary and other lifestyle factors that are known to increase the risk of cancer were evaluated, and especially that the net effects on health and mortality of light-to-moderate alcohol consumption were not commented on by the authors.
References from Forum critique
Prasad V. But how many people died? Health outcomes in perspective. Cleveland Clinic J Med 2015;82:146-150. doi:10.3949/ccjm.82a.14058.
Rehm J. Editorial: Light or moderate drinking is linked to alcohol related cancers, including breast cancer. BMJ 2015;351:h4400
This well-done analysis based on data from two very large cohort studies, the Nurses’ Health Study and the Health Professionals’ Follow-up Study, evaluated the association of alcohol consumption over many years with the risk of cancer. The authors conclude that men reporting an average intake of more than 15 grams of alcohol per day, but not less, have a significant increase in risk of alcohol-related cancers. In women, however, even the consumption of an average of 5.0 – 14.9 grams of alcohol per day (the equivalent of between ½ and 1 ½ typical drinks) was associated with a slight increase in total cancer risk, stated to be primarily from an increase in the risk of breast cancer. For both genders, there seemed to be a dose-response increase in risk of cancer with larger amounts of alcohol. At the same time, smoking was identified as an even more important risk factor than alcohol for these cancers.
Forum members considered this to be an important study presenting data that are of relevance to individuals and agencies providing advice regarding alcohol consumption. There is little question that heavy drinking markedly increases the risk of upper aero-digestive cancers, and no physician or agency advises people to consume in excess of the typical limits of no more than one or two drinks/day. Hence the finding in this study of an increase in some cancers among women (especially breast cancer) even for lighter drinking has implications for alcohol policy.
The Forum was disappointed that all dietary and other lifestyle factors that have been shown to increase the risk of cancer (data that are known to be available to these investigators) were not included in their report. Members were especially concerned that the net effects on health and mortality of light-to-moderate alcohol consumption were not commented upon in the paper. Given that the risk of the most common causes of death, especially cardiovascular diseases, are reduced by moderate drinking, it seems that the authors should have included data, or at least mentioned, the effects of alcohol on cancer mortality and, especially, on total mortality. It has recently been emphasized that cohort studies that report on only one outcome (in this case, cancer) but do not mention other outcomes that are affected by the same exposure, especially total mortality, do not present the full picture, and can lead to biased recommendations. The inclusion of such information from these cohorts could provide even more valuable data upon which to develop appropriate guidelines for alcohol use by individual patients and the public.
* * * * * * * *
Comments on this paper have been provided by the following members of the International Scientific Forum on Alcohol Research:
Elizabeth Barrett-Connor, MD, Distinguished Professor, Division of Epidemiology, Department of Family Medicine and Public Health and Department of Medicine, University of California, San Diego, La Jolla, CA USA
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
R. Curtis Ellison, MD. Section of Preventive Medicine & Epidemiology, Department of Medicine, Boston University School of Medicine, Boston, MA, USA
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Dominique Lanzmann-Petithory,MD, PhD, Nutrition/Cardiology, Praticien Hospitalier Hôpital Emile Roux, Paris, France
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia.
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Fulvio Ursini, MD, Dept. of Biological Chemistry, Universityof Padova, Padova, Italy
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA