Critique 258.  Do the calories in alcoholic beverages increase the risk of obesity?                   November 2022

Bowden J, Harrison NJ, Caruso J, Room R, Pettigrew S, Olver I, Miller C. Which drinkers have changed their alcohol consumption due to energy content concerns? An Australian survey.  BMC Public Health 20222; 22:1775.  https://doi.org/10.1186/s12889-022-14159-9

Authors’ Abstract

Background: Alcohol is a discretionary, energy dense, dietary component. Compared to non-drinkers, people who consume alcohol report higher total energy intake and may be at increased risk of weight gain, overweight, and obesity, which are key preventable risk factors for illness. However, accurate consumer knowledge of the energy content in alcohol is low. To inform future behaviour change interventions among drinkers, this study investigated individual characteristics associated with changing alcohol consumption due to energy-related concerns.

Methods: An online survey was undertaken with 801 Australian adult drinkers (18–59 years, 50.2% female), i.e. who consumed alcohol at least monthly. In addition to demographic and health-related characteristics, participants reported past-year alcohol consumption, past-year reductions in alcohol consumption, frequency of harm minimisation strategy use (when consuming alcohol), and frequency of changing alcohol consumption behaviours because of energy-related concerns.

Results: When prompted, 62.5% of participants reported changing alcohol consumption for energy-related reasons at least ‘sometimes’. Women, those aged 30–44 years, metropolitan residents, those with household income $80,001– 120,000, and risky/more frequent drinkers had increased odds of changing consumption because of energy-related concerns, and unemployed respondents had reduced odds.

Conclusions: Results indicate that some sociodemographic groups are changing alcohol consumption for energy-related reasons, but others are not, representing an underutilised opportunity for health promotion communication. Further research should investigate whether messaging to increase awareness of alcohol energy content, including through systems-based policy actions such as nutritional/energy product labelling, would motivate reduced consumption across a broader range of drinkers.

Forum Comments

There have been numerous studies on the relation of alcohol consumption to body weight and obesity.  Given that alcohol contains calories, it might be assumed that the consumption of alcoholic beverages would lead to increased calories, and increased body weight.  However, essentially all observational studies tend to show that compared with abstainers, moderate drinkers have the same or lower body weight while only heavier drinkers tend to show higher weight and more obesity.  Further, in a limited number of clinical trials, the addition of moderate amounts of alcohol to the diet has not led to increase in body weight.  It is appreciated that the risk of obesity is influenced by many other lifestyle habits and social influences, as well as genetics, and we still do not have a clear picture of the relation between alcohol consumption and body weight and mechanisms involved.  These factors are not taken into account by the authors of the present paper, who seem to believe that any alcohol intake increases people’s risk of obesity, which is not supported by current data.

Forum member Ellison was concerned that the authors consider only adverse effects of alcohol consumption, and do not consider clear beneficial effects of moderate drinking, especially in reducing the risk of cardiovascular diseases, diabetes, and total mortality.  “The authors do not discuss that increases in several of the common diseases of ageing would be predicted to occur if all moderate drinkers stopped consuming alcohol.  Further, they consider all alcoholic beverages (beer, wine, spirits) the same, which is not usually found to be the case.  They do not take into account the important role of the pattern of drinking (regular versus episodic, in binges, with or without food, etc.) in terms of the effects on health.  Also, given the very unusual way the authors used to select subjects for the present analyses (the subjects had previously agreed to take part in ongoing research before being recruited for the present study), it is difficult to predict how the findings in this analysis may relate to the general population.  The main comparisons of the project are based on separating people concerned with possible effects of alcohol intake on body weight from those not expressing such a concern; the methods described raise concerns about misclassification of the ‘exposure’ for these analyses.   

“It seems unusual that, while the authors had detailed data on the number of drinks reported by subjects, they ended up consolidating drinking levels into only two groups: ‘daily or weekly’ and ‘monthly.’  There is further concern that the specific questions asked of subjects relating to the perceived effects of alcohol suggest that they would ‘guide’ the subject’s responses.  Also, while BMI would be assumed to be an important variable to use in determining the characteristics of the two groups of subjects being compared, the investigators based estimates of BMI on self-assessments by the participants of their height and weight (and such data were not reported by 19% who admitted that they ‘did not know’ what their height or weight was).  In Table 1 of the paper, those determined to be obese by calculated BMI were actually less likely than others to state that they had “changed alcohol consumption because of energy-related concern.’  Further, it appears that concerns about weight gain was not just in obese subjects, as 62.1% of those considered by the calculated BMI to be underweight or of healthy weight stated that they had changed their consumption due to such concerns.  These and many other characteristics of this study make it difficult to know how such data could be used in determining appropriate guidelines for the general population.”   

Reviewer Skovenborg agreed with these concerns, and has identified a number of other problems with the present paper.  “The first issue is that the authors provide no reservations regarding their statement that ‘these findings highlight weight gain risk for the population majority consuming moderate alcohol amounts, rather than only at an extreme of consumption.’  The findings relevant to effects of moderate alcohol consumption in this paper are references 7, 8 and 9.  However, the conclusion of their reference 7 (Ohlsson & Manjer) is ‘high intake of alcohol per occasion associated with obesity; high frequency of drinking is associated with reduced risk of obesity’ and the conclusion of reference 8 (Golzarand et al) is ‘only heavy drinking was positively associated with overweight. The association between alcohol intake and obesity was U-shaped such that moderate drinkers had lower odds of obesity compared to non-, light, and heavy drinkers.’  And finally, the conclusion of reference 9 (Dionisi & Addolorato) is: “small quantities of alcohol seem to have no effect.’  It looks like the fundamental premise of the paper ─ a moderate alcohol intake would increase the risk of overweight and obesity ─ is built on conflicting evidence and that the conflicting evidence of their own references regarding moderate alcohol consumption and obesity is disregarded by the authors.

“In addition, the meta-analysis paper (Golzarand et al) draws attention to two caveats: (1) interpretation of the results is difficult due to high heterogeneity between-studies; and (2) the association between alcohol intake and weight is complex and is mediated by several factors, such as physical activity, dietary habits, energy intake, genetics, sex, and smoking.

“The second issue, as mentioned, is drinking pattern – an important factor in relation to alcohol and health and also regarding alcohol and weight.  For a given level of total alcohol intake, obesity was inversely associated with drinking frequency in the Danish Diet, Cancer and Health Study, whereas the amount of alcohol intake was positively associated with obesity (Tolstrup et al).  The observation of an inverse association between BMI and frequent alcohol consumption in the authors’ reference 7 was explained by ‘a less healthy diet and malnutrition.’  The finding in the present paper of a lower BMI in patients who frequently consume alcohol is in accordance with previous studies, which show that frequent alcohol consumption was associated with a less healthy diet, malnutrition, and weight loss.  The conclusion of their Reference 12 (Addolorato et al) was the following: ‘Despite the higher total energy intake of alcohol consumers, because alcohol was added to the diet, body weight increased less than expected.  BMI was only marginally higher in men and it was significantly lower in women with high alcohol intake.’  The principal finding of their reference 29 (from the Australia New Zealand Food Standards Code, based on a study of older Japanese adults) was that ‘men and women in the lowest BMI quartile had lower total energy intake but higher alcohol intake than men in the other BMI quartiles.’  Overall, there is inconsistency between the conclusions of the authors of this paper and the authors of the papers that were cited.   Further, around 40–70% of inter-individual variability in BMI, commonly used to assess obesity, has been attributed to genetic factors (Maes et al), a fact not discussed in the present paper.  

“A third issue relates to the culture of shame: the pervasive misconception that obesity is primarily a lifestyle-related condition and the simplistic belief that weight reduction is just  adopting permanent caloric restriction, and the failure of a person with obesity to achieve and maintain weight loss is caused by a lack of discipline (Hagan et al).  The authors of the present paper call for ‘energy product labelling’ and ‘media campaigns highlighting the link between alcohol and weight gain’ is just more of the same public health policy, the failure of which has been proven over decades.  The intention of the present paper is in reality a suggestion to extend the shame of obesity to include a shame when drinking beer or wine, even in moderation.”

Forum member Lanzmann pointed out that another highly relevant paper on the topic was ignored by the authors: “The Arif & Rohrer study of NHANES III on 8,236 men and women who were non-smokers reported that the risk ratios for obesity were 0.46 for subjects reporting 1 drink/day (versus abstainers), 0.59 for 2 drinks/day, 1.07 for 3 drinks/day, and 1.77 for subjects reporting binge drinking versus no binge drinking.  (Results for the two lower consuming groups and the binge drinking group were statistically significant.)  Such data strongly support the evidence of many studies that there is no increase in risk of obesity (or even a decrease in risk) for moderate drinkers but an increase in risk among heavy drinkers.”

Forum member Waterhouse commented: “The authors seem either oblivious to the data on alcohol calories and obesity or are deliberately ignoring it.  Instead, their only concern appears to be to increase the incorrect perception that any alcohol consumption will increase obesity.  Then, by disseminating that misinformation to new populations, rely on the fear of obesity to decrease alcohol consumption.   It is a pity that some scientists and journals are so cavalier with the truth these days.” 

Reviewer Mattivi had similar concerns: “What can I say?  It is difficult to comment on a study that assumes that the energy intake provided by alcohol makes one fat in general, whereas there is a broad consensus that the effect of alcohol consumption is much more complex and such that an unambiguous and certain relationship cannot be reached.  For example, the conclusions of the detailed study by Canadian scientists (Traversy & Chaput) published in 2015 was ignored by the Australian authors.  Overall, the present paper appears to me to be of modest relevance — and perhaps, misleading — in the sense that it diverts the attention of experts, and potentially the population at risk, from the far more important and above all actionable effort to recommend moderation in drinking associated with a healthy and active lifestyle that does not lead to weight gain.”

Forum member Boban agreed: “I fully support what has been noted by other reviewers. This study is an example how misconceptions about alcohol and obesity are kept alive.  I have added a reference that supports statements in the Forum Comments about observational studies showing that moderate drinking may be associated with lower body weight, and a clinical trial showing that introduction of alcohol to the diet did not promote weight gain or abdominal adiposity (Wang et al, Golan et al).”   

Reviewer Goldfinger wrote: “As stated, the authors cited a study by Wang et al that makes it clear that moderate alcohol consumption, particularly wine, when the statistics are looked at, is associated with a reduced incidence of obesity.  Further, I echo the concern raised by Waterhouse that this study begins with the biased assumption that moderate alcohol consumption, based on intake of calories, is a significant risk factor for obesity.  Even if, in fact, their bias was correct, which it is not, the consistent findings of a lower risk of cardiovascular events, ischemic stroke, and total mortally in so many studies should have been on their radar, and leads to questions about the general health premise they presented.” 

Reviewer Finkel reflected the concerns of Forum members in a closing statement: “As these comments confirm, this paper fulfills my pessimistic age-old prediction about studies of alcohol and weight.  It continues to retail a number of misconceptions, and uses them to indict drinking.”

The conclusion section is misleading in stating that the association between alcohol consumption and weight gain is given. Also, it has not been shown that alcoholic beverages are the main contributor to the discretionary energy intake for this specific population. Following up on that issue, I would have liked to see a more extensive evaluation of other options for people to reduce their weight: it is unclear if  this study population preferred to reduce alcohol calories over e.g. other dietary calories or preferred alcohol consumption reduction over being more physically active. Such insights may be relevant for the efficacy of the educational campaigns proposed.

The following is an additional invited comment from Henk Hendriks, PhD, Senior Researcher, Centre for Earth, Environmental and Life Sciences, Zeist, Netherlands, “The conclusion section is misleading in stating that the association between alcohol consumption and weight gain is given. Also, it has not been shown that alcoholic beverages are the main contributor to the discretionary energy intake for this specific population. Following up on that issue, I would have liked to see a more extensive evaluation of other options for people to reduce their weight: it is unclear if  this study population preferred to reduce alcohol calories over e.g. other dietary calories or preferred alcohol consumption reduction over being more physically active. Such insights may be relevant for the efficacy of the educational campaigns proposed.”

“Policy recommendations are easily made, but are based on a very selective group which, according to the authors themselves, is not representative and would need confirmation of high quality studies with random sampling or demographic quotas consistent with national population characteristics”.

References from Forum critique

Addolorato A, Capristo E, Greco AV, Stefanini GF, Gasbarrini G.  Influence of chronic alcohol abuse on body weight and energy metabolism: is excess ethanol consumption a risk factor for obesity or malnutrition? J Intern Med 1998;244:387–395.

Arif AA, Rohrer JE.  Patterns of alcohol drinking and its association with obesity: data from the third national health and nutrition examination survey, 1988–1994.  BMC Public Health 2005; 5:126.  doi:10.1186/1471-2458-5-126.

Australia New Zealand Food Standards Code – Standard 1.2.7 – Nutrition, health and related claims. https://www.legislation.gov.au/Details/F2018 C00942.

Dionisi T, Addolorato G.  Effect of alcohol combined with high fat diet: Two wrongs don’t make a right but they could make a good excuse. Alcohol Alcohol 2020;56:348–350.

Golan R, Shelef I, Shemesh E, et al.  Effects of initiating moderate wine intake on abdominal adipose tissue in adults with type 2 diabetes: a 2-year randomized controlled trial.  Public Health Nutrition 2016;20:549–555.  doi:10.1017/S1368980016002597.

Golzarand M, Salari-Moghaddam A, Mirmiran P: Association between alcohol intake and overweight and obesity: a systematic review and dose-response meta-analysis of 127 observational studies. Crit Rev Food Sci Nutr. 2021. https://doi.org/10.1080/10408398.2021.1925221.

Hagan S. E66.01 and our culture of shame. N Engl J Med 2021;385:2307-2309.

Maes HH, Neale MC, Eaves LJ.  Genetic and environmental factors in relative body weight and human adiposity. Behav Genet 1997;27:325–351.

Ohlsson B, Manjer J. Sociodemographic and lifestyle factors in relation to overweight defined by BMI and “normal-weight obesity.” J Obes 2020;2020:2070297.

Tolstrup JS, Heitmann BL, Tjønneland AM, et al. The relation between drinking pattern and body mass index and waist and hip circumference.  Int J Obes 2005;29:490-497.

Traversy G, Chaput J-P.  Alcohol Consumption and Obesity: An Update.  Current Obesity Reports 2015;4:122–130.  DOI 10.1007/s13679-014-0129-4

Wang L, Lee I-M, Manson JE, Buring JE, Sesso HD.  Alcohol consumption, weight gain, and risk of becoming overweight in middle-aged and older women.   Arch Intern Med 2010;170:453-461 

Forum Summary

This study was designed to determine if greater emphasis on the role of alcohol consumption as related to body weight might lead to less drinking and have a beneficial effect on subsequent weight and obesity.  It was based on a survey of a select group of Australian adults who had previously volunteered to take part in medical research (not a population-based cohort). The authors took data from subjects who agreed with a question that asked if the subject ever thought about decreasing alcohol intake to lose weight, comparing them with those who did not answer the question in the affirmative.  They compared the two groups (those answering yes versus those answering no to this question) and found some relation with body weight and with BMI (the latter of which was based on self-report of subjects regarding their height and weight).

Unfortunately, the premise that any alcohol increases body weight is erroneous, as the investigators ignored current scientific data that has consistently shown that light-to-moderate alcohol consumption does not increase body weight (which is seen generally only with excessive alcohol intake or binge drinking).  While the authors had detailed data on the number of drinks reported by subjects, they ended up categorizing drinking levels into only two groups: ‘daily or weekly,’ or ‘monthly,’ which make it impossible to evaluate results of realistic drinking levels.  They present no data on the pattern of drinking (e.g., with or without food, on a regular basis or in binges) or effects of different types of alcohol.  They do not consider the complexity of factors related to obesity, and do not mention the effects of diet, exercise, smoking, or genetics on the risk of obesity.  They misquoted the results of several previous studies to support their premise that lowering alcohol intake for everybody in the population would lead to less obesity, while a number of these studies actually showed the opposite.  

Thus, the authors seem either oblivious to the data on alcohol calories and obesity or are deliberately ignoring it.  Instead, their only concern appears to be to increase the incorrect perception that decreasing alcohol consumption among everybody, regardless of their intake, will lower the risk of obesity.  Perhaps of even greater importance, they do not consider that a message to the entire population to decrease or stop alcohol consumption could result in increases in many diseases of ageing (e.g., coronary heart disease, ischemic stroke, diabetes, premature mortality risk) that are reduced by regular, moderate intake of alcohol, especially when consumed with food.

They end with conclusions that relate obesity to the culture of shame: the pervasive misconception that obesity is primarily a lifestyle-related condition and the simplistic belief that weight reduction is just adopting permanent caloric restriction, and the failure of a person with obesity to achieve and maintain weight loss is caused by a lack of discipline. 

Overall, the present paper appears to the Forum as potentially misleading in the sense that it diverts the attention of experts, and potentially the population at risk, from the far more important and above all actionable effort to recommend moderation in drinking associated with a healthy and active lifestyle that does not lead to weight gain.

*********************************************************************************************

Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

Mladen Boban, MD, PhD, Professor and Head of the Department of Pharmacology, University of Split School of Medicine, Croatia

R. Curtis Ellison, MD, Professor of Medicine, Emeritus; Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA

Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University Medical Center, Boston, MA, USA)

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, University of Arizona School of Medicine, Tucson, AZ, USA

Dominique Lanzmann, MD, PhD, Nutrition Geriatrics, Hôpital Emile Roux, APHP Paris, Limeil-Brévannes, France

Fulvio Mattivi, MSc, C3A – Center Agriculture, Food, Environment, San Michele all’Adige, Italy

Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark

Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Principal, Stockley Health and Regulatory Solutions; Adjunct Senior Lecturer, The University of Adelaide, Adelaide, Australia

Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France 

Fulvio Ursini, MD, Dept. of Biological Chemistry, University of Padova, Padova, Italy

Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA