Critique 259: Importance of consuming alcohol with meals in relation to the risk of type 2 diabetes mellitus – 30 November 2022
Ma H, Wang X, Li X, Heianza Y, Qi1 L. Moderate alcohol drinking with meals is related to lower incidence of type 2 diabetes. Am J Clin Nutr 2022;0:1–8.
Background: Previous studies on alcohol drinking and health largely have ignored the potential impact of the timing of drinking.
Objectives: We aimed to investigate the joint associations of the timing of alcohol intake with respect to meals (i.e., with meals or outside of meals) and the amount of alcohol consumed with the risk of type 2 diabetes (T2D).
Methods: A total of 312,388 current drinkers from the UK Biobank without T2D at baseline were included. Cox proportional hazards models were used to examine the association between the timing of alcohol intake with respect to meals and the risk of T2D.
Results: During a median of 10.9 y of follow-up, 8598 incident cases of T2D were documented. After adjustment for covariates and the amount of alcohol consumed, consuming alcohol with meals was significantly associated with a 12% lower risk of T2D (HR: 0.88; 95% CI: 0.83, 0.93) than was consuming alcohol outside of meals. In addition, we found that the timing of alcohol intake with respect to meals significantly modified the relations between the amount of alcohol consumed and risk of T2D (Pinteraction = 0.017); the beneficial association of moderate drinking with T2D risk was only observed in participants who consumed alcohol with meals, but not in others. Further analyses on various types of alcoholic beverages indicated that the beneficial associations between alcohol drinking with meals and T2D were mainly driven by wine consumption. Moreover, we found that when consumed together with meals, drinking more wine, rather than other alcoholic beverages, was related to lower concentrations of C-reactive protein.
Conclusions: In current drinkers, moderate drinking of alcohol, especially wine, with meals is associated with a lower risk of T2D.
For decades, members of this Forum have emphasized that the pattern of drinking, usually even more so that the amount of alcohol consumed, has the most striking effects on health outcomes. For example, an individual consuming 20 grams of alcohol each evening with dinner would have a total alcohol intake of 140 grams/week. At the same time, someone who has 14 drinks only on the weekend would have the same average weekly intake, but the health consequences could be very different.
The present analysis provides key data supporting the advantages of consuming alcoholic beverages with meals, rather than on an empty stomach, in terms of the risk of diabetes mellitus. It also shows that, in terms of diabetes, wine consumption provides greater health benefits than those associated with other beverages.
The paper is based on information from a huge dataset (from more than 300,000 subjects) collected for the UK Biobank. The authors had very reasonable estimates of alcohol intake as well as good ascertainment of health outcomes; 8,598 cases of diabetes were ascertained over a period of 10.9 years. They also had considerable data on many lifestyle and social factors that relate to the development of diabetes so that confounding and modification of effect by them could be dealt with in the analyses.
These analyses were conducted only among subjects reporting that they were “drinkers.” Thus, the investigators do not present data on the risk of diabetes among abstaining subjects. (Based on extensive previous reports from cohort studies, it would be expected that the risk of diabetes would have been considerably higher among abstainers than among drinkers.) However, it is interesting that even among the present cohort, all of whom consumed at least some amount of alcohol, the advantages from consuming alcohol with food versus without food were very clear and statistically significant.
Previous research has provided some excellent data on the relation between alcohol consumption and many health outcomes. Streppel et al reported that wine consumption, in particular, plays an important role in the health benefits of alcoholic beverages. In their meta-analysis they state: “Light to moderate alcohol intake (<20 g/day) was strongly, inversely, and significantly associated with cerebrovascular (HR=0.43), total cardiovascular (HR=0.70), and all-cause mortality (HR=0.75). Independent of total alcohol intake, long-term wine consumption of up to one half of a glass daily was strongly and inversely associated with coronary heart disease (HR=0.61), total cardiovascular disease (HR=0.68), and all-cause mortality (HR=0.73).” These authors stated that their results could not be explained by differences in socioeconomic status, and concluded that life expectancy was about 5 years longer in men who consumed wine compared with abstainers.
In a large meta-analysis, Li, et al found a U-shaped relation between alcohol intake and risk of diabetes, with the nadir of risk at about 25 g/day; the point estimate of risk of diabetes for subjects reporting 60 g/day was still lower than that of abstainers. Among subjects in the Nurses’ Health study, women reporting > 0.5 drinks/day had approximately 40% lower risk of diabetes than non-drinkers (Hu et al),
While the present paper did not relate alcohol consumption to cardiovascular disease, subjects with diabetes are known to be at a greatly increased risk of coronary heart disease, the leading cause of death throughout the world. In the Health Professional Study, Tanasescu et al reported that the risk of coronary heart disease among adults with type 2 diabetes was reduced in a linear fashion with alcohol intake, with the lowest risk (HR=0.48) among men reporting >2 drinks/day.
Specific comments on the present paper: Reviewer Ellison considered that the large database and analytic methods provided strong support from this study for a beneficial effect of wine with meals. However, Forum member Djoussé stated: “It is an Interesting paper, however, I have serious concerns on the approach as follows:
- Concern about non-comparability of subjects across 3 groups (alcohol intake with meals, outside meals, and not sure) based on Table 1. For example, the 5th and 95th percentiles of total alcohol do not have adequate overlap between extreme groups; similar observations were seen for physical activity and other potential confounders. In the absence of adequate overlap, serious confounding cannot be excluded as a potential explanation for observed results. Simply adding a variable (or variables) without adequate overlap in a multivariable model will not adequately address the confounding issue.
- Data are reported as drinkers of wine, beer, and spirits for beverage-specific analyses, yet the authors do not provide details under alcohol assessment how they defined beverage-specific categories. I doubt that the number reported for wine corresponds to people that drank wine only; it is possible that a larger proportion of their total alcohol intake was provided by wine, but with some amounts from beer and/or spirits. The basis for classification is unclear and thus, the conclusion is a bit misleading if we are looking at people that preferred wine (but consumed to some extent beer and/or spirits in smaller amounts).
Forum member Skovenborg wrote: “I have a comment on a possible association between wine preference and diet quality as a confounding factor. The specific ingredients of the meals ingested with the alcoholic beverages were not collected in this study. Some studies have found that wine drinkers have a healthier diet than people who drink beer or spirits, and variation in diet associated with the preferred drink may explain to some extent why wine has an additional beneficial effect on health. A cross-sectional study of transactions from Danish supermarkets (Johansen et al) found that wine buyers bought more olives, fruit and vegetables, poultry, cooking oil, and low fat cheese, milk, and meat than buyers of beer. Beer buyers bought more ready-cooked dishes, sugar, cold cuts, chips, pork, butter or margarine, sausages, lamb, and soft drinks than wine buyers. Similar observations of a relationship between purchases of beer and the less healthy, traditional-type diet and between purchases of wine and the healthier Mediterranean-type diet was found in a French supermarket study (Hansel et al).
“In an intervention study in which young male volunteers were given Mediterranean or occidental (high-fat) diets accompanied by 240 mL of red wine daily over a 4-week period, total phenols in plasma were elevated after wine consumption (Leighton et al). In that study, total plasma antioxidant capacity (total antioxidant reactivity) increased 28% above basal levels in the Mediterranean diet group, but not in the high-fat diet group. In both groups, wine induced a marked increase in total antioxidant reactivity above basal levels, 56% and 23%, respectively.”
Forum member Estruch followed up with the following: “I fully agree with the comments from Skovenborg regarding this paper. It is unbelievable that a study on alcoholic beverages and risk of type II-diabetes does not take into account the dietary pattern of the participants. In the Predimed trial, we observed that participants who were allocated (at random) to a Mediterranean diet plus extra-virgin olive oil (that also included moderate wine intake) showed a significant reduction in the incidence of type II diabetes compared to those assigned to the low-fat diet: (HR 0.60 [95% CI 0.43 to 0.85] (Salas-Salvador et al). The results were confirmed in a Systematic Review with Metaanalyses performed by Exposito et al.”
Reviewer Ellison noted: “Despite these identified limitations, the overall pattern shown in these analyses support the authors’ conclusions. Further, it is especially interesting to note that adding total alcohol intake and drinking frequency to the equation relating the pattern of drinking and risk of diabetes added very little; the overall hazard ratios are almost the same as when taking only the drinking pattern into consideration. (These findings would indicate that research results based only on total alcohol intake may have serious limitations for setting alcohol drinking policy.)
“The greater health benefits from wine, rather than beer or spirits, shown in these analyses also supports extensive prior research. It is interesting that the biomarkers related to diabetes that were studied (CRP, HbA1c, HDL-cholesterol) were also clearly better for subjects who consumed alcohol with, rather than without, the intake of food. Again, the beneficial associations with wine were much better than those seen with other beverages, even if the methods for assigning drinkers to categories by type of beverage were not described.”
Potential Mechanisms for wine and alcohol’s beneficial effect on the risk of diabetes. The authors of the present paper describe a number of potential mechanisms for the beneficial effects of alcohol on the risk of diabetes. Slowing the absorption of alcohol is one possibility, as peak blood ethanol levels are dramatically lower when a given amount of alcohol is consumed with food in comparison with the same amount of alcohol being given without food (Jones & Jönsson). The authors of the present paper also suggest that consuming alcohol with meals might more quickly and effectively reduce the oxidative stress caused by meals, and oxidative stress is closely related to glucose metabolism, lipid metabolism, and inflammation. As for reasons why wine consumers tend to have greater benefit in this regard, the authors mention the polyphenols in wine, but also state that other lifestyle and social factors among regular moderate wine consumers may play a role, even though they were adjusted for (as far as possible) in the present analyses. It has been pointed out by others that the conclusions would have been better supported if dietary factors could have been included in the analyses.
Reviewer de Gaetano recalls the results of previous studies on potential mechanisms for wine and alcohol’s beneficial effects in relation to meal consumption. “Ursini and his coworkers reported in several papers that consuming the meal with a suitable source of antioxidants, such as those found in red wine, minimizes documented postprandial oxidative stress. The inhibition of peroxidation of lipids present in the meal during digestion is a possible mechanism for the observed protection of LDL. These modified particles could account for a relevant link between nutrition and early biological processes that foster the development of atherosclerosis and diabetes.
“Mattivi’s group performed two different studies showing that consumption of wine with the meal could prevent the postprandial increase in plasma cholesterol oxidation products (Natella et al). In their studies, twelve healthy volunteers consumed the same test meal rich in oxidised and oxidisable lipids (a double cheeseburger), with 300 ml of water (control) or 300 ml of red wine (wine). The postprandial plasma concentration of cholesterol oxidation products was measured by GC-MS. The control meal induced a significant increase in the plasma concentration of lipid hydroperoxides and of two cholesterol oxidation products, 7-β-hydroxycholesterol and 7-ketocholesterol. The postprandial increase in lipid hydroperoxides and cholesterol oxidation products was fully prevented by wine when consumed with the meal.”
De Gaetano continued: “Ramon Estruch and I conducted a randomized, crossover, single-blinded trial to compare the effects of wine and gin on inflammatory biomarkers of atherosclerosis. Forty healthy men (mean age, 37.6 years) consumed 30 g ethanol per day as either wine or gin for 28 days. After either gin or wine consumption, plasma fibrinogen decreased by 5% and 9%, respectively, and cytokine IL-1alpha by 23% and 21%. The expression of LFA-1 (-27%), Mac-1 (-27%), VLA-4 (- 32%) and MCP-1 (-46%) decreased significantly after wine, but not after gin. Wine reduced the serum concentrations of hs-CRP (-21%), VCAM-1 (-17%) and ICAM-1 (-9%). Both wine and gin showed anti-inflammatory effects by reducing plasma fibrinogen and IL-1 alpha levels. However, wine had the additional effect of decreasing C-reactive protein (hs-CRP), as well as a number of monocyte and endothelial adhesion molecules.”
Forum member Hendricks added: “It is surprising that beverage preference makes such a difference in the beneficial effect of moderate drinking with a meal on diabetes type 2 hazard ratios. In intervention studies (reviewed in a.o. Diabetes Care 2015;38:723–732 | DOI: 10.2337/dc14-1556) all three alcoholic beverages were used and compared to their alcohol free counterpart. All gave similar results in that biomarkers for diabetes type 2 risk, as analysed in the study by Ma, showed similar trends, namely CRP and HbA1c being negatively associated and HDL cholesterol positively. I agree with other Forum members that meal quality will make a difference, which in this study was only evaluated as a healthy diet score. Also, the proportion of wine drinkers drinking with meals was substantially larger than the proportion of beer and liquor drinkers drinking with meals, which may explain some of the results.”
Where do we now stand on the relation of alcohol to diabetes? Our current knowledge based on extensive research leads us to the conclusions that
(1) drinkers in general are at a lower risk of developing diabetes than are abstainers (from extensive prior research);
(2) among drinkers, those who consume their alcohol with food have additional benefits in terms of cardiovascular disease, other diseases of ageing, and total mortality (as shown in earlier research) and now, from this study, they also have a lower risk of diabetes; and
(3) the consumption of wine, rather than beer or spirits, with food almost always seems to relate to the strongest beneficial effects on health outcomes.
Overall, despite some limitations noted, the Forum considered this to be an important study. Reviewer Finkel stated: “If there’s one thing we’ve learned in reviewing epidemiological studies it’s that the effects of drinking are not simple. What, how, and when we drink count a great deal, often eclipsing how much. This paper reminds us that drinking wine with meals is a different experience for the body, one much more healthful than naked consumption of, say spirits, without food. The good life has good sense behind it.”
Forum member Ursini added: “It is really a satisfaction hearing that modern epidemiological studies ‘discover’ that wine with food every day is more healthy than spirits without food only on the weekend. However, this notion has a kind of ‘prior art’: Arnoldo of Villanova, indeed, published in 1480 a sort of edited version of Regimen Sanitatis Salernitanum, a didactic poem collecting rules about food and lifestyle to be followed to preserve health. The rules encompassed popular wisdom, folk culture, and medical indication collected and recorded since at least the Second Century B.C. Thus, we have known for many centuries that wine with food is generally not only safe, it is healthy. Not always is epidemiology indispensable for discriminating between what is good and what is bad for you.”
Forum member Parente explored Professor Ursini’s reference to Regimen Sanitatis Salernitanum which she found in Dr. John Ordronaux’s 1871 Latin-English translation, digitized & uploaded from the Harvard University library:
Dr. Parente commented: “Very entertaining and written in verse; for example, on pdf page 86, in On the Preservation of Health, section 38 Method of Eating and Drinking:
‘At meals to sipping only, cling perforce,
And for health’s sake drink not between each course.’
“And later, from an 1861 addition to a Paris edition:
‘Old wine is apt to burn and dessicate,
Make bile, and e’en, ‘tis said, to constipate;
But, sparingly diluted, quickly gives
Fair sustenance to everything that lives.'”
References from Forum critique
Estruch R, Sacanella E, Badia E, Antúnez E, Nicolás JM,Fernández-Solá J. Rotilio E,de Gaetano G, Rubin E, Urbano-Márquez A. Different effects of red wine and gin consumption on inflammatory biomarkers of atherosclerosis: a prospective randomized crossover trial. Effects of wine on inflammatory markers. Atherosclerosis 2004;175:117-123.
Esposito K, Maiorino MI, Bellastella G, et al. A journey into a Mediterranean diet and type 2 diabetes: a systematic review with meta-analyses. Diabetes and Endocrinology. BMJ Open 2015;5:e008222. http://dx.doi.org/10.1136/bmjopen-2015-008222.
Hansel B, Roussel R, Diguet V, et al. Relationships between consumption of alcoholic beverages and healthy foods: the French supermarket cohort of 196,000 subjects. Eur J Prev Cardiol 2015;22:215-222.
Hu FB. Manson JE, Stampfer MJ, et al. Diet, Lifestyle, and the Risk of Type 2 Diabetes Mellitus in Women. N Engl J Med 2001;345:790-797. DOI: 10.1056/NEJMoa010492
Johansen D, Friis K, Skovenborg E, et al. Food buying habits of people who buy wine or beer: cross sectional study. BMJ 2006;332:519-522.
John Ordronaux’s 1871 Latin-English translation of Regimen Sanitatis Salernitanum. Accessed on November 29, 2022, from Harvard University Library website: (https://ia804506.us.archive.org/16/items/codehealthschoo00salegoog/codehealthschoo00salegoog.pdf
Jones AW, Jönsson KA. Food-induced lowering of blood-ethanol profiles and increased rate of elimination immediately after a meal. J Forensic Sci 1994;39:1084-1093.
Leighton F, Cuevas A, Guasch V, Pérez DD, et al. Plasma polyphenols and antioxidants, oxidative DNA damage and endothelial function in a diet and wine intervention study in humans. Drugs Under Exp Clin Res 1999;25:133-141.
Li X-H, Yu F-F, Zhou Y-H, He J. Association between alcohol consumption and the risk of incident type 2 diabetes: a systematic review and dose-response meta-analysis. Am J Clin Nutr 2016;103:818–829, https://doi.org/10.3945/ajcn.115.114389
Natella F, Macone A, Ramberti A, Forte M, Mattivi F, Matarese RM, Scaccini C. Red wine prevents the postprandial increase in plasma cholesterol oxidation products: a pilot study. Br J Nutr 2011;105:1718-1723.
Salas-Salvadó J, Bullo M, Estruch R, et al. Prevention of diabetes with Mediterranean diets: a subgroup analysis of a randomized trial. Ann Intern Med 2014;160:1-10.
Streppel MT, Ocke MC, Boshuizen HCk K9k FJ, Kromhout D. Long-term wine consumption is related to cardiovascular mortality and life expectancy independently of moderate alcohol intake: the Zutphen Study.
Tanasescu M, Hu FB, Willett WC, Stampfer MJ, Rimm EB. Alcohol consumption and risk of coronary heart disease among men with type 2 diabetes mellitus. Am Coll Cardiol 2001;38:1836-1642.
Ursini F, Sevanian A. Postprandial oxidative stress. Biol Chem 2002;383:599-605.
Essentially all previous epidemiologic studies have shown that subjects who consume alcohol are at a lower risk than abstainers of developing Type 2 diabetes mellitus. The present analysis provides new information from more than 300,000 subjects in the UK BioBank who reported that they were consumers of alcohol; it demonstrates how the pattern of consuming alcohol plays an important role in the effect of alcohol on the risk of diabetes. Consuming an alcoholic beverage with food, versus drinking it at other times, is associated with a considerably lower risk of diabetes mellitus. This effect in independent of the amount of alcohol consumed, which suggests that all alcohol policy should be based only on research in which the pattern of drinking is known.
The study also demonstrated that the protective effects against diabetes was greater for consumers of wine rather than for those consuming beer or spirits. These associations are consistent with most previous epidemiologic studies, and now supported from these analysis of a very large cohort.
The authors had very reasonable estimates of alcohol intake as well as good ascertainment of health outcomes. They also had considerable data on many lifestyle and social factors that relate to the development of diabetes, so that confounding and modification of effect by them could be dealt with in the analyses. Forum members noted some limitations to the analytic methods regarding categorization of subjects by the type of alcohol they consumed, and the failure to adjust for diet; however, overall the results appear to support well the conclusions of the authors that there are key health advantages in terms of the risk of developing diabetes of consuming alcoholic beverages (especially wine) with meals, rather than on an empty stomach.
Reference: Ma H, Wang X, Li X, Heianza Y, Qi1 L. Moderate alcohol drinking with meals is related to lower incidence of type 2 diabetes. Am J Clin Nutr 2022;0:1–8.
Comments on this paper by the International Scientific Forum on Alcohol Research were provided by the following members:
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
Luc Djoussé, MD, DSc, Dept. of Medicine, Division of Aging, Brigham & Women’s Hospital and Harvard Medical School, Boston, MA, USA
R. Curtis Ellison, MD, Professor of Medicine, Emeritus; Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
Ramon Estruch, MD, PhD, Hospital Clinic, IDIBAPS, Associate Professor of Medicine, University of Barcelona, Spain
Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University Medical Center, Boston, MA, USA)
Henk Hendriks, PhD, Netherlands
Matilda Parente, MD, consultant in molecular pathology/genetics and emerging technologies, San Diego, CA, USA
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France
Fulvio Ursini MD, Emeritus Professor of Biochemistry, School of medicine, University of Padova, Italy
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South
Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA