Wine Information Council

Wine Information Council

23 September 2021 In Latest Scientific News 0 comment

The health effects of alcoholic beverages cannot be assessed only by the amount consumed. This is the conclusion of a new study from the UK where the researchers - for the first time - examined how the mortality risk is affected when considering the drinking pattern.

Just to be clear: we are not talking about excessive drinking amounts which - without any doubts - reduce the life expectancy. It is about moderate consumption, which has been associated with a reduced mortality risk in some studies that, unlike other ones, were taking into account the type of alcoholic beverage, the frequency and drinking pattern, and not only assessing the amount of alcohol consumed.

In the UK Biobank study, the type of alcoholic beverage was not considered but for the first time, it was analysed how the drinking pattern affected/influenced the relationship of the drinking amount and mortality risk of more than 300,000 participants of the study.

From the nearly half a million participants (between37 and 73 years) of this observational study, the researchers only selected those who indicated that they consume alcoholic beverages and that did not have coronary heart diseases and cancer. Participants were asked about their drinking habits from where the researchers calculated their weekly consumption, that resulted between 6.4 g and 518.4 g of alcohol per week.

In addition to the amount, the participants indicated whether they consumed the alcoholic beverages regularly and if they were consumed with the meals or not. To better capture their overall drinking habits, the scientists developed a score (Drinking Habits Score) that combined the frequency of alcohol intake and whether it was consumed with the meals. Points were given for each favorable drinking habit: one point for regular consumption (at least 3 times per week) and one for consumption with a meal.

During the 9 year follow-up period, 8600 individuals died and based on their data, the researchers could not only asses the association between mortality and the amount of alcoholic beverages that was consumed weekly but also if and how healthy drinking patterns can affect the mortality risks.


From linear associations to J-shaped curves

The influence of favourable drinking patterns was considerable:

  • When the alcoholic beverages were consumed with a meal, not only the mortality from cardiovascular diseases, cancer and other causes decreased significantly, but also the total mortality (risk of dying from any cause).
  • The relation between the amount of alcohol consumed and the risk of all-cause mortality was different, on the basis of the drinking habits:

When only the drinking amounts were considered, the total mortality risk started to increase significantly above 300 g of alcohol weekly. With healthy drinking patterns – regular drinking and drinking with the meal – a U-shaped association was found, with a consistent and significantly lower risk in participants consuming between 50 g and 300 g per week and no increased risk up to 400 g per week than those who had the lowest amount of alcohol and unfavourable drinking habits.

  • Similar effects were observed for the different specific mortality rates such as for cardiovascular diseases, cancer and from other causes.
  • Also, the cancer mortality was affected by favourable drinking patterns: for the cancer mortality risk, a positive and linear association was observed in participants with unfavourable drinking habits. In contrast, a U-shaped association was seen in individuals with favourable drinking habits, where a moderate intake of alcoholic beverages (50-200g of alcohol/week) was not related to any increased cancer mortality risk.


For the first time, the researchers reported that drinking habits significantly modified the relationship between the actual amount of alcohol intake and the risk of mortality from all causes.

Since it is an observational study, no cause-and-effect relationship can be derived. However, the results are important for the communication of responsible drinking patterns: not only the amount but also the drinking pattern (and the type of alcoholic beverage) are critical for the health effects.


Source: Ma H, Li X, Zhou T, Sun D, Shai I, Heianza Y, Rimm EB, Manson JE, Qi L. Alcohol Consumption Levels as Compared With Drinking Habits in Predicting All-Cause Mortality and Cause-Specific Mortality in Current Drinkers. Mayo Clin Proc. 2021 Jul;96(7):1758-1769. doi: 10.1016/j.mayocp.2021.02.011. PMID: 34218856; PMCID: PMC8262073.

For more information about this abstract, click here.

When results of observational studies show health benefits for moderate wine drinkers and an increased risk for teetotallers (“J curve”), criticism often arises, assuming that methodological limitations might be responsible for the results. Constantly, critics question that a J-shaped relationship between moderate consumption of alcoholic beverages, in particular wine, and various health benefits exists. In some international reports, it is being communicated that any consumption of alcoholic beverages is harmful to health and a complete abstinence is preferable. The association between the consumption of alcoholic beverages and chronic diseases is indeed complex, particularly due to possible bias/confounding with the selection of the reference group, pattern of drinking, type of alcoholic beverages, etc. However, with a careful analysis of the data, it can be reproducibly shown that the known J curve is valid.

One of the most recent study on the matter, an elaborate analysis of data from 16 - mostly European - studies (MORGAM project) with almost 150.000 participants and around 16.000 deaths, represent additional supporting evidence. The MORGAM study examined the relationship between the volume of alcohol consumed and the total and cause-specific mortality, taking into account the drinking pattern and drinking frequency.


Hard endpoints, careful analysis of the data

The international research team chose total mortality as well as cardiovascular and cancer deaths and deaths from other causes as important endpoints. As opposed to using risk factors such as high blood pressure or cholesterol levels as endpoints, these are considered “hard” endpoints because of their high health significance. 

Furthermore, the scientists made sure to carefully subdivide those participants who identified as non-drinkers in two sub-categories: lifetime abstainers and former drinkers. The reason being that lifetime abstainers have a completely different health profile and often also a different lifestyle than former drinkers, who quit drinking alcoholic beverages because of health problems. Not making this distinction leads to the so called “sick quitter effect”, that distorts the results of those studies, which pooled all abstainers. Being a valid point of criticism of older studies, most recent studies are taking care of making this distinction among non-drinkers.


J-curve – once again - confirmed

The results of the current study confirm that a moderate consumption (up to 20 g of alcohol/day) is associated with a lower mortality risk than abstinence while, with a higher consumption, the risk increases: without any doubt, a J curved association. The study also showed that, in comparison with lifetime abstainers, ex-drinkers had a higher mortality risk, proving once again how important it is to make a distinction among non-drinkers in the studies. Furthermore, light to moderate consumption was not associated with cancer mortality, what cannot be said in the case of excessive consumption and with former drinkers, where the risk significantly increased.


Best results for wine drinkers

Such a J-shaped association of alcoholic beverages intake with mortality was observed for all European countries but the magnitude somewhat varied according to countries, which might be an indication of different drinking patterns and beverage preference. In Mediterranean countries, alcoholic beverages are typically consumed during main meals and largely in the form of wine. When the researchers subdivided the data according to beverage preference, among those who preferred drinking wine, such as in Italy and France, considerable differences were observed: the area (of the J curve) of improved life expectancy was much wider than in countries where other alcoholic beverages were consumed (up to approx. 40 g of alcohol per day). The authors concluded that the intake of alcoholic beverages was associated with a reduced mortality risk, resulting more evident in individuals preferring wine. At light to moderate intake (but not zero intake) the balance is in favor of total, cardiovascular and “other cause” mortality risk reduction, without any apparent increased cancer mortality risk.


Di Castelnuovo, A et al., 2021, Alcohol Intake and Total Mortality in 142,960 Individuals from the MORGAM Project: a population-based study. Addiction; doi:10.1111/add.15593

For more information about this abstract, click here.

This UK study attempts to explain the inequalities in alcohol-related harm among lower socio-economic individuals.

Alcohol-related harm was found to be higher in disadvantaged groups, despite similar alcohol consumption as advantaged groups. This is known as the alcohol harm paradox.

This study investigated whether and to what extent individual alcohol consumption by type of beverage, smoking, BMI and other factors could account for alcohol-related hospital admissions. In addition, it was examined how the patterns of consumption by beverage type differed by socioeconomic group.

The results showed that the beverage type could not explain the inequalities in alcohol-related harm. Smoking and the body weight partly explained some of these differences, but deprived groups still had a persistently higher risk of alcohol-related hospital admissions. Deprived drinkers drank more beer and in most age groups more spirits, but less wine compared to less deprived drinkers. Furthermore, the hospital admissions were greater for spirits consumers and occurred primarily in the youngest age group (age 16-29).


Gartner A et al., Drinking beer, wine or spirits – does it matter for inequalities in alcohol-related hospital admission? A record-linked longitudinal study in Wales, BMC 2019;19:1651, doi:10.1186/s12889-019-8015-3.

In this prospective study, the risk of chronic kidney disease was lower among moderate drinkers compared to non-drinkers.

Because coronary heart disease (CHD) and chronic kidney disease (CKD) share many risk factors, it is possible that the moderate consumption of alcoholic beverages may also reduce the risk of CKD. However, the relationship is complex as alcohol increases blood pressure and may thus increase the risk of hypertension which is a major risk factor for CKD. On the other hand, alcohol may be protective of CHD, heart attacks and diabetes which are major risk factors of CKD.

In this prospective analysis which studied the 12,000 participants of the Atherosclerosis Risk in Communities (ARIC) study, both their consumption of alcoholic beverages and their estimated glomerular filtration rate kidneys were assessed. During an average follow-up of 24 years, 3,664 cases of CKD were observed.

The results show that moderate drinkers (from < 1 drink to up to 15 drinks per week) had a lower risk of chronic kidney disease compared to never drinkers after adjusting for potential confounding factors.

The mechanisms for this inverse association are not fully understood but the authors explain that the mechanism of alcohol on CKD may be similar to the mechanism between alcohol and CHD since they share similar pathways. It has been proposed that the “good” HDL cholesterol may increase the transport rate of lipoproteins. Since low HDL cholesterol levels may increase the risk of renal dysfunction, the inverse relationship between alcohol and CKD may be mediated by HDL cholesterol.


1 drink = 4-ounce of wine, 12-ounce of beer, 1 ½ ounce shot of hard liquor


Hu E, Alcohol consumption and incident kidney disease: Results from the Atherosclerosis Risk in communities Study,J of Renal Nutrition 2019: 1-9.

For more information about this article, read the scientific abstract here. 


Page 6 of 17


The authors have taken reasonable care in ensuring the accuracy of the information herein at the time of publication and are not responsible for any errors or omissions. Read more on our disclaimer and Privacy Policy.