What decades of wine research can, and cannot, tell us about health

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Aug 25 2026

A new review revisits the J-shaped curve, wine polyphenols, cancer risk, and the modeling choices behind increasingly restrictive alcohol guidelines.

Review: Interpreting the Evidence on Wine Consumption and Health: A Narrative Review of the J-Shaped Association and Contemporary Alcohol Guidelines. Image Credit: Evgeny Karandaev / Shutterstock

In a recent review published in the journal Nutrients, researchers in Australia and Croatia synthesized evidence on wine consumption and health and examined how methodological approaches have influenced the interpretation of the J-shaped association and contemporary alcohol guidelines.

Background

Could a glass of wine have different health implications depending on how much is consumed? Studies have reported a J-shaped association between wine intake and total mortality, with the lowest observed risk at light-to-moderate consumption.

Cardiovascular disease has been a major driver, alongside reported metabolic, endothelial, anti-inflammatory, and antioxidant effects. At the same time, alcohol guidance in countries such as Australia, Canada, and the United Kingdom has become more restrictive, alongside increasing public-health emphasis that no level of alcohol consumption is entirely risk-free.

Wine-specific interpretation is difficult because alcohol, drinking patterns, diet, and lifestyle factors overlap. Research is needed to clarify causal relationships and differences between populations.

Review methodology

A structured literature search was conducted in PubMed/MEDical Literature Analysis and Retrieval System Online (MEDLINE), Scopus, and Web of Science between January and March 2026, primarily covering studies published from 1980 to March 2026 while also including earlier landmark research.

Searches addressed wine and mortality, cardiovascular outcomes, wine polyphenols, resveratrol and hormesis, alcohol-guideline methodology, J-shaped epidemiology, alcohol modeling, and cancer risk.

Only peer-reviewed English-language publications were included. Eligibility was organized using the Population, Intervention/Exposure, Comparator, Outcomes, and Study design (PICOS) framework.

Studies without separable wine data, where beverage-specific evidence was central to the review, conference abstracts, unpublished reports, unsupported opinion articles, single-case reports, and studies focused exclusively on heavy or binge drinking that did not address questions about moderate wine consumption were excluded. 

The review followed the Scale for the Assessment of Narrative Review Articles (SANRA) and did not undertake a formal systematic review or meta-analysis.

Biological and epidemiological evidence

Wine contains ethanol and grape-derived phenolic compounds, including resveratrol, quercetin, catechins, anthocyanins, and procyanidins. Evidence describes hormesis, in which lower exposures can produce effects that differ from those at higher doses.

Wine-related mechanisms include improved endothelial function through greater nitric oxide (NO) bioavailability and endothelial nitric oxide synthase (eNOS) activity, reduced inflammation, lower platelet aggregation, enhanced fibrinolysis, and improved glucose metabolism.

Reported inflammatory effects include reductions in C-reactive protein, interleukin-6, tumor necrosis factor-alpha, and endothelial adhesion molecules. 

The review also suggests that some polyphenols may stimulate endogenous antioxidant defenses through adaptive cellular responses rather than acting mainly as direct free-radical scavengers.

Animal and cellular evidence provides additional biological context. In a study examining four weeks of moderate white wine or ethanol consumption in rats subjected to surgically induced myocardial infarction, survival was highest in the white-wine group, followed by ethanol and water controls.

Wine was also associated with a more favorable inflammatory environment in infarcted cardiac tissue. Resveratrol has demonstrated antioxidant, anti-inflammatory, immunomodulatory, cardioprotective, and anticancer effects in experimental models.

Many experimental studies have used concentrations exceeding those achievable through moderate wine consumption, so direct comparisons require caution. Its effects also show a biphasic, hormetic pattern, with higher concentrations potentially becoming toxic.

Epidemiological evidence has repeatedly reported a J-shaped association between wine consumption and all-cause mortality. Light-to-moderate intake has been associated with the lowest mortality risk, whereas risk rises among heavy drinkers.

The review broadly defines moderate consumption as intake near the lowest point of this curve, generally around 10-30 g of ethanol per day, although this varies between populations and outcomes. 

Moderate wine consumption has also been associated with lower rates of coronary heart disease (CHD), stroke, heart failure, cardiovascular mortality, and type 2 diabetes, as well as better insulin sensitivity and glucose levels.

Studies have reported associations with lower risks of cognitive decline, dementia, and Alzheimer's disease. However, some cardiovascular associations have also been reported with beer and spirits, suggesting that ethanol itself may account for part of the apparent benefit, while any additional wine-specific advantage remains uncertain. 

Observational evidence cannot establish causality. Healthy-user bias, residual confounding, reverse causation, former-drinker bias, exposure misclassification, and differences in lifestyle may contribute to observed associations.

Cancer risk and alcohol guidelines

The review acknowledges that alcohol consumption is causally linked to several cancers, although risk varies by cancer site and level of intake. Evidence is strongest for breast cancer, for which risk appears to increase approximately linearly with alcohol intake.

Associations have also been reported for cancers of the oral cavity, pharynx, larynx, esophagus, colorectum, and liver, particularly at higher consumption levels. Absolute increases at low consumption are generally small but vary by cancer site, age, sex, and baseline risk.

Alcohol accounts for approximately 4-5% of incident cancers and ranks third among modifiable risk factors after tobacco smoking and excess body weight.

Canada’s 2023 guidance recommends no more than two drinks per week for low-risk drinking. Australia’s 2020 guideline set an upper limit of 100 g of ethanol per week, while the United Kingdom’s 2016 guideline used 112 g per week for both sexes.

United States cardiovascular prevention guidance proposes no more than two standard drinks per day for men and one for women. European recommendations generally use lower limits, although national guidance varies. 

The authors also note broad agreement that people should not start drinking alcohol for health benefits and that abstinence is appropriate for those with contraindications.

Methodological and public health considerations

Differences in guidelines may reflect how evidence is modeled as well as the evidence itself. Important choices include which disease endpoints are included, whether wine and other alcoholic beverages are treated as equivalent exposures, whether relative or absolute risk is emphasized, how age and competing risks are considered, and whether drinking patterns are distinguished.

Some models place greater emphasis on alcohol-attributable cancers while giving less weight to cardioprotective outcomes or all-cause mortality. These choices can yield different risk estimates at lower consumption levels. 

The review also considered Mendelian randomization studies, which have challenged causal interpretations of observational associations but may not fully capture beverage type, drinking patterns, or drinking context.

Effects were described as context-dependent, varying by age group and baseline risk. Findings from Mediterranean populations should not be automatically generalized because dietary composition, drinking culture, and lifestyle differ across settings.

Public health strategies should prioritize reducing heavy and episodic drinking while providing nuanced communication about uncertainty, drinking patterns, beverage type, and dietary context.

Conclusion

The evidence reviewed supports the biological plausibility and epidemiological consistency of a reported J-shaped association between moderate wine consumption and several outcomes, particularly cardiovascular disease and all-cause mortality.

Wine’s ethanol and phenolic constituents may influence endothelial function, inflammation, platelet activity, oxidative stress, and metabolism. However, observational findings remain vulnerable to confounding, healthy-user bias, reverse causation, and other limitations, preventing definitive causal conclusions.

Mechanistic and intervention studies are also generally shorter and often rely on surrogate endpoints rather than long-term clinical outcomes. Guideline differences may also reflect modeling choices concerning outcomes, beverage type, drinking patterns, age, and competing risks. 

Future research should integrate epidemiological, clinical, dietary, biomarker, and genetic evidence to clarify population-specific risk-benefit profiles.

Overall, the authors conclude that moderate, regular wine consumption with meals may be compatible with a healthy dietary pattern for some adults, rather than establishing that wine consumption is beneficial or should be recommended.

The review received no external funding. The authors disclosed any current or prior research funding, employment, consultancy work, or professional affiliations with wine- and alcohol-related organizations.