Is red wine good for your heart? Should you start drinking?
A red-wine trial changed some blood markers, but that is not proof of heart protection. Israeli and Chinese studies explain why starting for health is unsupported.

Complementary human studies · Published 2015–2019
The idea that a glass of red wine protects the heart has an unusually comfortable story attached to it: a relaxed dinner, a Mediterranean table, and permission to enjoy something pleasant. The difficulty is that a pleasant story and a proven prevention strategy are not the same thing.
You do not need to start drinking to make a diet healthier. To understand why, it helps to separate three questions: can wine change a blood marker, do wine drinkers differ from non-drinkers, and does starting wine actually prevent illness? Evidence for the first two does not automatically answer the third.
What did the red-wine trial actually show?
In the CASCADE trial, adults in Israel with well-controlled type 2 diabetes were assigned red wine, white wine or water with dinner for two years. Everyone received Mediterranean dietary guidance.
The red-wine group had a modestly greater rise in HDL cholesterol than the water group. That is a measurable finding. It is not a count of heart attacks prevented, and the trial did not establish that the overall balance of drinking benefits and harms was favourable for everyone.
Its participants were carefully selected. People with several relevant risk histories were excluded, and the results should not be casually transferred to someone with different medications, a history of addiction, pregnancy or another condition. Even in the selected group, two years and a few hundred people cannot settle all possible long-term harms.
The paper also attracted a published critique about differences between registered and reported outcomes, followed by an editorial response. That reporting dispute is another reason to treat its attractive biomarker results as part of the evidence, rather than a definitive prescription.
Why “wine drinkers live longer” can mislead
People do not randomly choose their lifelong drinking habits. Health, income, social circumstances and other behaviours can influence both drinking and disease risk. Some non-drinkers have stopped because they are already unwell. Comparing these groups can make modest drinking look protective even when the drink is not the cause.
Statistical adjustment helps, but cannot guarantee that every relevant difference has been removed. A headline based on an association should therefore leave room for other explanations instead of becoming “a daily glass prevents heart disease.”
A different test in China
A large China Kadoorie Biobank study followed more than half a million adults and used genetic variants affecting alcohol metabolism to examine alcohol exposure from another angle. The usual observational comparisons appeared to favour moderate drinking; the genetic comparisons did not support that apparent stroke protection.
The genetic analysis associated greater predicted alcohol intake with higher blood pressure and stroke risk. The heart-attack estimate was uncertain rather than a clear demonstration of protection or harm. Participants mainly drank spirits, so this was not a direct trial of red wine.
Those details matter. The study helps challenge a broad alcohol-health story, but it cannot rank every drink or calculate what one specific glass will do to you. Its genetic approach also depends on assumptions about how the variants influence health.
A marker is not the whole health outcome
Measured in the trial
Lipids and glucose-control markers in selected adults with diabetes.
Not established
Fewer heart attacks, strokes or deaths from starting red wine.
Different lens
Chinese genetic evidence challenges apparent stroke protection.
Practical answer
Do not begin drinking for an unproven health benefit.
What about antioxidants and resveratrol?
Red wine contains non-alcoholic compounds, but the presence of an interesting compound does not prove the complete drink improves long-term health. A benefit in a laboratory experiment, a supplement study and a glass at dinner are three different exposures.
You can eat a varied diet containing fruits, vegetables and other plant foods without adding alcohol. The American Heart Association notes that potentially useful plant compounds are found in other foods too. There is no nutritional requirement to obtain them through wine.
Likewise, calling a meal Mediterranean does not make every item on the table essential. The dietary pattern can be discussed without treating an alcoholic drink as the ingredient that unlocks its benefits.
The practical decision
The American Heart Association does not recommend starting alcohol for health benefits. If you already drink, the relevant discussion is your individual risk, amount, pattern and medical context—not whether an article grants wine a healthy-food label.
Someone taking medication or managing a medical condition should ask their clinician about alcohol rather than copying a trial dose. The amount used in a study is a research exposure, not a personalised recommendation.
CASCADE received European Foundation for the Study of Diabetes support; its paper disclosed an author's pharmaceutical relationships outside the study. The Chinese study reported charitable, public, research and pharmaceutical support. Funding does not erase findings, but endpoints, selection and uncertainty still govern what they mean. For the direct question—should you begin red wine for your heart—the evidence does not justify that step.
Sources & further reading
- Israel · CASCADE wine trial (2015)
- China Kadoorie Biobank conventional and genetic evidence
- Published critique of CASCADE outcome reporting
- American Heart Association alcohol guidance
Study years differ from this article’s publication date. We reviewed full primary papers and relevant supporting material; access limitations are kept out of efficacy claims. Original illustrations and diagrams; no publisher figures reproduced. Educational writing, not clinical review.
Published 11 October 2026 · Sources checked 11 October 2026. Suggest a correction.