Alcohol and Your Heart: A San Diego Cardiologist's Honest Look at the Evidence

For most of my career, patients have asked me a version of the same question: "Is a glass of wine with dinner good for my heart?" And for most of my career, the honest answer was "probably, in moderation." That answer has changed. The 2025 American Heart Association (AHA) scientific statement on alcohol and cardiovascular disease represents a real shift in how cardiologists talk about drinking. I want to walk you through what the best available evidence now says, where the older "J-curve" belief came from, why it has weakened under closer scrutiny, and what I tell my patients in clinic in Encinitas today.

This is a long read because the topic deserves nuance. Alcohol interacts with blood pressure, atrial fibrillation, coronary artery disease, heart failure, stroke, and overall mortality in ways that are not identical. A single guideline number cannot capture that. I will give you the numbers where the evidence supports them and the uncertainty where it does not.

The Short Version

If you are pressed for time, here is what I want you to take away. Mendelian randomization studies, which use genetic variation as a natural experiment, suggest that alcohol has harmful or neutral effects on cardiovascular disease at every consumption level. The older observational "J-curve" showing a benefit at one to two drinks a day appears to reflect the fact that light drinkers tend to have healthier lifestyles, not that alcohol is protective. Alcohol raises blood pressure in a dose-dependent way starting around 12 grams per day (one drink). It increases the risk of atrial fibrillation with each additional drink. Heavy drinking (at least three drinks a day or any binge pattern) worsens outcomes across every cardiovascular entity studied. No major cardiology society now recommends drinking for heart benefit, and the 2025 AHA statement explicitly says clinicians should reinforce physical activity, avoiding tobacco, and maintaining healthy body weight rather than endorse alcohol.

Where the "J-Curve" Came From

Starting in the late 1980s and through the 1990s, large observational cohort studies reported a curious pattern. When researchers plotted alcohol intake on the x-axis and cardiovascular mortality on the y-axis, the curve dipped in the middle. Light to moderate drinkers, usually defined as about one drink per day for women and up to two for men, appeared to have lower rates of coronary events and lower all-cause mortality than both heavy drinkers and people who did not drink at all. This pattern was nicknamed the J-curve because of its shape, and it was reinforced by mechanistic studies showing that alcohol raised HDL cholesterol and modestly affected platelet aggregation.

The J-curve quickly became common knowledge. Guidelines from that era avoided endorsing drinking outright but acknowledged that "moderate consumption" appeared associated with lower heart risk. Patients took the nuance and collapsed it. I still have patients in their seventies who tell me their doctor advised them to "have a glass of red wine" for their heart. They are not misremembering. That was the culture of the time.

Why the J-Curve Is Weakening

The problem with the original J-curve studies was never the statistics. It was the comparison group. When you ask a large population "how many drinks per week do you have," the people who answer "zero" are not a random sample of healthy nondrinkers. That group is a mix of lifelong abstainers, former heavy drinkers who quit because they got sick, people with active illness who were told to stop, and people on medications that forbid alcohol. In other words, the nondrinker reference group was skewed toward worse baseline health. Light drinkers, by contrast, tend to have higher income, better diets, more physical activity, and access to more preventive care. When you compare light drinkers to a sick reference group, light drinkers look healthier. That does not mean alcohol made them healthier.

Two streams of newer research have changed the picture. The first is individual-level meta-analyses that correct for misclassification of former drinkers as "never drinkers." When those corrections are applied, the apparent benefit at low doses shrinks or disappears. The second, and more persuasive, is Mendelian randomization (MR). Certain genetic variants, especially in the ALDH2 and ADH1B genes that govern alcohol metabolism, strongly influence how much a person drinks across a lifetime. These variants are not linked to socioeconomic status, diet, or exercise. By comparing outcomes across people grouped by these genetic variants, researchers approximate a natural randomized experiment.

A landmark 2022 JAMA Network Open study by Biddinger and colleagues used this method in nearly 372,000 UK Biobank participants. It found no causal protective effect of light or moderate drinking for coronary heart disease or myocardial infarction. Every level of alcohol intake above abstinence was associated with higher cardiovascular risk, and risk rose steeply past moderate amounts. Subsequent Mendelian randomization work across European, African American, and Hispanic American populations has reinforced the same pattern and found no causal association between alcohol consumption and reduced cardiometabolic disease risk.

The 2025 AHA scientific statement synthesizes both streams. It concludes that while observational studies still suggest possible risk reduction with one to two drinks per day for coronary disease and possibly heart failure, the current evidence challenges the J-curve hypothesis, and clinicians should not recommend alcohol for cardiovascular protection. That is a significant change in posture from statements a decade ago.

What Counts as a "Drink"

Before I go further, I need to pin down the unit. In the United States, a standard drink is about 14 grams of pure alcohol. That is 12 ounces of regular beer at about 5 percent alcohol, 5 ounces of wine at about 12 percent, or 1.5 ounces of 80-proof spirits. Those pours are smaller than most people serve at home. A generous glass of wine at dinner is often closer to 8 ounces, which is one and a half to two standard drinks. A strong craft beer at 8 percent alcohol delivers about 1.6 standard drinks in a 12-ounce can. Almost every patient I see underestimates their weekly intake. When we count honestly, the numbers climb.

Alcohol and Blood Pressure

The effect of alcohol on blood pressure is the clearest of all the cardiovascular effects. It is dose-dependent, reproducible across studies, and begins at amounts people consider moderate.

In a meta-analysis of cohort studies cited by the 2025 AHA statement, people consuming one drink per day had systolic blood pressure 1.25 mmHg higher than nondrinkers, and those consuming three drinks per day had systolic pressure about 4.9 mmHg higher. The association starts to appear above approximately 12 grams per day, or one drink. The relationship is nearly linear in men and shows a threshold effect in women, with steeper increases above 12 grams per day.

In clinic, I see two common patterns. The first is a patient in their fifties with "stubborn" hypertension on two or three medications, and when we review an honest drinking history we find 15 to 25 drinks a week. When they cut back to five or fewer drinks per week for eight weeks, systolic pressure often drops 5 to 10 mmHg without any medication change. The second is a patient who drinks only on weekends but does so heavily. Weekday readings look great and weekend readings are awful, and ambulatory monitoring shows alcohol-related surges that can be mistaken for white-coat hypertension.

If you have hypertension and drink more than a few times a week, reducing alcohol is one of the most reliably effective lifestyle changes you can make for blood pressure. It compares favorably with most single medications in terms of the millimeter-of-mercury impact.

Alcohol and Atrial Fibrillation

Alcohol causes atrial fibrillation (AF). It does so through multiple mechanisms: direct electrical effects on atrial tissue, neurohormonal activation through the sympathetic nervous system, sleep disruption that further destabilizes the atria, and over years structural changes including atrial enlargement and fibrosis. "Holiday heart syndrome" (AF triggered by binge drinking) has been recognized for decades. What has sharpened more recently is how low the chronic threshold is.

Each additional drink per day increases AF risk by approximately 8 percent in a dose-dependent fashion. Binge drinking patterns are particularly important: in pooled analyses, moderate drinkers (less than 30 grams per day) who also had heavy episodic (binge) drinking episodes had a relative risk of 1.12, while moderate drinkers without binge episodes had a pooled relative risk of 0.64. In other words, how you drink matters as much as how much you drink on average. A single binge can undo the "averaging" that makes moderate intake look safer on paper.

What I tell my AF patients is simple. Reducing alcohol, and ideally eliminating binge patterns, is one of the highest-yield interventions we have for keeping you in sinus rhythm. It is every bit as important as your rate or rhythm medication, and often more so.

Alcohol and Coronary Artery Disease

This is where the J-curve lived longest and where the newer data have been most disruptive. The original premise was that moderate alcohol raised HDL, reduced platelet stickiness, and thereby lowered heart-attack risk. Observational studies for decades supported it.

Mendelian randomization has weakened that story substantially. Biddinger's 2022 analysis found no causal protective effect at low doses and rising risk at higher levels. Observational studies continue to report a J-curve, but MR and corrected meta-analyses suggest those apparent benefits reflect confounding rather than a real biologic effect. The 2025 AHA statement reflects this shift: the cardiovascular effects at very low levels are at best uncertain, and at moderate-to-heavy levels they are clearly net harmful. The idea that moderate drinking should be counted in the same column as exercise or Mediterranean-style eating has been retired.

If you are at elevated risk for coronary disease or already have it, I would not start drinking for protection. If you already drink at very low levels, it is unlikely to be the dominant driver of your cardiac risk. Your LDL, blood pressure, blood sugar, smoking status, and activity matter far more.

Alcohol and Heart Failure

The heart failure picture has important thresholds. At one to two drinks per day, observational data do not show increased heart failure risk. At 21 drinks per week or more, heart failure risk rises by roughly 50 percent. Classic alcoholic cardiomyopathy develops after consuming seven to fifteen drinks per day over five to fifteen years, though susceptibility varies with genetics and nutrition. Recent data suggest even four drinks per week may increase diastolic dysfunction in some populations.

The most striking finding applies to patients who already have structural or functional cardiac abnormalities (for example, prior MI with reduced ejection fraction, or left ventricular hypertrophy). In that group, as little as five drinks per week is associated with a five-fold increased risk of progression to symptomatic heart failure. If your heart is already vulnerable, the safety margin is much narrower.

Alcoholic cardiomyopathy can partially or fully reverse with sustained abstinence. I have watched ejection fractions climb from the low 20s back to the mid 40s over a year of sobriety combined with guideline-directed medical therapy. For patients with established heart failure, current practice is to recommend minimal or no alcohol.

Alcohol and Stroke

Stroke risk with alcohol is dose-dependent and type-specific. Heavy consumption (more than four drinks per day) consistently increases the risk of all stroke types. Among current drinkers, each additional 100 grams of alcohol per week (roughly one drink per day) is associated with a 13 percent increase in ischemic stroke risk, a 17 percent increase in intracerebral hemorrhage, and a 9 percent increase in subarachnoid hemorrhage.

Mendelian randomization studies have confirmed positive associations between genetically predicted alcohol consumption and stroke risk. The evidence for any protective effect at low-to-moderate intake is insufficient and conflicting.

Alcohol-related atrial fibrillation is a separate stroke pathway. If you develop AF and do not know it (which happens often), you are at risk for embolic stroke. AF can raise stroke risk several-fold depending on your CHA2DS2-VASc score. If alcohol is contributing to AF, it is indirectly contributing to stroke risk as well.

Alcohol and All-Cause Mortality

Traditional observational studies have shown a J-curve for all-cause mortality with a nadir at one to two drinks per day. Mendelian randomization challenges this, finding no causal association between low-to-moderate alcohol consumption and reduced mortality once confounding is handled. The 2025 AHA statement now takes the position that it remains unknown whether any level of drinking is part of a healthy lifestyle.

Cancer is part of the mortality picture too. Alcohol is a Group 1 carcinogen and contributes to cancers of the oropharynx, esophagus, larynx, liver, colon, and female breast. Even at one drink a day, breast cancer risk rises by several percent. For a cardiovascular patient also thinking about cancer, this is relevant context.

So What Is a Reasonable Threshold?

The 2025 AHA statement does not set a specific numeric upper limit because the writing group felt the evidence was not strong enough. What it does is retire the claim that moderate drinking is cardioprotective, emphasize that clinicians should reinforce established healthy behaviors (physical activity, not smoking, weight, diet) rather than alcohol, and flag binge drinking and average intake of three or more drinks per day as consistently associated with worse outcomes across every cardiovascular entity studied.

My practical framing, keeping cardiology in mind, is this. Fewer than three drinks per week carries a small cardiovascular risk premium but is unlikely to be a dominant driver of your overall heart risk if your other numbers are controlled. Three to seven drinks per week begins to show measurable effects on blood pressure and AF. More than seven drinks per week, and especially more than fourteen, is associated with meaningful cardiovascular harm. Binge drinking, typically defined as four or more drinks in two hours for women and five or more for men, is particularly dangerous regardless of weekly average because of the acute AF and stroke risk.

What I Actually Say in Clinic

I do not moralize about alcohol. Patients have their own reasons for drinking and their own relationships with it. My job is to be accurate about the tradeoffs.

For a patient with new atrial fibrillation, I am direct. Cutting alcohol to three or fewer drinks per week, and ideally eliminating binge episodes entirely, will do more for your rhythm than most of the pills we can prescribe. If you value staying in sinus rhythm, this is a high-leverage lifestyle change.

For a patient with hypertension, especially resistant hypertension on multiple medications, I ask detailed questions about drinking and suggest a six- to eight-week trial of fewer than three drinks per week with home BP monitoring. The drop is often substantial, consistent with the dose-response numbers above.

For a patient with coronary disease, I do not ask them to stop drinking if they are at very low levels. But I do not endorse drinking as cardioprotective, and I make clear that LDL, weight, blood pressure, blood sugar, and exercise matter far more.

For a patient with any structural heart disease or heart failure, the five-fold progression signal at five or more drinks per week gets my full attention. I recommend minimal to no alcohol.

For a patient who does not currently drink and asks whether they should start for their heart, the answer is no. There is no credible cardiology evidence that initiating alcohol improves your outcomes.

Red Wine, Resveratrol, and the Mediterranean Diet

A natural follow-up question is whether red wine specifically offers protection that other alcoholic drinks do not, given its association with the Mediterranean diet. The evidence has always been thin. Resveratrol, the polyphenol often cited for red wine's benefits, is present at such small concentrations in a glass of wine that you would need to drink roughly 40 liters per day to approach the doses used in animal studies. The Mediterranean diet is associated with lower cardiovascular risk, but the benefit comes from olive oil, nuts, vegetables, fish, and whole grains. Any residual "red wine is healthy" story should be treated as marketing more than medicine.

Alcohol, Sleep, and the Heart

One underappreciated path by which alcohol damages heart health is through sleep. Even a single drink within three hours of bedtime reduces REM sleep, raises nocturnal heart rate, increases sleep-related blood pressure, and worsens obstructive sleep apnea by relaxing upper airway muscles. Wearable heart-rate data from large cohorts show that even "moderate" drinkers have persistently higher nocturnal heart rates than nondrinkers, and reduced resting heart rate variability. Over years, poor sleep architecture and untreated or worsened sleep apnea contribute to hypertension, AF, and cardiovascular mortality. When I counsel patients on reducing alcohol, I often frame the immediate win not as decades-out cancer risk but as the sleep and resting heart rate they can feel within two weeks.

Special Populations

Patients on anticoagulation, whether warfarin or a DOAC such as apixaban or rivaroxaban, should drink minimally. Alcohol affects warfarin metabolism and can swing the INR unpredictably. For DOAC users, alcohol raises bleeding risk including GI bleeding and intracranial hemorrhage, especially in patients over 75 or those prone to falls.

Patients with prior alcoholic cardiomyopathy or alcohol use disorder should not drink at all; reversal of cardiomyopathy depends on sustained abstinence. Patients on amiodarone, statins, or regular acetaminophen should be careful about hepatotoxic combinations, especially at higher intakes.

Women metabolize alcohol differently than men because of lower body water content and lower gastric alcohol dehydrogenase activity. Blood alcohol levels for the same number of drinks run higher in women, and cardiovascular and cancer effects track at lower doses. This is reflected in the threshold effect women show for blood pressure above 12 grams per day, with steeper slopes than men.

Cutting Back: Practical Strategies

For patients who want to reduce their drinking, I suggest a few anchors. Track your drinks for two weeks before trying to change anything, because the awareness alone often reduces intake by 20 to 30 percent. Set a weekly budget, not a daily one, because daily limits get broken more easily and weekly targets give flexibility. Build in at least three alcohol-free days per week; this alone correlates with meaningfully lower AF and blood pressure effects. When you do drink, eat first and hydrate alongside. Alternating water with alcoholic drinks reduces the total consumed and blunts the cardiovascular effects. Consider substitutes: non-alcoholic beer and spirits have grown into a robust category and many patients find them acceptable at social occasions.

If you suspect your drinking is beyond what you can adjust on your own, talk to your primary care physician or your cardiologist. Medications such as naltrexone and acamprosate are effective and underused. Structured programs, whether through a therapist, Alcoholics Anonymous, or SMART Recovery, help many people. There is no virtue in struggling alone.

Bottom Line From My Clinic

If you are a patient in my practice and you ask me today, "Should I drink for my heart?" I will say no. If you ask me, "I enjoy a glass of wine a few times a week. Is that OK?" I will say it is a small cardiovascular risk you are accepting, probably not decisive if your other numbers are good, but I will not tell you it is protective. If you ask me, "I have AF and I want to stay in sinus rhythm," cutting alcohol to three or fewer drinks a week, and ideally eliminating binge patterns, is among the highest-yield moves you can make. If you ask me, "I have high blood pressure on three medications and I drink most days," I will ask you to try eight weeks under three drinks a week and check your pressure at home. If you ask me, "I have structural heart disease or heart failure," I will recommend little to no alcohol given the five-fold progression signal in that group.

The evidence has moved. The old comfortable answer, "a glass a day for your heart," does not survive contact with modern methods. That is hard to hear, because wine with dinner is tied to meaningful social and emotional rituals for many people. I respect that. My job is to give you accurate data so that when you make your choices, you do so with your eyes open.

References

Piano, Mariann R., Gregory M. Marcus, Dana M. Aycock, et al. "Alcohol Use and Cardiovascular Disease: A Scientific Statement From the American Heart Association." Circulation (2025).

Biddinger, Kiran J., Connor A. Emdin, Mary E. Haas, et al. "Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease." JAMA Network Open 5, no. 3 (2022): e223849.

"Alcohol Intake and Risk of Hypertension: A Systematic Review and Dose-response Meta-analysis." Hypertension (meta-analysis cited in 2025 AHA Scientific Statement).

Voskoboinik, Aleksandr, Jonathan M. Kalman, Alex De Silva, et al. "Alcohol Abstinence in Drinkers With Atrial Fibrillation." New England Journal of Medicine 382 (2020): 20–28.

Van Gelder, Isabelle C., Michiel Rienstra, Karina V. Bunting, et al. "2024 ESC Guidelines for the Management of Atrial Fibrillation." European Heart Journal 45, no. 36 (2024): 3314–3414.

Joglar, Jose A., Mina K. Chung, Anastasia L. Armbruster, et al. "2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation." Circulation 149, no. 1 (2024): e1–e156.

Mendelian randomization evidence: Alcohol, Clinical & Experimental Research (multi-population MR study cited in 2025 AHA Scientific Statement).

Published on damianrasch.com. This article is for educational purposes and is not medical advice. Always discuss your individual situation with your physician.