Things to Remember
-
The "wine is good for your heart" advice is outdated: For years, doctors said moderate drinking might protect your heart, but newer research shows this was flawed. Many "non-drinkers" in those studies had actually quit because they were already sick, making moderate drinkers look healthier by comparison. When scientists account for this, the heart benefits largely disappear.
-
Alcohol is a proven cancer-causing substance: The WHO classified alcohol the same way as tobacco back in 1988. When you drink, your liver breaks alcohol down into a toxic chemical called acetaldehyde that damages your DNA - this is what increases cancer risk, especially for breast, throat, esophageal, liver, and colon cancers.
-
Even light drinking increases breast cancer risk: Just one drink per day increases breast cancer risk by about 7-10%. This matters especially if you already have risk factors like family history or genetic mutations (BRCA genes). The risk builds up over years of drinking.
-
Recent government reports are giving mixed messages: Three major health reports came out in late 2024/early 2025 with slightly different conclusions, which is confusing. The clearest message: there's no truly "safe" amount of alcohol when it comes to cancer risk, even if cardiovascular effects are less clear-cut.
-
"Moderate drinking" might not mean what you think: One standard drink is a 12 oz beer, 5 oz glass of wine, or 1.5 oz of liquor. Many people dramatically underestimate how much they're actually drinking - pouring larger servings or not realizing that craft beers often have higher alcohol content.
-
Bottom line: Your doctor can't give you a simple "yes, it's safe" anymore. The decision to drink is about weighing personal risks (especially cancer) against enjoyment, and understanding that any amount carries some risk - especially as it adds up over time.
This article examines what recent research actually shows about alcohol's health effects and why decades-old advice about "safe" drinking levels is being reconsidered.
Someone asked me last week if having a glass of wine with dinner every night was okay. Before I could answer, they added: "My doctor said red wine is good for the heart."
Alcohol Consumption Levels and Associated Cancer Risks
| Consumption Level | Amount Per Day | Cardiovascular Impact | Cancer Risk Increase | Key Cancers Affected |
|---|---|---|---|---|
| None (Abstainer) | 0 drinks | No alcohol-related risk | Baseline (0%) | N/A |
| Light Drinking | <1 drink (women) 1-2 drinks (men) |
Previously thought protective; now shown neutral or harmful |
7-10% per drink | Breast, colorectal, oropharyngeal |
| Moderate Drinking | 1-2 drinks (women) 2-3 drinks (men) |
No cardioprotective benefit when biases removed |
20-30% increase | Breast, liver, esophageal, colorectal |
| Heavy Drinking | 3+ drinks (women) 4+ drinks (men) |
Increased hypertension, stroke, cardiomyopathy |
50%+ increase | All seven alcohol-related cancers significantly elevated |
Note: One standard drink = 14g alcohol (5 oz wine, 12 oz beer, 1.5 oz spirits)
The 7 Cancers Causally Linked to Alcohol Consumption
- Breast Cancer
- Risk level: Increased even at <1 drink/day
- Mechanism: Estrogen metabolism disruption, acetaldehyde DNA damage
-
Who's most affected: Women with BRCA1/2 mutations or family history
-
Liver Cancer
- Risk level: Dramatically elevated with heavy use
- Mechanism: Cirrhosis progression, chronic inflammation
-
Who's most affected: Those with hepatitis B/C, fatty liver disease
-
Colorectal Cancer
- Risk level: 20% increase at moderate consumption
- Mechanism: Acetaldehyde exposure to gut lining, folate depletion
-
Who's most affected: Men over 50, those with inflammatory bowel disease
-
Esophageal Cancer
- Risk level: One of the strongest associations
- Mechanism: Direct acetaldehyde contact with esophageal tissue
-
Who's most affected: People with ALDH2 deficiency (common in East Asians)
-
Oropharyngeal Cancer (mouth and throat)
- Risk level: Synergistic with tobacco use
- Mechanism: Local acetaldehyde exposure, immune suppression
-
Who's most affected: Smokers who also drink
-
Laryngeal Cancer
- Risk level: Dose-dependent increase
- Mechanism: Direct tissue exposure, chronic inflammation
-
Who's most affected: Heavy drinkers, especially smokers
-
Stomach Cancer
- Risk level: Moderate at heavy consumption
- Mechanism: Gastric lining damage, H. pylori interaction
- Who's most affected: Those with chronic gastritis or H. pylori infection
I paused. Not because I didn't know what to say, but because I was watching in real time how deeply that old story - alcohol is good for you - has embedded itself in how we think about health.
The truth is more complicated. And less reassuring.
What Changed (And What We're Still Catching Up To)
For decades, the medical story around alcohol was relatively straightforward: a little is fine, maybe even beneficial. The French Paradox - how could a population eating butter and cream have such low heart disease rates? - was explained away with red wine and resveratrol. Moderate drinking, we were told, might lower cardiovascular risk. Two drinks for men, one for women. Stay within those bounds and you're probably helping your heart.
That advice is unraveling.
Not because the data was fabricated, but because we're understanding it differently now. The early observational studies that suggested moderate drinking reduced heart disease risk had a problem: they compared drinkers to non-drinkers. But who are the non-drinkers? Some are lifelong abstainers in excellent health. But many are people who stopped drinking - often because they already had serious health problems. When you compare moderate drinkers to a group that includes people who quit alcohol due to liver disease, heart failure, or cancer, of course the moderate drinkers look healthier.
It's a bias we call the "sick-quitter effect." And once you account for it, the cardioprotective benefit of alcohol starts to disappear.
A 2023 meta-analysis in JAMA Network Open reanalyzed decades of alcohol research, controlling for these biases. The conclusion: there is no safe level of alcohol consumption when it comes to overall mortality. Not zero drinks. Not one drink. The J-shaped curve we used to draw - where mortality risk dipped slightly for moderate drinkers - flattens out. Or worse, it doesn't dip at all.
At the same time, evidence linking alcohol to cancer has been accumulating for years. Not new evidence - clearer evidence. The kind that's harder to argue with.
The Cancer Link We've Been Ignoring
Alcohol is a Group 1 carcinogen. The World Health Organization classified it that way in 1988 - the same category as tobacco and asbestos. But unlike tobacco, we've been slow to talk about it publicly.
Ethanol itself isn't the problem - it's what happens after you drink it. Your liver breaks ethanol down into acetaldehyde, a toxic compound that damages DNA and interferes with cellular repair. Acetaldehyde is also a Group 1 carcinogen. So when you drink alcohol, you're essentially introducing a DNA-damaging agent into your bloodstream, and your body scrambles to metabolize it before it does too much harm.
Sometimes, it doesn't scramble fast enough.
Alcohol consumption is causally linked to at least seven types of cancer: oropharyngeal, laryngeal, esophageal, liver, colorectal, and - most notably for the conversation we're having now - breast cancer.
Breast cancer is where the data gets uncomfortable. Even low levels of alcohol consumption - less than one drink per day - are associated with an increased risk. A 2023 systematic review published in The Lancet Oncology found that each 10 grams of alcohol per day (roughly one standard drink) increases breast cancer risk by about 7-10%. For women already at elevated risk due to family history or genetic factors like BRCA1/2 mutations, the increase is steeper.
It's not a huge risk. But it's not negligible either. And it accumulates over time.
The Reports That Tried to Say This Out Loud
In late 2024 and early 2025, three major reports were published within a month of each other, all attempting to update public health guidance on alcohol. They didn't fully agree, which created confusion.
The National Academy of Sciences report, released December 17, 2024, found that drinking less than two drinks per day for men (or one for women) was associated with lower all-cause mortality and cardiovascular mortality - "with moderate certainty." But it also acknowledged a 10% increased risk of breast cancer at those levels.
On January 3, 2025, the U.S. Surgeon General issued a report emphasizing the causal link between alcohol and cancer, particularly breast cancer, even at low levels of consumption. The message was clearer and less equivocal: there is no truly safe threshold for alcohol when cancer risk is considered.
Eleven days later, the HHS Interagency Coordinating Committee on the Prevention of Underage Drinking (ICCPUD) released its own report, which also highlighted cancer risks and recommended against starting to drink for health reasons.
Why the discrepancies? Partly methodological - different ways of analyzing risk, different populations studied, different endpoints prioritized (mortality vs. cancer incidence vs. cardiovascular events). But also philosophical: how much risk are we willing to accept? And what counts as "low risk" when the risk is cancer?
I think about this when people ask me about drinking. They want a clean answer: is it safe or not? But the real answer is: it depends on what you're willing to risk, and what you value most.
What "Moderate" Drinking Actually Means (And Why It Matters)
I saw someone recently who was confused about whether their drinking was "moderate." Their partner had told them it was fine - two tequilas and six beers most nights. They genuinely believed that fell within the guidelines.
It doesn't. Not even close.
Standard drink definitions vary slightly by country, but in the U.S., one standard drink is:
- 12 ounces of beer (5% alcohol)
- 5 ounces of wine (12% alcohol)
- 1.5 ounces of distilled spirits (40% alcohol)
Each contains roughly 14 grams of pure ethanol.
The old "moderate drinking" guidelines suggested up to two drinks per day for men, one for women. But those guidelines were based on cardiovascular risk reduction - a benefit that, as I mentioned, is now being questioned. And they never accounted for cancer risk, which increases linearly. There's no threshold below which cancer risk disappears. Even one drink per day nudges the risk upward.
Canada updated its alcohol guidance in 2023, lowering recommended limits significantly. Their new advice: no more than two standard drinks per week for low risk. Not per day. Per week. Anything above that increases cancer and mortality risk in measurable ways.
Other countries are following suit. Ireland and South Korea now require cancer warnings on alcohol labels - similar to tobacco warnings. The evidence that such warnings reduce consumption is solid. A 2024 study in Lancet Public Health found that clear, visible warnings reduced alcohol purchases by 7-12% in the months following implementation.
But in the U.S., we're not there yet. The warning labels we do have - small print about birth defects and impaired driving - don't mention cancer. Most people still don't know alcohol is a carcinogen.
Why We're So Reluctant to Talk About This
Alcohol is woven into the social fabric in ways tobacco never was. We celebrate with it. We relax with it. We bond over it. To say "alcohol causes cancer" feels like saying "celebrations cause cancer," which sounds absurd even though it's technically true.
There's also a cultural narrative - rooted in decades of mixed messages from medicine - that moderate drinking is healthy. The French Paradox. The Mediterranean diet. Red wine and longevity. These stories are hard to let go of, even when the evidence shifts.
And then there's the question of magnitude. If you're a 40-year-old woman with no family history of breast cancer, having two glasses of wine per week might increase your lifetime breast cancer risk by a fraction of a percent. That's a real increase. But is it enough to change your behavior? For some people, yes. For others, no.
This is where medicine gets messy. We can quantify risk, but we can't tell you how to feel about that risk. A 7% increase in breast cancer risk might sound trivial or terrifying depending on who you are, what you've experienced, and what you value.
What I Tell People Now
I don't tell people to stop drinking. That's not my job. But I do tell them what we know.
Alcohol increases cancer risk, even at low levels. That risk is small for any individual drink, but it accumulates over time. If you drink daily, the cumulative effect matters. If you drink heavily - more than two drinks per day - the risks escalate sharply: liver disease, pancreatitis, cardiomyopathy, neurological damage, and significantly higher cancer rates.
The cardiovascular benefits we used to talk about? They're either smaller than we thought, or they don't exist once you correct for confounding factors. No major health organization currently recommends starting to drink for health reasons. If you don't drink, don't start.
If you do drink, the question becomes: how much risk are you comfortable with?
For some people, that means cutting back significantly. For others, it means being more intentional - drinking less frequently, choosing not to drink at all on certain days, or setting clearer limits. And for a few, it means quitting entirely, especially if there's a family history of cancer or liver disease.
I also talk about what counts as a drink. Because the gap between what people think is moderate and what actually qualifies can be enormous. A pint of craft beer (7-9% alcohol) is often 1.5 to 2 standard drinks. A large glass of wine poured at home can easily be 2-3 drinks. People underestimate their consumption more often than they overestimate it.
The Bigger Picture: Public Health and Personal Choice
Public health guidance is always a balance between individual autonomy and collective risk. We don't ban alcohol. We don't criminalize drinking. But we do try to make sure people understand what they're choosing.
The Surgeon General's report wasn't calling for prohibition. It was calling for better information. Warning labels. Clearer guidelines. More honest conversations between doctors and patients.
That last part is hard. It's easier to say "a glass of wine is fine" than to explain the nuances of cancer risk, the limitations of observational data, and the uncertainty still embedded in the evidence. But I think people deserve the nuance. They deserve to know that the story is changing.
Some days I wonder if we'll look back on this period the way we look back on smoking in the 1950s - when doctors endorsed cigarettes and we genuinely didn't know the full extent of the harm. Maybe in 30 years, drinking alcohol will seem as obviously risky as smoking does now. Or maybe we'll find that low-level consumption really is fine for most people, and the cancer risk is manageable.
I don't know yet. Neither does anyone else. But we know more now than we did ten years ago, and that should matter.
The light was fading when I finished that conversation last week. The person sitting across from me nodded slowly, processing. They didn't say they'd quit. But they said they'd think about it. Sometimes that's enough.
