日本語
Fact checkChecked 2026-09-30

Does the Type of Alcohol Change Cancer Risk? Which Cancers Rise, Which Show Lower Numbers, and Whether Drinking Has Benefits, Checked Against International Assessments and Studies

PR This article contains affiliate links to Rakuten Ichiba and A8.net (“See on Rakuten”, a test-kit link). What is written here is based on assessments by international bodies, national guidelines, and studies, regardless of whether there are ads. The “See on Amazon.co.jp” and “See on Yahoo! Shopping” links are also affiliate links.

“Shochu is better for you than beer.” “Red wine is fine because it has polyphenols.” “People who drink a little live longer.” There are many theories about alcohol.

In this article, we check the following four points against assessments by international bodies and studies from around the world and Japan:

  • Does the type of alcohol (beer, wine, sake, shochu, whisky, and so on) change how likely you are to get cancer?
  • Which cancers increase with drinking? Conversely, are there cancers that show lower numbers?
  • The link with the “facial flushing constitution” that is common among Japanese people
  • Does alcohol have any benefits (heart disease, lifespan, and so on)?

For pancreatic cancer, we look in detail in the article on alcohol and pancreatic cancer.

Conclusion

  • What determines cancer risk is not the type of alcohol but the amount of alcohol you drink. Both the World Cancer Research Fund (WCRF) and the U.S. Surgeon General’s advisory (2025) say there is no difference among beer, wine, and spirits
  • The cancers that are convincingly increased by alcohol are cancers of the mouth, throat, and larynx, the esophagus, the liver, the colorectum, and the breast. In heavy drinkers, oral/throat and esophageal cancers are about 5 times as common. Stomach cancer is also “probably increased”
  • Kidney cancer, thyroid cancer, and malignant lymphoma show lower numbers in drinkers. The WCRF judges that for kidney cancer, “up to 30 g a day probably lowers the risk.” However, the reason is unclear, and the possibility of distortion in the studies has also been pointed out. For cancer overall, the risk rises as the amount increases
  • People with the facial flushing constitution (about 41% of Japanese people) face a very high risk of esophageal cancer and oral cancer if they keep drinking
  • The theory that “people who drink a little live longer” becomes unclear once the distortion in studies is corrected. Some studies show lower numbers for myocardial infarction, gallstones, and type 2 diabetes in women, but stroke and high blood pressure rise from small amounts. The WCRF says, “for cancer prevention, not drinking is best”

Assessments by national and international bodies

Body Judgment on alcohol and cancer
IARC (International Agency for Research on Cancer, a specialized agency of the WHO) Alcohol is “carcinogenic to humans (Group 1).” Acetaldehyde, which is formed when alcohol is broken down, is also Group 1. Stopping or reducing drinking lowers the risk of oral and esophageal cancer (2023 assessment)
WCRF / AICR (World Cancer Research Fund) Convincingly increases: mouth, throat, and larynx; esophagus (squamous cell carcinoma); liver; colorectum; breast (postmenopausal). Probably increases: stomach; breast (premenopausal). Probably decreases: kidney (up to 30 g a day). Regardless of the type of drink, the cause is ethanol itself
U.S. Surgeon General’s advisory (January 2025) A cause of at least 7 cancers (breast, colorectum, esophagus, liver, mouth, throat, larynx). Linked to about 100,000 cancer cases and about 20,000 cancer deaths a year in the U.S. The link holds regardless of the type of drink: beer, wine, or spirits
National Cancer Center Japan (assessment of studies on Japanese people, May 2026 edition) Convincing: cancer overall, liver, colorectum, esophagus, head and neck (mouth, throat, etc.). Probable: stomach (men), breast (premenopausal). Lung, pancreas, prostate, bladder, and others are “insufficient data”
MHLW “Guidelines on Alcohol Consumption with Consideration for Health” (2024) Gives, for each disease, the amount of drinking above which the risk rises (table below). Pure alcohol of 40 g or more per day for men and 20 g or more for women is defined as “an amount that raises the risk of lifestyle-related diseases”

Which cancers increase? Risk by site

A study pooling 572 studies worldwide (486,538 people who developed cancer) (Bagnardi 2015) found the following risks by cancer site, compared with non-drinkers.

Alcohol amount and cancer risk by site (ratio versus non-drinkers = 1)Results of Bagnardi 2015, which pooled 572 studies (486,538 cancer cases). Relative risk versus non-drinkers for heavy drinking (over 50 g a day): mouth and throat 5.13×, esophagus (squamous cell carcinoma) 4.95×, larynx 2.65×, gallbladder 2.64×, liver 2.07×, breast 1.61×, colorectum 1.44×, stomach 1.21×, pancreas 1.19×, lung 1.15×. Even with light drinking (12.5 g a day or less): mouth and throat 1.13×, esophagus 1.26×, breast 1.04×. Kidney (moderate 0.79×), thyroid (0.81×), and malignant lymphoma (non-Hodgkin 0.75 to 0.88×, Hodgkin 0.63 to 0.73×) showed lower numbers in drinkers.Light (≤12.5 g/day)Moderate (≤50 g)Heavy (over 50 g)Hollow = not statistically clear (within chance)0.5×1×2×4×← LowerHigher →Ratio versus non-drinkers = 1 (log scale)Higher riskMouth and throatMouth and throat, light (up to 12.5 g/day): 1.13×Mouth and throat, moderate (up to 50 g): 1.83×Mouth and throat, heavy (over 50 g): 5.13×5.13Esophagus (squamous)Esophagus (squamous), light (up to 12.5 g/day): 1.26×Esophagus (squamous), moderate (up to 50 g): 2.23×Esophagus (squamous), heavy (over 50 g): 4.95×4.95Larynx (voice box)Larynx (voice box), light (up to 12.5 g/day): 0.87× (not clear)Larynx (voice box), moderate (up to 50 g): 1.44×Larynx (voice box), heavy (over 50 g): 2.65×2.65GallbladderGallbladder, light (up to 12.5 g/day): 1.23× (not clear)Gallbladder, moderate (up to 50 g): 0.88× (not clear)Gallbladder, heavy (over 50 g): 2.64×2.64LiverLiver, light (up to 12.5 g/day): 1.00× (not clear)Liver, moderate (up to 50 g): 1.08× (not clear)Liver, heavy (over 50 g): 2.07×2.07Breast (women)Breast (women), light (up to 12.5 g/day): 1.04×Breast (women), moderate (up to 50 g): 1.23×Breast (women), heavy (over 50 g): 1.61×1.61ColorectumColorectum, light (up to 12.5 g/day): 0.99× (not clear)Colorectum, moderate (up to 50 g): 1.17×Colorectum, heavy (over 50 g): 1.44×1.44StomachStomach, light (up to 12.5 g/day): 0.99× (not clear)Stomach, moderate (up to 50 g): 0.97× (not clear)Stomach, heavy (over 50 g): 1.21×PancreasPancreas, light (up to 12.5 g/day): 0.95× (not clear)Pancreas, moderate (up to 50 g): 1.03× (not clear)Pancreas, heavy (over 50 g): 1.19×LungLung, light (up to 12.5 g/day): 0.84×Lung, moderate (up to 50 g): 0.98× (not clear)Lung, heavy (over 50 g): 1.15×Lower numbers (reason unclear)KidneyKidney, light (up to 12.5 g/day): 0.92×Kidney, moderate (up to 50 g): 0.79×Kidney, heavy (over 50 g): 0.80× (not clear)ThyroidThyroid, light (up to 12.5 g/day): 0.81×Thyroid, moderate (up to 50 g): 0.81×Non-Hodgkin lymphomaNon-Hodgkin lymphoma, light (up to 12.5 g/day): 0.88×Non-Hodgkin lymphoma, moderate (up to 50 g): 0.87×Non-Hodgkin lymphoma, heavy (over 50 g): 0.75×Hodgkin lymphomaHodgkin lymphoma, light (up to 12.5 g/day): 0.73×Hodgkin lymphoma, moderate (up to 50 g): 0.73×Hodgkin lymphoma, heavy (over 50 g): 0.63×
Made from Figure 2 of Bagnardi V et al. (2015), Br J Cancer. Light = 12.5 g or less of pure alcohol a day, moderate = up to 50 g, heavy = over 50 g. The comparison group is non-drinkers and occasional drinkers. Hollow circles are those whose 95% confidence interval spans 1 (not beyond the range of chance). Sites with few studies (small intestine, cervix, brain, etc.) and sites with no association seen (endometrium, ovary, bladder, etc.) are omitted.
  • Mouth and throat, and esophagus (squamous cell carcinoma) start to increase at small amounts and become about 5 times as common at heavy amounts. These are the sites most affected
  • Breast also increases little by little from small amounts (1.04×). A newer pooled study found a risk of 1.05× in women even at 0.5 drinks (5 g) a day (Sohi 2024)
  • Colorectum and larynx increase from moderate amounts, and liver, stomach, pancreas, and gallbladder increase at heavy amounts
  • Lung was 1.15× at heavy amounts and showed a lower number (0.84×) at small amounts. However, the authors note that people who drink a lot often smoke too, so the effect of smoking may not have been fully removed. In studies of only non-smokers, no link between drinking and lung cancer was seen

“From how much” does it increase?

The WCRF says the amount at which risk starts to rise differs by cancer site.

Cancer Amount at which risk starts to rise (WCRF)
Breast; esophagus (squamous cell carcinoma) No amount has been found that can be called “safe below this”
Colorectum Above about 30 g a day (1.5 tall cans of beer)
Liver; stomach Above about 45 g a day (a little over 2 tall cans)

The table in the national guideline, based on studies of Japanese people, is as follows (the per-day amount is a reference value: the weekly amount divided by 7).

Disease Men Women
Esophageal cancer Rises with any drinking No data
Stomach cancer Rises with any drinking 150 g/week (20 g/day)
Colorectal cancer 150 g/week (20 g/day) 150 g/week (20 g/day)
Liver cancer 450 g/week (60 g/day) 150 g/week (20 g/day)
Breast cancer ― 100 g/week (14 g/day)
Prostate cancer (advanced) 150 g/week (20 g/day) No data
Lung cancer (smokers) 300 g/week (40 g/day) No data
High blood pressure Rises with any drinking Rises with any drinking
Stroke (hemorrhagic) 150 g/week (20 g/day) Rises with any drinking

(Excerpted from Table 1 of the MHLW “Guidelines on Alcohol Consumption with Consideration for Health.” Ischemic heart disease is “under study”)

Does the type of alcohol make a difference?

The answer from international bodies: no difference

The WCRF writes in its report: “Alcohol causes various cancers regardless of type. The cause is ethanol itself, and how much it causes cancer is determined by how much and how often you drink.” The U.S. advisory likewise says the link holds “regardless of type, such as beer, wine, and spirits.”

What studies show

There is a pooled analysis of studies on cancers of the mouth, throat, and larynx (head and neck cancer) that separated people who drank only beer, only spirits, or only wine (Purdue 2009, 15 studies, 9,107 patients).

Likelihood of cancers of the mouth, throat, and larynx by type of alcoholResults of Purdue 2009, which pooled 15 studies (9,107 people with head and neck cancer). Odds ratios with never-drinkers = 1: beer only 1.6, 1.9, 2.2, and 5.4; spirits only 1.6, 1.5, 2.3, and 3.6; wine only 1.1, 1.2, 1.9, and 6.3 (up to 5 drinks, 6 to 15, 16 to 30, and 31 or more per week). For every type, the more people drank, the higher the ratio, and among the heaviest drinkers wine is just as high.0×1×2×3×4×5×6×7×Non-drinkers≤5 drinks6–1516–3031+Drinks per week (adjusted for the alcohol content of each type)Beer only, up to 5 drinks/wk: 1.6×Beer only, 6–15 drinks/wk: 1.9×Beer only, 16–30 drinks/wk: 2.2×Beer only, 31+ drinks/wk: 5.4×Spirits only, up to 5 drinks/wk: 1.6×Spirits only, 6–15 drinks/wk: 1.5×Spirits only, 16–30 drinks/wk: 2.3×Spirits only, 31+ drinks/wk: 3.6×Wine only, up to 5 drinks/wk: 1.1×Wine only, 6–15 drinks/wk: 1.2×Wine only, 16–30 drinks/wk: 1.9×Wine only, 31+ drinks/wk: 6.3×5.43.66.3Beer onlySpirits onlyWine only
Made from the abstract of Purdue MP et al. (2009), Am J Epidemiol. Odds ratios with people who have never drunk = 1. Compared by “number of drinks,” which equalizes the difference in alcohol content among types of alcohol.
  • Beer and spirits increased in about the same way at the same number of drinks
  • For people who drank only wine, the increase appeared smaller at low numbers of drinks. However, the authors say they cannot rule out the influence of the fact that wine drinkers differ in diet and lifestyle. Among the heaviest drinkers, wine was also 6.3×

For other sites as well, differences by type were not clear.

Study What was found
Colorectal cancer (Cho 2004, 8 follow-up studies, about 490,000 people) Risk rises from 30 g a day or more. No clear difference by type of alcohol
Red wine and white wine (Lim 2025, 42 studies) For cancer overall, no link for either red or white (red 0.98×, white 1.00×). Does not support the theory that “red wine is better for you.” One result for women showed 1.26× for white wine
Kidney cancer (Lee 2007, 12 follow-up studies, about 760,000 people) No difference among beer, wine, and spirits
Pancreatic cancer (Naudin 2025) Slightly higher with beer and spirits, no link with wine. However, it is hard to separate out the effects of differences in diet and lifestyle (article on alcohol and pancreatic cancer)

Some studies show “lower with wine,” but wine drinkers often differ from drinkers of other alcohol in diet, income, smoking, and so on, so it cannot be said to be a difference in the alcohol itself. Nor has the cancer-preventing effect of the polyphenols in red wine been confirmed by studies.

What to compare is “grams of pure alcohol”

More important than the type is the amount of pure alcohol you drink. The national guideline calculates it with this formula.

Pure alcohol (g) = amount of drink (ml) × alcohol by volume (%) ÷ 100 × 0.8

Amounts of alcohol that come to about 20 g of pure alcoholBeer at 5% is 500 ml, canned chuhai at 7% is 350 ml, sake at 15% is 180 ml (1 gō), wine at 12% is 200 ml, shochu at 25% is 100 ml, and whisky at 43% is 60 ml; each is about 20 g of pure alcohol. Pure alcohol (g) = amount (ml) × alcohol by volume ÷ 100 × 0.8.Each is about 20 g of pure alcoholBeer (5%) 500 ml = about 20 g of pure alcohol500mlBeer5%1 tall canCanned chuhai (7%) 350 ml = about 20 g of pure alcohol350mlCanned chuhai7%1 can (350 ml)Sake (15%) 180 ml = about 22 g of pure alcohol180mlSake15%1 gōWine (12%) 200 ml = about 19 g of pure alcohol200mlWine12%1.5 glassesShochu (25%) 100 ml = about 20 g of pure alcohol100mlShochu25%Half a glassWhisky (43%) 60 ml = about 21 g of pure alcohol60mlWhisky43%1 double
Made using the formula in the MHLW “Guidelines on Alcohol Consumption with Consideration for Health.” The alcohol percentages are typical examples and vary by product.

People sometimes say “shochu is better for you than beer,” but 100 ml of shochu and 500 ml of beer both contain about 20 g of pure alcohol. Even a drink with a low alcohol content is the same if the amount is large. We check the sugar and purines in shochu and the “blood-thinning” theory in the article on shochu and health.

The “facial flushing constitution” common among Japanese people

How alcohol is broken down in the body, and the facial flushing constitutionThe ethanol in alcohol is turned into acetaldehyde by enzymes such as ADH1B, then into acetic acid by the enzyme ALDH2, and finally becomes water and carbon dioxide. The IARC classifies acetaldehyde as Group 1 (carcinogenic to humans). People whose ALDH2 works weakly build up acetaldehyde easily and their faces turn red (the flushing reaction). In Japan this is said to be about 41%.EthanolAlcohol in drinksAcetaldehydeCarcinogen (Group 1)Acetic acid→ water and CO₂ADH1B etc.ALDH2Mainly in the liverPeople with weak ALDH2 activity (about 41% in Japan)• Face flushes and heart races after a little alcohol (flushing)• Acetaldehyde stays in the body longer• Continued drinking: very high risk of esophageal and oral cancer• You do not become “stronger” by drinking (set by genes)
Made from the MHLW “Guidelines on Alcohol Consumption with Consideration for Health” and the classification of the International Agency for Research on Cancer (IARC).

The alcohol in a drink is broken down in the body into acetaldehyde. Acetaldehyde is a substance the IARC classifies as Group 1 (carcinogenic to humans).

People whose enzyme ALDH2, which breaks down acetaldehyde, works weakly get a red face after drinking only a little (the flushing reaction). According to the national guideline, about 41% of people in Japan have this constitution.

Studies on Japanese people What was found
JPHC Study (Ono 2020, about 79,000 people followed for an average of 17 years) Half of men and 36% of women answered that their face turns red. In men whose face turns red, the risk of alcohol-related cancer was 1.63× at 150–299 g a week (about 21–43 g a day) and 2.02× at 300–449 g. Even in people whose face does not turn red, those who drink a lot have a higher risk
JPHC Study (Ishiguro 2009, about 45,000 men) In people with the flushing constitution who drink a lot and also smoke a lot, the rise in esophageal cancer risk was especially large
JPHC Study (Inoue 2005, about 73,000 people) In men, the risk of cancer overall rose in proportion to the amount drunk (1.18× at 1–149 g a week, 1.61× at 450 g or more). About 13% of cancers in men were estimated to be due to drinking 300 g or more a week
Nationwide hospital survey (Zaitsu 2020, 63,232 cancer patients) The risk of cancer was lowest in “lifelong non-drinkers.” Even at small amounts of 2 drinks a day or less, the risk of cancer overall was raised

A National Cancer Center Japan group estimates that 59,838 people (6.2%) of those newly diagnosed with cancer in Japan in 2015 had cancer caused by alcohol (Hirabayashi 2021).

The national guideline warns that even for people with the flushing constitution “who have drunk for many years and have become able to drink without discomfort,” the risk of oral cancer and esophageal cancer is very high. Even if you feel you have become stronger by drinking, your constitution has not changed.

Do the cancers with lower numbers decrease with alcohol?

As Figure 1 shows, kidney cancer, thyroid cancer, and malignant lymphoma show lower numbers in drinkers.

Cancer Study results Caution
Kidney The WCRF judges that “up to 30 g a day probably lowers the risk.” In a pooled analysis of 12 follow-up studies, 0.72× at 15 g a day or more (Lee 2007) For amounts above 30 g, there are no studies from which to judge. Hypotheses such as an effect on how insulin works have been put forward as the reason, but nothing is certain
Thyroid 0.74× in a pooled analysis of 33 studies (Hong 2017) Bagnardi 2015 says “the number of studies is small, so no conclusion can be drawn”
Malignant lymphoma 0.6–0.9× for both Hodgkin and non-Hodgkin (Bagnardi 2015) People who quit drinking because of early symptoms of the disease may have been counted among “non-drinkers,” making the numbers look lower. In a study of about 120,000 people in Europe, no link was seen

None of these cancers leads to a conclusion strong enough to be a reason to drink. Even after giving its judgment on kidney cancer, the WCRF says that its recommendation for cancer prevention is “not drinking is best. If you drink, follow the national guidelines.”

Does alcohol have benefits?

Is it true that “people who drink a little live longer”?

Until now, many studies found that people who drink a little have a lower risk of death than people who do not drink at all (the J-shaped curve). However, if the “non-drinkers” group includes people who quit drinking because of illness, the risk of non-drinkers looks higher than it really is.

Is “people who drink a little live longer” true? Risk of death compared with non-drinkersRisk of death from any cause, with non-drinkers = 1. In Stockwell 2016, people who drink a small amount were 0.86 before adjustment, but after correcting the bias of mixing people who used to drink and quit into the “non-drinker” group, it was 0.97 and the difference disappeared. Adjusted results of Zhao 2023: occasional 0.96, 1.3 to 24 g a day 0.93 (neither clearly different), 25 to 44 g 1.05, 45 to 64 g 1.19, and 65 g or more 1.35.0.80.91.01.11.21.31.4Ratio versus people who have never drunk = 1Stockwell 2016 (87 studies), light drinking (1.3–24.9 g/day)Before adjustmentBefore adjustment: 0.860.86Looks 14% lowerAfter adjustmentAfter adjustment: 0.970.97No clear differenceZhao 2023 (107 studies, about 4.84 million people), adjustedOccasional (under 1.3 g)Occasional (under 1.3 g): 0.960.96Light (1.3–24 g)Light (1.3–24 g): 0.930.9325–44 g25–44 g: 1.051.0545–64 g45–64 g: 1.191.1965 g or more65 g or more: 1.351.35
Made from the abstracts of Stockwell T et al. (2016), J Stud Alcohol Drugs, and Zhao J et al. (2023), JAMA Netw Open. Red = statistically higher; gray = no clear difference.
  • In Stockwell 2016, which pooled 87 studies, people who drink a little were 0.86× before adjustment, but after adjustment it was 0.97×, and the difference disappeared. People who had quit drinking were 1.22×, pushing up the non-drinker group
  • In Zhao 2023, which pooled 107 studies (about 4.84 million people), after adjustment there was no clear difference for people drinking up to 24 g a day, and at 45 g or more a day the risk rose to 1.19–1.35×

Heart and brain diseases

Study What was found
About 600,000 drinkers (Wood 2018) The risk of death was lowest at 100 g a week or less. As the amount rises, stroke (1.14× per 100 g) and heart failure (1.09×) go up. Only myocardial infarction goes down (0.94×)
About 510,000 people in China (Millwood 2019) By self-reported drinking, a “U shape” appeared, with lower stroke and myocardial infarction at around 100 g a week. But when analyzed by genetically determined drinking amount, the U shape disappeared: stroke rose the more people drank, and there was no link with myocardial infarction
About 260,000 people of European ancestry (Holmes 2014) People with genes that make them drink less alcohol have less heart disease (0.90×). The conclusion was that even for people who drink small to moderate amounts, cutting down is better for the heart
National drinking guideline (2024) For high blood pressure, “rises with any drinking” for both men and women. Ischemic heart disease is “under study”

Analysis using genes (Mendelian randomization) is a method that is less affected by differences in lifestyle. This method has not shown that small amounts of drinking protect the heart or brain.

Other conditions with lower numbers

Disease Study results
Gallstones In a pooled analysis of 18 studies, the heaviest drinkers were 0.62×. 0.88× for every 10 g a day (Wang 2017)
Type 2 diabetes In women, 0.69× at around 16 g a day (J shape). No link in men. Even in women, the lower number appeared only in people who were somewhat overweight (Llamosas-Falcón 2023)

The answer changes with age

A study summarizing the global burden of disease (GBD 2020) calculated the “amount at which the health risk is smallest” by age.

  • Ages 15–39: 0–0.6 drinks a day (0–6 g of pure alcohol). Young people suffer greater harm from injuries, accidents, and so on, so an amount close to not drinking is best
  • Ages 40 and over: 0.1–1.9 drinks a day (1–19 g), depending on the region. In the age group where heart disease and similar conditions increase, a J shape appeared in which small amounts lowered the risk

(One drink in the GBD is 10 g of pure alcohol)

In summary

  • There are diseases that show lower numbers with small amounts of drinking (myocardial infarction, gallstones, type 2 diabetes in women)
  • However, cancer, stroke, and high blood pressure increase from small amounts, and “people who drink a little live longer” becomes unclear once the distortion in studies is corrected
  • The enjoyment of unwinding and of time with other people is hard to show in study numbers. But it can be said that the grounds for “drinking because it is good for you” are weak. There is no reason for a non-drinker to start drinking for their health

What drinkers can do

What to do Why
Know the amount of pure alcohol Risk is determined by amount, not by type. The national guideline’s benchmark is not to exceed 40 g a day for men and 20 g for women
Decide the amount in advance before drinking Recommended by the national guideline. Alcohol poured at home is hard to judge by eye, so a measuring cup helps
Mix in non-alcoholic drinks Recommended by the national guideline. Replacing one drink removes that much pure alcohol entirely
Do not drink a lot at one time 60 g or more on one occasion is “heavy episodic drinking,” which should be avoided
Make non-drinking days in the week Lowers the weekly amount. Drinking every day can lead to dependence
If your face turns red, drink less The risk of esophageal and oral cancer is especially high
If you also smoke, cut down on both When drinking and smoking overlap, the risk of esophageal cancer and others rises further
Quit or cut down For oral and esophageal cancer, quitting or cutting down lowers the risk (IARC 2023)

For those who want to drink less (Rakuten Ichiba)

Non-alcoholic drinks

アルプス ヴァン フリー 赤 サントリー のんある酒場 レモンサワー
Type Wine-taste drink (90% grape juice) Carbonated drink (lemon sour taste)
Alcohol content 0.00% Non-alcoholic
Volume 720 ml 350 ml × 24 cans
Price (tax included) 1,091 yen + 598 yen shipping 3,059 yen (free shipping)
From the label No antioxidants added. Made by a winery in Shiojiri, Nagano (株式会社アルプス) Ingredients: lemon juice, shochu extract (non-alcoholic) / acidulant, carbonation, flavoring, antioxidant (vitamin C), sweeteners (acesulfame K, sucralose)
Shop 酒類の総合専門店 フェリシティー (a liquor retailer) 楽天24 ドリンク館
Buy See on Rakuten (フェリシティー) PR
See on Amazon.co.jp PR
See on Yahoo! Shopping PR
See on Rakuten (楽天24) PR
See on Amazon.co.jp PR
See on Yahoo! Shopping PR

For non-alcoholic beer, we compare 3 products in the article on alcohol and pancreatic cancer.

Measuring how much you drink at home

高桑金属 メジャーカップ M
Volume 45 ml / 30 ml (both sides)
Material and origin 18-8 stainless steel, made in Japan
Price (tax included) 1,780 yen (free shipping)
Benchmark 45 ml of whisky (43%) is about 15 g of pure alcohol; 45 ml of shochu (25%) is about 9 g
Shop リコメン堂生活館
Buy See on Rakuten (リコメン堂生活館) PR
See on Yahoo! Shopping PR

Knowing your constitution

ビズジーン アルコール分解型判定キット(VGAL-01) ハーセリーズ アルコール感受性遺伝子検査キット
What it tests The genes for the enzymes that break down alcohol (ADH1B and ALDH2). Results are given in 5 types The same two genes (ADH1B and ALDH2). Results are given in 9 types
Method Rub the inside of your mouth with a dedicated swab and mail it in. Results on the web in about 15 business days Take cells from the lining of your mouth with the included swab and mail it in. A results report in about 15 business days (a paper report can also be chosen). In principle 18 and older
Price 3,980 yen (tax included) 5,278 yen (free shipping)
Shop ビズジーン (the maker’s Rakuten Ichiba store) The official page of ハーセリーズ (株式会社イービーエス)
Buy See on Rakuten (ビズジーン) PR
See on Amazon.co.jp PR
See on Yahoo! Shopping PR
『アルコール感受性遺伝子分析キット』 PR (A8.net)

(Checked on September 30, 2026 on the Rakuten Ichiba and Herselys sales pages. Prices and shipping may change. Non-alcoholic drinks cannot be bought by people under 20)

Note The products introduced here do not prevent cancer. They are introduced as tools for drinking less and for knowing how much you drink. Genetic testing is not a medical diagnosis. Whether your face turns red can be roughly known, without any test, from your own reaction when you drink a small amount of alcohol. If you want to quit drinking but cannot, you can consult a medical institution, a public health center, or a mental health and welfare center.

Summary

  • Cancer risk is determined by the amount, not the type, of alcohol. Whether it is beer, wine, or shochu, think in grams of pure alcohol
  • The cancers that increase are those of the mouth, throat, and larynx, the esophagus, the liver, the colorectum, and the breast (convincing) and the stomach (probably). For the mouth, throat, and esophagus, heavy drinking brings about 5 times the risk
  • Kidney cancer, thyroid cancer, and malignant lymphoma show lower numbers, but the reason is unclear and it is no basis for recommending alcohol
  • People with the facial flushing constitution (about 41% of Japanese people) have an especially high risk of esophageal and oral cancer
  • “People who drink a little live longer” becomes unclear once the distortion in studies is corrected. Lower numbers appear for myocardial infarction and others, but stroke and high blood pressure increase from small amounts
  • If you drink, do not exceed 40 g a day for men and 20 g for women, do not drink 60 g or more at one time, and make non-drinking days. Quitting or cutting down lowers the risk of oral and esophageal cancer

Sources

Related articles and databases

This article organizes labeling rules and evaluations based on public agencies' documents and research. It does not judge whether any particular food or additive is safe or dangerous. Rules in Japan and abroad change over time (checked 2026-09-30). This is a translation of our Japanese article; if they differ, the Japanese article is the reference.