Scotch: the glass is ethanol, and so is the evidence

Nothing on this page was measured with scotch in the glass. What was measured is ethanol: a laboratory dose given by body weight, and vodka with cola. In a randomized crossover trial, 21 healthy adults drank 0.3 to 0.35 g of alcohol per kilogram of body weight on four separate occasions. Fasted, their peak blood alcohol reached 0.064 percent. After a 635 calories multicomponent meal the same dose peaked at 0.020 percent. That trial followed them by breathalyser for 90 minutes and measured nothing else, so it describes how fast a drink arrives rather than what it does. In a separate trial, twelve volunteers drank 0.5 g of alcohol per kilogram as vodka with regular or with diet cola. Peak breath alcohol was 38.3 micrograms per 3.4 fl oz (100 mL) with the diet mixer against 34.8 with the sugared one, and the authors suggest the sugar-free drink may leave the stomach faster. Twelve people, open label, vodka rather than whisky. Trust the direction more than the digits. None of this sets a limit. The safety section below carries a Group 1 carcinogen classification and the position that no safe amount of alcohol for cancer can be established. It also names who should not drink at all: pregnant women, people with alcohol-associated liver disease or cirrhosis, people recovering from alcohol use disorder or unable to control their drinking, and people on medicines that interact with alcohol.

As people eat it, per 3.5 oz (100 g)

What a real portion looks like

• The US alcohol institute (NIAAA) defines one standard drink as about 14 g of pure alcohol, the amount in a 1.5-ounce shot of distilled spirits such as whiskey at 40% alcohol. — US drinkers Evidence E source
  One standard drink = about 14 g pure alcohol = a 1.5 fl oz shot of distilled spirits such as whiskey at about 40% alcohol

What your body absorbs

• Eating before the glass changes how drunk the same measure makes you. In 21 fasted adults, peak blood alcohol was 0.064% with no food and 0.020% after a multicomponent meal. — 21 overnight fasted healthy adults, 10 male and 11 female, given 0.3 to 0.35 g of alcohol per kg body weight (n = 21) Evidence B source
  Peak BAC 0.064% no food, 0.047% savory snack mix, 0.031% reduced-bioavailability bar, 0.020% multicomponent meal; alcohol bioavailability over 90 min reduced by 22.0%, 45.0% and 67.9% against no food
• The mixer matters too. Twelve volunteers drinking the same spirit dose reached a higher peak breath alcohol with a diet mixer than with a sugared one, 38.3 against 34.8 micrograms per 3.4 fl oz (100 mL), and the authors suggest the sugar-free drink may leave the stomach faster. — 12 healthy participants aged 19 to 64, 8 male and 4 female, given 0.5 g of alcohol per kg body weight as vodka with regular or diet cola (n = 12) Evidence B source
  Peak breath alcohol 38.3 ± 9.45 versus 34.8 ± 6.82 mcg/100 mL and a larger area under the breath ethanol curve with the diet mixer; gastric half-emptying 100.09 versus 110.74 min

What it does to your health

• The heart benefit that made a nightly dram respectable comes from observational studies, and the scoping review behind the Nordic Nutrition Recommendations 2023 reports that Mendelian randomization analyses do not fully support it. — evidence base reviewed for the Nordic and Baltic countries up to 31 May 2021 Evidence C source
  Observational data suggest protection against myocardial infarction and type 2 diabetes, which Mendelian randomization does not fully support

Safety, limits and interactions

• The US alcohol institute (NIAAA) lists fast heartbeat, blood pressure swings, vomiting and flushing when alcohol meets infection medicines such as metronidazole, stomach bleeding and ulcers with pain relievers and liver damage with acetaminophen, and abnormally low blood sugar with diabetes medicines. — people taking infection medicines including metronidazole (Flagyl), pain relievers including acetaminophen, or diabetes medicines Evidence E source
  Infection medicines: fast heartbeat, blood pressure changes, vomiting, flushing; pain relievers: bleeding, ulcers, liver damage with acetaminophen; diabetes medicines: abnormally low blood sugar
• NIAAA lists drowsiness, dizziness, slowed or difficult breathing, impaired motor control and memory problems when alcohol is mixed with sleep medicines such as zolpidem. — people taking sleep medicines such as zolpidem (Ambien) Evidence E source
  Drowsiness, dizziness, slowed or difficult breathing, impaired motor control, unusual behavior, memory problems
• Alcoholic beverages have been classified as a Group 1 carcinogen since 1988, with sufficient evidence that they cause cancer in humans at seven sites including the mouth, esophagus, liver, colorectum and female breast. — general population, as assessed by the IARC Monographs program Evidence E source
  Group 1 classification since 1988, with cancers of the oral cavity, pharynx, larynx, esophagus, colorectum, liver and female breast causally linked
• Alcohol is a drug interaction of its own. A systematic review of nine second generation antihistamines found that food, fruit juices or alcohol can significantly change how well and how safely these everyday allergy drugs work. — nine second generation H1-antihistamines: bilastine, cetirizine, desloratadine, ebastine, fexofenadine, levocetirizine, loratadine, mizolastine and rupatadine (n = 2326 publications screened) Evidence D source
  Food, fruit juices or alcohol may significantly impact efficacy and safety for selected H1-antihistamines

What people believe that the data does not support

• The risk is not limited to heavy drinking: WHO and IARC state that light or moderate drinking also raises cancer risk and that no safe amount for cancer can be set. — general population, including light and moderate drinkers Evidence E source
  No safe amount of alcohol consumption for cancers can be established

Who this works differently for

• The CDC says there is no known safe amount of alcohol during pregnancy and no safe time in pregnancy to drink, and it names liquor alongside wine and beer as able to cause harm. — pregnant women Evidence E source
  No known safe amount, no safe time; liquor named among harmful drinks
• The 2025-2030 Dietary Guidelines for Americans say to drink less alcohol. Pregnant women, people recovering from alcohol use disorder or unable to control their drinking, and people with interacting medicines or medical conditions should avoid it completely. — US adults, with named groups who should not drink at all Evidence E source
  Drink less; avoid completely if pregnant, recovering from alcohol use disorder or unable to control intake, or on interacting medicines or with interacting conditions; family history of alcoholism: be mindful
• NIDDK says doctors will tell people with alcohol-associated liver disease to stop drinking completely, and tells anyone with cirrhosis not to drink alcohol. — people with alcohol-associated liver disease or cirrhosis Evidence E source
  Stop drinking completely; do not drink alcohol with cirrhosis
• The US alcohol institute (NIAAA) describes alcohol use disorder as an impaired ability to stop or control drinking despite harm, and says the risk of developing it depends partly on how much, how often and how quickly a person drinks. — people who drink Evidence E source
  AUD defined as impaired ability to stop or control drinking despite harm; risk depends partly on how much, how often and how quickly one drinks
• The same review reports that light to moderate drinking does not raise all-cause mortality in middle-aged and older adults who avoid binge drinking, while in young adults the lowest mortality risk belongs to those who do not drink at all. — middle-aged and older adults, and young adults, in observational cohorts Evidence C source
  No increase in all-cause mortality with light to moderate intake in middle-aged and older adults without binge drinking; lowest mortality risk with total abstinence in young adults

What is still unknown

• The same review names why the numbers on moderate drinking stay contested: intake is hard to ascertain, comparison groups are poorly chosen, and confounders, colliders and mediators are insufficiently controlled. — observational studies of alcohol consumption reviewed for the Nordic recommendations Evidence C source
  Ascertainment of intake, selection of exposure groups and control of confounders, colliders and mediators named as inherent methodological issues

The bottom line

One part of this is not open to argument. Alcoholic beverages have been classified as a Group 1 carcinogen since 1988, with sufficient evidence that they cause cancer in humans at the oral cavity, pharynx, larynx, esophagus, colorectum, liver and female breast. WHO and IARC also state that there is clear evidence of increased cancer risk from light or moderate drinking, and that no safe amount of alcohol consumption for cancers can be established. A hazard classification of this kind says that alcohol causes cancer without putting a figure on what one glass adds. It is not a statement that the risk begins above some amount, because the same source says no safe amount can be set. The nightly dram earned its reputation elsewhere, in observational work. The scoping review behind the Nordic Nutrition Recommendations 2023 reports that regular moderate drinking looks protective against myocardial infarction and type 2 diabetes in those studies, and that Mendelian randomization analyses do not fully support the finding. The same review names why such numbers keep moving. Intake is hard to ascertain, exposure groups are chosen badly, and confounders, colliders and mediators are insufficiently controlled. Its reading of mortality splits by age. Light to moderate intake was not linked to higher all-cause mortality in middle-aged and older adults who avoid binge drinking, while among young adults the lowest mortality risk belonged to those who drank nothing. Those are observational comparisons from the same hampered evidence base, so read them as an open question rather than a settled one, and read them alongside the classification above rather than instead of it. Some people should not drink at all, and the sources are plain about it. The CDC says there is no known safe amount of alcohol during pregnancy and no safe time in pregnancy to drink, and it names liquor alongside wine and beer. The 2025-2030 Dietary Guidelines for Americans say to drink less, and say that pregnant women, people recovering from alcohol use disorder or unable to control how much they drink, and people with interacting medicines or medical conditions should avoid alcohol completely. NIDDK says doctors will tell people with alcohol-associated liver disease to stop drinking completely, and tells anyone with cirrhosis not to drink. NIAAA describes alcohol use disorder as an impaired ability to stop or control drinking despite harm, and says the risk of developing it depends partly on how much, how often and how quickly a person drinks. None of these sources gives a number of drinks that is safe. Medicines are the other line. NIAAA lists fast heartbeat, blood pressure swings, vomiting and flushing when alcohol meets infection medicines such as metronidazole, stomach bleeding and ulcers with pain relievers, liver damage with acetaminophen, and abnormally low blood sugar with diabetes medicines. With sleep medicines such as zolpidem it lists drowsiness, dizziness, slowed or difficult breathing, impaired motor control and memory problems. A systematic review of nine second generation antihistamines, screening 2326 publications, found that food, fruit juices or alcohol may significantly change their efficacy and safety. These are lists of what can happen, not rates, so any regular medicine is worth raising with whoever prescribes it. For scale, NIAAA defines one US standard drink as about 14 g of pure alcohol, the amount in a 1.5-ounce shot of spirits such as whiskey at 40% alcohol. That is a unit for counting, not a safe amount. We hold no USDA composition record under this name, and nothing above was tested with scotch in the glass.

Sources

1. hf2-scot-7 · agency definition
   In the United States, one standard drink contains about 14 grams, or about 0.6 fluid ounces, of pure alcohol. That is the amount of alcohol in: ... A 1.5-ounce shot glass of distilled spirits at 40% alcohol by volume ... 1.5 fl oz shot of distilled spirits (gin, rum, tequila, vodka, whiskey, etc.)
   NIAAA, What Is A Standard Drink? (updated December 2025) · checked 2026-10-08
2. scot-r4-03 · randomized crossover trial
   The pBAC of each group was different (P < .001) from all other groups (NF = 0.064 ± 0.003, SSM = 0.047 ± 0.002, RABB = 0.031 ± 0.002, MCM = 0.020 ± 0.002%; mean ± standard error of the mean).
   J Med Food, 2020 · checked 2026-09-24
3. scot-r4-04 · randomized crossover trial
   Diet coke increased both the peak BrAC (38.3 ± 9.45 vs. 34.8 ± 6.82 μg/100 mL)...The lack of sucrose in diet mixers may cause faster stomach emptying of alcohol, increasing its absorption rate into the blood, resulting in higher peak BrAC and increased exposure to other alcohol-related dangers.
   Toxicol Lett, 2025 · checked 2026-09-24
4. scot-r4-05 · scoping review
   Current evidence from mainly observational epidemiological studies suggests that regular, moderate alcohol consumption may confer protective effects against myocardial infarction (MI) and type 2 diabetes. Mendelian randomization analyses do not fully support these findings, possibly because these analyses may fail to identify low alcohol intake.
   Food Nutr Res, 2024 · checked 2026-09-24
5. hf2-scot-5 · agency advice
   Fast heartbeat, sudden changes in blood pressure; stomach pain, upset stomach, vomiting, headache, or flushing or redness of the face; liver damage (isoniazid, ketoconazole) ... Stomach upset, bleeding and ulcers; liver damage (acetaminophen); rapid heartbeat ... Abnormally low blood sugar levels, flushing reaction (nausea, vomiting, headache, rapid heartbeat, sudden changes in blood pressure); symptoms of nausea and weakness may occur (metformin)
   NIAAA, Harmful Interactions: Mixing Alcohol with Medicines (brochure, revised 2014) · checked 2026-10-08
6. hf2-scot-6 · agency advice
   Drowsiness, sleepiness, dizziness; slowed or difficulty breathing; impaired motor control; unusual behavior; memory problems
   NIAAA, Harmful Interactions: Mixing Alcohol with Medicines (brochure, revised 2014) · checked 2026-10-08
7. scot-r4-01 · agency position
   In 1988 alcoholic beverages were identified as a Group 1 carcinogen, with sufficient evidence that they cause cancer in humans....Cancers at the following sites have been causally linked to alcohol consumption: oral cavity, pharynx, larynx, oesophagus, colorectum, liver and female breast.
   WHO Regional Office for Europe and IARC, joint statement to the European Parliament on alcohol and cancer, 2023 · checked 2026-09-24
8. scot-r4-08 · systematic review of pharmacokinetic studies
   For selected H1-antihistamines food, fruit juices or alcohol consumption may significantly impact the efficacy and safety of the therapy.
   Biomed Pharmacother, 2017 · checked 2026-09-24
9. scot-r4-02 · agency position
   There is also clear evidence of an increased risk of cancer from light or moderate alcohol drinking. As such, no safe amount of alcohol consumption for cancers can be established.
   WHO Regional Office for Europe and IARC, joint statement to the European Parliament on alcohol and cancer, 2023 · checked 2026-09-24
10. hf2-scot-1 · agency advice
    There is no known safe amount of alcohol use during pregnancy. ... There is no safe time during pregnancy to drink alcohol. ... All types of alcohol can be harmful, including red or white wine, beer, and liquor.
    US CDC, About Alcohol Use During Pregnancy (page reviewed August 14, 2026) · checked 2026-10-08
11. hf2-scot-2 · dietary guideline
    Consume less alcohol for better overall health. People who should completely avoid alcohol include pregnant women, people who are recovering from alcohol use disorder or are unable to control the amount they drink, and people taking medications or with medical conditions that can interact with alcohol. For those with a family history of alcoholism, be mindful of alcohol consumption and associated addictive behaviors.
    Dietary Guidelines for Americans, 2025-2030 (HHS, USDA), January 2026 · checked 2026-10-08
12. hf2-scot-3 · agency advice
    For people who have alcohol-associated liver disease, or damage to the liver and its function from drinking too much alcohol, doctors will recommend that they completely stop drinking alcohol and may refer them for alcohol treatment. ... If you have cirrhosis, you can take steps to help keep cirrhosis from getting worse. ... Do not drink alcohol or use illegal drugs.
    NIDDK, Treatment for Cirrhosis (last reviewed June 2023) · checked 2026-10-08
13. hf2-scot-4 · agency advice
    Alcohol use disorder (AUD) is a medical condition characterized by an impaired ability to stop or control alcohol use despite adverse social, occupational, or health consequences. ... A person’s risk for developing AUD depends in part on how much, how often, and how quickly they consume alcohol.
    NIAAA, Understanding Alcohol Use Disorder (updated January 2025) · checked 2026-10-08
14. scot-r4-07 · scoping review
    All-cause mortality is not increased with light to moderate alcohol intake in middle-aged and older adults who do not engage in binge drinking. Total abstinence is associated with the lowest risk of mortality in young adults.
    Food Nutr Res, 2024 · checked 2026-09-24
15. scot-r4-06 · scoping review
    Observational studies on alcohol consumption are hampered by a number of inherent methodological issues such as ascertainment of alcohol intake, selection of appropriate exposure groups, and insufficient control of confounding variables, colliders, and mediators.
    Food Nutr Res, 2024 · checked 2026-09-24