Alcohol Units, AUDIT-C and Dependence: The Evidence Behind the Pathway

A 48-year-old autorickshaw driver from Manipal comes to the OPD with tiredness. His liver tests are abnormal: AST 156 U/L, ALT 64 U/L, and a raised GGT. His MCV is 104 fL. When you ask about alcohol, he says, “Only a little, doctor. Just a quarter in the evening, like everybody.” This is an illustrative case, not a real patient.

By the end of this article you will know how much “a quarter” really is, and whether he is dependent. You will also know what his blood tests are telling you, and the two mistakes that hurt patients like him most.

1. Count in Units, Not in Drinks

A unit measures the ethanol itself, not the drink it arrived in. That is the whole point of it: whisky, beer and toddy become comparable, and a history written in units can be read by whoever sees the patient next. One unit is 10 millilitres, or 8 grams, of pure ethanol.

The formula is short: units = (volume in ml × ABV %) / 1000, where ABV is the alcohol by volume printed on the bottle. The UK Chief Medical Officers’ 2016 guidance recommends no more than 14 units a week, spread over 3 or more days, with several alcohol-free days. India sets no unit-based limit of its own.

“Why not just write ‘drinks occasionally’?” Because the spirits drunk here are stronger than the ones in Western textbooks, and the pours are larger. A history taken in pegs and bottles, converted with the wrong assumptions, puts the patient in a lower risk band than the one he belongs in.

• A large peg of IMFL (Indian Made Foreign Liquor), 60 ml at 42.8%, is 2.6 units: nearly a fifth of the weekly limit.
• A “quarter” or pauwa, 180 ml at 42.8%, is 7.7 units: a heavy binge dose.
• A 650 ml bottle of strong beer at 8.0% is 5.2 units.
• Country liquor or arrack, 100 ml at about 35%, is 3.5 units, and carries a high risk of contaminants.

Our driver drinks a quarter on six evenings a week. That is 180 × 42.8 / 1000, or 7.7 units a night, and 46.2 units a week: more than three times the limit. “Only a little” was not a lie. It was simply not counted.

2. The AUDIT-C: Three Questions

The AUDIT-C is the consumption part of the WHO’s AUDIT questionnaire: three questions, scored 0 to 12, validated for identifying hazardous drinking or an active alcohol use disorder. A score of 4 or more in men, or 3 or more in women, is a positive screen. A positive AUDIT-C obliges the full 10-question AUDIT and a physical examination.

A common mistake is to apply the male cut-off of 4 to everyone. It is easy to see why: one number is easier to remember than two. But a woman scoring 3 is positive and needs full evaluation, and the single cut-off misses hazardous drinking in women.

3. Is He Dependent? CRAVE

The tool uses five bedside clues, remembered as CRAVE: Compulsion (a strong, persistent urge to drink), Relapse (repeated failed attempts to cut down), Autonomic withdrawal (morning tremor, sweating, “eye-openers”), Volume tolerance (needing markedly more for the same effect) and Erosion of function (social, occupational or family breakdown). Two or more features, especially A and V together, indicate probable dependence, which is severe alcohol use disorder by DSM-5.

“Why are A and V the strongest pair?” Because both are signs that the brain itself has changed. On tolerance, the tool used to add: It marks GABA and NMDA receptor adaptation. Put simply, alcohol calms the brain through GABA, its main braking system, and blocks NMDA, one of its accelerators. Given alcohol every day, the brain answers by weakening its brakes and strengthening its accelerators, so more alcohol is needed for the same effect.

Think of a car driven every day with the handbrake half on. The driver learns to press the accelerator harder. Now release the handbrake suddenly: the car lurches forward. That lurch is withdrawal, the tremor and sweating on waking that a morning drink relieves. It is the single most telling finding in the history. The comparison stops at one point: a car settles once the brake is off, but the brain in withdrawal can go on to seizures and delirium tremens.

Asked gently, our driver admits that his hands shake in the morning until he has had “a small one”, and that he now needs a full quarter where a half once did. That is A and V together: probable dependence.

4. What the Blood Tests Are Telling You

The De Ritis ratio. In alcoholic liver disease, AST is raised out of proportion to ALT, and a ratio above 2:1 strongly suggests alcohol as the cause. The tool used to explain why: It follows from alcohol-induced mitochondrial damage and a relative deficiency of pyridoxal-5'-phosphate, the active form of vitamin B6. ALT needs more of that vitamin than AST does, so in a drinker short of B6, ALT is held down while AST, released from damaged mitochondria, climbs.

Our driver’s AST of 156 against an ALT of 64 is a ratio of about 2.4. That is above 2:1.

Macrocytosis. A raised MCV comes from direct bone marrow toxicity, often independent of B12 or folate deficiency. An MCV above 100 fL in a drinker should prompt a frank conversation, even if the patient denies heavy drinking. His is 104.

The tool used to set the three common causes of liver disease side by side in one table:

No single row decides it. But a ratio above 2:1, a GGT several times the upper limit and a raised MCV, together with the history, point one way. Hepatitis B and C still need to be excluded with HBsAg and anti-HCV, because a drinker can have both.

5. Two Errors Worth Preventing

Stopping suddenly, with “SOS” tablets. Never advise a dependent patient to stop abruptly. He needs a severity assessment against a scale (CIWA-Ar) and a written tapering schedule of a benzodiazepine, to prevent seizures and delirium tremens. A few chlordiazepoxide tablets “SOS” for sleep or tremor is not a plan. Delirium tremens can declare itself 48 to 72 hours after the last drink, by which time nobody is watching.

Glucose before thiamine. A glucose load drives the enzymes for which thiamine is the cofactor, so dextrose given first to a thiamine-deplete patient can use up what little remains and precipitate Wernicke encephalopathy. Thiamine goes in first. The exception is the patient who is hypoglycaemic now: treat the hypoglycaemia at once and give the thiamine immediately afterwards. Where injectable thiamine is not stocked, start oral thiamine 100 mg three times daily at once and procure the injectable form urgently for anyone in whom Wernicke encephalopathy is suspected. The classic triad of confusion, ataxia and ophthalmoplegia is complete in only 16% of cases, so treat on suspicion.

Back to Our Patient

He drinks 46.2 units a week, more than three times the limit. His AUDIT-C will be positive (drinking on four or more days a week scores 4 on its first question alone), and he has two CRAVE features, A and V: probable dependence. His AST:ALT ratio of about 2.4 and his MCV of 104 fit alcohol as the cause of his liver disease, though hepatitis B and C still need excluding. He must not be told simply to stop. He needs a CIWA-Ar assessment, a written taper and thiamine, and the conversation about his drinking starts today, in units he can count himself.

Try This

A 38-year-old teacher scores 3 on the AUDIT-C. Her colleague says it is negative, “because the cut-off is 4”. Who is right, and what next? Illustrative case.

Answer. For a woman the cut-off is 3, so her screen is positive. She needs the full 10-question AUDIT and a physical examination.

Take-Home Points

• Count in units: volume in ml times ABV per cent, divided by 1000. A quarter of IMFL is 7.7 units.
• The AUDIT-C is positive at 4 in men and 3 in women; a positive screen needs the full AUDIT.
• Two CRAVE features, especially withdrawal and tolerance, mean probable dependence.
• Never advise abrupt cessation: assess with CIWA-Ar and write a taper.
• Thiamine before glucose, except in hypoglycaemia; treat Wernicke encephalopathy on suspicion.

For the withdrawal itself, see Alcohol Withdrawal: The Evidence Behind the Pathway; for staging the liver, FIB-4 and Elastography: The Evidence Behind the Thresholds; for the pancreas, Acute Pancreatitis Severity: The Evidence Behind the Scores; and for patients who ask about fatty liver without drinking, Why Fatty Liver When I Don’t Even Drink?

Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Evidence tab on 5 November 2026, each from the source the tool names. The illustrative cases follow the tool’s rules; check the tool’s result for a real patient.

References

1. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Arch Intern Med. 1998;158(16):1789-1795.
2. Bradley KA, DeBenedetti AF, Volk RJ, Williams EC, Frank D, Kivlahan DR. AUDIT-C as a brief screen for alcohol misuse in primary care. Alcohol Clin Exp Res. 2007;31(7):1208-1217.
3. Indian Council of Medical Research (ICMR). Standard Treatment Workflows (STW) - Management of Alcohol Use Disorders. New Delhi, India; 2022.
4. Munjal YP, Nadkar MY, Shah SN, Kamath SA, editors. API Textbook of Medicine, 12th Edition. Mumbai: Association of Physicians of India; 2022.
5. National Institute of Mental Health and Neuro-Sciences (NIMHANS). Clinical Practice Guidelines for the Management of Substance Use Disorders. Bengaluru, India; 2016.
6. World Health Organization. Global status report on alcohol and health 2018. Geneva: WHO; 2018.
7. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. AUDIT: The Alcohol Use Disorders Identification Test. Guidelines for Use in Primary Care. 2nd ed. Geneva: World Health Organization; 2001.
8. Department of Health, United Kingdom. UK Chief Medical Officers’ Low Risk Drinking Guidelines. London: Department of Health; 2016.
9. Indian Psychiatric Society. Clinical Practice Guidelines for the Management of Substance Use Disorders. Basu D, Ghosh A, Subodh BN, Mattoo SK. The same guideline is cited by the Alcohol Withdrawal: CIWA-Ar and Thiamine Protocol, so the two tools rest on one national source.
10. Harper CG, Giles M, Finlay-Jones R. Clinical signs in the Wernicke-Korsakoff complex: a retrospective analysis of 131 cases diagnosed at necropsy. J Neurol Neurosurg Psychiatry. 1986;49(4):341-345.