AUDIT-C Alcohol Use Screening — 3-Item Short Form Test
This tool implements the standard three-item AUDIT-C alcohol use screening test, the validated short form of the World Health Organization (WHO) Alcohol Use Disorders Identification Test (AUDIT), as originally derived and validated by Bush, Kivlahan, McDonell, Fihn, and Bradley (1998) in Archives of Internal Medicine. The three AUDIT-C items cover (1) frequency of alcohol consumption, (2) typical number of standard drinks on drinking days, and (3) frequency of heavy episodic (binge) drinking defined as six or more drinks on one occasion per the original AUDIT-C and WHO AUDIT wording. Each item is scored 0–4 using standard ordinal response anchors, summing to a total score in the 0–12 range. Standard evidence-based screening-positive cutoffs are reported separately by sex per NIAAA and WHO guidance: ≥3 for women and ≥4 for men indicate possible at-risk or hazardous drinking warranting further evaluation. AUDIT-C sensitivity for identifying alcohol use disorders across primary care samples is ≥90% in meta-analytic data (Aalto et al., 2011).
AUDIT-C 3-Item Screening — Your Usual Drinking Pattern
AUDIT-C Instrument: Origins, Scoring, and Screening-Positive Cutoffs
From WHO AUDIT (10 Items) to the AUDIT-C 3-Item Short Form
The full 10-item Alcohol Use Disorders Identification Test (AUDIT) was developed by a World Health Organization collaborative multinational study published in 1993, with formal guidelines for use in primary care released in 2001 (WHO, 2001). AUDIT was intentionally designed as a cross-culturally applicable screening instrument for use in general medical and primary care settings, rather than specialized addiction clinics, with the goal of identifying the full spectrum of hazardous and harmful alcohol consumption patterns — not only severe alcohol dependence. The 10 items fall into three conceptual domains: alcohol consumption (Items 1–3, covering frequency, typical quantity, and heavy episodic drinking frequency), dependence symptoms (Items 4–6), and alcohol-related consequences (Items 7–10). In 1998, Bush, Kivlahan, McDonell, Fihn, and Bradley at the University of Washington and VA Puget Sound Health Care System published the seminal validation of a three-item ultra-short screening form labeled AUDIT-C (C for consumption), consisting of only the first three AUDIT consumption items. The AUDIT-C was specifically calibrated to retain very high sensitivity for identifying DSM-IV alcohol abuse or dependence (current AUD) while substantially reducing administration time to under one minute in busy primary care settings. In the original VA primary care sample of 4,895 male patients, AUDIT-C at a cutoff of ≥4 achieved 97% sensitivity and 78% specificity for current AUD. Multiple subsequent validations in female, adolescent, geriatric, civilian, and cross-national samples replicated high sensitivity, leading to NIAAA, WHO, USPSTF, and ASAM endorsement of AUDIT-C as the preferred first-stage brief alcohol screen in primary care.
Three-Item Scoring Algorithm and Sex-Differentiated Cutoffs
Each of the three AUDIT-C items uses five ordinal response anchors scored 0, 1, 2, 3, and 4 from lowest to highest alcohol exposure. The total AUDIT-C score is the simple arithmetic sum of the three item scores, yielding a closed integer range of 0 to 12. Item 1 (drinking frequency) anchors: 0 = Never; 1 = Monthly or less; 2 = 2 to 4 times a month; 3 = 2 to 3 times a week; 4 = 4 or more times a week. Item 2 (typical drinks on drinking days) anchors: 0 = 1 or 2; 1 = 3 or 4; 2 = 5 or 6; 3 = 7 to 9; 4 = 10 or more, measured in standard drinks of ~14 g pure alcohol. Item 3 (heavy episodic / binge frequency, original wording "6 or more drinks on one occasion") anchors: 0 = Never; 1 = Less than monthly; 2 = Monthly; 3 = Weekly; 4 = Daily or almost daily. Standard screening-positive cutoffs differ by sex per NIAAA 2005 clinician guide and Aalto 2011 systematic review: for cisgender and transgender women, score ≥3 is screening-positive; for cisgender and transgender men, score ≥4 is screening-positive. This sex-differentiated threshold accounts for pharmacokinetic differences (smaller volume of distribution, lower gastric alcohol dehydrogenase activity in women producing higher BAC per drink) and NIAAA low-risk limits (≤7 drinks/week and ≤4/occasion women vs. ≤14 drinks/week and ≤5/occasion men). For pregnant people and individuals under 21, any nonzero score should prompt discussion with a clinician because no safe lower limit of alcohol consumption exists in those populations.
Test Properties: Sensitivity ≥90%, Specificity, and Screening Rationale
AUDIT-C is unusual among brief screening instruments for its very high pooled sensitivity in identifying DSM-IV/DSM-5 Alcohol Use Disorder (AUD) across primary care and general medical samples. Aalto, Seppä, and Tuunanen's 2011 systematic review and meta-analysis in Alcohol and Alcoholism of 28 AUDIT-C validation studies with 15,522 participants reported a pooled sensitivity of 0.92 (95% CI 0.89–0.94) at the standard ≥4 cutoff for men and ≥3 for women, meaning that AUDIT-C correctly identifies ≥90% of individuals meeting formal AUD diagnostic criteria on structured clinical interview. By comparison, the widely used CAGE questionnaire achieves sensitivity of only ~60–75% for AUD because it primarily captures dependence symptoms rather than hazardous consumption patterns. The tradeoff for very high sensitivity is more moderate specificity: Aalto 2011 pooled AUDIT-C specificity of 0.74 (95% CI 0.69–0.78), meaning that approximately 25% of individuals without AUD will screen positive and would require further confirmatory evaluation (e.g., full 10-item AUDIT, Structured Clinical Interview for DSM-5 AUD criteria, or clinical conversation about alcohol-related consequences). Clinicians generally accept this high-sensitivity, moderate-specificity profile because the downstream consequences of missing at-risk drinking are substantially greater than the burden of confirmatory follow-up on screening-positive individuals. Importantly, AUDIT-C scores also function as continuous dose-response variables: scores of 0–2 have very low AUD probability, 3–6 indicate low-to-moderate elevated risk, 7–9 indicate substantially elevated probability of moderate-severity AUD, and scores of 10–12 indicate very high probability of severe Alcohol Use Disorder with probable dependence features and urgent clinical referral indicated.
| AUDIT-C Score Range | Screening Interpretation (Reference Labels) | Approximate AUD Probability Gradient & Action Reference |
|---|---|---|
| 0 – 2 | Low-risk drinking pattern (Screening-negative) | Scores in this band correspond to very low probability of meeting DSM-5 Alcohol Use Disorder criteria on structured diagnostic interview. NIAAA low-risk drinking limits are ≤7 drinks/wk and ≤4/occasion for women or ≤14 drinks/wk and ≤5/occasion for men. Brief standard preventive counseling about safe limits remains appropriate. Exceptions: any nonzero score during pregnancy warrants referral to an obstetric or addiction medicine clinician, and any nonzero score for individuals under the legal minimum drinking age warrants confidential discussion with a pediatric or adolescent medicine provider. |
| ≥ 3 for women; ≥ 4 for men (and 3–6 overall) | Screening-positive — Possible at-risk / hazardous drinking or mild AUD | AUDIT-C scores of 3 (women) or 4 (men) through 6 meet the screening-positive threshold for at-risk or hazardous drinking per WHO, NIAAA, and USPSTF primary care alcohol screening guidelines. Meta-analytic AUD probability at these scores is moderate (roughly 20–50% depending on population base rate). Standard next-step references: (a) complete full 10-item AUDIT to assess dependence symptoms and consequences, (b) brief motivational intervention or structured clinician conversation about alcohol-related consequences, (c) discuss reduction goals aligned with NIAAA low-risk limits, (d) consider structured DSM-5 diagnostic interview if clinically indicated. |
| 7 – 9 | Screening-positive — Elevated probability of moderate-severity AUD | Scores of 7–9 on the 12-point AUDIT-C scale correspond in primary care validation data to substantially increased likelihood of meeting DSM-5 Moderate Alcohol Use Disorder criteria (i.e., 4–5 of 11 DSM-5 AUD criteria present). In the original Bush 1998 VA sample, scores ≥7 were associated with high prevalence of alcohol-related consequences and probable dependence features. Standard next-step references: (a) structured DSM-5 diagnostic assessment by a qualified addiction medicine or mental health clinician; (b) laboratory or biomarker panel if clinically indicated (e.g., liver enzymes, MCV, phosphatidylethanol PEth, gamma-glutamyl transferase GGT); (c) discuss evidence-based pharmacotherapy options for AUD (naltrexone, acamprosate, disulfiram) with a prescribing physician; (d) consider referral to specialty substance use disorder treatment. |
| 10 – 12 | Screening-positive — Very high probability of severe Alcohol Use Disorder (possible dependence) | Maximum-range AUDIT-C scores of 10–12 are associated in validation samples with a very high probability of meeting DSM-5 Severe Alcohol Use Disorder criteria (6+ of 11 criteria present), including probable pharmacological tolerance and withdrawal features, repeated failed attempts to cut down, and continued use despite recurrent physical or psychological consequences caused by alcohol. Standard next-step references: (a) urgent referral to a qualified addiction medicine physician, psychiatrist, or licensed SUD treatment program for formal diagnostic evaluation and management; (b) medical assessment for alcohol withdrawal syndrome risk (CIWA-Ar protocol) because abrupt cessation after chronic heavy consumption may require medically supervised detoxification with benzodiazepine or other symptom-triggered pharmacotherapy to prevent delirium tremens, seizure, or other potentially life-threatening withdrawal complications; (c) comprehensive biopsychosocial treatment planning including FDA/EMA-approved AUD pharmacotherapies, structured psychotherapies (CBT, motivational enhancement therapy, 12-step facilitation), and peer recovery support as clinically appropriate. |
Data Source and References
- [1] World Health Organization (2001). The Alcohol Use Disorders Identification Test (AUDIT): Guidelines for Use in Primary Care (2nd edition). Geneva, Switzerland: WHO Department of Mental Health and Substance Dependence. Official 10-item AUDIT instrument, scoring, and primary care implementation framework.
- [2] Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA (1998). The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Archives of Internal Medicine, 158(16):1789-1795. doi: 10.1001/archinte.158.16.1789. Original AUDIT-C 3-item validation; 97% sensitivity for current AUD at ≥4 cutoff in VA primary care sample.
- [3] Aalto M, Seppä K, Tuunanen S (2011). A systematic review of alcohol screening questionnaires: AUDIT and its derivatives in primary health care. Alcohol and Alcoholism, 46(1):81-88. doi: 10.1093/alcalc/agq074. Meta-analysis of 28 AUDIT-C validation studies; pooled sensitivity 0.92 (≥90%) and pooled specificity 0.74 across 15,522 participants.
- [4] National Institute on Alcohol Abuse and Alcoholism (NIAAA) (2005, updated 2023). Helping Patients Who Drink Too Much: A Clinician's Guide. NIH Publication No. 06–3769. Binge thresholds: men ≥5 drinks, women ≥4 drinks in approximately 2 hours; low-risk weekly limits and AUDIT-C screening-positive cutoffs (≥3 ♀ / ≥4 ♂).
Frequently Asked Questions
-
The AUDIT-C alcohol screening test produces a total integer score in the 0 to 12 range, derived as the arithmetic sum of three 0–4 items. Item 1 (drinking frequency) contributes 0–4 points, Item 2 (typical standard drinks per drinking day) contributes 0–4 points, and Item 3 (frequency of heavy episodic drinking, originally "6 or more drinks on one occasion") contributes 0–4 points. Standard evidence-based screening-positive cutoffs were established in the original Bush et al. 1998 Archives of Internal Medicine validation study and subsequently refined by NIAAA, WHO, and multiple systematic reviews: for men, a total score of 4 or higher (≥4) is the standard screening-positive cutoff for at-risk or hazardous drinking and probable alcohol use disorder warranting further clinical evaluation; for women, a lower total score of 3 or higher (≥3) is the standard screening-positive cutoff. This sex-differentiated threshold reflects differences in alcohol pharmacokinetics (smaller volume of distribution and lower gastric alcohol dehydrogenase activity in women producing higher blood alcohol concentrations per standard drink) and the NIAAA 2005 definition of binge drinking as ≥4 drinks per occasion for women versus ≥5 drinks per occasion for men. Scores of 0–2 are conventionally classified as Low-risk reference. These cutoffs are descriptive reference labels for screening purposes only and do not constitute a diagnosis.
-
No. AUDIT-C is explicitly a brief screening instrument, not a diagnostic instrument. It was derived by Bush et al. (1998) from the first three consumption items of the full 10-item WHO Alcohol Use Disorders Identification Test (AUDIT) published by WHO in 2001, and is designed for rapid identification in primary care and general medical settings of patients who may have hazardous or harmful alcohol use and who warrant further confirmatory evaluation. The American Psychiatric Association Diagnostic and Statistical Manual (DSM-5) diagnosis of Alcohol Use Disorder (AUD) — Mild, Moderate, or Severe — requires structured clinical assessment of 11 specific diagnostic criteria covering impaired control, social impairment, risky use, and pharmacological criteria (tolerance and withdrawal). A positive AUDIT-C screen (≥3 women, ≥4 men) indicates elevated probability of meeting criteria for AUD or at-risk drinking, but a substantial minority of screening-positive individuals will not meet formal AUD criteria on structured diagnostic interview (AUDIT-C achieves sensitivity ≥90% but specificity is more moderate, typically 60–85% depending on population and cutoff per the Aalto 2011 Alcohol and Alcoholism systematic review). A definitive diagnosis of AUD or any substance use disorder must be made by a qualified physician, psychiatrist, licensed psychologist, or addiction medicine specialist using DSM-5 or ICD-11 diagnostic criteria.
-
AUDIT-C and most alcohol screening instruments apply sex-differentiated screening-positive cutoffs (≥3 for women, ≥4 for men) for two converging reasons: pharmacokinetic and clinical-epidemiological. First, on pharmacokinetic grounds, women on average have lower total body water and smaller volume of distribution for water-soluble ethanol, lower gastric mucosal alcohol dehydrogenase activity (first-pass metabolism), and higher relative blood alcohol concentration per gram of consumed alcohol per kilogram of body weight compared with men. A given quantity of alcohol therefore produces greater physiological exposure and greater toxicity risk in women at the same self-reported drink count. Second, on clinical-epidemiological grounds, the U.S. National Institute on Alcohol Abuse and Alcoholism (NIAAA) 2005 clinician's guide and subsequent updates define low-risk drinking limits separately by sex: for women, no more than 3 standard drinks on any single day and no more than 7 drinks per week; for men, no more than 4 standard drinks on any single day and no more than 14 drinks per week. NIAAA similarly defines binge drinking thresholds separately: ≥4 standard drinks in approximately 2 hours for women, versus ≥5 drinks for men (WHO AUDIT original wording used "6 or more drinks" as a universal heavy occasion question, which AUDIT-C retains as Item 3 text per the Bush 1998 original wording while applying sex-differentiated cutoff interpretation). Validation studies consistently confirm that the ≥3 / ≥4 split maximizes combined sensitivity and specificity for at-risk drinking across sexes relative to a universal single cutoff.
-
AUDIT-C Item 2 queries the number of "standard drinks containing alcohol" on a typical drinking day using a 5-point scale anchored at 0 (1 or 2 drinks), 1 (3 or 4), 2 (5 or 6), 3 (7 to 9), 4 (10 or more). A standard drink in the WHO AUDIT framework and NIAAA/U.S. definitions is defined as any beverage containing approximately 14 grams (0.6 fluid ounces or ~17.7 mL) of pure ethyl alcohol. Common U.S. examples of one standard drink include: 12 fluid ounces (355 mL) of regular 5% ABV beer; 5 fluid ounces (148 mL) of 12% ABV table wine; 1.5 fluid ounces (44 mL, one shot) of 40% ABV (80-proof) distilled spirits (whiskey, vodka, gin, tequila, rum, etc.); and 8–9 fluid ounces (237–266 mL) of 7% ABV malt liquor. Mixed drinks and cocktails often contain 1.5–3 standard drinks depending on recipe and pour size. Standard drink definitions vary slightly across jurisdictions (e.g., 10 g in the UK and many EU countries, 12 g in Australia and Japan), but AUDIT-C response anchors are intentionally framed as ordinal quantity bands ("3 or 4," "5 or 6," etc.) rather than gram-level quantities, making the scoring scale approximately comparable across countries despite minor jurisdictional standard-drink gram differences. Respondents should count each standard-drink-equivalent serving separately when tallying typical quantity per drinking day.