CAGE is a 4-question yes/no alcohol screener developed by Dr. John A. Ewing at the University of North Carolina and published in JAMA in 1984. Each item asks about lifetime drinking patterns: Cut down, Annoyed by criticism, Guilty about drinking, and morning Eye-opener. Total score 0–4. Score ≥2 is the standard clinical cutoff: sensitivity 71%, specificity 90% for alcohol use disorder. CAGE is public-domain, recommended by NIAAA and SAMHSA as a brief alternative to the longer AUDIT-10. It is a screening, not a diagnosis — only a licensed clinician can diagnose AUD using DSM-5 criteria.
On this page
Have you ever…
What the CAGE measures and why NIAAA and SAMHSA endorse it
CAGE is a 4-item yes/no alcohol screener developed by Dr. John A. Ewing at the Bowles Center for Alcohol Studies, University of North Carolina at Chapel Hill, and published in 1984 in the Journal of the American Medical Association (JAMA). The acronym stands for the four questions: Cut down, Annoyed by criticism, Guilty about drinking, and morning Eye-opener. Validation across more than fifty primary-care, emergency-department, and hospital studies showed sensitivity 71% and specificity 90% for alcohol use disorder at the cutoff score of 2 (Ewing 1984).
CAGE lives inside the NIAAA Core Resource on Alcohol and the SAMHSA SBIRT workflow as a brief alternative to the 10-question AUDIT. CAGE matches workflows where time is tight (emergency departments, hospital admissions) and AUDIT-C (the 3-question consumption subset of AUDIT) covers the same niche when frequency-quantity data is also needed. CAGE is public-domain since 1984 — no copyright restriction, no permission required.
The 4 items capture different psychosocial markers of alcohol use disorder: perceived need to reduce drinking (Q1, awareness of problem), social criticism (Q2, interpersonal impact), guilt (Q3, internal recognition), and physiological dependence (Q4, morning eye-opener — the strongest single positive predictor of AUD in CAGE). Each item asks about lifetime drinking ("ever") rather than a recent recall window, which is one of CAGE's known limitations: it does not distinguish past resolved problems from current ones.
CAGE has documented lower sensitivity in women, college-age populations, and the elderly than in the general adult population. For these groups, AUDIT-10 or AUDIT-C is generally preferred. The TWEAK (a 5-question CAGE-derivative) was specifically validated for pregnant women. CAGE remains the most-used quick alcohol screen in US primary care, EDs, and hospital admission workflows because of its brevity and easy memorization.
What the 3 score bands mean
CAGE score interpretation comes directly from Ewing's 1984 JAMA publication: 0 is "negative screen," 1 is "borderline" (possible concern), and 2–4 is "positive screen" (clinically significant indication of alcohol use disorder). Each band below explains what the score reflects clinically, what guidance Ewing attached to it, and how the band typically maps to ASAM levels of care in current US practice.
A CAGE score of 0 means none of the four lifetime markers are present. Alcohol use disorder is unlikely under Ewing's framework, but CAGE is a screener, not a diagnosis. Specifically: a CAGE of 0 has roughly 90% specificity, so most people with negative scores really do not have AUD — but the 71% sensitivity at cutoff 2 means roughly one in three actual AUD cases can score below the cutoff, especially in women, college students, and the elderly. If a specific concern prompted you to take this screen, consider following up with the longer AUDIT-10 which catches hazardous drinking earlier (cutoff 8 instead of 2 with a cleaner past-12-month recall).
A CAGE score of 1 falls between Ewing's negative and positive thresholds. One of the four lifetime markers is present — common patterns include past attempts to cut down without current dependence, or remembered guilt about drinking from a specific past episode. At this band the standard recommendation is follow up with the full AUDIT-10, which uses past-12-month recall and a finer 0–3 frequency scale to distinguish current concerns from historical ones. SAMHSA's primary-care SBIRT workflow typically triggers a brief educational conversation (advice rather than full brief intervention) at this band. NIAAA "low-risk" guidance: ≤14 drinks/week and ≤4/day for men, ≤7/week and ≤3/day for women.
A CAGE score of 2 or higher is the standard clinical cutoff for positive screen — sensitivity 71%, specificity 90% for alcohol use disorder. Two or more of four lifetime markers indicate clinically significant patterns warranting full DSM-5 evaluation. Treatment guidance follows the AUDIT framework: brief intervention at minimum, structured outpatient (ASAM Level 1.0) or intensive outpatient (Level 2.1 IOP) depending on dependence severity. A score of 3 or 4 typically maps to higher-intensity care including possible medically supervised detox (ASAM Level 3.7-WM) — alcohol withdrawal can be life-threatening (seizures, delirium tremens) after weeks of heavy daily use. Three FDA-approved medications for alcohol use disorder — naltrexone, acamprosate, disulfiram — substantially reduce relapse mortality when combined with counselling.
From a CAGE score to a real next step: AUDIT-10 follow-up, SAMHSA helpline, ASAM levels, and AUD medications
If your CAGE score is 2 or higher, Ewing's 1984 cutoff recommends clinical follow-up. A CAGE score alone is not a treatment plan but a structured, evidence-based starting point. Here is how the instrument's recommendations translate to concrete actions in the United States today.
If you scored 1
- Take the full AUDIT-10 next. AUDIT uses past-12-month recall and a finer scale, catching hazardous drinking that CAGE may miss. AUDIT cutoff 8 is more sensitive for current concerns than CAGE cutoff 2 for lifetime markers.
- Track NIAAA low-risk limits. ≤14 drinks/week and ≤4/day for men; ≤7/week and ≤3/day for women. Most US adults exceed these at some point; persistent excess is the early signal for AUD development. NIAAA's Rethinking Drinking has a built-in calculator.
- Re-screen in 3–6 months if drinking continues at any concerning level.
If you scored 2 or higher
- Talk to a primary-care clinician and bring your CAGE score. CAGE lives in primary-care SBIRT screening as a quick triage; a 2+ is a direct reason for longer conversation about drinking and treatment options including medication and counselling.
- Call SAMHSA's free 24/7 helpline at 1-800-662-4357. Counsellors provide information, insurance navigation, and warm transfers to local treatment. Free, confidential, English and Spanish.
- FDA-approved medications work. Naltrexone (oral or monthly injectable) reduces craving and heavy-drinking days; acamprosate supports abstinence after withdrawal; disulfiram creates a deterrent reaction with alcohol. STAR-D-equivalent trials show medication plus counselling outperforms either alone.
- Search SAMHSA's treatment locator by ZIP for AUD-experienced providers, or browse our directory by state.
If you scored 3 or 4 — read this first
Heavy daily drinking can produce life-threatening withdrawal. Seizures and delirium tremens are documented after abrupt cessation in heavy daily users (typically >8 standard drinks per day for >3 weeks). Do not quit alone if any of these apply to you. Medically supervised detox at ASAM Level 3.7-WM (medically managed inpatient withdrawal) is indicated — benzodiazepines (diazepam, lorazepam) under medical supervision are the standard treatment for alcohol withdrawal in this setting. Once detox is complete, naltrexone or acamprosate plus structured treatment (IOP or PHP) substantially reduces relapse mortality. Call SAMHSA 1-800-662-4357 and describe your daily intake honestly — they route to the appropriate detox level. Immediate medical distress: 911.
If CAGE is not the right fit: AUDIT-10, DAST-10, PHQ-9, and GAD-7
CAGE screens alcohol with 4 yes/no lifetime questions. If you want a deeper alcohol picture (past-12-month, frequency-quantity scoring), drug use is a separate concern, or co-occurring depression or anxiety needs a screen, the validated instruments below cover those distinct questions. Pair CAGE with PHQ-9 + GAD-7 for the standard primary-care behavioral panel.
AUDIT-10 (Alcohol Use Disorders Identification Test)
10-question WHO-validated alcohol screener. Pair with PHQ-9 for a full alcohol-plus-depression picture — the combination is the standard primary-care SBIRT workflow.
DAST-10 (Drug Abuse Screening Test)
10-question Skinner-1982 drug-use screener (excluding alcohol and tobacco). Pair with PHQ-9 if substance use other than alcohol is a concern.
PHQ-9 (Patient Health Questionnaire-9)
9-question Kroenke-2001 depression severity screener. Depression co-occurs with substance use disorder in 30–40% of cases — integrated treatment is best evidence.
GAD-7 (Generalized Anxiety Disorder)
7-question anxiety severity screener (Spitzer 2006 — same author family as PHQ-9). Anxiety co-occurs with depression in roughly 50% of cases. At cutoff ≥10: 89% sensitivity, 82% specificity for GAD.
PHQ-2
2-item subset of PHQ-9 (anhedonia + depressed mood). Faster pre-screen used in primary care before triggering full PHQ-9.
ACE (Adverse Childhood Experiences)
10-question retrospective measure of childhood adversity (Felitti / CDC 1998). Public-domain. ACE ≥4 increases adult substance use disorder risk by 7×. Trauma-informed counterpart to current-symptom screeners.
PCL-5 (PTSD Checklist for DSM-5)
20-question current-symptom screener developed by VA National Center for PTSD (Weathers 2013). Public-domain. DSM-5-aligned with 4 cluster sub-scores. Pairs with ACE (past exposure + current symptoms = standard trauma workup). Cutoff ≥31 for probable PTSD.
WHO ASSIST
Longer 8-item WHO instrument covering alcohol, tobacco, cannabis, cocaine, stimulants, inhalants, sedatives, hallucinogens, opioids. For one-shot multi-substance screening. WHO ASSIST manual.
CAGE terms: AUD, MAT, ASAM, DSM-5, SBIRT, AUDIT-C
Six terms appear repeatedly across this page — the instrument itself (CAGE), the diagnosis it screens for (AUD via DSM-5), the FDA-approved medications for AUD (MAT — naltrexone, acamprosate, disulfiram), the level-of-care framework (ASAM), the public-health workflow CAGE fits inside (SBIRT), and the close peer instrument that often replaces or follows CAGE (AUDIT-C). Plain definitions follow.
- CAGE
- 4-question alcohol screener developed by Dr. John A. Ewing at UNC Chapel Hill and published in JAMA 1984. Public-domain. Score 0–4. Cutoff ≥2 for positive screen (sensitivity 71%, specificity 90% for AUD).
- AUD · Alcohol use disorder
- DSM-5 clinical diagnosis with 11 criteria graded mild (2–3), moderate (4–5), or severe (6+) over a 12-month period. Includes physical dependence, tolerance, craving, and social/occupational impact. AUDIT-10 and CAGE both screen for AUD; clinical interview is the diagnostic standard.
- MAT · Medications for alcohol use disorder
- Three FDA-approved medications: naltrexone (oral or monthly Vivitrol injectable; opioid-receptor antagonist that reduces craving and heavy-drinking days), acamprosate (supports abstinence after withdrawal), and disulfiram (Antabuse, creates a deterrent acetaldehyde reaction with alcohol). All three combined with counselling outperform counselling alone in randomized trials. See our MAT treatment page.
- ASAM level of care
- The American Society of Addiction Medicine framework classifying treatment intensity from Level 0.5 through Level 4. CAGE positive screens typically map to Level 1.0 outpatient or Level 2.1 IOP (intensive outpatient). High scores (3–4) plus heavy daily intake may indicate Level 3.7-WM medically supervised withdrawal management. See our treatment types index.
- SBIRT
- Screening, Brief Intervention, and Referral to Treatment — SAMHSA public-health workflow. CAGE is one of the brief screens (alongside AUDIT-C) that triggers brief intervention at positive scores. Brief intervention is a 5–15 minute structured conversation with a clinician.
- AUDIT-C
- 3-item subset of AUDIT-10 covering only consumption frequency, typical quantity, and binge frequency. Faster than full AUDIT, more sensitive than CAGE for current drinking patterns, less sensitive than CAGE for established AUD. NIAAA recommends CAGE or AUDIT-C as quick screens; full AUDIT-10 if more granular assessment is needed.
CAGE FAQ: anonymity, scoring, AUDIT comparison, withdrawal safety
Ten common questions about CAGE — whether the quiz is truly anonymous, what specific scores mean clinically, how CAGE differs from AUDIT-10 (the longer alcohol screener) and DSM-5 alcohol use disorder diagnosis, when to use a different instrument (AUDIT for past-12-month current drinking, TWEAK for pregnancy), and the alcohol-specific withdrawal safety facts that matter most at high scores.
Is this CAGE quiz anonymous?
Yes. Answers and scores are computed entirely in your browser. Nothing is sent to our servers. No email required. No account. No tracking of individual responses.
Is the CAGE a diagnosis?
No. CAGE is a brief screening tool, not a diagnosis. A score of 2 or higher is the standard cutoff for clinically significant indication of alcohol use disorder. Only a licensed healthcare professional can diagnose AUD using DSM-5 criteria and a clinical interview.
Who developed the CAGE?
Dr. John A. Ewing developed CAGE at the Bowles Center for Alcohol Studies, University of North Carolina at Chapel Hill, and published it in JAMA in 1984. The instrument is public-domain — no copyright restriction. NIAAA and SAMHSA list CAGE alongside AUDIT-C as a standard brief alcohol screen.
What does a CAGE score of 2 or higher mean?
A score of 2 to 4 is the clinical cutoff for positive screen, with sensitivity 71% and specificity 90% for alcohol use disorder. At a positive screen, follow up with the full AUDIT-10 (more sensitive at lower drinking levels), talk with a clinician, or call SAMHSA 1-800-662-4357. Severe scores (3-4) warrant full DSM-5 evaluation and possibly medically supervised detox.
How is CAGE different from AUDIT-10?
CAGE is 4 yes/no questions about lifetime drinking patterns; AUDIT-10 is 10 multi-point questions about the past 12 months. CAGE is faster but less sensitive in women, college students, and the elderly. AUDIT-10 catches hazardous drinking earlier (cutoff 8 vs CAGE cutoff 2). Best practice: CAGE as quick triage in time-pressured settings (ED, hospital admission); AUDIT-10 if CAGE positive or for routine primary-care screening.
Can this quiz replace a clinical evaluation?
No. A clinician will use CAGE results alongside a full clinical interview, medical history, and DSM-5 criteria. This screener is a useful starting point, not a substitute.
Is my data saved anywhere?
No. The quiz uses client-side JavaScript only. Questions and answers never leave the page. You can take a screenshot or print the page to share with a clinician, but nothing is stored on our servers.
What about alcohol withdrawal if I score 3 or 4 and want to stop?
Alcohol withdrawal can be life-threatening — seizures and delirium tremens are documented after abrupt cessation in heavy daily users (typically >8 standard drinks/day for >3 weeks). Do not quit alone. Three FDA-approved medications for alcohol use disorder (naltrexone, acamprosate, disulfiram) combined with medically supervised withdrawal at ASAM Level 3.7-WM substantially reduce relapse and mortality. Call SAMHSA 1-800-662-4357 to find a medically supervised detox program.
Does CAGE work for women, the elderly, or college students?
CAGE has documented lower sensitivity in women, the elderly, and college-age populations than in the general adult population. For these groups, AUDIT-10 or AUDIT-C is generally preferred. The TWEAK (5 questions) is a CAGE-derivative validated specifically for pregnant women. A CAGE score of 0 in these groups should not rule out alcohol concern if other clinical signs are present.
Is this the same CAGE my primary-care doctor might use?
Yes — exactly the same 4 items with the same scoring. CAGE lives inside the SAMHSA SBIRT toolkit, the NIAAA Core Resource on Alcohol, and most major US EHR systems. It is among the most-used quick alcohol screens in US primary care, emergency departments, and hospital admission workflows because brevity matches workflow.
CAGE primary sources: Ewing 1984 JAMA, NIAAA, SAMHSA, ASAM, NIDA
Every fact and threshold on this page traces to one of the 9 sources below: Ewing's original 1984 JAMA publication of CAGE, the NIAAA Core Resource on Alcohol and Rethinking Drinking guides, the SAMHSA helpline and SBIRT workflow, ASAM's national practice guideline for alcohol withdrawal management, and the SAMHSA treatment locator.
- Ewing JA. Detecting alcoholism: the CAGE questionnaire. JAMA. 1984;252(14):1905-1907.
- NIAAA Core Resource on Alcohol — Screen and Assess: Use Quick, Effective Methods.
- NIAAA Rethinking Drinking (low-risk limit calculator).
- 988 Suicide & Crisis Lifeline.
- SAMHSA National Helpline (1-800-662-4357).
- SAMHSA — Screening, Brief Intervention, and Referral to Treatment (SBIRT).
- SAMHSA treatment locator (findtreatment.gov).
- ASAM National Practice Guideline.
- NIDA — Comorbidity: Substance Use Disorders and Other Mental Illnesses.
reviewedBy schema appears in the structured data, and this notice stays visible. We do not provide medical advice and we do not replace evaluation by a licensed clinician. Medical emergency: 911. Crisis: call or text 988. Free, confidential substance-use guidance 24/7: SAMHSA 1-800-662-4357. Alcohol-use-disorder treatment records are protected under 42 CFR Part 2 and HIPAA.
