Maybe you've been wondering for a while. You've tried cutting back and it hasn't stuck. Or someone close to you said something, and now you can't stop turning it over. Or you're just quietly asking yourself: would this actually count? That's a fair question — and it has a real, clinical answer.

Alcohol use disorder (AUD) is diagnosed against a specific checklist of 11 criteria. You don't need a doctor to read through them first. Understanding what clinicians are actually looking for can help you figure out whether what you're experiencing lines up with a diagnosable condition — and whether it's time to talk to someone about it.

Key Takeaways

• Two criteria is the threshold. Under DSM-5, meeting just 2 of 11 specific criteria within a 12-month period is enough to qualify for an AUD diagnosis — most people are surprised by how low that bar is.
• Screening tools don't diagnose — they flag. A positive AUDIT or AUDIT-C score means you should get a full assessment, not that you definitely have AUD; the formal diagnosis requires evaluating all 11 DSM-5 criteria.
• Severity is graded mild, moderate, or severe. Meeting 2–3 criteria is mild AUD, 4–5 is moderate, and 6 or more is severe — but which criteria you meet matters as much as how many.
• Withdrawal is the highest-risk single criterion. Someone with only mild AUD who experiences withdrawal has a dramatically worse prognosis than someone with mild AUD who doesn't — the label alone doesn't tell the whole story.
• A formal diagnosis dramatically increases your odds of getting help. In a large study, people who received an official AUD diagnosis were more than ten times as likely to be offered medication treatment than those who only screened positive [1]✓ Verified knowledgeGreen et al. (2023) — Closing care gap.
• Many people who meet criteria are never told. Only about 1 in 10 people who screen positive for unhealthy alcohol use end up with a documented diagnosis — the gap between screening and care is a systemic problem, not a personal failing [1]✓ Verified knowledgeGreen et al. (2023) — Closing care gap.

How do you know when drinking has become AUD?

The current clinical standard — the one your doctor, a psychiatrist, or an addiction specialist would use — is the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). It replaced an older system that split alcohol problems into two separate diagnoses: "alcohol abuse" and "alcohol dependence." If you've seen both terms and wondered how they relate, the distinction between alcohol abuse and alcoholism is worth understanding, because the language shifted significantly in 2013.

Under DSM-5, there's one diagnosis — AUD — and it's graded by how many of 11 criteria you meet within a 12-month period. Two criteria is the minimum. That's it. You don't need to be drinking every day, you don't need to have hit a dramatic "rock bottom," and you don't need to have lost a job or a relationship. Two criteria, same 12 months.

The 11 criteria fall into four conceptual clusters. Reading through them honestly is the closest thing to a self-assessment that the clinical framework offers.

The 11 DSM-5 criteria — what clinicians are actually asking

Loss of control

• Drinking more or longer than you intended. You plan on two drinks and end up having six. You meant to drink only on weekends and find the pattern expanding. This criterion captures what people often describe as the internal "stop" signal going quiet.
• Repeated failed attempts to cut back. You've tried — maybe many times — and haven't been able to sustain the change. This isn't about not wanting to stop; it requires that you've actually made attempts that didn't hold.
• A great deal of time spent obtaining, using, or recovering. When a significant portion of your day is organized around drinking — getting it, doing it, or recovering from it — this criterion is met. In severe cases, most waking hours follow this pattern.
• Craving — a strong urge to drink. This criterion was added to DSM-5 (it didn't exist in the older framework) based on neuroscience showing that craving reflects a distinct brain process separate from physical dependence. If you find yourself preoccupied with thoughts of drinking when you're not, this is what that criterion is pointing at.

Social and role impairment

• Failing to meet major responsibilities at work, school, or home. Missing work because of hangovers, neglecting family obligations, or falling behind in school — and this happening more than once.
• Continuing to drink despite ongoing relationship or social problems it's causing. Arguments with a partner about your drinking, distance from family, lost friendships — and you recognize the connection but continue anyway.
• Giving up or cutting back on things that used to matter. Hobbies abandoned, social events avoided, opportunities declined — because drinking has become the organizing priority.

Risky use

• Drinking in physically dangerous situations. Driving while impaired, operating machinery, or drinking in contexts where impairment creates direct physical risk — and doing it repeatedly.
• Continuing to drink despite knowing it's making a physical or mental health problem worse. A doctor has told you your liver, your depression, or your stomach is being damaged by alcohol — and you continue. This criterion requires awareness, not just the presence of a medical problem.

Physical dependence

• Tolerance — needing more to get the same effect, or getting less effect from the same amount. The person who once felt intoxicated after three drinks now needs eight to feel it. Or their usual amount just doesn't do what it used to. Tolerance is worth understanding carefully: in one prospective study, it was by far the most commonly endorsed criterion (50.3% of participants), and removing it from the criterion set reduced lifetime AUD prevalence estimates by roughly 40% [2]✓ Verified knowledgeSubramaniam et al. (2022) — Differences prevalence profile. That raises real questions about whether tolerance, as currently defined, always signals a clinical problem versus a physiological adaptation in heavy drinkers.
• Withdrawal — physical symptoms when you stop or cut back, or drinking to prevent them. Tremor, sweating, anxiety, nausea, and in severe cases, seizures. Someone who drinks in the morning to stop shaking meets this criterion even if they've never gone through a formal withdrawal episode. Withdrawal is the single highest-risk criterion in the entire list. Among people with mild-to-moderate AUD, the presence of withdrawal was associated with an adjusted hazard ratio of 11.62 for progression to severe AUD [3]✓ Verified knowledgeYue et al. (2026) — Rates diagnosis treatment — nearly double the risk seen in mild-to-moderate AUD without withdrawal. If you're experiencing withdrawal symptoms, that changes the clinical picture significantly, and it's a medical situation worth taking seriously.

If you recognize several of these in your own life, you're not alone — and you're not broken. These are clinical criteria, not character flaws. The pattern has a name, and it has treatments.

How severity is graded — and why it's more than a number

The DSM-5 uses criterion count to assign a severity level:

These cutoffs are useful shorthand, but they have a documented limitation: two people can share the same severity label and have very different clinical trajectories. A person with mild AUD (2–3 criteria) who endorses withdrawal has a dramatically worse prognosis than someone with mild AUD who endorses only tolerance and craving [3]✓ Verified knowledgeYue et al. (2026) — Rates diagnosis treatment. The which matters as much as the how many.

This is one reason that understanding the stages of alcohol use disorder — not just the severity label — gives a fuller picture of where someone is and where they're likely headed without intervention.

What screening tools actually measure (and what they don't)

You may have taken a questionnaire at a doctor's office or online and gotten a score. It's important to understand what that score does and doesn't tell you.

Screening tools identify people who may have AUD. They don't diagnose it. A positive screen is the beginning of the assessment process, not the end.

The AUDIT

The Alcohol Use Disorders Identification Test (AUDIT) is a 10-item questionnaire developed by the World Health Organization and validated across primary care, emergency departments, and community settings. It covers how often and how much you drink, whether you've experienced loss of control, and whether drinking has caused problems in your life. Scores run from 0 to 40.

• A score of 8 or higher suggests hazardous or harmful drinking and warrants a closer look.
• A score of 15 or higher suggests likely alcohol dependence in most populations.

A systematic review of 35 studies (nearly 80,000 participants) found that an AUDIT score of 8 or higher produces a likelihood ratio of 6.5 for DSM-5 AUD — meaning a positive result makes the diagnosis meaningfully more probable [4]✓ Verified knowledgeCaneva et al. (2020) — Cognitive impairments early. Notably, the AUDIT performs better for identifying AUD in women (likelihood ratio 6.9) than in men (likelihood ratio 3.8) [4]✓ Verified knowledgeCaneva et al. (2020) — Cognitive impairments early — a clinically important difference that means the same score carries different weight depending on sex.

The AUDIT-C

The AUDIT-C is a three-item version of the AUDIT that keeps only the consumption questions. It's widely embedded in electronic health records because it's fast. Scores of 3 or higher for women and 4 or higher for men suggest hazardous drinking.

The AUDIT-C is a reasonable first flag, but it's considerably less useful than the full AUDIT for identifying AUD specifically. Its likelihood ratio for AUD at standard cutoffs is only about 1.8–2.0 [4]✓ Verified knowledgeCaneva et al. (2020) — Cognitive impairments early — a modest signal. The AUDIT-C was designed to detect excessive drinking, not DSM-5 AUD. These are different constructs. A positive AUDIT-C means you should get a fuller assessment, not that you have a disorder.

Other tools for specific situations

• CAGE (four questions about cutting down, annoyance at criticism, guilt, and eye-opener drinking) is quick and widely recognized, but it was developed before DSM-5 and performs best for identifying established dependence. It's less sensitive for mild-to-moderate AUD.
• T-ACE and TWEAK were specifically validated for use during pregnancy, where any alcohol use carries fetal risk and standard AUDIT cutoffs don't apply. If you're pregnant, these are the appropriate tools.
• CRAFFT is the validated screening tool for adolescents aged 12–21. Standard adult cutoffs should not be applied to younger people.
• MAST (Michigan Alcoholism Screening Test) and its shorter versions may be more sensitive for detecting long-standing AUD in older adults, whose alcohol pharmacokinetics differ — they experience greater impairment at lower blood alcohol concentrations.

Visual suggestion: A comparison chart showing AUDIT vs. AUDIT-C vs. CAGE — what each measures, validated populations, and cutoff scores — would help readers quickly understand which tool applies to their situation.

What happens between a positive screen and a real diagnosis

Here's where a lot of people fall through the cracks. In a cohort of more than 114,000 people who screened positive for unhealthy alcohol use, only about 10% had a formal AUD diagnosis documented in their medical record [1]✓ Verified knowledgeGreen et al. (2023) — Closing care gap. That's not because 90% of them didn't have AUD. It's because the system often stops at the screening step.

The SBIRT framework — Screening, Brief Intervention, and Referral to Treatment — is designed to connect a positive screen to a graduated clinical response:

1. Screening identifies people with hazardous drinking or possible AUD using a validated tool.
2. Brief Intervention is a structured 5–15 minute motivational conversation for people whose drinking is risky but who don't meet AUD criteria. It's not a treatment for AUD — it's an intervention for risky drinking.
3. Referral to Treatment connects people who do meet AUD criteria to appropriate care — which may include medication, behavioral therapy, or a higher level of structured treatment.

In practice, many clinical systems record the AUDIT-C score and move on, without completing the diagnostic evaluation needed to generate a formal diagnosis or initiate treatment. The result: a documented positive screen with no clinical follow-through.

Why does the formal diagnosis matter so much? Because it's a gateway. In the same large cohort, receiving an AUD diagnosis increased the adjusted odds of being prescribed medication by more than tenfold (aOR = 10.68) and the odds of receiving psychotherapy by over 50% (aOR = 1.57) [1]✓ Verified knowledgeGreen et al. (2023) — Closing care gap. The act of diagnosis is itself a clinical intervention.

Heavy drinking vs. AUD — they're not the same thing

You can drink at levels that carry significant health risk without meeting criteria for AUD. The NIAAA defines hazardous drinking as more than 4 drinks on any single day or more than 14 drinks per week for men; more than 3 drinks on any single day or more than 7 per week for women. Drinking above those limits is a target for intervention — but it doesn't constitute AUD unless at least 2 DSM-5 criteria are also present [5]✓ Verified knowledgeGhandour et al. (2020) — Impact dsm classification.

A person who regularly drinks 5–6 drinks per evening but has never experienced loss of control, craving, withdrawal, or functional impairment may be at real health risk — liver disease, cardiovascular disease, cancer — without having AUD. They need a different kind of help than someone with moderate or severe AUD. Understanding the warning signs of a developing problem can help clarify where on that spectrum your drinking falls.

The formal diagnosis requires assessing all 11 DSM-5 criteria. One validated tool for doing this in routine care is the Alcohol Symptom Checklist, which patients can complete themselves and which demonstrated strong test-retest reliability in primary care settings (intraclass correlation coefficient = 0.82) [5]✓ Verified knowledgeGhandour et al. (2020) — Impact dsm classification. It's the kind of structured self-report that can move a clinical conversation from vague concern to specific criteria.

Visual suggestion: A DSM-5 severity ladder diagram — showing the 11 criteria organized by domain, with the 2/4/6 threshold markers for mild/moderate/severe — would give readers a clear visual anchor for the assessment framework.

What remission looks like — and what the research says about recovery

If you've been in a period of not drinking, or drinking significantly less, the DSM-5 has formal remission specifiers that matter for insurance coverage, disability determinations, and treatment planning:

• Early remission: No criteria met (except craving) for at least 3 months but less than 12 months.
• Sustained remission: No criteria met (except craving) for 12 months or longer.
• In a controlled environment: Abstinence in a setting where alcohol access is restricted — incarceration, residential treatment — which doesn't carry the same prognostic weight as community-based remission.
• On maintenance therapy: Taking a prescribed medication (naltrexone, acamprosate) as part of treatment.

The natural history data are worth knowing. A prospective cohort study tracking DSM-5 AUD symptoms from late adolescence to age 42 found that symptoms peaked at age 24 and that 67% of people remitted by age 42 [6]✓ Verified knowledgeKerr et al. (2025) — Natural history dsm. Most people do get better. At the same time, 11–13% showed persistent symptoms from late adolescence onward [6]✓ Verified knowledgeKerr et al. (2025) — Natural history dsm — a subgroup that needs sustained, serious support. Recovery is the most common outcome, but it's not automatic, and for some people it requires real clinical engagement.

Who gets diagnosed — and who gets missed

The gap between who has AUD and who gets diagnosed isn't random. It falls along predictable lines. Research documents lower odds of receiving a formal AUD diagnosis among women, racial and ethnic minorities, younger adults, and people in economically disadvantaged areas — despite comparable or higher rates of screening positivity in some of these groups [1]✓ Verified knowledgeGreen et al. (2023) — Closing care gap.

In the Veterans Health Administration — a system with mandatory universal annual alcohol screening — the largest gaps between survey-based AUD prevalence and documented diagnosis rates were among patients aged 18–34 (prevalence 22.4% vs. diagnosis rate 6.9%) and Hispanic/Latinx patients (17.7% vs. 7.6%) [3]✓ Verified knowledgeYue et al. (2026) — Rates diagnosis treatment. Even when screening is universal, the pathway from positive screen to formal diagnosis to treatment breaks down unevenly.

This matters for anyone reading this page. If you've been to a doctor, screened positive on a questionnaire, and never had a follow-up conversation about what that actually means — that's a gap in the system, not evidence that your drinking isn't a problem. You can ask directly: "Do I meet criteria for alcohol use disorder?" That question deserves a real answer.

Visual suggestion: A simple flowchart showing the pathway from screening → diagnostic assessment → diagnosis → treatment, with the documented drop-off points marked, would help readers understand where the system commonly fails and what to advocate for.

When does severity actually change what treatment looks like?

Severity matters — but not in the way most people assume. The mild/moderate/severe labels are useful starting points, but the specific criteria you meet shape the clinical picture more than the count alone. Withdrawal in a person with mild AUD changes the risk profile dramatically. Craving and tolerance in the absence of other criteria raise questions about whether the pattern will escalate.

For people with moderate or severe AUD, structured treatment options — including FDA-approved medications like naltrexone and acamprosate, behavioral therapies, and varying levels of care from outpatient to residential — are evidence-based and effective. The right level of care depends on the full clinical picture, not just the severity label.

Understanding what alcohol use disorder actually is — its neuroscience, its health consequences, and its treatability — is a useful foundation before that conversation with a clinician. The assessment is where that conversation starts.