AUDIT-C: three-item alcohol screen

What this assessment is

AUDIT-C is a three-item screen for risky alcohol consumption. It uses the consumption questions from the full 10-item AUDIT.

Katharine Bush and colleagues tested the brief form in a 1998 validation study to identify heavy drinking and active alcohol problems with fewer questions.

AUDIT-C is not the full AUDIT. It does not directly assess dependence-related experiences or alcohol-related harm.

Not a diagnosis.

Who it was designed for

The original study involved men receiving care in three U.S. Veterans Affairs clinics. Later studies expanded the evidence to women, general primary care, and other settings.

That history matters. A cutoff that worked in one sample may not give the same balance of missed cases and false positives elsewhere.

Use a threshold validated for the exact form, population, and purpose whenever possible.

How it is administered

AUDIT-C asks about alcohol use during the past year. It can be self-completed or asked as part of a clinical interview.

There is more than one current form. The original third question uses one high-quantity drinking threshold. The current U.S. Department of Veterans Affairs form changes that threshold according to sex and age.

That is a substantive difference, not a formatting detail.

Form matters.

A report should identify which wording and policy were used. Scores from different forms should not be treated as interchangeable without review.

What it measures

AUDIT-C covers three parts of alcohol consumption:

• how often a person drinks
• the usual amount consumed on a drinking day
• how often higher-quantity drinking occurs

It does not ask the seven full-AUDIT questions about impaired control, morning drinking, guilt, memory loss, injury, or concern from other people.

A brief result can start a conversation. It cannot describe the whole pattern.

How scoring works

Each response scores from 0 to 4. The three values are added for a total from 0 to 12.

Higher scores indicate a greater likelihood of risky alcohol use. There are no severity bands that work across every AUDIT-C form and population.

Complete all three questions when possible. Do not replace a missing answer with zero unless a defined protocol supports that choice.

How to interpret an AUDIT-C score

Interpretation starts with the form and its policy.

The AHRQ-hosted clinical form calls a result positive above 4 for men and above 2 for women. In other words, its decision points are 5 or higher for men and 3 or higher for women.

The current VA form uses a modified third question and a single positive cutoff of 5 or higher. The current VA and Department of Defense guideline uses the same policy.

Neither rule should be presented as universal. A service should choose its form and threshold before screening, document both, and keep them stable when comparing results over time.

Keep them separate.

A positive result means that further assessment is appropriate. It does not establish an alcohol use disorder.

What the score cannot tell you

AUDIT-C cannot show whether alcohol has caused injury, withdrawal, loss of control, relationship problems, or difficulty meeting responsibilities.

It cannot determine whether stopping suddenly could be dangerous. It also cannot choose treatment.

Follow-up should review drinking quantity and pattern, dependence-related features, harm, physical and mental health, medications, pregnancy, and the person's priorities.

Evidence and limitations

In the original VA study, AUDIT-C performed well as a brief screen for heavy drinking or active alcohol problems. The sample excluded women, which limits how directly those findings apply to other groups.

A primary care validation study found useful performance in men and women but supported different thresholds by sex. Other healthcare systems have chosen different policies to fit their population and workflow.

The main strength is brevity. The main limit is the same: three consumption questions leave out important information that the full AUDIT and a clinical assessment can provide.