The CAGE questionnaire is a four-question screening tool designed to identify potential alcohol problems. Each letter in “CAGE” stands for a key word in one of the questions: Cut down, Annoyed, Guilty, and Eye-opener. It was one of the most widely used alcohol screening instruments in medicine for decades, though major health organizations now consider it outdated for routine screening.
The Four CAGE Questions
The questionnaire, developed by Dr. John Ewing in 1984, asks:
- C – Cut down: Have you ever felt you should cut down on your drinking?
- A – Annoyed: Have people annoyed you by criticizing your drinking?
- G – Guilty: Have you ever felt bad or guilty about your drinking?
- E – Eye-opener: Have you ever had a drink first thing in the morning to steady your nerves or get rid of a hangover?
Unlike questions that ask how many drinks you have per day, the CAGE focuses on behavioral and emotional consequences of drinking. The idea is that someone who feels guilty about their drinking, or who drinks first thing in the morning, is showing patterns that point toward a problem, regardless of how many drinks they report.
How Scoring Works
Each “yes” answer scores one point, and each “no” scores zero, giving a total between 0 and 4. A score of 2 or higher is traditionally considered clinically significant, meaning it warrants further evaluation for an alcohol use disorder. Some clinicians use a lower threshold of just 1 positive answer to cast a wider net and catch more people who might benefit from help.
A positive CAGE score is not a diagnosis. It signals that a more thorough assessment is appropriate. Someone scoring 2 or above has a higher likelihood of having an alcohol problem, but the screening alone can’t confirm one.
Accuracy and Known Limitations
The CAGE’s sensitivity (its ability to correctly identify people with alcohol problems) ranges from 43% to 94% across studies, while its specificity (correctly identifying people without problems) ranges from 70% to 97%. That’s an unusually wide spread, and it reflects a core weakness: the questionnaire performs inconsistently across different populations.
A study of over 1,300 primary care patients found that women, Mexican-American patients, and higher-income patients were consistently less likely to answer “yes” to CAGE questions even when they had a diagnosable alcohol use disorder. Gender and cultural differences in how people perceive problem drinking, and in the social consequences they experience, appear to drive this gap. Of the four questions, only the “cut down” question reliably predicted an alcohol use disorder across all demographic subgroups. The “guilty” question contributed almost nothing to prediction, and the “annoyed” and “eye-opener” questions were inconsistent.
These biases make the CAGE a poor “rule-out” test, meaning a negative result doesn’t reliably mean someone is in the clear.
Why It’s Considered Outdated
The National Institute on Alcohol Abuse and Alcoholism (NIAAA) now advises clinicians to avoid using the CAGE as a screening tool. The U.S. Preventive Services Task Force also does not recommend it, because the questionnaire only captures people who are already experiencing consequences of heavy drinking. It misses those who are drinking at risky levels but haven’t yet developed the guilt, criticism from others, or morning drinking that the questions ask about. In other words, it catches problems late rather than early.
More current tools like the AUDIT (Alcohol Use Disorders Identification Test) and its shorter version, the AUDIT-C, ask about drinking patterns and quantity in addition to consequences. This makes them better at identifying risky drinking before it escalates. The full AUDIT has 10 questions, which has historically made it harder to work into a brief office visit. But the AUDIT-C condenses the most useful items into just three questions, making it a practical alternative that outperforms the CAGE in most clinical settings.
The CAGE-AID: An Adaptation for Drug Use
A modified version called the CAGE-AID (Adapted to Include Drugs) adjusts each question to cover drug use alongside alcohol. For example, “Have you ever felt bad or guilty about your drinking?” becomes “Have you ever felt bad or guilty about your drinking or drug use?” The same four-question structure and scoring apply.
The CAGE-AID is more sensitive than the original CAGE, meaning it catches more people with substance use problems. This improvement holds across different sexes, income levels, and education backgrounds. The tradeoff is lower specificity: it flags more people who don’t actually have a disorder. Using a threshold of 1 positive answer instead of 2 increases sensitivity further but also increases false positives.
What to Know If You’re Asked These Questions
If a healthcare provider asks you CAGE questions during a visit, they’re conducting a brief screen, not making a judgment. The questions are designed to open a conversation about drinking patterns, and honest answers lead to better care. Clinicians sometimes encounter evasiveness around these topics, which is understandable given the stigma around alcohol use. But the screening exists to connect people with support early, not to label them.
You might also encounter these questions on intake forms at a new doctor’s office or during a routine checkup. If you score positive, expect a follow-up conversation rather than an immediate diagnosis. Your provider will likely ask more detailed questions about how much and how often you drink, whether it’s affecting your health or daily life, and whether you’ve tried to change your habits.