The CAGE-AID is a four-item conjoint screen for problematic alcohol and other drug use. It takes the four classic CAGE questions - feeling the need to cut down, being annoyed by criticism, feeling guilty, and using first thing in the morning - and widens each one from drinking alone to drinking or drug use. A single instrument therefore covers both alcohol and drugs, including illegal drugs and prescription medicines used other than as prescribed.
It is a lifetime screen, not a severity measure. A positive result says that a substance problem may be present and warrants proper assessment; it does not say which substance, how severe the problem is, or whether it is current. Because the questions refer to any point in the person''s life, the score does not fall when someone recovers - it is a gate to further assessment, never a monitoring tool.
02 - Origin & purpose
Where it comes from.
The parent instrument, the CAGE, was described by John Ewing in 1984 as a brief, non-confrontational way of raising alcohol use in general medical settings. Its four questions had been circulating since the late 1960s and were first validated by Mayfield and colleagues in 1974. The CAGE became one of the most widely used alcohol screens in medicine precisely because it is short, memorable and easy to embed in a consultation.
In 1995 Richard Brown and Laura Rounds at the University of Wisconsin adapted the questions to include drugs, reasoning that primary care patients with drug problems were being missed when clinicians only asked about alcohol. They validated the conjoint version against DSM-III-R substance use disorder diagnoses in 124 patients of a community family practice, and found it more sensitive - though somewhat less specific - than the CAGE for the combined detection of alcohol and drug disorders.
03 - Scoring & cutoffs
How scoring works.
Each of the four questions is answered yes or no, and each yes scores one point, giving a total from 0 to 4. Two thresholds are in use. The conventional cutoff is 2 or more, which in the original validation gave a sensitivity of 0.70 and a specificity of 0.85. In primary care, a cutoff of 1 or more is often preferred because it raises sensitivity to 0.79 (at the cost of specificity, 0.77), and a screen is usually the wrong place to miss cases. A positive screen is not a diagnosis: it should prompt a fuller assessment - a substance-focused history, and instruments such as the AUDIT for alcohol or the DAST-10 for drugs.
Score
Severity
Interpretation
0–1
Low concern
Below the threshold that warrants further assessment.
2–4
Clinically significant
Two or more positive answers suggest a clinically significant substance problem; assess further.
04 - Validation evidence
How well it performs.
The evidence base rests on the original criterion-validity study by Brown and Rounds (1995), which compared the CAGE-AID against DSM-III-R substance use disorder diagnoses in 124 primary care patients. The conjoint version detected more true cases than the alcohol-only CAGE across sex, income and education groups, at the cost of some specificity. Formal reliability statistics (internal consistency, test-retest) have not been widely reported for the conjoint version; the psychometric case rests on criterion validity.
0.79
Sensitivity (cutoff >=1)
Against DSM-III-R substance use disorder diagnoses; Brown & Rounds 1995
0.77
Specificity (cutoff >=1)
Same criterion study, 124 primary care patients
0.70 / 0.85
Sensitivity / specificity (cutoff >=2)
The conventional threshold trades sensitivity for specificity
DSM-III-R
Criterion standard
Structured diagnostic comparison in a community family practice
-Distinguishing alcohol from drug problems - the conjoint wording deliberately blurs them
-Detecting hazardous drinking below the disorder threshold - the AUDIT covers this, the CAGE family does not
07 - Confidence & precision
Reading the score with care.
The CAGE-AID is a binary screen, so the usual precision statistics do not apply: no standard error of measurement or minimally important change has been established, and none would be meaningful for a four-point lifetime score. Precision here means the trade-off between the two cutoffs - at 1 or more you accept roughly one false positive in four to catch about four cases in five; at 2 or more you miss more cases but cut false positives. Choose the threshold by what a missed case costs in your setting, and treat every positive as the start of an assessment, not a conclusion.
08 - Limitations
What it cannot tell you.
The lifetime timeframe cannot distinguish a resolved problem from a current one, and makes the score useless for monitoring.
A positive screen does not identify which substance is involved.
Criterion validation rests on a single primary care sample of 124 patients using DSM-III-R criteria.
The parent CAGE is known to be less sensitive in women and older adults, and the conjoint version has not resolved this.
Face-valid yes/no items are easy to answer defensively when someone does not want the problem found.
Not validated in adolescents or in pregnancy-specific screening.
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.