CAGE Score Calculator

CAGE asks nothing about how much a person drinks. Its four yes or no items instead probe the behavioural and psychological signature of dependence: whether the person has felt they should Cut down, whether they have been Annoyed by others criticising their drinking, whether they have felt Guilty about it, and whether they have needed a morning Eye-opener. Each yes scores 1 point, giving a total of 0 to 4, and 2 or more is the conventional positive screen. At that cutoff the questionnaire performs at roughly 86 percent sensitivity and 93 percent specificity in typical primary care samples, though reported ranges across settings are wider, from about 71 to 93 percent sensitivity and 77 to 97 percent specificity, because predictive value tracks the underlying prevalence of alcohol use disorder in the population being screened. The Eye-opener item carries the most diagnostic weight on its own, because drinking on waking to settle tremor or nausea implies overnight withdrawal and therefore physiological dependence rather than merely hazardous use. The main structural limitation is the time frame. The questions are phrased over a lifetime, so someone with ten years of stable sobriety will still screen positive, and someone drinking heavily but without social friction or guilt can screen negative. CAGE is also insensitive to binge patterns and to drinking that is harmful but not yet dependent, and validation was strongest in middle-aged men, with weaker performance reported in women and younger adults. Some settings, including obstetric care, drop the threshold to a single positive answer.

John Ewing, founding director of the Center for Alcohol Studies at the University of North Carolina at Chapel Hill, developed the four questions with Beatrice Rouse from interviews with hospital patients in the late 1960s. The goal was a screen brief enough that a physician would actually use it during an ordinary consultation, and unthreatening enough that patients would answer honestly. The pair presented the instrument in 1970, but the definitive description appeared only in 1984, when Ewing published it in the Journal of the American Medical Association along with the mnemonic that made it memorable. Publication in JAMA gave the questionnaire immediate reach, and it was validated repeatedly through the 1980s and 1990s against DSM-III and later DSM criteria across primary care, emergency, psychiatric, and surgical populations, with translations into more than twenty languages. Variants followed as its limits became clear. CAGE-AID, published by Richard Brown and James Rounds in 1995, extends each question to cover drug use as well as alcohol. The T-ACE, described by Robert Sokol and colleagues in 1989, replaced the guilt item with a tolerance question for use in pregnancy, where sensitivity matters more than specificity. Since the 1990s the World Health Organization AUDIT and its three-item consumption subset have largely displaced CAGE for detecting hazardous drinking before dependence develops, while CAGE retains a role as a rapid probe for established alcohol use disorder.