Validation of a single screening question for problem drinking

The researchers hoped to confirm the sensitivity and specificity of a single screening question for problem drinking: "When was the last time you had more than X drinks in 1 day?", where X=4 for women and X=5 for men. Cross-sectional study. Adult patients presenting to 3 emergency departments in Boone County, Missouri, for care within 48 hours of an injury. The answers to the question were coded as never, more than 12 months ago, 3 to 12 months ago, and within the past 3 months. Problematic drinking was defined as either hazardous drinking (identified by a 29-day retrospective interview) or a past-year alcohol use disorder (defined by questions from the Diagnostic Interview Schedule). There was a 70% participation rate. Of 2517 interviewed patients: 29% were hazardous drinkers; 20% had a past-year alcohol use disorder; and 35% had either or both. Considering "within the last 3 months" as positive, the sensitivity of the single question was 86%, and the specificity was 86%. In men (n=1432), sensitivity and specificity were 88% and 81%; in women, 83% and 91%. Using the 4 answer options for the question, the area under the receiver-operating characteristic curve was 0.90. Controlling for age, sex, tobacco use, injury severity, and breath alcohol level in logistic regression models changed the findings minimally. A single question about the last episode of heavy drinking has clinically useful sensitivity and specificity in detecting hazardous drinking and alcohol use disorders.