Using the alcohol use disorders identification test to predict hospital admission for alcohol‐related conditions in the Danish general population: a record‐linkage study

BACKGROUND AND AIMS: Most studies validating the alcohol use disorders identification test (AUDIT) have either assessed its factor structure and/or test–retest reliability or used diagnostic interviews as validators of current alcohol use disorders. The aim of the present study was to determine whether AUDIT and AUDIT‐Consumption (AUDIT‐C) scores are associated with subsequent risk of hospital admission for alcohol‐related disorders and diseases (ARDDs). DESIGN: We used a historical cohort study. Using national registers, survey respondents were tracked from 1 September 2011 to hospitalization for an ARDD, emigration, death, or 31 December 2018, whichever occurred first. SETTING: Denmark. PARTICIPANTS: Respondents (n = 4522) from a Danish national survey conducted in autumn 2011. MEASUREMENTS: Outcome was incident ARDD admission recorded in the National Patient Register. Predictors were AUDIT and AUDIT‐C scores, and covariates were age, gender, highest level of education and previous psychiatric disorder. FINDINGS: During the study period, 56 respondents had a first‐time ARDD admission. Respondents who scored above the 8‐point AUDIT cut‐off and respondents who scored above the 5‐point AUDIT‐C cut‐off had a significantly increased risk of being admitted for an ARDD compared with respondents who scored below the cut‐offs, (AUDIT: hazard ratio (HR), 4.72; 95% CI, 2.59–8.60; AUDIT‐C: HR, 7.97; 95% CI, 3.66–17.31). CONCLUSIONS: Scores above alcohol use disorders identification test (AUDIT) and AUDIT‐Consumption (AUDIT‐C) cut‐offs are associated with an increased risk of long‐term alcohol‐related hospital admissions. At widely used cut‐offs, the AUDIT‐C is a better predictor of alcohol‐related hospitalizations among members of the general population than the full AUDIT.