Development and Validation of a Simple Risk Calculator for Alcohol-Related Adverse Outcomes: Results from a Composite Scale Approach in a Nationally Representative Sample

Abstract

We tested how 5 simple variables, routinely collected in primary care (i.e., the 3 AUDIT-C items, age, and sex), could constitute a clinician-friendly risk calculator tool of the 3-year risk of several alcohol-related adverse outcomes (i.e., alcohol use disorder, withdrawal symptoms, occurrence of tremors or seizures, and alcohol related interpersonal relationship problems, legal problems, and psychological problems) among non-alcohol-dependent individuals consuming alcohol. Data were drawn from a nationally representative sample of US adults, the National Epidemiologic Survey on Alcohol and Related Conditions (wave 1, 2001–2002; wave 2, 2004–2005). We used scaled and rounded composite scoring systems to combine information derived from these predictors and quantify the risks for each subject. Analyses were limited to 16,710 respondents without a lifetime history of alcohol use disorder who completed both interviews and had consumed alcohol during the year prior to wave 1 (development sample N = 8355, validation sample N = 8355). The risk equations calibrated well (Hosmer and Lemeshow test p values ≥ 0.072) and showed good predictive values (C indices ranging from 0.727 to 0.872) in the validation sample. This risk calculator can be clinically useful in primary care for identification of at-risk individuals, encourage respondents to reduce their drinking, and improve prevention.