The Alcohol Use Disorders and Associated Disabilities Interview Schedule (AUDADIS; (Grant et al., 1995; Grant et al., 2003) was used to diagnose lifetime alcohol dependence (AD) following the DSM-IV guidelines, i.e., three or more dependence criteria within a 12-month period (American Psychiatric Association, 2000). Since DSM-5 will combine AD and alcohol abuse (AA) (American Psychiatric Association, 2010), we also included “any AUD”, based on presence of AD or AA (one or more abuse criteria in the absence of lifetime AD). As reviewed previously (Hasin et al., 2006), reliability and validity of AUDADIS-IV alcohol diagnoses in international general population samples ranges from good to excellent. We also included count phenotypes: AD severity (number of AD criteria endorsed, range 0-7) and AUD severity (number of AD and AA criteria endorsed, range 0-11). Test-retest reliability of AUDADIS count variables is good to excellent in general and clinical samples (intraclass correlation coefficients [ICC]=0.63-0.86;(Grant et al., 1995; Hasin et al., 1997a). Combining AD and AA criteria was supported this sample, showing that the alcohol criteria were unidimensional (Shmulewitz et al., 2010).
