The SSADDA was administered in a computer-assisted format to obtain lifetime psychiatric diagnoses. The SSADDA has been shown to have good inter-rater reliability (κ = 0.50–0.60, with the exception of bipolar disorder, κ = 0.43) and good-to-excellent retest reliability at an approximately two-week interval (κ = 0.50–0.76, with the exception of MDD, κ = 0.49) for the five index psychiatric diagnoses in this study. The SSADDA had excellent inter-rater reliability for cocaine and opioid dependence (κ = 0.83 and 0.91, respectively) and fair-togood inter-rater reliability for alcohol, cannabis, and stimulant dependence (κ = 0.48–0.66), as well as good-to-excellent retest reliability (κ = 0.66–0.94) for these diagnoses (Pierucci-Lagha et al., 2005, 2007). Although designed based on the specific diagnostic criteria of the American Psychiatric Association’s (1994) DSM-IV for each psychiatric disorder, the validity of the SSADDA for psychiatric disorders and substance dependence disorders has not been established. In addition to using the five psychiatric diagnostic categories as variables, a separate variable was constructed based on the presence of more than one psychiatric disorder (i.e., multiple psychiatric comorbidity), in light of findings from Mills and colleagues (2007).
