Validation of a DIS-III-R rescoring algorithm for DSM-IV alcohol-use disorders

The aim of this study was to evaluate the validity of the Diagnostic Interview Schedule, Version III, Revised (DIS-IIIR), rescoring algorithm when compared with Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), diagnoses established using the Composite International Diagnostic Interview (CIDI). The research question is "Does our DIS-III-R rescore accurately reflect DSM-IV diagnoses established independently using the CIDI?"

One hundred subjects were administered the CIDI and the DISIII-R alcohol-use modules. The DIS-III-R criteria were rescored to the DSM-IV diagnosis using our published algorithm. The specificity of the rescore, along with the sensitivity for abuse and dependence diagnoses, was computed.

The rescoring algorithm showed good sensitivity for lifetime alcohol abuse (60.4%) and dependence (80.0%) but poor specificity (0.0%). Current alcohol dependence was diagnosed with high sensitivity (100.0%) and moderate specificity (40.2%). Current alcohol abuse was diagnosed with poor sensitivity (18.2%).

Our algorithm can be used to rescore DIS-III-R data to DSM-IV alcohol-use diagnoses, but one must use caution with respect to the timeframe of alcohol-use disorders and the specific diagnosis.