DIAGNOSING ANTISOCIAL PERSONALITY DISORDER AMONG SUBSTANCE ABUSERS: THE SCID VERSUS THE MCMI-II

There is much controversy among social scientists and clinicians over the proper measurement of antisocial personality disorder (ASPD). The degree to which various diagnostic measures differ in their assessment of ASPD among substance abusers is not known. This study assessed the degree of agreement between a semistructured clinical interview and a self-report inventory on a diagnosis of ASPD among substance abusers. The Structured Clinical Interview for DSM-III-R (SCID-II), a clinically generated instrument, and the Millon Clinical Multiaxial Inventory (MCMI-II), a self-report inventory, were administered to 275 clients randomly assigned to two therapeutic communities (TCs). Based on the limited existing literature, it was hypothesized that there would be minimal agreement between the diagnosis of ASPD by the two scales. This hypothesis was supported. The kappa statistic indicated low agreement between the scales (kappa = 0.27), with the MCMI-II diagnosing ASPD more often than the SCID-II. The low agreement on a diagnosis of ASPD may be due to the different types of information collected by the two scales. The SCID-II emphasizes observable behavioral criteria, while the MCMI-II emphasizes pathological personality traits. The focus of the MCMI-II on pathological personality traits may more accurately diagnose ASPD in substance-abusing populations in which the majority of the clients have extensive criminal histories. Definite conclusions regarding the proper measurement of ASPD in substance-abusing samples is difficult without additional empirical evidence.