Previous studies have documented high but variable rates of DSM personality disorders (axis II) in clinical samples of substance abusers. Distinguishing between personality disorder symptoms that are independent versus substance-related (SR) is a particular challenge for diagnosing comorbid axis II disorders in substance abusers. DSM-IV guidelines currently recommend excluding axis II symptoms that are accounted for by an axis I disorder, including a substance use disorder. In this study, axis II diagnoses were made on a heterogenous clinical sample of 370 patients entering treatment for substance use disorders. Axis II diagnoses were made according to DSM-III-R criteria using the Structured Clinical Interview for DSM-III-R (SCID-II), which was modified to determine, on an item-by-item basis, whether symptoms were attributed to subjects' substance use disorders or independent of these disorders. The majority (57.0%) of substance use disorder patients met criteria for at least one comorbid axis II disorder, with cluster B (45.7%) being particularly prominent, especially antisocial personality disorder (ASP) (27.0%) and borderline personality disorder (BPD) (18.4%). Notably, inclusion of SR symptoms led to a substantial number of newly diagnosed cases, especially for ASP (19.2%) and BPD (11.4%). Including SR symptoms improved the reliability of ASP and did not change the reliability of BPD diagnoses. Generally, patients with SR and independent personality disorders had a similar clinical profile.