An analysis of the relationship between the borderline, schizotypal and paranoid personality disorders

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)



First Committee Member

Catherine J. Green - Committee Chair


This study investigated factors related to severity, overlap, and ability to distinguish between the Borderline, Schizotypal and Paranoid personality disorders. Clinicians completed a Millon Diagnostic Personality Schedule (MDPS) and a DSM-III clinical diagnosis on 307 patients while the patients completed a Millon Clinical Multiaxial Inventory (MCMI).In relation to the Borderline disorder, subjects with elevations on the MCMI Borderline scale were found to have a significantly higher level of clinical symptomatology than subjects with mild personality disorders. An association was found between the Borderline and the Passive-Aggressive, Dependent, Histrionic, and Avoidant personality disorders. In addition, a high degree of convergence between the three assessment approaches was found for the Borderline personality disorder. Last, the MDPS Borderline attributes were able to differentiate subjects with Borderline from Schizotypal and Paranoid disorders.For subjects with an MCMI Schizotypal diagnosis, a significantly higher level of clinical symptomatology was found when compared to subjects with mild personality disorders. A high level of covariance was found between the Schizotypal, and Avoidant and Schizoid disorders. A relatively high level of convergence between assessment approaches was found when mean base rate scores on the Schizotypal scales of the MCMI and MDPS were examined. Last, MDPS attributes for the Schizotypal disorder were able to differentiate subjects with a Schizotypal diagnosis from those with a Paranoid diagnosis, while some overlap was found with the Borderline disorder.The findings for the Paranoid disorder are unclear in regards to its level of severity, overlap with other personality disorders, and its ability to be distinguished from other disorders. The results obtained in this study seem to be a function of the lack of consensus regarding the definition of the Paranoid disorder, the state of the art of instrumentation for the diagnosis of the disorder, sample size, and patient characteristics which influence test results and accuracy of clinician diagnosis (i.e., defensiveness).


Psychology, Clinical

Link to Full Text


Link to Full Text