Test-retest reliability of the Personality Diagnostic Questionnaire: Revised

Abstract

We describe the test-retest reliability of the Personality Diagnostic Questionnaire: Revised (PDQ-R), which was developed to assess DSM-III-R personality disorders (PD). Thirty patients, diagnosed clinically with PD, filled out the PDQ-R twice at 1 month intervals after having undergone 8 weeks of treatment leading to moderate resolution of their acute axis-I symptomatology. The chance-corrected agreement was almost perfect for all DSM-III-R PD (κ coefficients: 0.76-1.00). Dimensional scores for individual PD and the total PDQ-Q score were highly correlated (Pearson's correlation coefficients: 0.62-0.99) and the mean of scores did not significantly differ between test and retest. Based on these results, we propose that highly reliable PD diagnoses are possible, even using a self-report questionnaire, if the axis-I symptomatology of depression and anxiety have moderately subsided. These results clarify controversy over how and when reliable PD diagnoses may be obtained via a self-report questionnaire in a clinical sample.