Background. The International Personality Disorder Examination (IPDE) has been developed as a standardized interview for personality disorders. While it has good psychometric properties, its length makes it difficult to use in the community in population research, particularly outside psychiatric settings. The informant-based Standard Assessment of Personality (SAP), which has been in use since 1981, could serve as a valid screen to detect likely personality disordered individuals who would then receive a definitive diagnosis by IPDE. This study aimed to compare the two instruments in their capacity to detect personality disorder according to ICD-10 taxonomy and to estimate the efficiency of the use of the two together in a case-finding exercise.

Method. Ninety psychiatric out-patients in Bangalore, India, were assessed for personality disorder using the two methods. Assessment was conducted by a pair of trained interviewers in random order and by random allocation to interviewer.

Results. Overall agreement between the two instruments in the detection of ICD-10 personality disorder was modest (kappa = 0·4). The level of agreement varied according to personality category, ranging from kappa 07·66 (dependent) to kappa 0·09 (dyssocial). The SAP proved to have a high negative predictive value (97%) for IPDE as the gold standard, suggesting its potential as a screen in samples where the expected prevalence of personality disorder is low.

Conclusion. A two-stage approach to epidemiological studies of personality disorder may be practicable.