Measuring psychosis proneness in a nonclinical Korean population: is the Peters et al Delusions Inventory useful for assessing high-risk individuals?

Abstract

OBJECTIVES: We examined the psychometric properties of the Korean version of the Peters et al Delusions Inventory (PDI) 40 and investigated the distribution of delusional ideation in a nonclinical population. We also used the item response theory to evaluate the usefulness of the PDI in measuring the risk for psychosis. METHODS: A total of 310 nonclinical individuals completed the Korean PDI-40, the Magical Ideation Scale (MIS), and the Schizotypal Personality Scale (STA). In addition, 60 psychotic inpatients with delusions completed the PDI-40. Among 310 individuals, 124 participated in a follow-up study 6 months after completing their original questionnaire. RESULTS: The PDI-40 revealed a slightly skewed distribution, but the score range was similar to that of the British population. Scores were negatively correlated with age, but no sex differences were found. The Korean PDI-40 exhibited good internal consistency and test-retest reliability. The PDI was significantly correlated with the MIS and the STA. Ten components were extracted through a principal component analysis with varimax rotation. The test results using item response theory revealed 39 items as the items which individuals with very high level of psychosis proneness will answer as "yes." Moreover, all items yielded "above moderate" discrimination in terms of psychosis proneness. CONCLUSIONS: We confirmed the reliability and validity of the Korean PDI-40. The usefulness of the PDI-40 in a nonclinical population was replicated in the Korean sample. The PDI-40 can be used as an informative device when investigating "psychosis proneness" in a group at high risk for psychosis.