Discrepancy between self-reported and interviewed psychosis risk symptoms: auditory distortions are the most reliably reported symptom by self-report.

Aim: In the recent literature, there has been growing interest in assessment methods for detecting increased risk of developing psychosis. Self-report methods are popular but may lead to different results compared to clinical interviews.

Methods: The difference in psychosis risk scores was tested between self-reported psychosis risk symptoms (PROD-SR) and self-reported symptoms additionally confirmed by interview (PROD-SR + I). The symptom categories were derived from 12 common psychosis risk symptoms included in the PROD screening instrument. The data were collected by questionnaires and interviews conducted with 395 adolescents (mean age 15.3 years) in an early intervention and detection team, JERI, at Helsinki University Central Hospital, Finland.

Results: The results show a significant difference between the PROD-SR risk symptom sum scores and the PROD-SR + I risk symptom sum scores (N = 395; Z = -15.123; P < 0.001). In an item-by-item analysis, the item 'Disorders in connection with hearing' had the strongest kappa value (0.827) agreement between an interviewed and self-report psychosis risk item. Agreement in most items remained between slight and substantial (kappa values from 0.082 to 0.649).

Conclusions: The results suggest that there is a significant difference between psychosis risk symptom responses collected by self-report and self-report responses which are additionally confirmed by interview. Auditory disorders are the most reliably reported item with self-report.