The validity of self-rated psychotic symptoms in depressed inpatients

A B S T R A C T Background: Self-ratings of psychotic experiences might be biased by depressive symptoms. Method: Data from a large naturalistic multicentre trial on depressed inpatients (n = 488) who were assessed on a biweekly basis until discharge were analyzed. Self-rated psychotic symptoms as assessed with the 90-Item Symptom Checklist (SCL-90) were correlated with the SCL-90 total score, the SCL-90 depression score, the Beck Depression Inventory (BDI), the Hamilton Depression Rating Scale 21 item (HAMD-21) total score, the Montgomery Asberg Depression Rating Scale (MADRS) total score and the clinician-rated paranoid-hallucinatory score of the Association for Methodology and Documentation in Psychiatry (AMDP) scale. Results: At discharge the SCL-90 psychosis score correlated highest with the SCL-90 depression score (0.78, P < 0.001) and with the BDI total score (0.64, P < 0.001). Moderate correlations were found for the MADRS (0.34, P < 0.001), HAMD (0.37, P < 0.001) and AMDP depression score (0.33, P < 0.001). Only a weak correlation was found between the SCL-90 psychosis score and the AMDP paranoid-hallucinatory syndrome score (0.15, P < 0.001). Linear regression showed that change in self-rated psychotic symptoms over the treatment course was best explained by a change in the SCL-90 depression score (P < 0.001). The change in clinician-rated AMDP paranoid-hallucinatory score had lesser influence (P = 0.02). Conclusions: In depressed patients self-rated psychotic symptoms correlate poorly with clinician-rated psychotic symptoms. Caution is warranted when interpreting results from epidemiological surveys using self-rated psychotic symptom questionnaires as indicators of psychotic symptoms. Depressive symptoms which are highly prevalent in the general population might influence such self-ratings.