AIM: The stability of delusional themes in psychotic disorders is only sparsely lit. The study aims to investigate the stability of delusional themes over a period of time in patients with first-episode non-affective psychosis.

METHODS: The data were drawn from a randomized controlled trial of 547 first-episode patients with schizophrenia spectrum disorders, the OPUS trial. A subsample of 411 patients met our inclusion criteria. They were interviewed at the time of first treatment contact, and at years 1, 2 and 5 after onset. The delusional themes of the 411 patients were classified by applying principal component analysis to the patients' scores on 12 items of delusions from the Scale for the Assessment of Positive Symptoms in Schizophrenia, SAPS. The course of the patients' predominant delusional themes was analysed afterwards.

RESULTS: The 12 delusional themes from SAPS were converted into five groups using principal component analysis. At all three follow ups there were almost equal proportions of patients with the predominant delusion being from the same group of delusional themes as at baseline and of patients with the predominant delusion being from another group of delusional themes than at baseline.

CONCLUSIONS: The predominant delusional theme at the time of first treatment contact was seen to be only of some predictive value to the predominant delusional theme at the later follow-up points in patients with first-episode non-affective psychosis. This stresses the importance of a systematic assessment of different delusional themes in the continued monitoring of early signs of psychotic relapse and at each psychotic episode.