Objective: The course over time of religious delusions (RDs) in late-life schizophrenia and psychotic depression may be relevant to know how long certain aspects of RDs may affect treatment. The present study examines (1) the 1-year follow-up of RDs and other prevalent delusions, (2) the association between RDs and the clinical course of psychotic depression and schizophrenia compared to those without RDs, and (3) associations of RDs and other prevalent delusions with "indicators of complexity"(e.g., suicidality, refusing medication). Design: Prospective study (half year and 1-year follow-up combined). Setting: Outpatients and inpatients in Geriatric Psychiatry Institution of Yulius, South-Holland, the Netherlands. Participants: One hundred and thirty seven older adult patients, mean age 76.3 (s.d. 8.1). Intervention: Natural follow-up study. Measurements: Diagnostic interview measures included Schedules for Clinical Assessment in Neuropsychiatry (SCAN 2.1), positive psychosis items of the Community Assessment of Psychic Experiences-42 (CAPE), and the 20-item measures from the Centre for Epidemiologic Studies Depression Scale (CES-D). Results: Although RDs in older adults decline in the clinical course of psychotic depression, the course is unfavorable compared to psychotic depression without RDs with regard to depressive symptom severity as measured by CES-D. No significant differences were noted in relation to clinical course of positive psychotic symptoms for both psychotic depression and schizophrenia. In schizophrenia, RDs persist more frequently compared to the most prevalent delusions. No significant difference was observed between patients with RDs compared to patients without RDs regarding indicators of clinical complexity. Conclusions: RDs predicting a less favorable course over time in psychotic depression. In schizophrenia, RDs appears to be relatively pervasive.