Objective: To compare the clinical and neurocognitive profile of early-onset (EOP, <40 years), late-onset (LOP, 40e59 years) and very-late-onset (VLOP, >= 60 years) psychosis. Design: Cross-sectional observational study. Setting: Secondary, tertiary, and community mental health care. Participants: Patients with a DSM-IV diagnosis of non-affective psychotic disorder were included from two complementary studies (GROUP and PSITE) on genetic and environmental risk factors of psychosis in the Netherlands and Belgium. Measurements: Main outcome measures were the severity of positive and negative symptoms, quality of life, and age-corrected scores on measures of general intelligence, verbal memory, attention, and executive function. One-year follow-up data were used to validate diagnoses and exclude participants with possible or probable dementia. Results: 286 EOP (85%), 24 LOP (7%) and 28 VLOP (8%) participated. VLOP patients reported significantly more positive symptoms than EOP patients. Age-at-onset groups had similar age-corrected scores on IQ, verbal memory, attention and executive functions. A significantly better performance was found in VLOP compared with LOP on the CAMCOG total score, though scores were still within the normal range. After controlling for possible confounding, however, VLOP differed significantly on an attention accuracy task compared with LOP patients. Re-entering data for probable dementia patients (N = 4) did change the results regarding cognition outcomes. Conclusions: VLOP patients show more positive symptoms but do not appear to differ on neuropsychological tests from EOP and LOP when age is controlled for. This questions the idea that VLOP is the expression of underlying neurodegeneration.
- Late-onset schizophrenia-spectrum disorders
- late-onset psychosis
- clinical phenotype