The age of onset and cognitive impairment at the early stage of schizophrenia

In schizophrenia, the age of first episode onset can reflect genetic loading and predict prognosis. Little is known about the association between the age of onset and cognition among individuals with early-stage schizophrenia. We aimed to compare the pre-treatment neurocognition profile between individuals with early-onset schizophrenia (EOS, the age of onset < 18 years), typical-onset schizophrenia (TOS, the age of onset between 18 and 39 years), and late-onset schizophrenia (LOS, the age of onset between 40 and 59 years). We included individuals with a current diagnosis of schizophrenia within 3 years and medication naive or less than 2 weeks of cumulative antipsychotic exposure and current daily antipsychotic dosage equivalent to ≤ 15 mg of olanzapine. Assessments included the MATRICS Consensus Cognitive Battery (MCCB) and the Positive and Negative Syndrome Scale (PANSS). We used linear regression to compare the difference between age-of-onset groups. We included 356 participants (67 EOS, 195 TOS, and 94 LOS). Compared with LOS, TOS was associated with lower scores in the verbal learning scores of the MCCB after adjusting for education years and the subscale scores of the PANSS (45.5 ± 12.9 vs. 40.5 ± 14.1, adjusted B = − 5.79, p = 0.001). The three groups had no difference in other cognitive domain scores. The association between the age of onset and MCCB verbal memory was U-shape (square of the age of onset, adjusted B = 0.02, p = 0.003). Patients with LOS had a better verbal learning function compared with individuals with TOS. These findings suggest that involvement of cognition assessment and rehabilitation training is necessary for patients with TOS.