• schizophrenia;
  • bipolar disorder;
  • cognition;
  • remission;
  • elderly


Evidence in younger populations suggests quantitative but not categorical differences in cognitive impairments between schizophrenia and bipolar disorder. It is uncertain whether a similar distinction applies to patients in later life.


We compared the cognitive abilities of older, community-living schizophrenia patients, controlling for their state of symptomatic remission, with those of older euthymic patients with bipolar I disorder. The study included 67 patients with schizophrenia (20 in symptomatic remission, 47 not in symptomatic remission; mean age 68 years) and 74 euthymic bipolar I patients (mean age 70 years), who were compared using analysis of covariance on clinical and neuropsychological variables (e.g., attention/working memory, verbal memory, executive function and verbal fluency) and contrasted with 69 healthy controls.


Remitted (SR) and non-remitted (SN) schizophrenia patients and bipolar I (BP) patients were impaired relative to healthy controls, with mostly large effect sizes for verbal memory (Cohen's d: SR 1.34, SN 1.48, BP 1.09), executive function (Cohen's d: SR 0.87, SN 1.29, BP 0.71) and verbal fluency (Cohen's d: SR 1.09, SN 1.25, BP 0.88), but smaller effect sizes for the domain of attention/working memory (Cohen's d: SR 0.26, SN 0.18, BP 0.52). Differences in cognitive performance between the remitted schizophrenia patients and the bipolar I patients were not significant.


In both older patients with schizophrenia and with bipolar disorder, serious and pervasive cognitive deficits can be demonstrated. Trait-related cognitive deficits in schizophrenia and bipolar disorder may share major phenotypic similarity in later life. Copyright © 2012 John Wiley & Sons, Ltd.