OBJECTIVE: To determine whether there are differences in verbal working memory amongst subjects with schizophrenia, their first degree relatives and controls, and to evaluate the influence of symptoms on these differences, as an initial step to assess whether this cognitive function is an endophenotype. METHODS: We examined 197 cases with schizophrenia, 197 first degree relatives and 200 controls through psychiatric interviews and the Letters and Numbers Sequencing test (LNS). Performance was compared among the three groups adjusting for age, sex and education level. Adjustment for “negative symptoms” and “disorganization” was performed afterwards. RESULTS: Subjects with schizophrenia showed lower performance in the LNS than their first degree relatives and the healthy controls; the effect sizes were 0.75 and 1.18 respectively. There was a small difference between relatives and controls (effect size =0.38). These differences were significant after adjustment for negative and disorganized symptoms, but the effect sizes became smaller: 0.26 for relatives vs. subjects with schizophrenia, 0.56 for controls vs. subjects with schizophrenia and 0.33 for relatives vs. controls. Among individuals with schizophrenia, performance in the LNS was not associated with disorder duration, disease onset age, antipsychotics, history of depressive episodes or substance use disorders. CONCLUSION: Results suggest verbal working memory may be considered as an endophenotype in schizophrenia.
OBJECTIVE: To determine whether there are differences in verbal working memory amongst subjects with schizophrenia, their first degree relatives and controls, and to evaluate the influence of symptoms on these differences, as an initial step to assess whether this cognitive function is an endophenotype. METHODS: We examined 197 cases with schizophrenia, 197 first degree relatives and 200 controls through psychiatric interviews and the Letters and Numbers Sequencing test (LNS). Performance was compared among the three groups adjusting for age, sex and education level. Adjustment for “negative symptoms” and “disorganization” was performed afterwards. RESULTS: Subjects with schizophrenia showed lower performance in the LNS than their first degree relatives and the healthy controls; the effect sizes were 0.75 and 1.18 respectively. There was a small difference between relatives and controls (effect size =0.38). These differences were significant after adjustment for negative and disorganized symptoms, but the effect sizes became smaller: 0.26 for relatives vs. subjects with schizophrenia, 0.56 for controls vs. subjects with schizophrenia and 0.33 for relatives vs. controls. Among individuals with schizophrenia, performance in the LNS was not associated with disorder duration, disease onset age, antipsychotics, history of depressive episodes or substance use disorders. CONCLUSION: Results suggest verbal working memory may be considered as an endophenotype in schizophrenia.