A longitudinal analysis of memory in patients with schizophrenia.

Memory deficits are widely reported in patients with schizophrenia, but uncertainties remain about the extent and the longitudinal course of these deficits. Twenty-eight patients with a DSM-IV diagnosis of schizophrenia were tested on multiple aspects of memory at baseline, 9- and 18-month follow-up. Measures included: digit span, the Rivermead Behavioural Memory test (RBMT) battery, the Graded Naming Test (GNT) and several computerized memory tests from the Cambridge Automated Neuropsychological Testing Battery (CANTAB). A group of healthy controls (N=17) was tested on the CANTAB battery at baseline and 9-month follow up. The patients performed significantly poorer than controls on all CANTAB measures; however, there was no difference in change between groups over a 9-month period. Within-group patient comparisons revealed that symptoms reduced significantly over the study period, but had no association with memory. Significant improvements were observed for patients on two verbal memory tasks: the GNT and digit span, but not on any other measure. Interestingly, these were the only two tests on which patients were within normal limits at baseline. This study shows that patients with schizophrenia have deficits in multiple aspects of memory which remain stable over long periods of time. In addition, patients showed a tendency to improve on memory tasks which contained a verbal component.