An evaluation of longitudinal neurocognitive performance among middle-aged and older schizophrenia patients: use of mixed-model analyses.

The purpose of this study was to evaluate the trajectory of neuropsychological performance among middle-aged and older patients with schizophrenia using hierarchical linear modeling (HLM). Using the Mattis Dementia Rating Scale (DRS), we assessed cognitive performance of 143 schizophrenia outpatients (Mean age=53.1 years, SD=8.8) and 66 healthy comparison subjects (Mean age=61.8 years, SD=8.8). All participants had DRS data from two or more testing sessions with average intervals between sessions being 15 months (range=6-55). Using HLM, we found cognitive performance to be stable over time in both schizophrenia patients and healthy comparison subjects. Age, gender, ethnicity, and education did not affect trajectories. We also analyzed the data using more traditional statistical approaches to longitudinal data and found similar results. In sum, we found cognitive impairment to be stable over time among outpatients with schizophrenia, regardless of the statistical methodology used to analyze longitudinal data.