Screening Instruments for the Early Detection of Cognitive Impairment in Patients with Multiple Sclerosis.

Abstract

Cognitive impairments are common in individuals with MS and adversely affect functioning. Early detection of cognitive impairment, therefore, would enable earlier, and possibly more effective, treatment. We sought to compare self-reports with a short neuropsychological test as possible screening tools for cognitive impairment. One hundred patients with MS were tested with the Minimal Assessment of Cognitive Function in Multiple Sclerosis; z scores were used to derive the Cognitive Index (CI). Receiver operator characteristic curve analyses were performed, with criteria for impairment set at -1.5 and -2.0 SD below the mean. Scores from two self-reports (the Multiple Sclerosis Neuropsychological Screening Questionnaire-Patient Version and the Behavior Rating Inventory of Executive Function-Adult Version [BRIEF-A]) and a neuropsychological test (the Symbol Digit Modalities Test [SDMT]) were entered as test variables. Exploratory regression analyses were conducted with 1) CI and self-reports and 2) CI and the Problem-Solving Inventory (PSI). Classification accuracy was high or moderately high for SDMT when the criterion was -2.0 or -1.5 SD, respectively, but low for the self-reports. Hierarchical linear regression showed that the SDMT alone was the best predictor of cognitive impairment; adding the self-reports did not improve the model. Exploratory analyses indicated that certain self-reports (BRIEF-A, PSI) provided some explanatory power in separate models. The SDMT is a more accurate screening tool for cognitive impairment; however, self-reports provide additional information and may complement objective testing. Results suggest that screening for cognitive impairment may require a multidimensional approach.