Determining empirically based self-reported cognitive change: Development of reliable change indices and standardized regression-based change norms for the multiple abilities self-report questionnaire in an epilepsy sample

Abstract

Purpose Reliable change indices (RCIs) and standardized regression-based (SRB) change score norms were calculated for a measure of self-reported cognitive function, the Multiple Abilities Self-Report Questionnaire (MASQ), in patients with complex partial seizures. Establishment of such standardized change scores could be useful in determining the magnitude and direction of self-appraised cognitive change after epilepsy surgery or other treatment interventions. The primary study objective was to calculate RCI and SRB values for the MASQ. A secondary objective was to report SRB change scores in patients who had undergone anterior temporal lobectomy (ATL) and to assess relationships between self-reported cognitive change, seizure outcome, objective memory test performance, and mood. Methods The MASQ was administered to 36 patients with complex partial seizures on two occasions (mean test–retest interval, 6 months). This group did not have major psychopathology and were on stable antiepileptic drugs. RCI and SRB change scores were calculated. Adjustments for baseline ratings, age, education, gender, age at seizure onset, and seizure duration were made with the SRB method. A confidence interval cutoff score (90% level) was calculated for the five MASQ cognitive domains (Language, Visual Perception, Verbal Memory, Visual–Spatial Memory, Attention/Concentration). MASQ SRB scores were computed for a second sample of 50 patients who had undergone ATL. Results Test–retest reliabilities for the MASQ domains ranged from a low of 0.63 (Attention/Concentration) to a high of 0.87 (total score). Baseline MASQ score was the single largest contributor to the regression equations. Left and right ATL groups demonstrated similar magnitudes of self-reported cognitive change across all five MASQ domains. Individual base rate change distributions were similar across four of the five domains. with a higher proportion of right ATL patients reporting worsening attention function. Both postoperative mood and SRB-based verbal memory outcome were significantly correlated to self-reported cognitive change in the patients who had undergone ATL. Conclusions SRB methodology provides a standardized technique with which to establish patient perception of cognitive change and may be of use when examining change across individual- and group-level ratings of cognitive functioning in clinical and research settings. These techniques also provide a common metric for direct comparison between subjective self-report ratings of cognitive function and objective cognitive test instruments.