Reliability and clinical usefulness of the short-forms of the Wechsler memory scale (revised) in patients with epilepsy.

ABSTRACT

PURPOSE: The Wechsler Memory Scale-R (WMS-R) is often used for the evaluation of patients with epilepsy, but is time consuming. Two shortened forms of the WMS-R have been published in an effort to reduce the time involved in the test. The purpose of this study is to cross-validate these abbreviated forms in patients with epilepsy. METHODS: We assessed 103 consecutive patients with drug-resistant partial epilepsy being treated in our epilepsy unit. The WMS-R was administered to all patients. The reliability of two shortened versions of the WMS-R were evaluated. The appraisal was performed taking into account the side of the epileptogenic region. The statistical analyses were carried out with the Pearson's correlation and with the intraclass correlation coefficient. RESULTS: Both shortened formulas showed good reliability coefficients for predicting the General Memory and Delayed Recall Indices. In the overall accuracy of the predictions by both short forms no significant differences were found among the three study. groups (right, left or bilateral). However, when considering the predictive error only the three-subtest formulas comprising logical memory, verbal paired associates and visual reproduction efficiently predicted performance, regardless of the side of the epileptogenic region. Of the predicted scores in the overall sample, 92 and 97% fell below the standard error of measurement for general and delayed memory index, respectively. Both short forms correctly classified 89-94% of the performances at the average or impaired level. CONCLUSIONS: The short form of the M-MS-R using three subtests (logical memory, verbal paired associates and visual reproduction) is reliable and time-efficient for estimating the General and Delayed Recall Memory indexes in patients with drug-resistant partial epilepsy.