Neuropsychological outcomes in randomized controlled trials of antiepileptic drugs: a systematic review of methodology and reporting standards.

Abstract

To systematically review the methodology and use of neuropsychological tests in randomized controlled trials (RCTs) of antiepileptic drugs (AEDs) in patients with epilepsy. Trial reports were found by searching Medline 1966-1996 and searching through journals by hand. Inclusion and exclusion criteria were applied, and methodological and neuropsychological test data was extracted by using a proforma. 43 reports met our inclusion criteria, representing 40 RCTs. as three RCTs had generated two reports. Twenty-two were actively controlled, and 18 were placebo-controlled studies. Reporting of basic methods such as randomization method was poor. There has been no uniform approach to the use of neuropsychological tests, and a total of 87 has been used. The Stroop Colour Word Test and the Finger Tapping Test were most commonly used, at 13 times each, but were not used or reported in a uniform manner. Poor reporting of methods and the use of a plethora of neuropsychological tests create great difficulties for anyone wishing to make sense of currently available data. If we are better to understand the neuropsychological effects of AEDs, a more rational approach is needed, for which recommendations are made.