The A-B neuropsychological assessment schedule (ABNAS): the relationship between patient-perceived drug related cognitive impairment and results of neuropsychological tests

Our intention was to evaluate the relationships between the A-B neuropsychological assessment schedule (ABNAS) as a measure of patient-perceived cognitive effects of antiepileptic drugs (AEDs) and the results of neuropsychological tests. The measure was developed specifically to assess patient-perceived cognitive effects of AED treatment. Evidence of its reliability and validity has been previously documented. In this study 96 patients were included using stratified inclusion-criteria to guarantee variability of performance: 55 patients were included from a 'low risk condition' with respect to possible cognitive effect (i.e. monotherapy carbamazepine within a dose range of 600-1200 mg/day) and 41 patients were included from a 'high risk condition' (i.e. polytherapy of three or two AEDs including either phenytoin, phenobarbitone or a benzodiazepine; treatment with topiramate with a titration speed using 50 mg or higher increments per week and within the first 6 months of treatment). All patients were prospectively assessed using the ABNAS and five neuropsychological tests (all part of the FePsy test system) with proven sensitivity of cognitive effects of antiepileptics: three tasks using reaction-time to measure speed ('simple (visual) reaction-time measurement', 'the binary choice reaction test' and 'the computerized visual searching task'); one test measuring motor speed ('the finger tapping task'); and a memory test ('recognition of words'). The three reaction-time tasks and the finger tapping test were significantly correlated with the ABNAS-score with correlations ranging from 0.22 to 0.35. The highest correlation was with 'simple (visual) reaction-time measurement' (0.35). Discriminant analysis showed that with the neuropsychological tests 61.5% of the patients were correctly identified as having high/low ABNAS-scores. The ABNAS underestimated impairment in 17.8% of the patients ( = low ABNAS-score but impairment on the neuropsychological tests). The present study contributes to the already existing evidence of validity of the ABNAS as a screening instrument for clinical practice as the relationship between the ABNAS-score and results of neuropsychological tests can help to identify who is at risk and needs further referral for neuropsychological assessment. Moreover the correlation between ABNAS-score and those neuropsychological tests that are sensitive for drug-effects may provide a sensitive instrument in early drug-development phases while keeping the burden on financial and time resources to a minimum.