Abstract
BACKGROUND: The present study aimed at evaluating the diagnostic properties of the Frontal Assessment Battery (FAB) in non-demented ALS patients by addressing the Edinburgh Cognitive Behavioural ALS Screen (ECAS) as the gold standard, as well as by examining the association between its administrability and scores with motor-functional measures. MATERIALS: N = 348 consecutive patients were administered the ECAS and FAB. Disease severity (ALSFRS-R), duration, progression rate (ΔFS), and stages (via King's and Milano-Torino systems) were considered. Administrability rates and prevalence of below-cut-off FAB scores were compared across clinical stages; regression models allowed to test whether, net of the ECAS-Total, motor features predicted the probability of the FAB not being administrable and of a defective FAB score. Intrinsic and post-test diagnostics were explored against a combined defective ECAS-Executive and ECAS-Fluency scores. RESULTS: 85.3% of patients managed to complete the FAB. FAB administrability rates decreased with advanced clinical stages, whereas the prevalence of below-cut-off FAB scores did not. The probability of the FAB not being administrable was predicted only by lower ALSFRS-R-bulbar and ALSFRS-R-upper-limb scores; no motor features, but the ECAS-Total, predicted a below-cut-off performance on the FAB. Raw and adjusted FAB scores showed high accuracy (AUC = .85 and .81, respectively) and good intrinsic and post-test properties. DISCUSSION: The FAB is featured by optimal diagnostics for detecting executive deficits in ALS, provided that it can be administered according to its original, standardized procedure, and thus that patients have sufficiently spared motor abilities to complete the test.