Our objective was to assess the validity of the Edinburgh Cognitive and Behaviour ALS Screen (ECAS), a multi-domain screen designed to detect cognitive deficits in patients with motor disorders. Forty ALS patients (without pre-diagnosed dementia) and 40, age-, gender- and education-matched healthy controls were recruited. All participants underwent extensive neuropsychological assessment and the ECAS. Performance at neuropsychological assessment across five domains (fluency, executive function, language, memory and visuospatial function) was compared to the ECAS ALS-Specific (fluency, executive functions and social cognition, language), ALS Non-specific (memory, visuospatial functions), and Total scores. Data from the healthy controls produced population-based abnormality cut-offs: composite score performance ≤ 2 SD in any domain classified impairment at neuropsychological assessment. Thirty-three percent of patients were impaired, most commonly in a single domain (executive or language dysfunction). Receiver Operator Curve (ROC) analyses using ECAS Total scores and ALS-Specific scores revealed 85% sensitivity and 85% specificity in the detection of cognitive impairment characteristic of ALS (fluency, executive function, language). A five-point borderline range produced optimal values (ALS-Specific Score 77-82, and ECAS-Total Score 105-110). In conclusion, validation against gold standard extensive neuropsychology demonstrated that the ECAS is a screening tool with high sensitivity and specificity to impairment characteristic of ALS.
Keywords: Cognition; executive functions; fluency; language; screen.