ABSTRACT
Objectives: To adapt and translate the Edinburgh Cognitive and behavioural amyotrophic lateral sclerosis screen (ECAS); to generate preliminary normative data for three language groups in South Africa (SA); to assess the convergent validity of the ECAS in SA samples. Methods: The ECAS was linguistically and culturally adapted for Afrikaans-, isiXhosa-, and English-speaking SA adults (n = 108, 100, and 53, respectively). Each language group was stratified by age and educational level. Cutoff scores for cognitive impairment were set at the group mean minus two standard deviations (SDs). A pilot sample of ALS patients and controls (n = 21 each) were administered the ECAS and an extensive neuropsychological evaluation (NPE) and the Montreal Cognitive Assessment (MoCA) to assess convergent validity. Results: Across the three language groups, the total ECAS cutoff scores ranged from 68 to 97. The ECAS score correlated significantly positively with educational level (p < 0.001) and negatively with age (p < 0.005). The restricted letter fluency task demonstrated a floor effect, particularly in Afrikaans-speakers. The mean total ECAS score (±SD) was similar in ALS patients (103.52 ± 11.90) and controls (100.67 ± 20.49; p = 0.58). Three (14.3%) ALS patients scored below the cutoff for cognitive impairment. Correlations between individual ECAS subtests and analogous NPE tests ranged from weak to moderate. The MoCA score was significantly positively correlated with the ECAS total score (r = 0.59; p = < 0.001). Conclusions: The adapted ECAS and associated normative data will aid cognitive screening of African ALS patients. Larger participant numbers are needed to assess the validity of the adapted instrument.
