Impairments in executive functions four years after a stroke

Post-stroke executive functions (EF) are understudied, particularly in the chronic phase. This study assesses impairment rates in Working Memory, Inhibition, and Shifting, examines the correlations between domain-specific EF measured with performance tests and self-report, and explores whether vascular risk factors and initial stroke characteristics predict specific EF 4 years post-stroke. Sixty-five stroke survivors (48 men, mean age = 64 years), with mild (55%) or moderate (42%) strokes at onset performed subtests from the Delis-Kaplan Executive Function System, Wechsler Adult Intelligence Scale-IV, Conners Continuous Performance Test 3rd ed., and the self-reported Behavior Rating Inventory of Executive Functions. Subtests were combined into EF domain scores and global composite scores, calculated separately for self-reported and performance-based tests. Correlational analyses and multivariate models were employed to probe associations. Aside from the Digit Span Backwards, all test scores were below average. Impairment rates varied from 11 to 49% on individual tests but were reduced in domain and global composites. No significant correlations were found between self-reported and performance-based measures within an EF domain. Predictor analyses showed that atrial fibrillation was associated with poorer Shifting, diabetes with poorer Inhibition, and shorter education with poorer Working Memory, all with moderate effect sizes but only for performance EF. EF impairments were common, varying by domain and method of measurement. Performance-based EF domains, unlike the self-reported domains, were linked to vascular risk factors and education level. Further research is needed to identify optimal EF tests for stroke populations. In clinical practice, EF assessment remains paramount, even after mild strokes.