Objective: To investigate the relationship between cognitive impairments an
d rated activity restrictions. Design: Comparison of neuropsychological tes
ts and activity questionnaires. Activity ratings were made by staff and str
oke patients. Setting: Two neurological rehabilitation hospitals. Participa
nts: Eighty-seven stroke patients. Results: Moderately high correlations we
re found between cognitive test scores and activity ratings made by staff m
embers. In contrast, correlations between ratings made by patients and cogn
itive rests were much lower. There was also little agreement between the st
affs' ratings and the patients' own ratings. Multiple regression analyses i
ndicate that cognitive impairments account for 28.9% of the variance in the
activin rating made by the staff members. The coefficient of determination
was slightly higher when age, time since onset of illness, and depression
scores were included as predictors. Conclusions: These findings provide par
tial support fur a hierarchical structure of the "International Classificat
ion of Impairment, Activities and Participation" model of the World Health
Organization.