Validation of the Addenbrooke's Cognitive Examination-III for detecting vascular dementia in Iranian patients with stroke: A secondary data analysis.

Abstract

The current study was conducted with the aim of evaluating the third version of Addenbrooke's Cognitive Examination (ACE-III), and exploring its diagnostic power for a sample of stroke patients in the Iranian population. This was a cross-sectional analytical study, in which 206 patients with stroke were compared with 200 normal individuals as the control group. The patients were diagnosed based on the findings of neuroimaging and clinical examination by a neurologist. ACE-III, Montreal Cognitive Assessment (MoCA), Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), and The Structured Clinical Interview for DSM-5 (SCID-5) were used to gather the data and assess the vascular dementia in the patients. Furthermore, Cronbach's alpha, Pearson correlation coefficient, discriminant function analysis, and the receiver operating characteristic (ROC) curve were used to respectively measure internal consistency, convergent validity, discriminant validity, sensitivity, specificity, and the cutoff point of ACE-III. Internal consistency of ACE-III was excellent (α = 0.92 - 0.95), and convergent validity was measured through calculating the correlation between the scores of ACE-III and MoCA, which was very high (r = 0.957, P < 0.0001). Moreover, overall classification accuracy of ACE-III revealed that it is able to differentiate 87% of patients with vascular dementia from other patients. The area under the ROC curve was found to be 0.84, and cutoff point was 45/46, at which sensitivity and specificity were obtained as 0.72 and 0.90, respectively. ACE-III is a rapid, inexpensive, and efficient tool for evaluating cognitive deficits in specialized neurology clinics to provide a clinical and differential diagnosis of vascular dementia after stroke.