Development of scoring criteria for the clock drawing task in Alzheimer's disease.

OBJECTIVE: To investigate the reliability and validity of free-hand clock drawings, a frequently used measure of constructional apraxia, in patients with Alzheimer's disease. DESIGN: Survey for the purpose of testing reliability and validity of a new scale. SETTING: Memory Disorder Clinic at a university-affiliated hospital in the Upper Midwest. PATIENTS: Forty-six patients were diagnosed with clinically probable dementia of the Alzheimer type after a dementia evaluation, and 26 normal elderly controls were research volunteers without a history of cognitive dysfunction. MEASUREMENTS: Neuropsychological tests, dementia-related scales, and clock drawings rated by a new 20-item Clock Drawing Interpretation Scale. Reliability measures, correlations, and clustering of items in the CDIS. RESULTS: The CDIS had inter-rater reliability (r = .94), internal consistence (rtt = .95), and reproducibility over a 6-month interval. CDIS scores were significantly correlated with two dementia-related scales and all neuropsychological tests and had the highest correlations with other measures of constructional apraxia. All but four Alzheimer patients (91%) and none of the controls had CDIS scores of 18 or less. CONCLUSION: Clinicians may reliably screen patients with Alzheimer's disease with the clock-drawing task, a measure sensitive to deficits in constructional apraxia.