Objective: To examine the clinical utility of the Clock Test for identifying dementia.
Design: Group comparisons.
Setting: A hospital-based out-patient diagnostic clinic.
Patients: Volunteer sample of elderly individuals (normal elderly, NE, n = 62) and a referred sample of probable Alzheimer Disease (AD, n = 58) patients meeting NINCDS-ADRDA criteria.
Main outcome measure: The Clock Test is composed of three components: Clock Drawing, Clock Setting, and Clock Reading. A detailed scoring system for qualitative as well as quantitative evaluation of Clock Drawing errors was used. Five time settings, varying in level of complexity, were used to evaluate Clock Setting and Clock Reading.
Results: The groups differed significantly on Clock Drawing, Clock Setting, and Clock Reading (P less than 0.001). On Clock Drawing, the AD group made significantly more errors of omission and misplacement of numbers than the NE group (P less than 0.001). Using cut-off scores derived to maximize separation between the groups to define deficits in performance, the sensitivity and specificity for the diagnosis of AD of Clock Drawing, Clock Setting, and Clock Reading were 92% and 86%, 87% and 97%, 92% and 85%, respectively. Using a criterion of deficits on two or more of the three components, sensitivity and specificity increased to 94% and 93%, respectively.
Conclusions: Deficits on clock drawing in AD may be reflective of a generalized disturbance in the conceptualization of time rather than constructional apraxia, per se. The functionally relevant components of Clock Setting and Clock Reading combined with Clock Drawing make the Clock Test particularly useful as a screening and research tool for AD.