[The Clock Test: drawing a clock for detection of cognitive disorders in geriatric patients].

Patients of a geriatric hospital (n = 263; 145 women, 118 men) had the task of drawing a clock and indicating a given time by the placements of hands. Errors were classified hierarchically by using a five-category-panel based on defined criteria. Its use as a "first-line" screening test for cognitive disorders in old age was evaluated. Retest- (r = 0.89) and interrater- correlations (r = 0.81) were satisfying. The results of this "clock drawing test" were compared both to the outcome in the Abbreviated Mental Test (AMT) and to the general judgment of the patients' cognitive status, based on the history, the report of informants, on clinical observation and examination. Patients with faultless clock drawings performed well in the AMT. However, prediction of the outcome in the AMT gradually became impossible with poorer performance in the clock drawing test, resulting in a low specificity (0.74) for a normal AMT. Specificity of the clock drawing test increased (0.79) for the general judgment of cognitive disorder, with sensitivity remaining constant (0.84). Sudden onset of cognitive disorder (most commonly confusion) in case of acute disease was discovered by the clock drawing test. Focusing on all patients without apparent cognitive disorder and with normal AMT-result, the subgroup of patients with faultless clock drawings was significantly younger (t = 5.0); p < 0.001). It is suggested to use the clock drawing test in addition to conventional psychometric screening tests because it requires visuospatial skills and conceptual thinking in addition to mere mnestic and verbal qualities.(ABSTRACT TRUNCATED AT 250 WORDS)