Figure drawing tasks are commonly used standalone or as part of broader screening tests to detect cognitive impairment. Only one study has compared the classification accuracy of three common drawing tasks—overlapping infinity loops, wire cube, and the clock drawing task (CDT)—in mild cognitive impairment (MCI) and dementia, but age and education, which impact performance, were not accounted for. We replicated the research, adjusting for age and education and, for the first time, assessed subjective cognitive decline (SCD) too. Participants were recruited from the Tasmanian ISLAND Cognitive Clinic and healthy controls from a community sample. All participants completed the three figure drawing tasks. The clinic patients were categorised according to interdisciplinary consensus diagnosis. Binomial logistic regression and area under ROC curves (AUC) were calculated to determine the discriminatory ability of each drawing task. Overall, 112 adults were recruited; 51 had normal cognition (NC), 21 SCD, 24 MCI, and 16 had dementia. The infinity loops test did not discriminate any of the groups, casting some doubt on its usefulness. The wire cube discriminated NC from dementia (AUC 0.7; p < 0.05). The CDT discriminated NC from dementia (AUC 0.77; p < 0.01), NC from cognitive impairment (dementia + MCI; AUC 0.59; p < 0.05), and MCI from dementia (AUC 0.76; p < 0.01). None of the tests discriminated NC from MCI or NC from SCD. The CDT was the most discriminatory test, followed by the wire cube. This may help guide clinicians who often choose just one figure drawing task due to time constraints or patient fatigue.
Keywords: Addenbrooke’s Cognitive Examination; clock drawing test; dementia; figure drawing task; infinity loops; mild cognitive impairment; subjective cognitive decline; wire cube.