A comparison of six clock-drawing test scoring methods in a nursing home

Background Clock-drawing test (CDT) is widely used but lack of a suitable scoring method. Aims To compare the validity of six common CDT scoring methods and to find out the best one. Methods The drawing CDT was administered in a Chinese nursing-home inhabitants living on the mainland including 110 dementia, 118 MCI (mild cognitive impairment), and 133 random normal. We calculated the sensitivity and specificity of six scoring methods and applied the receiver-o perating characteristic (ROC) curve statistic, including determining the area under the curve (AUC). Results (1) All six CDT scoring methods had a value of sensitivity higher than 80% and a specificity of 60% except Jouk and Tuokko. Freund got the highest sensitivity (92.73%) of that five for the testing of dementia and high sensitivity (82.20%) for MCI with an acceptable specificity (70.68%). (2) The AUC (area under the ROC curve) of all six CDT methods was over 0.8 for dementia, and for MCI, only Jouk and Tuokko were lower than 0.8. Mendez had the largest AUC of 0.872 for MCI, which closely followed by Freund with 0.859. (3) Freund predicted dementia best but had no significant difference (p > 0.05); it only had significant difference with Jouk and Tuokko (p < 0.001) and the method in MoCA (p < 0.05) for both MCI and cognitive impairment. Conclusions Our study suggests that Freund scoring method could be the best one among the six evaluated scoring methods within our setting.