Objective To determine the validity of the Clock Drawing Test (CDT) and the Mini-Mental State Examination (MMSE) respectively or in combination for differentiating Vascular Cognitive Impairment No Dementia (V-CIND) from normal subjects. Methods Eighty V-CIND patients and 80 healthy control subjects were blindly evaluated with MMSE, CDT, and additional neuropsychological tests. CDT was scored according to the Rouleau method and AD Cooperative Study method. Sensitivities and specificities of the two CDT measures and MMSE for identifying V-CIND patients were determined. The Areas Under the Receiver Operating Characteristic Curve (AUCs) were compared, and the sensitivity of the combination of CDT with MMSE calculated. Results V-CIND group performed worse than controls on both MMSE (p < 0.0001) and the two CDTs (p < 0.0001). In differentiating V-CIND patients from normal subjects, the two CDT measures provided sensitivities of 68.7% and 65.0%, and specificities of 78.7% and 86.2% respectively at optimal cutoff scores, which did no better than MMSE (sensitivity 80%, specificity 70%) (comparison of the AUCs, p = 0.992 and 0.428). The sensitivity of MMSE was marginally higher than that of CDT scored with AD Cooperative Study method (p = 0.053). By combining the two CDT measures with MMSE, the sensitivity was improved to 93.7% and 92.5% respectively. Conclusions Compared with MMSE, CDT is of only similar or even weaker ability for identifying V-CIND. MMSE at a cutoff of 28 may be of some value in detecting V-CIND patients. CDT and MMSE in combination provide a valid instrument for V-CIND screening. Copyright (C) 2007 John Wiley & Sons, Ltd.