Clinical utility of the K-T cancellation test in a memory clinic population

The K-T cancellation test (K-T) has been validated as a measure of executive functions (EF) but its clinical utility has not yet been examined. This study aimed to validate K-T in a memory clinic setting by examining its capacity to discriminate older adults with normal cognition (NC) from those with mild cognitive impairment (MCI) and Alzheimer's disease (AD).

K-T was administered to 120 NC subjects, 146 patients with MCI, and 93 patients with AD. A one-way analysis of covariance was used to compare the correct cancellations of K-T between the groups. Linear regressions were run to identify significant demographic predictors of K-T for NC subjects and to determine the equation to calculate z scores for all subjects. The area under the curve (AUC), sensitivity (Se), specificity (Sp), and positive (PPV) and negative (NPV) predictive values were assessed to compare the diagnostic performance between K-T and the Mini-Mental State Examination (MMSE) for discrimination between NC subjects and patients with cognitive impairment.

After adjusting for age, education, and gender, the groups were significantly different from each other regarding the number of correct cancellations of K-T, F(2, 353) = 116.6, p < .001, η(2)p = .40. Compared to the NC group (Z = 0, SD = 1), the mean z score was -1.52 for the MCI group and -2.53 for the AD group, suggesting impaired performance for the patient groups. K-T showed a better diagnostic performance for discrimination between the NC subjects and the patients with MCI (AUC = .83; 95% CI [.79, .88]; Se = .79; Sp = .74; PPV = .79; NPV = .74), compared to that of MMSE (AUC = .74, 95% CI [.68, .80]; Se = .68; Sp = .73; PPV = .79; NPV = .64).

The K-T cancellation test showed a good diagnostic performance in discriminating cognitively normal older adults from cognitively impaired patients. Our findings support the clinical utility of K-T in geriatric neuropsychological assessment for detection of early cognitive impairment.