Abstract

Naming difficulty is one of the common cognitive impairments demonstrated by vascular dementia (VaD) as well as Alzheimer's disease (AD). The current study was conducted to develop a short form of the 60-item Korean-Boston Naming Test (K.-BNT; Kim & Na, 1997) empirically for using in dementia patients. First, the K-BNT was administered to 130 dementia patients (75 AD and 55 VaD) and 130 healthy normal elderly whose age, sex, and education level were matched with the dementia patients. Based on the difference between groups in the proportion providing a correct response to each item and the results of discriminant functional analysis, 15 good predictor items to differentiate dementia patients from normal adults were found. The internal consistency (Cronbach's &#x03B1;) for these 15-items was .86. The correlation between the 60-item K-BNT and the 15-item short form (S-K-BNT) was as high as .94 (p<.001). To compare the probabilities of correctly identifying dementia, the areas under Receiver Operating Characteristic (ROC) curves of the S-K-BNT and the K-BNT were compared. It was found that the S-K-BNT is the more efficient measure for identifying dementia than the K-BNT. Second, age and education-stratified norms on the S-K-BNT are presented for 565 healthy normal elderly aged 55-84 years. Third, the S-K-BNT was cross-validated in an independent sample that consisted of 68 dementia patients (39 AD and 29 VaD) and 68 healthy elderly. Correct classification rates were compared between the S-K-BNT and the K-BNT. They were 75.7% and 70.6%, respectively. It confirmed that the 15-item S-K-BNT is as efficient as the 60-item K-BNT in identifying dementia. Finally, several precautionary notes relating to the use of S-K-BNT were discussed.