Development of the Korean version of Alzheimer's Disease Assessment Scale (ADAS-K).

OBJECTIVE The purpose of this study was the development of the Korean Version of Alzheimer's Disease Assessment Scale (ADAS-K). METHOD ADAS-K was administrated to 84 AD patients as well as 105 non-demented control subjects. Three aspects of reliability were tested. To evaluate the validity of ADAS-K, discriminant validity and concurrent validity were tested. To evaluate the sensitivity of ADAS-K to disease severity, all subjects, AD patients and control subjects, were grouped by CDR scale and their mean scores on ADAS-K were compared. RESULT ADAS-K demonstrated high levels of reliability. Mean ADAS-K scores for AD patients were significantly different from the control group (p < 0.01). Furthermore, ADAS-K exhibited significant correlations with other tests and scales (range 0.45-0.85, p < 0.01). In ROC curve analysis, ADAS-K displayed high diagnostic efficacy and the optimal cut-off point was selected between 18/19. ADAS-K was able to discriminate the degree of AD severity according to CDR classification. Our results suggested that ADAS-K-cog was sensitive to very mild AD. CONCLUSION We demonstrated that ADAS-K is a reliable and valid instrument not only for AD diagnosis but also for evaluation of its severity.