Reliability and Validity of the Korean Version of Revised form of Hasegawa Dementia Scale (K-HDS)

ABSTRACT

BACKGROUND: The revised version of the Hasegawa Dementia Scale (HDS-R) is a useful dementia screening tool with a test for frontal lobe function and is relatively less influenced by education level and linguistic ability. We developed a Korean version of HDS-R (K-HDS) by translating the HDS-R to screen dementia patients in the Korean elderly. METHODS: The basic structure of the HDS-R was preserved but some questions were modified for lingual and cultural difference. It was administrated along with the Korean version of the MMSE, Korean Dementia Screening Questionnaire, Short form Samsung Dementia Questionnaire and Clinical Dementia Rating (CDR) scales, to 151 patients (55 Alzheimer's disease, 73 vascular dementia, 23 others) with mild to moderate dementia and to 225 elderly control subjects. To screen dementia, the optimal cut-off score was estimated by receiver operating characteristic (ROC) curve analysis. By comparing the Area Under the Curve, the diagnostic efficiency of K-HDS was compared with that of K-MMSE. RESULTS: The K-HDS had good internal consistency (Crohnbach's alpha coefficient=0.66), inter-rater reliability (r=0.95), and test-retest reliability (r=0.92). K-HDS was well correlated with the K-MMSE (r=0.84) and CDR (r=-0.67), which confirms the validity of this test. The optimal cut-off score was different according to educational level. In patients with an educational level less than 10 years, the cut-off score was 20 with the sensitivity of 87.0% and the specificity of 83%. With an educational level of 10 years or more, the cut-off score was 22 with the sensitivity of 93.0% and the specificity of 89.6%. The overall diagnostic efficiency of K-HDS was superior to that of K-MMSE especially in patients with an educational level of less than 10 years. CONCLUSIONS: The K-HDS is a reliable, valid and useful tool to screen dementia in the Korean elderly.