Evaluating instrumental activities of daily living (IADL) is an important part of procedure to diagnose dementia. The Korean-Instrumental Activities of Daily Living (K-IADL) has been used extensively in Korea. However, its cut-off score has not been reformulated since 2002. The purpose of this study was to yield a new optimal cut-off score for the K-IADL and confirm the validity of this new cut-off score with various dementia groups.

We retrospectively collected a total of 2,347 patients' K-IADL data from 6 general hospitals in Korea. These patients had mild cognitive impairment (MCI) or dementia with various etiologies for cognitive impairment. We also recruited a normal control group ( n =254) from the community. Korean-Mini Mental State Examination, Short version of the Geriatric Depression Scale, Clinical Dementia Rating, and Global Deterioration Scale were administered to all participants. Caregivers completed K-IADL and Barthel Index.

K-IADL scores were significantly different among dementia subgroups, but not significantly different among MCI subgroups. Based on internal consistency, correlations with other scales, and factor analysis, K-IADL showed excellent reliability and validity. The new optimal cut-off score to diagnose dementia was 0.40, which gave a sensitivity of 0.901 and a specificity of 0.916. Positive predictive value for dementia using the new cut-off score was 94.2% for Alzheimer's disease, 100% for vascular dementia, and 84% for Parkinson's disease.

Our results illustrate that the new K-IADL cut-off score of 0.40 is reliable and valid for screening impairments of daily functioning resulting from various etiologies.