A Short form of the Samsung Dementia Questionnaire ( S-SDQ ): development and cross-validation

ABSTRACT

BACKGROUND: The Samsung Dementia Questionnaire (SDQ) has recently been developed for screening dementia. Objectives of this study are to develop a short form of SDQ (S-SDQ) and to cross-validate S-SDQ in a new sample. METHODS: Factorial analysis of SDQ items was used to develop the S-SDQ. The S-SDQ was administered to informants of 100 dementia patients (61 Alzheimer's disease, 39 vascular dementia) and to those of 100 hospital control subjects. Two groups were matched in terms of age and sex. The retest of S-SDQ was conducted in 75 informants to determine the reliability of tests and retests which were spanned average 23.9 days. RESULTS: The S-SDQ was not influenced by education, age or sex. S-SDQ scores negatively correlated with the K-MMSE (r=-0.84). The S-SDQ showed stepwise scaling with dementia severity. The area under the ROC curve for the S-SDQ was 0.974 (S.E.=0.084). With respect to a diagnosis of dementia, the SDQ (cut-off point 8) had a sensitivity of 94% and a specificity of 90%. The S-SDQ was found to have a high test-retest reliability (r=0.97). CONCLUSIONS: An abbreviated version of SDQ (S-SDQ) had high sensitivity, specificity and test-retest reliability which were comparable to those of original SDQ. Thus, the S-SDQ can be a useful, brief dementia screening questionnaire.