Simplified Questionnaire for Early Detection of Dementia

Abstract

Background: Patients with dementia may have already progressed to an advanced stage of illness before they receive a proper diagnosis in a medical institution. Early detection of dementia is critical, and simplified screening methods are needed to support this goal.
Methods: An easy-to-complete questionnaire, consisting of 15 items that assess behaviors associated with dementia, was used for this study. Respondents were asked to fill out both the questionnaire and the Hasegawa Dementia Rating Scale-Revised (HDS-R). Multivariate regression analyses were used to identify which of the 15 items were significantly related to the HDS-R.
Results: The response rate was 74% (217/293). Using multivariate logistic regression, 4 items from the questionnaire were identified as predictors of an HDS-R score ≤20 (“Needs to be diagnosed”). These items include: “Cannot perform multiple tasks at the same time”, “Unable to count money”, “Unable to choose one’s seasonal clothes”, and “Repeatedly buying the same thing” (all ps < 0.05). Age and gender were also significant predictors in the model. A revised questionnaire based on these results combines the 6 significant predictors; sensitivity and specificity were 93.9% and 82.1%, respectively.
Conclusions: This work developed a simplified screening method for dementia that can be implemented in primary care settings.