Background: In persons free of dementia, accelerated cognitive decline might be an indicator of pathological impairment and a potential predictor of future dementia. However, the reliable detection of cognitive decline is a challenging task in some cases. Changes in psychometric test scores do not necessarily result from 'true' clinical change and may result from factors like measurement error and regression to the mean. Reliable Change Indices (RCI) facilitate the interpretation of changes in psychometric test scores. However, it is unknown whether RCI predict future dementia in persons who are presently free of dementia.
Aim: To examine for a widely used screening instrument for cognitive impairment and dementia (SIDAM) whether RCI methods contribute to the prediction of dementia diagnosis in older adults who are presently free of dementia.
Methods: A sample of 554 elderly individuals aged 75 and over who participated in the Leipzig Longitudinal Study of the Aged (LEILA 75+) and had no diagnosis of dementia at the first two visits. Participants have been tested with the test part of the SIDAM at maximal six visits with 1.5 year intervals. In all participants, RCI for change in SIDAM score (Time 1 to Time 2) have been computed. The main outcome measure was the diagnosis of dementia.
Results: ROC analyses showed that RCI were significant predictors of future dementia. Participants who deteriorated in SIDAM score -4 points or more had a three-fold increased risk of developing dementia (odds ratio 2.71, CI 1.6 to 4.6). However, RCI were not independent predictors of dementia in multivariate analysis.
Conclusion: RCI may support the early diagnosis of dementia.