|Research Centre||DCRC Assessment and Better Care|
On average cognition declines with age but there is wide variability (Wilson et al., 1999).
Cognitive decline is typically studied by comparing those who show major cognitive decline to people showing normal age associated decline (e.g., Wilson et al., 2002).
In mild cognitive impairment (MCI) neuropsychological test performance is used to operationalise the distinction between normal age-associated cognitive decline and abnormal cognitive decline.
By definition, those with MCI perform worse in one or more cognitive domains compared to those without impairment. However, this approach does not examine the difference, if any, amongst those considered to be ageing normally.
The first aim of this project is to model cognitive decline trajectories in the Sydney MAS sample and assess the associates of variability in these paths.
A secondary aim of the project is to apply item response theory (IRT) analysis to produce psychometrically matched domain measures. IRT has been used to scale test performance and create domain measures that are sensitive to performance across ability levels (e.g. Mungas et al., 2000; 2003). The suitability of this approach for generating outcomes for the cognitive decline analysis will be evaluated.