Off-campus WSU users: To download campus access dissertations, please use the following link to log into our proxy server with your WSU access ID and password, then click the "Off-campus Download" button below.
Non-WSU users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Date of Award
John L. Woodard
Compounding research has demonstrated high reversion rates from mild cognitive impairment (MCI) to normal cognition (4%-55%), leaving the potential for false positive MCI diagnoses. Notably, common limitations exist in the MCI literature, such as using few longitudinal data points or only the first and last data point, and subjectively defining cognitive trajectories.
The current study statistically determined classes of cognitive trajectories for 6,795 Medicare beneficiaries, aged 65-105, from the National Health and Aging Trends Study (NHATS). MCI status, impairment in one or more of three cognitive domains, was determined at five annual time-points. Four multinomial logistic regressions examined either distal/chronic or proximal/situational latent factors, and their association with cognitive trajectory class, directly, and through the mediator, baseline cognition.
A latent class growth analysis (LCGA) identified a three-class, cognitive trajectory model: no-impairment (noMCI, 84%), fluctuating/stable MCI (fMCI, 9%), and progressing MCI/dementia (pMCI, 7%). The distal/chronic model was significant for a direct relationship between cardiovascular risk and disease (CVF) and trajectory class (fMCI/noMCI, pMCI/noMCI, pMCI/fMCI); early life factors (ELF) was not associated with LCGA class. The proximal/situational model was significant for a direct relationship between RHF and LCGA class (fMCI/noMCI, pMCI/fMCI); psychoemotional factors (PEF) was not associated with LCGA class. Baseline cognition was a significant mediator for all latent factor and LCGA class relationships.
This study strongly supports the importance of considering instability of the MCI diagnosis in addition to detection of MCI conversion. It highlights the importance of considering statistically-derived cognitive trajectories rather than subjectively-defined trajectories, examining cognitive trajectories in an SEM framework, examining, baseline cognition as a mediator, rather than a covariate.
Norman, Andria L., "Predicting Mci Cognitive Course: Examination Of Distal And Proximal Theoretical Models In The National Health And Aging Trends Study" (2018). Wayne State University Dissertations. 2122.