Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Stark_Masters_5.6_FINAL_updated.pdf (1.23 MB)

ETD Abstract Container

Abstract Header

Using Partial Least Squares Analyses to Explore the Relationship between Alzheimer’s Disease Biomarkers, Modifiable Health Variables, and Cognition in Older Adults with Mild Cognitive Impairment

Stark, Jessica Hana
http://rave.ohiolink.edu/etdc/view?acc_num=osu1620642072303

Abstract Details

2021, Master of Science, Ohio State University, Psychology.
Objective: This thesis aims to identify novel relationships between modifiable physical and health variables, Alzheimer’s disease (AD) biomarkers, and cognitive function in a cohort of older adults with mild cognitive impairment (MCI). Methods: Metrics of cardiometabolic risk (e.g., body mass index), stress (e.g., cortisol), inflammation (e.g., c-reactive protein), neurotrophic/growth factors (e.g., brain-derived neurotrophic factor), and AD (e.g., plasma tau) were assessed in 154 MCI participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) at baseline (mean age = 74.1; sd =7.5; mean education = 16.0; sd = 2.9). Of these 154 participants, 126 had 2-year follow-up data available for analyses (mean age = 74.0; sd = 7.6; mean education = 16.0; sd = 2.9). Participants also completed a comprehensive neuropsychological battery. Individual test scores and composite scores of memory and executive function published by ADNI were assessed. Partial least squares correlation (PLSC), an unbiased and flexible multivariate technique, was employed to examine cross-sectional associations among these physiological variables and cognition. Partial least squares regression (PLSR), a multivariate technique that defines optimal combinations of variables that best predict an outcome, was used to identify which, if any, of these physiological variables are important in predicting memory or executive function at 2-year follow-up. Results: The PLSC analysis revealed a latent variable describing a unique combination of AD biomarkers, neurotrophic/growth factors, education, and stress that were significantly associated with specific domains of cognitive function, including episodic memory, executive function, processing speed, and language, representing 45.2% of the covariance in the data. Age, BMI, and tests of basic attention and premorbid IQ were not significant. The PLSR analyses revealed that baseline metrics of cardiometabolic function, inflammation, and AD biomarkers were important in predicting memory and executive function performance at 2-year follow-up. Baseline education was important in predicting memory but not executive function performance at 2-year follow-up. Our two best models predicted 65.1% and 63.7% of the variance in memory and executive function respectively at 2-year follow-up. Conclusion: Our data-driven analysis highlights the significant cross-sectional relationships between metrics associated with AD-pathology, neuroprotection, and neuroplasticity primarily with tasks requiring higher order cognitive abilities (episodic memory, executive function, verbal fluency), rather than cognitive tasks that do not require mental manipulation (premorbid IQ and basic attention). Baseline metrics of cardiometabolic function, inflammation, and AD pathology were statistically important in predicting future memory and executive function performance at 2-year follow-up, suggesting that variables associated with neuroprotection and neuroplasticity (such as brain-derived neurotrophic factor and platelet-derived growth factor) may hold relatively less importance in predicting future cognition.
Scott Hayes, Ph.D. (Advisor)
Jasmeet Hayes, Ph.D. (Committee Member)
Ruchika Prakash, Ph.D. (Committee Member)
79 p.
Aging; Clinical Psychology
Neuropsychology; Aging; Mild Cognitive Impairment; Memory; Executive Function; Alzheimers Disease; Neuroplasticity; Neuroprotection; AD Biomarkers; Growth Factors; Healthy Aging

Recommended Citations

Citations

  • Stark, J. H. (2021). Using Partial Least Squares Analyses to Explore the Relationship between Alzheimer’s Disease Biomarkers, Modifiable Health Variables, and Cognition in Older Adults with Mild Cognitive Impairment [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1620642072303

    APA Style (7th edition)

  • Stark, Jessica. Using Partial Least Squares Analyses to Explore the Relationship between Alzheimer’s Disease Biomarkers, Modifiable Health Variables, and Cognition in Older Adults with Mild Cognitive Impairment. 2021. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1620642072303.

    MLA Style (8th edition)

  • Stark, Jessica. "Using Partial Least Squares Analyses to Explore the Relationship between Alzheimer’s Disease Biomarkers, Modifiable Health Variables, and Cognition in Older Adults with Mild Cognitive Impairment." Master's thesis, Ohio State University, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=osu1620642072303

    Chicago Manual of Style (17th edition)

osu1620642072303
64
© 2021, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.