Alvarez, Karina
(2012)
Application of Advanced Statistical Methods in an Aging Dataset.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
The focus of this thesis was to explore the application of advanced statistical methods in the Ginkgo Evaluation of Memory (GEM) Study. GEMS enrolled 3,069 participants age 75 or older with normal cognition or mild cognitive impairment. Those with dementia were excluded from participation. After extensive medical and neuropsychological screening, participants were randomly assigned to receive twice-daily doses of either 120 milligrams of ginkgo extract or an identical-appearing placebo. The 240 milligrams daily dose of ginkgo was selected based on current dosage recommendations and prior clinical studies indicating possible effectiveness at this dosage. The products used in the study were supplied by Schwabe Pharmaceuticals, a German company. We focused on two methods, a flexible Cox model (Gray’s model) and a trajectory procedure based on a mixture model that is implemented in the SAS procedure PROC TRAJ. The spline-based extension of the Cox model was applied to biomarker data; specifically: Cystatin-C, Beta Amyloid 40, Beta Amyloid 42, and a ratio of Beta Amyloid 42 over Beta Amyloid 40. We wanted to determine if the estimate of the log-hazard ratio changed over time for each of the biological measures. The trajectory analysis was used to determine if a patient’s illness trajectory continued on the same path towards demented or non-demented before experiencing a pneumonia event. The trajectory analysis was applied to the longitudinal trajectories of activities of daily living (ADL), independent activities of daily living (IADL) and modified mini-mental status exam (3MSE). The Cox Spline analysis resulted in no statistically significant information added to the models using the spline analysis. Trajectory analysis concluded that patients on a downward trajectory at baseline only escalated before the pneumonia event. As the average life expectancy continues in increase in humans, it is important to evaluate statistical methods in the elderly population to identify subpopulations that need more medical attention than the population at large. Thus, the public health significance of this thesis is that by identifying these subgroups that are distinctly different from the overall population, we can provide preventative care where needed more efficiently.
Share
Citation/Export: |
|
Social Networking: |
|
Details
Item Type: |
University of Pittsburgh ETD
|
Status: |
Unpublished |
Creators/Authors: |
|
ETD Committee: |
|
Date: |
29 June 2012 |
Date Type: |
Completion |
Defense Date: |
13 April 2012 |
Approval Date: |
29 June 2012 |
Submission Date: |
25 April 2012 |
Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
Number of Pages: |
49 |
Institution: |
University of Pittsburgh |
Schools and Programs: |
School of Public Health > Biostatistics |
Degree: |
MS - Master of Science |
Thesis Type: |
Master's Thesis |
Refereed: |
Yes |
Uncontrolled Keywords: |
GEMS, Ginkgo biloba, ADL, IADL, 3MSE, biomarkers, Alzheimer's, dementia, pneumonia, trajectory analysis, Gray's model, flexible cox model, spline analysis |
Date Deposited: |
29 Jun 2012 15:57 |
Last Modified: |
19 Dec 2016 14:38 |
URI: |
http://d-scholarship.pitt.edu/id/eprint/12014 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
|
View Item |