Estimating Alzheimer's disease progression and life expectancies with Markov
Transition Probabilities
 
   Lead Investigator:    Rachel Kukielka
   Institution      :    University of Minnesota- Twin Cities
   E-Mail           :    kukie011@umn.edu
   Proposal ID      :    1336



   Proposal Description:
   1) Outline method for transforming longitudinal Alzheimer's data into transition probabilities, providing estimates of life expectancies based on this method. Impact on research field(s) includes an analysis of previous and novel methods for transition probability estimating, and propose novel estimates of life expectancies. The methods I'm planning on using to transform the data into Markov transition probabilities are statistical analysis methods that I haven't confirmed at the moment, but will probability utilize R. The hypothesis I'll test will involve life expectancies of Alzheimer's patients across different treatment methods, and different health states. The outcomes will most likely be a comparison of different treatments and the impact on life expectancy based on Markov Chain analysis. The types of data used will most likely be probabilities/rates of disease progression across a patient population sample, which could be extracted from the amount of patients falling into different Alzheimer's health states divided by the total sample size if direct rates aren't given. At the moment, my methods and hypotheses aren't solidified, but this a loose outline of what I plan/hope to accomplish.     2) Create a new method for estimating transitional probabilities