Semi-Markov Models for Interval-Censored Transient Cognitive States with Competing Risk

• Activity Number: 256
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Shaoceng Wei*+ and Richard J. Kryscio
• Companies: University of Kentucky and Center on Aging and Departments of Statistics and Biostatistics, University of Kentucky
• Keywords: semi-Markov ; Weibull ; interval censoring ; panel data ; transitions ; competing risk

Abstract

Continuous-time multi-state stochastic processes are useful for modeling the flow of subjects from intact cognition to dementia with mild cognitive impairment and global impairment as intervening transient, cognitive states and death as a competing risk. Each subject cognition is assessed only periodically resulting in interval censoring for the cognitive states; death without a dementia is not interval censored. Markov chains are often applied to this type of panel data. Here we assume a Semi-Markov process in which we assume (i) the time from intact cognition to first cognitive impairment depends on age and other risk factors and (ii) the time is Weibull distributed. All other waiting times are assumed to be exponentially distributed except death. We apply our model to a real dataset, the Nun Study, a cohort of 461 participants.