Term: Markov Model
Last Updated: 2015-12-03
The Markov model is a modelling approach that predicts how likely an event is to enter a specific state in the future based on the current state of this event and chance (Last, 2001). This approach can use constant likelihood of each event state (stationary model) or changing likelihood of each event state (non-stationary model) (Schaubel et al., 1998). A non-stationary model can account for the influence of factors that might impact the likelihood of future events.
The Markov model can evaluate the outcomes of a process, such as whether a patient diagnosed with kidney failure who enters the Manitoba Renal Program (MRP) will receive hemodialysis, peritoneal dialysis, home hemodialysis, kidney transplant, or will die. For more information on how this modelling approach was used in Chatier et al. (2015), please read Appendix 2: Markov Model Transitional Probability Matrix in this report.
- Last JM. A Dictionary of Epidemiology. 4th Edition. In: Spasoff, RA, et. al. (eds). New York, New York: Oxford University Press; 2001. 0-0.(View)
- Schaubel DE, Morrison HI, Desmeules M, Parsons D, Fenton SS. End-stage renal disease projections for Canada to 2005 using Poisson and Markov models. Int J Epidemiol 1998;27(2):274-281. [Abstract] (View)
- Chartier M, Dart A, Tangri N, Komenda P, Walld R, Bogdanovic B, Burchill C, Koseva I, McGowan K-L, Rajotte L. Care of Manitobans Living with Chronic Kidney Disease. Winnipeg, MB: Manitoba Centre for Health Policy, 2015. [Report] [Summary] [Updates and Errata] [Additional Materials] (View)
Manitoba Centre for Health Policy
Community Health Sciences, Max Rady College of Medicine,
Rady Faculty of Health Sciences,
Room 408-727 McDermot Ave.
University of Manitoba
Winnipeg, MB R3E 3P5 Canada