# Difference between revisions of "Markov - Cohort simulation"

ArchivedUser (talk | contribs) |
|||

Line 1: | Line 1: | ||

{{not_approved_content}} |
{{not_approved_content}} |
||

− | Markov Models are often used in economic evaluations. It includes clinical and epidemiological applications, costs and outcomes. Generally these models are used to represent stochastic processes, |
+ | Markov Models are often used in economic evaluations. It includes clinical and epidemiological applications, costs and outcomes. Generally these models are used to represent stochastic processes, for medical purposes you can model the course of chronic disease. The model is subclassified in the different disease states, connected by transition probabilities. The first task is to define the disease states, these states are economically and clinically important events in the disease process. |

[[Image:Marcov_Model.jpg]] |
[[Image:Marcov_Model.jpg]] |
||

Line 7: | Line 7: | ||

− | In |
+ | In this representation of a Markov Process there is a hypothetical cohort of patients. Each patient is distributed to one of the possible states of the Markov Model. The transition from one state to an other is based/dependent on the transition probabilities. |

For example my hypothetical cohort has a value of 10 000 patients and my transition probabilities from state "Asymatic Disease" to the state "Progressiv Disease" is 0,7 and from "Progressiv Disease" to "Death" 0,1. |
For example my hypothetical cohort has a value of 10 000 patients and my transition probabilities from state "Asymatic Disease" to the state "Progressiv Disease" is 0,7 and from "Progressiv Disease" to "Death" 0,1. |
||

## Latest revision as of 18:07, 19 June 2008

**This page contains development contents which may not yet represent final PROSIT Content.**

Markov Models are often used in economic evaluations. It includes clinical and epidemiological applications, costs and outcomes. Generally these models are used to represent stochastic processes, for medical purposes you can model the course of chronic disease. The model is subclassified in the different disease states, connected by transition probabilities. The first task is to define the disease states, these states are economically and clinically important events in the disease process.

**Cohort simulation**

In this representation of a Markov Process there is a hypothetical cohort of patients. Each patient is distributed to one of the possible states of the Markov Model. The transition from one state to an other is based/dependent on the transition probabilities.
For example my hypothetical cohort has a value of 10 000 patients and my transition probabilities from state "Asymatic Disease" to the state "Progressiv Disease" is 0,7 and from "Progressiv Disease" to "Death" 0,1.

__At cycle x we have :__

0 :10 000 patients in the state "Asymatic Disease" 0 patients in the state "Progressive Disease" and 0 patients in state "Death".

1 :3 000 patients in the state "Asymatic Disease" 7000 patients in the state "Progressive Disease" and 0 patients in state "Death".

2 : 900 patients in the state "Asymatic Disease" 8400 patients in the state "Progressive Disease" and 700 patients in state "Death".

and so on ...