I am writing because I have a problem when I try to perform a budget impact analysis. When the transition probabilities depend on time ("non-homogeneous Markov models"), I cannot compute them, taking into account the cycle in which the patient enters the model. By default, the transition probabilities count from cycle one, when the patient may have entered a subsequent cycle.
I am writing because I have a problem when I try to perform a budget impact analysis. When the transition probabilities depend on time ("non-homogeneous Markov models"), I cannot compute them, taking into account the cycle in which the patient enters the model. By default, the transition probabilities count from cycle one, when the patient may have entered a subsequent cycle.
Thanks for your help,