Define different Initial Counts for two strategies when running a model

[kleanthisk10]

Hello, is it possible to run different initial counts for two strategies?

Let's say we have the example provided from ?run_model

mod1 <- define_strategy( transition = define_transition( .5, .5, .1, .9 ), define_state( cost = 543, ly = 1 ), define_state( cost = 432, ly = 1 ) )

and

mod2 <- define_strategy( transition = define_transition( .5, .5, .1, .9 ), define_state( cost = 789, ly = 1 ), define_state( cost = 456, ly = 1 ) )

How can we change

res2 <- run_model( mod1, mod2, cycles = 10, cost = cost, effect = ly)

and set mod1 c(80,20) and for mod2 c(60,40) initial counts?

Thank you

[KZARCA]

Hi,
We've got the argument init to specify the initial counts for each state, but nothing to change the counts for different strategies.
Can you please provide us with an example of the use cases for which it would be helpful to have such an option ?

[kleanthisk10]

For example you can see this paper, https://academic.oup.com/eurheartj/article/34/3/220/550289, where
there is a two-part cost-effectiveness model for the first strategy, which includes a short-term decision tree and a long-term Markov structure that was used to estimate long-term costs and health outcomes (Figure 1). The long-term Markov model is a logical extension of the short-term decision tree.

In the first strategy, the initial counts of the markov model is based on the probabilities derived from the short-term decision tree, meaning that the number of individuals in each state at the beginning is different from the other one where all the individuals start from a specific state.

[KZARCA]

I see.
It is for now not possible to set different initial counts. If you are comfortable with R, I think that you need to change the run_model_ function and eval_strategy so that they can accept a list of initial counts. Once you're done, please submit a pull request.