36.2 Entry state - adding to the cohort

The Entry node type provides a mechanism to add to the size of the cohort with each cycle. The Entry node will be automatically executed at the start of each cycle, so it does not require an Initial Probability. Entry nodes are followed by subsequent branches with probabilities which specify the additional population added to the cohort at each cycle.

Consider the Healthcare tutorial example tree, Dynamic Cohort - Markov Cancer.trex first described in the prior section. Note that the Markov node is already set to allow non-coherent branch probabilities (see prior section). This section will focus on the Entry node for adding to the cohort.

To build a model adding members to the cohort:

• To add additional members into the cohort, insert an Entry Node as a direct branch of the Markov node.

  • The Entry node does not require an initial probability.

  • The Entry node will be automatically run every cycle.

  • Entry nodes automatically allow non-coherent branch probabilities.

• Add branches of the entry node to add to the cohort by sending new entries to one of the true health states via jump states at terminal nodes. The new entries will appear in those health states at the start of the next cycle.

The figure below shows the Entry node with the red flag indicating non-coherence at the node.

In this model, the Entry node is called New Patients and the subtree associated with the Entry node adds nNewLocal members to the Local Cancer and nNewMets members to the Metastases state each cycle.

The "health state" represented by the Entry node will not be included in most of the Markov Cohort reports, but the new additions to the cohort will appear in the healthe states in the next cycle. In this example the same number of people are added in every cycle, but nNewLocal and/or nNewMets could refer to formulas that change by cycle.