On 2014 May 07, T Eugene Day commented:

This paper addresses a critical aspect of modeling complex systems in healthcare. Dynamic human decision makers exhibit important complexities within systems, and there are demands on their time and attention that should be included in rigorous models. This is especially true in academic environments where attending physicians are required to oversee and instruct residents and fellows. Crucial systemic outcomes depend on these interactions, as well as medical outcomes.

The validation of this model could be more informative if the authors had compared the simulated data to real-world data from their modeled ED. This would allow us to examine how well the inclusion of physician-delegate interactions allowed them to mimic the performance of a real-world ED. This would provide stronger evidence that this model can be used to inform practice.

Ever more lifelike models of human behavior will be required to make useful predictions of systems performance, and this is an important step on that road. The use of pseudo-agents, rather than a strict resource/entity model, allows greater flexibility and autonomy on the part of the simulated human resources.

On 2014 May 07, T Eugene Day commented:

This paper addresses a critical aspect of modeling complex systems in healthcare. Dynamic human decision makers exhibit important complexities within systems, and there are demands on their time and attention that should be included in rigorous models. This is especially true in academic environments where attending physicians are required to oversee and instruct residents and fellows. Crucial systemic outcomes depend on these interactions, as well as medical outcomes.

The validation of this model could be more informative if the authors had compared the simulated data to real-world data from their modeled ED. This would allow us to examine how well the inclusion of physician-delegate interactions allowed them to mimic the performance of a real-world ED. This would provide stronger evidence that this model can be used to inform practice.

Ever more lifelike models of human behavior will be required to make useful predictions of systems performance, and this is an important step on that road. The use of pseudo-agents, rather than a strict resource/entity model, allows greater flexibility and autonomy on the part of the simulated human resources.