Simulation models: to reuse or not to reuse?

Simulation Explained, Simulation in Healthcare



Whether simulation models can be reused for similar problems is a controversial issue in literature. Robinson et al (2004) make the point that anyone reusing a simulation model has to trust it. In order to trust it, they need to understand it and validate it for their local circumstances – and that may take the same amount of time as it would have taken to build a simulation from scratch. If the point of simulation models is to help make better, more evidence-based decisions, then does a ready-made simulation prevent stakeholders from getting to grips with understanding the problem?

On the other hand, it takes time to build a simulation model for the first time. Our experience in healthcare is that expert simulation modelers working in the sector are relatively rare – more so in the UK than in the US. The chances are that if a problem needs a quantified solution, simulation may not be used – even if it is the best technique. Naturally it would take longer at first to build up the skills but in the longer term simulation could be the most effective. Robinson et al suggest building simple simulations that teach lessons about possible solutions and promote understanding of the system. The simulation modeler is used as a facilitator of improvement rather than providing “the answer”.

Aware of these caveats, I wanted to reflect on our experience of re-using simulation models particularly in modeling pathways of patient care. The parameters of these kinds of simulations are variously based on:

  • Population demand forecasting – how many patients of what type are expected to need services?
  • Service utilization – what is the flow of patients from service to service/provider to provider, depending on need?
  • Clinical best practice – what is the impact of implementation for all patients?
  • Disease progression – how does this affect the way in which patients need services, and their likely outcomes if they do or do not receive them?

While all of these perspectives are important, the constant in any care pathway is the way in which a disease progresses. This is the same regardless of healthcare provider or network. The data on disease progression can also be sourced in public health, health economics and epidemiology peer reviewed journal papers, which means it can be a desktop exercise at least initially.

We have found that basing a patient care pathway simulation on the way in which a disease behaves can be reused as a base simulation and is easily referenced from the literature. Once stakeholders have understood and trust the underlying disease pathway simulation – and clinical endorsement and trust of this progression simulation is crucial at this point – then service utilization can be validated against organizational baseline data and service improvements and interventions can be tested.

A great example of reusing simulations is our recent sepsis work. This approach worked well and delivered great benefits. You can hear more about the benefits in the below video where I show an example of a disease pathway simulation, focusing on sepsis – from initial infection through to sepsis shock.

 

Learn more about using SIMUL8 for healthcare process improvement

Find out more about how simulation is used by healthcare providers, read more case studies and access a range of learning resources.

 

S. Robinson et al. / Simulation Modelling Practice and Theory 12 (2004) 479–494