What happens in the Southern Hemisphere usually provides healthcare experts with a preview of what will happen in the Northern Hemisphere. Unfortunately, if early data from Australia holds true, the 2022-2023 flu season in the U.S. could be one of the worst on record. Pair that with RSV and still circulating COVID-19 cases and the prospect of new variants, and a possible “tripledemic” this winter begins to look more likely.
Hospitals that engage in predictive analytics and specifically discrete-event simulation for process optimization will be far better prepared to handle the influx of patients. But what is discrete-event simulation? How does it work? How can it help hospitals proactively plan staffing and resource needs, this winter and beyond?
What is discrete-event simulation?
Discrete-event simulation is a process used to replicate and simulate real-life operational behavior patterns. Essentially, discrete-event simulation takes events that occur at specific times and in a complex environment – like a hospital or Integrated Delivery Network (IDN) – and uses that information to determine what is likely to happen next.
Discrete-event simulation can be instrumental in giving nurses managers and patient flow administrators strong insights into how patients are likely to progress throughout their facilities, thereby aiding in capacity planning and needed staffing levels. For instance, each patient that enters a hospital is modeled based on their personal electronic health records and historical data pulled from cohorts of similar patients. The system uses data analytics to predict and simulate where that patient is likely to go, how long they’re likely to stay in the hospital, and more.
How can discrete-event simulation be used to predict capacity?
When paired with a digital twin, the simulation provides healthcare workers with a completely new level of insight into their censuses and capacity.
A digital twin is an accurate visual representation of what a hospital looks and behaves like at any point in time—right now, later today, tomorrow, next week, the next few months or years. The digital twin can show simulations of patient movements, including which care levels and rooms they’re likely to be in and when they’re likely to be discharged.
Users can select a timeframe to view these movements and see an accurate representation of how many patients will be in the hospital during that period and where they’re likely to go. 
For example, a nurse manager attempting to get an idea of how many staff they’ll need for a specific week in January could set the dates and view the predictive analytics software models that have been built through discrete-event simulation. They can then use this information to plan their staffing needs, ensure they have the right number of beds and equipment, and more.
How is discrete-event simulation used to predict “what if” scenarios?
A discrete-event simulation can be used in a What If Scenario Analysis (WISA) to show potential outcomes. WISA is used to determine the impact of different variables and process changes can have on operations.
For instance, a hospital can use discrete-event simulation to test different levels of patient surges and determine how each might impact operational excellence, hospital staffing needs, and other factors. The hospital can use the models resulting from the analysis to plan accordingly so that they are not caught short-staffed, or resource-constrained during a potential surge.
A WISA can be used not only to see how surges can impact operations but also to test contingency plans. The benefits of WISA are twofold:
- Hospitals can determine how different variables, like a census surge, can impact bed capacity.
- Hospitals can see how their plans will impact operations. For example, a WISA will show the impact of an action plan and how it affects patient flow.
Through a WISA, patient flow leaders can be sure that their plans are robust enough to work ahead of time. They’ll have a playbook that they know will work.
Why is it important to model individual patients?
Discrete-event simulation treats each patient as a unique individual with their own diagnoses, acuity levels, and testing needs. It considers the patient’s status and requirements, layers these factors on top of broader population data, and applies them over time.
With this information, hospitals can understand what will most likely happen to that patient when they’re in the hospital, including:
- What unit will they be placed in?
- What services will they likely need?
- How long will their length of stay be?
The ability of the system to treat every patient as an individual is critical, for a couple of reasons. First, patient care is not static, and different people require different levels of service. The simulation software needs to be flexible enough to build in multiple iterations and outcomes for each patient, or else the census predictions may not be accurate. Also, the more the system learns about different patients and outcomes, the smarter it becomes, and the more accurate future simulations will be.
Discrete-event simulation: a must-have for the coming flu/COVID-19 season
Discrete-event simulation is a must-have tool for every hospital or IDN at any time, but it’s especially important now, as health organizations prepare for the upcoming flu/COVID-19 season. If the season is as difficult as many predict it will be, healthcare facilities will need to start preparing now. Discrete-event simulation can give them insights into what the future holds so they can arrange to have the right number of resources for the right number of patients.
For more information on how discrete-event simulation can help hospitals proactively plan their censuses, download our latest whitepaper: Preparing for a Tripledemic: Resource Planning in the Face of Covid-19 / Flu / RSV.
Predictive AI census planning for your hospital system
FutureFlow Rx leverages predictive analytics and a digital twin of your hospital to provide an early warning system for census & capacity surges. Learn more about our AI software solutions for hospitals.