Seasonal Disease Outbreaks

"Flu season" is something we deal with every winter. Cyclical or seasonal outbreaks of an infectious disease can occur for a variety of reasons, such as seasonal changes in host behavior (for example, more people staying indoors), the seasonal abundance of disease carriers (for example, mosquitos), and many others.

Another factor that can play a role in cyclical outbreaks occurs when the natural immunity to a disease has a short duration. This enables each individual in the population to repeatedly contract the disease. Graph 1 shows a hypothetical disease with periodic waves of infection made possible by the host's short period of natural immunity.

The simulation in Graph 2 has a lower Basic Reproduction Number (R₀) than Graph 1 and no other changes. The simulation still shows cycles of infection, but they are less intense and become smaller over time.

In the Sandbox you can press the Customize Settings button to model how disease variables, such as the Basic Reproduction Number (R₀), Duration of Natural Immunity, Duration of Infectiousness, Latent Period, and others can interact to create cyclical outbreaks.

Read the FAQ if you have questions about how to use the Sandbox, how to save and share your work, or what the different settings mean.

Tips and hints:

• During your exploration phase, slowly change one disease variable at a time to see and record its impact. Return to the settings for a common starting point (such as Graph 1) before exploring a different disease variable (use Customize Settings – Reset to Defaults). Only after you examine changing one variable at a time should you start combining changes to multiple variables.
• You can use Export to Excel to dump the simulation output. In a spreadsheet program you can do more sophisticated analysis than you can by just looking at the SimPandemic graphs.
• When you interpret your simulation output, remember that if the Latent Period is greater than zero, you will always have some infected individuals that don't display symptoms until their latent period ends. Therefore, the Symptoms curve will be lower than the Infected curve.

Some possible questions to explore:

• Which are the most important disease variables? Which play no role?
• For a given set of inputs, can you predict the period between outbreaks?
• What parameters can cause a disease to settle into a "steady state" where it doesn't result in outbreaks, but it never goes away?
• What happens if everyone contracts a disease before immunity disappears?

NOTE: If you press Re-run, you will notice (mostly) small changes to the output from run to run. This is because just like in the real world, many events are based on chance (e.g., who infects who and when). See more in the FAQ.