Skip to main content
Data Source
Any data publicly available
Organizer
Dr Kathryn Morrison with (Dr. Chel Hee Lee, Dr. Ehsan Karim,and Dr. Zhaozhi Fan)

Background

Compartment models refer to a broad class of infectious disease models that seek to understand how infectious diseases spread throughout susceptible populations. The basic SIR model stands for susceptible, infected, and recovered, indicating the states (or compartments) of individuals at a given point in time. More complex versions of the model include, for example, compartments where individuals are exposed but not yet infectious (a latency period) or a return to susceptibility after recovery. The rate at which individuals move between states is usually modelled by ordinary differential equations, though both deterministic and stochastic versions of these models exist. The goal for the application of compartment models is not to realistically model or predict the outbreak exactly, but rather to explore plausible scenarios for infections given various input parameters and constraints.

 

Case study objectives

The goal of this case study is to explore the dynamics of SARS-CoV-2 in Canada with a compartmental model, and define one to three clear research questions that your model seeks to address.  You may choose any type of compartment model within this broad class, and use any data or parameters found from publicly available sources and/or published literature to run your simulations.

 

Research Question

Some example research questions are included below; you may select one or choose your own. If you are unsure whether your research question(s) would meet the case study competition criteria, please check with Dr. Morrison.

 

Sample research question 1: What would have been the impact of fewer or no public health interventions / shut-downs on the total number of cases and deaths of COVID-19 in Canada during 2020? (You could instead focus on a particular geographic area within Canada, such as a city or province).

 

Sample research question 2: The initial stages of the COVID-10 outbreak in Canada were challenging for public health for many reasons, one being a lack of sufficient tests to know the true number of infections in the community. Given the number of deaths that occurred in Canada (or a given geographic area of your choosing), and the best estimates of the mortality rate of COVID-19, what were the likely true number of infections over time given the early stage of the outbreak in March and April 2020?

 

Your case study report and poster must include:

  • The research question(s) you sought to address with your analysis.
  • The source of all parameters used in your simulation models (for example, population numbers, reproductive numbers, incubation period length).
  • A discussion on the impact of your assumptions and parameters and the limitations of these types of models.
  • At least one visualization of your simulated data over time.
  • A summary of the key takeaways from your analysis.

 

Evaluation & grading points

Each team should design a poster that evaluates their research question(s), and present their results for approximately 10 minutes (plus an additional 8-10 minutes for discussion). The case study competition will be evaluated as follows:

§  Creative visualizations of the data (20%)

§  Thoughtfully curated input data for the given research question (10%)

§  Appropriateness, creativity, and understanding of the strengths and limitations of the model proposed (50%)

§  Quality and clarity of presentation (20%)

 

Resources for getting started

There are many existing packages to support compartment modelling, and many tutorials for understanding how the various versions of these models operate.

 

Useful papers

  • He, Shaobo, Yuexi Peng, and Kehui Sun.
    "SEIR modeling of the COVID-19 and its dynamics." Nonlinear Dynamics 101.3 (2020): 1667-1680.
     
  • Iwata, Kentaro, and Chisato Miyakoshi.
    "A Simulation on Potential Secondary Spread of Novel Coronavirus in an Exported Country Using a Stochastic Epidemic SEIR Model." Journal of Clinical Medicine 9.4 (2020): 944.
     
  • Ganyani, Tapiwa, Christel Faes, and Niel Hens.
    "Simulation and Analysis Methods for Stochastic Compartmental Epidemic Models." Annual Review of Statistics and Its Application 8 (2020).

 

Variables

Any variables available from the database chosen for the study.

 

Data Access

 Open access.

R Packages

 

Python Packages

 

Organizer contact information

This case study was prepared by Dr. Kathryn Morrison with help and guidance from the other members of the case study committee of the Statistical Society of Canada (Dr. Chel Hee Lee, Dr. Ehsan Karim, and Dr. Zhaozhi Fan). Any questions or concerns can be directed to: kathryn@precision-analytics.ca.