Sorting the pandemic scenarios

Hey there, time traveller!
This article was published 03/12/2020 (1441 days ago), so information in it may no longer be current.

Imagine playing four versions of open-ended video game SimCity at once, with 1.4 million different characters and 165 different parameters.

That’s how a mathematical modeller projects where Manitoba’s COVID-19 cases numbers are headed under four different scenarios.

The province’s “agent-based” pandemic modelling (made public Friday) simulates what’s probably going to happen, depending on how many and when restrictions are put in place, and how many and when people comply with them.

It creates 1.4 million “agents” or unique entities, who will make decisions based on probabilities. The modelling uses data from Manitoba Health, the census program, peer-reviewed journals, expert assessments, and a few educated guesses to decide things such as: how many are health-care workers, what are their chances of them catching and spreading the novel coronavirus, how many will be symptomatic, how many people will they infect over a certain period of time, how many will end up in hospital and for how low, and how many will require intensive care?

Meanwhile, it doesn’t include schools nor project what’s likely to happen if K-12 students are in class or stay home, or what to expect when they return after the holiday break.

The modeller inputs the data and uses computer clusters to run at least 100 versions of each of the four scenarios simultaneously, keeping an eye on each of them and recalibrating one parameter at a time. The process can take up to 15 days.

“Modelling helps us better understand how the COVID-19 virus spreads and how we can target public health measures,” chief provincial public health officer Dr. Brent Roussin said in news release Friday.

“This allows us to take steps to prevent periods of extreme growth in COVID-19 activity while balancing the social and economic impacts on Manitobans.

The four scenarios modelled:

• Extreme: minimal public health restrictions are introduced and compliance with basic public health recommendations is low, resulting in significant COVID-19 activity, including a rapid rise in cases.

• Severe: some public health restrictions are introduced, but compliance with fundamental public health recommendations remains low, resulting in increased cases though at a less rapid rate than Scenario 1.

• Moderate: additional public health restrictions are implemented and compliance improves, with individuals beginning to adhere to fundamental recommendations, resulting in a level of COVID-19 activity which, if sustained, is manageable for the health-care system.

• Controlled: strong public health restrictions are put in place and compliance is observed as individuals follow the fundamental recommendations, resulting in the lowest possible level of COVID-19 activity.

December COVID update

carol.sanders@freepress.mb.ca