Math saves lives: How scientists use modelling to guide COVID-19 decision-making

For those of us who struggled through high school math classes, it’s hard to imagine it could actually be used to save lives! But that is exactly what scientists at the Public Health Agency of Canada’s (PHAC) National Microbiology Laboratory (NML) are doing—using data analysis to create forecasts based on mathematical models. This data, presented to Dr. Theresa Tam, Canada’s Chief Public Health Officer and the Council of Chief Medical Officers of Health, helps guide COVID-19 public health decision-making.

Bridging the past, present and future

Dr. Michael Li, NML Senior Scientist, describes modelling as a “way to bridge information from the past, present and future.”

NML scientists use different types of mathematical models to help them answer questions about how COVID-19 spreads over time based on information such as case counts, demographics, hospitalizations, deaths, testing, variants of concern and exposures. Models create projections for the future based on past and present data and help decision makers to evaluate and, if need be, adjust public health measures to control the spread of COVID-19. These models help answer complicated questions such as the most favourable time to reopen after a lockdown, how long to maintain individual precautions like physical distancing and the risk associated with different gathering sizes. Decision makers at various jurisdictional levels use a combination of results from different models to inform public health actions in order to reduce the burden on our hospitals and help save lives.

The more you know, the better

Dr. Li maintains one of the models at PHAC that provides weekly forecasts for all the provinces in Canada and monthly long-range forecasts. In addition to Canadian data, Dr. Li uses COVID-19 data on variants from the UK to create projections with the assumption that Canada would follow similar patterns. For example, data showing how fast a variant of concern such as B.1.1.7 Alpha variant at the end of 2020 and now the B.1.617.2 Delta variant is growing in the UK helps inform what could happen in Canada. In February 2021, while overall cases in Canada were trending down because of public health measures, Dr. Li’s projections showed it was just the beginning of the third wave. While there was some skepticism, it turned out to be an accurate projection. It is in this way that modelling provides a snapshot of what might happen in the future giving officials more time to prepare.

Dr. David Champredon, NML Senior Scientist, also integrates wastewater data that the NML gathers into the modelling exercises. Sampling wastewater is a relatively new method for detecting community cases of COVID-19 (and other infectious diseases). By using this data in models, it helps scientists gain a better understanding of the dynamics of the spread of the virus allowing them to make projections that better inform local decision makers. Most models only rely on data from traditional COVID-19 testing methods, which can’t account for everyone because testing capacities are usually limited. Hence, wastewater data is very valuable because it accounts for even asymptomatic and mildly symptomatic people who may not get tested, but who still shed the virus through their waste. (To learn more, read our blog on wastewater surveillance.)

“If you want to control the disease by putting measures in place, the more you know the better,” says Dr. Champredon. “The great thing about wastewater is that you can sample a whole city, and it is an efficient way to detect COVID-19 in the community and provide early warnings of possible outbreaks.”

Dr. Antoinette Ludwig, NML Veterinary Epidemiologist, and her team use a model that separates the population into groups based on disease status, such as asymptomatic, symptomatic and recovered cases. This simple, efficient way of representing the spread of the disease in the population, is used to show the impact of different public health strategies throughout the course of the pandemic, such as quarantining, level of case detection and vaccination. In one study, scientists explored the use of the less sensitive rapid antigen tests in a wider range of the population (including asymptomatic people) to increase detection of cases. The antigen tests were modelled as a complement to traditional polymerase chain reaction (PCR) molecular testing to detect possible COVID-19 infections. The models showed that antigen tests would contribute to reduce the impact of the third wave of the pandemic in some settings. This was confirmed by Canadian experience. In Nova Scotia for example, community rapid testing is believed to have contributed to “bending the curve” of the third wave. They were also successfully used in Canadian workplaces and university residences to identify asymptomatic cases and stop the spread, preventing further outbreaks.

The data era

Modelling is still a relatively new field and advancements in technology with COVID-19 are pushing it even further along.

“In just over a year, 10-20 years of research was produced. The work done over this year will help us to be better prepared and more efficient at providing projections for possible outcomes when the next pandemic occurs,” says Dr. Champredon.

Since the start of COVID-19, collaboration has greatly increased between scientists across government departments and within the scientific and medical communities across Canada.

“It was an opportunity to shed some light on our expertise in public health. We are constantly being asked questions to explain our projections to public health officials and working under pressure, but it is very important for us to contribute to the solution and help manage this pandemic to protect the lives of Canadians,” says Dr. Ludwig.

Now, enormous data sets exist that provide information about how the virus spreads based on our behaviour such as wearing masks or staying home when sick. The integration of geographical data from cell phones, along with testing, has helped tremendously during the pandemic. Now there is a need to pull all of this data together to make sense of it all.

“We’re heading into a new technological era where everything is data-driven,” says Dr. Li. “Modelling plays an important role in controlling disease spread, and we can use it to help link all of this information together to come up with better overall public health strategies.”

To see all the models that have been presented by the Chief Public Health Office please visit