Article 52QW4 Predicting COVID-19’s path is tough. So is it fair to say Ottawa ‘botched’ it?

Predicting COVID-19’s path is tough. So is it fair to say Ottawa ‘botched’ it?

by
Tonda MacCharles - Ottawa Bureau
from on (#52QW4)
theresa_tam.jpg

OTTAWA-Is Canada flattening the coronavirus curve? Are we still on the way up? Or on the way down? Where are we headed? Who knows for sure.

The federal government was slated to release new national modelling projections Friday but did not.

Health Minister Patty Hajdu's office said the latest forecasts aren't ready with officials still working on them.

However the Star has learned that the last time Ottawa released what it called the government of Canada's "national modelling," the package was not Ottawa's own projections based on data it has at hand, but a "synopsis" of scenarios produced by internal and external experts.

In response to questions from the Star, Health Canada said it produced "a synopsis of modelling studies including those conducted by the Public Health Agency of Canada and by other epidemiologists and modellers in Canada and elsewhere in the world."

The department pointed the Star to a couple of published outside models. Yet one of those, published in the Canadian Medical Association Journal, projected a much higher infection rate than what chief public health officer Dr. Theresa Tam's team did.

The study suggested infection rates, even with strong controls over a 12-month period, could see 60 per cent or more of the population infected, not 10 per cent as Ottawa had projected.

Its authors underlined that "Although events in China, Singapore, Hong Kong and elsewhere have shown that COVID-19 epidemics can be contained, the seeding of epidemics in countries around the globe, many with weak health systems means that reintroduction of COVID-19 will continue to occur for some time."

In any event, for all the experts that Health Canada says it consulted, it produced a forecast on April 9 that was quickly outdated.

The document used by Dr. Tam showed Canada had 401 COVID-19 deaths on April 8. She told reporters that by April 16, the following week, between 500 and 700 people would likely die.

By April 16 a week later, in fact, there were 1,048 deaths, with close to half linked to long-term-care homes.

Tam defended her agency's projection when asked by the Star this week why it presented a synopsis.

"Modelling is for planning and this is not real life," Tam said. "It's simulations for planning purposes and they do have to be updated."

Tam said modelling is complicated by the fact that "we don't know enough about the trajectory of this virus. We're learning every day." She pointed to the alarming clusters in long-term-care homes.

Professor Amir Attaran, who teaches at the University of Ottawa's law and medical schools, doesn't buy it.

He says the department's admission that it presented a synopsis not an actual mathematical model shows what he has argued before the Commons health committee: that the federal government has "botched" its modelling work.

Attaran said with provinces now seeking to reopen parts of their economy, good modelling is not just useful, it's "indispensable." They are the compass to guide whether to open schools early, how soon to open restaurants and other sectors, how to staff airlines once commercial flights become available.

"Everyone knows that models are not perfect forecasts but that is not an excuse not to do them. The weather forecast isn't perfect, the forecast of inflation isn't perfect. The Bank of Canada and Environment Canada persist anyway. The Public Health Agency of Canada is simply failing to provide a model forecast " They're trying to make a synopsis of multiple models, but there is no scientific way to synopsize models."

"Anything less than the disclosure of a model, the input data, the methodology and the output is scientifically incomplete and untrustworthy," he said.

Attaran conceded the federal job is made much harder by the fact that provinces do not all gather and report data in a consistent way, nor do they provide the data to Ottawa in real-time - a long-standing recommendation from past reviews of Canada's health system, including after SARS and H1N1 outbreaks.

Even in its daily epidemiology updates, Ottawa concedes it is working with data on fewer than 60 per cent of reported COVID-19 cases, or about 24,000 of 41,000 confirmed cases reported Friday.

Attaran says it's infuriating that even once Ottawa receives data, the federal government strips out a lot of information - such as province of origin - from the microdata it does provide outside scientists who are also trying to model the trajectory of the pandemic in Canada.

Dionne Aleman, an industrial engineer who teaches at the University of Toronto and models health systems and pandemics, said in an interview that she understands why the federal government needed to consult outside experts.

Modelling is complex, subject to many variables, and requires sophisticated academic approaches that may not be available within the public service, and is subject to many variables.

"It's not just 'here's the model, plug in the numbers and then we're done,'" she said. "It's a whole estimation of possibilities that could happen."

But Aleman agrees it's important for governments to disclose not only the data they have but also the data they are lacking.

Both are important for public policy decision making and to let people know where we are on the disease trajectory, she said, "because at least in my mind"the most stressful thing about this long self-isolation physical distancing hardship that we're all going through is just not knowing when it's going to end."

Even knowing what the information the government does not have "helps put the models in the right context," and could fuel some "political willpower to institute data collection systems that will allow us to have that information in the future," she said.

One thing Aleman, Attaran and Tam all agree on is the need for all governments, and Canadians, to be careful.

Aleman says looking back at the week that has just passed, Canada may well have hit the plateau, or be in the middle of flattening the curve of the disease's trajectory, at least of a first wave.

"Every piece of information that we've seen about COVID-19 in every country is a cautionary tale about the importance of maintaining strong physical distancing measures, the importance of contact tracing, the importance of testing lots of the population, of having good border controls," she said.

Tonda MacCharles is an Ottawa-based reporter covering federal politics for the Star. Follow her on Twitter: @tondamacc

External Content
Source RSS or Atom Feed
Feed Location https://www.thespec.com/rss/article?category=news
Feed Title
Feed Link https://www.thespec.com/
Reply 0 comments