Early 'excess death' data for Canada suggests decrease in deaths in early days of pandemic
Preliminary StatsCan figures capture January-March, before the peak hit
Data released Wednesday by Statistics Canada suggests that fewer Canadians died in the first three months of 2020 than in the same period in previous years, despite the onset of the coronavirus pandemic.
In a first attempt to shed light on the real death toll of COVID-19, the agency published data from eight provinces and one territory on "excess deaths" — an increase over what would be expected for the same period of time based on historical data.
But the numbers available so far — 87,186 deaths in the three-month period, a decrease of 1,145 compared to last year— suggest there were fewer deaths overall in Canada, with excess deaths detected only in Alberta and British Columbia. Experts say that's because of the patchwork of reporting methods across jurisdictions. It also suggests lockdown measures may have reduced the number of non-COVID-related deaths, they said.
- INTERACTIVE | The latest data on coronavirus cases in Canada
In countries such as the United States and the U.K., similar analyses revealed hundreds, sometimes thousands, of excess deaths. Experts say excess deaths are a more reliable measurement because they capture not only deaths directly linked to COVID-19 but also those that might be indirectly linked, such as heart attacks or strokes that weren't treated in time because of pressure on the health system.
Canada's first confirmed COVID-19-related death was on March 8 in British Columbia. By the end of March, there were 126 confirmed deaths.
The bulk of confirmed deaths happened since then, most of them in Quebec and Ontario.
"The data are provisional; there are deaths missing from there,'' said Owen Phillips, senior analyst with Statistics Canada's vital statistics program.
"As more data become available, and we are able to attach a cause of death and look at mortality rates across different ages and sexes, then we'll be able to get a much better understanding of how Canadians have been affected.''
WATCH | What has two months of physical distancing accomplished?
One researcher explained why the numbers so far don't tell the full story.
"People generally die about four weeks after initial infection, so to die by March 28, you would have to be infected at the beginning of March, and we had only a handful of cases at that time," said Jay Kaufman, a professor in the department of epidemiology, biostatistics and occupational health at McGill University in Montreal.
"This means that the impact on mortality isn't yet detectable."
Confusing figures in Quebec
Quebec has the largest reported number of COVID-19 deaths in Canada: As of Wednesday, there were 3,220 deaths recorded in that province — compared to 5,389 in the country, according to CBC's tally based on provincial data, regional health information and CBC's own reporting.
The death count reported by Statistics Canada raises more questions than answers, according to one expert.
Prabhat Jha, professor of epidemiology at the Dalla Lana School of Public Health at the University of Toronto, compared the number of deaths in Quebec reported by Statistics Canada with data obtained from the Quebec Ministry of Labour, Employment and Social Solidarity by the National Post.
The data from the Quebec government shows an increase of 214 in deaths in March, not a decrease of 563 as in the StatsCan data. Jha sees this as a flaw in how data is shared between provinces and the federal government.
"Both can't be right," Jha said. "This needs to be sorted out. This is a call for better co-ordination between provinces and feds during a pandemic."
No numbers for Ontario and N.B.
Another reason the preliminary numbers don't paint a complete picture is because they don't include Ontario, Canada's second hardest-hit province, or New Brunswick. Data for Nunavut and Yukon is also missing.
According to CBC's tally, 71 COVID-19-related deaths were recorded in Ontario during the month of March.
Phillips says that in recent years, it has taken Ontario and New Brunswick more than 60 days to share their monthly death reports with the agency.
He says gathering data on reported deaths in all provinces and territories is a "complex, decentralized process" that comes with several challenges — even in a non-pandemic year.
WATCH | Canada's contact tracers make difficult calls:
"We rely on vital statistics offices. They have different legislations, different mandates, different capacity issues, as well as different data collection methods. So we are up against 13 different data collection processes,'' Phillips said.
According to Statistics Canada's own methodological notes, numbers from Nunavut were suppressed "owing to their incompleteness." The agency also notes that this data has not been received from Yukon since 2017.
To Colin Furness, an infection control epidemiologist and assistant professor at the University of Toronto, this reveals flaws in how data is shared in a timely manner among provinces and the federal government.
"I think in some ways this is a story about bad data," Furness said in an email to Radio-Canada. "But we should wait until April data is available and, hopefully, Ontario will participate."
Lower death numbers could still be good news
The StatsCan data may still offer some insights, according to Greta Bauer, a professor in the department of epidemiology and biostatistics at Western University in London, Ont. For one thing, it can show that some non-COVID deaths were prevented in the early days of the pandemic.
Physical distancing and confinement measures adopted by some provinces as early as mid-March could have reduced the number of people dying from seasonal flu and traffic accidents because Canadians were staying at home.
It's also possible there were fewer suicides, she said.
"We're changing the death risk by having people socially distant," Bauer said. "What we know about suicide risk may not be valid in a pandemic."
For instance, unemployment is a high suicide risk, but in the context of COVID-19, it could be viewed differently.
"People are getting more financial support, and maybe there's less self-blame," she said.
However, without knowing the causes of the deaths, these remain purely conjecture.
Statistics Canada promises "a clearer picture will emerge" soon, as more data becomes available. The agency will release a similar analysis next month, which will capture all deaths reported by provinces and territories in the month of April.