Flowers are shown outside Maison Herron, a long-term care home in the Montreal suburb of Dorval, on April 12, 2020.Graham Hughes/The Canadian Press
The loss of life in the pandemic’s first nine months was more widespread than official numbers indicated, with mortality spiking sharply even in provinces that reported relatively low death tolls from the virus, according to newly available national data.
A revised geography of the pandemic in Canada emerges from an analysis of the figures by Tara Moriarty, a professor of laboratory medicine and pathobiology at the University of Toronto.
Official COVID-19 deaths in the spring and summer of 2020 were mostly limited to Ontario and Quebec. However, the rate of “excess deaths” – those over and above what would have been expected given the trend in recent years – was much more pronounced in Western Canada than virus fatalities alone can account for.
How many coronavirus cases are there in Canada?
The statistics suggest some provinces may have inadvertently undercounted COVID-19 deaths by a wide margin, Dr. Moriarty said. Provincial governments themselves have argued the disparity could be caused by hundreds of ailing people dying at home while avoiding hospitals for fear of the virus.
Whatever their exact cause, the number of excess deaths shows a country that suffered much more equally than previously believed. (The analysis found there was virtually no gap between COVID-19 deaths and excess deaths in Atlantic Canada and the North.)
Estimated excess deaths versus reported
COVID-19 deaths, select provinces
March 7, 2020 to Jan. 2, 2021
B.C.
Alta.
Sask.
Man.
Ont.
Que.
Cumulative excess
deaths rate per 100,000
Cumulative COVID-19
deaths rate per 100,000
100
80
60
40
20
0
April
2020
Aug.
Dec.
April
2020
Aug.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE:
STATISTICS CANADA; PROVINCIAL GOVERNMENTS
Estimated excess deaths versus reported
COVID-19 deaths, select provinces
March 7, 2020 to Jan. 2, 2021
B.C.
Alta.
Sask.
Man.
Ont.
Que.
Cumulative excess
deaths rate per 100,000
Cumulative COVID-19
deaths rate per 100,000
100
80
60
40
20
0
April
2020
June
Aug.
Oct.
Dec.
April
2020
June
Aug.
Oct.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE:
STATISTICS CANADA; PROVINCIAL GOVERNMENTS
Estimated excess deaths versus reported COVID-19 deaths, select provinces
March 7, 2020 to Jan. 2, 2021
B.C.
Alta.
Sask.
Man.
Ont.
Que.
Cumulative excess
deaths rate per 100,000
Cumulative COVID-19
deaths rate per 100,000
100
80
60
40
20
0
April
2020
June
Aug.
Oct.
Dec.
April
2020
June
Aug.
Oct.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE: STATSCAN; PROVINCIAL GOVERNMENTS
Between March and November last year, British Columbia saw about six times as many “excess deaths” as reported COVID-19 fatalities, Dr. Moriarty found. Saskatchewan saw 10 times as many. The figures suggest hundreds if not thousands more people may have died in Western Canada as a result of the pandemic than official estimates recorded.
“I think we’ve had a much bigger epidemic than we know,” she said.
To gain a clearer picture of the virus’s real toll, Dr. Moriarty parsed data compiled by Statistics Canada dating from the beginning of the pandemic to mid-November, the most recent point for which fairly complete numbers are available for all provinces.
Dr. Moriarty also subtracted the growth in toxic drug deaths from her analysis, to make sure she wasn’t simply capturing the human cost of Canada’s parallel opioid-poisoning epidemic. The numbers that remained were stark and surprising.
Even controlling for overdoses in this way, B.C. still had 1,650 excess deaths in those nine months, compared with 290 official COVID-19 deaths. Among adults over 64 – those most vulnerable to COVID-19 – there were 1,099 excess deaths.
In Saskatchewan, the disparity was even more dramatic: 293 excess deaths against just 29 COVID-19 fatalities. Alberta (1049 vs. 401) and Ontario (4,037 vs. 3,372) also had considerably higher excess death tallies than their official pandemic death toll. All told, Dr. Moriarty’s calculations show more than 3,000 excess deaths, over and above those attributed to COVID-19, outside of the Quebec epicentre.
“We’re missing something,” she said.
Difference in estimated excess deaths
vs. reported COVID-19 deaths, Canada
Weekly deaths, March 7, 2020 to Jan. 2, 2021
Estimated excess deaths
COVID-19 deaths undercounted
Reported COVID-19 deaths
COVID-19 deaths overcounted
1,400
1,200
1,000
800
600
400
200
0
-200
April
2020
June
Aug.
Oct.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE:
STATISTICS CANADA, PROVINCIAL GOVERNMENTS
Difference in estimated excess deaths versus
reported COVID-19 deaths, Canada
Weekly deaths, March 7, 2020 to Jan. 2, 2021
Estimated excess deaths
COVID-19 deaths undercounted
Reported COVID-19 deaths
COVID-19 deaths overcounted
1,400
1,200
1,000
800
600
400
200
0
-200
April
2020
June
Aug.
Oct.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE:
STATISTICS CANADA, PROVINCIAL GOVERNMENTS
Difference in estimated excess deaths versus reported COVID-19 deaths, Canada
Weekly deaths, March 7, 2020 to Jan. 2, 2021
Estimated excess deaths
Reported COVID-19 deaths
COVID-19 deaths undercounted
COVID-19 deaths overcounted
1,400
1,200
1,000
800
600
400
200
0
-200
April
2020
June
Aug.
Oct.
Dec.
MURAT YÜKSELIR / THE GLOBE AND MAIL, SOURCE: STATISTICS CANADA, PROVINCIAL GOVERNMENTS
If there has been widespread undercounting of COVID-19 deaths, it would fit the international pattern. A U.S. study in September found virus deaths in that country may have been undercounted by as much as 36 per cent. The Office for National Statistics in Britain has acknowledged excess deaths in the first wave significantly outran the official pandemic death count, because limited testing and inexperience with a new disease led health authorities to miss many cases.
Canadian researchers such as Prabhat Jha, director of the Centre for Global Health Research at St. Michael’s Hospital in Toronto, have been raising the prospect of undercounting since last spring.
The most recent figures suggest that is likely what happened in Canada, Dr. Moriarty says. “If anything, this is the norm – undercounting of COVID,” she said. “There’s no reason to think we’re immune from this.”
How could so many Canadians have died uncounted amid such a well-publicized pandemic? In one scenario that haunts Dr. Moriarty, and which other researchers agree is plausible, a frail elderly person living at home in rural B.C. or Saskatchewan develops the sniffles, but none of the classic COVID-19 symptoms such as a cough, fever or shortness of breath.
Frail seniors are now known to mount a weaker immune response to the coronavirus, so “present atypically,” a phenomenon that wasn’t well understood earlier in the pandemic, said Janet McElhaney, a geriatrician and professor at the Northern Ontario School of Medicine.
The possibility that certain provinces “didn’t cast a wide enough net” through testing, causing them to miss cases and deaths, is a credible theory, said Lynora Saxinger, an associate professor of infectious diseases at the University of Alberta.
“That story makes sense to me,” she said. “It tells us that the shape of the first wave might have been different.”
But Dr. Saxinger, who is also co-chair of the COVID-19 Scientific Advisory Group for Alberta Health Services, said other factors could be at play, such as people with chronic conditions avoiding medical care because of the virus.
A recent report from the Canadian Institute for Health Information pointed to a sharp decline in medical care for nursing-home residents during the pandemic’s first wave, which could have caused deaths from other conditions.
Several provinces argued in written responses to The Globe and Mail that indirect casualties of the pandemic were a significant factor in the gap between COVID-19 and excess deaths. They played down the possibility of undercounted virus deaths.
In B.C., the share of people who died at home rose from 18 per centto 26 per cent between February and April last year, said Sarah Henderson, scientific director of Environmental Health Services for the BC Centre for Disease Control. That could suggest the pandemic “prevented people from seeking potentially life-saving acute care,” she wrote.
Still, widespread undercounting of COVID-19 deaths remains the most credible explanation, Dr. Moriarty says. More deaths at home could equally support the undercounting theory, she argued, if seniors in the community were being felled by the virus before reaching hospital because of unusual symptoms and low testing rates.
Indeed, B.C. performed among the fewest tests per capita last year, and also had the most unaccountable excess deaths, raising the possibility that the virus was circulating more widely than case counts suggested.
Another factor separating provinces is some may have been more rigorous about attributing deaths to COVID-19. Most use a variation of the Public Health Agency of Canada definition to tally deaths from the virus (“A probable or confirmed COVID-19 case whose death resulted from a clinically compatible illness, unless there is a clear alternative cause of death”).
But some provinces have a slightly more restrictive definition: Public Health Ontario only counts COVID-19 deaths among “confirmed” cases of the virus, for example, not “probable” cases, as the federal agency recommends.
Curiously, in Quebec, the story was reversed: Authorities reported more deaths from COVID-19 last year than total excess deaths. Dr. Moriarty says she believes this anomaly could be a product of the extremely advanced age of many of the province’s pandemic victims, along with more stringent reporting standards.
Quebec’s Director of Public Health, Horacio Arruda, has long argued that Quebec counts mortality from the virus more aggressively than other provinces. This could be because of more thorough postmortem contact tracing or a tendency to ascribe deaths to COVID-19 when there are other potential causes such as cancer, explained Rodica Gilca, a public-health specialist with the Institut national de santé publique du Québec.
“I think there’s a better method of capturing here than in other provinces,” said Dr. Gilca, who is also an associate professor in the faculty of medicine at Laval University. “Even if we overdeclare a bit … we’ll have the most exhaustive possible portrait.”
The remaining uncertainty about what exactly caused the spike in excess deaths during the pandemic points to the need for more detailed numbers about mortality in Canada, Dr. Moriarty says. Provinces often take months to release the cause of deaths to Statistics Canada, for example.
Although the agency has sped the release of mortality data during the pandemic, the country still has no clear picture of deaths during the country’s devastating second wave, and remains months slower in producing such figures than countries such as New Zealand and Britain.
That has left researchers such as Dr. Moriarty to analyze what numbers we have and raise questions about how exactly thousands of Canadians died at the beginning of the pandemic.
“I think it’s worth having a national conversation about this,” she said.
Sign up for the Coronavirus Update newsletter to read the day’s essential coronavirus news, features and explainers written by Globe reporters and editors.