Provinces with least-strict COVID-19 restrictions this summer saw sharp case spikes in second wave, data show

Hundreds of people wait in line to enter McArthur Glen Designer Outlet, due to the shopping centre being at capacity, on Boxing Day in Richmond, B.C., on Saturday, December 26, 2020.DARRYL DYCK/The Canadian Press

The two Canadian provinces with the least-strict approaches to controlling COVID-19 over the summer have had some of the highest rates of infections and hospitalizations in the country during the second wave of the pandemic, according to data compiled by the University of Oxford.

Alberta and Saskatchewan opened up wider and stayed open longer than other provinces, according to the data, and then saw sharp spikes in COVID-19 rates in recent months.

Conversely, Atlantic provinces put in strong measures early on, including a travel bubble, and were able to keep COVID-19 rates near-negligible compared with elsewhere in the country. That quick intervention appeared to allow those provinces to ease up over the summer without seeing significant outbreaks.

How many coronavirus cases are there in Canada, by province, and worldwide? The latest maps and charts

When will Canadians get COVID-19 vaccines? The federal and provincial rollout plans so far

Is my city going back into lockdown? A guide to COVID-19 restrictions across Canada

The data were released by Oxford’s Blavatnik School of Government as part of its government response tracker, which attempts to quantify how seriously – or not – jurisdictions around the world are approaching COVID-19 restrictions. Previous data focusing on places like the United States show a strong relationship between the stringency of infection-control measures and the severity of outbreaks.

The project recently released provincial and territorial data in Canada, scoring each government’s pandemic measures in a variety of areas, including mask mandates, work-from-home orders, the closing of businesses such as restaurants, and travel restrictions.

Provinces are then ranked on a 100-point scale based on how stringent their overall response has been. The data track those scores over time.

RESPONSES TO COVID-19,

BY PROVINCE OR TERRITORY

University of Oxford’s stringency index

No data

B.C.

Alta.

Sask.

Man.

Ont.

Que.

N.B.

N.S.

PEI

N.L.

Yukon

NWT

Nunavut

Canada

Jan.

Mar.

May

July

Sept.

Nov.

MURAT yükselir / the globe and mail,

source: University of Oxford

RESPONSES TO COVID-19,

BY PROVINCE OR TERRITORY

University of Oxford’s stringency index

No data

B.C.

Alta.

Sask.

Man.

Ont.

Que.

N.B.

N.S.

PEI

N.L.

Yukon

NWT

Nunavut

Canada

Jan.

Mar.

May

July

Sept.

Nov.

MURAT yükselir / the globe and mail,

source: University of Oxford

RESPONSES TO COVID-19, BY PROVINCE OR TERRITORY

University of Oxford’s stringency index

No data

B.C.

Alta.

Sask.

Man.

Ont.

Que.

N.B.

N.S.

PEI

N.L.

Yukon

NWT

Nunavut

Canada

Jan.

Mar.

May

July

Sept.

Nov.

MURAT yükselir / the globe and mail, source: University of Oxford

The data show that every jurisdiction in Canada implemented strong measures in March, as businesses were closed and many people worked from home for most of the spring – though there were variations. Saskatchewan had the least strict measures, even in March and April, according to the data, while the Atlantic provinces were among the most strict.

For most of the summer, Alberta and Saskatchewan had the lowest scores in Canada, averaging 50 and 46, respectively, from July 1 to Nov. 1.

Those two provinces also had significant surges in recent months.

GOVERNMENT RESPONSE TO COVID-19

Using a stringency index developed by the University of Oxford’s Blavatnik School of Government, this chart shows hospitalization rates across Canada have changed in relation to government’s response to COVID-19. Vertical movement indicates change in hospitalization rate, while horizontal movement indicates change in stringency

Two-week averages:

At summer’s end

At mid-fall

At fall’s end

More stringent

Less stringent

By the end of summer, Alberta and Saskatchewan remained the least stringent and saw sharp spikes in hospitalization rates by the end of fall. Manitoba, which had the worst experience in the second wave, appears to be an outlier.

Man.

HOSPITALIZATION RATE PER 100,000

Alta.

Sask.

STRINGENCY INDEX

More stringent

Less stringent

B.C., Ontario and Quebec ended the summer with relatively higher stringency scores and their hospitalization rates managed to remain near or below the national average. Quebec and Ontario have recently seen increases in infections but remain below the worst of the second wave so far in the Prairie provinces.

Man.

HOSPITALIZATION RATE PER 100,000

Que.

Canada

B.C.

Ont.

STRINGENCY INDEX

More stringent

Less stringent

A strong response early on in Atlantic Canada allowed those provinces to ease up over the summer without seeing significant outbreaks.

HOSPITALIZATION RATE PER 100,000

N.B.

N.S.

N.L.

PEI

STRINGENCY INDEX

Note: Territories were excluded because of low population

and case levels.

MURAT yükselir / the globe and mail, source:

University of Oxford; provincial governments

GOVERNMENT RESPONSE TO COVID-19

Two-week averages:

At summer’s end

At mid-fall

At fall’s end

More stringent

Less stringent

Man.

HOSPITALIZATION RATE PER 100,000

Alta.

Sask.

STRINGENCY INDEX

More stringent

Less stringent

Man.

HOSPITALIZATION RATE PER 100,000

Que.

Canada

B.C.

Ont.

STRINGENCY INDEX

More stringent

Less stringent

A strong response early on in Atlantic Canada allowed those provinces to ease up over the summer without seeing significant outbreaks.

HOSPITALIZATION RATE PER 100,000

N.B.

N.S.

N.L.

PEI

STRINGENCY INDEX

Note: Territories were excluded because of low population and case levels.

MURAT yükselir / the globe and mail, source:

University of Oxford; provincial governments

GOVERNMENT RESPONSE TO COVID-19

Two-week averages:

At summer’s end

At mid-fall

At fall’s end

More stringent

Less stringent

Man.

HOSPITALIZATION RATE PER 100,000

Alta.

Sask.

STRINGENCY INDEX

More stringent

Less stringent

Man.

HOSPITALIZATION RATE PER 100,000

Que.

Canada

B.C.

Ont.

STRINGENCY INDEX

More stringent

Less stringent

A strong response early on in Atlantic Canada allowed those provinces to ease up over the summer without seeing significant outbreaks.

HOSPITALIZATION RATE PER 100,000

N.B.

N.S.

N.L.

PEI

STRINGENCY INDEX

Note: Territories were excluded because of low population and case levels.

MURAT yükselir / the globe and mail, source: University of Oxford;

provincial governments

Alberta saw a sudden spike in October and November that gave the province the highest rates of new infections and among the highest for hospitalizations and ICU admissions per capita for much of the fall. Alberta now leads the country in hospital admissions.

In Saskatchewan, the situation was not as severe, but the province saw significant increases in December. It had the second-highest rates of infections for parts of mid-December, behind Alberta and about even with Manitoba, another hot spot. Saskatchewan now has the third-highest rates of hospitalizations – a figure that is increasing.

Both provinces waited until late November and early December to put significant infection-control measures in place. Alberta closed in-person dining, restricted retail capacity, cancelled in-person classes for some students and banned all social gatherings. Saskatchewan’s response was less strict, prohibiting private gatherings with some exceptions outdoors while allowing restaurants, bars and retailers to remain open under capacity restrictions.

Manitoba, which had by far the worst experience in the second wave in terms of infections, hospitalizations and deaths, appears to be an outlier. The Oxford study gives the province among the highest scores for stringency despite the significant toll of COVID-19 in the fall. The Oxford data put the province high on the stringency index even before strict measures, including a ban on all but essential indoor shopping, took effect in November.

A woman wearing a mask walks past the Peter Lougheed Hospital in Calgary on Dec. 3, 2020.Jeff McIntosh/The Canadian Press

Ontario, Quebec and B.C. were in the middle of the pack when it comes to the Oxford stringency scores over the summer.

Ontario implemented regional lockdowns in Toronto and Peel Region in late November, which expanded to the entire province last week. In Ontario, the province has seen record-setting infection rates in recent weeks but has largely kept below the national average and well behind Alberta for new cases, hospitalizations and deaths per capita throughout the fall.

Quebec put in significant restrictions beginning in October, which included strict limits on businesses, public gatherings and social activity after a surge in cases in September shortly after relaxing public-health measures. The province’s infection rates initially slowed after those measures were implemented but have since been increasing rapidly in recent weeks.

Even now, Ontario and Quebec are doing considerably better than Manitoba or Alberta at the worst parts of their second waves.

In B.C., the province has the lowest infection rates outside of Atlantic Canada and roughly the same hospitalization rates as Ontario.

In Atlantic Canada, the four provinces were in a travel bubble for most of the pandemic until Prince Edward Island and Newfoundland tightened their travel rules by pulling out of the bubble in late November. The Atlantic provinces maintained comparatively high stringency scores in late spring as other provinces eased up considerably.

Those provinces gently eased up on restrictions as well, particularly Nova Scotia, but their success at driving down infections in the community early on appeared to set them up to keep COVID-19 rates at such a low level that three of those provinces had only a single person in hospital as of Thursday even in the second wave. There were no hospitalizations in P.E.I.

The territories have taken a fairly strict approach to COVID-19, focusing heavily on travel restrictions and quarantine requirements, and those policies are reflected in the Oxford index. Nunavut, for example, is at about 73 on the index right now – the highest of any jurisdiction – after responding to a significant spike in infections in November, and the Northwest Territories had among the most stringent measures through much of the summer.

Yukon is currently ranked with the lowest stringency in the country, but the territory has maintained quarantine requirements for anyone coming in from elsewhere in Canada. The territory has reported only 60 cases throughout the entire pandemic and a single death. There have been 24 cases in the Northwest Territories. Of the three territories, Nunavut has had the worst experience, with a sudden spike in infections six weeks ago that was followed by a significant escalation in infection-control measures. Nunavut has reported only seven new infections in the past two weeks.

The territories’ small populations have meant that even slight changes in infection or hospitalization numbers produce wild swings in their per capita rates.

Ashleigh Tuite, assistant professor at University of Toronto’s Dalla Lana School of Public Health, said the Oxford data largely line up with what epidemiologists like her have been saying all along: acting quickly with strict measures can slow the spread of the virus.

“I think the data basically bear out what I think most people intuitively understand, which is that to get transmission under control, you need to respond early, and you need to respond strongly,” Dr. Tuite said. “The more quickly you act, the less pain everyone experiences, because you can get this under control.”

Dr. Tuite said it can be difficult to predict the impact of any single policy but that it would be a mistake to focus on outliers as proof that infection-control measures aren’t necessarily effective.

“There’s a little bit of randomness in all of that, so there’s always going to be a bit of a challenge to separate what the trend is from the noise,” she said. “There are always these exceptions, but we look at the general trend – which is that a stronger, faster response results in a better-controlled epidemic.”

In the Oxford data, the picture in Manitoba is the most confounding. The Oxford researchers gave the province some of the highest stringency scores out of any province for most of the year.

Despite that, when the second wave of the pandemic began to pick up steam in September, Manitoba was the first to see significant increases in COVID-19 rates. The province had by far the highest rates in the country in early November and, while new infections have decreased owing to provincewide restrictions on businesses and gatherings, the province continues to have the highest fatality and hospitalization rates.

Winnipeg epidemiologist Cynthia Carr, founder of EPI Research, said models such as the Oxford government response tracker can only measure government policies, not whether they are enforced or if people are following the rules.

“These models are interesting but they don’t necessarily 100 per cent apply to every situation, because human behaviour is an element – we’re the wild cards,” Ms. Carr said. “We certainly saw wildcard behaviour going on here in Manitoba, and it did impact our rates. You can have as many measures as you want – it depends if you’re complying with it.”

She also said the pandemic has affected people in different parts of Canada, or even different areas of the same city, in sometimes vastly different ways. For example, rates of COVID-19 have been found to be worse in lower-income communities.

The relationship between restrictions and infection rates is less stark in the Canadian data than in the United States. The Oxford team released state-level data in early December, which showed dramatic differences between the least-strict states and the places that put in more severe measures.

Some of the highest infection rates in the country – and the world – have been in states that took very few steps to limit activities, restrict businesses or require masks.

States such as South and North Dakota have had stringency scores of less than 40 for much of the year and haven’t pushed passed 50 since the spring. That puts them well behind even the least-stringent places in Canada.

Emily Cameron-Blake, one of the Oxford researchers behind the tracking project, said it’s difficult to look only at the stringency levels without also considering the individual policies. For example, she said, in other jurisdictions limits on travel appear to have a significant impact on the ability to slow infections and keep them contained, which could explain much of the success of the Atlantic provinces.

“The ability to actually seal yourself off, as PEI has done – or put in mandatory quarantine at least – discourages travel between [the areas], so you could probably contain an outbreak better if you’ve got a travel restriction in place,” she said.

Ms. Cameron-Blake said her team has yet to do its own in-depth analysis of the Canadian data, preferring instead to release it and let others come to their own conclusions. She hopes governments look at the data to inform their own decisions as they manage the pandemic.

“We’re trying to take apples and oranges and compare them in a systematic way,” she said.

“We’re hoping that policy makers will use it as a tool – not only to see what’s happening in real time and what isn’t working, but when all is said and done, we can go back and compare all these countries … and make better decisions in the future on how to prepare.”

Sign up for the Coronavirus Update newsletter to read the day’s essential coronavirus news, features and explainers written by Globe reporters.

Your Globe

Follow the author of this article:

Follow topics related to this article: