Where are COVID-19 people taking? Community transmission is still a conundrum in Ontario

Nine months after the start of the COVID-19 pandemic, health officials still face a puzzle as troublesome as it was at the start of the outbreak: defining how cases are spread in the community.

In Ontario, up to 60% of cases have unknown sources of transmission.

The lack of data has, in part, led to our current lockdown, as government advisers look to the US for information on how far it needs to go to shut down parts of the economy.

“The restaurants did everything that was asked for and more. Officials make decisions based on theories and assumptions. But the financial and reputational damages of this sector are not theoretical. They’re all too real, ”said James Rilett, spokesperson for Restaurants Canada, a non-profit that supports the restaurant industry.

Frustration over the province’s decision to shut down Toronto and the Peel region spurred backlash from business owners this week, with some openings clashing with sanitation measures and others calling for the closure of department stores deemed essential services.

One of the reasons for the lack of data on where community spread is occurring is that it is extremely difficult to trace the contacts of strangers in crowded places like restaurants, bars, and gyms. The growing number of cases in recent weeks has made it nearly impossible to find contacts in some areas, such as Toronto and Peel. (In early October, Toronto Public Health announced it was suspending contact tracing outside of high-risk situations and institutional outbreaks due to the growing number of cases in the wider community.)

But researchers from the COVID-19 Science Advisory Table, a group of scientists and health policy experts who analyze trends and advise the provincial government, say there is enough evidence from U.S. studies and local epidemic data to make plausible assumptions about where the virus is most likely to spread.

“It is very likely that a lot of what is not detected in statistics and contact tracking, even if people try hard, happens in those internal spaces where people meet the most,” said Dr. Peter Juni, scientific director of the scientific advisory board and a professor of medicine and epidemiology at the University of Toronto.

“Indoor meetings where people are relatively close to each other – it can be you and your friends; it doesn’t have to be other people you don’t know and where people stay for an extended period of time, which is more than 15 minutes, this is actually what is happening with indoor meals. “

In a report earlier this month, the advisory table established a link between the frequency of visits to restaurants with indoor dining and the possibility of spreading the virus if those establishments reopen.

The report also shows that while less than 10% of COVID-19 cases in Ontario are linked to outbreaks, restaurants, bars, ceremonies and religious services make up a total of nine% of these outbreaks. The contexts with the highest number of epidemics are nursing homes and long-term rest homes (26.5%), schools and nurseries (22.2%).

“It does not mean that the proportion shown in that data for restaurants covers everyone who has been infected in a restaurant; it only covers the ones we know, “said Beate Sander, co-chair of the COVID-19 Modeling Consensus Table of Ontario.” This large percentage of community transmission occurs in schools, restaurants and gyms and workplaces. .. everywhere, in practice. “

A snapshot of data from Toronto Public Health shows that 45 active outbreaks occurred in the workplace during a week in September, 18 of them in restaurants and bars, one in a nightclub and another in an adult entertainment club. .

“Together, these outbreaks accounted for 44% of our community outbreak investigations alone,” Toronto Associate Health Officer Dr. Vinita Dubey said in an email, adding that the investigations “have been incredibly high. resource intensity “.

“Bars and restaurants have large volumes of contacts to track,” he said, “with some of these places having more than 500 contacts to notify and one with 1,700 customers to reach.”

The province closed restaurants, bars, gyms and indoor theaters in Toronto, Peel and Ottawa on October 10. The gyms were cleared to reopen earlier this month before being closed again in Toronto and Peel under the new block, which also closed outdoor dining, hair salons and non-essential retail.

Dubey said the strategy behind the current blockade is to limit people’s ability to spend time together in high-risk environments where the possibility of infection has increased.

“This is a difficult but necessary choice right now,” he said.

Asked if there was evidence to close restaurant patios, as Toronto and Peel did this week, the Ministry of Health said in an email that the lockdown was necessary to limit community transmission “in order to maintain open schools, safeguard the capacity of the health system and protect the most vulnerable populations of the province “.

“In collaboration with the Chief Health Medical Officer and our local health officers, we continue to closely monitor the evolving situation to advise on whether and when public health measures need to be adequate,” a spokesman for the ministry said.

A study by the United States Centers for Disease Control and Prevention in September found that adults who confirmed they had COVID-19 were about twice as likely to have reported having dinner at a restaurant in the two weeks before getting sick than they were. those that tested negative.

Another US study, which appeared this month in the British scientific journal Nature and which the Ontario Scientific Advisory Table drew on, suggests a link between where people go and how they are infected.

The study used anonymous cellular data to track the movement of 98 million people in large American cities from neighborhoods to hundreds of thousands of what they called “points of interest,” which included restaurants, grocery stores and churches, from March to May.

The researchers ran a simulation to predict, based on visits, where residents were infected and found that a small fraction of points of interest accounted for a large percentage of infections. The simulation results produced infection rates close to what was happening in cities at the time.

Those spots were typically places that “tend to have denser visitors and visitors tend to stay longer,” like restaurants, said Emma Pierson, who co-authored the study while working on her Ph.D. at Stanford University. Pierson holds a Masters in Statistics from Oxford University, where she was a Rhodes Scholar, and is now a Senior Research Fellow at Microsoft.

Density was also key as to why the model predicted higher infection rates in less white neighborhoods, which is consistent with what really happened during the first wave, Pierson said.

The model showed that the people of these neighborhoods were unable to reduce their mobility that much, partly because they were more likely to be essential workers.

“It’s hard for them to work from home,” Pierson said.

The places they went to typically had more visitors and were “therefore more dangerous,” Pierson said, noting that in the United States there is a correlation between higher population density and lower socio-economic areas.

Although the Nature study did not assess whether individual restaurants were riskier than other points of interest, a simulation to reopen each sector by category while keeping the others closed showed that reopening restaurants would lead to hundreds of thousands of infections. in the Chicago area.

The reopening of full-service restaurants was particularly risky, according to the study, with researchers predicting about 600,000 additional infections in the Chicago area by the end of May, more than triple the next category of establishment.

Pierson said part of the reason may simply be the number of restaurants.

Across all metropolitan areas used in the study, full-service restaurants, along with gyms, hotels, cafes, religious organizations, and limited-service restaurants, produced the largest expected increase in infections when they reopened.

Nature study statistics used by the Ontario Scientific Advisory Table show that additional COVID-19 infections associated with reopening of locations (after a lockdown, for example) are much higher for the poorest 10% of income earners than for 10 percent.

The risk of becoming infected in restaurants, places of worship and grocery stores was disproportionately high for the lowest 10% of workers.

“These are the essential workers. It’s not the people who buy their organic bananas. It’s the people who work there who get infected, “Juni said.” This is like the perfect storm for transmission. They’re exposed because they’re essential workers. “

Further research in the United States could lead to understanding how retail and restaurants can reopen more safely in the event of a pandemic.

The study in Nature found that if points of interest in Chicago limited maximum occupancy to 20 percent, it would cut new infections by more than 80 percent, reducing overall visits by just 40 percent.

“If you reduce the maximum occupancy, it can potentially allow you to reopen more efficiently because it creates a large reduction in infections for a relatively small reduction in visits,” Pierson said. “The reason is that it only hits places when they are busiest and most dangerous.”

The authors also believe that the study results show that the reopening has a disparate impact on at-risk neighborhoods, a risk that could be mitigated through strict occupancy limits, food distribution in areas with high-risk shops, more testing and leave or income. improved paid supports for essential workers as well as personal protective equipment.

Toronto has had a higher incidence of the disease in the northeastern and northwestern parts of the city.

Juni said he hopes the scientific advisory board will find ways to make restaurants and other places safer.

“Basically, try to wear a mask as often as possible, try to ventilate as best you can in these places,” he said. “Keep the tables separate from each other so that at least the groups of people don’t interact with each other. They will all be useful in the future.”

“The problem is that only if you limit relatively hard (you will) will you have this thing under control because it’s so contagious.”