Where are people getting COVID-19? A global crowdsourced database provides some clues

Yorkregion.com
November 2, 2020

Inside an opulent Vaughan ballroom in early October, about a hundred guests gathered under the chandeliers for a fall wedding. Many had also been there for the first part of the event, a few days before.

They were screened at the door for symptoms and given a temperature check, said Calvin Htut of the Avani Event Centre. Tables were spaced out inside the venue, which had the capacity for over 600.

But someone had COVID-19. Now almost 50 cases, across seven health units in the province, are linked to the two-part event. York Region Public Health took the rare step of issuing a public notice to tell anyone who was in attendance to self-isolate and seek testing, and is reviewing whether charges will be laid. Htut said the venue did everything right and followed all public health guidelines.

Superspreader events like the Vaughan wedding often capture the public’s attention, and scorn, and can easily overwhelm local public health units. There’s a growing consensus that containing and preventing them are key parts of keeping a lid on the virus.

But a lack of reliable local and provincial data leaves the public guessing about just where they are happening.

How often do wedding guests get sick? What about people in bars? Workplaces? Gyms? Many are asking, especially as entire categories of businesses in hot spots are shut down by a province trying to control a second wave. A global crowd-sourced database of these events can provide some clues into where people are getting infected.

Students and professors at the London School of Hygiene & Tropical Medicine and an independent researcher from the Netherlands living in Toronto teamed up this summer on a Google doc, where they track every superspreader event they can find, worldwide. Volunteers help comb through government and media reports, and members of the public can send them events for a rich, real-time record.

The hope, said researcher Koen Swinkels in an email, is to help identify when and where these events tend to happen.

“We can see what kinds of features they might have in common, and that knowledge could help public health officials, researchers, organizations and the public devise better, more targeted strategies of protection against the virus,” he said.

“And it might inform academic research of the virus and the pandemic.”

So far they’ve noticed, for example, that superspreader events are “vastly more common indoors than outdoors.” There’s not one clear example of a purely outdoor event in the database, but some involved both indoor and outdoor settings, and it’s not clear where transmission happened.

The “vast majority” of the event transmission they found “took place in settings where people were essentially confined together for a prolonged period,” such as nursing homes, prisons, cruise ships or worker housing, Swinkels said.

There’s no minimum number of infected people used by researchers to define a superspreader, but most events have 10 or more, he added.

The database contains many reports of some of the more high-profile outbreaks at long-term-care homes in Canada, but not all. Toronto Public Health keeps a complete list of outbreaks in long-term-care homes and hospitals on its website. The province also has an online tracker for school and daycare outbreaks.

The team found that food processing plants, where temperatures are very low, seemed particularly vulnerable compared to other factories, added Swinkels.

“There are some settings where superspreading events seem surprisingly rare,” he said. There are no movie theatres, theme parks or libraries in the database, and only a few flights.

But he cautions that the database captures a fraction of the events around the world and has some serious limitations.

Since many countries have had strict lockdowns, some high-risk spots could be “significantly underrepresented” as they were closed for much of the pandemic. For example, there are no indoor concerts in the database, but that doesn’t mean they aren’t very risky, Swinkels said.

Infection could have occurred before or after an event. The team is also limited to events reported in the media or by health authorities. In Ontario, many outbreaks are not ever reported, unless there is a serious risk to the public like the Vaughan wedding, for privacy reasons or to protect business from stigma.

Ashleigh Tuite, an epidemiologist with the University of Toronto’s Dalla Lana School of Public Health, said despite limitations, such a database is useful.

“We haven’t been that effective in getting that information from local public health, so I think this is the best we can do right now,” she said.

One of her biggest takeaways is the lack of transmission outside.

“It really highlights the fact that whatever the factors are that enhance the likelihood of it being a superspreader event, one of those is clearly being indoors vs. out, and that has implications as well as we move into the winter.”

This makes sense given what scientists know about how the virus spreads, through larger respiratory droplets and smaller ones that, according to a growing body of evidence, can linger in the air and travel past two metres.

For a superspreader event, you need a person who is contagious. But in plenty of examples patient zero doesn’t appear to have shown symptoms, like a recent outbreak at a Hamilton Spin Studio, where owners were reportedly following the rules. There’s a risk anytime you’re sharing air without masks, but it goes up with more people in the same space.

Certain activities, like singing, shouting, or heavy breathing during exercise, can spread more aerosol particles and droplets.

Many public health units warn people to stay away from some version of the three Cs. Crowds, closed spaces--particularly indoor spots with poor ventilation--and close contact.

Htut of the Avani Event Centre said in an email the centre “strongly” maintains it “abided by all public health rules, guidelines and procedures.”

Staff are required to wear masks, and are screened for symptoms. Hand sanitizer is available.

“We also prominently display signage on the front entrance, throughout the premises and even projected on a large screen requesting that guests socially distance, do not crowd or congregate, and wear masks,” he added.

York Region Public Health issued a public alert because of the risk to the public, and because the cluster involves “numerous public health units and regions,” said Patrick Casey, director of corporate communications for the region, in an email.

A number of people attended both wedding events, and “with approximately 100 attendees, this is a high attack rate of nearly 46 per cent,” he said. “Masks were not being used consistently and people were not maintaining physical distancing.”

Such an event is even riskier when people remove masks, he added, and eat and drink together, “especially within 2 metres of each other for 10 minutes or more.”

Provincial regulations allow for up to 30 per cent of the room capacity to gather for indoor religious ceremonies, such as weddings. But social gatherings such as receptions before or after are limited. At the time of the wedding, York Region was in Phase 3, limited to 50 people in banquet halls. It’s back to a modified Phase 2 and the maximum is 10.

York is investigating the venue, and looking into possible charges, said Casey. Htut said there was no reception, only a ceremony.

Weddings are high-risk as they “tend to be very social, happy events” where people can get too close, especially if they’re drinking, said Tuite.

In September, York Region Public Health said at least 23 people had tested positive for COVID-19 after attending a series of wedding events in Markham, Whitchurch-Stoufille and Toronto. There was also an outbreak linked to a wedding in Oshawa later that month.

The snapshot of Canadian superspreader events that the database shows two other weddings, as well as a funeral, linked to multiple cases. It lists two spin studio outbreaks, including the one in Hamilton, four superspreader events at meat packing plants and three at religious gatherings.

This might relate to the kinds of activities that happen at religious ceremonies, such as singing or chanting, and rituals that involve touching or close contact, Tuite said. Just this week, more than 30 cases were tied to a church in Chatham-Kent, Ont.

The database only shows one superspreader event at a bar, a Quebec karaoke bar, which Tuite said is a limitation. Bars are often not named by public health authorities and media reports. Private parties also rarely make the news unless “you have 100 people to your house,” she said.

But one could infect a lot of people even if it’s the same family, or close friends, and the cases from several smaller ones can add up.

Public health officials and politicians often blame COVID’s spread on family gatherings. But a patchwork of publicly available local and provincial data has made it tough to sell decisions on new lockdowns, and even guidance on avoiding family and friends.

Forty-four per cent of Toronto’s outbreaks between Sept. 20 and 26 were related to restaurants, bars and entertainment venues, Toronto Public Health said last week.

“Bars and restaurants have large volumes of contacts to trace, with some of these venues having more than 500 contacts to notify, and with one having 1,700 patrons to reach,” said spokesperson Dr. Vinita Dubey.

Twenty-one cases were linked to adult recreational hockey in Toronto in early October, and in late September, 18 cases were linked to a fitness centre, with 76 contacts.

The city’s overwhelmed public health department has had to shift resources away from contact tracing in the community, toward outbreaks in high-risk institutional settings such as hospitals, long-term-care homes and schools.

The province released pie charts breaking down outbreaks by broad categories this past week, by hotspot health unit, showing three per cent traced to gyms and 14 per cent to bars, restaurants and clubs in Toronto. But many more cases can’t be traced to any source, about 65 per cent in Toronto.

Bars, restaurants and gyms (excluding dance studios) are closed in Toronto, Peel, York and Ottawa, but Toronto Public Health staff are no longer able to trace contacts that might be related to a private party or wedding.

That’s why it’s useful to look at larger patterns, like the ones in the database, Tuite said.

“It’s important information, I think right now, in particular when we’re having a lot of conversations and debates about why we’re closing certain businesses, and why we’re preventing certain activities,” she said.

“These patterns are not unique to where we are locally. They’re being seen over and over again around the world.”