Editor’s note: The Salt Lake Tribune is providing free access to critical stories about the coronavirus. Sign up for our Top Stories newsletter, sent to your inbox every weekday morning. To support journalism like this, please donate or become a subscriber.
Given the popularity of my last article on how the coronavirus can spread in everyday scenarios, I thought I’d add to the list of places we go, including some summertime favorites like the outdoors and weddings.
Remember how one London School of Hygiene analysis suggested that 80% of coronavirus transmissions were caused by 10% of people? That means superspreading events are pretty important to understand for those trying to avoid the virus.
Given that, here are more places we either have or have not discovered superspreading events, and what we can learn from each example.
I wrote about the Wenzhou, China shopping mall outbreak in last week’s article. In this case, 34 people were infected in a shopping mall, 18 had offices on the seventh floor or interacted with those workers. The remaining 16 worked or shopped on the third floor or below.
Here’s a fact buried in the U.S. Centers for Disease Control and Prevention (CDC) report that I missed the first time I read it: all 16 of the people who caught the disease on the lower floors were women. Now, women are more likely to go to malls than men, so it’s possible but unlikely that this happened by chance. There are stores that orient their business towards women, but the employees who caught the virus worked on different floors and in different stores.
The more likely source of the transmission was a female bathroom. One of the patients who caught the virus was a restroom cleaner, further sending our suspicions in that direction.
That being said, that’s the only documented superspreading example tied to the possibility of significant bathroom spread. In general, people don’t spend very long in bathrooms, so they tend to get smaller doses of whatever is in there, which probably makes it less likely that they’ll develop the virus. They also don’t frequently have face-to-face conversations, lowering the possibility of droplet transmission.
There are a lot of dirty surfaces in bathrooms, and of course, there is poop. But the CDC now considers surface transmission to be relatively rare, and we still have no documented cases of fecal-oral transmission of COVID-19. Of course, thoroughly washing your hands will prevent nearly all surface transmission cases anyway.
• A bathroom superspreading event appears to be relatively unlikely.
Well, we certainly know a lot about COVID-19 transmission on cruise ships.
Because they were early hotspots, and a relatively natural petri dish, cruise ships have been studied perhaps more than any other environment for the disease. The problem is so widespread that Wikipedia has its own page dedicated to cruise ships and coronavirus, which shows that over 2,000 people have died across 40 different boats.
You likely have heard of the Diamond Princess and the Grand Princess, and the CDC outlined what happened on those ships. But I want to tell you about an even more dire situation: that of a 21-day trek to Antarctica that was designed to follow the route of 20th century explorer Ernest Shackleton.
The writing is significantly above-par when compared to the average scientific paper, so check it out if you have time. It reports this: 223 passengers and crew set sail from Argentina headed for Antarctica in mid-March. Before boarding, all were given temperature and symptom checks. No passengers were allowed who had been in China, Macau, Hong Kong, Taiwan, Japan, South Korea or Iran over the previous three weeks. Hand sanitizer stations were placed throughout the ship.
The crew continued to check on possible symptoms for everyone regularly, and early in the cruise, all were fine. On Day 8, though, a passenger had a fever. All guests were confined to their cabins. The crew delivered meals three times a day wearing PPE, including N95 masks. The cruise ship turned around, and started to call Southern Hemisphere countries to ask where they could dock and get everyone home. Uruguay looked most promising.
On Day 14, VivaDiag rapid tests, which purports to give results within 15 minutes, were delivered to the boat and administered to six people who now had fevers. All came up negative. Still, all passengers were instructed to stay in their cabins. Despite this, Uruguay insisted that all aboard take standard virus swab tests before accepting the passengers.
This was a good call by Uruguay. Swab tests were delivered and administered on Day 20 — 128 people tested positive for the coronavirus, or 59%. That includes all six people who tested negative on the VivaDiag rapid test. Interestingly, of the 128 positive cases, only 24 had any symptoms. Of those 24, eight were medically evacuated.
Eventually, Uruguay allowed the ship to dock and passengers to disembark on Day 32.
• Seriously, don’t go on cruises right now. As the CDC puts it, “All persons should defer all cruise travel worldwide during the COVID-19 pandemic.”
• Temperature and symptom checks are not enough to ensure the virus doesn’t make it into a large-population closed environment.
Carnivals and festivals
On Feb. 15, Gangelt, Germany (a town of about 12,000) hosted a celebration of Carnival in an indoor hall and about 300 people attended. A 47-year-old man danced on stage as part of the festivities, and he would test positive 10 days later.
Later, German officials would antibody and swab test 919 people who lived in the area. Of those who went to Carnival, 21.3% were infected. Of those who didn’t, 9.5% were.
But that’s not all: they also counted symptoms among the positive cases. People who went to the celebration had over four coronavirus symptoms on average, while those who got it elsewhere only averaged 2.5. That was true even while adjusting for age.
That led the researchers to hypothesize that those who went to Carnival were not only more likely to get the virus, they also got higher viral loads than those who caught the virus in other ways. Now, we do know that higher initial viral loads lead to higher symptom severity in influenza, but we don’t know that happens in this coronavirus, it’s just a hypothesis. But it is a good one.
The CDC also reported that Mardi Gras quickened the spread of coronavirus in New Orleans, one of our nation’s hardest-hit areas.
• Avoid festivals and carnivals with hundreds of people in a small space.
• Those who get high viral load may be more likely to have more symptoms.
[Read Andy Larsen’s first look at how the virus spreads in everyday places. That column explores bars, buffets, buses, choirs, church, family gatherings, grocery stores, gyms, malls, offices, planes, polling places, schools and sporting venues]
Speaking of getting hundreds of people together, conferences have also been significant sites of superspreader events.
A company conference in Singapore led to 20 people being infected. This makes sense. At the conference, there was a buffet lunch, a Chinese banquet dinner, four-hour breakout sessions, and 2.5-hour team building activities, which did include some physical touching. Then, one infected person went back and celebrated the Chinese new year in Malaysia, infecting two. Another then went to France, Spain, and the U.K., infecting people in all three countries.
At a Biogen conference in Boston, at least 89 people caught the disease.
In a smart move, a conference on coronaviruses and similar pathogens was cancelled due to the coronavirus threat.
• Conferences pack hundreds of people in close quarters, and should be avoided.
• Conferences usually involve long-distance travelers, who are more likely to bring the virus in and more likely to spread it wide once they leave.
There just aren’t superspreading events documented outdoors.
In a systematic review of 1,245 cases in China, only one outdoor transmission was found: a conversation that also may have included handshakes. I was able to only find one other clear-cut example of outdoor transmission: a marathon runner who gave it to a friend he went running with.
There are a few cases where transmission could have happened outdoors, but it isn’t clear. For example, four of Singapore’s clusters happened at construction sites, but it’s not clear if the workers were inside or outside of the building.
Being outdoors has a lot going for it: UV light kills the virus quickly, and wind currents allow the droplets to disperse or evaporate. It is also easier to stay distanced.
And even bringing in just a little bit of outdoors has the potential to help in a big way. One study looked at droplet spread in three rooms: one with no ventilation, one with mechanical ventilation only, and one in which a small window was opened. The results were clear: in the stale room, the droplets linger, but in the room with good ventilation, they go away very quickly.
One other note: another study found reduced droplet spread in high-humidity conditions. The droplets from humans that potentially contain the virus quickly bond to droplets in the air and fall to the ground.
• Outdoors is largely safe, barring close face-to-face conversations or heavy exercise over time.
• Opening a window and improving ventilation has clear impacts on droplet duration.
In January, a family from Wuhan traveled to Guangzhou, China. The next day, they had lunch at a restaurant. All were asymptomatic at the time, but later that same day, one person started feeling feverish. Remember, there’s a 2-14 day delay between when someone is infected and when someone shows symptoms, so this person was infected in Wuhan.
At the lunch, nine others contracted the virus: four people sitting at the same table, and five people at other tables. The person who contracted the disease first is labeled A1 below, with red circles indicating the other people who got COVID-19. Families A and B were seated at adjacent tables for 53 minutes, A and C for 73 minutes.
I should note that a few of the people who contracted the disease were sitting more than 6 feet away, leading researchers to hypothesize that the virus-carrying droplets were spread further than usual thanks to air conditioning currents in the room — a poor-circulation situation. The only people who contracted COVID-19 were those at tables in the circulating path of the air-conditioner in an otherwise stuffy dining area.
• Six feet of distance doesn’t totally prevent the spread of the coronavirus. On the other hand, one infected patron probably won’t affect tables on the other side of the restaurant.
• When possible, restaurants should be cognizant of air traffic patterns when deciding where to place tables — open air dispersion is probably best.
Ski resort towns were immediately hard-hit by the coronavirus in February and March.
Here in Utah, Park City and Summit County’s early troubles are well-documented. U.S. counties with major ski resorts held four of the top 10 spots for coronavirus cases per capita at the end of march. Colorado’s Gunnison and Eagle County, as well as Idaho’s Blaine County, joined Summit County in that unfortunate list.
It’s not just in the U.S. In Austria, the town of Ischgl is in real hot water after resisting shutdowns for nine days after initial warnings. Like Park City, the first case of community spread was found in a bar employee.
And that makes sense: after the skiing is done, there’s a lot of time spent in cozy indoor areas. One of the earliest clusters was discovered in a chalet in the French Alps where 16 people stayed for four days and eventually 12 of those people tested positive.
• Places where people gather after hitting the slopes, like bars and chalets, appear to be particularly vulnerable to wide-scale spread.
In mid-March, the nation of Jordan had only one confirmed coronavirus case. That wouldn’t last.
That’s because on March 13, a two-hour indoor wedding was held in the northern part of the country. About 360 people came, including the bride’s father, who traveled from Spain.
He’s the suspected patient zero, and he developed a fever, cough, and runny nose two days before the big day. But he wasn’t going to miss his daughter’s wedding. That would also mean he’d stand in the reception line and greet those who came, including with hugs, cheek kisses, and handshakes. Afterwards, people danced.
Of those 360 attendees, 350 were tested. It turns out that 76 of them had the coronavirus, and nine more close contacts of those who attended tested positive in the immediate aftermath. Of the 76 positive attendees, 40 were symptomatic when tested. They developed symptoms an average of five days after the wedding, but with a range of anywhere between two days and 13 days later.
• Indoor weddings can be significant sources of coronavirus spread.
• If you’re sick, don’t put yourself in situations where you’re going to be in close contact with hundreds of people.
Andy Larsen is a Tribune sports reporter who covers the Utah Jazz. During this crisis, he has been assigned to dig into the numbers surrounding the coronavirus. You can reach Andy at email@example.com or on Twitter at @andyblarsen.