Back in January, before it was officially declared a pandemic — when roughly 500 people were believed to be infected and fewer than 20 had died — one patient in Wuhan, China, reportedly spread the disease to 14 health care workers. More recently, in March, a single member of a choir in Washington state was found to have infected 52 others during a 2 1/2-hour practice session. And in early May, after South Korea began to relax social-distancing restrictions, dozens of new infections emerged around Seoul; the cluster was linked to a 29-year-old man who tested positive after visiting five nightclubs and bars in one night.
These people have been called “super spreaders,” but are they really different from the rest of us?
“It’s not a great term to be labeled with on individual level,” said Dr. Jared Baeten, a professor of public health and epidemiology, and the vice dean of the School of Public Health at the University of Washington. “But I think it helps people think about how a virus like this can make its way quickly through communities.”
It may be that a so-called super spreader is the right person in the wrong place at the wrong time.
“Some individuals are either particularly infectious or that the [amount of] time or number of people that they spend time with is enough to make their effect on spreading the virus to be disproportionately greater than the average,” Baeten said.
Measuring how viruses spread
First, a little science. Infectious disease specialists and epidemiologists, have a number to describe how infectious a disease is. It’s called the reproduction number or R-naught, written R0. It’s basically a measurement of the average number of people an infected person goes on to infect.
“How infectious a disease is, you grade it by this R-naught factor. The R-naught factor is somewhere between 2 and 3 for coronavirus. For measles it’s much higher — it’s between 16 and 32,” explains Dr. Rob Murphy, a professor of infectious diseases and of biomedical engineering, and the executive director of the Institute for Global Health, at Northwestern University Feinberg School of Medicine.
A disease’s reproduction number can change as time goes on; it can also be different in different places. The goal is to get it as low as possible.
“All you have to do to get rid of an epidemic is to have the R-naught factor be less than one. As soon as it’s less than one — each person gives it to less than one other person — it dies off,” Murphy said. “The problem with the R-naught factor is, it’s very nice to describe the average. However, in medicine, as in many things, if you find an average person let me know; I want to go meet them.”
In other words, experts think that most people infected with the virus are so-called “dead ends,” infecting, on average, fewer than one, and a small proportion of people infected with the coronavirus that causes Covid-19 are infecting many more people than average.
What makes super spreading possible?
One reason someone might infect more people than others comes down to individual bodies.
“Everybody is different, they’re handling the virus in their bodies differently,” said Murphy. “So certain individuals, they have what we call a high viral load — that’s just the amount of virus per droplet. Some people have 10 or 100 times more virus in their droplets than other people.”
Baeten agrees. “There are things we don’t fully understand about this virus, but usually with viral infections, there’s a range of viral loads and some people — either because of the way the virus interacts with their immune system or their genetics — have a higher viral load than others. And higher viral load probably means that they’re more infectious to people they come around.” He adds that the duration of the contact is also very important.
Both Murphy and Baeten say behavioral aspects can also play a part.
Does someone speak loudly? People who speak loudly (or sing or huff-and-puff during exercise) are thought to release more virus-laden droplets into the air because of what they are doing or how they are doing it.
Is that person an extrovert who enjoys the company of others or the huggy-feely sort who gets physically close to others? By now, we’ve all been schooled on the benefits of isolating and social distancing to keep infection rates down — but even during a pandemic, that doesn’t always happen.
Another factor that might lead to super spreading is timing. Studies have found that people start shedding virus particles about two days before they start showing symptoms, when they’re pre-symptomatic; others never experience symptoms of Covid-19 at all — asymptomatic — but are still shedding virus. People in both these groups circulate throughout the day unknowingly infecting others.
“The unfortunate thing for transmitting disease is that right when the infection takes off, just before a person becomes symptomatic, they probably have the highest level [of virus],” said Murphy. “So, the day they start having symptoms, and the day before and maybe the day before that is when they have the highest viral load.”
Location, location, location
Also important is the opportunity to spread. “Just because someone has a high viral load doesn’t make them a super spreader if they’re behind a mask [or] at home,” Baeten said. “Because the super spreading happens when someone who is able to transmit the virus is in contact with enough people that they end up spreading the virus to more people than the average.”
That’s where locations — like a crowded night club, a church service or a meatpacking plant — come into play. “Those are all the kind of events that can be a setup for having a large spread,” says Baeten. “Introduce someone who’s infectious — and maybe not even super infectious — into a crowd of people in close quarters for extended duration of time — like a funeral, like a sporting event, like a nursing home –and that can amplify the spread.”
Epidemiologists have a way of capturing that type of disease clustering across locations — it’s called the dispersion (or “K”) factor.
“In a way it’s a measure to give you a sense of, is this a disease that is randomly occurring over time and place, or is it one that is clustered in time and place,” explains Michael Osterholm director of the Center for Infectious Disease Research and Policy at the University of Minnesota. “If you have 100 cases, they are going to be randomly distributed much like every foot apart … Or you can have all the same 100 cases in the same space but bunched together, so one cluster here [of] 30 people very closely together, another over there. So that’s an attempt to measure is how randomly [or clustered] they occur in that space.”
Another way of thinking about it is, the dispersion factor measures how many cases of infection are responsible for all the transmissions. The lower the number, the more the disease “clusters” around super spreading events or locations.
It turns out that the coronavirus clusters a lot, according to several studies. One not-yet-published study out of the University of Minnesota conducted by Osterholm’s colleagues found that 20% of cases in Hong Kong, all involving social gatherings, were responsible for 80% of transmissions. (That translates to a dispersion factor of about 0.45, for anyone keeping track.)
Gwenan Knight and colleagues at the London School of Hygiene & Tropical Medicine have been studying settings linked to the transmission of the coronavirus. The results of a study by her colleagues, also still a preprint that has not yet been peer-reviewed, point to an even larger clustering effect.
“We know that transmission of many infectious diseases is heterogeneous — different people transmit more, and different settings are conducive to transmission than others (e.g. indoor more than outdoor). This seems especially so for this new coronavirus with the dispersion factor analysis suggesting that 10% of cases cause 80% of infections,” she wrote in an email.
As for where these clusters have occurred, “In our database of settings linked to transmission, we found the usual suspects (healthcare and cruise ships, etc.) but some odd things, such as several meat processing plants, and fewer schools than we’d expect.” She adds to the list “large, shared accommodations” like worker dormitories and refugee camps.
“We have not found protests to yet be associated but we are really looking at settings rather than events,” Knight says. “I would imagine that protests might not be so risky as they are outdoor but then the individuals may be involved in shouting (which may be risky for transmission) and they may be in close contact for long periods of time.”
Putting the pieces together
Knowing that coronavirus transmissions often occur in clusters linked to particular settings and through certain individuals can help inform how governments, public health experts and society try to control it.
The experts all agree that frequent hand washing, masks and social distancing are key to keeping down the rate of infection. Studies back that up: A meta-analysis of more than 150 studies, recently published in The Lancet, found that the chance of transmission when people are less than about 3.3 feet apart is almost 13%, but that drops to 2.6% when they are more than 3.3 feet apart, and double that is even better. And the chance of transmission without a mask is greater than 17%, compared with 3% when a mask is worn.
Murphy adds that “massive testing — five times more than what we have” is also needed to confirm who is sick with Covid-19 as well as to identify who may be infected and walking around without symptoms. “You have to identify them,” he said. “So, basically, test everyone, and then quarantine the people who are infectious. That would do it alone.”
But he points out that the barriers to getting tested (such as cost and accessibility) need to be eliminated or “people will just drop out of the system.” He foresees a time not too far in the future where people can be tested and receive their result on the spot, or even test themselves at home.
Moving the focus from the action of individuals to locations, Knight says, “We may need to tailor exit strategies from lockdown to target some places more than others — e.g. riskier settings should be kept closed or visits to these settings should be monitored in contact tracing apps. By understanding where and from who transmission occurs, we can intervene to stop it.”
Osterholm adds contact tracing to the list of control measures. The Center for Infectious Disease Research and Policy, which he heads, recently put out a paper on the topic, outlining the right and wrong ways to go about contact tracing without alienating people and squandering resources.
He argues for more targeted approach.
“Understanding that the whole thesis of contact tracing may just fall on this very issue [of super spreaders],” he says. “If you have a small number of cases, you can clearly try to contact trace all of them, and you’re not wondering if they’re super spreaders or not. With many thousands of cases, you’re always going to be trying to determine: what are the highest priority? If you don’t have a sense of who the super spreaders are, there’s no way you can [factor] that in.” But, therein lies the rub, he says: “Contact tracing won’t identify these people themselves until long after they are sick.” And by then, the damage will have been done.
No one knows who is super spreading — even the person spreading the virus — until it is too late
“If we could identify super spreaders early that would help but at this point we don’t have any reasonable way to do that,” said Osterholm. “There’s no big double S on their forehead that says ‘Super Spreader’ so that we can go find them. And that’s the challenge.”
Murphy has a slightly different take. “If they’re super spreaders or just regular spreaders, it doesn’t really make any difference — they are spreading,” he said.
Ultimately, said Baeten, it comes down to all of us, because we are in this together.
“I think it’s incumbent on all of us to do our best to not be a super spreader by staying away from large crowds and large attractions where we might be able to spread the virus, and by wearing a mask.”