If germs hung a recruiting sign for their hosts, it would probably be a version of the World War I poster of Uncle Sam pointing: We want YOU to help us reproduce. All hosts were equally eligible for service, infectious-disease researchers thought. Assuming the recruits weren’t immune due to a prior infection or vaccination, anyone should have roughly the same potential to spread a disease’s pathogens. But then came severe acute respiratory syndrome, or SARS.
This pandemic started as just another strange pneumonia from southern China, but in 2003 it turned into a global outbreak that infected 8,098 people and killed 774. Key to the disease’s spread, researchers found, was a small but crucial portion of the population that became known as “superspreaders,” people who transmitted the infection to a much greater than expected number of new hosts. The more scientists learn about superspreaders, the more they are beginning to realize that this tiny segment of the population is the driving force behind the emergence and spread of infectious diseases.
Perhaps the most infamous superspreader in history was Mary Mallon, aka Typhoid Mary. An Irish cook in New York City in the 1900s, she was chronically infected with Salmonella typhi. Although the infection didn’t cause any symptoms for Mallon, she did excrete large numbers of typhoid bacteria in her feces. Her career as a cook made it easy to transmit fecal bacteria to customers through the food she prepared. She infected about 50 people and killed several (official counts vary) before she was arrested and jailed for refusing to give up her career as a cook.
To epidemiologists, people like Typhoid Mary were seen as an anomaly and not the main drivers of infectious disease spread. Instead, epidemiologists focused on a number they called R0, the average number of people a single person could infect. The value of R0 depends on three main factors: the number of susceptible individuals in the population, the number currently infected, and those resistant to infection. If the value of R0 is less than one, it means that each individual infects less than one other person, and the outbreak will ultimately die out. However, if the value of R0 is greater than one, the disease has the potential to spread.
When you look across a large population, R0 is a good estimate of whether and how a particular infection might spread. At an individual level, it turns out, R0 is less accurate. Most people won’t spread the disease at all, but a very few people will spread the infection to tens or even hundreds of others.
As computer models of infectious disease grew more powerful and precise, scientists began to realize that a lot of infectious disease spread is due to superspreaders. Researchers call it the 80/20 rule: Twenty percent of the population is responsible for 80 percent of disease spread. The key to understanding and stopping outbreaks of infectious disease means homing in on this small portion of the population that drives the majority of transmission.
Accounting for the presence of superspreaders also means accounting for their much more common counterparts: people who don’t spread the disease to anyone. Their R0 is zero. Taken together, that means a disease is much more likely to be introduced into new areas by superspreaders, but it is also more likely to fizzle out in those places due to nonspreaders. Only occasionally, as with SARS, can the pathogen be introduced to somewhere new and then, like a spark on dry kindling, ignite a massive blaze of infection.
On Feb. 21, 2003, a 64-year-old doctor checked into Room 911 at the Hotel Metropole in Hong Kong. He was already sick with SARS but felt well enough to travel. Public health officials documented a total of seven people directly infected by the doctor, making him a superspreader. Several of the people he infected were also superspreaders. One young man infected every doctor, nurse, and medical student who examined him in a Hong Kong hospital—more than 100 health care workers in total. A flight attendant from Singapore went on to infect at least 160 others. These travelers then brought the SARS virus home with them as a very unwelcome souvenir. One of these infected individuals arrived in Toronto. Several days after she returned home, she developed symptoms consistent with SARS and ultimately died. Nine other SARS cases and three deaths were directly linked to her, and the outbreak ultimately spread to 257 others. This chain of contagion accounted for more than 70 percent of Toronto’s total SARS cases. Although Toronto was besieged by SARS in 2003, Vancouver (with a similar population) saw only a handful of cases. The reason? No superspreaders were infected in Vancouver, said infectious disease expert James Lloyd-Smith of the University of California, Los Angeles. “SARS made the superspreader phenomenon so obvious we could no longer ignore it,” he said. But “SARS is just one point along a continuum—all infectious show this [pattern] to some degree.”
Superspreaders are not limited to humans. When researchers began to deconstruct the outbreaks of West Nile virus that hit the United States in the first decade of the 21st century, they found that certain areas, like the western suburbs of Chicago, were hit especially hard. Tony Goldberg, an epidemiologist at the University of Wisconsin-Madison, began to study the mosquitoes that transmitted West Nile virus in an attempt to figure out why some places had so many more cases than others.
Using traps and a device that operates like a giant vacuum cleaner, Goldberg and colleagues collected mosquitoes from around Chicago. A few of these mosquitoes had blood in their abdomen from a recent meal, which allowed researchers to identify which animals the mosquitoes had been biting. Overwhelmingly, the mosquitoes had been feeding on Turdus migratorius, the American robin.
West Nile virus is traditionally associated with crows, blue jays, and other corvids, Goldberg said. These are the birds that get sick and die in large numbers during West Nile outbreaks. Public health officials knew to brace themselves for a West Nile outbreak in humans this summer when they began to see large numbers of dead crows. Although researchers knew that other birds could become infected, they were thought to be fairly minor contributors to West Nile spread. The fact that robins don’t die or become noticeably ill made them both invisible to public health authorities and remarkably good spreaders of West Nile. When crows and related birds die, they are unable to transmit the virus to other birds. They become literal dead ends. Robins, on the other hand, continue the everyday activities that bring them in contact with new mosquitoes, which drink the virus from the birds’ blood and continue the spread of the disease.
Whether they are individuals or an entire species, superspreaders have certain traits that make them efficient engines of infection. According to disease ecologist Sara Paull at the University of Colorado, Boulder, superspreaders share three major qualities. They shed large quantities of the pathogen. They transmit it to a large number of people. And they do so for a long period of time. A combination of an individual’s physiology and behavior determines whether he or she will become a superspreader.
Take Typhoid Mary. If she hadn’t been a cook and hadn’t harbored an asymptomatic infection of Salmonella typhi in her gall bladder, she wouldn’t have infected many other people. In the case of SARS, many local outbreaks centered on hospitals. A health care worker who hacked up large amounts of virus had a much greater potential of spreading SARS to a wide range of people, many of whom had impaired immune systems, compared to someone who rarely had contact with the outside world. Scientists still don’t know why SARS superspreaders were so effective at transmitting the disease, although a major factor appears to be invasive respiratory procedures like intubation and nebulizer treatments that aerosolized large amounts of virus.
Identifying superspreaders seems like it would be a logical place for public health officials to start fighting disease outbreaks. “If you just apply infectious disease control measures randomly, you may not be able to actually eliminate an infectious disease. But if you target the most relevant 20 percent, then you can vaccinate or treat enough of these individuals to eliminate the disease much more quickly,” said Virginia Tech disease ecologist Dana Hawley.
In order to figure out who these superspreaders really are, scientists need to define superspreading in a concrete way that can be applied to multiple different diseases. Thus far, noted Lloyd-Smith, this hasn’t been done. “This term is being bandied about, and everyone is using their own personal definition,” Lloyd-Smith said. The basic definition Lloyd-Smith developed was a person who spreads an infection to a significantly higher number of contacts than average. But no one knows exactly how much higher that number needs to be before someone crosses the line from unlucky to superspreader. Nor do researchers understand why some diseases are more prone to superspreading than others.
In the meantime, scientists have begun to look at superspreaders’ mirror image—those people who don’t transmit infections to anyone else—as a way to halt disease outbreaks. Are their immune systems fundamentally different? What about their behavior? Perhaps through these individuals, researchers will crack the mystery of superspreading and ultimately stop local infections from becoming pandemics.