When the first case of Ebola was diagnosed in the United States this week, Northeastern University researcher Alessandro Vespignani was not surprised. For weeks, the data scientist who designs and runs big computer simulations of infectious disease outbreaks has been carefully tweaking a model to project the spread of the virus.
The current risk of an infected person arriving in the US is around 10 percent, a chance that is projected to grow over the next month to about a 25 percent probability by the third week of October.
“This is like getting lottery tickets, so when the probability is very small you don’t expect to win the lottery,” Vespignani said. But at a level of 10 percent risk, it becomes only a matter of time before an infected person shows up, “and this probability is continuously growing in time.”
To make predictions, Vespignani and colleagues create a synthetic version of the world within a computer, drawing on data that includes international air traffic patterns and models of how thousands of groups of people in 220 countries interact. The researchers run this model thousands of times and then count up how many times the disease spreads to a particular country to calculate the odds virus will arrive, and the number of people that will be infected.
People in the United States should not panic for several reasons, Vespignani said. Two of the three major avenues for transmission in Africa -- in hospitals and at funerals -- are not considered conduits for disease here, because of better infection control practices. In models in which the disease does show up in the US, as it now has in Texas, the models also show the outbreak will be contained because the spread will occur when people are initially ill and undiagnosed, limiting possible infection to family members or other people in close contact.
“We should be able to contain those types of outbreaks very well. Our estimates give us a median [average] outbreak of less than three cases in the US, a very limited outbreak,” Vespignani said. Those could likely be family members or people who interact with a person before it is clear he or she is infected.
But the appearance of the deadly virus in the U.S. is extremely serious, and the odds of more people coming over with the disease only increases as the outbreak rages on in Africa. The most effective way to reduce the probability of more cases is to contain the virus where it is spreading, Vespignani said.
Cutting back on flights to the afflicted region by 80 percent can decrease the probability of a case being imported, but even such extreme measures only delay the importation of the virus by about three or four few weeks, Vespignani said.
The biggest worry may be that his model shows increasing probability the disease will spread to nations such as Ghana, Ivory Coast, Mali, Gambia, and Morocco. In the United States, where there is a sizeable probability a case could arrive through an infected traveler, there is confidence in the health care system’s ability to contain outbreaks. But in other countries, those systems may not bein place. Vespignani said that his models show that if the virus continues to spread, the probability stretches to countries that now have very small chances of having an Ebola case, such as countries in Asia.
“This creates a kind of chain of events; every time a country has a potential outbreak, if it is not able to contain it, that increases the number of people who will travel,” Vespignani said.
The solution, he said, is not to panic about Ebola symptoms in the U.S., but for countries to work to help contain the outbreak in Africa to reduce their risk.