Predicting how long it takes for an infectious disease outbreak to begin may help health officials develop an early warning system, according to the authors of a new study on lag times in epidemics.
To prevent large scale disease outbreaks, any health official would love to have a crystal ball capable of forecasting epidemics. In reality, predicting an outbreak is difficult and involves many factors, but a group of researchers have recently developed a model that could help us create an early warning system for infectious diseases.
According to the World Health Organization (WHO), while noncommunicable diseases such as cancer and cardiovascular disease together cause the vast majority of deaths around the world, infectious diseases caused by bacterial, viral, parasitical, and fungal pathogens, as a whole, pose a major public health threat. Some of the biggest infectious threats around the world include tuberculosis, measles, malaria, and HIV/AIDS. Most recently and notably, the Zika outbreak in South America and Central America has gained the attention of infectious disease specialists. Public health agencies such as the Centers for Disease Control and Prevention (CDC) aim to track emerging diseases through surveillance efforts at national, state, local, and territorial levels to help them respond to and prevent outbreaks that put the population at increased risk. However, even with the best data and monitoring, knowing that there will be an outbreak before it happens and understanding the timing of an epidemic are difficult factors for health experts to predict.
In a new study, researchers at the University of Georgia aimed to understand what impacts the waiting time between when the possibility of an infectious disease outbreak emerges and when an epidemic actually takes place. The research was conducted as part of a National Institutes of Health-funded study called Project AERO, focused on anticipating emerging and re-emerging outbreaks. The authors note that infectious diseases begin as “self-limiting, stuttering chains of infection” which tend to die off, and can develop into the more “sustained chains of human-to-human transmission” that lead to larger outbreaks. The transition from the former to the latter becomes possible when an outbreak reaches its tipping point, which, in regards to infectious diseases, occurs when every person infected with a disease is in a susceptible population and has transmitted the infection to more than one other person.
In any given population, there are individuals who are susceptible to a disease, those who are presently infected, and those who have recovered from infection. The researchers created a model to determine the parameters impacting what they call bifurcation delay, the waiting time between the tipping point and the start of an actual outbreak. They noted that outbreaks do not begin as soon as they are possible; creating a model to understand and forecast that delay could help create an early warning system for coming outbreaks.
By running thousands of computer simulations, the research team found that two things create the biggest impact on bifurcation delay. One is what the authors call the “sparking rate,” which is the rate at which infections are introduced from outside the population. The second, known as the “sweeping rate,” is the speed of basic reproductive ratio, or the number of new infections an infected individual is expected to cause. As the sparking rate becomes relatively lower, the sweep rate becomes less important, according to the authors. However, greater sparking rates lead to smaller delays in infectious disease outbreaks. In cases of slow sweep rates, adding an infectious individual to a population when basic reproductive ratio is low leads to low likelihood of a major outbreak.
“Whether a newly arriving infection gives rise to an epidemic depends on so many factors; individual contact rates, who gets contacted and for how long, population density, the rate of vaccination, whether there's an adequate public health response, the time of year, and even time of day, as viral loads can differ dramatically from morning to evening,” said study author, Christopher J. Dibble, PhD. “Project AERO focuses on forecasting the transition to supercriticality - the theoretical point at which a disease outbreak becomes possible. This happens when each infection leads to, on average, more than one new infection. Below one, and the disease will go extinct in the population. Above one, and it can increase very rapidly.”
The work, Dr. Dibble admitted, is still in its theoretical stages, and without more research could be prone to error. “The first problem is that an outbreak could happen well before we say it will. But in that case, we've at least forecasted the transition, and given folks some advance notice,” he explained. “What happens, though, if the outbreak never happens? That's good, right? Well, it depends. It could be that all of our interventions actually reduced the risk of an epidemic, and we went from a bad situation to a better one, or it could be that we were wrong, and raised the public alarm for no good reason. A major problem is that we'll never know whether we're right or wrong, unless an epidemic happens. That's one of the reasons we're focused so much on getting the theory right, through math and simulation; it gives us more confidence that any future forecasts will be robust.”