The Ebola outbreak has been devastating. As of April 1, 2015 there have been 25,178 Ebola cases and some 10,000 deaths in just three nations: Sierra Leone, Liberia, and Guinea. This is according to the World Health Organization (WHO) figures.
While these numbers are very high, they are also far lower than the projections made earlier last last year by some reputable agencies, including the WHO Ebola Response Team, the U,S. Centers for Disease Control and Prevention (CDC), and others. Some of the more alarming reports predicted that some 1.4 million people would be infected by the end of January 2015.
Commenting on the difference in the “actual” and predicted figures, Aaron King, an associate professor of ecology, evolutionary biology, and mathematics at the University of Michigan, said via his university: “Those predictions proved to be wrong, and it was not only because of the successful intervention in West Africa. It’s also because the methods people were using to make the forecasts were inappropriate.”
In a new paper, published in Proceedings of the Royal Society B, King has outlined the porblems of standard epidemic modeling methods (such as those used by WHO and CDC). As an alternative, King points out ways to generate more accurate predictions, including incorporating disease transmission equations. Kings suggests that the off-the-shelf deterministic models, used by the major health agencies, do not account for elements of randomness and uncertainty.
“Attention to uncertainty helps one prepare rationally for a range of likely scenarios,” King has said in an interview to the website All Africa. he adds: “If followed, our proposals will help to reduce the chance that we make large mistakes.”
King’s research paper is titled “Avoidable errors in the modelling of outbreaks of emerging pathogens, with special reference to Ebola.” It is available for free.
In related news, a new study suggests that the Ebola virus may not be evolving as quickly as a previous research groups have estimated.
