A team of scientists at Columbia University’s Mailman School of Public Health recently discovered a new method to accurately predict the timing and intensity of West Nile virus outbreaks.
The method uses a computer model to forecast multiple situations that imitate potential behaviors of an outbreak to generate a prediction. Previously, Columbia researchers have used computer models to predict weekly forecasts for seasonal flu. They also used similar methods following the Ebola outbreak in West Africa.
The forecasting model draws data from mosquito infection rates, reported human cases, and transmission rates between mosquitoes and birds, which then make contact with humans. In one instance, the researchers used their models to create retrospective predictions for West Nile outbreaks in Suffolk County, New York between 2001 and 2014, which accurately predicted the total number of seasonal cases of the virus in humans up to nine weeks prior to the last reported case.
“There is a great deal of variation in outbreak intensity and duration year to year,” study co-author Nicholas D. DeFelice, a post-doctoral researcher at Columbia, said. “Absent a computer model, it’s difficult to predict the impact of an outbreak, even once the outbreak is underway, and thus it is important that robust quantitative decision tools are developed to help guide control efforts.”
The researchers said their forecasting models would become more accurate as more years of data are made available. They also said they are currently working on a real-time forecasting system.
The study was supported by grants from the National Institutes of Health. Data from the study was made available in a recent issue of the journal Nature Communications.