Researchers are working to enable influenza outbreaks to be predicted in much the same way as meteorologists forecast weather.
The research led by RMIT offers a potential boon for public health officials and the general public, by supporting improved management of flu outbreaks.
RMIT University's Professor Lewi Stone, from the School of Mathematical and Geospatial Sciences and the Platform Technologies Research Institute, was lead investigator in the recently published study.
Professor Stone collaborated with researchers from Tel Aviv University (Israel), Princeton University and National Institutes of Health (USA) to develop a simple human influenza epidemiological model that predicts multi-annual outbreaks.
"Influenza kills 250,000 to 500,000 people each year around the globe, with 2,500 Australians dying annually from the virus," he said.
"We know generally that winter is peak season for influenza, but any long-term planning is difficult because outbreaks are so hard to predict.
"If we can help health authorities better understand when influenza epidemics will strike, they can better prepare and ensure treatments are targeted where they will be most effective in saving lives."
Human influenza occurs annually in most temperate climatic zones of the world, with epidemics peaking in the cold winter months.
Seasonal drivers include weather variables, such as temperature and humidity, and social factors such as the frequency of close contact between individuals.
Recurrent influenza epidemics need a sufficient and continuous source of new susceptible individuals arising in a population, enough to fuel each new outbreak.
In the case of influenza, infected individuals recover with immunity but eventually become susceptible again because of the rapidly evolving nature of the virus.
"This creates a renewed source of susceptible individuals, also known as antigenic drift," Professor Stone said.
"Attempts to make long-term predictions of infectious diseases are hampered by our inability to understand the complex interplay of these dynamics."
The research team designed a simple epidemiological model based on 12 years of Israeli influenza surveillance data, resulting in a remarkable level of prediction accuracy that has not yet been achieved elsewhere.
The study, published in the Proceedings of the National Academy of Sciences, looked at Tel Aviv, Israel's largest city, using data from June 2001 to January 2013.
The data was obtained from the Maccabi Health Maintenance Organization, whose medical surveillance covers about 45 per cent of the Tel Aviv population.
Professor Stone said that the dataset strongly matched the incidence of confirmed influenza cases.
"The high level of coverage and the data quality makes this one of the finest available influenza surveillance datasets by world standards," he said.
"We used the classical Susceptible-Infected-Recovered-Susceptible (SIRS) epidemic model of a large city population.
"The model captures the complex interaction between the changing supply of new susceptible individuals arising due to loss of immunity in the population through antigenic drift, the strong transient dynamics following the appearance of a new strain, and the timing of the climatic cycle each year."
Professor John Hearne, Head of the School of Mathematical and Geospatial Sciences, said successful forecasting could improve preparation and management of annual flu outbreaks in Australia and the United States.
"If the forecasts are reasonably accurate, they could help public health officials target vaccines and anti-viral drugs to areas of greatest need," Professor Hearne said.
"With expertise ranging from information security and network science to risk and community safety and system modelling - in addition to our strong links with industry and government - RMIT's researchers continue to strive to meet the changing needs of our local and global communities."