Friday, 5 June 2009

Anti-viral drugs are only one weapon in the fight against pandemics such as ‘swine flu' (H1N1).Another key to helping policy makers such as public health authorities decide how best to respond is data produced by computer modelling. And that modelling has shown that social distancing is the only guaranteed defence against a new strain of influenza in the early stages of a pandemic.

Professor George Milne, Dr Joel Kelso and Associate Professor Heath Kelly in the School of Computer Science and Software Engineering have created a computer-based model which simulates the transmission of an infectious disease in a community of 30,000 people in the developed world.

Dr Kelso said this wasn't the first time modelling had been used to make sense of and predict health emergencies. During the 1920s, mathematicians used differential equations to work out the likely percentages of people to be infected and how long epidemics would last.

He said such equations had been used to make sense of the physical world since Isaac Newton's day, explaining everything from the solar system, electricity and magnetism to the way fluids flow around the wing of an aeroplane. Computers could be used to make predictions from highly complex cases, using a different language to that familiar to the pre-computer mathematicians.

Professor Milne's team used data pertinent to Albany from the Australian Bureau of Statistics and the Departments of Education and Training and Planning and Infrastructure. They also precisely simulated movement within the community, and individual contact patterns based on information from other surveys and mobile phone companies.

Theirs is the first model of the spread of infectious disease to focus on the precise replication of a small community, down to individual schools, employers and the exact make-up of households. It is the most detailed replication of an actual population centre used to quantify the impact of alternative intervention strategies. It models the spread of pandemic influenza and the effect on daily and final attack rates of four social distancing methods: school closure, isolating infected family members inside the home, staying away from the workplace, and avoiding community contact.

Professor Milne said the model showed that non-pharmaceutical social distancing interventions could reduce the rate of development and the overall burden of epidemics. Together with the use of anti-viral drugs, social distancing is the only guaranteed defence against a new strain of influenza in the early phases of a pandemic and could be readily activated as a first line weapon in developing and developed countries.

"While such draconian measures seem unlikely to be mandated given their impact on personal freedom, they appear to play a key role in delaying the development of a ‘worst case' influenza epidemic," he said. "Social distancing may be critical in holding back an epidemic until vaccines are deployed on a sufficient scale that subsequent relaxation of these measures won't result in an acceleration of the outbreak.

"The measures must be employed as soon as possible if they are to have maximum effect. We found that for an outbreak as infectious as the 1918 Spanish Flu pandemic, a combination of all intervention measures must be introduced within two weeks of the first case appearing in a town or city to prevent an epidemic developing. Delays of two, three and four weeks resulted in final attack rates of seven per cent, 21 per cent and 45 per cent respectively."

The UWA team has provided reports based on their two published papers to the Federal Department of Health and Ageing and the World Health Organisation based on its computer modelling. It received funding from the National Health and Medical Research Council, DoHA and WHO.

Tags

Groups
ECM Faculty Focus