Mac mathematician predicts epidemics
Suzanne Morrison, Medical Research Reporter
The Spectator
UPDATED: Fri Jan 28, 2000 02:39 AM
- David Earn, a professor of applied mathematics at McMaster University, has helped refine a formula to predict epidemics
David Earn has discovered increases or decreases in birth or vaccination rates cause dramatic changes in patterns of epidemics.
The research is published in today's journal Science.
Earn, a professor of applied mathematics, said dramatic changes in the patterns of epidemics, particularly measles, have puzzled epidemiologists and ecologists.
"We now have a simple model that explains the transitions in patterns of epidemics in the last century and we can attribute them to changes in birth rate and/or changes in vaccination levels," said Earn, who has spent the past 10 years at Cambridge University in England.
Earn said nobody has shown before that birth rates need to be considered.
"For example, if you increase vaccination levels but the birth rate also goes up, then you may not have much of an impact. You need to bear these things in mind."
For the mathematically inept, Earn explained his formula as "the birth rate times one minus the vaccination rate.
"That's because it's one minus the vaccination rate that gives you individuals leftover who are susceptible after your vaccination program. That and the birth rate tell you the rate at which susceptible individuals are coming into the population."
Earn said the practical application means that if the number of children who get vaccinated for measles, mumps and rubella decreased -- as is happening in Britain because of vaccine scares -- his model lets scientists predict what patterns of epidemics will emerge.
That means doctors can be alerted and get prepared when more measles cases will be cropping up.
Earn developed the mathematical model with a team of researchers at the University of Cambridge, England and Benjamin Bolker, assistant professor of zoology at the University of Florida.
They studied historical data on outbreaks of measles in London, Liverpool, New York and Baltimore, then examined historical records of births, vaccination and cases of measles.
Patterns of measles epidemics in those cities range from similar outbreaks every year, to large or small outbreaks in alternate years, to very irregular outbreaks of varying size.
In each city, numerous shifts occurred between various epidemic patterns. The research discovered the causes were changes in birth rates and vaccination rates.
"The key thing is that we are now able to explain these changing patterns of epidemics which, in the past, have been mysterious," he said."And (that) ... will certainly help us in trying to understand patterns of epidemics in the future."
Earn said previous work predicting epidemics used complex models, such as patterns of transmission of a virus between people of different ages."The model we propose is much simpler."
The same approach could be used for chicken pox, polio and whooping cough because they involve the same virus which remains consistent over time.
It's not applicable to flu outbreaks because that virus is constantly changing.
Earn returned to McMaster today from a flu workshop in Mexico where scientists and mathematicians tried to understand why the flu does what it does and how to prevent it from causing so much damage in the future.
"I think it will be a long time before we have an understanding of the flu the way we do of measles.
"It's a much harder system to understand."
How did a mathematician get involved in studying epidemics?
Earn works on dynamic systems -- anything that changes in time -- and he says he's always been interested in epidemics.
"If you are a mathematician you have a bag of tricks in your pocket that you can apply to all sorts of problems," he said.
"This was one that had not been thought about in the way that I thought about it so it led to some interesting progress."
Legal Notice: Contents copyright © 1996-2001, The Hamilton Spectator. All rights reserved. Distribution, transmission or republication of any material from www.hamiltonspectator.com is strictly prohibited without the prior written permission of The Hamilton Spectator. For information contact us or send email to jaussem@hamiltonspectator.com