|
|
 |
Columns::March 31, 2003
Worth repeating
This years Odum Lecture was delivered by Brian Grenfell, professor of population biology at the University of Cambridge. A specialist in disease ecology, he discussed the use of data about childhood epidemics to derive principles that can be applied to other disease epidemics. Some excerpts:
What I want to do first is talk about childhood epidemics. . . . What I want to do particularly is show the very rich data that were privileged to have for these diseases, particularly for measles. . . .
For about 60 or 70 years, people have been making models of this oscillation. . . . The classic pattern of an epidemic is each case causes a number of other cases. . . . And each of those causes a number of other cases, so you get an exponentially increasing epidemic. . . . This progression uses up the susceptibles very rapidly . . . and that means the epidemic eventually turns over and dies away. . . .
Can we look at this across time and space? . . . Theres tremendous synchrony, but theres also a wave-like pattern. . . . The epidemic seems to start in London and move away from London in a wave, and it also starts in the northwest. It starts in big places and moves to the surrounding hinterland. So there is synchrony, but superimposed on the synchrony are these waves. And were trying to explain those patterns. . . .
So we can do statistical analyses, and these are spatial correlations. And they just show that the correlations of the raw data, which Ive called synchrony, and of these phases, drop with distance, so the correlation of the epidemic drops with distance and that gives us the idea that overall theres some hierarchical pattern of waves. . . . And things were much less synchronized in the vaccination era. . . .
|
|
|
|
|