epidemics & epidemiological models
The spread of ideas is often likened, not always successfully, to the spread of viruses, hence the term viral marketing. (The striking images left are 3D hyperbolic graphs of
Internet topology, created using the Walrus
visualisation tool developed by Young Hyun at CAIDA. See copyright note below.)
When considering how ideas are spread, an understanding of some key concepts in epidemiology can give new insights and useful understanding.
Epidemiology is the study of infection and disease in populations.
All epidemics at the population level have the same bell-curve pattern. What we don't know is the scaling.
When R is the rate of infection, and R = <1, there is no chance of an epidemic. When R is greater than one, then there is a chance of an epidemic.
Below are two scenarios:
There are four factors for disease epidemics (i.e. when R = >1) and a change in any one of them changes the outcome:
- Probability of transmission: dependent on infectivity (the capability of entering, surviving and multiplying in a host), virulence (the ability to cause disease) and pathogenicity (disease-producing capacity)
- Duration of infectivity: (e.g. Ebola is 2-3 days, HIV is 10+ years because of treatment regimes)
- Contact rates: who comes into contact with the agent
- Susceptability: determinants can be changed through the physical environment (e.g. temperature, nutrients, toxins), interventions (e.g.vaccines, animal dips, movement controls, quarantines) or other biological condition (e.g. being in utero, immunosuppression).
Susceptibility is the fuel by which epidemics run . . . but it gets even more complicated than this. A host's susceptibility depends on all the factors above and upon whether or not the agent is itself changing. If it stays the same, the epidemic follows what's known as the SIR-model; if the agent changes, it follows the SIS-model.
The image at the top of this webpage is copyright 2002 The Regents of the University of California. All Rights Reserved. Permission to use, copy, modify and distribute any part of this "Walrus Image of Internet Topology from skitter data" for educational, research and non-profit purposes, without fee, and without a written agreement is hereby granted, provided that the above copyright notice, and this paragraph appear in or near all copies.