"Building models isn't just about how something looks. It's about how biologically accurate it is."

The above quote highlights the importance of observing field data to build disease models, rather than building complex maths models per se. Maths is a tool to assist epidemiologists, but it should not replace the epidemiology, nor become more important than the epidemiology. For example, in the 2001 UK foot-and-mouth disease (FMD) epidemic, much was made of slaughtering animals promptly in order to achieve a rapid control of the disease spread, because some models had inaccurately predicted that the speed of slaughter was essential. This proved to be a misconception since speed of slaughter was not relevant in 50 per cent of infected herds [10].

Field data:

Speed of slaughter (for animals) or the speed of treatment (in humans) only applies to one section within a population, irrespective of whether that population is vaccinated or unvaccinated [11]. This is because for many diseases there is a subclinical form - a form that initially shows no clincial signs, and eventually when clinical signs are shown, they are only mild. The undetected seeding of disease means that any speed of slaughter is irrelevant for disease control. What is relevant for subclinical disease, is to carefully target and extend the surveillance zones around areas of subcinical infection ie. to detect the disease that has already spread.

In areas where only acute disease is found, then the speed of treatment (or slaughter) would affect the control of disease spread.