If there is over-dispersion, fewer people are responsible for more of the overall transmission. For instance, a lower percentage of cases cause 80% of onward transmissions. When extreme, these are sometimes called superspreaders. 2/n 
With more over-dispersion, it is less likely that any one imported case will cause additional cases in a new country (though those that do will cause more cases). That means any one introduction is not likely to cause an epidemic, but those that occur might be very explosive. 3/n
Reductions in R have a similar effect. Reductions might be because health officials catch cases and prevent them from transmitting, or due to differences in how well the virus transmits in different places. 4/n
So we tried to figure out at various of levels of R, what amount of over-dispersion best explains the data with the goal of figuring what levels of over-dispersion were needed to explain the lack of onward transmission assuming various reductions in R. 5/n
Does it matter for control? Yes. If you don't know who might be a superspreader and randomly allocate control, over dispersed epidemics are harder to control. But... 6/n
...if we can identify the contexts when superspreading is most likely to occur, the epidemic becomes easier to control. For SARS a lot of the transmission was happening in hospitals, so good infection control measures helped eliminate superspreading and stop the epidemic. 7/7
