The big dog right now seems to be the folks at Virginia Tech. They seem to have the most sophisticated models, which require clusters of supercomputers to execute and are actually contracted by the Department of Health and Human Services to advise the Federal government in situations like this. Actual experiments are useful, but you can simulate things which are impossible to study in real life and you can tweak things over and over again to your heart's content.
One strategy which stood out as particularly effective in stopping a fast spreading disease like a pandemic flu in simulations is "social distancing," particularly of school age children, since they are prolific vectors of the flu. The most effective way to implement closing schools is right at the very beginning of an outbreak.
I can tell you that the little VT airport has been hopping lately with government jets, and the Hokie-birds (the twin jets that are used as Blacksburg-DC shuttles) are in the air almost constantly. The modeling is mainly contracted by the DHHS, but the whole governmental alphabet is involved, including the folks in uniform.
The coolest thing about the VT models is that you can tweak them on the fly until you find the best response. It also includes social factors, which is really what this thread is about. You can close schools, but unless it's accompanied by a curfew or lock-down there will still be some interaction. There's not much the government can do about that without scaring people or forcing them to stay home. As data comes in from all directions (sociology, psychology, etc.), such as "What percentage of people will ignore an order to stay home?" it's added to the model. Notice is being taken. If the H1N1 stays at the same level of threat, it is going to be a very useful real-world data stream for the models, and they will be even better the next time they're needed.
) 
Previous Topic
Index
