GovernmentExecutive.com

Return to Article: Software allows agencies to model disease, disaster response

• 47934
  

  Pandemic modeling has been discussed at some length in other venues. A good place to start might be at the Effect Measure blog and at fluwikie.com. Search either for "model". In short, the models are only as good as the assumptions that go into them. IMHO, the assumption of a 75% reduction in interactions is unrealistic.

  MoM Posted April 4, 2008 12:39 PM
• 47322
  

  A more interesting point about the "prevented the outbreak from becoming an epidemic" statement is the 75% reduction in agent interactions for a month. In Singapore when the SARS infection occurred, this highly disciplined and authortarian society managed to contain the virus by preventing movement. How will a society based on individual freedom and distrust of central government react to being told to reaminnin thier homes for a month, with no outside contact whatsoever? Does the Brookins Institute really think that preventing three quarters of the population of the USA from moving around for a month is really feasible, or have they not thought past the numbers and the elegant graphics?

  Jeff Goatcher Posted March 27, 2008 6:15 AM
• 47068
  

  From the article, "But only now have they become big enough and flexible enough to model and predict human behavior on a large scale." Reading this, I thought this is the beginning of the "Foundation Trilogy" by Isaac Asimov. From Wikipedia, "The premise of the series is that scientist Hari Seldon spent his life developing a branch of mathematics known as psychohistory". "Using the law of mass action, it can predict the future, but only on a large scale; it is error-prone for anything smaller than a planet or an empire. It works on the principle that the behavior of a mass of people is predictable. And the quantity of this mass tends to be very huge (equal to the population of the galaxy). The larger the mass the more predictable is the future."

  Just curious, and as sensible as this sounds: "In the LA-based flu scenario, the model showed that reducing interactions among agents by 75 percent for one month prevented the outbreak from becoming an epidemic." Does that mean that we might see practice exercises where the government restricts travel? I think that national ID card might go a long way towards helping this.

  With the current national atmosphere of fear (spoken of in the book "State of Fear" by Michael Crichton, albeit from yet another possible threat to human life) could we "see the effects of specific interventions from government or other organizations" in the near future?

  "the university center of excellence is autonomous and retains the right to choose what it studies so the model could not be commandeered by a federal agency." And wasn't atomic research to be used for peaceful energy uses only? If the government feels the need; how long will it be before terrorist scenarios, like a dirt bomb, are tested? Or even the social-economic impact of a flat tax? Certain laws such as the PATRIOT Act, or the FISA? I can't see the university center of excellence being a stalwart of independence should the DHS consider this software to be a national security resource. Can you?

  Tip off Posted March 24, 2008 9:28 AM
• 47053
  

  This article doesn't make clear what these videos are showing. Like, what's with the pretty colors in the first one. WHERE is the second one?

  Jack S Posted March 24, 2008 7:41 AM
• 46980
  

  While this model is a true tour de force, it is not the only approach to this topic. In addition, although the inventor and correspondent make the approach out as an excellent model because it's intuitive, one needs to validate the model correctness before making major decisions based on its predictions. For instance, the spread of the great influenza of 1920, mapped after the fact, also showed a good image of the passenger rail system of the time. More recently, the spread of SARS was very much tied into the connectivity of regions by air. The transportation system has its own dynamics and models, and these may overwhelm (in a real incident) the results of the agent-based model shown. On the other hand, a less complex disease model combined with transportation models might show a much more useful picture, since it would suggest a mechanism to cut disease lines without shutting down society (something which could be impossible in any case). The cautionary note is not to treat an interesting invention as a panacea. There are many approaches with their strengths and weaknesses, and we must be very thorough in our analysis of these -- including how well they would have functioned as predictors for infections in history.

  Richard Wright Posted March 21, 2008 11:59 AM