Application of an individual-based model with real data for transportation mode and location to pandemic influenza

Abstract

Currently, an individual-based model is a basic tool for creating a plan to prepare for the outbreak of pandemic influenza. However, even if we can construct the model as finely as possible, it cannot mimic the real world precisely. Therefore, we should use real data for transportation modes and locations, and simulate the diffusion of an infectious disease into that real data. In the present study, we obtained data on the transportation modes and locations of 0.88 million persons a day in the Tokyo metropolitan area. First, we defined the location of all individuals in the data set every 6 min. Second, we determined how many people they came in contact with in their household, in each area, and on the train, and then we assumed that a certain percentage of those contacted would become infected and transmit the disease. Data for natural history and other parameters were taken from previous research. The average number of contacts in each area was 51 748 (95% confidence intervals [CI],46 846-56 650]), at home it was 246 (95% CI, 232-260), and on the train it was 91 (95% CI, 81-101). The number of newly infected people was estimated to be 3032 on day 7 and 126 951 on day 10. The geographic diffusion on day 7, the day when the earliest response would have started, expanded to the whole of the Tokyo metropolitan area. We were able to realize the speed and geographic spread of infection with the highest reality. Therefore, we can use this model for making preparedness plans.