Linking contact behavior and droplet patterns to dynamically model indoor respiratory infections among schoolchildren.

BACKGROUND: We used the results of a contact behavior survey in conjunction with droplet pattern measurement to investigate the indoor population transmission dynamics of respiratory infections. METHODS: A total of 404 questionnaires on all contact behaviors were distributed to junior high school students. Droplet number concentration and size distribution generated by coughing and talking were measured by droplet experimentation. A deterministic susceptible-exposed-infected-recovery (SEIR) model was used to simulate the indoor transmission dynamics of influenza infection among schoolchildren. RESULTS: Results indicated that the average contact rates ranged from 9.44 to 11.18 person(-1) day(-1) for grades 7 to 9. We showed that total median droplet number concentrations were 9.01 × 10(7) and 8.23 × 10(7) droplets per cubic meter for coughing and talking, respectively. Population dynamic simulations indicated that the size-dependent median number of droplets per person resulted in a maximum of 8 and 10 infected persons on day 4, respectively, for talking and coughing activities. CONCLUSIONS: Human contact behavior and airborne droplet characteristics may substantially change predicted indoor population transmission dynamics of influenza infection.