Dynamic Model On Strategies For Containing Influenza Outbreaks In Schools

OBJECTIVE:To simulate the outbreak of influenza in a school in Changsha using a dynamical model, and evaluate the effectiveness of different school-closure strategies to control the outbreak in school and the pandemic in community, and assess the effectiveness of individual strategies and combination strategies, including antiviral treatment, isolation, antiviral prophylaxis, school closure, vaccination before outbreak, vaccination in outbreak.METHODS:Assumed all students of a middle school in Changsha were the boarder, and based on the history of influenza, we set up a susceptible-exposed-infectious/asymptomatic-removed (SEIAR) model to simulate an outbreak of influenza in a school in Changsha. The outbreak was simulated by the software of Berkeley Madonna8.3.18. Indicators including total attack rate (TAR), total asymptomatic infection rate (AIR) and total infection rate (TIR) were estimated to assess the effectiveness of all the strategies for controlling the outbreak of influenza.RESULTS:The basic reproduction number (ko) of influenza in the school was4.40.When the school was not closed, TAR, TIR of school influenza were46.32%and98.18%respectively, and TAR, TIR of community influenza were0, and TAR, TIR of total influenza were2.12%and4.51%respectively. After the implementation of different school-closures, the TAR and TIR of influenza in school were only reduced by0.81%-13.82%, and TAR, TIR of community influenza were23.04%-24.66%,48.97%-52.39%, respectively, and TAR, TIR of total influenza were24.05%-25.34%s51.11%-53.86%.When assessed the individual strategies, the most effective strategy was vaccination before outbreak with70%susceptible vaccinated per day. The TAR, AIR and TIR reduced by79.74%.When two kinds of interventions combined, the most effective strategy was the strategy21, which was "one-week school-closure+ vaccination before outbreak with70%susceptible vaccinated per day (S1w+VP70)", with the TAR, AIR and TIR reduced by91.02%, and the duration of influenza shortened25days.When three kinds of interventions combined, the most effective strategy was the strategy31, which was "antiviral prophylaxis+isolation+vaccination before outbreak with70%susceptible vaccinated per day (P+Q+Vp70)", with the TAR, AIR and TIR reduced by91.12%, and the duration of influenza shortened20days.When four kinds of interventions combined, the effectiveness of these combination strategies which included vaccination were better and changed little, with the TAR, AIR and TIR reduced by89.51%-91.38%, and the duration of influenza shortened20-27days.When five kinds of interventions combined, the effectiveness of all combination strategies were better and changed little, with the TAR, AIR and TIR reduced by90.84%-91.53%, and the duration of influenza shortened24-29days.When six kinds of interventions combined, that was "antiviral prophylaxis+antiviral treatment+isolation+one-week school-closure+vaccination before outbreak+vaccination in outbreak (P+T+Q+S1w+VP70+V170)", the TAR, AIR and TIR reduced by91.53%, and the duration of influenza was shortened29days.CONCLUSIONS:SEIAR model is stable when used to simulate the outbreak of influenza in a school. Compared to be not school-closure, the effectiveness of different school-closures is small, and school-closure would cause an increase epidemic of community influenza and total influenza. Considering the difficulty of implementing interventions by health department and fewer numbers of combination strategy, so "antiviral prophylaxis+isolation+vaccination before outbreak" seemed to be a more appropriate combination strategy. The effetiveness of combination strategies with vaccination was better than combination strategies without vaccination. The effectiveness of individual strategy, which was vaccination before outbreak with70%susceptible vaccinated per day, was better than the all combination strategies of "isolation","antiviral treatment","antiviral prophylaxis" and "one-week school closure".