Please use this identifier to cite or link to this item:
|Title:||Contact network based framework for infectious disease interventions||Authors:||Zhang, Tianyou||Keywords:||DRNTU::Engineering||Issue Date:||2014||Source:||Zhang, T. (2014). Contact network based framework for infectious disease interventions. Doctoral thesis, Nanyang Technological University, Singapore.||Abstract:||Infectious diseases are life-threatening and often incur the enormous amounts of economic cost and social cost all around the world. As a densely populated hub city, Singapore has suffered from almost every pandemic in the recent decades. Controlling the spread of infectious diseases is challenging to the public health system and there is a need for policy makers to make wise decisions on the choices of intervention measures as well as the implementation in a timely manner. However, there is the lack of sound contact network model for Singapore community and there are few quantitative evaluations to support the decisions on public health interventions as well. In this thesis, we address these problems by introducing contact network based simulation framework for evaluating epidemic interventions in Singapore. The framework consists of a contact network generator “HPCgen” and an intervention-oriented simulator “IntSim”. HPCgen is a fast and scalable contact network generator for urban cities. We demonstrate that HPCgen is able to generate a labelled contact network of 13.4 million population in 7.27 minutes. IntSim is an extremely efficient multi-agent simulator that is able to simulate cyclic interventions, multi-level interventions and combined interventions. Using our framework for a practical study on influenza outbreak, we first create an idealised contact network based on real-life data in Singapore, including demographics, social structure information and contact behaviour surveys. Running IntSim with the idealised contact network for Singapore, we simulate the spread of influenza under multi-level school closure, cyclic workforce shift and their combination, with varying temporal parameters and transmissibilities. Our results show that social distancing is sensitive to temporal factors as well as intervention scale and frequency. The effect of all-school closure and workforce shift is saturated at 8 and 6 weeks respectively. All-school closure of the duration shorter than 6 weeks tends to be more effective if starting later in an epidemic and the closure of the duration longer than 6 weeks is wise to start as early as reasonable. Moreover, individual class closure is observed to excel in reducing the overall attack rate. All-school closure is the most effective to lower the peak incidence. We also discover if duration is longer than 6 weeks or school closure is triggered at prevalence of symptomatic infection equal to 5%, combined interventions outperform each individual intervention working alone. Combined interventions tends to be more effective when either school closure starts first and lasts for less than 4 weeks or workforce shift starts first and lasts for more than 4 weeks.||URI:||http://hdl.handle.net/10356/61718||DOI:||10.32657/10356/61718||Fulltext Permission:||open||Fulltext Availability:||With Fulltext|
|Appears in Collections:||SCSE Theses|
Page view(s) 20284
checked on Oct 28, 2020
checked on Oct 28, 2020
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.