A Bayesian ensemble approach for epidemiological projections.
A Bayesian ensemble approach for epidemiological projections.
复制标题
DOI：
10.1371/journal.pcbi.1004187
复制
发表时间：
2015-04
影响因子：
4.3
通讯作者：
Webb C
中科院分区：
文献类型：
--
作者：
Lindström T;Tildesley M;Webb C
数学模型是流行病学的强大工具，可用于比较控制措施。然而，不同的模型和模型参数化可能会提供不同的结果预测。在其他研究领域，集成建模已被用来组合多个投影。我们通过采用为气候预测开发的贝叶斯技术来探索将此类方法应用于流行病学的可能性。我们基于针对 2001 年英国口蹄疫爆发运行的沃里克模型的不同参数化，举例说明了单一模型集成的实施，并比较了不同控制措施的效果。这使我们能够研究基于不同建模假设的预测之间的差异对整体预测的影响。敏感性分析表明，先验的选择会对感兴趣数量的后验估计产生显着影响，特别是对于预测之间存在较大差异的集合。然而，通过使用该方法的分层扩展，我们表明可以规避先前的敏感性。我们进一步扩展该方法以包括关于不同建模假设的先验信念，并证明这种效果可能会根据预测之间的差异产生不同的后果。我们认为该方法是疾病爆发集成建模的一种有前途的分析工具。应对突发疾病爆发的政策决策使用模拟模型来告知不同控制行动的效率。然而，根据模型和参数化的选择，可能会做出不同的预测。集成建模提供了组合多个投影的能力，并已在其他研究领域成功使用。集成建模的一个中心问题是在组合投影时如何对它们进行加权。为此，我们在此调整并扩展了气候预测中使用的加权方法，以便将其用于流行病学考虑。我们通过将该方法应用于 2001 年英国口蹄疫爆发的整体预测来研究该方法的表现。我们得出的结论是，该方法是疾病爆发整体建模的一种有前途的分析工具。
Mathematical models are powerful tools for epidemiology and can be used to compare control actions. However, different models and model parameterizations may provide different prediction of outcomes. In other fields of research, ensemble modeling has been used to combine multiple projections. We explore the possibility of applying such methods to epidemiology by adapting Bayesian techniques developed for climate forecasting. We exemplify the implementation with single model ensembles based on different parameterizations of the Warwick model run for the 2001 United Kingdom foot and mouth disease outbreak and compare the efficacy of different control actions. This allows us to investigate the effect that discrepancy among projections based on different modeling assumptions has on the ensemble prediction. A sensitivity analysis showed that the choice of prior can have a pronounced effect on the posterior estimates of quantities of interest, in particular for ensembles with large discrepancy among projections. However, by using a hierarchical extension of the method we show that prior sensitivity can be circumvented. We further extend the method to include a priori beliefs about different modeling assumptions and demonstrate that the effect of this can have different consequences depending on the discrepancy among projections. We propose that the method is a promising analytical tool for ensemble modeling of disease outbreaks. Policy decisions in response to emergent disease outbreaks use simulation models to inform the efficiency of different control actions. However, different projections may be made, depending on the choice of models and parameterizations. Ensemble modeling offers the ability to combine multiple projections and has been used successfully within other fields of research. A central issue in ensemble modeling is how to weight the projections when they are combined. For this purpose, we here adapt and extend a weighting method used in climate forecasting such that it can be used for epidemiological considerations. We investigate how the method performs by applying it to ensembles of projections for the UK foot and mouth disease outbreak in UK, 2001. We conclude that the method is a promising analytical tool for ensemble modeling of disease outbreaks.
登录
查看更多内容
影响因子：
2.6
作者：
Backer, J. A.;Hagenaars, T. J.;van Roermund, H. J. W.
通讯作者：
van Roermund, H. J. W.
影响因子：
2.2
作者：
Gloster, John;Jones, Andrew;Paton, David
通讯作者：
Paton, David
影响因子：
4.3
作者：
Chao DL;Halloran ME;Obenchain VJ;Longini IM Jr
通讯作者：
Longini IM Jr
影响因子：
1.1
作者：
Garner, MG;Beckett, SD
通讯作者：
Beckett, SD
影响因子：
56.9
作者：
Ferguson, NM;Donnelly, CA;Anderson, RM
通讯作者：
Anderson, RM