چکیده (انگلیسی):

Recent events have thrown the spotlight on infectious disease outbreak response. We developed adata-driven method, EpiGro, which can be applied to cumulative case reports to estimate the order ofmagnitude of the duration, peak and ultimate size of an ongoing outbreak. It is based on a surprisinglysimple mathematical property of many epidemiological data sets, does not require knowledge or estima-tion of disease transmission parameters, is robust to noise and to small data sets, and runs quickly dueto its mathematical simplicity. Using data from historic and ongoing epidemics, we present the model.We also provide modeling considerations that justify this approach and discuss its limitations. In theabsence of other information or in conjunction with other models, EpiGro may be useful to public healthresponders.© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).1. IntroductionAs infectious diseases are identified for the first time or emergein new populations, researchers increasingly use mathematicalmodels to describe observed patterns and to plan and evaluatepublic health responses (Anderson and May, 1992; Grassly andFraser, 2008; Keeling and Danon, 2009; Anderson et al., 2015).These models vary in complexity and scale, from simple compart-mental models (Hethcote, 2000) to complex stochastic agent-basedand metapopulation approaches that include external informationlike transportation networks (Rvachev and Longini, 1985; Hufnagelet al., 2004; Eubank et al., 2004; Ferguson et al., 2006; Balcan et al.,2010; Ajelli et al., 2010; Van den Broeck et al., 2011). The latter havebeen shown to efficiently capture the real-time spread of epidemics(Tizzoni et al., 2012), but often require large amounts of informa-tion. Key parameters need to be estimated from epidemiologicaldata, which may be accomplished by maximum likelihood estima-tion (Ionides et al., 2006; Bretó et al., 2009; King et al., 2015) ordata assimilation (Rhodes and Hollingsworth, 2009; Shaman andKarspeck, 2012). However, for newly emerging infections or whenestimating the impact of bioterrorism events (Walden and Kaplan,2004; Rotz and Hughes, 2004), such information may not alwaysbe available. Sometimes, the community is able to quickly compile∗Corresponding author at: Department of Mathematics, University of Arizona,617 N. Santa Rita Avenue, Tucson, AZ 85721, USA.E-mail address: lega@math.arizona.edu (J. Lega).and share epidemiological parameters, as was for instance the casefor the devastating 2014/2015 Ebola outbreak (Van Kerkhove et al.,2015; Chowell et al., 2014). It is nevertheless expected that modelchoices reflect the balance between data availability and the needsof the public health community (Keeling and Danon, 2009). More-over, since the accuracy of predictions depends heavily on modelingassumptions (Keeling and Danon, 2009; Wearing et al., 2005), it isalso important to balance the need for detailed, realistic modelsagainst limitations in parameter information (May, 2004).Knowing how many cases to expect, as well as when they willpeak, before an outbreak has run its course is central to preparing apublic health response (Flu Activity Forecasting Website Launched,2016). Entire epidemiological curves can often be fitted with stan-dard functions, such as for instance a logistic curve or the Richardsmodel (Tjørve and Tjørve, 2010; Peleg and Corradini, 2011; Wanget al., 2012; Ma et al., 2014), but are only effective late into theoutbreak. Conversely, time series approaches allow forecasting,but are considered accurate only for short-term prediction. Forinstance, using only case data and an autoregressive integratedmoving average (ARIMA) model, researchers were able to fore-cast hospital bed utilization during the severe acute respiratorysyndrome (SARS) outbreak in Singapore up to three days forward(Earnest et al., 2005). Additional information is usually required forlonger forecasts (see e.g. 3-month dengue forecasting using climatedata (Gharbi et al., 2011)), limiting the utility of such approachesfor newly emerging diseases, when many associated risk factors arestill unknown.
http://dx.