Bayesian Spatio-Temporal Prediction and Counterfactual Generation: An Application in Non-Pharmaceutical Interventions in COVID-19

The spatio-temporal course of an epidemic (such as Covid-19) can be significantly affected by non-pharmaceutical interventions (NPIs), such as full or partial lockdowns. Bayesian Susceptible-Infected-Removed (SIR) models can be applied to the spatio-temporal spread of infectious disease (STIF) (such as Covid-19). In causal inference it is classically of interest to investigate counterfactuals. In the context of STIF it is possible to use nowcasting to assess the possible counterfactual realization of disease in incidence that would have been evidenced with no NPI. Classic lagged dependency spatio-temporal IF models will be discussed and the importance of the ST component in nowcasting will be assessed. The real example of lockdowns for Covid-19 in two US states during 2020 and 2021 is provided. The degeneracy in prediction in longer time periods is highlighted and the wide confidence intervals characterize the forecasts.

[1]  D. Chumachenko,et al.  Investigation of Statistical Machine Learning Models for COVID-19 Epidemic Process Simulation: Random Forest, K-Nearest Neighbors, Gradient Boosting , 2022, Comput..

[2]  Todd E. Clark,et al.  Nowcasting Tail Risk to Economic Activity at a Weekly Frequency , 2022, Journal of Applied Econometrics.

[3]  A. Lawson,et al.  Bayesian space-time SIR modeling of Covid-19 in two US states during the 2020–2021 pandemic , 2022, medRxiv.

[4]  Jue Liu,et al.  Global Percentage of Asymptomatic SARS-CoV-2 Infections Among the Tested Population and Individuals With Confirmed COVID-19 Diagnosis , 2021, JAMA network open.

[5]  R. Doorley,et al.  Using mobile phone data to estimate dynamic population changes and improve the understanding of a pandemic: A case study in Andorra , 2021, medRxiv.

[6]  A. Galvani,et al.  Asymptomatic SARS-CoV-2 infection: A systematic review and meta-analysis , 2021, Proceedings of the National Academy of Sciences.

[7]  T. Middleton,et al.  Modeling the Economic and Societal Impact of Non‐Pharmaceutical Interventions During the COVID‐19 Pandemic , 2021 .

[8]  Maria L. Daza-Torres,et al.  Bayesian sequential data assimilation for COVID-19 forecasting , 2021, Epidemics.

[9]  A. Lawson,et al.  Space-time covid-19 Bayesian SIR modeling in South Carolina , 2020, medRxiv.

[10]  Katriona Shea,et al.  The SEIRS model for infectious disease dynamics , 2020, Nature Methods.

[11]  D. Fisman,et al.  Mathematical modeling of COVID-19 transmission and mitigation strategies in the population of Ontario, Canada , 2020 .

[12]  Chawarat Rotejanaprasert,et al.  Bayesian spatiotemporal modeling with sliding windows to correct reporting delays for real-time dengue surveillance in Thailand , 2020, International Journal of Health Geographics.

[13]  Nicolas A. Menzies,et al.  Nowcasting by Bayesian Smoothing: A flexible, generalizable model for real-time epidemic tracking , 2019, bioRxiv.

[14]  Duncan Temple Lang,et al.  Programming With Models: Writing Statistical Algorithms for General Model Structures With NIMBLE , 2015, 1505.05093.

[15]  Michael A. Johansson,et al.  Nowcasting the Spread of Chikungunya Virus in the Americas , 2014, PloS one.

[16]  Andrew B. Lawson,et al.  Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology , 2008 .