The development and deployment of a model for hospital-level COVID-19 associated patient demand intervals from consistent estimators (DICE)

Hospitals commonly project demand for their services by combining their historical share of regional demand with forecasts of total regional demand. Hospital-specific forecasts of demand that provide prediction intervals, rather than point estimates, may facilitate better managerial decisions, especially when demand overage and underage are associated with high, asymmetric costs. Regional forecasts of patient demand are commonly available as a Poisson random variable, e.g., for the number of people requiring hospitalization due to an epidemic such as COVID-19. However, even in this common setting, no probabilistic, consistent, computationally tractable forecast is available for the fraction of patients in a region that a particular institution should expect. We introduce such a forecast, DICE (Demand Intervals from Consistent Estimators). We describe its development and deployment at an academic medical center in California during the `second wave' of COVID-19 in the Unite States. We show that DICE is consistent under mild assumptions and suitable for use with perfect, biased, unbiased regional forecasts. We evaluate its performance on empirical data from a large academic medical center as well as on synthetic data.

[1]  Scott Levin,et al.  Discrete Event Simulation for Healthcare Organizations: A Tool for Decision Making , 2013, Journal of healthcare management / American College of Healthcare Executives.

[2]  R. Weron,et al.  Forecasting spot electricity prices: A comparison of parametric and semiparametric time series models , 2008 .

[3]  J. Crilly,et al.  Prediction and surveillance of influenza epidemics , 2011, The Medical journal of Australia.

[4]  Michael Y. Li,et al.  Global stability for the SEIR model in epidemiology. , 1995, Mathematical biosciences.

[5]  S. Pei,et al.  Initial Simulation of SARS-CoV2 Spread and Intervention Effects in the Continental US , 2020, medRxiv.

[6]  COVIDSurg Collaborative Elective surgery cancellations due to the COVID-19 pandemic: global predictive modelling to inform surgical recovery plans: Elective surgery during the SARS-CoV-2 pandemic , 2020 .

[7]  Pierre Lanchantin,et al.  Unsupervised restoration of hidden nonstationary Markov chains using evidential priors , 2005, IEEE Transactions on Signal Processing.

[8]  Ulrike Goldschmidt,et al.  An Introduction To The Theory Of Point Processes , 2016 .

[10]  P. Glynn,et al.  A model to forecast regional demand for COVID-19 related hospital beds , 2020, medRxiv.

[11]  D. Khullar,et al.  COVID-19 and the Financial Health of US Hospitals. , 2020, JAMA.

[12]  Ophir Frieder,et al.  Coughing, sneezing, and aching online: Twitter and the volume of influenza-like illness in a pediatric hospital , 2017, PloS one.

[13]  Mark Bartlett,et al.  Deterministic and Stochastic Models for Recurrent Epidemics , 1956 .

[14]  Naren Ramakrishnan,et al.  Collaborative efforts to forecast seasonal influenza in the United States, 2015–2016 , 2019, Scientific Reports.

[15]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[16]  Y. Gel,et al.  Influenza Forecasting with Google Flu Trends , 2013, PloS one.

[17]  Locally Informed Simulation to Predict Hospital Capacity Needs During the COVID-19 Pandemic , 2020, Annals of Internal Medicine.

[18]  Jonathan D. Cryer,et al.  Time Series Analysis , 1986 .

[19]  Craig O'Neill,et al.  Computer Modeling of Patient Flow in a Pediatric Emergency Department Using Discrete Event Simulation , 2007, Pediatric emergency care.

[20]  Chun-Liang Li,et al.  Interpretable Sequence Learning for COVID-19 Forecasting , 2020, NeurIPS.

[21]  Edward Livingston,et al.  Coronavirus Disease 2019 (COVID-19) in Italy. , 2020, JAMA.

[22]  Gary E. Weissman,et al.  Locally Informed Simulation to Predict Hospital Capacity Needs During the COVID-19 Pandemic , 2020, Annals of Internal Medicine.

[23]  Yu Wang,et al.  Curating a COVID-19 data repository and forecasting county-level death counts in the United States , 2020, Harvard Data Science Review.

[24]  W. O. Kermack,et al.  Contributions to the mathematical theory of epidemics—I , 1991, Bulletin of mathematical biology.

[25]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[26]  R. Rosenfeld,et al.  A collaborative multiyear, multimodel assessment of seasonal influenza forecasting in the United States , 2019, Proceedings of the National Academy of Sciences.

[27]  T. Anderson Statistical analysis of time series , 1974 .

[28]  Gary Alan Bass,et al.  Elective surgery cancellations due to the COVID-19 pandemic: global predictive modelling to inform surgical recovery plans , 2020, The British journal of surgery.

[29]  H. Akaike Fitting autoregressive models for prediction , 1969 .

[30]  L. Allen Some discrete-time SI, SIR, and SIS epidemic models. , 1994, Mathematical biosciences.

[31]  Guoqiang Peter Zhang,et al.  An investigation of model selection criteria for neural network time series forecasting , 2001, Eur. J. Oper. Res..

[32]  Y. Wang,et al.  Ensemble Forecasts of Coronavirus Disease 2019 (COVID-19) in the U.S. , 2020, medRxiv.

[33]  Ahmed Tealab,et al.  Time series forecasting using artificial neural networks methodologies: A systematic review , 2018, Future Computing and Informatics Journal.

[34]  R. Harrison,et al.  Estimation and forecasting hospital admissions due to Influenza: Planning for winter pressure. The case of the West Midlands, UK , 2005 .

[35]  S. T. Buckland,et al.  An Introduction to the Bootstrap. , 1994 .

[36]  Emmanuel Chazard,et al.  Real Time Influenza Monitoring Using Hospital Big Data in Combination with Machine Learning Methods: Comparison Study , 2018, JMIR public health and surveillance.

[37]  H. Akaike Fitting autoregressive models for prediction , 1969 .

[38]  L. Rabiner,et al.  An introduction to hidden Markov models , 1986, IEEE ASSP Magazine.

[39]  Hannah R. Meredith,et al.  A scenario modeling pipeline for COVID-19 emergency planning , 2020, Scientific Reports.

[40]  Alex Bateman,et al.  An introduction to hidden Markov models. , 2007, Current protocols in bioinformatics.

[41]  Jeffrey D. Scargle,et al.  An Introduction to the Theory of Point Processes, Vol. I: Elementary Theory and Methods , 2004, Technometrics.

[42]  Ozgur M. Araz,et al.  Using Google Flu Trends data in forecasting influenza-like-illness related ED visits in Omaha, Nebraska. , 2014, The American journal of emergency medicine.