Atlantic Canada and the territories have experienced a qualitatively different COVID-19 epidemic than the other Canadian provinces. For example, as of September 22, 2020, in Newfoundland and Labrador (NL) there have been 90 days with fewer than 1 active case per 100,000 people, while in Ontario there have been no such days since March 15. However, do these differences suggest different best approaches to public health policy? Imported infections are a large fraction of all infections when infection prevalence is low. Travel restrictions, including the Atlantic bubble, have reduced the infection importation rate to NL, and we use a continuous time branching process model to show that the implementation of NL’s travel restrictions reduced COVID-19 infections by 92%. To understand how travel restrictions might affect other aspects of public health policy, we consider 3 alert levels: strict, moderate, and relaxed, each corresponding to progressively less demanding restrictions on schools, businesses, and events. Using stochastic dynamic programming, we find that the optimal strategy for alert level implementation depends on local characteristics such as the number of ports of entry, thus suggesting qualitatively different approaches to a comprehensive public health strategy in low importation regions.

This is joint research with Drs. Amy Hurford Proton Rahman, Troy Day, Julien Arino, and Maria Martignoni.

Dr. Loredo-Osti is a Professor and chair of the department of Mathematics and Statistics at Memorial University. He completed his Ph.D. in statistics from Dalhousie University (2009), postdoctoral scholar and research associate at McGill University. In 2005 joined department of Mathematics and Statistics of Memorial University. Dr. Loredo-Osti has published papers in statistical inference, population genetics, genetic epidemiology, time series and mixed models. His interests are in population genetics and genetic epidemiology, statistical inference, applied stochastic processes, in particularly, extreme and/or rare events and alpha-stable processes. In the recent months his work has focused on stochastic processes applied to modelling of Covid-19.

Since 2014, Dr. Loredo-Osti is a SNI-Conacyt National Researcher Level II (Sistema Nacional de Investigadores, Consejo Nacional de Ciencia y Tecnologia, Mexico).

https://www.medrxiv.org/content/10.1101/2020.09.02.20186874v1