[medRxiv, 11 July 2022]

Even with daily case counts, the true scope of the COVID-19 pandemic in Canada is unknown due to undetected cases. We estimate the pandemic scope through a new multi-site model using publicly available disease count data including detected cases, recoveries among detected cases, and total deaths. These counts are used to estimate the case detection probability, the infection fatality rate through time, as well as the probability of recovery, and several important population parameters including the rate of spread, and importation of external cases. We also estimate the total number of active COVID-19 cases per region of Canada for each reporting interval. We applied this multi-site model Canada-wide to all provinces and territories, providing an estimate of the total COVID-19 burden for the 90 weeks from 23 Apr 2020 to 6 Jan 2022. We also applied this model to the five Health Authority regions of British Columbia, Canada, describing the pandemic in B.C. over the 31 weeks from 2 Apr 2020 to 30 Oct 2020.

[Journal of Biological Dynamics, 2 Jun 2022]

Contact tracing is an important intervention measure to control infectious diseases. We present a new approach that borrows the edge dynamics idea from network models to track contacts included in a compartmental SIR model for an epidemic spreading in a randomly mixed population. Unlike network models, our approach does not require statistical information of the contact network, data that are usually not readily available. The model resulting from this new approach allows us to study the effect of contact tracing and isolation of diagnosed patients on the control reproduction number and number of infected individuals. We estimate the effects of tracing coverage and capacity on the effectiveness of contact tracing. Our approach can be extended to more realistic models that incorporate latent and asymptomatic compartments.

[Infectious Disease Modelling, 20 June 2023]

The perceived infection risk changes individual behaviors, which further affects the disease dynamics. This perception is influenced by social communication, including surveying their social network neighbors about the fraction of infected neighbors and averaging their neighbors’ perception of the risk. We model the interaction of disease dynamics and risk perception on a two-layer random network that combines a social network layer with a contact network layer. We found that if information spreads much faster than disease, then all individuals converge on the true prevalence of the disease. On the other hand, if the two dynamics have comparable speeds, the risk perception still converges to a value uniformly on the network. However, the perception lags behind the true prevalence and has a lower peak value. We also study the behavior change caused by the perception of infection risk. This behavior change may affect the disease dynamics by reducing the transmission rate along the edges of the contact network or by breaking edges and isolating the infectious individuals. The effects on the basic reproduction number, the peak size, and the final size are studied. We found that these two effects give the same basic reproduction number. We find edge-breaking has a larger effect on reducing the final size, while reducing the transmission rate has a larger

[Mathematical Biosciences and Engineering, 19 June 2023]

We propose a new method to estimate the change of the effective reproduction number with time, due to either disease control measures or seasonally varying transmission rate. We validate our method using a simulated epidemic curve and show that our method can effectively estimate both sudden changes and gradual changes in the reproduction number. We apply our method to the COVID-19 case counts in British Columbia, Canada in 2020, and we show that strengthening control measures had a significant effect on the reproduction number, while relaxations in May (business reopening) and September (school reopening) had significantly increased the reproduction number from around 1 to around 1.7 at its peak value. Our method can be applied to other infectious diseases, such as pandemics and seasonal influenza.

[Physica D: Nonlinear Phenomena, 17 February 2022]

The basic reproduction number R0 is an important indicator of the severity for an epidemic outbreak, but it may not be obtained easily for heterogeneous populations especially with biased mixing contacts. Usually, explicit formula for the basic reproduction number in degree correlated networks is difficult to get, thus it is very useful to understand the relationship with their uncorrelated counterparts. It seems that the assortative mixing increases the basic reproduction number of susceptible–infected–recovered (SIR) epidemic, whereas disassortative mixing decreases it, or that assortative mixing enhances the percolation, whereas disassortative mixing weakens it. This result is obtained for degree correlated networks based on a small deviation from the uncorrelated networks, and may not be universal. In this paper, we show rigorously that the result is true for degree correlated networks with arbitrary bimodal degree distribution and for assortative mixing networks with arbitrary degree distribution. However, this may not be always true for general disassortative mixing networks. We also introduce a numerical algorithm to construct an arbitrary joint degree distribution and generate a counterexample of disassortative mixing network that the result fails. Finally, a sufficient condition is given to guarantee that the SIR model in disassortative mixing networks yields smaller R0.