The BC COVID-19 Modelling Group works on rapid response modelling of the COVID-19 pandemic, with a special focus on British Columbia and Canada.

The interdisciplinary group, working independently from Government, includes experts in epidemiology, mathematics, and data analysis from UBC, SFU, UVic, and the private sector, with support from the Pacific Institute for the Mathematical Sciences.

Short-term forecasts of traditional streams from public health reporting (such as cases, hospitalizations, and deaths) are a key input to public health decision-making during a pandemic. Since early 2020, our research group has worked with data partners to collect, curate, and make publicly available numerous real-time COVID-19 indicators, providing multiple views of pandemic activity in the United States. This paper studies the utility of five such indicators—derived from deidentified medical insurance claims, self-reported symptoms from online surveys, and COVID-related Google search activity—from a forecasting perspective. For each indicator, we ask whether its inclusion in an autoregressive (AR) model leads to improved predictive accuracy relative to the same model excluding it. Such an AR model, without external features, is already competitive with many top COVID-19 forecasting models in use today. Our analysis reveals that 1) inclusion of each of these five indicators improves on the overall predictive accuracy of the AR model; 2) predictive gains are in general most pronounced during times in which COVID cases are trending in “flat” or “down” directions; and 3) one indicator, based on Google searches, seems to be particularly helpful during “up” trends.

Estimates of the basic reproduction number (R0) for COVID-19 are particularly variable in the context of transmission within locations such as long-term healthcare (LTHC) facilities. We sought to characterize the heterogeneity of R0 across known outbreaks within these facilities. We used a unique comprehensive dataset of all outbreaks that occurred within LTHC facilities in British Columbia, Canada as of 21 September 2020. We estimated R0 in 18 LTHC outbreaks with a novel Bayesian hierarchical dynamic model of susceptible, exposed, infected and recovered individuals, incorporating heterogeneity of R0 between facilities. We further compared these estimates to those obtained with standard methods that use the exponential growth rate and maximum likelihood. The total size of outbreaks varied dramatically, with range of attack rates 2%–86%. The Bayesian analysis provided an overall estimate of R0 = 2.51 (90% credible interval 0.47–9.0), with individual facility estimates ranging between 0.56 and 9.17. Uncertainty in these estimates was more constrained than standard methods, particularly for smaller outbreaks informed by the population-level model. We further estimated that intervention led to 61% (52%–69%) of all potential cases being averted within the LTHC facilities, or 75% (68%–79%) when using a model with multi-level intervention effect. Understanding of transmission risks and impact of intervention are essential in planning during the ongoing global pandemic, particularly in high-risk environments such as LTHC facilities.

Importance: An effective vaccine against SARS-CoV-2 will reduce morbidity and mortality and allow substantial relaxation of physical distancing policies. However, the ability of a vaccine to prevent infection or disease depends critically on protecting older individuals, who are at highest risk of severe disease.

Objective: We quantitatively estimated the relative benefits of COVID-19 vaccines, in terms of preventing infection and death, with a particular focus on effectiveness in elderly people.

Design: We applied compartmental mathematical modelling to determine the relative effects of vaccines that block infection and onward transmission, and those that prevent severe disease. We assumed that vaccines showing high efficacy in adults would be deployed, and examined the effects of lower vaccine efficacy among the elderly population.

Setting and participants: Our mathematical model was calibrated to simulate the course of an epidemic among the entire population of British Columbia, Canada. Within our model, the population was structured by age and levels of contact.

Main outcome(s) and measure(s): We assessed the effectiveness of possible vaccines in terms of the predicted number of infections within the entire population, and deaths among people aged 65 years and over.

Results: In order to reduce the overall rate of infections in the population, high rates of deployment to all age groups will be critical. However, to substantially reduce mortality among people aged 65 years and over, a vaccine must directly protect a high proportion of people in that group.

Conclusions and relevance: Effective vaccines deployed to a large fraction of the population are projected to substantially reduce infection in an otherwise susceptible population. However, even if transmission were blocked highly effectively by vaccination of children and younger adults, overall mortality would not be substantially reduced unless the vaccine is also directly protective in elderly people. We strongly recommend: (i) the inclusion of people aged 65 years and over in future trials of COVID-19 vaccine candidates; (ii) careful monitoring of vaccine efficacy in older age groups following vaccination.

Serology tests for SARS-CoV-2 provide a paradigm for estimating the number of individuals who have had infection in the past (including cases that are not detected by routine testing, which has varied over the course of the pandemic and between jurisdictions). Classical statistical approaches to such estimation do not incorporate case counts over time, and may be inaccurate due to uncertainty about the sensitivity and specificity of the serology test. In this work, we provide a joint Bayesian model for case counts and serological data, integrating uncertainty through priors on the sensitivity and specificity. We also model the Phases of the pandemic with exponential growth and decay. This model improves upon maximum likelihood estimates by conditioning on more data, and by taking into account the epidemiological trajectory. We apply our model to the greater Vancouver area, British Columbia, Canada with data acquired during Phase 1 of the pandemic.

South Africa is currently experiencing a second wave of resurgence in COVID-19 infection. In this modelling study, we use a Bayesian compartmental model to project possible spread of the second wave of COVID-19 in South Africa under various levels of lockdown restrictions. Our model suggests that strict lockdown restrictions will have to be in place up to the end of March 2021 before cases can drop to levels observed, in September to early November 2020, after the first wave. On the one hand, extended lockdown restrictions have negative consequences – albeit effective, they are not sustainable over extended periods. On the other hand, short lockdown restrictions over a few weeks will not have a lasting effect on the spread of the disease. Lockdown restrictions need to be supplemented with increased rapid testing, palliative support for the vulnerable, and implementations of other non-pharmaceutical interventions (NPIs) such as mask mandate. These multifaceted approaches could help keep cases under control until vaccines are widely available.

Under the implementation of non-pharmaceutical interventions such as social distancing and lockdowns, household transmission has been shown to be significant for COVID-19, posing challenges for reducing incidence in settings where people are asked to self-isolate at home and to spend increasing amounts of time at home due to distancing measures. Accordingly, characteristics of households in a region have been shown to relate to transmission heterogeneity of the virus. We introduce a stochastic epidemiological model to examine the impact of the household size distribution in a region on the transmission dynamics. We choose parameters to reflect incidence in two health regions of the Greater Vancouver area in British Columbia and simulate the impact of distancing measures on transmission, with household size distribution the only different parameter between simulations for the two regions. Our result suggests that the dissimilarity in household size distribution alone can cause significant differences in incidence of the two regions, and the distributions drive distinct dynamics that match reported cases. Furthermore, our model suggests that offering individuals a place to isolate outside their household can speed the decline in cases, and does so more effectively where there are more larger households.

Contact tracing has played a central role in COVID-19 control in many jurisdictions and is often used in conjunction with other measures such as travel restrictions and social distancing mandates. Contact tracing is made ineffective, however, by delays in testing, calling, and isolating. Even if delays are minimized, contact tracing triggered by testing of symptomatic individuals can only prevent a fraction of onward transmissions from contacts. Without other measures in place, contact tracing alone is insufficient to prevent exponential growth in the number of cases in a population with little immunity. Even when used effectively with other measures, occasional bursts in call loads can overwhelm contact tracing systems and lead to a loss of control. We propose embracing approaches to COVID-19 contact tracing that broadly test individuals without symptoms, in whatever way is economically feasible—either with fast and cheap tests that can be deployed widely, with pooled testing, or with screening of judiciously chosen groups of high-risk individuals. These considerations are important both in regions where widespread vaccination has been deployed and in those where few residents have been immunized.

Background: Antibodies to Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) have been shown to neutralize the virus in-vitro. Similarly, animal challenge models suggest that neutralizing antibodies isolated from SARS-CoV-2 infected individuals prevent against disease upon re-exposure to the virus. Understanding the nature and duration of the antibody response following SARS-CoV-2 infection is therefore critically important.

Methods: Between April and October 2020 we undertook a prospective cohort study of 3555 healthcare workers in order to elucidate the duration and dynamics of antibody responses following infection with SARS-CoV-2. After a formal performance evaluation against 169 PCR confirmed cases and negative controls, the Meso-Scale Discovery assay was used to quantify in parallel, antibody titers to the SARS-CoV-2 nucleoprotein (N), spike (S) protein and the receptor-binding-domain (RBD) of the S-protein. All seropositive participants were followed up monthly for a maximum of 7 months; those participants that were symptomatic, with known dates of symptom-onset, seropositive by the MSD assay and who provided 2 or more monthly samples were included in the analysis. Survival analysis was used to determine the proportion of sero-reversion (switching from positive to negative) from the raw data. In order to predict long-term antibody dynamics, two hierarchical longitudinal Gamma models were implemented to provide predictions for the lower bound (continuous antibody decay to zero, “Gamma-decay”) and upper bound (decay-to-plateau due to long lived plasma cells, “Gamma-plateau”) long-term antibody titers.

Results: A total of 1163 samples were provided from 349 of 3555 recruited participants who were symptomatic, seropositive by the MSD assay, and were followed up with 2 or more monthly samples. At 200 days post symptom onset, 99% of participants had detectable S-antibody whereas only 75% of participants had detectable N-antibody. Even under our most pessimistic assumption of persistent negative exponential decay, the S-antibody was predicted to remain detectable in 95% of participants until 465 days [95% CI 370-575] after symptom onset. Under the Gamma-plateau model, the entire posterior distribution of S-antibody titers at plateau remained above the threshold for detection indefinitely. Surrogate neutralization assays demonstrated a strong positive correlation between antibody titers to the S-protein and blocking of the ACE-2 receptor in-vitro [R2=0.72, p<0.001]. By contrast, the N-antibody waned rapidly with a half-life of 60 days [95% CI 52-68].

Discussion: This study has demonstrated persistence of the spike antibody in 99% of participants at 200 days following SARS-CoV-2 symptoms and rapid decay of the nucleoprotein antibody. Diagnostic tests or studies that rely on the N-antibody as a measure of seroprevalence must be interpreted with caution. Our lowest bound prediction for duration of the spike antibody was 465 days and our upper bound predicted spike antibody to remain indefinitely in line with the long-term seropositivity reported for SARS-CoV infection. The long-term persistence of the S-antibody, together with the strong positive correlation between the S-antibody and viral surrogate neutralization in-vitro, has important implications for the duration of functional immunity following SARS-CoV-2 infection.

Widespread school closures occurred during the COVID-19 pandemic. Because closures are costly and damaging, many jurisdictions have since reopened schools with control measures in place. Early evidence indicated that schools were low risk and children were unlikely to be very infectious, but it is becoming clear that children and youth can acquire and transmit COVID-19 in school settings and that transmission clusters and outbreaks can be large. We describe the contrasting literature on school transmission, and argue that the apparent discrepancy can be reconciled by heterogeneity, or “overdispersion” in transmission, with many exposures yielding little to no risk of onward transmission, but some unfortunate exposures causing sizeable onward transmission. In addition, respiratory viral loads are as high in children and youth as in adults, pre- and asymptomatic transmission occur, and the possibility of aerosol transmission has been established. We use a stochastic individual-based model to find the implications of these combined observations for cluster sizes and control measures. We consider both individual and environment/activity contributions to the transmission rate, as both are known to contribute to variability in transmission. We find that even small heterogeneities in these contributions result in highly variable transmission cluster sizes in the classroom setting, with clusters ranging from 1 to 20 individuals in a class of 25. None of the mitigation protocols we modeled, initiated by a positive test in a symptomatic individual, are able to prevent large transmission clusters unless the transmission rate is low (in which case large clusters do not occur in any case). Among the measures we modeled, only rapid universal monitoring (for example by regular, onsite, pooled testing) accomplished this prevention. We suggest approaches and the rationale for mitigating these larger clusters, even if they are expected to be rare.

COVID-19 is a global pandemic with over 25 million cases worldwide. Currently, treatments are limited, and there is no approved vaccine. Interventions such as handwashing, masks, social distancing, and “social bubbles” are used to limit community transmission, but it is challenging to choose the best interventions for a given activity. Here, we provide a quantitative framework to determine which interventions are likely to have the most impact in which settings. We introduce the concept of “event R,” the expected number of new infections due to the presence of a single infectious individual at an event. We obtain a fundamental relationship between event R and four parameters: transmission intensity, duration of exposure, the proximity of individuals, and the degree of mixing. We use reports of small outbreaks to establish event R and transmission intensity in a range of settings. We identify principles that guide whether physical distancing, masks and other barriers to transmission, or social bubbles will be most effective. We outline how this information can be obtained and used to reopen economies with principled measures to reduce COVID-19 transmission.

Introduction Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) specific antibodies have been shown to neutralize the virus in-vitro. Understanding antibody dynamics following SARS-CoV-2 infection is therefore crucial. Sensitive measurement of SARS-CoV-2 antibodies is also vital for large seroprevalence surveys which inform government policies and public health interventions. However, rapidly waning antibodies following SARS-CoV-2 infection could jeopardize the sensitivity of serological testing on which these surveys depend.Methods This prospective cohort study of SARS-CoV-2 humoral dynamics in a central London hospital analyzed 137 serial samples collected from 67 participants seropositive to SARS-CoV-2 by the Meso-Scale Discovery assay. Antibody titers were quantified to the SARS-CoV-2 nucleoprotein (N), spike (S-)protein and the receptor-binding-domain (RBD) of the S-protein. Titers were log-transformed and a multivariate log-linear model with time-since-infection and clinical variables was fitted by Bayesian methods.Results The mean estimated half-life of the N-antibody was 52 days (95% CI 42-65). The S- and RBD-antibody had significantly longer mean half-lives of 81 days (95% CI 61-111) and 83 days (95% CI 55-137) respectively. An ACE-2-receptor competition assay demonstrated significant correlation between the S and RBD-antibody titers and ACE2-receptor blocking in-vitro. The time-to-a-negative N-antibody test for 50% of the seropositive population was predicted to be 195 days (95% CI 163-236).Discussion After SARS-CoV-2 infection, the predicted half-life of N-antibody was 52 days with 50% of seropositive participants becoming seronegative to this antibody at 195 days. Widely used serological tests that depend on the N-antibody will therefore significantly underestimate the prevalence of infection following the majority of infections.Significance statement We believe that our study has significant and urgent public health and translational impact. Firstly, our findings demonstrate that the half-life of the SARS-CoV-2 nucleoprotein antibody is only 52 days. This has immediate and important implications for large-scale seroprevalence surveys, government policy and mathematical modelling predictions which rely on serological tests that target this antibody. Secondly, the slower decay of the SARS-CoV-2 spike protein antibody identified in this study makes assays to the spike protein a more reliable target for serological assays in the longer term. We demonstrate a strong positive linear correlation between spike/RBD antibody and ACE-2 receptor binding in vitro. Our findings are therefore likely to reflect the time to loss of a functional antibody response in SARS-CoV-2.

Since the beginning of the epidemic, daily reports of CoViD-19 cases, hospitalizations, and deaths from around the world have been publicly available. This paper describes methods to characterize broad features of the spread of the disease, with relatively long periods of constant transmission rates, using a new population modeling framework based on discrete-time difference equations. Comparative parameters are chosen for their weak dependence on model assumptions. Approaches for their point and interval estimation, accounting for additional sources of variance in the case data, are presented. These methods provide a basis to quantitatively assess the impact of changes to social distancing policies using publicly available data. As examples, data from Ontario and German states are analyzed using this framework. German case data show a small increase in transmission rates following the relaxation of lock-down rules on May 6, 2020. By combining case and death data from Germany, the mean and standard deviation of the time from infection to death are estimated.

Background: Many countries have implemented population-wide interventions such as physical distancing measures, in efforts to control COVID-19. The extent and success of such measures has varied. Many jurisdictions with declines in reported COVID-19 cases are moving to relax measures, while others are continuing to intensify efforts to reduce transmission.

Aim: We aim to determine the time frame between a change in COVID-19 measures at the population level and the observable impact of such a change on cases.

Methods: We examine how long it takes for there to be a substantial difference between the cases that occur following a change in control measures and those that would have occurred at baseline. We then examine how long it takes to detect a difference, given delays and noise in reported cases. We use changes in population-level (e.g., distancing) control measures informed by data and estimates from British Columbia, Canada.

Results: We find that the time frames are long: it takes three weeks or more before we might expect a substantial difference in cases given a change in population-level COVID-19 control, and it takes slightly longer to detect the impacts of the change. The time frames are shorter (11-15 days) for dramatic changes in control, and they are impacted by noise and delays in the testing and reporting process, with delays reaching up to 25-40 days.

Conclusion: The time until a change in broad control measures has an observed impact is longer than is typically understood, and is longer than the mean incubation period (time between exposure than onset) and the often used 14 day time period. Policy makers and public health planners should consider this when assessing the impact of policy change, and efforts should be made to develop rapid, consistent real-time COVID-19 surveillance.

Following successful non-pharmaceutical interventions (NPI) aiming to control COVID-19, many jurisdictions reopened their economies and borders. As little immunity had developed in most populations, re-establishing higher contact carried substantial risks, and therefore many locations began to see resurgence in COVID-19 cases. We present a Bayesian method to estimate the leeway to reopen, or alternatively the strength of change required to re-establish COVID-19 control, in a range of jurisdictions experiencing different COVID-19 epidemics. We estimated the timing and strength of initial control measures such as widespread distancing and compared the leeway jurisdictions had to reopen immediately after NPI measures to later estimates of leeway. Finally, we quantified risks associated with reopening and the likely burden of new cases due to introductions from other jurisdictions. We found widely varying leeway to reopen. After initial NPI measures took effect, some jurisdictions had substantial leeway (e.g., Japan, New Zealand, Germany) with > 0.99 probability that contact rates were below 80% of the threshold for epidemic growth. Others had little leeway (e.g., the United Kingdom, Washington State) and some had none (e.g., Sweden, California). For most such regions, increases in contact rate of 1.5–2 fold would have had high (> 0.7) probability of exceeding past peak sizes. Most jurisdictions experienced June–August trajectories consistent with our projections of contact rate increases of 1–2-fold. Under such relaxation scenarios for some regions, we projected up to ∼100 additional cases if just one case were imported per week over six weeks, even between jurisdictions with comparable COVID-19 risk. We provide an R package covidseir to enable jurisdictions to estimate leeway and forecast cases under different future contact patterns. Estimates of leeway can establish a quantitative basis for decisions about reopening. We recommend a cautious approach to reopening economies and borders, coupled with strong monitoring for changes in transmission.

[Current Biology, 3 August 2020]

There is no doubt that the novel coronavirus SARS-CoV-2 that causes COVID-19 is mutating and thus has the potential to adapt during the current pandemic. Whether this evolution will lead to changes in the transmission, the duration, or the severity of the disease is not clear. This has led to considerable scientific and media debate, from raising alarms about evolutionary change to dismissing it. Here we review what little is currently known about the evolution of SARS-CoV-2 and extend existing evolutionary theory to consider how selection might be acting upon the virus during the COVID-19 pandemic. Although there is currently no definitive evidence that SARS-CoV-2 is undergoing further adaptation, continued evidence-based analysis of evolutionary change is important so that public health measures can be adjusted in response to substantive changes in the infectivity or severity of COVID-19.

We collated contact tracing data from COVID-19 clusters in Singapore and Tianjin, China and estimated the extent of pre-symptomatic transmission by estimating incubation periods and serial intervals. The mean incubation periods accounting for intermediate cases were 4.91 days (95%CI 4.35, 5.69) and 7.54 (95%CI 6.76, 8.56) days for Singapore and Tianjin, respectively. The mean serial interval was 4.17 (95%CI 2.44, 5.89) and 4.31 (95%CI 2.91, 5.72) days (Singapore, Tianjin). The serial intervals are shorter than incubation periods, suggesting that pre-symptomatic transmission may occur in a large proportion of transmission events (0.4–0.5 in Singapore and 0.6–0.8 in Tianjin, in our analysis with intermediate cases, and more without intermediates). Given the evidence for pre-symptomatic transmission, it is vital that even individuals who appear healthy abide by public health measures to control COVID-19.

Extensive non-pharmaceutical and physical distancing measures are currently the primary interventions against coronavirus disease 2019 (COVID-19) worldwide. It is therefore urgent to estimate the impact such measures are having. We introduce a Bayesian epidemiological model in which a proportion of individuals are willing and able to participate in distancing, with the timing of distancing measures informed by survey data on attitudes to distancing and COVID-19. We fit our model to reported COVID-19 cases in British Columbia (BC), Canada, and five other jurisdictions, using an observation model that accounts for both underestimation and the delay between symptom onset and reporting. We estimated the impact that physical distancing (social distancing) has had on the contact rate and examined the projected impact of relaxing distancing measures. We found that, as of April 11 2020, distancing had a strong impact in BC, consistent with declines in reported cases and in hospitalization and intensive care unit numbers; individuals practising physical distancing experienced approximately 0.22 (0.11–0.34 90% CI [credible interval]) of their normal contact rate. The threshold above which prevalence was expected to grow was 0.55. We define the “contact ratio” to be the ratio of the estimated contact rate to the threshold rate at which cases are expected to grow; we estimated this contact ratio to be 0.40 (0.19–0.60) in BC. We developed an R package ‘covidseir’ to make our model available, and used it to quantify the impact of distancing in five additional jurisdictions. As of May 7, 2020, we estimated that New Zealand was well below its threshold value (contact ratio of 0.22 [0.11–0.34]), New York (0.60 [0.43–0.74]), Washington (0.84 [0.79–0.90]) and Florida (0.86 [0.76–0.96]) were progressively closer to theirs yet still below, but California (1.15 [1.07–1.23]) was above its threshold overall, with cases still rising. Accordingly, we found that BC, New Zealand, and New York may have had more room to relax distancing measures than the other jurisdictions, though this would need to be done cautiously and with total case volumes in mind. Our projections indicate that intermittent distancing measures—if sufficiently strong and robustly followed—could control COVID-19 transmission. This approach provides a useful tool for jurisdictions to monitor and assess current levels of distancing relative to their threshold, which will continue to be essential through subsequent waves of this pandemic.