Evaluating COVID-19 vaccine allocation policies using Bayesian $m$-top exploration (preprint)

Individual-based epidemiological models support the study of fine-grained preventive measures, such as tailored vaccine allocation policies, in silico. As individual-based models are computationally intensive, it is pivotal to identify optimal strategies within a reasonable computational budget. Moreover, due to the high societal impact associated with the implementation of preventive strategies, uncertainty regarding decisions should be communicated to policy makers, which is naturally embedded in a Bayesian approach. We present a novel technique for evaluating vaccine allocation strategies using a multi-armed bandit framework in combination with a Bayesian anytime $m$-top exploration algorithm. $m$-top exploration allows the algorithm to learn $m$ policies for which it expects the highest utility, enabling experts to inspect this small set of alternative strategies, along with their quantified uncertainty. The anytime component provides policy advisors with flexibility regarding the computation time and the desired confidence, which is important as it is difficult to make this trade-off beforehand. We consider the Belgian COVID-19 epidemic using the individual-based model STRIDE, where we learn a set of vaccination policies that minimize the number of infections and hospitalisations. Through experiments we show that our method can efficiently identify the $m$-top policies, which is validated in a scenario where the ground truth is available. Finally, we explore how vaccination policies can best be organised under different contact reduction schemes. Through these experiments, we show that the top policies follow a clear trend regarding the prioritised age groups and assigned vaccine type, which provides insights for future vaccination campaigns.

Key performance indicators of COVID-19 contact tracing in Belgium from September 2020 to December 2021 (preprint)

Background Contact tracing aims to prevent onward transmission of infectious diseases and data obtained during tracing provide unique information on transmission characteristics. A key performance indicator that has been proposed to evaluate contact tracing is the proportion of cases arising from known contacts. However, few empirical studies have investigated the effectiveness of contact tracing. Methods Using data collected between September 2020 and December 2021 in Belgium, we investigated the impact of contact tracing on SARS-CoV-2 transmission. We compared confirmed cases that were previously identified as a close contact to those that were not yet known, in terms of their traced contacts and secondary cases as well as the serial interval. In addition, we established contact and transmission patterns by age. Findings Previously traced, hence 'known', cases comprised 20% of all cases and they were linked to relatively fewer close contacts as well as fewer secondary cases and a lower secondary attack rate compared to cases that were not already known. In addition we observed a shorter serial interval for 'known' cases. There was a relative increase in transmission from children to adults during circulation of the Delta and Omicron variants, without an increase in the extent of contact between these age groups. Interpretation These results suggest that contact tracing in Belgium has been effective in reducing onward transmission and that individuals aware of their exposure to SARSCoV- 2 seemed more reserved in their social contact behaviour. Data from a reference period or region are needed to measure the impact of contact tracing in terms of the number of cases and deaths averted.

Exploring the Pareto front of multi-objective COVID-19 mitigation policies using reinforcement learning (preprint)

Infectious disease outbreaks can have a disruptive impact on public health and societal processes. As decision making in the context of epidemic mitigation is hard, reinforcement learning provides a methodology to automatically learn prevention strategies in combination with complex epidemic models. Current research focuses on optimizing policies w.r.t. a single objective, such as the pathogen's attack rate. However, as the mitigation of epidemics involves distinct, and possibly conflicting criteria (i.a., prevalence, mortality, morbidity, cost), a multi-objective approach is warranted to learn balanced policies. To lift this decision-making process to real-world epidemic models, we apply deep multi-objective reinforcement learning and build upon a state-of-the-art algorithm, Pareto Conditioned Networks (PCN), to learn a set of solutions that approximates the Pareto front of the decision problem. We consider the first wave of the Belgian COVID-19 epidemic, which was mitigated by a lockdown, and study different deconfinement strategies, aiming to minimize both COVID-19 cases (i.e., infections and hospitalizations) and the societal burden that is induced by the applied mitigation measures. We contribute a multi-objective Markov decision process that encapsulates the stochastic compartment model that was used to inform policy makers during the COVID-19 epidemic. As these social mitigation measures are implemented in a continuous action space that modulates the contact matrix of the age-structured epidemic model, we extend PCN to this setting. We evaluate the solution returned by PCN, and observe that it correctly learns to reduce the social burden whenever the hospitalization rates are sufficiently low. In this work, we thus show that multi-objective reinforcement learning is attainable in complex epidemiological models and provides essential insights to balance complex mitigation policies.

Different forms of superspreading lead to different outcomes: heterogeneity in infectiousness and contact behavior relevant for the case of SARS-CoV-2 (preprint)

Superspreading events play an important role in the spread of SARS-CoV-2 and several other pathogens. Hence, while the basic reproduction number of the original Wuhan SARS-CoV-2 is estimated to be about 3 for Belgium, there is substantial inter-individual variation in the number of secondary cases each infected individual causes. Multiple factors contribute to the occurrence of superspreading events: heterogeneity in infectiousness and susceptibility, variations in contact behavior, and the environment in which transmission takes place. While superspreading has been included in several infectious disease transmission models, our understanding of the effect that these different forms of superspreading have on the spread of pathogens and the effectiveness of control measures remains limited. To disentangle the effects of infectiousness-related heterogeneity on the one hand and contact-related heterogeneity on the other, we implemented both forms of superspreading in an individual-based model describing the transmission and spread of SARS-CoV-2 in the Belgian population. We considered its impact on viral spread as well as on the effectiveness of social distancing. We found that the effects of superspreading driven by heterogeneity in infectiousness are very different from the effects of superspreading driven by heterogeneity in contact behavior. On the one hand, a higher level of infectiousness-related heterogeneity results in less outbreaks occurring following the introduction of one infected individual. Outbreaks were also slower, with a lower peak which occurred at a later point in time, and a lower herd immunity threshold. Finally, the risk of resurgence of an outbreak following a period of lockdown decreased. On the other hand, when contact-related heterogeneity was high, this also led to smaller final sizes, but caused outbreaks to be more explosive in regard to other aspects (such as higher peaks which occurred earlier, and a higher herd immunity threshold). Finally, the risk of resurgence of an outbreak following a period of lockdown increased. Determining the contribution of both source of heterogeneity is therefore important but left to be explored further. Author summaryTo investigate the effect of different sources of superspreading on disease dynamics, we implemented superspreading driven by heterogeneity in infectiousness and heterogeneity in contact behavior into an individual-based model for the transmission of SARS-CoV-2 in the Belgian population. We compared the impact of both forms of superspreading in a scenario without interventions as well as in a scenario in which a period of strict social distancing (i.e. a lockdown) is followed by a period of partial release. We found that both forms of superspreading have very different effects. On the one hand, increasing the level of infectiousness-related heterogeneity led to less outbreaks being observed following the introduction of one infected individual in the population. Furthermore, final outbreak sizes decreased, and outbreaks became slower, with lower and later peaks, and a lower herd immunity threshold. Finally, the risk for resurgence of an outbreak following a period of lockdown also decreased. On the other hand, when contact-related heterogeneity was high, this also led to smaller final sizes, but caused outbreaks to be more explosive regarding other aspects (such as higher peaks that occurred earlier). The herd immunity threshold also increased, as did the risk of resurgence of outbreaks.

Controlling SARS-CoV-2 in schools using repetitive testing strategies (preprint)

SARS-CoV-2 remains a worldwide emergency. While vaccines have been approved and are widely administered, these are only available to adults and adolescents in Europe. Therefore, in order to mitigate the spread of more transmissible SARS-CoV-2 variants among children, the use of non-pharmaceutical interventions is still warranted. We investigate the impact of different testing strategies on the SARS-CoV-2 infection dynamics in a primary school environment, using an individual-based modelling approach. Specifically, we consider three testing strategies: 1) symptomatic isolation, where we test symptomatic individuals and isolate them when they test positive, 2) reactive screening, where a class is screened once one symptomatic individual was identified, and 3) repetitive screening, where the school in its entirety is screened on regular time intervals. Through this analysis, we demonstrate that repetitive testing strategies can significantly reduce the attack rate in schools, contrary to a reactive screening approach. Furthermore, we investigate the impact of these testing strategies on the average number of school days lost per child.

The influence of risk perceptions on close contact frequency during the SARS-CoV-2 pandemic (preprint)

Human behaviour is known to be crucial in the propagation of infectious diseases through respiratory or close-contact routes like the current SARS-CoV-2 virus. Intervention measures implemented to curb the spread of the virus mainly aim at limiting the number of close contacts, until vaccine roll-out is complete. Our main objective was to assess the relationships between SARS-CoV-2 perceptions and social contact behaviour in Belgium. Understanding these relationships is crucial to maximize interventions' effectiveness, e.g. by tailoring public health communication campaigns. In this study, we surveyed a representative sample of adults in Belgium in two longitudinal surveys (8 waves of survey 1 in April 2020 to August 2020, and 11 waves of survey 2 in November 2020 to April 2021). Generalized linear mixed effects models were used to analyse the two surveys. Participants with low and neutral perceptions on perceived severity made a significantly higher number of social contacts as compared to participants with high levels of perceived severity after controlling for other variables. Furthermore, participants with higher levels of perceived effectiveness of measures and perceived adherence to measures made fewer contacts. However, the differences were small. Our results highlight the key role of perceived severity on social contact behaviour during a pandemic. Nevertheless, additional research is required to investigate the impact of public health communication on severity of COVID-19 in terms of changes in social contact behaviour.

Inferring age-specific differences in susceptibility to and infectiousness upon SARS-CoV-2 infection based on Belgian social contact data (preprint)

Several important aspects related to SARS-CoV-2 transmission are not well known due to a lack of appropriate data. However, mathematical and computational tools can be used to extract part of this information from the available data, like some hidden age-related characteristics. In this paper, we investigate age-specific differences in susceptibility to and infectiousness upon contracting SARS-CoV-2 infection. More specifically, we use panel-based social contact data from diary-based surveys conducted in Belgium combined with the next generation principle to infer the relative incidence and we compare this to real-life incidence data. Comparing these two allows for the estimation of age-specific transmission parameters. Our analysis implies the susceptibility in children to be around half of the susceptibility in adults, and even lower for very young children (preschooler). However, the probability of adults and the elderly to contract the infection is decreasing throughout the vaccination campaign, thereby modifying the picture over time.

Assessing the feasibility and effectiveness of household-pooled universal testing to control COVID-19 epidemics (preprint)

Current outbreaks of SARS-CoV-2 are threatening the health care systems of several countries around the world. The control of SARS-CoV-2 epidemics currently relies on non-pharmaceutical interventions, such as social distancing, teleworking, mouth masks and contact tracing. However, as pre-symptomatic transmission remains an important driver of the epidemic, contact tracing efforts struggle to fully control SARS-CoV-2 epidemics. Therefore, in this work, we investigate to what extent the use of universal testing, i.e., an approach in which we screen the entire population, can be utilized to mitigate this epidemic. To this end, we rely on PCR test pooling of individuals that belong to the same households, to allow for a universal testing procedure that is feasible with the current testing capacity. We evaluate two isolation strategies: on the one hand pool isolation, where we isolate all individuals that belong to a positive PCR test pool, and on the other hand individual isolation, where we determine which of the individuals that belong to the positive PCR pool are positive, through an additional testing step. We evaluate this universal testing approach in the STRIDE individual-based epidemiological model in the context of the Belgian COVID-19 epidemic. As the organisation of universal testing will be challenging, we discuss the different aspects related to sample extraction and PCR testing, to demonstrate the feasibility of universal testing when a decentralized testing approach is used. We show through simulation, that weekly universal testing is able to control the epidemic, even when many of the contact reductions are relieved. Finally, our model shows that the use of universal testing in combination with stringent contact reductions could be considered as a strategy to eradicate the virus.

CoMix: comparing mixing patterns in the Belgian population during and after lockdown (preprint)

Background The COVID-19 pandemic has shown how a newly emergent communicable disease can lay considerable burden on public health. To avoid system collapse, governments have resorted to several social distancing measures. In Belgium, this included a lockdown and a following period of phased re-opening. Methods A representative sample of Belgian adults was asked about their contact behaviour from mid-April to mid-July, during different stages of the intervention measures in Belgium. Use of personal protection equipment (face masks) and compliance to hygienic measures was also reported. We estimated the expected reproduction number computing the ratio of R 0 with respect to pre-pandemic data. Findings During the first two waves (the first month) of the survey, the reduction in the average number of contacts was around 80% and was quite consistent across all age-classes. The average number of contacts increased over time, particularly for the younger age classes, still remaining significantly lower than pre-pandemic values. Since the end of May, the estimated reproduction number has a median value larger than one, although with a wide dispersion. Conclusions We have shown how a rapidly deployed survey can measure compliance to social distancing and assess its impact on COVID-19 spread. Monitoring the effectiveness of social distancing recommendations is of paramount importance to avoid further waves of COVID-19.

A data-driven metapopulation model for the Belgian COVID-19 epidemic: assessing the impact of lockdown and exit strategies (preprint)

BackgroundIn response to the ongoing COVID-19 pandemic, several countries adopted measures of social distancing to a different degree. For many countries, after successfully curbing the initial wave, lockdown measures were gradually lifted. In Belgium, such relief started on May 4th with phase 1, followed by several subsequent phases over the next few weeks. MethodsWe analysed the expected impact of relaxing stringent lockdown measures taken according to the phased Belgian exit strategy. We developed a stochastic, data-informed, meta-population model that accounts for mixing and mobility of the age-structured population of Belgium. The model is calibrated to daily hospitalization data and serological data and is able to reproduce the outbreak at the national level. We consider different scenarios for relieving the lockdown, quantified in terms of relative reductions in pre-pandemic social mixing and mobility. We validate our assumptions by making comparisons with social contact data collected during and after the lockdown. ResultsOur model is able to successfully describe the initial wave of COVID-19 in Belgium and identifies interactions during leisure/other activities as pivotal in the exit strategy. Indeed, we find a smaller impact of school re-openings as compared to restarting leisure activities and re-openings of work places. We also assess the impact of case isolation of new (suspected) infections, and find that it allows re-establishing relatively more social interactions while still ensuring epidemic control. Scenarios predicting a second wave of hospitalizations were not observed, suggesting that the per-contact probability of infection has changed with respect to the pre-lockdown period. ConclusionsCommunity contacts are found to be most influential, followed by professional contacts and school contacts, respectively, for an impending second wave of COVID-19. Regular re-assessment is crucial to adjust to evolving behavioral changes that can affect epidemic diffusion. In addition to social distancing, sufficient capacity for extensive testing and contact tracing is essential for successful mitigation.

The impact of contact tracing and household bubbles on deconfinement strategies for COVID-19: an individual-based modelling study (preprint)

Background. The rising COVID-19 pandemic caused many governments to impose policies restricting social interactions. These policies have slowed down the spread of the SARS-CoV-2 virus to the extent that restric- tions can be gradually lifted. Models can be useful to assess the consequences of deconfinement strategies with respect to business, school and leisure activities. Methods. We adapted the individual-based model "STRIDE" to simulate interactions between the 11 million inhabitants of Belgium at the levels of households, workplaces, schools and communities. We calibrated our model to observed hospital incidence and seroprevalence data. STRIDE can explore contact tracing options and account for repetitive leisure contacts in extended household settings (so called "household bubbles") with varying levels of connectivity. Findings. Household bubbles have the potential to reduce the number of COVID-19 hospital admissions by up to 90%. The effectiveness of contact tracing depends on its timing, as it becomes futile more than 4 days after the index case developed symptoms. Assuming that children have a lower level of susceptibility and lower probability to experience symptomatic SARS-CoV-2 infection, (partial) school closure options have relatively little impact on COVID-19 burden. Interpretation. Not only the absolute number and intensity of physical contacts drive the transmission dynamics and COVID-19 burden, also their repetitiveness is influential. Contact tracing seems essential for a controlled and persistent release of lockdown measures, but requires timely compliance to testing, reporting and self-isolation. Rapid tracing and testing, and communication ensuring continued involvement of the population are therefore essential.

Modeling the early phase of the Belgian COVID-19 epidemic using a stochastic compartmental model and studying its implied future trajectories (preprint)

Following the onset of the ongoing COVID-19 pandemic throughout the world, a large fraction of the global population is or has been under strict measures of physical distancing and quarantine, with many countries being in partial or full lockdown. These measures are imposed in order to reduce the spread of the disease and to lift the pressure on healthcare systems. Estimating the impact of such interventions as well as monitoring the gradual relaxing of these stringent measures is quintessential to understand how resurgence of the COVID-19 epidemic can be controlled for in the future. In this paper we use a stochastic age-structured discrete time compartmental model to describe the transmission of COVID-19 in Belgium. Our model explicitly accounts for age-structure by integrating data on social contacts to (i) assess the impact of the lockdown as implemented on March 13, 2020 on the number of new hospitalizations in Belgium; (ii) conduct a scenario analysis estimating the impact of possible exit strategies on potential future COVID-19 waves. More specifically, the aforementioned model is fitted to hospital admission data, data on the daily number of COVID-19 deaths and serial serological survey data informing the (sero)prevalence of the disease in the population while relying on a Bayesian MCMC approach. Our age-structured stochastic model describes the observed outbreak data well, both in terms of hospitalizations as well as COVID-19 related deaths in the Belgian population. Despite an extensive exploration of various projections for the future course of the epidemic, based on the impact of adherence to measures of physical distancing and a potential increase in contacts as a result of the relaxation of the stringent lockdown measures, a lot of uncertainty remains about the evolution of the epidemic in the next months.

Belgian Covid-19 Mortality, Excess Deaths, Number of Deaths per Million, and Infection Fatality Rates (8 March - 9 May 2020) (preprint)

Objective. Scrutiny of COVID-19 mortality in Belgium over the period 8 March-9 May 2020 (Weeks 11-19), using number of deaths per million, infection fatality rates, and the relation between COVID-19 mortality and excess death rates. Data. Publicly available COVID-19 mortality (2020); overall mortality (2009-2020) data in Belgium and demographic data on the Belgian population; data on the nursing home population; results of repeated sero-prevalence surveys in March-April 2020. Statistical methods. Reweighing, missing-data handling, rate estimation, visualization. Results. Belgium has virtually no discrepancy between COVID-19 reported mortality (confirmed and possible cases) and excess mortality. There is a sharp excess death peak over the study period; the total number of excess deaths makes April 2020 the deadliest month of April since WWII, with excess deaths far larger than in early 2017 or 2018, even though influenza-induced January 1951 and February 1960 number of excess deaths were similar in magnitude. Using various sero-prevalence estimates, infection fatality rates (IFRs; fraction of deaths among infected cases) are estimated at 0.38-0.73% for males and 0.20-0.39% for females in the non-nursing home population (non-NHP), and at 0.79-1.52% for males and 0.88-1.31% for females in the entire population. Estimates for the NHP range from 38 to 73% for males and over 22 to 37% for females. The IFRs rise from nearly 0% under 45 years, to 4.3% and 13.2% for males in the non-NHP and the general population, respectively, and to 1.5% and 11.1% for females in the non-NHP and general population, respectively. The IFR and number of deaths per million is strongly influenced by extensive reporting and the fact that 66.0% of the deaths concerned NH residents. At 764 (our re-estimation of the figure 735, presented by "Our World in Data"), the number of COVID-19 deaths per million led the international ranking on May 9, 2020, but drops to 262 in the non-NHP. The NHP is very specific: age-related increased risk; highly prevalent comorbidities that, while non-fatal in themselves, exacerbate COVID-19; larger collective households that share inadvertent vectors such as caregivers and favor clustered outbreaks; initial lack of protective equipment, etc. High-quality health care countries have a relatively older but also more frail population [1], which is likely to contribute to this result.

Intervention strategies against COVID-19 and their estimated impact on Swedish healthcare capacity (preprint)

Objectives: During March 2020, the COVID-19 pandemic has rapidly spread globally, and non-pharmaceutical interventions are being used to reduce both the load on the healthcare system as well as overall mortality. Design: Individual-based transmission modelling using Swedish demographic and Geographical Information System data and conservative COVID-19 epidemiological parameters. Setting: Sweden Participants: A model to simulate all 10.09 million Swedish residents. Interventions: 5 different non-pharmaceutical public-health interventions including the mitigation strategy of the Swedish government as of 10 April; isolation of the entire household of confirmed cases; closure of schools and non-essential businesses with or without strict social distancing; and strict social distancing with closure of schools and non-essential businesses. Main outcome measures: Estimated acute care and intensive care hospitalisations, COVID-19 attributable deaths, and infections among healthcare workers from 10 April until 29 June. Findings: Our model for Sweden shows that, under conservative epidemiological parameter estimates, the current Swedish public-health strategy will result in a peak intensive-care load in May that exceeds pre-pandemic capacity by over 40-fold, with a median mortality of 96,000 (95% CI 52,000 to 183,000). The most stringent public-health measures examined are predicted to reduce mortality by approximately three-fold. Intensive-care load at the peak could be reduced by over two-fold with a shorter period at peak pandemic capacity. Conclusions: Our results predict that, under conservative epidemiological parameter estimates, current measures in Sweden will result in at least 40-fold over-subscription of pre-pandemic Swedish intensive care capacity, with 15.8 percent of Swedish healthcare workers unable to work at the pandemic peak. Modifications to ICU admission criteria from international norms would further increase mortality.

SOCRATES: An online tool leveraging a social contact data sharing initiative to assess mitigation strategies for COVID-19 (preprint)

Objective: Establishing a social contact data sharing initiative and an interactive tool to assess mitigation strategies for COVID-19. Results: We organized data sharing of published social contact surveys via online repositories and formatting guidelines. We analyzed this social contact data in terms of weighted social contact matrices, next generation matrices, relative incidence and R0. We incorporated location-specific isolation measures (e.g. school closure or telework) and capture their effect on transmission dynamics. All methods have been implemented in an online application based on R Shiny and applied to COVID-19 with age-specific susceptibility and infectiousness. Using our online tool with the available social contact data, we illustrate that social distancing could have a considerable impact on reducing transmission for COVID-19. The effect itself depends on assumptions made about disease-specific characteristics and the choice of intervention(s). Keywords: social contact data, user interface, transmission dynamics, infectious diseases, epidemics, social distancing, behavioral changes, data sharing initiative, open-source, COVID-19