Incorporating social contact data in spatio-temporal models for infectious disease spread

Transmission matrices used in epidemiologic modelling

Mixing matrices are included in infectious disease models to reflect transmission opportunities between population strata. These matrices were originally constructed on the basis of theoretical considerations and most of the early work in this area originates from research on sexually transferred diseases in the 80s, in response to AIDS. Later work in the 90s populated these matrices on the basis of survey data gathered to capture transmission risks for respiratory diseases. We provide an overview of developments in the construction of matrices for capturing transmission opportunities in populations. Such transmission matrices are useful for epidemiologic modelling to capture within and between stratum transmission and can be informed from theoretical mixing assumptions, informed by empirical evidence gathered through investigation as well as generated on the basis of data. Links to summary measures and threshold conditions are also provided.

Predicting subnational incidence of COVID-19 cases and deaths in EU countries

BACKGROUND

Recurring COVID-19 waves highlight the need for tools able to quantify transmission risk, and identify geographical areas at risk of outbreaks. Local outbreak risk depends on complex immunity patterns resulting from previous infections, vaccination, waning and immune escape, alongside other factors (population density, social contact patterns). Immunity patterns are spatially and demographically heterogeneous, and are challenging to capture in country-level forecast models.

METHODS

We used a spatiotemporal regression model to forecast subnational case and death counts and applied it to three EU countries as test cases: France, Czechia, and Italy. Cases in local regions arise from importations or local transmission. Our model produces age-stratified forecasts given age-stratified data, and links reported case counts to routinely collected covariates (e.g. test number, vaccine coverage). We assessed the predictive performance of our model up to four weeks ahead using proper scoring rules and compared it to the European COVID-19 Forecast Hub ensemble model. Using simulations, we evaluated the impact of variations in transmission on the forecasts. We developed an open-source RShiny App to visualise the forecasts and scenarios.

RESULTS

At a national level, the median relative difference between our median weekly case forecasts and the data up to four weeks ahead was 25% (IQR: 12-50%) over the prediction period. The accuracy decreased as the forecast horizon increased (on average 24% increase in the median ranked probability score per added week), while the accuracy of death forecasts was more stable. Beyond two weeks, the model generated a narrow range of likely transmission dynamics. The median national case forecasts showed similar accuracy to forecasts from the European COVID-19 Forecast Hub ensemble model, but the prediction interval was narrower in our model. Generating forecasts under alternative transmission scenarios was therefore key to capturing the range of possible short-term transmission dynamics.

DISCUSSION

Our model captures changes in local COVID-19 outbreak dynamics, and enables quantification of short-term transmission risk at a subnational level. The outputs of the model improve our ability to identify areas where outbreaks are most likely, and are available to a wide range of public health professionals through the Shiny App we developed.

Comparison of statistical approaches to predicting norovirus laboratory reports before and during COVID-19: insights to inform public health surveillance

Social distancing interrupted transmission patterns of contact-driven infectious agents such as norovirus during the Covid-19 pandemic. Since routine surveillance of norovirus was additionally disrupted during the pandemic, traditional naïve forecasts that rely only on past public health surveillance data may not reliably represent norovirus activity. This study investigates the use of statistical modelling to predict the number of norovirus laboratory reports in England 4-weeks ahead of time before and during Covid-19 pandemic thus providing insights to inform existing practices in norovirus surveillance in England. We compare the predictive performance from three forecasting approaches that assume different underlying structure of the norovirus data and utilized various external data sources including mobility, air temperature and relative internet searches (Time Series and Regularized Generalized Linear Model, and Quantile Regression Forest). The performance of each approach was evaluated using multiple metrics, including a relative prediction error against the traditional naive forecast of a five-season mean. Our data suggest that all three forecasting approaches improve predictive performance over the naïve forecasts, especially in the 2020/21 season (30-45% relative improvement) when the number of norovirus reports reduced. The improvement ranged from 7 to 22% before the pandemic. However, performance varied: regularized regression incorporating internet searches showed the best forecasting score pre-pandemic and the time series approach achieved the best results post pandemic onset without external data. Overall, our results demonstrate that there is a significant value for public health in considering the adoption of more sophisticated forecasting tools, moving beyond traditional naïve methods, and utilizing available software to enhance the precision and timeliness of norovirus surveillance in England.

The impact of national and international travel on spatio-temporal transmission of SARS-CoV-2 in Belgium in 2021

BACKGROUND

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has rapidly spread over the world and caused tremendous impacts on global health. Understanding the mechanism responsible for the spread of this pathogen and the impact of specific factors, such as human mobility, will help authorities to tailor interventions for future SARS-CoV-2 waves or newly emerging airborne infections. In this study, we aim to analyze the spatio-temporal transmission of SARS-CoV-2 in Belgium at municipality level between January and December 2021 and explore the effect of different levels of human travel on disease incidence through the use of counterfactual scenarios.

METHODS

We applied the endemic-epidemic modelling framework, in which the disease incidence decomposes into endemic, autoregressive and neighbourhood components. The spatial dependencies among areas are adjusted based on actual connectivity through mobile network data. We also took into account other important factors such as international mobility, vaccination coverage, population size and the stringency of restriction measures.

RESULTS

The results demonstrate the aggravating effect of international travel on the incidence, and simulated counterfactual scenarios further stress the alleviating impact of a reduction in national and international travel on epidemic growth. It is also clear that local transmission contributed the most during 2021, and municipalities with a larger population tended to attract a higher number of cases from neighboring areas.

CONCLUSIONS

Although transmission between municipalities was observed, local transmission was dominant. We highlight the positive association between the mobility data and the infection spread over time. Our study provides insight to assist health authorities in decision-making, particularly when the disease is airborne and therefore likely influenced by human movement.

Exploring human mixing patterns based on time use and social contact data and their implications for infectious disease transmission models

BACKGROUND

The increasing availability of data on social contact patterns and time use provides invaluable information for studying transmission dynamics of infectious diseases. Social contact data provide information on the interaction of people in a population whereas the value of time use data lies in the quantification of exposure patterns. Both have been used as proxies for transmission risks within in a population and the combination of both sources has led to investigate which contacts are more suitable to describe these transmission risks.

METHODS

We used social contact and time use data from 1707 participants from a survey conducted in Flanders, Belgium in 2010-2011. We calculated weighted exposure time and social contact matrices to analyze age- and gender-specific mixing patterns and to quantify behavioral changes by distance from home. We compared the value of both separate and combined data sources for explaining seroprevalence and incidence data on parvovirus-B19, Varicella-Zoster virus (VZV) and influenza like illnesses (ILI), respectively.

RESULTS

Assortative mixing and inter-generational interaction is more pronounced in the exposure matrix due to the high proportion of time spent at home. This pattern is less pronounced in the social contact matrix, which is more impacted by the reported contacts at school and work. The average number of contacts declined with distance. On the individual-level, we observed an increase in the number of contacts and the transmission potential by distance when travelling. We found that both social contact data and time use data provide a good match with the seroprevalence and incidence data at hand. When comparing the use of different combinations of both data sources, we found that the social contact matrix based on close contacts of at least 4 h appeared to be the best proxy for parvovirus-B19 transmission. Social contacts and exposure time were both on their own able to explain VZV seroprevalence data though combining both scored best. Compared with the contact approach, the time use approach provided the better fit to the ILI incidence data.

CONCLUSIONS

Our work emphasises the common and complementary value of time use and social contact data for analysing mixing behavior and analysing infectious disease transmission. We derived spatial, temporal, age-, gender- and distance-specific mixing patterns, which are informative for future modelling studies.

On the interplay of regional mobility, social connectedness and the spread of COVID-19 in Germany

Since the primary mode of respiratory virus transmission is person-to-person interaction, we are required to reconsider physical interaction patterns to mitigate the number of people infected with COVID-19. While research has shown that non-pharmaceutical interventions (NPI) had an evident impact on national mobility patterns, we investigate the relative regional mobility behaviour to assess the effect of human movement on the spread of COVID-19. In particular, we explore the impact of human mobility and social connectivity derived from Facebook activities on the weekly rate of new infections in Germany between 3 March and 22 June 2020. Our results confirm that reduced social activity lowers the infection rate, accounting for regional and temporal patterns. The extent of social distancing, quantified by the percentage of people staying put within a federal administrative district, has an overall negative effect on the incidence of infections. Additionally, our results show spatial infection patterns based on geographical as well as social distances.

A Review of Spatiotemporal Models for Count Data in R Packages. A Case Study of COVID-19 Data

Spatiotemporal models for count data are required in a wide range of scientific fields, and they have become particularly crucial today because of their ability to analyze COVID-19-related data. The main objective of this paper is to present a review describing the most important approaches, and we monitor their performance under the same dataset. For this review, we focus on the three R-packages that can be used for this purpose, and the different models assessed are representative of the two most widespread methodologies used to analyze spatiotemporal count data: the classical approach and the Bayesian point of view. A COVID-19-related case study is analyzed as an illustration of these different methodologies. Because of the current urgent need for monitoring and predicting data in the COVID-19 pandemic, this case study is, in itself, of particular importance and can be considered the secondary objective of this work. Satisfactory and promising results have been obtained in this second goal. With respect to the main objective, it has been seen that, although the three models provide similar results in our case study, their different properties and flexibility allow us to choose the model depending on the application at hand.

Understanding norovirus reporting patterns in England: a mixed model approach

BACKGROUND

Norovirus has a higher level of under-reporting in England compared to other intestinal infectious agents such as Campylobacter or Salmonella, despite being recognised as the most common cause of gastroenteritis globally. In England, this under-reporting is a consequence of the frequently mild/self-limiting nature of the disease, combined with the passive surveillance system for infectious diseases reporting. We investigated heterogeneity in passive surveillance system in order to improve understanding of differences in reporting and laboratory testing practices of norovirus in England.

METHODS

The reporting patterns of norovirus relating to age and geographical region of England were investigated using a multivariate negative binomial model. Multiple model formulations were compared, and the best performing model was determined by proper scoring rules based on one-week-ahead predictions. The reporting patterns are represented by epidemic and endemic random intercepts; values close to one and less than one imply a lower number of reports than expected in the given region and age-group.

RESULTS

The best performing model highlighted atypically large and small amounts of reporting by comparison with the average in England. Endemic random intercept varied from the lowest in East Midlands in those in the under 5 year age-group (0.36, CI 0.18-0.72) to the highest in the same age group in South West (3.00, CI 1.68-5.35) and Yorkshire & the Humber (2.93, CI 1.74-4.94). Reporting by age groups showed the highest variability in young children.

CONCLUSION

We identified substantial variability in reporting patterns of norovirus by age and by region of England. Our findings highlight the importance of considering uncertainty in the design of forecasting tools for norovirus, and to inform the development of more targeted risk management approaches for norovirus disease.

The geography of COVID-19 and the structure of local economies: The case of Italy

The aim of this article is to analyze the subnational spread of COVID-19 in Italy using an economic geography perspective. The striking spatial unevenness of COVID-19 suggests that the infection has hit economic core locations harder, and this raises questions about whether, and how, the subnational geography of the disease is connected to the economic base of localities. We provide some first evidence consistent with the possibility that the local specialization in geographically concentrated economic activities acts as a vehicle of disease transmission. This could generate a core-periphery pattern in the spatiality of COVID-19, which might follow the lines of the local economic landscape and the tradability of its outputs.

Simulation of the COVID-19 epidemic on the social network of Slovenia: Estimating the intrinsic forecast uncertainty

In the article a virus transmission model is constructed on a simplified social network. The social network consists of more than 2 million nodes, each representing an inhabitant of Slovenia. The nodes are organised and interconnected according to the real household and elderly-care center distribution, while their connections outside these clusters are semi-randomly distributed and undirected. The virus spread model is coupled to the disease progression model. The ensemble approach with the perturbed transmission and disease parameters is used to quantify the ensemble spread, a proxy for the forecast uncertainty. The presented ongoing forecasts of COVID-19 epidemic in Slovenia are compared with the collected Slovenian data. Results show that at the end of the first epidemic wave, the infection was twice more likely to transmit within households/elderly care centers than outside them. We use an ensemble of simulations (N = 1000) and data assimilation approach to estimate the COVID-19 forecast uncertainty and to inversely obtain posterior distributions of model parameters. We found that in the uncontrolled epidemic, the intrinsic uncertainty mostly originates from the uncertainty of the virus biology, i.e. its reproduction number. In the controlled epidemic with low ratio of infected population, the randomness of the social network becomes the major source of forecast uncertainty, particularly for the short-range forecasts. Virus transmission models with accurate social network models are thus essential for improving epidemics forecasting.

Foodborne viral outbreaks associated with frozen produce

Over the past decade, frozen fruits have been a major vehicle of foodborne illnesses mainly attributed to norovirus (NoV) and hepatitis A virus (HAV) infections. Fresh produce may acquire viral contamination by direct contact with contaminated surface, water or hands, and is then frozen without undergoing proper decontamination. Due to their structural integrity, foodborne viruses are able to withstand hostile conditions such as desiccation and freezing, and endure for a long period of time without losing their infectivity. Additionally, these foods are often consumed raw or undercooked, which increases the risk of infection. Herein, we searched published literature and databases of reported outbreaks as well as the databases of news articles for the viral outbreaks associated with the consumption of frozen produce between January 2008 and December 2018; recorded the worldwide distribution of these outbreaks; and analysed the implication of consumption of different types of contaminated frozen food. In addition, we have briefly discussed the factors that contribute to an increased risk of foodborne viral infection following the consumption of frozen produce. Our results revealed that frozen fruits, especially berries and pomegranate arils, contributed to the majority of the outbreaks, and that most outbreaks were reported in industrialised countries.

Combining serological and contact data to derive target immunity levels for achieving and maintaining measles elimination

BACKGROUND

Vaccination has reduced the global incidence of measles to the lowest rates in history. However, local interruption of measles virus transmission requires sustained high levels of population immunity that can be challenging to achieve and maintain. The herd immunity threshold for measles is typically stipulated at 90-95%. This figure does not easily translate into age-specific immunity levels required to interrupt transmission. Previous estimates of such levels were based on speculative contact patterns based on historical data from high-income countries. The aim of this study was to determine age-specific immunity levels that would ensure elimination of measles when taking into account empirically observed contact patterns.

METHODS

We combined estimated immunity levels from serological data in 17 countries with studies of age-specific mixing patterns to derive contact-adjusted immunity levels. We then compared these to case data from the 10 years following the seroprevalence studies to establish a contact-adjusted immunity threshold for elimination. We lastly combined a range of hypothetical immunity profiles with contact data from a wide range of socioeconomic and demographic settings to determine whether they would be sufficient for elimination.

RESULTS

We found that contact-adjusted immunity levels were able to predict whether countries would experience outbreaks in the decade following the serological studies in about 70% of countries. The corresponding threshold level of contact-adjusted immunity was found to be 93%, corresponding to an average basic reproduction number of approximately 14. Testing different scenarios of immunity with this threshold level using contact studies from around the world, we found that 95% immunity would have to be achieved by the age of five and maintained across older age groups to guarantee elimination. This reflects a greater level of immunity required in 5-9-year-olds than established previously.

CONCLUSIONS

The immunity levels we found necessary for measles elimination are higher than previous guidance. The importance of achieving high immunity levels in 5-9-year-olds presents both a challenge and an opportunity. While such high levels can be difficult to achieve, school entry provides an opportunity to ensure sufficient vaccination coverage. Combined with observations of contact patterns, further national and sub-national serological studies could serve to highlight key gaps in immunity that need to be filled in order to achieve national and regional measles elimination.

Networks of face-to-face social contacts in Niakhar, Senegal

We present the first analysis of face-to-face contact network data from Niakhar, Senegal. Participants in a cluster-randomized influenza vaccine trial were interviewed about their contact patterns when they reported symptoms during their weekly household surveillance visit. We employ a negative binomial model to estimate effects of covariates on contact degree. We estimate the mean contact degree for asymptomatic Niakhar residents to be 16.5 (95% C.I. 14.3, 18.7) in the morning and 14.8 in the afternoon (95% C.I. 12.7, 16.9). We estimate that symptomatic people make 10% fewer contacts than asymptomatic people (95% C.I. 5%, 16%; p = 0.006), and those aged 0-5 make 33% fewer contacts than adults (95% C.I. 29%, 37%; p < 0.001). By explicitly modelling the partial rounding pattern observed in our data, we make inference for both the underlying (true) distribution of contacts as well as for the reported distribution. We created an estimator for homophily by compound (household) membership and estimate that 48% of contacts by symptomatic people are made to their own compound members in the morning (95% CI, 45%, 52%) and 60% in the afternoon/evening (95% CI, 56%, 64%). We did not find a significant effect of symptom status on compound homophily. We compare our findings to those from other countries and make design recommendations for future surveys.

Estimating age-stratified influenza-associated invasive pneumococcal disease in England: A time-series model based on population surveillance data

BACKGROUND

Measures of the contribution of influenza to Streptococcus pneumoniae infections, both in the seasonal and pandemic setting, are needed to predict the burden of secondary bacterial infections in future pandemics to inform stockpiling. The magnitude of the interaction between these two pathogens has been difficult to quantify because both infections are mainly clinically diagnosed based on signs and symptoms; a combined viral-bacterial testing is rarely performed in routine clinical practice; and surveillance data suffer from confounding problems common to all ecological studies. We proposed a novel multivariate model for age-stratified disease incidence, incorporating contact patterns and estimating disease transmission within and across groups.

METHODS AND FINDINGS

We used surveillance data from England over the years 2009 to 2017. Influenza infections were identified through the virological testing of samples taken from patients diagnosed with influenza-like illness (ILI) within the sentinel scheme run by the Royal College of General Practitioners (RCGP). Invasive pneumococcal disease (IPD) cases were routinely reported to Public Health England (PHE) by all the microbiology laboratories included in the national surveillance system. IPD counts at week t, conditional on the previous time point t-1, were assumed to be negative binomially distributed. Influenza counts were linearly included in the model for the mean IPD counts along with an endemic component describing some seasonal background and an autoregressive component mimicking pneumococcal transmission. Using age-specific counts, Akaike information criterion (AIC)-based model selection suggested that the best fit was obtained when the endemic component was expressed as a function of observed temperature and rainfall. Pneumococcal transmission within the same age group was estimated to explain 33.0% (confidence interval [CI] 24.9%-39.9%) of new cases in the elderly, whereas 50.7% (CI 38.8%-63.2%) of incidence in adults aged 15-44 years was attributed to transmission from another age group. The contribution of influenza on IPD during the 2009 pandemic also appeared to vary greatly across subgroups, being highest in school-age children and adults (18.3%, CI 9.4%-28.2%, and 6.07%, CI 2.83%-9.76%, respectively). Other viral infections, such as respiratory syncytial virus (RSV) and rhinovirus, also seemed to have an impact on IPD: RSV contributed 1.87% (CI 0.89%-3.08%) to pneumococcal infections in the 65+ group, whereas 2.14% (CI 0.87%-3.57%) of cases in the group of 45- to 64-year-olds were attributed to rhinovirus. The validity of this modelling strategy relies on the assumption that viral surveillance adequately represents the true incidence of influenza in the population, whereas the small numbers of IPD cases observed in the younger age groups led to significant uncertainty around some parameter estimates.

CONCLUSIONS

Our estimates suggested that a pandemic wave of influenza A/H1N1 with comparable severity to the 2009 pandemic could have a modest impact on school-age children and adults in terms of IPD and a small to negligible impact on infants and the elderly. The seasonal impact of other viruses such as RSV and rhinovirus was instead more important in the older population groups.

sPop: Age-structured discrete-time population dynamics model in C, Python, and R

This article describes the sPop packages implementing the deterministic and stochastic versions of an age-structured discrete-time population dynamics model. The packages enable mechanistic modelling of a population by monitoring the age and development stage of each individual. Survival and development are included as the main effectors and they progress at a user-defined pace: follow a fixed rate, delay for a given time, or progress at an age-dependent manner. The model is implemented in C, Python, and R with a uniform design to ease usage and facilitate adoption. Early versions of the model were previously employed for investigating climate-driven population dynamics of the tiger mosquito and the chikungunya disease spread by this vector. The sPop packages presented in this article enable the use of the model in a range of applications extending from vector-borne diseases towards any age-structured population including plant and animal populations, microbial dynamics, host-pathogen interactions, infectious diseases, and other time-dependent epidemiological processes.