Outbreak analytics: a developing data science for informing the response to emerging pathogens

Mechanistic Models of Influenza Transmission in Commercial Swine Populations: A Systematic Review

Influenza in commercial swine populations leads to reduced gain in fattening pigs and reproductive issues in sows. This literature review aims to analyze the contributions of mathematical modeling in understanding influenza transmission and control among domestic swine. Twenty-two full-text research articles from seven databases were reviewed, categorized into swine-only (n = 13), swine-avian (n = 3), and swine-human models (n = 6). Strains of influenza models were limited to H1N1 (n = 7) and H3N2 (n = 1), with many studies generalizing the disease as influenza A. Half of the studies (n = 14) considered at least one control strategy, with vaccination being the primary investigated strategy. Vaccination was shown to reduce disease prevalence in single animal cohorts. With a continuous flow of new susceptible animals, such as in farrow-to-finish farms, it was shown that influenza became endemic despite vaccination strategies such as mass or batch-to-batch vaccination. Human vaccination was shown to be effective at mitigating human-to-human influenza transmission and to reduce spillover events from pigs. Current control strategies cannot stop influenza in livestock or prevent viral reassortment in swine, so mechanistic models are crucial for developing and testing new biosecurity measures to prevent future swine pandemics.

Continued improvement in the development of the SARS-CoV-2 whole genome sequencing proficiency testing program

Application of whole genome sequencing (WGS) has allowed monitoring of the emergence of variants of concern (VOC) of severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) globally. Genomic investigation of emerging variants and surveillance of clinical progress has reduced the public health impact of infection during the COVID-19 pandemic. These steps required developing and implementing a proficiency testing program (PTP), as WGS has been incorporated into routine reference laboratory practice. In this study, we describe how the PTP evaluated the capacity and capability of one New Zealand and 14 Australian public health laboratories to perform WGS of SARS-CoV-2 in 2022. The participants' performances in characterising a specimen panel of known SARS-CoV-2 isolates in the PTP were assessed based on: (1) genome coverage, (2) Pango lineage, and (3) sequence quality, with the choice of assessment metrics refined based on a previously reported assessment conducted in 2021. The participants' performances in 2021 and 2022 were also compared after reassessing the 2021 results using the more stringent metrics adopted in 2022. We found that more participants would have failed the 2021 assessment for all survey samples and a significantly higher fail rate per sample in 2021 compared to 2022. This study highlights the importance of choosing appropriate performance metrics to reflect better the laboratories' capacity to perform SARS-CoV-2 WGS, as was done in the 2022 PTP. It also displays the need for a PTP for WGS of SARS-CoV-2 to be available to public health laboratories ongoing, with continuous refinements in the design and provision of the PTP to account for the dynamic nature of the COVID-19 pandemic as SARS-CoV-2 continues to evolve.

Using real-time modelling to inform the 2017 Ebola outbreak response in DR Congo

Important policy questions during infections disease outbreaks include: i) How effective are particular interventions?; ii) When can resource-intensive interventions be removed? We used mathematical modelling to address these questions during the 2017 Ebola outbreak in Likati Health Zone, Democratic Republic of the Congo (DRC). Eight cases occurred before 15 May 2017, when the Ebola Response Team (ERT; co-ordinated by the World Health Organisation and DRC Ministry of Health) was deployed to reduce transmission. We used a branching process model to estimate that, pre-ERT arrival, the reproduction number was R = 1.49 (95% credible interval ( 0.67, 2.81 ) ). The risk of further cases occurring without the ERT was estimated to be 0.97 (97%). However, no cases materialised, suggesting that the ERT's measures were effective. We also estimated the risk of withdrawing the ERT in real-time. By the actual ERT withdrawal date (2 July 2017), the risk of future cases without the ERT was only 0.01, indicating that the ERT withdrawal decision was safe. We evaluated the sensitivity of our results to the estimated R value and considered different criteria for determining the ERT withdrawal date. This research provides an extensible modelling framework that can be used to guide decisions about when to relax interventions during future outbreaks.

Aptamers: A Cutting-Edge Approach for Gram-Negative Bacterial Pathogen Identification

Early and accurate diagnoses of pathogenic microorganisms is essential to correctly identify diseases, treating infections, and tracking disease outbreaks associated with microbial infections, to develop precautionary measures that allow a fast and effective response in epidemics and pandemics, thus improving public health. Aptamers are a class of synthetic nucleic acid molecules with the potential to be used for medical purposes, since they can be directed towards any target molecule. Currently, the use of aptamers has increased because they are a useful tool in the detection of specific targets. We present a brief review of the use of aptamers to detect and identify bacteria or even some toxins with clinical importance. This work describes the advances in the technology of aptamers, with the purpose of providing knowledge to develop new aptamers for diagnoses and treatment of different diseases caused by infectious microorganisms.

Spatial modeling of Mycobacterium tuberculosis transmission with dyadic genetic relatedness data

Understanding factors that contribute to the increased likelihood of pathogen transmission between two individuals is important for infection control. However, analyzing measures of pathogen relatedness to estimate these associations is complicated due to correlation arising from the presence of the same individual across multiple dyadic outcomes, potential spatial correlation caused by unmeasured transmission dynamics, and the distinctive distributional characteristics of some of the outcomes. We develop two novel hierarchical Bayesian spatial methods for analyzing dyadic pathogen genetic relatedness data, in the form of patristic distances and transmission probabilities, that simultaneously address each of these complications. Using individual-level spatially correlated random effect parameters, we account for multiple sources of correlation between the outcomes as well as other important features of their distribution. Through simulation, we show the limitations of existing approaches in terms of estimating key associations of interest, and the ability of the new methodology to correct for these issues across datasets with different levels of correlation. All methods are applied to Mycobacterium tuberculosis data from the Republic of Moldova, where we identify previously unknown factors associated with disease transmission and, through analysis of the random effect parameters, key individuals, and areas with increased transmission activity. Model comparisons show the importance of the new methodology in this setting. The methods are implemented in the R package GenePair.

A COVID-19 Outbreak in a Large Meat-Processing Plant in England: Transmission Risk Factors and Controls

The meat-processing industry had frequent COVID-19 outbreaks reported worldwide. In May 2021, a large meat-processing plant in the UK had an outbreak affecting 4.1% (63/1541) of workers. A rapid on-site investigation was conducted to understand the virus transmission risk factors and control measures. This included observational assessments of work activities, control measures, real-time environmental measurements and surface microbial sampling. The production night-shift attack rate (11.6%, 44/380) was nearly five times higher than the production day-shift (2.4%, 9/380). Shared work transport was provided to 150 staff per dayshift and 104 per nightshift. Production areas were noisy (≥80 dB(A)) and physical distancing was difficult to maintain. Face visors were mandatory, additional face coverings were required for some activities but not always worn. The refrigeration system continuously recirculated chilled air. In some areas, the mean temperature was as low as 4.5 °C and mean relative humidity (RH) was as high as 96%. The adequacy of ventilation in the production areas could not be assessed reliably using CO2, due to the use of CO2 in the packaging process. While there were challenges in the production areas, the observed COVID-19 control measures were generally implemented well in the non-production areas. Sixty surface samples from all areas were tested for SARS-CoV-2 RNA and 11.7% were positive. Multi-layered measures, informed by a workplace specific risk assessment, are required to prevent and control workplace outbreaks of COVID-19 or other similar respiratory infectious diseases.

A Mobile Health Application Using Geolocation for Behavioral Activity Tracking

The increasing popularity of mHealth presents an opportunity for collecting rich datasets using mobile phone applications (apps). Our health-monitoring mobile application uses motion detection to track an individual's physical activity and location. The data collected are used to improve health outcomes, such as reducing the risk of chronic diseases and promoting healthier lifestyles through analyzing physical activity patterns. Using smartphone motion detection sensors and GPS receivers, we implemented an energy-efficient tracking algorithm that captures user locations whenever they are in motion. To ensure security and efficiency in data collection and storage, encryption algorithms are used with serverless and scalable cloud storage design. The database schema is designed around Mobile Advertising ID (MAID) as a unique identifier for each device, allowing for accurate tracking and high data quality. Our application uses Google's Activity Recognition Application Programming Interface (API) on Android OS or geofencing and motion sensors on iOS to track most smartphones available. In addition, our app leverages blockchain and traditional payments to streamline the compensations and has an intuitive user interface to encourage participation in research. The mobile tracking app was tested for 20 days on an iPhone 14 Pro Max, finding that it accurately captured location during movement and promptly resumed tracking after inactivity periods, while consuming a low percentage of battery life while running in the background.

Praemonitus praemunitus: can we forecast and prepare for future viral disease outbreaks?

Understanding the origins of past and present viral epidemics is critical in preparing for future outbreaks. Many viruses, including SARS-CoV-2, have led to significant consequences not only due to their virulence, but also because we were unprepared for their emergence. We need to learn from large amounts of data accumulated from well-studied, past pandemics and employ modern informatics and therapeutic development technologies to forecast future pandemics and help minimize their potential impacts. While acknowledging the complexity and difficulties associated with establishing reliable outbreak predictions, herein we provide a perspective on the regions of the world that are most likely to be impacted by future outbreaks. We specifically focus on viruses with epidemic potential, namely SARS-CoV-2, MERS-CoV, DENV, ZIKV, MAYV, LASV, noroviruses, influenza, Nipah virus, hantaviruses, Oropouche virus, MARV, and Ebola virus, which all require attention from both the public and scientific community to avoid societal catastrophes like COVID-19. Based on our literature review, data analysis, and outbreak simulations, we posit that these future viral epidemics are unavoidable, but that their societal impacts can be minimized by strategic investment into basic virology research, epidemiological studies of neglected viral diseases, and antiviral drug discovery.

Go.Data as a digital tool for case investigation and contact tracing in the context of COVID-19: a mixed-methods study

BACKGROUND

A manual approach to case investigation and contact tracing can introduce delays in response and challenges for field teams. Go.Data, an outbreak response tool developed by the World Health Organization (WHO) in collaboration with the Global Outbreak Alert and Response Network, streamlines data collection and analysis during outbreaks. This study aimed to characterize Go.Data use during COVID-19, elicit shared benefits and challenges, and highlight key opportunities for enhancement.

METHODS

This study utilized mixed methods through qualitative interviews and a quantitative survey with Go.Data implementors on their experiences during COVID-19. Survey data was analyzed for basic univariate statistics. Interview data were coded using deductive and inductive reasoning and thematic analysis of categories. Overarching themes were triangulated with survey data to clarify key findings.

RESULTS

From April to June 2022, the research team conducted 33 interviews and collected 41 survey responses. Participants were distributed across all six WHO regions and 28 countries. While most implementations represented government actors at national or subnational levels, additional inputs were collected from United Nations agencies and universities. Results highlighted WHO endorsement, accessibility, adaptability, and flexible support modalities as main enabling factors. Formalization and standardization of data systems and people processes to prepare for future outbreaks were a welcomed byproduct of implementation, as 76% used paper-based reporting prior and benefited from increased coordination around a shared platform. Several challenges surfaced, including shortage of the appropriate personnel and skill-mix within teams to ensure smooth implementation. Among opportunities for enhancements were improved product documentation and features to improve usability with large data volumes.

CONCLUSIONS

This study was the first to provide a comprehensive picture of Go.Data implementations during COVID-19 and what joint lessons could be learned. It ultimately demonstrated that Go.Data was a useful complement to responses across diverse contexts, and helped set a reproducible foundation for future outbreaks. Concerted preparedness efforts across the domains of workforce composition, data architecture and political sensitization should be prioritized as key ingredients for future Go.Data implementations. While major developments in Go.Data functionality have addressed some key gaps highlighted during the pandemic, continued dialogue between WHO and implementors, including cross-country experience sharing, is needed ensure the tool is reactive to evolving user needs.

Prehospital Cardiac Arrest Should be Considered When Evaluating Coronavirus Disease 2019 Mortality in the United States

BACKGROUND

Public health emergencies leave little time to develop novel surveillance efforts. Understanding which preexisting clinical datasets are fit for surveillance use is of high value. Coronavirus disease 2019 (COVID-19) offers a natural applied informatics experiment to understand the fitness of clinical datasets for use in disease surveillance.

OBJECTIVES

This study evaluates the agreement between legacy surveillance time series data and discovers their relative fitness for use in understanding the severity of the COVID-19 emergency. Here fitness for use means the statistical agreement between events across series.

METHODS

Thirteen weekly clinical event series from before and during the COVID-19 era for the United States were collected and integrated into a (multi) time series event data model. The Centers for Disease Control and Prevention (CDC) COVID-19 attributable mortality, CDC's excess mortality model, national Emergency Medical Services (EMS) calls, and Medicare encounter level claims were the data sources considered in this study. Cases were indexed by week from January 2015 through June of 2021 and fit to Distributed Random Forest models. Models returned the variable importance when predicting the series of interest from the remaining time series.

RESULTS

Model r2 statistics ranged from 0.78 to 0.99 for the share of the volumes predicted correctly. Prehospital data were of high value, and cardiac arrest (CA) prior to EMS arrival was on average the best predictor (tied with study week). COVID-19 Medicare claims volumes can predict COVID-19 death certificates (agreement), while viral respiratory Medicare claim volumes cannot predict Medicare COVID-19 claims (disagreement).

CONCLUSION

Prehospital EMS data should be considered when evaluating the severity of COVID-19 because prehospital CA known to EMS was the strongest predictor on average across indices.

Systematic review of global hepatitis E outbreaks to inform response and coordination initiatives

INTRODUCTION

Hepatitis E virus (HEV) is the most common cause of acute hepatitis. While symptoms are generally mild and resolve within weeks, some populations (e.g., pregnant women, immunocompromised adults) are at high-risk of severe HEV-related morbidity and mortality. There has not been a recent comprehensive review of contemporary HEV outbreaks, which limits the validity of current disease burden estimates. Therefore, we aimed to characterize global HEV outbreaks and describe data gaps to inform HEV outbreak prevention and response initiatives.

METHODS

We performed a systematic review of peer-reviewed (PubMed, Embase) and gray literature (ProMED) to identify reports of outbreaks published between 2011 and 2022. We included (1) reports with ≥ 5 cases of HEV, and/or (2) reports with 1.5 times the baseline incidence of HEV in a specific population, and (3) all reports with suspected (e.g., clinical case definition) or confirmed (e.g., ELISA or PCR test) cases if they met criterium 1 and/or 2. We describe key outbreak epidemiological, prevention and response characteristics and major data gaps.

RESULTS

We identified 907 records from PubMed, 468 from Embase, and 247 from ProMED. We screened 1,362 potentially relevant records after deduplication. Seventy-one reports were synthesized, representing 44 HEV outbreaks in 19 countries. The populations at risk, case fatalities, and outbreak durations were not reported in 66% of outbreak reports. No reports described using HEV vaccines. Reported intervention efforts included improving sanitation and hygiene, contact tracing/case surveillance, chlorinating boreholes, and advising residents to boil water. Commonly missing data elements included specific case definitions used, testing strategy and methods, seroprevalence, impacts of interventions, and outbreak response costs. Approximately 20% of HEV outbreaks we found were not published in the peer-reviewed literature.

CONCLUSION

HEV represents a significant public health problem. Unfortunately, extensive data shortages and a lack of standardized reporting make it difficult to estimate the HEV disease burden accurately and to implement effective prevention and response activities. Our study has identified major gaps to guide future studies and outbreak reporting systems. Our results support the development of standardized reporting procedures/platforms for HEV outbreaks to ensure accurate and timely data distribution, including active and passive coordinated surveillance systems, particularly among high-risk populations.

A community-based contact isolation strategy to reduce the spread of Ebola virus disease: an analysis of the 2018-2020 outbreak in the Democratic Republic of the Congo

INTRODUCTION

Despite tremendous progress in the development of diagnostics, vaccines and therapeutics for Ebola virus disease (EVD), challenges remain in the implementation of holistic strategies to rapidly curtail outbreaks. We investigated the effectiveness of a community-based contact isolation strategy to limit the spread of the disease in the Democratic Republic of Congo (DRC).

METHODS

We did a quasi-experimental comparison study. Eligible participants were EVD contacts registered from 12 June 2019 to 18 May 2020 in Beni and Mabalako Health Zones. Intervention group participants were isolated to specific community sites for the duration of their follow-up. Comparison group participants underwent contact tracing without isolation. The primary outcome was measured as the reproduction number (R) in the two groups. Secondary outcomes were the delay from symptom onset to isolation and case management, case fatality rate (CFR) and vaccination uptake.

RESULTS

27 324 EVD contacts were included in the study; 585 in the intervention group and 26 739 in the comparison group. The intervention group generated 32 confirmed cases (5.5%) in the first generation, while the comparison group generated 87 (0.3%). However, the 32 confirmed cases arising from the intervention contacts did not generate any additional transmission (R=0.00), whereas the 87 confirmed cases arising from the comparison group generated 99 secondary cases (R=1.14). The average delay between symptom onset and case isolation was shorter (1.3 vs 4.8 days; p<0.0001), CFR lower (12.5% vs 48.4%; p=0.0001) and postexposure vaccination uptake higher (86.0% vs 56.8%; p<0.0001) in the intervention group compared with the comparison group. A significant difference was also found between intervention and comparison groups in survival rate at the discharge of hospitalised confirmed patients (87.9% vs 47.7%, respectively; p=0.0004).

CONCLUSION

The community-based contact isolation strategy used in DRC shows promise as a potentially effective approach for the rapid cessation of EVD transmission, highlighting the importance of rapidly implemented, community-oriented and trust-building control strategies.

An analytical tool to support public policies and isolation barriers against SARS-CoV-2 based on mobility patterns and socio-economic aspects

It is crucial to develop spatiotemporal analysis tools to mitigate risks during a pandemic. Many dashboards encountered in the literature do not consider how the geolocation characteristics and travel patterns may influence the spread of the virus. This work brings an interactive tool that is capable of crossing information about mobility patterns, geolocation characteristics and epidemiologic variables. To do so, our system uses a mobility network, generated through anonymized mobile location data, which enables the division of a region into representative clusters. The clusters' aggregated socioeconomic, and epidemiologic indicators can be analyzed through multiple coordinated views. The proposal is to enable users to understand how different locations commute citizens, monitor risk over time, and understand what locations need more assistance, considering different layers of visualization, such as clusters and individual locations. The main novelty is the interactive way to construct the mobility network that defines the social distancing level and the way that risks are managed, since many different geolocation characteristics can be considered and visualized, such as socioeconomic indicators of a location, the economic importance of a set of locations, and the connection of important neighborhoods of a city with other cities. The proposed tool was built and verified by experts assembled to give scientific recommendations to the city administration of Recife, the capital city of Pernambuco. Our analysis shows how a policymaker could use the tool to evaluate different isolation scenarios considering the trade-off between economic activity and contamination risk, where the practical insights can also be used to tighten and relax mitigation measures in other phases of a pandemic.

Data-driven scenario-based model projections and management of the May 2021 COVID-19 resurgence in India

The resurgence of the May 2021 COVID-19 wave in India not only pointed to the explosive speed with which SARS-CoV-2 can spread in vulnerable populations if unchecked, but also to the gross misreading of the status of the pandemic when decisions to reopen the economy were made in March 2021. In this combined modelling and scenario-based analysis, we isolated the population and policy-related factors underlying the May 2021 viral resurgence by projecting the growth and magnitude of the health impact and demand for hospital care that would have arisen if the spread was not impeded, and by evaluating the intervention options best able to curb the observed rapidly developing contagion. We show that only by immediately re-introducing a moderately high level of social mitigation over a medium-term period alongside a swift ramping up of vaccinations could the country be able to contain and ultimately end the pandemic safely. We also show that delaying the delivery of the 2nd dose of the Astra Zeneca vaccine, as proposed by the Government of India, would have had only slightly more deleterious impacts, supporting the government's decision to vaccinate a greater fraction of the population with at least a single dose as rapidly as possible. Our projections of the scale of the virus resurgence based on the observed May 2021 growth in cases and impacts of intervention scenarios to control the wave, along with the diverse range of variable control actions taken by state authorities, also exemplify the importance of shifting from the use of science and knowledge in an ad hoc reactive fashion to a more effective proactive strategy for assessing and managing the risk of fast-changing hazards, like a pandemic. We show that epidemic models parameterized with data can be used in combination with plausible intervention scenarios to enable such policy-making.

A scoping review of foot-and-mouth disease risk, based on spatial and spatio-temporal analysis of outbreaks in endemic settings

Foot-and-mouth disease (FMD) is one of the most important transboundary animal diseases affecting livestock and wildlife species worldwide. Sustained viral circulation, as evidenced by serological surveys and the recurrence of outbreaks, suggests endemic transmission cycles in some parts of Africa, Asia and the Middle East. This is the result of a complex process in which multiple serotypes, multi-host interactions and numerous socio-epidemiological factors converge to facilitate disease introduction, survival and spread. Spatial and spatio-temporal analyses have been increasingly used to explore the burden of the disease by identifying high-risk areas, analysing temporal trends and exploring the factors that contribute to the outbreaks. We systematically retrieved spatial and spatial-temporal studies on FMD outbreaks to summarize variations on their methodological approaches and identify the epidemiological factors associated with the outbreaks in endemic contexts. Fifty-one studies were included in the final review. A high proportion of papers described and visualized the outbreaks (72.5%) and 49.0% used one or more approaches to study their spatial, temporal and spatio-temporal aggregation. The epidemiological aspects commonly linked to FMD risk are broadly categorizable into themes such as (a) animal demographics and interactions, (b) spatial accessibility, (c) trade, (d) socio-economic and (e) environmental factors. The consistency of these themes across studies underlines the different pathways in which the virus is sustained in endemic areas, with the potential to exploit them to design tailored evidence based-control programmes for the local needs. There was limited data linking the socio-economics of communities and modelled FMD outbreaks, leaving a gap in the current knowledge. A thorough analysis of FMD outbreaks requires a systemic view as multiple epidemiological factors contribute to viral circulation and may improve the accuracy of disease mapping. Future studies should explore the links between socio-economic and epidemiological factors as a foundation for translating the identified opportunities into interventions to improve the outcomes of FMD surveillance and control initiatives in endemic contexts.

Exploring Socioeconomic Status as a Global Determinant of COVID-19 Prevalence, Using Exploratory Data Analytic and Supervised Machine Learning Techniques: Algorithm Development and Validation Study

BACKGROUND

The COVID-19 pandemic represents the most unprecedented global challenge in recent times. As the global community attempts to manage the pandemic in the long term, it is pivotal to understand what factors drive prevalence rates and to predict the future trajectory of the virus.

OBJECTIVE

This study had 2 objectives. First, it tested the statistical relationship between socioeconomic status and COVID-19 prevalence. Second, it used machine learning techniques to predict cumulative COVID-19 cases in a multicountry sample of 182 countries. Taken together, these objectives will shed light on socioeconomic status as a global risk factor of the COVID-19 pandemic.

METHODS

This research used exploratory data analysis and supervised machine learning methods. Exploratory analysis included variable distribution, variable correlations, and outlier detection. Following this, the following 3 supervised regression techniques were applied: linear regression, random forest, and adaptive boosting (AdaBoost). Results were evaluated using k-fold cross-validation and subsequently compared to analyze algorithmic suitability. The analysis involved 2 models. First, the algorithms were trained to predict 2021 COVID-19 prevalence using only 2020 reported case data. Following this, socioeconomic indicators were added as features and the algorithms were trained again. The Human Development Index (HDI) metrics of life expectancy, mean years of schooling, expected years of schooling, and gross national income were used to approximate socioeconomic status.

RESULTS

All variables correlated positively with the 2021 COVID-19 prevalence, with R2 values ranging from 0.55 to 0.85. Using socioeconomic indicators, COVID-19 prevalence was predicted with a reasonable degree of accuracy. Using 2020 reported case rates as a lone predictor to predict 2021 prevalence rates, the average predictive accuracy of the algorithms was low (R2=0.543). When socioeconomic indicators were added alongside 2020 prevalence rates as features, the average predictive performance improved considerably (R2=0.721) and all error statistics decreased. Thus, adding socioeconomic indicators alongside 2020 reported case data optimized the prediction of COVID-19 prevalence to a considerable degree. Linear regression was the strongest learner with R2=0.693 on the first model and R2=0.763 on the second model, followed by random forest (0.481 and 0.722) and AdaBoost (0.454 and 0.679). Following this, the second model was retrained using a selection of additional COVID-19 risk factors (population density, median age, and vaccination uptake) instead of the HDI metrics. However, average accuracy dropped to 0.649, which highlights the value of socioeconomic status as a predictor of COVID-19 cases in the chosen sample.

CONCLUSIONS

The results show that socioeconomic status is an important variable to consider in future epidemiological modeling, and highlights the reality of the COVID-19 pandemic as a social phenomenon and a health care phenomenon. This paper also puts forward new considerations about the application of statistical and machine learning techniques to understand and combat the COVID-19 pandemic.

Quantifying the information in noisy epidemic curves

Reliably estimating the dynamics of transmissible diseases from noisy surveillance data is an enduring problem in modern epidemiology. Key parameters are often inferred from incident time series, with the aim of informing policy-makers on the growth rate of outbreaks or testing hypotheses about the effectiveness of public health interventions. However, the reliability of these inferences depends critically on reporting errors and latencies innate to the time series. Here, we develop an analytical framework to quantify the uncertainty induced by under-reporting and delays in reporting infections, as well as a metric for ranking surveillance data informativeness. We apply this metric to two primary data sources for inferring the instantaneous reproduction number: epidemic case and death curves. We find that the assumption of death curves as more reliable, commonly made for acute infectious diseases such as COVID-19 and influenza, is not obvious and possibly untrue in many settings. Our framework clarifies and quantifies how actionable information about pathogen transmissibility is lost due to surveillance limitations.

Proficiency testing for SARS-CoV-2 whole genome sequencing

Extensive studies and analyses into the molecular features of severe acute respiratory syndrome related coronavirus 2 (SARS-CoV-2) have enhanced the surveillance and investigation of its clusters and transmission worldwide. The whole genome sequencing (WGS) approach is crucial in identifying the source of infection and transmission routes by monitoring the emergence of variants over time and through communities. Varying SARS-CoV-2 genomics capacity and capability levels have been established in public health laboratories across different Australian states and territories. Therefore, laboratories performing SARS-CoV-2 WGS for public health purposes are recommended to participate in an external proficiency testing program (PTP). This study describes the development of a SARS-CoV-2 WGS PTP. The PTP assessed the performance of laboratories while providing valuable insight into the current state of SARS-CoV-2 genomics in public health across Australia. Part 1 of the PTP contained eight simulated SARS-CoV-2 positive and negative specimens to assess laboratories' wet and dry laboratory capacity. Part 2 involved the analysis of a genomic dataset that consisted of a multi-FASTA file of 70 consensus genomes of SARS-CoV-2. Participating laboratories were required to (1) submit raw data for independent bioinformatics analysis, (2) analyse the data with their processes, and (3) answer relevant questions about the data. The performance of the laboratories was commendable, despite some variation in the reported results due to the different sequencing and bioinformatics approaches used by laboratories. The overall outcome is positive and demonstrates the critical role of the PTP in supporting the implementation and validation of SARS-CoV-2 WGS processes. The data derived from this PTP will contribute to the development of SARS-CoV-2 bioinformatic quality control (QC) and performance benchmarking for accreditation.

Uncovering chains of infections through spatio-temporal and visual analysis of COVID-19 contact traces

A major challenge for departments of public health (DPHs) in dealing with the ongoing COVID-19 pandemic is tracing contacts in exponentially growing SARS-CoV-2 infection clusters. Prevention of further disease spread requires a comprehensive registration of the connections between individuals and clusters. Due to the high number of infections with unknown origin, the healthcare analysts need to identify connected cases and clusters through accumulated epidemiological knowledge and the metadata of the infections in their database. Here we contribute a visual analytics dashboard to identify, assess and visualize clusters in COVID-19 contact tracing networks. Additionally, we demonstrate how graph-based machine learning methods can be used to find missing links between infection clusters and thus support the mission to get a comprehensive view on infection events. This work was developed through close collaboration with DPHs in Germany. We argue how our dashboard supports the identification of clusters by public health experts, discuss ongoing developments and possible extensions.

Gecko: A time-series model for COVID-19 hospital admission forecasting

During the COVID-19 pandemic, concerns about hospital capacity in the United States led to a demand for models that forecast COVID-19 hospital admissions. These short-term forecasts were needed to support planning efforts by providing decision-makers with insight about future demands for health care capacity and resources. We present a SARIMA time-series model called Gecko developed for this purpose. We evaluate its historical performance using metrics such as mean absolute error, predictive interval coverage, and weighted interval scores, and compare to alternative hospital admission forecasting models. We find that Gecko outperformed baseline approaches and was among the most accurate models for forecasting hospital admissions at the state and national levels from January–May 2021. This work suggests that simple statistical methods can provide a viable alternative to traditional epidemic models for short-term forecasting.

Measuring the unknown: An estimator and simulation study for assessing case reporting during epidemics

The fraction of cases reported, known as 'reporting', is a key performance indicator in an outbreak response, and an essential factor to consider when modelling epidemics and assessing their impact on populations. Unfortunately, its estimation is inherently difficult, as it relates to the part of an epidemic which is, by definition, not observed. We introduce a simple statistical method for estimating reporting, initially developed for the response to Ebola in Eastern Democratic Republic of the Congo (DRC), 2018-2020. This approach uses transmission chain data typically gathered through case investigation and contact tracing, and uses the proportion of investigated cases with a known, reported infector as a proxy for reporting. Using simulated epidemics, we study how this method performs for different outbreak sizes and reporting levels. Results suggest that our method has low bias, reasonable precision, and despite sub-optimal coverage, usually provides estimates within close range (5-10%) of the true value. Being fast and simple, this method could be useful for estimating reporting in real-time in settings where person-to-person transmission is the main driver of the epidemic, and where case investigation is routinely performed as part of surveillance and contact tracing activities.

When Is Mass Prophylaxis Cost-Effective for Epidemic Control? A Comparison of Decision Approaches

BACKGROUND

For certain communicable disease outbreaks, mass prophylaxis of uninfected individuals can curtail new infections. When an outbreak emerges, decision makers could benefit from methods to quickly determine whether mass prophylaxis is cost-effective. We consider 2 approaches: a simple decision model and machine learning meta-models. The motivating example is plague in Madagascar.

METHODS

We use a susceptible-exposed-infectious-removed (SEIR) epidemic model to derive a decision rule based on the fraction of the population infected, effective reproduction ratio, infection fatality rate, quality-adjusted life-year loss associated with death, prophylaxis effectiveness and cost, time horizon, and willingness-to-pay threshold. We also develop machine learning meta-models of a detailed model of plague in Madagascar using logistic regression, random forest, and neural network models. In numerical experiments, we compare results using the decision rule and the meta-models to results obtained using the simulation model. We vary the initial fraction of the population infected, the effective reproduction ratio, the intervention start date and duration, and the cost of prophylaxis.

LIMITATIONS

We assume homogeneous mixing and no negative side effects due to antibiotic prophylaxis.

RESULTS

The simple decision rule matched the SEIR model outcome in 85.4% of scenarios. Using data for a 2017 plague outbreak in Madagascar, the decision rule correctly indicated that mass prophylaxis was not cost-effective. The meta-models were significantly more accurate, with an accuracy of 92.8% for logistic regression, 95.8% for the neural network model, and 96.9% for the random forest model.

CONCLUSIONS

A simple decision rule using minimal information about an outbreak can accurately evaluate the cost-effectiveness of mass prophylaxis for outbreak mitigation. Meta-models of a complex disease simulation can achieve higher accuracy but with greater computational and data requirements and less interpretability.

HIGHLIGHTS

We use a susceptible-exposed-infectious-removed model and net monetary benefit to derive a simple decision rule to evaluate the cost-effectiveness of mass prophylaxis.We use the example of plague in Madagascar to compare the performance of the analytically derived decision rule to that of machine learning meta-models trained on a stochastic dynamic transmission model.We assess the accuracy of each approach for different combinations of disease dynamics and intervention scenarios.The machine learning meta-models are more accurate predictors of mass prophylaxis cost-effectiveness. However, the simple decision rule is also accurate and may be a preferred substitute in low-resource settings.

Telemedicine medical evaluation of low-risk patients with dengue during an outbreak may be an option in reducing the need for on-site physicians

OBJECTIVE

To analyze the effectiveness of telemedicine consultations during an outbreak in reducing the need for face-to-face consultations at a field hospital for patients with dengue.

METHODS

We performed a retrospective unicentric study between April and May 2015 with 4626 patients (≥15 years old) who spontaneously sought care at an emergency field hospital (Sāo Paulo/Brazil). A nurse initially assessed all patients with dengue through rapid diagnostic testing, automated complete blood count, and risk stratification. During overcrowded situations, a video-based telemedicine consultation was provided as an option to all low-to-moderate risk patients who tested positive. The management was carried out according to current dengue guidelines. The primary end point was a referral to immediate face-to-face medical evaluation.

RESULTS

Of all patients suspected of dengue infection, 2003 presented positive testing, 1978 were classified as low-moderate risk, and 267 patients with dengue were evaluated by telemedicine. The mean age was 38.17 ± 13.7 years (54.6% female). Oral medications were recommended in 169 (63.3%), intravenous hydration or symptomatic drugs in 96 (36%), 252 (94.4%) were discharged after telemedicine assessment, and only 15 (5.6%) were referred to immediate face-to-face medical evaluation. No adverse events were recorded.

CONCLUSION

Telemedicine medical assessment of low-to-moderate risk patients with dengue previously screened by nursing triage is effective in replacing the face-to-face evaluation in a field hospital. Telemedicine may be reinforced in epidemiological outbreak scenarios as a cost-effective strategy for the initial assessment of acute patients.

Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the United States

Short-term probabilistic forecasts of the trajectory of the COVID-19 pandemic in the United States have served as a visible and important communication channel between the scientific modeling community and both the general public and decision-makers. Forecasting models provide specific, quantitative, and evaluable predictions that inform short-term decisions such as healthcare staffing needs, school closures, and allocation of medical supplies. Starting in April 2020, the US COVID-19 Forecast Hub (https://covid19forecasthub.org/) collected, disseminated, and synthesized tens of millions of specific predictions from more than 90 different academic, industry, and independent research groups. A multimodel ensemble forecast that combined predictions from dozens of groups every week provided the most consistently accurate probabilistic forecasts of incident deaths due to COVID-19 at the state and national level from April 2020 through October 2021. The performance of 27 individual models that submitted complete forecasts of COVID-19 deaths consistently throughout this year showed high variability in forecast skill across time, geospatial units, and forecast horizons. Two-thirds of the models evaluated showed better accuracy than a naïve baseline model. Forecast accuracy degraded as models made predictions further into the future, with probabilistic error at a 20-wk horizon three to five times larger than when predicting at a 1-wk horizon. This project underscores the role that collaboration and active coordination between governmental public-health agencies, academic modeling teams, and industry partners can play in developing modern modeling capabilities to support local, state, and federal response to outbreaks.

Fundamental limits on inferring epidemic resurgence in real time using effective reproduction numbers

We find that epidemic resurgence, defined as an upswing in the effective reproduction number (R) of the contagion from subcritical to supercritical values, is fundamentally difficult to detect in real time. Inherent latencies in pathogen transmission, coupled with smaller and intrinsically noisier case incidence across periods of subcritical spread, mean that resurgence cannot be reliably detected without significant delays of the order of the generation time of the disease, even when case reporting is perfect. In contrast, epidemic suppression (where R falls from supercritical to subcritical values) may be ascertained 5-10 times faster due to the naturally larger incidence at which control actions are generally applied. We prove that these innate limits on detecting resurgence only worsen when spatial or demographic heterogeneities are incorporated. Consequently, we argue that resurgence is more effectively handled proactively, potentially at the expense of false alarms. Timely responses to recrudescent infections or emerging variants of concern are more likely to be possible when policy is informed by a greater quality and diversity of surveillance data than by further optimisation of the statistical models used to process routine outbreak data.

Effectiveness of Human Mobility Change in Reducing the Spread of COVID-19: Ecological Study of Kingdom of Saudi Arabia

Non-pharmacological interventions including mobility restriction have been developed to curb transmission of SARS-CoV-2. We provided precise estimates of disease burden and examined the impact of mobility restriction on reducing the COVID-19 effective reproduction number in the Kingdom of Saudi Arabia. This study involved secondary analysis of open-access COVID-19 data obtained from different sources between 2 March and 26 December 2020. The dependent and main independent variables of interest were the effective reproduction number and anonymized mobility indices, respectively. Multiple linear regression was used to investigate the relationship between the community mobility change and the effective reproduction number for COVID-19. By 26 December 2020, the total number of COVID-19 cases in Saudi Arabia reached 360,690, with a cumulative incidence rate of 105.41/10,000 population. Al Jouf, Northern Border, and Jazan regions were ≥2.5 times (OR = 2.93; 95% CI: 1.29–6.64), (OR = 2.50; 95% CI: 1.08–5.81), and (OR = 2.51; 95% CI: 1.09–5.79) more likely to have a higher case fatality rate than Riyadh, the capital. Mobility changes in public and residential areas were significant predictors of the COVID-19 effective reproduction number. This study demonstrated that community mobility restrictions effectively control transmission of the COVID-19 virus.

Iterative data-driven forecasting of the transmission and management of SARS-CoV-2/COVID-19 using social interventions at the county-level

The control of the initial outbreak and spread of SARS-CoV-2/COVID-19 via the application of population-wide non-pharmaceutical mitigation measures have led to remarkable successes in dampening the pandemic globally. However, with countries beginning to ease or lift these measures fully to restart activities, concern is growing regarding the impacts that such reopening of societies could have on the subsequent transmission of the virus. While mathematical models of COVID-19 transmission have played important roles in evaluating the impacts of these measures for curbing virus transmission, a key need is for models that are able to effectively capture the effects of the spatial and social heterogeneities that drive the epidemic dynamics observed at the local community level. Iterative forecasting that uses new incoming epidemiological and social behavioral data to sequentially update locally-applicable transmission models can overcome this gap, potentially resulting in better predictions and policy actions. Here, we present the development of one such data-driven iterative modelling tool based on publicly available data and an extended SEIR model for forecasting SARS-CoV-2 at the county level in the United States. Using data from the state of Florida, we demonstrate the utility of such a system for exploring the outcomes of the social measures proposed by policy makers for containing the course of the pandemic. We provide comprehensive results showing how the locally identified models could be employed for accessing the impacts and societal tradeoffs of using specific social protective strategies. We conclude that it could have been possible to lift the more disruptive social interventions related to movement restriction/social distancing measures earlier if these were accompanied by widespread testing and contact tracing. These intensified social interventions could have potentially also brought about the control of the epidemic in low- and some medium-incidence county settings first, supporting the development and deployment of a geographically-phased approach to reopening the economy of Florida. We have made our data-driven forecasting system publicly available for policymakers and health officials to use in their own locales, so that a more efficient coordinated strategy for controlling SARS-CoV-2 region-wide can be developed and successfully implemented.

A data-driven approach to measuring epidemiological susceptibility risk around the world

Epidemic outbreaks are extreme events that become more frequent and severe, associated with large social and real costs. It is therefore important to assess whether countries are prepared to manage epidemiological risks. We use a fully data-driven approach to measure epidemiological susceptibility risk at the country level using time-varying information. We apply both principal component analysis (PCA) and dynamic factor model (DFM) to deal with the presence of strong cross-section dependence in the data. We conduct extensive in-sample model evaluations of 168 countries covering 17 indicators for the 2010-2019 period. The results show that the robust PCA method accounts for about 90% of total variability, whilst the DFM accounts for about 76% of the total variability. Our index could therefore provide the basis for developing risk assessments of epidemiological risk contagion. It could be also used by organizations to assess likely real consequences of epidemics with useful managerial implications.

Assessment of Food and Waterborne Viral Outbreaks by Using Field Epidemiologic, Modern Laboratory and Statistical Methods-Lessons Learnt from Seven Major Norovirus Outbreaks in Finland

Seven major food- and waterborne norovirus outbreaks in Western Finland during 2014–2018 were re-analysed. The aim was to assess the effectiveness of outbreak investigation tools and evaluate the Kaplan criteria. We summarised epidemiological and microbiological findings from seven outbreaks. To evaluate the Kaplan criteria, a one-stage meta-analysis of data from seven cohort studies was performed. The case was defined as a person attending an implicated function with diarrhoea, vomiting or two other symptoms. Altogether, 22% (386/1794) of persons met the case definition. Overall adjusted, 73% of norovirus patients were vomiting, the mean incubation period was 44 h (4 h to 4 days) and the median duration of illness was 46 h. As vomiting was a more common symptom in children (96%, 143/149) and diarrhoea among the elderly (92%, 24/26), symptom and age presentation should drive hypothesis formulation. The Kaplan criteria were useful in initial outbreak assessments prior to faecal results. Rapid food control inspections enabled evidence-based, public-health-driven risk assessments. This led to probability-based vehicle identification and aided in resolving the outbreak event mechanism rather than implementing potentially ineffective, large-scale public health actions such as the withdrawal of extensive food lots. Asymptomatic food handlers should be ideally withdrawn from high-risk work for five days instead of the current two days. Food and environmental samples often remain negative with norovirus, highlighting the importance of research collaborations. Electronic questionnaire and open-source novel statistical programmes provided time and resource savings. The public health approach proved useful within the environmental health area with shoe leather field epidemiology, combined with statistical analysis and mathematical reasoning.

Data collection for outbreak investigations: process for defining a minimal data set using a Delphi approach

Background

Timely but accurate data collection is needed during health emergencies to inform public health responses. Often, an abundance of data is collected but not used. When outbreaks and other health events occur in remote and complex settings, operatives on the ground are often required to cover multiple tasks whilst working with limited resources. Tools that facilitate the collection of essential data during the early investigations of a potential public health event can support effective public health decision-making. We proposed to define the minimum set of quantitative information to collect whilst using electronic device or not. Here we present the process used to select the minimum information required to describe an outbreak of any cause during its initial stages and occurring in remote settings.

Methods

A working group of epidemiologists took part in two rounds of a Delphi process to categorise the variables to be included in an initial outbreak investigation form. This took place between January–June 2019 using an online survey.

Results

At a threshold of 75 %, consensus was reached for nineteen (23.2%) variables which were all classified as ‘essential’. This increased to twenty-six (31.7%) variables when the threshold was reduced to 60% with all but one variable classified as ‘essential’. Twenty-five of these variables were included in the ‘Time zero initial case investigation’ ‘(T0)’ form which was shared with the members of the Rapid Response Team Knowledge Network for field testing and feedback. The form has been readily available online by WHO since September 2019.

Conclusion

This is the first known Delphi process used to determine the minimum variables needed for an outbreak investigation. The subsequent development of the T0 form should help to improve the efficiency and standardisation of data collection during emergencies and ultimately the quality of the data collected during field investigation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-12206-5.

Adaptive Preferential Sampling in Phylodynamics With an Application to SARS-CoV-2

Longitudinal molecular data of rapidly evolving viruses and pathogens provide information about disease spread and complement traditional surveillance approaches based on case count data. The coalescent is used to model the genealogy that represents the sample ancestral relationships. The basic assumption is that coalescent events occur at a rate inversely proportional to the effective population size N e (t), a time-varying measure of genetic diversity. When the sampling process (collection of samples over time) depends on N e (t), the coalescent and the sampling processes can be jointly modeled to improve estimation of N e (t). Failing to do so can lead to bias due to model misspecification. However, the way that the sampling process depends on the effective population size may vary over time. We introduce an approach where the sampling process is modeled as an inhomogeneous Poisson process with rate equal to the product of N e (t) and a time-varying coefficient, making minimal assumptions on their functional shapes via Markov random field priors. We provide efficient algorithms for inference, show the model performance vis-a-vis alternative methods in a simulation study, and apply our model to SARS-CoV-2 sequences from Los Angeles and Santa Clara counties. The methodology is implemented and available in the R package adapref. Supplementary files for this article are available online.

Novel Use of Capture-Recapture Methods to Estimate Completeness of Contact Tracing during an Ebola Outbreak, Democratic Republic of the Congo, 2018-2020

Despite its critical role in containing outbreaks, the efficacy of contact tracing, measured as the sensitivity of case detection, remains an elusive metric. We estimated the sensitivity of contact tracing by applying unilist capture-recapture methods on data from the 2018–2020 outbreak of Ebola virus disease in the Democratic Republic of the Congo. To compute sensitivity, we applied different distributional assumptions to the zero-truncated count data to estimate the number of unobserved case-patients with any contacts and infected contacts. Geometric distributions were the best-fitting models. Our results indicate that contact tracing efforts identified almost all (n = 792, 99%) of case-patients with any contacts but only half (n = 207, 48%) of case-patients with infected contacts, suggesting that contact tracing efforts performed well at identifying contacts during the listing stage but performed poorly during the contact follow-up stage. We discuss extensions to our work and potential applications for the ongoing coronavirus pandemic.

The Epidemic Failure Cycle hypothesis: Towards understanding the global community's recent failures in responding to an epidemic

BACKGROUND

Within a few years, the global community has failed twice in responding to large viral infection outbreaks: the Ebola epidemic in 2014 and the SARS-Cov-2 pandemic in 2020. There is, however, no systematic approach or research available that analyses the repeated failures with regard to an adequate response to an epidemic.

METHODS

For a better understanding of failing societal responses, we have analysed the available research literature on societal responses to epidemics and we propose a framework called the 'Epidemic Failure Cycle' (EFC).

RESULTS

The EFC consists of four phases: Negligence, Arrogance/Denial, Panic and Analysis/Self-criticism. These phases fit largely with the current World Health Organization pandemic influenza phases: Interpandemic, Alert, Pandemic, Transition. By utilizing the Ebola epidemic and the SARS-Cov-2 pandemic as case studies, we show striking similarities in the response to these outbreaks during both crises. Finally, we suggest three major areas to be of utmost importance for triggering and maintaining the EFC. In terms of ecology, zoonoses, supposed to be the main biological origin for virus epidemics, have been largely neglected by politicians, the media and the scientific community. Socioeconomic and cultural conditions such as harsh living and working conditions as well as conspiracy theories hinder effective preventive and counter measures against epidemics. Lastly, in terms of epistemology, the reliance on knowledge about previous outbreaks has led to slow and inadequate decisions.

CONCLUSIONS

We conclude that any current society has to be aware of the risks of repeating responses to epidemics that will fail. Being aware of the societal mechanisms that trigger inadequate responses may help to get to more appropriate decisions in the face of an epidemic.

Electronic data collection, management and analysis tools used for outbreak response in low- and middle-income countries: a systematic review and stakeholder survey

BACKGROUND

Use of electronic data collection, management and analysis tools to support outbreak response is limited, especially in low income countries. This can hamper timely decision-making during outbreak response. Identifying available tools and assessing their functions in the context of outbreak response would support appropriate selection and use, and likely more timely data-driven decision-making during outbreaks.

METHODS

We conducted a systematic review and a stakeholder survey of the Global Outbreak Alert and Response Network and other partners to identify and describe the use of, and technical characteristics of, electronic data tools used for outbreak response in low- and middle-income countries. Databases included were MEDLINE, EMBASE, Global Health, Web of Science and CINAHL with publications related to tools for outbreak response included from January 2010-May 2020. Software tool websites of identified tools were also reviewed. Inclusion and exclusion criteria were applied and counts, and proportions of data obtained from the review or stakeholder survey were calculated.

RESULTS

We identified 75 electronic tools including for data collection (33/75), management (13/75) and analysis (49/75) based on data from the review and survey. Twenty-eight tools integrated all three functionalities upon collection of additional information from the tool developer websites. The majority were open source, capable of offline data collection and data visualisation. EpiInfo, KoBoCollect and Open Data Kit had the broadest use, including for health promotion, infection prevention and control, and surveillance data capture. Survey participants highlighted harmonisation of data tools as a key challenge in outbreaks and the need for preparedness through training front-line responders on data tools. In partnership with the Global Health Network, we created an online interactive decision-making tool using data derived from the survey and review.

CONCLUSIONS

Many electronic tools are available for data -collection, -management and -analysis in outbreak response, but appropriate tool selection depends on knowledge of tools' functionalities and capabilities. The online decision-making tool created to assist selection of the most appropriate tool(s) for outbreak response helps by matching requirements with functionality. Applying the tool together with harmonisation of data formats, and training of front-line responders outside of epidemic periods can support more timely data-driven decision making in outbreaks.

Scoping future outbreaks: a scoping review on the outbreak prediction of the WHO Blueprint list of priority diseases

BACKGROUND

The WHO's Research and Development Blueprint priority list designates emerging diseases with the potential to generate public health emergencies for which insufficient preventive solutions exist. The list aims to reduce the time to the availability of resources that can avert public health crises. The current SARS-CoV-2 pandemic illustrates that an effective method of mitigating such crises is the pre-emptive prediction of outbreaks. This scoping review thus aimed to map and identify the evidence available to predict future outbreaks of the Blueprint diseases.

METHODS

We conducted a scoping review of PubMed, Embase and Web of Science related to the evidence predicting future outbreaks of Ebola and Marburg virus, Zika virus, Lassa fever, Nipah and Henipaviral disease, Rift Valley fever, Crimean-Congo haemorrhagic fever, Severe acute respiratory syndrome, Middle East respiratory syndrome and Disease X. Prediction methods, outbreak features predicted and implementation of predictions were evaluated. We conducted a narrative and quantitative evidence synthesis to highlight prediction methods that could be further investigated for the prevention of Blueprint diseases and COVID-19 outbreaks.

RESULTS

Out of 3959 articles identified, we included 58 articles based on inclusion criteria. 5 major prediction methods emerged; the most frequent being spatio-temporal risk maps predicting outbreak risk periods and locations through vector and climate data. Stochastic models were predominant. Rift Valley fever was the most predicted disease. Diseases with complex sociocultural factors such as Ebola were often predicted through multifactorial risk-based estimations. 10% of models were implemented by health authorities. No article predicted Disease X outbreaks.

CONCLUSIONS

Spatiotemporal models for diseases with strong climatic and vectorial components, as in River Valley fever prediction, may currently best reduce the time to the availability of resources. A wide literature gap exists in the prediction of zoonoses with complex sociocultural and ecological dynamics such as Ebola, COVID-19 and especially Disease X.

How to improve outbreak response: a case study of integrated outbreak analytics from Ebola in Eastern Democratic Republic of the Congo

The emerging field of outbreak analytics calls attention to the need for data from multiple sources to inform evidence-based decision making in managing infectious diseases outbreaks. To date, these approaches have not systematically integrated evidence from social and behavioural sciences. During the 2018–2020 Ebola outbreak in Eastern Democratic Republic of the Congo, an innovative solution to systematic and timely generation of integrated and actionable social science evidence emerged in the form of the Cellulle d’Analyse en Sciences Sociales (Social Sciences Analytics Cell) (CASS), a social science analytical cell. CASS worked closely with data scientists and epidemiologists operating under the Epidemiological Cell to produce integrated outbreak analytics (IOA), where quantitative epidemiological analyses were complemented by behavioural field studies and social science analyses to help better explain and understand drivers and barriers to outbreak dynamics. The primary activity of the CASS was to conduct operational social science analyses that were useful to decision makers. This included ensuring that research questions were relevant, driven by epidemiological data from the field, that research could be conducted rapidly (ie, often within days), that findings were regularly and systematically presented to partners and that recommendations were co-developed with response actors. The implementation of the recommendations based on CASS analytics was also monitored over time, to measure their impact on response operations. This practice paper presents the CASS logic model, developed through a field-based externally led consultation, and documents key factors contributing to the usefulness and adaption of CASS and IOA to guide replication for future outbreaks.

The COVID-OUT study protocol: COVID-19 outbreak investigation to understand workplace SARS-CoV-2 transmission in the United Kingdom

Preventing SARS-CoV-2 transmission and protecting people from COVID-19 is the most significant public health challenge faced in recent years. COVID-19 outbreaks are occurring in workplaces and evidence is needed to support effective strategies to prevent and control these outbreaks. Investigations into these outbreaks are routinely undertaken by public health bodies and regulators in the United Kingdom (UK); however, such investigations are typically disparate in nature with a lack of consistency across all investigations, preventing meaningful analysis of the data collected. The COVID-OUT (COVID-19 Outbreak investigation to Understand Transmission) study aims to collect a consistent set of data in a systematic way from workplaces that are experiencing outbreaks, to understand SARS-CoV-2 transmission risk factors, transmission routes, and the role they play in the COVID-19 outbreaks. Suitable outbreak sites are identified from public health bodies. Following employer consent to participate, the study will recruit workers from workplaces where there are active outbreaks. The study will utilise data already collected as part of routine public health outbreak investigations and collect additional data through a comprehensive questionnaire, viral and serologic testing of workers, surface sampling, viral genome sequencing, and an environmental assessment of building plans, ventilation and current control measures. At each site, a detailed investigation will be carried out to evaluate transmission routes. A case-control approach will be used to compare workers who have and have not had SARS-CoV-2 infections during the outbreak period to assess transmission risk factors. Data from different outbreaks will be combined for pooled analyses to identify common risk factors, as well as factors that differ between outbreaks. The COVID-OUT study can contribute to a better understanding of why COVID-19 outbreaks associated with workplaces occur and how to prevent these outbreaks from happening in the future.

Real-time monitoring of COVID-19 dynamics using automated trend fitting and anomaly detection

As several countries gradually release social distancing measures, rapid detection of new localized COVID-19 hotspots and subsequent intervention will be key to avoiding large-scale resurgence of transmission. We introduce ASMODEE (automatic selection of models and outlier detection for epidemics), a new tool for detecting sudden changes in COVID-19 incidence. Our approach relies on automatically selecting the best (fitting or predicting) model from a range of user-defined time series models, excluding the most recent data points, to characterize the main trend in an incidence. We then derive prediction intervals and classify data points outside this interval as outliers, which provides an objective criterion for identifying departures from previous trends. We also provide a method for selecting the optimal breakpoints, used to define how many recent data points are to be excluded from the trend fitting procedure. The analysis of simulated COVID-19 outbreaks suggests ASMODEE compares favourably with a state-of-art outbreak-detection algorithm while being simpler and more flexible. As such, our method could be of wider use for infectious disease surveillance. We illustrate ASMODEE using publicly available data of National Health Service (NHS) Pathways reporting potential COVID-19 cases in England at a fine spatial scale, showing that the method would have enabled the early detection of the flare-ups in Leicester and Blackburn with Darwen, two to three weeks before their respective lockdown. ASMODEE is implemented in the free R package trendbreaker. This article is part of the theme issue 'Modelling that shaped the early COVID-19 pandemic response in the UK'.

A Comparative Analysis of Statistical Methods to Estimate the Reproduction Number in Emerging Epidemics, With Implications for the Current Coronavirus Disease 2019 (COVID-19) Pandemic

BACKGROUND

As the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic continues its rapid global spread, quantification of local transmission patterns has been, and will continue to be, critical for guiding the pandemic response. Understanding the accuracy and limitations of statistical methods to estimate the basic reproduction number, R0, in the context of emerging epidemics is therefore vital to ensure appropriate interpretation of results and the subsequent implications for control efforts.

METHODS

Using simulated epidemic data, we assess the performance of 7 commonly used statistical methods to estimate R0 as they would be applied in a real-time outbreak analysis scenario: fitting to an increasing number of data points over time and with varying levels of random noise in the data. Method comparison was also conducted on empirical outbreak data, using Zika surveillance data from the 2015-2016 epidemic in Latin America and the Caribbean.

RESULTS

We find that most methods considered here frequently overestimate R0 in the early stages of epidemic growth on simulated data, the magnitude of which decreases when fitted to an increasing number of time points. This trend of decreasing bias over time can easily lead to incorrect conclusions about the course of the epidemic or the need for control efforts.

CONCLUSIONS

We show that true changes in pathogen transmissibility can be difficult to disentangle from changes in methodological accuracy and precision in the early stages of epidemic growth, particularly for data with significant over-dispersion. As localized epidemics of SARS-CoV-2 take hold around the globe, awareness of this trend will be important for appropriately cautious interpretation of results and subsequent guidance for control efforts.

Impact of COVID-19 in Minas Gerais, Brazil: Excess deaths, sub-notified cases, geographic and ethnic distribution

By analysing the evolution of the COVID-19 epidemic in the state of Minas Gerais, Brazil, we showed the importance of considering the sub-notification not only of deaths but also of infected cases. It was shown that the largely used criteria of a historical all-deaths baseline are not approachable in this case, where most of the deaths are associated with causes that should decrease due to social distancing and reduction of economic activities. A quite simple and intuitive model based on the Gompertz function was applied to estimate excess deaths and excess of infected cases. It fits well the data and predicts the evolution of the epidemic adequately. Based on these analyses, an excess of 21.638 deaths and 557.216 infected cases is predicted until the end of 2020, with an upper bound of the case fatality rate of around 2.4% and a prevalence of 2.6%. The geographical distribution of cases and deaths and its ethnic correlation are also presented. This study points out the necessity of governmental and private organizations working together to improve public awareness and stimulate social distancing to curb the viral infection, especially in critical places with high poverty.

A review and agenda for integrated disease models including social and behavioural factors

Social and behavioural factors are critical to the emergence, spread and containment of human disease, and are key determinants of the course, duration and outcomes of disease outbreaks. Recent epidemics of Ebola in West Africa and coronavirus disease 2019 (COVID-19) globally have reinforced the importance of developing infectious disease models that better integrate social and behavioural dynamics and theories. Meanwhile, the growth in capacity, coordination and prioritization of social science research and of risk communication and community engagement (RCCE) practice within the current pandemic response provides an opportunity for collaboration among epidemiological modellers, social scientists and RCCE practitioners towards a mutually beneficial research and practice agenda. Here, we provide a review of the current modelling methodologies and describe the challenges and opportunities for integrating them with social science research and RCCE practice. Finally, we set out an agenda for advancing transdisciplinary collaboration for integrated disease modelling and for more robust policy and practice for reducing disease transmission.

What containment strategy leads us through the pandemic crisis? An empirical analysis of the measures against the COVID-19 pandemic

Since January 2020, the COVID-19 outbreak has been progressing at a rapid pace. To keep the pandemic at bay, countries have implemented various measures to interrupt the transmission of the virus from person to person and prevent an overload of their health systems. We analyze the impact of these measures implemented against the COVID-19 pandemic by using a sample of 68 countries, Puerto Rico and the 50 federal states of the United States of America, four federal states of Australia, and eight federal states of Canada, involving 6,941 daily observations. We show that measures are essential for containing the spread of the COVID-19 pandemic. After controlling for daily COVID-19 tests, we find evidence to suggest that school closures, shut-downs of non-essential business, mass gathering bans, travel restrictions in and out of risk areas, national border closures and/or complete entry bans, and nationwide curfews decrease the growth rate of the coronavirus and thus the peak of daily confirmed cases. We also find evidence to suggest that combinations of these measures decrease the daily growth rate at a level outweighing that of individual measures. Consequently, and despite extensive vaccinations, we contend that the implemented measures help contain the spread of the COVID-19 pandemic and ease the overstressed capacity of the healthcare systems.

Development of a Dissemination Platform for Spatiotemporal and Phylogenetic Analysis of Avian Infectious Bronchitis Virus

Infecting large portions of the global poultry populations, the avian infectious bronchitis virus (IBV) remains a major economic burden in North America. With more than 30 serotypes globally distributed, Arkansas, Connecticut, Delaware, Georgia, and Massachusetts are among the most predominant serotypes in the United States. Even though vaccination is widely used, the high mutation rate exhibited by IBV is continuously triggering the emergence of new viral strains and hindering control and prevention measures. For that reason, targeted strategies based on constantly updated information on the IBV circulation are necessary. Here, we sampled IBV-infected farms from one US state and collected and analyzed 65 genetic sequences coming from three different lineages along with the immunization information of each sampled farm. Phylodynamic analyses showed that IBV dispersal velocity was 12.3 km/year. The majority of IBV infections appeared to have derived from the introduction of the Arkansas DPI serotype, and the Arkansas DPI and Georgia 13 were the predominant serotypes. When analyzed against IBV sequences collected across the United States and deposited in the GenBank database, the most likely viral origin of our sequences was from the states of Alabama, Georgia, and Delaware. Information about vaccination showed that the MILDVAC-MASS+ARK vaccine was applied on 26% of the farms. Using a publicly accessible open-source tool for real-time interactive tracking of pathogen spread and evolution, we analyzed the spatiotemporal spread of IBV and developed an online reporting dashboard. Overall, our work demonstrates how the combination of genetic and spatial information could be used to track the spread and evolution of poultry diseases, providing timely information to the industry. Our results could allow producers and veterinarians to monitor in near-real time the current IBV strain circulating, making it more informative, for example, in vaccination-related decisions.

Epidemiological, clinical, and public health response characteristics of a large outbreak of diphtheria among the Rohingya population in Cox's Bazar, Bangladesh, 2017 to 2019: A retrospective study

BACKGROUND

Unrest in Myanmar in August 2017 resulted in the movement of over 700,000 Rohingya refugees to overcrowded camps in Cox's Bazar, Bangladesh. A large outbreak of diphtheria subsequently began in this population.

METHODS AND FINDINGS

Data were collected during mass vaccination campaigns (MVCs), contact tracing activities, and from 9 Diphtheria Treatment Centers (DTCs) operated by national and international organizations. These data were used to describe the epidemiological and clinical features and the control measures to prevent transmission, during the first 2 years of the outbreak. Between November 10, 2017 and November 9, 2019, 7,064 cases were reported: 285 (4.0%) laboratory-confirmed, 3,610 (51.1%) probable, and 3,169 (44.9%) suspected cases. The crude attack rate was 51.5 cases per 10,000 person-years, and epidemic doubling time was 4.4 days (95% confidence interval [CI] 4.2-4.7) during the exponential growth phase. The median age was 10 years (range 0-85), and 3,126 (44.3%) were male. The typical symptoms were sore throat (93.5%), fever (86.0%), pseudomembrane (34.7%), and gross cervical lymphadenopathy (GCL; 30.6%). Diphtheria antitoxin (DAT) was administered to 1,062 (89.0%) out of 1,193 eligible patients, with adverse reactions following among 229 (21.6%). There were 45 deaths (case fatality ratio [CFR] 0.6%). Household contacts for 5,702 (80.7%) of 7,064 cases were successfully traced. A total of 41,452 contacts were identified, of whom 40,364 (97.4%) consented to begin chemoprophylaxis; adherence was 55.0% (N = 22,218) at 3-day follow-up. Unvaccinated household contacts were vaccinated with 3 doses (with 4-week interval), while a booster dose was administered if the primary vaccination schedule had been completed. The proportion of contacts vaccinated was 64.7% overall. Three MVC rounds were conducted, with administrative coverage varying between 88.5% and 110.4%. Pentavalent vaccine was administered to those aged 6 weeks to 6 years, while tetanus and diphtheria (Td) vaccine was administered to those aged 7 years and older. Lack of adequate diagnostic capacity to confirm cases was the main limitation, with a majority of cases unconfirmed and the proportion of true diphtheria cases unknown.

CONCLUSIONS

To our knowledge, this is the largest reported diphtheria outbreak in refugee settings. We observed that high population density, poor living conditions, and fast growth rate were associated with explosive expansion of the outbreak during the initial exponential growth phase. Three rounds of mass vaccinations targeting those aged 6 weeks to 14 years were associated with only modestly reduced transmission, and additional public health measures were necessary to end the outbreak. This outbreak has a long-lasting tail, with Rt oscillating at around 1 for an extended period. An adequate global DAT stockpile needs to be maintained. All populations must have access to health services and routine vaccination, and this access must be maintained during humanitarian crises.

Systematic review of Integrated Disease Surveillance and Response (IDSR) implementation in the African region

BACKGROUND

The WHO African region frequently experiences outbreaks and epidemics of infectious diseases often exacerbated by weak health systems and infrastructure, late detection, and ineffective outbreak response. To address this, the WHO Regional Office for Africa developed and began implementing the Integrated Disease Surveillance and Response strategy in 1998.

OBJECTIVES

This systematic review aims to document the identified successes and challenges surrounding the implementation of IDSR in the region available in published literature to highlight areas for prioritization, further research, and to inform further strengthening of IDSR implementation.

METHODS

A systematic review of peer-reviewed literature published in English and French from 1 July 2012 to 13 November 2019 was conducted using PubMed and Web of Science. Included articles focused on the WHO African region and discussed the use of IDSR strategies and implementation, assessment of IDSR strategies, or surveillance of diseases covered in the IDSR framework. Data were analyzed descriptively using Microsoft Excel and Tableau Desktop 2019.

RESULTS

The number of peer-reviewed articles discussing IDSR remained low, with 47 included articles focused on 17 countries and regional level systems. Most commonly discussed topics were data reporting (n = 39) and challenges with IDSR implementation (n = 38). Barriers to effective implementation were identified across all IDSR core and support functions assessed in this review: priority disease detection; data reporting, management, and analysis; information dissemination; laboratory functionality; and staff training. Successful implementation was noted where existing surveillance systems and infrastructure were utilized and streamlined with efforts to increase access to healthcare.

CONCLUSIONS AND IMPLICATIONS OF FINDINGS

These findings highlighted areas where IDSR is performing well and where implementation remains weak. While challenges related to IDSR implementation since the first edition of the technical guidelines were released are not novel, adequately addressing them requires sustained investments in stronger national public health capabilities, infrastructure, and surveillance processes.

A Conceptual Model for Geo-Online Exploratory Data Visualization: The Case of the COVID-19 Pandemic

Responding to the recent COVID-19 outbreak, several organizations and private citizens considered the opportunity to design and publish online explanatory data visualization tools for the communication of disease data supported by a spatial dimension. They responded to the need of receiving instant information arising from the broad research community, the public health authorities, and the general public. In addition, the growing maturity of information and mapping technologies, as well as of social networks, has greatly supported the diffusion of web-based dashboards and infographics, blending geographical, graphical, and statistical representation approaches. We propose a broad conceptualization of Web visualization tools for geo-spatial information, exceptionally employed to communicate the current pandemic; to this end, we study a significant number of publicly available platforms that track, visualize, and communicate indicators related to COVID-19. Our methodology is based on (i) a preliminary systematization of actors, data types, providers, and visualization tools, and on (ii) the creation of a rich collection of relevant sites clustered according to significant parameters. Ultimately, the contribution of this work includes a critical analysis of collected evidence and an extensive modeling effort of Geo-Online Exploratory Data Visualization (Geo-OEDV) tools, synthesized in terms of an Entity-Relationship schema. The COVID-19 pandemic outbreak has offered a significant case to study how and how much modern public communication needs spatially related data and effective implementation of tools whose inspection can impact decision-making at different levels. Our resulting model will allow several stakeholders (general users, policy-makers, and researchers/analysts) to gain awareness on the assets of structured online communication and resource owners to direct future development of these important tools.