Monitoring Outbreak events for Disease surveillance in a data science context 2020-2024: A European epidemic intelligence platform to support infectious diseases surveillance and covariate data accessibility
Prepared by Beatriz Sarabia (MOOD Communications Officer, Opengeohub Foundation), Elena Arsevska (MOOD coordinator – CIRAD), Gülten Inan (Communications Assistant, Opengeohub Foundation)
The MOOD project in a nutshell:
There is a growing risk of the global spread of emerging zoonotic infectious disease (EIDs) pathogens due to climate change, animal and human mobility, growing populations and urbanisation. Detecting and assessing the risk of EIDs to public and animal health is thus crucial. The MOnitoring Outbreak Events for Disease Surveillance in a Data Science Context (MOOD) project, funded by the Horizon Europe 2020 Program (2020-2024), has focused on improving the utility of indicator-based (IBS) and event-based surveillance (EBS) epidemiological by combining with environmental, climate and host distribution factors to predict the risk of emergence in Europe. Over the past four years and in close collaboration with more than 20 partners, the MOOD project has aided in understanding the dynamics of modelling and risk assessment of infectious diseases in Europe.
Did you know that the MOOD project was one of the first to tackle the disease modelling challenges posed by the COVID-19 pandemic?
At the start of the COVID-19 pandemic in 2020, it became of utmost importance to the entire research community to provide science-based guidance for governments and health agencies on preventing and controlling the virus from spreading quickly and unpredictably. In its first year of implementation, the project tackled the challenges posed by the COVID-19 pandemic, demonstrating how it could support surveillance and control efforts for ongoing COVID-19 outbreaks in Europe and globally. Sixty-seven scientific publications were dedicated to COVID-19, including a dedicated working group and disease model named Disease X.
From IBS to EBS: A state-of-the-art approach for modelling emerging infectious diseases
Traditional IBS methods, though valuable, lack timeliness for reporting outbreaks and data openness, which are essential in preventing the introduction and spread of disease pathogens. MOOD addressed this gap by developing a robust framework and a platform to enhance the EU’s ability to respond to emerging health threats. The MOOD platform provides health agencies access to IBS and EBS data, various covariates, and other relevant tools for early detection and monitoring, modelling, and assessing health risks.
The MOOD project employed a comprehensive approach to better understand the spatiotemporal and climate change drivers of EIDs in Europe and beyond. By building detailed case studies, the MOOD project provides disease profiles, data, code, and model outputs for six disease groups integrated into the MOOD platform. By providing a complete portfolio per disease group, we facilitate health agencies and researchers’ better understanding of the disease burden, enhancing the overall effectiveness of surveillance and response efforts.
6 Different Case Studies/Profiles:
The first step in this endeavour was to build complete disease profiles/case studies for West Nile Virus Infection (WNV); Tick-borne Encephalitis (TBE); Antimicrobial Resistance (AMR), COVID-19, Chikungunya (CHIK) /Dengue (DEN) and Highly Pathogenic Avian Influenza (HPAI). These case studies offer a comprehensive and valuable overview of the current situation of the disease and perspectives on surveillance, control, and research (including gender aspects) within Europe due to their varied epidemiological profiles and significant impact on public health, the high economic costs associated with their medical care and outbreak control, and their sensitivity to climate and environmental changes.
These diseases also represent diverse transmission systems, i.e., air-borne, vector-borne, multi-transmission routes, and antimicrobial resistance, all of which require unique data streams for effective monitoring and early detection. By focusing on these strategically chosen EIDs, the MOOD project enhanced epidemic intelligence by identifying relevant data flows improving early detection, monitoring, and assessment of these health threats in Europe.
Each prototype disease was profiled based on information collected from a literature review and expert input. A “scoping reviews” scheme was adopted to gather and synthesise the available knowledge. All the disease profiles underwent peer review from experts of the MOOD project and, for some, also external to MOOD to ensure the accuracy and completeness of the information.
“According to the selection methods of the articles, we ensured a qualitative data extraction, and the final results were a summary of evidence from different papers. We calculated the extraction frequency for each covariate to provide their relative importance in disease dynamics. Moreover, data extracted were processed through peer review”, explains Annapaola Rizzoli, Director of the Research and Innovation Centre of the Edmund Mach Foundation. Similarly, the obtained datasets were also used for scientific publication in peer-reviewed journals (i.e., Dagostin et al., 2023 and Cataldo et al., still under review ). On the other hand, epidemiological data included in some of the profiles were obtained upon request from official sources (ECDC, WOAH, FAO, EFSA). To find more information, visit the MOOD project website.Combining case studies and data sets with environmental and socio-economic covariates provides a greater understanding of disease dynamics. Nonetheless, in modern disease surveillance, transparency and accessibility of data are essential for boosting data and tools’ usability. Hence, the MOOD project integrated Open Data principles and the FAIR (Findable, Accessible, Interoperable, and Reusable) principles at its core. Between 2020 and 2024, the MOOD project contributed to 146 publications and 161 datasets and codes, including 108 in top peer-reviewed journals with an impact factor > 10. These outputs demonstrate the project’s efficacy and ensure that the data and insights generated are widely usable and beneficial to various stakeholders in the long run.
The importance of Open Data for Public Health and the challenges of data standardisation
In 4 years, the MOOD project has produced 163 publications that directly contribute to the Pandemic Accord by the WHO, the One Health Plan by the EU, the SDGs, and other Epidemic Intelligence (EI) initiatives. However, its main contribution would be in terms of open data and application of FAIR principles, knowing that the MOOD project published at least 108 publications containing open data sets. This practice is still underdeveloped across the public health field and its interest groups.
Open Data’s potential can only be fully realised through effective data standardisation and integration. Standardisation ensures that data from diverse sources is uniformly formatted, making it easier to compare, combine, and analyse. Integration involves merging these standardised data sets to create a comprehensive and cohesive information resource. Together, these processes enable the seamless sharing and utilisation of data across different platforms and stakeholders.
Public health initiatives like the MOOD project can maximise their impact by ensuring that Open Data is standardised and integrated, providing more accurate and actionable insights for disease surveillance and response. In the MOOD project context, combining data from different sources means that end-users (mainly researchers, public health agencies, and researchers) can build and train their epidemiological models on more significant amounts of consistently compatible data with more relevant explanatory environmental variables. Ultimately, this leads to better and more helpful model projections, which in turn allows the building of improved detection and warning systems for pathogen outbreaks. Additionally, standardisation ensures that all project partners use consistent data, increasing efficiency by centralising data processing. It also improves data reusability, allowing data to be collected once and used multiple times, which is particularly valuable given the high cost of primary data collection.
Nonetheless, the challenges of data standardisation are enormous. Identifying relevant data and sources for spatial (environmental variables) and epidemiological (disease outbreak records) data was complex and required close collaboration with other work packages and end-user surveys. Standards had to be defined to homogenise heterogeneous data, including spatial extent, resolution, cartographic projection for spatial data, and taxonomy for epidemiological data. Developing data standardisation and dataset production pipelines posed technical difficulties, such as processing climatic variables and normalising geolocation for outbreak records needing more geographic coordinates, finally, making datasets and tools accessible involved using repositories like Figshare and Zenodo for environmental data and GitHub for epidemiological data tools, with dissemination via the MOOD platform for easy access and use by researchers.
An iterative process: from co-design to co-creation
A pivotal aspect of the MOOD project is its commitment to co-creating with potential end-users. These professionals, such as researchers, epidemiologists and risk assessors, are part of several public and animal health agencies at the national or European level.
These iterations took place in various scenarios, including interviews, online questionnaires during the annual meeting in June 2023, workshops, testing sessions, summer schools, email exchanges, and one-on-one or group meetings. These diverse formats ensured comprehensive feedback and engagement with potential end-users. Learning loops’ added value compared to top-down approaches lies in considering the innovation needs of European public health and animal health agencies.
By supporting changes in practices to improve EI and early warning of EIDs, the MOOD project addressed a range of expected changes, including technical, organisational, legal, data protection and ethical aspects. The strategy of case studies facilitated interactions with practitioners to adapt technical requirements to their needs. Consequently, the learning loops and initial user-needs assessment were the main drivers of the MOOD platform’s development, ensuring it was co-conceived and well-suited to user needs (Bouyer et al., 2024).Why is there a need for a new EI Platform? Practitioners face an extreme amount of data sourced from multiple sources and formats, and the difficulty of collecting consolidated datasets on disease outbreaks and covariates promptly allows them to proceed to rapid risk assessment and, eventually, decision-making quickly. Considering all the challenges, the platform has been developed to integrate and superpose epidemiological and covariate data layers, along with model results and outputs. Through the platform, practitioners will have access to timely documented high-resolution covariate data and estimates of the risk of disease occurrence.
The MOOD Platform: A Toolbox for Improved Epidemic Intelligence
Did you know that there is a platform that aims to improve the operational abilities of health agencies in detecting, monitoring, and assessing emerging infectious diseases? The vision behind this is the MOOD Platform.
The MOOD platform is a comprehensive toolbox designed to enhance EI by integrating various data sources, models, model outputs and tools to monitor and predict EIDs in Europe. It comprises several modules:
- Module 1: focuses on gathering and standardising data related to environmental, climatic, and socioeconomic factors that influence the spread of diseases and antimicrobial resistance, improving the understanding of drivers of disease emergence.
- Module 2: integrates event-based surveillance (EBS) using the PADI-Web tool to analyse reports and news about potential health risks.
- Module 3: combines environmental, demographic, and mobility data with epidemiological inputs to produce spatiotemporal predictions of disease occurrence in Europe.
To address its primary objectives, the platform focused on data integration and management, enabling large-scale data aggregation from diverse sources, real-time media monitoring to detect emerging threats from online news, a user-friendly interface for querying, visualising and downloading covariate data for public health officials, researchers, and policymakers, and advanced analytical tools, code and model outputs for multiple diseases of importance to Europe. Additionally, the platform prioritised interoperability with existing health systems and security measures to protect sensitive health data.These needs were identified through extensive stakeholder consultations, literature reviews, benchmarking existing platforms, and pilot testing, ensuring the MOOD platform effectively meets the challenges of modern disease surveillance.
The FUTURE of the MOOD:
The long-term sustainability of the MOOD’s work will be achieved by implementing the MOOD Epi-Platform International Non-Profit Association (INPA). The objectives of the INPA, which will shape the future of the MOOD project, can be listed as:
- Promote, maintain and further develop the standardised “MOOD Epi-Platform” that hosts the epidemiological e-tools developed by the MOOD project;
- Encourage the development, hosting and maintenance of additional state-of-the-art epidemiological e-tools and services by its members after the MOOD project is finished;
- Standardise and host relevant state-of-the-art epidemiological e-tools developed by third parties that wish to become a member of the INPA;
- Provide capacity building to support and promote the use of the tools.
Discover the produced materials:
Regarding the final output of a project being a mobile app, web app, platform, or other form of technology, the final results require a tailored communication strategy and action plan to reach its intended audience. For MOOD, capacity-building, hackathons, and summer school events were crucial for the outreach strategy. Not only the events but many other elements were produced as well. Such as:
- Scientific Publications: You can explore all publications the MOOD project contributed to.
- Tutorials from Summer school 2022 and 2023
- Webinars
- Blogs: You can read more than 30 blog posts covering different topics.
- Interviews with Experts: You can listen to the interviews from the International Workshop in Trento, interviews from the Science Webinars, and interviews from the MOOD Scientific Conference in Helsinki.
Last but not least, the contribution:
In conclusion, the MOOD project significantly responded to a need to facilitate data mining and analytical techniques on big data from multiple sources to improve the detection, monitoring, and assessment of EIDs in Europe.
MOOD has a broader scientific and public health impact by embedding Open Data and FAIR principles into its framework. This approach ensures that the data generated is valuable today and will continue to provide insights and drive innovation long into the future, ultimately enhancing the EU’s ability to respond to and manage emerging health threats.
The MOOD platform’s groundbreaking aspect was its ability to provide epidemiologists, risk assessors, decision-makers, disease specialists, and researchers with data and tools to perform EI better. This comprehensive integration allowed for a more holistic understanding of spatial risk mapping, enabling faster and more informed decision-making in the face of emerging health threats.
For more information, please visit our project website and the MOOD Platform and stay tuned for INPA’s (Epi-Platform International Non-Profit Association) activities. During the last week of November (26-27), the project will host its final conference in a public event in Rome, Italy, at the Istituto Superiore di Sanità (ISS), where partners and external participants will learn more about the project’s final outputs.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement MOOD N° 874850.