Our Data
Find out more about our data.
License Details
All visualisations and data produced by Pandemic PACT are open access under the Creative Commons BY license. You have permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited. All the software and code that we write is open source and made available via GitHub under the MIT license.
Citation
Pandemic PACT Research Programme, Grant and Evidence Gap Tracker by the Pandemic PACT Team with GloPID-R and UKCDR.
Academic: Norton, Alice; Sigfrid, Louise; Antonio, Emilia; Bucher, Adrian; Ndwandwe Duduzile et al. Improving coherence of global research funding: Pandemic PACT. The Lancet, Volume 403, Issue 10433, 1233.
Protocol for the Pandemic PACT Database and Dashboard
We have published a protocol on the Wellcome Open Research platform describing the metadata and database underpinning Pandemic PACT Research Funding Tracker, as well as outlining the plan for future publications, specifically, the baseline living mapping review analysis and subsequent update analyses. The protocol details our research-funding data collection process, which includes web scraping from agreed funders' websites, data extraction through APIs, direct data provision from select funding organisations. It describes the search strategy, the data extraction approach, and variables included in the database. The protocol further outlines the scope, content and methods used to prepare the living analyses. The research funding dataset and accompanying metadata are available through Figshare.
We are updating our protocol to reflect the expanded scope of the Pandemic PACT programme, including the collection, processing, classification and analysis of clinical research registration data from WHO ICTRP. The revised protocol will be linked here once published. The clinical research registrations dataset and accompanying metadata are available through a separate Clinical Research Registrations repository.
You can read the protocol here.
Pandemic PACT Grant Linkage to PMC Publications/Articles
The explore page of our database is powered by the Articles RESTful API from Europe PMC, which is linked to our collected grant data. This API is utilised to search for publications in the PMC Grist database that match our curated grant data.
We use ORCID IDs, Grant IDs, and/or the names of principal investigators (PIs) as identifiers. These identifiers are crucial for retrieving publications specific to a grant or a researcher, ensuring the explore page effectively showcases relevant information. The availability of these identifiers in our collected data is essential for the functionality of the explore page.
Data Submission Guidance
Data submission guide and template
The Data Submission Template is available here for funders wishing to share data directly with us. Included in the template is a self-explanatory "Data Dictionary (Guide)" tab, which assists data providers by explaining each field in detail. Access can be granted to direct data providers to our data file repository on Figshare, enabling users to submit/upload data seamlessly. For any issues or support related to this process, please contact Thomas Mendy at thomas.mendy@ndm.ox.ac.uk or pandemicpact.info@ndm.ox.ac.uk.
Current data included in the Pandemic PACT Research Funding database and dashboard
The Pandemic PACT research funding dashboard was last updated on 2nd September 2026. In response to the ongoing outbreak of Ebola Bundibugyo Disease, we are updating data for Ebola awards weekly. Ebola awards were last updated on 2nd September 2026.
The current coverage comprises data either scraped from publicly available sources based on search terms outlined in the Pandemic PACT protocol or provided directly by the relevant funders. We currently present data from 1st January 2020 onwards. Data from 350 global funders captured under the COVID CIRCLE initiative (2020 - 2022) are included in our database. We aim to provide a comprehensive representation of data from these funders across all the pandemic-prone diseases in scope despite limited publicly available data for the range of funders captured during the COVID-19 pandemic. We are prospectively collecting data for the full scope of diseases for the funders below. We frequently review various funding sources for inclusion. Further data is under production and the database and dashboard will continue to be expanded and updated
We collect data from a range of funders with different frequencies for updating their databases. Based on these update patterns, we have classified funders into four categories: high, moderate, or low frequency, and unclassified, as shown in the table below. We are monitoring patterns for the “unclassified” funders and will assign them to categories in due course. All funder websites are searched on a rolling four-week cycle*. Data obtained directly from funders is collected every six months.
While our dashboard is regularly updated with new data, there may be a slight delay between the time data is retrieved and when it becomes available on the dashboard (due to data processing). For any specific data queries, please contact the Pandemic PACT team at pandemicpact.info@ndm.ox.ac.uk. *During outbreaks, data is sourced more frequently to ensure timely updates.
Current data included in the Clinical Research Registrations database and dashboard
The pandemic PACT clinical research registrations dashboard was last updated on 2nd September 2026. The dashboard and its accompanying published dataset are normally refreshed monthly as part of the established Pandemic PACT data-update cycle.
The current coverage comprises clinical research registration data sourced from WHO ICTRP, based on search terms outlined in the Pandemic PACT protocol. Coverage spans clinical research registrations from 1 January 2020 onwards, focusing on high-PHEIC-risk pathogens identified by the WHO. COVID-19 is not included as a standalone pathogen. However, multipathogen records that include COVID-19 are captured within the dataset. Updated source files are received weekly from WHO ICTRP and incorporated into the established Pandemic PACT data pipeline. The data was last sourced from ICTRP on (insert date). The records undergo processing, validation, deduplication, coding and quality assurance before publication. Refreshed, coding-complete data are normally published monthly alongside the Research Funding data. During an active outbreak, relevant records are prioritised and published at least weekly, with more frequent updates where required and operationally feasible. These additional releases may also include other coding-complete records that are ready for publication. There may therefore be a delay between a record being added to or updated in WHO ICTRP and its appearance in the Pandemic PACT database and dashboard. For any specific data queries, please contact the Pandemic PACT team at pandemicpact.info@ndm.ox.ac.uk
Data Flow Diagram
Pandemic PACT collects research-funding data from funder websites and databases through web scraping, APIs and file downloads. In some cases, funders provide data directly through an agreed secure-transfer process. Clinical research registration data are sourced primarily from WHO ICTRP through registry searches and feeds, including weekly source files securely supplied to the Pandemic PACT team.
The Research Data Team oversee data intake. Records are processed through a Python-based automated workflow that performs data wrangling and transformation, including automated validation, cleaning, quality checks, standardisation, deduplication and ID assignment. This ensures that records are compatible with the Pandemic PACT standard schema before they are imported into REDCap, which is hosted within the University of Oxford’s network.
Within REDCap, the Research Coding Team reviews imported records and maps them to Pandemic PACT research categories and relevant policy roadmaps. Curated, coding-complete records are then published in their respective dataset folders through the University of Oxford’s Figshare repository for public download and reuse. The datasets, Data Dictionaries, data-flow diagram and related documentation are brought together within the overarching Pandemic PACT Figshare collection. Enovate retrieves published data through the Figshare API, maintains and develops the Pandemic PACT Tracker, and refreshes the dashboard with updated data. Below is the data flow:
Data Dictionary
Pandemic PACT maintains a Data Dictionary for each of its Research Funding and Clinical Research Registrations datasets. The two Data Dictionaries follow consistent naming conventions and share many standard variables, supporting alignment across the wider Pandemic PACT programme. Additional fields are included where required to capture information specific to each dataset, including clinical research registration variables.
Each Data Dictionary provides a comprehensive guide to the fields collected, processed and shared, including data types, possible values and explanations of what each field represents. These resources support the clear and consistent interpretation and reuse of Pandemic PACT data.
The latest Data Dictionaries can be viewed or downloaded from their respective Figshare pages:
You can also explore the Pandemic PACT collection on Figshare, which brings together the datasets, Data Dictionaries, data-flow diagram and related documentation.
Protocol for the Pandemic PACT RRNAs
To conduct this work, we established a platform to appraise existing evidence on WHO priority pathogens and diseases. The method utilises a novel 24-hour relay model with expert systematic reviewers based in different time zones and a software platform.
We have published a protocol on Wellcome Open Research describing the 'living' Rapid Research Needs Appraisal (RRNA) protocol for priority diseases. The protocol details our process, which includes systematically searching for published evidence on peer-reviewed databases and in grey literature, screening of publications for inclusion and data extraction in a global relay model. The protocol also outlines the responsibilities of the stakeholders involved, the research scope, and the software used. A training video explaining our methodology can be accesssed here. The RRNA project was completed in August 2026. Datasets can be downloaded from Figshare.
Implementation of the FAIR Data Principles
At the core of our mission, we embrace the FAIR data principles, ensuring that our data is Findable, Accessible, Interoperable, and Reusable. With support from the Go FAIR Foundation, we created a FAIR Implementation Plan (FIP; to be published soon), which is a list of declared technology choices outlining the intention to implement each of the FAIR Guiding Principles. This list serves as a reference for our data stewardship activities. The intention is that the FIP will be reused and repurposed by other researchers and alike setting up similar databases. These principles guide our approach to data management, making our research data openly available and reusable for the broader scientific community.
For enhanced data exploration all the metadata and the database itself were made machine-actionable. To embody this approach, we have developed a machine-findable and machine-readable Pandemic PACT FAIR Vocabulary, a cornerstone of our data schema for grants. This vocabulary is designed to facilitate a common understanding and enhance the usability of our data across different domains.
Furthermore, the version of metadata schema powering our vocabulary is also available for review in a human user-friendly format. This schema is pivotal in structuring our data to align with the FAIR principles, ensuring that it remains standardised and comprehensive. To access the metadata schema, visit our Figshare repository here.
It's important to note that both the FAIR Vocabulary and its underlying metadata schema are dynamic. Should there be changes in our main PACT schema or Dictionary, these updates will be reflected in the FAIR metadata schema and Vocabulary. This ensures that our resources remain aligned with our most up-to-date PACT schema or Dictionary, maintaining the integrity and utility of our data.