Cohort Discovery

The Cohort Discovery Service helps researchers quickly assess whether relevant patient cohorts exist across multiple datasets in one place. It provides a faster, more confident starting point for planning research before submitting a data access request.

Available through the , Cohort Discovery supports efficient discovery and responsible access to UK health data.

A smarter first step for research

Before applying for access to health data, researchers often need to answer some basic but critical questions:

Do the right patients exist?
Which datasets are relevant?
Which data custodians should I approach?

The Cohort Discovery Service helps answer these questions early. It enables researchers to assess whether relevant cohorts exist across multiple datasets in near real time, reducing uncertainty and making it easier to plan the next step.

Why use Cohort Discovery?

Run a single search query across multiple datasets
Get an early indication of approximate cohort size
Identify which data custodians hold relevant data for your research
Make more informed, targeted data access requests

1. Create an account: Sign up on the Health Data Research Gateway to begin.
2. Apply to access the service: Click the ‘Access Cohort Discovery’ button (top right of the page, ) and follow the prompts. You must be approved to use the service before being able to run any searches.
3. Define your cohort: Use the intuitive query builder to describe the patient population needed for your research.
4. Run your query: Your search is securely executed in real-time across multiple pseudonymised datasets available to you.
5. Review your results: Get rounded, privacy-safe counts to help inform your data access requests.
Cohort Discovery is designed to protect privacy at every stage:
- Researchers do not see patient-level data
- Queries are run on pseudonymised data that never moves or leaves its host organisation
- Results are returned as rounded, non-identifiable counts
- Access is permission-based and governed
The service is supported by a federated analytics ecosystem that enables queries to run securely within Secure Data Environments (Trusted Research Environments), including tools such as , developed by the University of Nottingham as part of the 51�� Federated Analytics programme.

This means researchers can explore potential cohorts safely while giving data custodians confidence in robust privacy safeguards.
Cohort Discovery is built on consistent data standards to enable reliable, scalable search across multiple datasets and environments. The service:
- Supports federated queries across multiple datasets
- Uses the OMOP Common Data Model to enable comparable cross-dataset searches
- Aligns with UK health data standards and governance frameworks
- Designed to scale across multiple secure data environments
Researchers

Cohort Discovery helps researchers:
- Validate research questions earlier
- Assess cohort availability across multiple datasets
- Reduce speculative enquiries
- Move forward more effectively with their research
The service is a simple, structured starting point for planning studies using UK health data.

Data custodians

By participating in Cohort Discovery, data custodians can help improve the efficiency and quality of research engagement. Benefits include:
- More targeted and relevant data access requests
- Reduced administrative burden
- Strong governance and privacy protections
- Increased visibility of datasets to the research community
We work with a range of trusted partners to deliver the Cohort Discovery Service, including:
- University of Nottingham (technology partner)
  Provides the to enable federated search queries.
- (technology partner)
  Provides the Insight tool to enable federated search queries
- The (collaborator)
- Data custodians (delivery partners)
  Organisations contributing datasets to Cohort Discovery.
- CO-CONNECT (foundational programme)
  The service builds on the CO-CONNECT programme, which enabled researchers to rapidly discover and access COVID-19 data while while ensuring patient information remained private and secure.