Getting started with the new Cohort Discovery Service
23 June 2026 | Author: Ruth Milne, Communications Manager, Infrastructure and Services
In this blog, we explore the latest improvements to the Cohort Discovery Service and how the redesigned experience helps researchers identify relevant patient cohorts more quickly, easily and efficiently.
Share this page
Finding out where the right patient cohort exists for your research can often feel complicated before a study has even begun.
The new Cohort Discovery Service, available through the , has been redesigned to make that process simpler, faster and more intuitive 鈥 whether you鈥檙e completely new to Cohort Discovery or an experienced researcher building complex queries.
Here鈥檚 a look at what鈥檚 changed, and how the new service is helping researchers move from initial idea to meaningful insights more quickly, and with less complexity along the way.
Start with everyday language, not technical terms
One of the biggest changes is the introduction of natural language search. Instead of needing detailed knowledge of the (OMOP CDM) terminology or complex query logic, researchers can now begin a search using plain English. For example, typing 鈥渁dults under 35 with asthma鈥 into the search bar.

The service automatically translates this into relevant clinical concepts and helps users refine the meaning as they go. From there, additional filters can be applied through the interface such as age ranges, event dates, inclusion or exclusion criteria, or additional conditions through the interface.
This means researchers can start exploring cohort feasibility immediately, without needing specialist technical knowledge to get started. Once a search has started, the platform also provides support throughout the query-building process.
Guidance that appears when you need it
The new interface includes built-in support throughout the query-building process. As users move through the service, the guidance panel updates dynamically depending on what they are doing. Whether refining a condition, adding filters or building more advanced queries, relevant help appears automatically alongside the workflow.

There are also videos, guidance panels, and help shortcuts available throughout the platform, helping users feel supported and making it easier for first-time users to explore the service independently.

Build simple or advanced queries
The redesigned query builder supports both simple and highly detailed cohort searches. Users can:
- start with natural language
- build queries manually
- combine multiple conditions
- select several matching search terms at once (for conditions, observations, etc.)
- apply age and date filters
- create grouped logic using AND/ OR rules
- use OMOP vocabulary-based searching where needed
This balance between simplicity and flexibility was a key focus of the redesign. A researcher can begin with a simple, plain English language query, whereas experienced users might build more advanced, logic-based searches using advanced tools when needed.
The redesign also focuses on helping researchers work more flexibly once queries have been created.
Compare and revisit searches more easily
The updated query experience also makes it easier to refine and compare cohort queries. Users can:
- save, export, and revisit previous searches
- open multiple query tabs at once to easily compare results
- duplicate and refine previous queries
- re-run previous queries against newly available and/ or updated datasets
This makes it easier to test ideas, refine questions, and return to previous work without needing to rebuild searches from scratch. The goal is to support rapid feasibility assessment, helping researchers understand whether suitable cohorts exist before beginning full data access requests.
A clearer path from discovery to data access
Once users identify relevant cohorts, they can move directly to the corresponding dataset metadata pages on the Gateway. The experience has been streamlined to make next steps clearer and more consistent, helping users move from discovering potential cohorts to exploring relevant datasets and beginning formal data access requests directly from the Gateway. Rather than presenting multiple possible actions, the service now guides users toward the most relevant next step in the research journey.

Open source and community-driven
Another significant development is that the new Cohort Discovery Service is now open source. Organisations can use the platform in either an integrated model connected to a Gateway-style system, or a standalone deployment within their own environment.
This means collaborators can adopt, extend, and improve the service for their own research ecosystems without needing to build a cohort discovery platform from scratch, for example for their own rapid internal feasibility searches in response to user queries.
Designed to keep improving
The new Cohort Discovery Service has been built to make cohort feasibility assessment more accessible, intuitive and efficient for researchers at every stage of the process. From natural language searching and dynamic guidance to flexible query building and improved workflows, the service is designed to help users spend less time navigating technical complexity and more time exploring research possibilities.
As the platform continues to evolve through user feedback and ongoing development, future enhancements will further support clinical trial planning, dataset exploration and collaborative research across the health data ecosystem.
Ultimately, the goal is simple: making it easier for researchers to discover whether the right data exists to support the next stage of a study.
Book a demo
Explore what’s possible with Cohort Discovery and book a personalised demo to learn how the service can help you build queries, explore potential patient populations and assess research feasibility with greater ease.
![]()