OMOP Frequently Asked Questions

The OMOP Common Data Model (CDM) is an open, international standard for organising health data, maintained by the OHDSI community. It maps each local code to one standard concept, so the same analysis code can run on every OMOP-mapped dataset. Read more about OMOP.

Consistent data makes analyses comparable and methods reusable across institutions. TRUST adopted the OMOP CDM in 2022, and the aim is for OMOP to become the default for research cohorts in Singapore where relevant, so that data is analysis-ready from day one.

OMOP CDM v5.5. The NDCT workflow and the Deploying OMOP guide are built on it. See the OMOP CDM v5.5 specification ↗ (opens in new tab).

No. Existing TRUST members can use OMOP Concepts Search, the OMOP Concept Search API and the Community of Practice pages with their current login.

Researchers from organisations with a Data Request Agreement with TRUST can register as a TRUST member. If you do not meet this requirement, contact us.

Through the TRUST Members' Portal ↗ (opens in new tab). OMOP Concepts Search is in the Metadata Catalogue, and the National Data Curation Team's pages, including the Community of Practice and the Technical Working Groups, are in the portal's Pages menu.

Singapore's first harmonised OMOP vocabulary, built with the research ecosystem. It brings together source-to-OMOP mappings contributed by participating institutions and projects, so programmes can reuse each other's work instead of mapping from scratch.

The search tool for the vocabulary, in the TRUST Metadata Catalogue. Search by concept name, description, local variable name or concept ID, and filter by domain and source vocabulary.

Version v1.1, index built 06 October 2026. The current version is always shown at the top of OMOP Concepts Search.

The local variables from Singapore cohorts that are already mapped to that concept. If one looks like your variable, you have probably found the right concept. Read more about how to read a concept record.

Work through this order. First try a broader or alternative term in OMOP Concepts Search, clear the domain filter, and search your variable name to see how other cohorts handled something similar. If there's no suitable Singapore mapping, search ATHENA ↗ (opens in new tab) for a valid, standard OMOP concept. For a long list of terms, Usagi ↗ (opens in new tab) can suggest concepts in bulk. If nothing suitable turns up, send it to NDCT at the NDCT common mailbox, with the source term or code, any existing mapping, your proposed alternative and why. More tips.

A way to look up concepts from your own mapping and ETL pipelines. It searches the same index as OMOP Concepts Search and returns results as JSON. Read more about the API.

Sign in to the members' portal, go to Account › API keys, and generate a key for OMOP concept search. Copy it straight away, as it is shown only once.

In the TRUST Members' Portal: OMOP Concept Search API reference ↗ (opens in new tab).

Use OMOP Concepts Search when you are mapping a handful of variables by hand. Use the API when you have many variables, or want lookups built into a pipeline.

The API is in beta and may change as it improves. Changes will be announced in the members' portal.

Read Deploying OMOP for an overview of the four steps: clean the data, map it, Extract, Transform and Load (including data quality checks), then maintain, share and reuse. Detailed guidance is in the members' portal.

Map to a valid, standard concept wherever possible, check that its domain fits what your variable records, and note how close the match is: equal, equivalent, wider or narrower. Have a domain expert review key mappings.

It depends on the domain. SNOMED for clinical conditions, most observations and procedures, and some measurements. LOINC for laboratory tests and measurements. RxNorm or RxNorm Extension for medications. UCUM for units of measurement.

It depends on what you are doing. Converting data into the OMOP CDM needs one standard concept per source value. Research analysis often needs a broader concept set, for example all concepts under a parent condition, to find the right population or events.

Follow its Maps to relationship in ATHENA to the standard concept, and use that. Data in the OMOP CDM needs standard concepts. If there's no Maps to relationship, send the case to NDCT at the NDCT common mailbox. NDCT works with the global OHDSI Vocabulary Working Group on new or promoted concepts.

The National Data Curation Team advises programmes that are ready to start. Email NDCT at the NDCT common mailbox for guidance on mapping, ETL design, data standardisation and data quality.

Yes. Open-source OHDSI tools: ATHENA and Usagi for mapping, WhiteRabbit and Rabbit-in-a-Hat for designing the Extract, Transform and Load process, and the Data Quality Dashboard for quality checks. They are listed in the portal's OMOP Resources ↗ (opens in new tab).

The community and outreach arm of the National Data Curation Team (NDCT). It brings together healthcare institutions, researchers, data engineers and informaticians through meetings, events and announcements, and is the way in to NDCT initiatives such as the Technical Working Groups.

The CoP is in the TRUST Members' Portal, so TRUST members already have access. To get involved, email the NDCT team.

Email NDCT at the NDCT common mailbox. The community welcomes source-to-OMOP mappings, technical guidance, ETL and data quality resources, research use cases, training materials and lessons learned.

Seven groups, run by NDCT, that bring members together on Data Standardisation, ETL & Implementation, Data Quality & Governance, Clinical Research, International Engagement, Education & Training, and Agentic AI. Members can sign up through the registration form on the Technical Working Groups page.