Skip to main content

Table 4 Summary of the main challenges involved in mapping registry data to the OMOP CDM, and their solutions

From: Standardizing registry data to the OMOP Common Data Model: experience from three pulmonary hypertension databases

Challenge Solution
Mapping incomplete dates Create standardized imputation rules to ‘restore’ missing dates
Mapping of MedDRA codes (classification concepts) to standard concepts Automated mapping from the unified medical language system to the ICD-10, ICD-10-Clinical Modification or SNOMED vocabulary with crosslinks and name matching, followed by further expert review and additional manual mapping
Mapping of free text e.g. from medication tables Free text extracted, custom mapped and contextualized via CONCEPT_RELATIONSHIP and CONCEPT_ANCESTOR tables
Mapping of non-existing OMOP vocabulary e.g. PH subgroups Used either a combination of SNOMED concepts (e.g. PAH and underlying cause), or a new custom concept (e.g. drug- and toxin-induced PAH)
Mapping information that is related to another piece of information e.g. severity of an event Separate clinical facts are stored in their appropriate domain and a link is added in the FACT_RELATIONSHIP table
Capturing the information that a procedure was not performed This information was excluding from mapping, per the OMOP convention, but is planned for mapping in a future update
  1. ICD International Classification of Disease; MedDRA Medical Dictionary for Regulatory Activities; OMOP CDM Observational Medical Outcomes Partnership common data model; SNOMED Systematized Nomenclature of Medicine Clinical Terms