Legacy conversion of daily general practice data registries: from the Belgian thesaurus to a SNOMED-CT GP Refset

BACKGROUND: Currently, a national thesaurus is implemented in all seven Belgian GP Electronic Health Records (EHRs). It was developed 20 years ago and consists of 50.000 Dutch and French clinical terms which are linked to ICPC-2 and ICD-10. However, since 2009 this thesaurus has not been maintained and the Belgian government has chosen SNOMED-CT to be the health care reference terminology.QUESTIONS: Challenges are to preserve the existing data in the EHRs and to ensure data interoperability for continuity of care through all medical professionals. The aim was to convert the most used clinical... Mehr ...

Verfasser: Fauquert, Benjamin
Erscheinungsdatum: 2023
Schlagwörter: Systematized Nomenclature of Medicine / Vocabulaire controllé
Sprache: Englisch
Permalink: https://search.fid-benelux.de/Record/base-27303471
Datenquelle: BASE; Originalkatalog
Powered By: BASE
Link(s) : http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/331255

BACKGROUND: Currently, a national thesaurus is implemented in all seven Belgian GP Electronic Health Records (EHRs). It was developed 20 years ago and consists of 50.000 Dutch and French clinical terms which are linked to ICPC-2 and ICD-10. However, since 2009 this thesaurus has not been maintained and the Belgian government has chosen SNOMED-CT to be the health care reference terminology.QUESTIONS: Challenges are to preserve the existing data in the EHRs and to ensure data interoperability for continuity of care through all medical professionals. The aim was to convert the most used clinical terms of 3BT to SNOMED-CT to create a Belgian SNOMED-CT General Practice Reference set (Belgian GP Refset).CONTENT: We extracted clinical terms that were registered more than once in three years within a sample of six primary care health centers. These terms were sorted according to the International Classification of Primary Care (ICPC-2) and mapped to SNOMED-CT using a blinded dual mapping process.In total, 8250 of 11099 (74,3%) extracted terms were exactly matched to SNOMED-CT concepts. The lowest percentage of exactly matched terms (54,2%, n=247) was observed for chapter Z (ICPC-2 social problems) and the highest (83,2%, n=781) for chapter N (neurological problems). These results were consistent with previous internationally published work. Incorporating non-exact matches (narrower, broader, or lacking mapping) is still in progress. Other work to ensure its usability and quality are to be done.Message for Practice: The GP Refset is distributed as part of the Belgian SNOMED-CT extension. ; info:eu-repo/semantics/published