Abstract
Health Data are collected or repurposed for research in many organisations across the UK. These data are held in many formats and at many scales. Such data can contain information on biometric measurements, medical conditions, medical procedures, demographics, and prescribed medications with several coding systems in use. For GDPR compliance, data are stored securely and de-identified prior to use in research making sharing and standardisation more difficult.
The Observational Health Data Sciences and Informatics (OHDSI) program’s Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) provides standard data objects and coding for health data.
CaRROT-CDM Open-Source tools were developed to define mappings and transform data from numerous disparate organisations (the Data Partners) to the OMOP standard to allow federated data queries via the Health Data Research UK (HDR-UK) gateway. These tools are CaRROT-Mapper and CaRROT-CDM, with the latter being the focus here. CaRROT-CDM takes mapping information from the CaRROT-Mapper and source data as input. It produces OMOP CDM format data as output. The tool was designed to run in Data Partners’ compute environments, reducing the governance burden for transforming data to the common standard. Sensitive, identifiable data are never seen by the development team.
Conclusions and Current Work The CaRROT-CDM tool is in use at Data Partner sites and has transformed over 20 COVID-19 and Pain data sets, so far, which are available for aggregated query via the HDR UK innovation gateway. Current work is focussed on re-engineering and testing the software for scaling to population-level health data.
The Observational Health Data Sciences and Informatics (OHDSI) program’s Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) provides standard data objects and coding for health data.
CaRROT-CDM Open-Source tools were developed to define mappings and transform data from numerous disparate organisations (the Data Partners) to the OMOP standard to allow federated data queries via the Health Data Research UK (HDR-UK) gateway. These tools are CaRROT-Mapper and CaRROT-CDM, with the latter being the focus here. CaRROT-CDM takes mapping information from the CaRROT-Mapper and source data as input. It produces OMOP CDM format data as output. The tool was designed to run in Data Partners’ compute environments, reducing the governance burden for transforming data to the common standard. Sensitive, identifiable data are never seen by the development team.
Conclusions and Current Work The CaRROT-CDM tool is in use at Data Partner sites and has transformed over 20 COVID-19 and Pain data sets, so far, which are available for aggregated query via the HDR UK innovation gateway. Current work is focussed on re-engineering and testing the software for scaling to population-level health data.
| Original language | English |
|---|---|
| DOIs | |
| Publication status | Published - 5 Sept 2023 |
| Event | Research Software Engineering Conference 2023 - Bay Campus, Swansea University, Swansea, United Kingdom Duration: 4 Sept 2023 → 7 Sept 2023 Conference number: 7 https://rsecon23.society-rse.org/ |
Conference
| Conference | Research Software Engineering Conference 2023 |
|---|---|
| Abbreviated title | RSECON23 |
| Country/Territory | United Kingdom |
| City | Swansea |
| Period | 4/09/23 → 7/09/23 |
| Other | RSECon23 has now finished! RSECon23 was held at the Great Hall on the Bay Campus of Swansea University, with Ed Bennett as our Programme Chair. The conference took place in the week beginning Monday 4th September with satellite events on the Monday and Friday, and the main conference starting on Tuesday 5th September. |
| Internet address |