Clinical trials have been in a long-term crisis, in terms of costs and poor recruitment, for the best part of a decade . There has been a steady move away from paper Case Report Forms (CRFs) towards both electronic maintenance of clinical trial data and electronic data capture systems (EDC) . The Learning Health System (LHS) envisages a health care organisation where routine EHR systems are the direct mediators of both re- search and knowledge translation activities .
Material and Methods
TRANSFoRm adopts a dual-level modelling ap- proach to maintain semantic interoperability:
- Clinical Research Information (CRIM) to guide research work ow
- Clinical Data Integration Model (CDIM) to describe and de ne the primary health care domain 
The Operational Data Model (ODM) is a re- search standard for data represenation in CRF. In order to pre-populate data in the ODM le, a query must be issued through the TRANSFoRm infrastructure to fetch data and return it in the right place in the ODM. By adding a <QueryId> child element, a unique identi er can be inserted to link the group with a related speci c query.
The anchor on the research side is the TRANSFoRm Study System (TSS) which is centrally hosted at a secure location. The TSS holds the study information and protocols, de ned via ODM les, as well as acting as the research repository.
When a patient arrives for a con- sultation, his existing EHR data is exported by the EHR to the data node connector (DNC) as an XML document. The XML is mapped through the interoperability framework to CDIM to express a common semantic. The DNC then applies the queries to the XML documents to compute eligibility.
The DNC retrieves the appropri- ate eCRF forms from the study system, transported as HTML forms parameterised for pre- loading and storage of eld values, together with the corre- sponding ODM document. Once, approved, the form is submitted to the DNC where data is in- serted in the ODM le which is sent to the TSS but also to the EHR for auditing purposes.
TRANSFoRm has shown that the process of integrating clinical trial processes and data management into the EHR system can be made simpler and more feasible for EHR vendors, lowering the barriers to adoption and increasing the uptake of the Learning Health System. One virtue of the approach is that it does not require modi cation of ODM, simply exploiting the existing ex- ibility of the standard and the ability to specify external bindings.
The approach we outlined is generalizable to other domains as di erent domain speci c models can be speci ed. We used higher level concepts to ensure future compatibility by using BRIDG for CRIM and building CDIM from middle level ontologies.
Acknowledgment and Contact
This work was supported in part by the European Commission – DG INFSO (FP7 247787).
Jean-François Ethier, [email protected], +33 (0)1 44 27 63 93
- Reynolds, T. Clinical trials: can technology solve the problem of low recruitment? BMJ 2011;34:d3662.
- http://www.cdisc.org/system/ les/all/reference_material_category/application/pdf/esdi.pdf.
- Brendan C. Delaney, Vasa Curcin, Anna Andreasson, et al., “Translational Medicine and Patient Safety in Europe: TRANSFoRm—Architecture for the Learning Health System in Europe,” BioMed Research International, Article ID 961526, in press
- Ethier JF, Curcin V, Barton A, et al. Clinical Data Integration Model: Core Interoperability Ontology for Research Using Primary Care Data. Methods Inf Med 2014; 54(1):16-23.
Saint-Malo 2015, France