TEHDAS - WP6 - D6.2 - Recommendations to enhance interoperability within HealthData@EU

Description

Abstract (English)

Executive summary

This report aims to provide guidance on the use of selected well-established interoperability standards, ultimately to enhance interoperability in the context of HealthData@EU. These recommendations are one of the pieces of the TEHDAS Data Quality Framework.

Interoperability standards play a significant role in some of the stages of the data life cycle; in particular, in the data preparation process, in the publication of data collections, and in the connection between nodes in the HealthData@EU. This report provides assessment and recommendations standards that:

1) provide a layer of semantic interoperability upon the different taxonomies and vocabularies used when collecting data for primary purposes;

2) facilitate the cataloguing of data sources and data collections in a way that enables findability in programmatic querying; and

3) allow communication with preserved meaning between nodes when health data access bodies grant access to data.

Finally, the recommendations of this report are primarily aimed at those institutions that are expected to prepare data for secondary use (e.g., data holders, and health data access bodies).

Methodology

After the selection of international well-established standards with potential for secondary use was decided, three observers assessed the principles of interoperability using the Common Assessment Method for Standards and Specifications (CAMSS), which is the European guide for assessing and selecting standards and specifications for eGovernment projects. CAMSS evaluates 16 criteria as openness, transparency, reusability, technical neutrality and portability; generic use and how they address users' needs and expectations; potential for cooperation; and the principles of interoperability in the European Interoperability Framework (EIF).

After assessment, each criterion and the standard itself achieve a score. These scores provide a notion of how well the standards address the interoperability requirements and, consequently, what the strengths and weaknesses are.

Results

This report evaluates nine standards for discoverability, five standards meant to enable semantic interoperability and five aiming the interoperable communication between nodes

Standards for discoverability

Overall, among the studied standards, DCAT-AP and INSPIRE were considered best equipped for data discoverability, as per CAMMS evaluation.

These standards, developed in the context of OpenData and Public Administration, are however not so well equipped to respond to the needs of the research communities. Conversely, standards developed in the context of research communities, that ranked lower in reusability or in the assessment of effectiveness and efficiency, ranked at the top with regard to user-centric approach. This finding highlights the actual relevance of these standards in data discoverability for specific research communities - i.e., data types such as biosamples, gene sequences, clinical research data, medical records, etc.

None of the above standards are part of a national framework or national strategy in the context of health data or has been widely adopted as a standard for discoverability. Importantly, DCAT-AP, INSPIRE and CESSDA have developed some mutual collaboration.

Standards enabling semantic interoperability

All the standards analysed got a similar interoperability overall score (around 80%), ranging between 74% in Orphanet standards and 82% in OMOP-CDM.

The adoption of these standards is quite uneven. While in the case of SNOMED CT and LOINC, there is wide experience across Europe, the case of Orphanet standards is more limited, and the case of OMOP-CDM is essentially linked to research projects on specific domains, although ranks the highest in the principles for cooperation among public institutions. CDISC-STDM is not used in any of the countries that provided insight.

Importantly, SNOMED CT is being mapped to Orphanet standards, OMOP-CDM, CDISC-SDTM, and LOINC have joined SNOMED CT. In addition, all the ICD and ICD-O are mapped to SNOMED CT, as well as, is being used in the Human Phenotype Ontology and in the GLobal Alliance for Genomics and Health. Finally, SNOMED CT is ISO-IDMP compliant allowing the extension of EMA case safety reports.

Standards meant interoperable communication

Overall, except for IDMP, the standards of communication scored 90% and over in accordance with the CAMSS assessment demonstrating suitability for interoperable communication within the HealthData@EU.

DICOM and HL7-FIHR both scored 99% while ISO-8000-110 reached 96% reflecting a lower score in principles for cooperation across institutions. IDMP´s low figures reflect issues concerning transparency, reusability, security and privacy and a lack of assessment of effectiveness and efficiency.

As per the adoption, DICOM has been widely adopted in all the countries surveyed in this study, and the same largely applies to HL7-FHIR. This is not the case for ISO 8000-110 adopted in just one institution of the surveyed countries. IDMP is used in the communication between manufacturers and EMA, but it is not mentioned at the country level. Importantly both DICOM and HL7-FHIR have set up collaborations between them and with the aforementioned semantic standards. IDMP has focused on SNOMED CT and HL7-FHIR. The level of cooperation is uneven. Finally, DICOM has established cooperation with the Cancer Genome Atlas program in the USA.

Other (English)

Disclaimer

The content of this deliverable represents the views of the author(s) only and is his/her/their sole responsibility; it cannot be considered to reflect the views of the European Commission and/or the Consumers, Health, Agriculture and Food Executive Agency or any other body of the European Union. The European Commission and the Agency do not accept any responsibility for the use of its contents.