Validating archetypes for the Multiple Sclerosis Functional Composite
1 Center for Medical Biometry and Medical Informatics, Medical Center – University of Freiburg, Stefan-Meier-Str. 26, 79104 Freiburg, Germany
2 NeuroCure Clinical Research Center, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany
3 Motognosis UG, Debenzer Str. 73, 12683 Berlin, Germany
4 Institute for Medical Informatics, Statistics and Documentation; Medical University of Graz, Auenbruggerplatz 2, 8036 Graz, Austria
BMC Medical Informatics and Decision Making 2014, 14:64 doi:10.1186/1472-6947-14-64Published: 3 August 2014
Numerous information models for electronic health records, such as openEHR archetypes are available. The quality of such clinical models is important to guarantee standardised semantics and to facilitate their interoperability. However, validation aspects are not regarded sufficiently yet. The objective of this report is to investigate the feasibility of archetype development and its community-based validation process, presuming that this review process is a practical way to ensure high-quality information models amending the formal reference model definitions.
A standard archetype development approach was applied on a case set of three clinical tests for multiple sclerosis assessment: After an analysis of the tests, the obtained data elements were organised and structured. The appropriate archetype class was selected and the data elements were implemented in an iterative refinement process. Clinical and information modelling experts validated the models in a structured review process.
Four new archetypes were developed and publicly deployed in the openEHR Clinical Knowledge Manager, an online platform provided by the openEHR Foundation. Afterwards, these four archetypes were validated by domain experts in a team review. The review was a formalised process, organised in the Clinical Knowledge Manager. Both, development and review process turned out to be time-consuming tasks, mostly due to difficult selection processes between alternative modelling approaches. The archetype review was a straightforward team process with the goal to validate archetypes pragmatically.
The quality of medical information models is crucial to guarantee standardised semantic representation in order to improve interoperability. The validation process is a practical way to better harmonise models that diverge due to necessary flexibility left open by the underlying formal reference model definitions.
This case study provides evidence that both community- and tool-enabled review processes, structured in the Clinical Knowledge Manager, ensure archetype quality. It offers a pragmatic but feasible way to reduce variation in the representation of clinical information models towards a more unified and interoperable model.