Domain Modeling and Application Development of an Archetype- and XML-based EHRS : Practical Experiences and Lessons Learnt.

Keywords
Electronic health record, EHR standards, medical informatics applications, health infromation exchange, openEHR, XML

Background
Access to patient data within the hospital or between hospitals is still problematic
since a variety of information systems is in use applying different vendor specific terminologies and
underlying knowledge models. Beyond, the development of electronic health record systems
(EHRSs) is time and resource consuming. Thus, there is a substantial need for a development strategy of standardized EHRSs. We are applying a reuse-oriented process model and demonstrate its feasibility and realization on a practical medical use case, which is an EHRS holding all relevant
data arising in the context of treatment of tumors of the sella region. In this paper, we describe the
development process and our practical experiences.

Methods
Requirements towards the development of the EHRS were collected by interviews with a
neurosurgeon and patient data analysis. For modelling of patient data, we selected openEHR as
standard and exploited the software tools provided by the openEHR foundation. The patient information model forms the core of the development process, which comprises the EHR generation and the implementation of an EHRS architecture. Moreover, a reuse-oriented process model from the business domain was adapted to the development of the EHRS.

Results
The reuse-oriented process model is a model for a suitable abstraction of both, modeling
and development of an EHR centralized EHRS. The information modeling process resulted in 18
archetypes that were aggregated in a template and built the boilerplate of the model driven development. The EHRs and the EHRS were developed by openEHR and W3C standards, tightly supported by well-established XML techniques. The GUI of the final EHRS integrates and visualizes information from various examinations, medical reports, findings and laboratory test results.

Conclusion
We conclude that the development of a standardized overarching EHR and an EHRS is
feasible using openEHR and W3C standards, enabling a high degree of semantic interoperability.
The standardized representation visualizes data and can in this way support the decision process of
clinicians.