Open data models for smart health interconnected applications: the example of openEHR.

BACKGROUND: Smart Health is known as a concept that enhances networking, intelligent data processing and combining patient data with other parameters. Open data models can play an important role in creating a framework for providing interoperable data services that support the development of innovative Smart Health applications profiting from data fusion and sharing. METHODS: This article describes a model-driven engineering approach based on standardized clinical information models and explores its application for the development of interoperable electronic health record systems. The following possible model-driven procedures were considered: provision of data schemes for data exchange, automated generation of artefacts for application development and native platforms that directly execute the models. The applicability of the approach in practice was examined using the openEHR framework as an example. RESULTS: A comprehensive infrastructure for model-driven engineering of electronic health records is presented using the example of the openEHR framework. It is shown that data schema definitions to be used in common practice software development processes can be derived from domain models. The capabilities for automatic creation of implementation artefacts (e.g., data entry forms) are demonstrated. Complementary programming libraries and frameworks that foster the use of open data models are introduced. Several compatible health data platforms are listed. They provide standard based interfaces for interconnecting with further applications. CONCLUSION: Open data models help build a framework for interoperable data services that support the development of innovative Smart Health applications. Related tools for model-driven application development foster semantic interoperability and interconnected innovative applications.

Clinical knowledge governance: the international perspective.

As a basis for semantic interoperability, ideally, a Clinical Knowledge Resource for a clinical concept should be defined formally and defined once in a way that all clinical professions and all countries can agree on. Clinical Knowledge Governance is required to create high-quality, reusable Clinical Knowledge Resources and achieve this aim. Traditionally, this is a time-consuming and cumbersome process, relying heavily on face-to-face meetings and being able to get sufficient input from clinicians. However, in a national or even international space, it is required to streamline the processes involved in creating Clinical Knowledge Resources. For this, a Web 2.0 tool that supports online collaboration of clinicians during their creation and publishing of Clinical Knowledge Resources has been developed. This tool is named the Clinical Knowledge Manager (CKM) and supports the development, review and publication of Clinical Knowledge Resources. Also, post-publication activities such as adding terminology bindings, translating the Clinical Knowledge Resource into another language and republishing it are supported. The acceptance of Clinical Knowledge Resources depends on their quality and being able to determine their quality, for example it is important to know that a broad umber of reviewers from various clinical disciplines have been involved in the development of the Clinical Knowledge Resource. We are still far from realizing the vision of a global repository of a great number of reusable, high-quality Clinical Knowledge Resources, which can provide the basis for broad semantic interoperability between systems. However progress towards this aim is being made around the world.

Facilitating secondary use of medical data by using openEHR archetypes.

Clinical trials are of high importance for medical progress. But even though more and more clinical data is available in electronic patient records (EPRs) and more and more electronic data capture (EDC) systems are used in trials, there is still a gap which makes EPR / EDC interoperability difficult and hampers secondary use of medical routine data. The openEHR architecture for Electronic Health Records is based on a two level modeling approach which makes use of 'archetypes'. We want to analyze whether archetypes can help to bridge this gap by building an integrated EPR / EDC system based on openEHR archetypes. We used the 'openEHR Reference Framework and Application' (Opereffa) and existing archetypes for medical data. Furthermore, we developed dedicated archetypes to document study meta data. We developed a first prototype implementation of an archetype based integrated EPR / EDC system. Next steps will be the evaluation of an extended prototype in a real clinical trial scenario. Opereffa was a good starting point for our work. OpenEHR archetypes proved useful for secondary use of health data.

Health informatics - an introduction.

This chapter gives an educational overview of: * the scope of the health informatics discipline * health informatics and e-health definitions * health informatics professional networks * potential benefits of applying health informatics technologies.

Change management - an overview.

This chapter gives an educational overview of: * The need for systematic and comprehensive change management in health that involves everybody * Various change management types, models, best practices, and techniques, and how to plan and execute change management * Some of the common mistakes with change management in healthcare.

Engaging clinicians in clinical content: herding cats or piece of cake?

It is essential that clinicians are able to contribute to the development of clinical content for electronic health records. Clinicians are able to participate meaningfully when the technical focus on the process and the clinical models are reduced or removed. In the openEHR approach - from the original design specification through to the clinical modelling tools - clinicians feature as the primary drivers of clinical content models. As tooling develops and matures to support authoring, reviewing and publishing of openEHR knowledge artefacts, clinicians are increasingly able to be involved - resulting in significant contributions to the development of standardized clinical content models which will support quality of care for their patients.

Archetype-based knowledge management for semantic interoperability of electronic health records.

Formal modeling of clinical content that can be made available internationally is one of the most promising pathways to semantic interoperability of health information. Drawing on the extensive experience from openEHR archetype research and implementation work, we present the latest research and development in this area to improve semantic interoperability of Electronic Health Records (EHRs) using openEHR (ISO 13606) archetypes. Archetypes as the formal definition of clinical content need to be of high technical and clinical quality. We will start with a brief introduction of the openEHR architecture followed by presentations on specific topics related to the management of a wide range of clinical knowledge artefacts. We will describe a web-based review process for archetypes that enables international involvement and ensures that released archetypes are technically and clinically correct. Tools for validation of archetypes will be presented, along with templates and compliance templates. All this in combination enables the openEHR computing platform to be the foundation for safely sharing the information clinicians need, using this information within computerized clinical guidelines, for decision support as well as migrating legacy data.

Towards a comprehensive electronic patient record to support an innovative individual care concept for premature infants using the openEHR approach.

PURPOSE: This paper introduces the modelling of a prototype neonatology electronic patient record (EPR) using openEHR archetypes. The EPR is necessary to support the complex communication tasks of the innovative concept of 'Developmental, Family-Centred, Individual Care of Premature Infants and Newborns' established for the Department of Neonatology at Heidelberg University Hospital. METHODS: The data to be documented was analysed and modelled using the five step openEHR data modelling approach (odma). RESULTS: The analysis revealed a total of 1818 items, which could be arranged into 70 clinical concepts. The items and concepts were then mapped to 132 openEHR archetypes. Fifty-eight of these archetypes could be reused either directly or via specialisation from the existing openEHR archetypes. A further 67 archetypes were newly developed. To combine and constrain archetypes for local settings 16 templates were developed. CONCLUSION: By using the five step openEHR data modelling approach, semantic interoperability, and a reduced need for repeated documentation of the same data can be realised. This is of major importance within the hospital as well as for trans-institutional data exchange.

An archetype-based testing framework.

With the introduction of EHR two-level modelling and archetype methodologies pioneered by openEHR and standardized by CEN/ISO, we are one step closer to semantic interoperability and future-proof adaptive healthcare information systems. Along with the opportunities, there are also challenges. Archetypes provide the full semantics of EHR data explicitly to surrounding systems in a platform-independent way, yet it is up to the receiving system to interpret the semantics and process the data accordingly. In this paper we propose a design of an archetype-based platform-independent testing framework for validating implementations of the openEHR archetype formalism as a means of improving quality and interoperability of EHRs.

Facilitating the openEHR approach - organizational structures for defining high-quality archetypes.

Using openEHR archetypes to establish an electronic patient record promises rapid development and system interoperability by using or adopting existing archetypes. However, internationally accepted, high quality archetypes which enable a comprehensive semantic interoperability require adequate development and maintenance processes. Therefore, structures have to be created involving different health professions. In the following we present a model which facilitates and governs distributed but cooperative development and adoption of archetypes by different professionals including peer reviews. Our model consists of a hierarchical structure of professional committees and descriptions of the archetype development process considering these different committees.

Electronic patient records: moving from islands and bridges towards electronic health records for continuity of care.

OBJECTIVES: Electronic patient record (EPR) systems are increasingly used and have matured sufficiently so as to contribute to high quality care and efficient patient management. Our objective is to summarize current trends and major achievements in the field of EPR in the last year and to discuss their future prospects. RESULTS: Integrating health data from a variety of sources in a comprehensive EPR is a major prerequisite for e-health and e-research. Current research continues to elaborate architectures, technologies and security concepts. To achieve semantic interoperability standards are developed on different levels, including basic data types, messages, services, architectures, terminologies, ontologies, scope and presentation of EPR content. Standards development organisations have started to harmonize their work to arrive at a consensus standard for EPR systems. Integrating the health care enterprise as a whole will optimize efficient use of resources, logistics and scheduling. CONCLUSIONS: The past few years have seen a myriad of developments of EPR systems. However, it is still a long way, until EPR systems can flexibly fulfill all user requirements and an EHR will become broadly accepted. Semantic interoperability will be a key to successful EPR use, especially to avoid double data entries and to better integrate data recording within local workflows. The patient will become an empowered partner, not only by giving him access to his health data. All this will result in enormous quantities of data. Thus, time has come to determine how relevant data can be presented to the stakeholders adequately.

Towards sustainability of health information systems: how can we define, measure and achieve it?

Health information systems (HIS) in their current form are rarely sustainable. In order to sustain our health information systems and with it our health systems, we need to focus on defining and maintaining sustainable Health Information System building blocks or components. These components need to be easily updatable when clinical knowledge (or anything else) changes, easily adaptable when business requirements or processes change, and easily exchangeable when technology advances. One major prerequisite for this is that we need to be able to define and measure sustainability, so that it can become one of the major business drivers in HIS development. Therefore, this paper analyses general definitions and indicators for sustainability, and analyses their applicability to HIS. We find that general 'Emergy analysis' is one possibility to measure sustainability for HIS. Based on this, we investigate major enablers and inhibitors to sustainability in a highlevel framework consisting of four pillars: clinical, technical, socio-technical, and political/business.

Towards semantic interoperability for electronic health records.

OBJECTIVES: In the field of open electronic health records (EHRs), openEHR as an archetype-based approach is being increasingly recognised. It is the objective of this paper to shortly describe this approach, and to analyse how openEHR archetypes impact on health professionals and semantic interoperability. METHODS: Analysis of current approaches to EHR systems, terminology and standards developments. In addition to literature reviews, we organised face-to-face and additional telephone interviews and tele-conferences with members of relevant organisations and committees. RESULTS: The openEHR archetypes approach enables syntactic interoperability and semantic interpretability -- both important prerequisites for semantic interoperability. Archetypes enable the formal definition of clinical content by clinicians. To enable comprehensive semantic interoperability, the development and maintenance of archetypes needs to be coordinated internationally and across health professions. Domain knowledge governance comprises a set of processes that enable the creation, development, organisation, sharing, dissemination, use and continuous maintenance of archetypes. It needs to be supported by information technology. CONCLUSIONS: To enable EHRs, semantic interoperability is essential. The openEHR archetypes approach enables syntactic interoperability and semantic interpretability. However, without coordinated archetype development and maintenance, 'rank growth' of archetypes would jeopardize semantic interoperability. We therefore believe that openEHR archetypes and domain knowledge governance together create the knowledge environment required to adopt EHRs.

Expressing clinical data sets with openEHR archetypes: a solid basis for ubiquitous computing.

PURPOSE: The purpose of this paper is to analyse the feasibility and usefulness of expressing clinical data sets (CDSs) as openEHR archetypes. For this, we present an approach to transform CDS into archetypes, and outline typical problems with CDS and analyse whether some of these problems can be overcome by the use of archetypes. METHODS: Literature review and analysis of a selection of existing Australian, German, other European and international CDSs; transfer of a CDS for Paediatric Oncology into openEHR archetypes; implementation of CDSs in application systems. RESULTS: To explore the feasibility of expressing CDS as archetypes an approach to transform existing CDSs into archetypes is presented in this paper. In case of the Paediatric Oncology CDS (which consists of 260 data items) this lead to the definition of 48 openEHR archetypes. To analyse the usefulness of expressing CDS as archetypes, we identified nine problems with CDS that currently remain unsolved without a common model underpinning the CDS. Typical problems include incompatible basic data types and overlapping and incompatible definitions of clinical content. A solution to most of these problems based on openEHR archetypes is motivated. With regard to integrity constraints, further research is required. CONCLUSIONS: While openEHR cannot overcome all barriers to Ubiquitous Computing, it can provide the common basis for ubiquitous presence of meaningful and computer-processable knowledge and information, which we believe is a basic requirement for Ubiquitous Computing. Expressing CDSs as openEHR archetypes is feasible and advantageous as it fosters semantic interoperability, supports ubiquitous computing, and helps to develop archetypes that are arguably of better quality than the original CDS.

Can design principles of traditional learning theories be fulfilled by computer-based training systems in medicine: the example of CAMPUS.

PURPOSE: Computer-based training (CBT) systems offer the potential to efficiently support modern teaching and learning. However, it is still unknown if a similar efficient learning experience built on sound learning theories and corresponding design principles can be created in the complex health care environment. The purpose of this paper is to analyse to what extent learning theories and corresponding design principles are relevant and can successfully be applied in computer-based training in medicine. METHODS: We use the case-based CBT system CAMPUS as an example for a CBT system currently used to enhance the medical teaching and learning experience. We apply two well-accepted learning theories (Bloom's taxonomy and practice fields) and related design principles to determine to what extent they are relevant and fulfilled in the context of CAMPUS. RESULTS: We demonstrate that in principle these learning theories and design principles can be implemented using computer-based training. However, not all design principles can be fulfilled by the system alone; rather the integration of the system into adequate -- traditional or virtual -- teaching and learning environments is essential. CONCLUSIONS: Traditional learning theories and design principles are a valuable means in designing adequate CBT systems in medicine. They can be successfully implemented in CBT systems for medical education if the system itself is adequately integrated into teaching and learning environments.

Systematic planning of patient records for cooperative care and multicenter research.

PURPOSE: The purpose of this paper is to introduce a method for systematically planning patient records for structured data entry that can be used in cooperative environments (e.g. cooperative care, multicenter trials) in a way that enables multipurpose use and shared data entry. METHODS: Design research, formal logic. RESULTS: The method suggests five steps: analyze the prevailing documentation infrastructure, provide terminology management system (TMS), provide documentation management system (DMS), plan the logical architecture, provide all necessary tools. CONCLUSIONS: The era of eHealth enables cooperative care and collaborative documentation. This can only be efficient if a multiple use and shared entry of data is realized. The task of the medical informatics community is to plan these environments systematically especially in complex environments which are enabled by emerging technologies.

Ubiquitous information for ubiquitous computing: expressing clinical data sets with openEHR archetypes.

Ubiquitous computing requires ubiquitous access to information and knowledge. With the release of openEHR Version 1.0 there is a common model available to solve some of the problems related to accessing information and knowledge by improving semantic interoperability between clinical systems. Considerable work has been undertaken by various bodies to standardise Clinical Data Sets. Notwithstanding their value, several problems remain unsolved with Clinical Data Sets without the use of a common model underpinning them. This paper outlines these problems like incompatible basic data types and overlapping and incompatible definitions of clinical content. A solution to this based on openEHR archetypes is motivated and an approach to transform existing Clinical Data Sets into archetypes is presented. To avoid significant overlaps and unnecessary effort during archetype development, archetype development needs to be coordinated nationwide and beyond and also across the various health professions in a formalized process.

Towards automatically generating graphical user interfaces from openEHR archetypes.

One of the main challenges in the field of Electronic Health Records (EHRs) is semantic interoperability. To utilise the full potential of interoperable EHR systems they have to be accepted by their users, the health care providers. Good Graphical User Interfaces (GUIs) that support customisation and data validation play a decisive role for user acceptance and data quality. This study investigates the use of openEHR archetypes to automatically generate coherent, customizable, data-validating GUIs. Using the Mozilla XML User Interface Language (XUL) a series of prototypes has been developed. The results show that the automatic generation of GUIs from openEHR archetypes is feasible in principle. Although XUL revealed some problems, the advantages of XML-based GUI languages are evident.

Building health informatics skills for health professionals: results from the Australian Health Informatics Skill Needs Survey.

OBJECTIVE: To ascertain health professionals' perceptions of health informatics skills required in their roles. DESIGN: A paper-based survey with a stratified random sample of Australian health professionals and a web-based survey open to all Australian health professionals were conducted. MEASUREMENT: A questionnaire on the health professionals' perceived degree of competency required for a total of 69 specific skills in five skill categories based on the International Medical Informatics Association's (IMIA) set of recommendations on education and IMIA's scientific map. RESULTS: 462 health professionals responded to the paper-based questionnaire, and 167 respondents to the Internet questionnaire. Internet respondents reported higher required degrees of competency for specific health informatics and information technology skills than paper respondents, while paper respondents valued clinical skills higher than the Internet respondents. CONCLUSION: Health professionals increasingly use information technology (IT), and some also deploy, research or develop health care IT. Consequently, they need to be adequately educated for their specific roles in health informatics. Our results inform developers of educational programs while acknowledging the diversity of roles in health informatics and the diversity of pathways towards a professional health informatics qualification.

Towards shared patient records: an architecture for using routine data for nationwide research.

Ubiquitous information is currently one of the most challenging slogans in medical informatics research. An adequate architecture for shared electronic patient records is needed which can use data for multiple purposes and which is extensible for new research questions. We introduce eardap as architecture for using routine data for nationwide clinical research in a multihospital environment. eardap can be characterized as terminology-based. Main advantage of our approach is the extensibility by new items and new research questions. Once the definition of items for a research question is finished, a consistent, corresponding database can be created without any informatics skills. Our experiences in pediatric oncology in Germany have shown the applicability of eardap. The functions of our core system were in routine clinical use in several hospitals. We validated the terminology management system (TMS) and the module generation tool with the basic data set of pediatric oncology. The multiple usability depends mainly on the quality of item planning in the TMS. High quality harmonization will lead to a higher amount of multiply used data. When using eardap, special emphasis is to be placed on interfaces to local hospital information systems and data security issues.