Georg Duftschmid

Verification of temporal scheduling constraints in clinical practice guidelines

The computerization of clinical practice guidelines is a significant scientific challenge for the medical informatics community. One frequently reported factor hindering this objective is the existence of deficiencies within guideline knowledge. In this paper, we focus on the detection of flaws within temporal scheduling constraints. Temporal scheduling constraints are important elements of therapy management, and are frequently incorporated in clinical practice guidelines. We present a suitable verification method that is based on calculating the minimal network of temporal constraints on the execution of guideline activities. Our method serves three purposes: (1) it checks whether temporal scheduling constraints are consistent with scheduling constraints implied by control flow operators and the hierarchical structuring of a guideline; (2) it yields suggestions for an equivalent, yet more explicit representation of non-minimal constraints; (3) it can be used by the guideline interpreter to assemble feasible time intervals for the execution of each guideline activity. We evaluate our approach by applying it to a guideline specified in the Asbru language. For this purpose, we implemented a prototype verifier. Although we concentrate on the guideline representation language Asbru as the demonstration medium of our method within this paper, our approach can be reused to verify several alternative guideline representation formats.

Knowledge-based verification of clinical guidelines by detection of anomalies

As shown in numerous studies, a significant part of published clinical guidelines is tainted with different types of semantical errors that interfere with their practical application. The adaptation of generic guidelines, necessitated by circumstances such as resource limitations within the applying organization or unexpected events arising in the course of patient care, further promotes the introduction of defects. Still, most current approaches for the automation of clinical guidelines are lacking mechanisms, which check the overall correctness of their output. In the domain of software engineering in general and in the domain of knowledge-based systems (KBS) in particular, a common strategy to examine a system for potential defects consists in its verification. The focus of this work is to present an approach, which helps to ensure the semantical correctness of clinical guidelines in a three-step process. We use a particular guideline specification language called Asbru to demonstrate our verification mechanism. A scenario-based evaluation of our method is provided based on a guideline for the artificial ventilation of newborn infants. The described approach is kept sufficiently general in order to allow its application to several other guideline representation formats.

Austria’s Path Toward Nationwide Electronic Health Records

OBJECTIVES This article discusses current planning activities in Austria after legislation has been passed to introduce the electronic health record (EHR). METHODS After describing similar activities in several other countries, the authors explore the current situation of healthcare telematics and imminent steps toward the implementation of a lifelong EHR. RESULTS Substantial efforts have been made to coordinate healthcare telematics in Austria since the mid-1990s. One result of these efforts was the definition of a framework for electronic data exchange. A number of standardization projects were also implemented. Major steps have been taken as part of an ongoing healthcare reform to promote the use of healthcare telematics. One important example is a national initiative whose objective is to implement the EHR. This initiative is extensively discussed along with other national activities related to healthcare telematics. CONCLUSION This EHR initiative has prepared the ground for extensive planning that is currently under way to implement a lifelong EHR in Austria on a national level. Introducing the EHR will have a strong impact on Austrian healthcare and should be performed in concert with international activities. The authors offer a number of practical recommendations for the implementation of an EHR on a national level.

Management of data from clinical trials using the ArchiMed system

Clinical trials constitute a key source of medical research and are therefore conducted on a regular basis at university hospitals. The professional execution of trials requires, among other things, a repertoire of tools that support efficient data management. Tasks that are essential for efficient data management in clinical trials include the following: the design of the trial database , the design of electronic case report forms , recruiting patients , collection of data , and statistical analysis . The present article reports the manner in which these tasks are supported by the ArchiMed system at the University of Vienna and Graz Medical Schools. ArchiMed is customized for clinical end users, allowing them to autonomously manage their clinical trials without having to consult computer experts. An evaluation of the ArchiMed system in 12 trials recently conducted at the University of Vienna Medical School shows that the individual system functions can be usefully applied for data management in clinical trials.

A retrieval system for the selection and statistical analysis of clinical data

Within the last years computer-aided clinical documentation has provided researchers and administrators with very large volumes of data for research. User-friendly retrieval tools are needed when processing these clinical databases. Clinical researchers require applications by means of which steps in selection and analysis can be performed in an iterative process. During the deduction of statistical parameters from routinely collected data a number of problems occur that do not appear in the analysis of data gathered within clinical studies. Unlike clinical studies, routine data have complex structures and must first be formatted and above all temporally synchronized. In this paper we will describe the medical retrieval system ArchiMed developed at the Vienna General Hospital. A main objective in the design of this system was to support a joint evaluation of data from clinical studies and routinely collected data. The retrieval system comprises the main functions: Selection of Patients; Selection and Joining of Variables; and Statistical Analysis.

A tool for the design of clinical forms supporting end-user integration

Computer-based forms are a common input medium for recording medical data by the electronic mode. Efficient documentation requires that the forms satisfy the specific needs and habits of clinical end-users in the collection of data. This can be effectively achieved by integrating the clinician in the process of designing the forms. However, most development environments used for the implementation of clinical forms are customized for computer specialists and are too technical in nature to be used by clinicians. We present a tool for the development of clinical forms, which supports the integration of end-users in the design process of the forms. The tool is customized for users from the medical domain, allowing interactive and intuitive development of forms based on the configuration of predefined components instead of programming. Clinicians may even design highly functional, complex forms autonomously without having to involve computer specialists at all. All collected data are stored on the basis of a generic data model favouring data retrieval. The described application has been successfully used at the University of Vienna and Graz medical schools since 1997 and 1999, respectively.

Analyzing healthcare provider centric networks through secondary use of health claims data

Using health claims data of the Austrian social security system we derived networks of healthcare providers who share patients. Hereby we focused on networks of primary care physicians. Based on the characteristics of these networks potential implications for the upcoming Austrian national EHR system ELGA were drawn. Amongst others we conclude from our analysis that Austrian urban primary care physicians will use ELGA more intensively and benefit more from it than their rural colleagues.

Representation of inter-patient relations within electronic healthcare record architectures

Inter-patient relations are relevant in numerous areas of medicine, such as in the examination of genetic disorders, environmental causes of diseases, and in epidemiology. Therefore, it is essential that inter-patient relations be well integrated into the electronic health care record (EHCR). In the present paper we will show how inter-patient relations can be modelled within the three important EHCR architectures CEN ENV 13606, HL7 Clinical Document Architecture, and GEHR. We will use a general model for the specification of inter-patient relations as our reference model, which covers genetic and non-genetic kinship relations as well as temporary communities of persons.

Incorporating family and social structures into clinical documentation and retrieval

Referencing familial and social relationships between patients supplies valuable information for the retrieval and interpretation of clinical data. We present a technique for the incorporation of patient relations into data retrieval that takes into account the specific properties of routinely collected clinical data. In most clinical databases, family relations are documented in a fragmentary manner at best. Furthermore, clinical retrieval systems do not support inter-patient queries in most cases. Our model is designed to formulate direct relations between patients and to identify patients as members of either temporary or persistent communities. In this way, the model supplies information on both genetic and social relations.Referencing familial and social relationships between patients supplies valuable information for the retrieval and interpretation of clinical data. We present a technique for the incorporation of patient relations into data retrieval that takes into account the specific properties of routinely collected clinical data. In most clinical databases, family relations are documented in a fragmentary manner at best. Furthermore, clinical retrieval systems do not support inter-patient queries in most cases. Our model is designed to formulate direct relations between patients and to identify patients as members of either temporary or persistent communities. In this way, the model supplies information on both genetic and social relations.