Ebook: Computer-Based Support for Clinical Guidelines and Protocols

In many countries medical care is increasingly based on standardised clinical guidelines, many of which have been derived from prospective clinical trials. In addition to the problem of how to obtain appropriate evidence for these standards, a major problem is how to make practical use of them and assure that patients receive adequate treatment. Addressing the latter problem, it will become more and more important to provide computer-based tools that assist physicians and health care providers in guideline and protocol-based disease management.

The design of computer-based clinical guideline support systems, requires that the following three fundamental aspects be addressed. Guideline development is usually an interdisciplinary effort requiring many specialists from different fields a broad range of knowledge types must be acquired, represented and integrated, and finally we must be aware that analysis of disease management requires extensive documentation, necessitating the support of data acquisition and data dependent consulting procedures.

Since the development of the classical rule-based frame-oriented approaches, new technologies originating from the field of knowledge and ontology engineering, workflow management, data warehousing and data mining have been used to support guideline-and protocol-based care.

It was the objective of the First European Workshop on Computer-Based Support for Clinical Guidelines and Protocols (held in Leipzig, Germany on November 13^th and 14^th 2000) to bring together leading experts in the field and discuss the present state of development and future perspectives. An impressive range of topics was covered. Among them were:

• Knowledge acquisition tools, knowledge bases and problem solving methods

• Ontological engineering and formal concepts

• Design and implementation of data dictionaries

• Multi-dimensional organization and online analytical processing

• Data mining techniques

• Integration of workflow systems and CSCW for guideline - and protocol-based health care

• Telematics-based infrastructures and knowledge distribution

The workshop demonstrated clear that we are witnessing the beginning of a fascinating new field of interdisciplinary science, bridging innovative disciplines in computer science, medical practice and disease management. There is no doubt that integrating such disciplines will reap long term benefits. However, it will be crucial to design and integrate systems carefully and to accumulate experience on a step by step basis.

Essential questions include to what degree we need, and how we can design, knowledge-based data dictionaries to provide a basis for these systems. Related to this is the problem of providing efficient tools for data and knowledge acquisition and representation. Most importantly, however, will be the need to define the role of the physician and the patient. Here we need to make use of practical experience and develop ways to measure success.

This volume collects together the most interesting papers submitted to the workshop. These papers were selected by peer review from about 20 submissions. A full set of abstracts can be obtained from ewglp@telematics.uni-leipzig.de.

The meeting was extremely encouraging and stimulated extensive discussions. It became apparent that it was the first meeting of its kind, and all participants agreed that it was a very timely subject to focus on. Indeed, plans for successor-workshops have already been discussed.

As the local convener of this meeting it is my privilege to thank all members of the local organizing staff who contributed with great enthusiasm to this meeting. These were Robert Müller, Kristin Lippoldt, Mónica Aguiar, Sebastian Dietzold, Michael Krüger and Roman Mishchenko. Jan Ramsch was very helpful in editing the manuscript of this issue. Particular merit goes to Barbara Heller, who initiated this meeting. Her ideas and her charm were infectious and left everybody enthused.

We hope the readers of this volume will find motivation for future work, and will be inspired to contribute to this field.

Leipzig, summer 2001

Markus Löffler

Institute of Medical Informatics, Statistics and Epidemiology

(www.imise.uni-leipzig.de)

Medical guidelines are constructed with the aim of assisting clinicians in making decisions that are informed by the best available medical evidence. In order to achieve this aim, they must be disseminated in a form that makes them easy for clinicians to use and easy for domain experts to critique. Furthermore, the language in which they are expressed must facilitate their transfer between institutions and their adaptation to local conditions. This paper describes PROforma, a knowledge representation language that attempts to meet these desiderata. PROforma is formal knowledge representation language designed to capture the content and structure of a clinical guideline in a form that can be interpreted by a computer. The language embodies many contemporary themes in machine interpretable guideline representation schemas whilst retaining some distinctive features. This paper describes the key features of the PROforma language in the context of recent trends and developments in the field of electronic guideline representation formats. We describe our experiences in applying PROforma to a range of clinical decision making and workflow problems, and the benefits and limitations of the current language specification are discussed. Finally, we outline plans for the further refinement of PROforma.

There is general recognition that the use of clinical test ordering protocols supported by Information Technology can enhance quality and proper usage of clinical laboratory resources. This effort aims at addressing the requirement for a generic framework and software environment for the specification, execution and management of clinical test ordering protocols. The approach taken for the representation of clinical test order protocols is based on the event-condition-action (ECA) rule paradigm as defined in active databases. In this paper, a relational database approach to maintaining the protocol specification is discussed. Emphasis is placed on the importance of the efficient and effective management of computer-based test ordering protocol specifications at both schema and instance levels. A framework for the execution and management of clinical test ordering protocols is presented. A distinction and separation is drawn between the static protocol (specification) and the dynamic patient test plan (protocol instance).

The PIERCE (Prototyping Informatics and Evaluation for Research in the Clinical Environment) is designed for the Québec, Canada Cardiology Network to enable a single working informatics environment to collect high quality patient and laboratory data within clinical, clinical research and clinical trial protocols, to enable secure data sharing and also sophisticated project tracking and data analysis. This paper describes the implementation of the PIERCE program to support clinical research into triglyceride disorders, an important risk factor for atheromatous heart disease with a strong genetic predisposition in Québec.

The need to improve the quality of health care has led to a strong demand for clinical protocols and computer systems supporting both their creation and execution. We need to build complex protocols, but also to reason about them in different ways in order to verify, select, and modify protocols, to consider the effects of different protocols over time, and to monitor the patient's health conditions and the protocols' execution in parallel. Techniques, like Workflow Management Systems or Planning are appropriate to implement well-structured observations and processes, but are not suitable for the dynamically changing nature of therapy planning. Because of the limitations of existing approaches, we propose plan management consisting of fully integrated and inter-leaved tasks, namely plan design (authoring and/or generating), verification, validation, selection, adaptation, execution, monitoring, modification, evaluation, critiquing, and visualization. The Asgaard-Asbru project outlines task-specific problem-solving methods to support design, implementation, and execution of clinical protocols represented as time-oriented, skeletal plans. We illustrate how the idea of plan management is implemented in our Asgaard-Asbru project.

Many published clinical trials are poorly designed, suggesting that the protocol was incomplete, disorganised or contained errors. This fact, doctors' limited statistical skills and the shortage of medical statisticians, prompted the development of a knowledge-based aid, Design-a-Trial, for authoring clinical trial protocols. Design-a-Trial interviews a physician, prompts and guides them through suitable design options, comments on the statistical rigour and feasibility of their proposed design, and generates a 6-page draft protocol document. This paper reviews the progress of the Design-a-Trial project by describing a working prototype and its recent and planned development.

Developed and experimented at the Service d'Oncologie Médicale Pitié-Salpêtrière (Paris, France) as a computer-based breast cancer guideline system, OncoDoc has already proven to improve physicians' compliance. The underlying methodology relies on a decision-tree knowledge base that the clinician interprets in the context of an actual patient. In order to test whether the implemented strategy and the encoded knowledge base could be reused and shared in another institution, we have experimented the system at the Institut Gustave Roussy with only minor but legitimate customizations. For each patient, clinicians had to record their a priori therapeutic attitude, to use the system, to evaluate the system, and to record their final decision. Adherence (79%) and compliance (88%) rates were even higher than those previously observed. In addition, the approach appeared to influence clinicians' behavior—16% of initial decisions were changed to comply with recommendations—and to increase clinical trial accrual (+50%).

The main difficulties of converting the original textual form of medical guidelines to a computer-tractable form are connected both with the ambiguity of the natural language text and with the complexity of the resulting formal (and operational) representation. Proceeding directly from one to the other is thus an extremely demanding task. The proposed Guide-X methodology addresses this problem by breaking the whole process of guideline operationalisation into several steps, each of which requires a different mixture of types (medical, knowledge representation, typographical) and degrees of expertise. The principal technology used is that of XML tagging (using both pre-existing and newly developed languages). The result of each step is connected, element-by-element, to the results of previous steps, thus making the verification and revision of the operationalisation process easier. The methodology is currently being tested in the field of hypertension treatment, within the framework of the Medical Guideline Technology project of the EU Fourth Framework Programme.

This paper (1) presents a set of patient-care tasks for which a computer-interpretable representation of clinical practice guideline and protocols can provide assistance, (2) surveys the formalisms and computational methods that have been proposed in relation to these tasks, and (3) describes a multi-method paradigm that we have been developing in the EON project to support these patient-care tasks.

The role of decision models in supporting automated clinical practice guideline development is gaining importance in recent years. Constructing decision models from scratch, however, is laborious, time-consuming, and knowledge-intensive. This paper presents a new practical methodology to facilitate effective dynamic decision model construction for clinical practice guideline development, updating and customization. The proposed approach uses existing paper-based guidelines in sharable formats as the main information sources, characterizes the information inherent in the decision models and the clinical practice guidelines, establishes the relevant correspondences, reduces the effort needed in decision model construction, and increases the reusability and flexibility of the knowledge captured in the resulting guidelines.