Ebook: Computer-based Medical Guidelines and Protocols: A Primer and Current Trends
This book brings together results from different branches of computer science (in particular, artificial intelligence), medical informatics and medicine to examine cutting edge approaches to computer-based guideline modeling, verification and interpretation. Different methods have been developed to support the development, deployment, maintenance and use of evidence-based guidelines, using techniques from artificial intelligence, software engineering, medical informatics and formal methods. Such methods employ different representation formalisms and computational techniques. As the guideline-related research spans a wide range of research communities, a comprehensive integration of the results of these communities was lacking. It is the intention of this book to fill this gap. It is the first book of its kind that partially has the nature of a textbook. The book consists of two parts. The first part consists of nine chapters which together offer a comprehensive overview of the most important medical and computer-science aspects of clinical guidelines and protocols. The second part of the book consists of chapters that are extended versions of selected papers that were originally submitted to the ECAI-2006 workshop ‘AI Techniques in Health Care: Evidence-based Guidelines and Protocols’. These chapters will provide the reader detailed information about actual research in the area by leading researchers.
This book, “Computer-based Clinical Guidelines and Protocols: a Primer and Current Trends”, is the result of the effort of the editors, started in 2006 with the organisation of an ECAI-2006 workshop at Riva del Garda titled “AI Techniques in Health Care: Evidence-based Guidelines and Protocols” and, subsequently, with the organisation of a workshop on “Computer-based Clinical Guidelines and Protocols (CCG'08)” at the Lorentz Centre of Leiden University at the beginning of 2008, to bring together researchers from the area of computer-based clinical guidelines and protocols with the aim of informing both researchers and others interested in clinical guidelines and protocols about the state of the art in this area. The ECAI-2006 workshop was a follow-up workshop from the “First European Workshop on Computerized Guidelines and Protocols” held in Leipzig, Germany in 2000 and the “Symposium on Computerized Guidelines and Protocols (CGP-2004)” held in Prague, Czech Republic in 2004.
With the current rise in the complexity and costs of health care, on the one hand, and increasing expectations of society about what health care is able to deliver, on the other hand, health-care professionals have developed a, sometimes urgent, need for care-practice support. Clinical guidelines and protocols have become the main instruments for disseminating best practices in health care. A clinical guideline gives general, usually nation wide, recommendations and instructions to assist the medical professional and the patient in decision making. In this book protocols are defined as local, specialised versions of guidelines, obtained in most cased by summarising information extracted from a guideline and by adding more detail, for example with regard to actual drugs or doses of drugs to be prescribed. As the detailed information may vary from hospital to hospital, the clinical protocols will reflect these differences between health care organisations.
Clinical guidelines and protocols promote safe practices, reduce inter-clinician practice variations and support decision-making in patient care while constraining the costs of care. In many cases, clinical guidelines and protocols have been useful in improving the quality and consistency of health care, by supporting health-care quality assessment and assurance, clinical decision making, work-flow and resource management. The benefits of having access to clinical guidelines and protocols are widely recognised, yet the guideline development process is time- and resource-consuming. In addition, the size and complexity of guidelines remains a major hurdle for effectively using them in clinical practice. Despite this, the number of clinical guidelines being developed and revised by professional health-care organisation has been rising steadily.
At the time when this preface was written, clinical guidelines were still textual documents, available in the form of booklets; it is only recent that these booklets have also become available in electronic form on the guideline-developers' world-wide web sites. Thus, present-day guidelines are still far removed from being ‘computer-based’. With the now almost ubiquitous presence of information technology in modern society it is likely that this will change, and that clinical guidelines will become computer-based in the very near future. This development was already foreseen by a small number of researchers, who started doing research in computer-based guidelines more than a decade ago.
It has taken a relatively long period of time in comparison to other areas, such as banking, before computers were accepted as valuable tools by medical doctors and nurses for the clinical management of disease of patients. Many countries are now on the brink of the wide-scale introduction of electronic patient records, which implies that, after many centuries, paper will no longer be used to store patient information and that computers will even become more important than they already are in health care. In this context, it seems even more likely that clinical guidelines will become computer-based, i.e., computer interpretable and executable. However, in order to make this happen, there is still a large gap between the current practice of guidelines development, on the one hand, and computer-based guidelines, on the other hand, that needs to be bridged. This issue is addressed by some of the chapters in this book.
Many researchers expect that the computer-based development, use and dissemination of guidelines will have a positive effect on the time required for the development of new guidelines and protocols, for the revision of existing ones, for deployment in daily care and dissemination. Furthermore, computer-based methods are indispensable for ensuring that guidelines are in agreement with the latest requirement for guideline development.
This book brings together results from different branches of computer science (in particular, artificial intelligence), medical informatics and medicine to examine cuttingedge approaches to computer-based guideline modelling, verification and interpretation. Different methods have been developed to support the development, deployment, maintenance and use of evidence-based guidelines, using techniques from artificial intelligence, software engineering, medical informatics and formal methods. Such methods employ different representation formalisms and computational techniques. As the guideline-related research spans a wide range of research communities, a comprehensive integration of the results of these communities was lacking. It is the intention of the publication of this book to fill this gap. It is the first book of its kind that partially has the nature of a textbook.
The book consists of two parts. The first part consists of 9 chapters which together offer a comprehensive overview of the most important medical and computer-science aspects of clinical guidelines and protocols. Not only are these chapters meant as a review of the state of the art, since, in addition, these chapters indicate cross links between topics and directions for future research. All chapters were written by authors with extensive expertise in the covered areas. Topics covered are: guideline development and deployment in medical practice, guideline representation languages, guideline modelling methods, use of formal methods in guideline development, temporal aspects of guidelines, planning, guideline adaptation, visualisation of guidelines and guideline compliance.
The second part of the book consists of chapters that are extended versions of selected papers that were originally submitted to the ECAI-2006 workshop mentioned at the beginning of this preface. These chapters will provide the reader detailed information about actual research in the area by leading researchers.
Chapters in both parts of the book have been extensively reviewed and profited from the feedback received in the writing process.
Thanks should go to the people–unfortunately too many to explicitly mention here–who helped in reviewing the various chapters included in the book and who provided very useful feedback to the authors. Finally, we are grateful to the Lorentz Centre at Leiden University for the facilities they offered in the process of completing the book, without which it would not have been possible to achieve the level of quality we were able to reach.
The Editors, 14th March, 2008
Annette ten Teije, Amsterdam
Silvia Miksch, Krems
Peter J.F. Lucas, Nijmegen
During the last decade many countries have become increasingly interested in the development and use of evidence-based practice guidelines, recognising that guidelines are key tools to improve the quality and appropriateness of health care. They are considered to be the ideal mediator for bridging the gap between the growing stream of research findings and actual clinical practice. Systematic reviews of guideline evaluations have shown that clinical practice guidelines can be an effective means of both changing the process of healthcare delivery and improving outcomes. A review of 59 guideline evaluation studies found that, in all but 4, statistically significant improvements occurred in clinical practice after implementation . A systematic review of 87 studies on the use of guidelines concluded that 81 studies revealed evidence of improved patient outcomes . Evidence-based guidelines are becoming an important and indispensable part of quality healthcare because of their potentials to improve quality and also reduce cost of health-care. Adherence to guidelines and protocols may reduce health-care costs up to a 25% . We will present an overview of the history of guidelinedevelopment and give some widely used definitions of guidelines. Guidelines are developed in a structured and systematic way, this process will be explained later. Also implementation tools necessary to put the guidelines into practice in an active way, will be discussed.
Implementing Computer-Interpretable Guidelines (CIGs) in active computer-based decision support systems promises to improve the acceptance and application of guidelines in daily practice. The model and underlying language are the core characteristics of every CIG approach. However, currently no standard model or language has been accepted by the CIG community. This aim of this chapter is to provide an overview of well-known approaches and to formulate a set of (minimal) requirements that can be used in the process of developing new CIG approaches or improving existing ones. It presents five CIG approaches (the Arden Syntax, GLIF, PROforma, Asbru and EON), followed by a general discussion of the strong points of each approach as well as their implications for future research.
Research on computer interpretable clinical guidelines has largely focused on individual points of care rather than processes of care. Whether we consider simple aids like clinical alerts and reminders or more sophisticated data interpretation and decision-making, guideline developers tend to focus on specific tasks rather than processes like care plans and pathways which are extended in time. In contrast, research on business process modelling has demonstrated notations and tools which deal directly with process modelling, but has not been concerned with problems like data interpretation and decision making. In this chapter we describe these two traditions, and compare some of their strengths and weaknesses. We also briefly discuss the distinct theoretical frameworks which have grown up around them, notably Petri nets for workflow modelling and mathematical logics for guidelines. We conclude that these offer complementary views of clinical processes and that a key research challenge is find a way of unifying them.
We wish to thank Natalie Mulyar, Marco Montali, Annette ten Teije and Mor Peleg for their helpful comments on earlier drafts of this discussion.
We wish to thank Natalie Mulyar, Marco Montali, Annette ten Teije and Mor Peleg for their helpful comments on earlier drafts of this discussion.
Formal methods play an important role in the development of software and hardware systems. In recent years, there has been a growing interest to apply these methods in the area of medical guidelines and protocols. This paper summarises these efforts, compares the approaches and discusses the role of formal methods in this area.
Temporal aspects play a major role within clinical guidelines. Temporal issues arise when considering both guidelines per se, and the application of guidelines to specific patients. As a matter of fact, guidelines per se specify different diagnostic and\or therapeutic patterns, and temporal constraints on the intended times of execution of the actions they contain are an intrinsic part of guidelines themselves. Moreover, guidelines must be executed on the basis of patients' data, which are intrinsically temporal data (consider, e.g., the time when symptoms hold). Devising suitable representation formalisms to properly model such pieces of temporal information is a challenging task, for which several solutions have been proposed in the last years. Besides representation formalisms, temporal reasoning methodologies are also needed. Temporal abstraction is needed in order to infer abstract temporal data (as described in guideline action conditions) from “raw” timestamped patient data. Moreover, temporal constraint propagationis also needed, both at acquisition and at execution time. During acquisition, temporal constraint propagation is used to detect whether the temporal constraints in the guideline are consistent. At execution time, it is needed in order to check whether the actual time of execution of actions has respected the temporal constraints in the guideline, and to detect which are the next candidate actions to be executed, on the basis of the temporal constraints in the guideline. This chapter sketches some of the most important recent results about the above issues.
A crucial feature of computerized clinical guidelines (CGs) lies in the fact that they may be used not only as conventional documents (as if they were just free text) describing general procedures that users have to follow. In fact, thanks to a description of their actions and control flow in some semiformal representation language, CGs can also take advantage of Computer Science methods and Information Technology infrastructures and techniques, to become executable documents, in the sense that they may support clinical decision making and clinical procedures execution. In order to reach this goal, some advanced planning techniques, originally developed within the Artificial Intelligence (AI) community, may be (at least partially) resorted too, after a proper adaptation to the specific CG needs has been carried out.
A rigorous development process of clinical practice guidelines through a systematic appraisal of available evidence is costly and time consuming. One way to reduce the costs and time, and avoid unnecessary duplication of effort of guideline development is by relying on a local adaptation approach of guidelines developed at the (inter)national level by expert groups. In this chapter we survey the work on guideline adaptation, which includes methodologies, case studies, assessment of effectiveness, and related work on guideline adaptation in the Artificial Intelligence community.
Authoring clinical guidelines as well as observing the execution and the maintenance of these is a time-consuming and cumbersome task. Usually, clinical guidelines are represented in conceptual models, which are very hard to understand by domain experts. Furthermore, to analyze the effectiveness and usefulness of clinical guidelines they need to be shown in connection with the patients' data. In this overview book chapter we present different methods to visualize clinical guidelines, patients' data, and the connection thereof. Finally, we illustrate how the different visualization methods support the various tasks in plan management.
Compliance with clinical practice guidelines is a challenging topic because it depends on a variety of factors, some related to guidelines themselves, some related to users, and some to the implementation context. Among the former are guideline quality, purpose and implementation modality. Among the userrelated factors are attitude to behavioural changes, authority interventions to foster adherence and eventually the type of users (general practitioners, hospital professionals, home caregivers, patients, etc.). Context is also crucial because organisational issues, such as lack of resources, can hamper guideline implementation and sometimes the original guideline intention is overridden by the guideline adaptation to a certain setting. This chapter analyses these factors and discusses their implications for the development of computerised decision support systems. Moreover, it gives examples of non-compliance detection and analysis in a specific real-world computerised guideline implementation, facing both methodological and practical issues.
Clinical guidelines and Careflow systems have been recently identified as a means to improve and standardize health care services. A number of ICT-based management solutions have been proposed, focussing on several aspects such as specification, process logs verification with respect to specification (compliance), enactment and administration of careflows.
In this paper we introduce the GPROVE framework, based on Computational Logic, and focused on the (formal) specification of careflows and on the compliance verification of the process executions w.r.t. the specified models. In particular, we show its application to the Cancer Screening Guideline used by the sanitary organization of the Emilia Romagna region, discussing its formalization in GPROVE and the results of the compliance checking applied to logs of the screening process.
This paper introduces our preliminary results in the modeling of Life Assistance Protocols, a new vision of medical guidelines and protocols through the lenses of p-Health. In this context the patient's role in the process is emphasized, the actions to be performed less defined and not only clinical situations considered, but also healthier lifestyle promotion processes accounted for, where the person's preferences and motivations play a key role.
We propose a complete framework, balancing on classical clinical guideline models and covering both the theoretical and the practical aspects of the problem, describing it from conceptualization to the execution environment.
Using machine-interpretable clinical guidelines to support evidencebased medicine promotes the quality of medical care. In this chapter, we present the Digital Electronic Guidelines Library (DeGeL), a comprehensive framework, including a Web-based guideline repository and a suite of tools, to support the use of automated guidelines for medical care, research, and quality assessment. Recently, we have developed a new version (DeGeL.NET) of the digital library and of its different tools. We intend to focus in our exposition on DeGeL's major tools, in particular for guideline specification in a Web-based and standalone fashion (Uruz and Gesher), tools for search and retrieval (Vaidurya and DeGeLookFor) and for runtime application (Spock); and to explain how these tools are combined within the typical lifecycle of a clinical guideline.
Medical guidelines and protocols are documents aimed at improving the quality of medical care by offering support in medical decision making in the form of management recommendations based on scientific evidence. Whereas medical guidelines are intended for nation-wide use, and thus omit medical management details that may differ among hospitals, medical protocols are aimed at local use, e.g., within hospitals, and, therefore, include more detailed information. Although a medical guideline and an associated protocol concerning the management of a particular disorder are related to each other, one question is to what extent they are different. Formal methods are applied to shed light on this issue. A Dutch medical guideline regarding the treatment of breast cancer, and a Dutch protocol based on it, are taken as an example.
This paper describes an infrastructure (TSNet) which can be used by geographically separated research groups to develop algorithms for the abstraction of complex time series data. The framework was specifically designed for the kinds of abstractions required for the application of clinical guidelines within intensive care.
PROforma is a formal language for modeling clinical processes that was developed by the Advanced Computation Laboratory of Cancer Research UK in 1996, together with associated tools, for creating decision support, care planning, clinical workflow management and other applications. The technology has been used to develop and evaluate a number of decision support applications in range of clinical settings. Clinical trials have been carried out and published for seven of these applications, all suggesting major positive benefits on a variety of outcome measures. The most recent and ongoing project called CREDO is an ambitious attempt to address the challenges in deploying sophisticated decision support systems in the intricate and convoluted management of chronic diseases, taking breast cancer as an example. In this chapter we describe the implementation of evidence based clinical guidelines within a complex care-pathway for patients with breast symptoms and analyse in detail the results of an evaluation study. Some important lessons learned during the process are shared and future directions are discussed.
Clinical guidelines usually need to be adapted to fit local practice before they can be actually used by clinicians. Reasons for adaptation include variations of institution setting such as type of practice and location, availability of resources, differences in patient populations, local policies, and practice patterns. When a guideline is implemented for clinical decision support and integrated with an institution's clinical information system, the data model of the local electronic medical record (EMR) and the data actually collected and stored in it also influence the guideline's adaptation. The purpose of this work is: (1) to characterize a tool-supported process for guideline encoding that addresses local adaptation and EMR integration, and (2) to identify the types of changes in guideline encoding during the local adaptation process.
We propose to use computerised medical guidelines as models for verification tools, so they can be validated with medical properties. To test the applicability we provide an implementation of the semantics of the medical planning language Asbru and also provide a formalised guideline for the treatment of breast cancer. With this case study we conduct experiments testing different proof techniques to cope with several challenges which guidelines provide.
The execution of clinical guidelines and protocols (CGPs) is a challenging task in high-frequency domains such as Intensive Care Units. On the one hand, sophisticated temporal data abstraction is required to match the low-level information from monitoring devices and electronic patient records with the high-level concepts in the CGPs. On the other hand, the frequency of the data delivered by monitoring devices mandates a highly efficient implementation of the reasoning engine which handles both data abstraction and execution of the guideline.
The language Asbru represents CGPs as a hierarchy of skeletal plans and integrates intelligent temporal data abstraction with plan execution to bridge the gap between measurements and concepts in CGPs.
We present our Asbru interpreter, which compiles abstraction rules and plans into a network of abstraction modules by the system. This network performs the content of the plans triggered by the arriving patient data. Our approach evaluated to be efficient enough to handle high-frequency data while coping with complex guidelines and temporal data abstraction.
We present GLARE, a domain-independent system for acquiring, representing and executing clinical guidelines (GL). GLARE is characterized by the adoption of Artificial Intelligence (AI) techniques in the definition and implementation of the system. First of all, a high-level and user-friendly knowledge representation language has been designed. Second, a user-friendly acquisition tool, which provides expert physicians with various forms of help, has been implemented. Third, a tool for executing GL on a specific patient has been made available. At all the levels above, advanced AI techniques have been exploited, in order to enhance flexibility and user-friendliness and to provide decision support. Specifically, this chapter focuses on the methods we have developed in order to cope with (i) automatic resource-based adaptation of GL, (ii) representation and reasoning about temporal constraints in GL, (iii) decision making support, and (iv) model-based verification. We stress that, although we have devised such techniques within the GLARE project, they are mostly systemindependent, so that they might be applied to other guideline management systems.