Using informatics to help implement clinical guidelines

Health Informatics Journal http://jhi.sagepub.com

Using informatics to help implement clinical guidelines L.A. Duff and A. Casey HEALTH INFORMATICS J 1999; 5; 90 The online version of this article can be found at: http://jhi.sagepub.com/cgi/content/abstract/5/2/90

Published by: http://www.sagepublications.com

Additional services and information for Health Informatics Journal can be found at: Email Alerts: http://jhi.sagepub.com/cgi/alerts Subscriptions: http://jhi.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Using informatics to help implement clinical guidelines L.A. Duff and A. Casey Clinical guidelines are a positive contribution to improving the quality of care and assuring its effectiveness. However, clinical guidelines need to be integrated with other quality improvement initiatives to fulfil their potential. We propose a model of how informatics can support the implementation of clinical guidelines and their integration into systems for decision support and clinical audit. Each element of the model is discussed in turn and particular attention is paid to how informatics can also facilitate the involvement of patients in developing and using clinical guidelines. The word ‘patients’ is used to describe all users of health services.

INTRODUCTION

Lesley Duff RGN MA Research and Development Officer Dynamic Quality Improvement Programme Royal College of Nursing, UK 20 Cavendish Square London, W1M 0AB, UK Tel: +44 (0)171 647 3828 Fax: + 44 (0)171 647 3427 Email: [email protected] Anne Casey RSCN Msc Royal College of Nursing, UK 20 Cavendish Square London, W1M 0AB, UK Email: [email protected] Correspondence to: Lesley Duff

Clinical guidelines are believed to have the potential to influence positively the quality of care received by our patients. This paper describes what clinical guidelines are, their use in identifying patient preferences and how they can facilitate evidence-based, collaborative decision-making and evaluation by healthcare professionals and patients to achieve clinically effective care. A key focus of the paper is the potential informatics has to make patient involvement in the development, use and evaluation of clinical guidelines a reality. The paper outlines how informatics can provide strategies that support the access, communication and evaluation of clinical guidelines through its three functions: knowledge browsing, messaging and counting [1]. Knowledge browsing describes the use of informatics to access information from a knowledge base. Messaging describes the way in which informatics is used to communicate information, for example through records, assessments, referrals and so on. Counting describes the use of informatics to generate and analyse data on the impact of clinical guidelines on practice and care quality.

Health Informatics Journal (1999) 5, 90-97

The challenge for informatics specialists is to turn the disparate functions of knowledge browsing, messaging and counting into integrated support systems that enable practitioners and patients to access clinical guidelines – or other forms of evidence about best practice – and to use the evidence contained within them in messaging systems already in use to plan and carry out care (care planning systems, referral systems and the like). It would also be helpful if messaging systems included educational highlights indicating decision points where reference to a guideline could be helpful: for example, with the prescription of the contraceptive pill for which a number of alternatives are possible. Reminders about key guideline recommendations within patient records are also an important contribution to guideline implementation [2]. Finally, the messaging and counting functions of an integrated support system would ideally enable practitioners to integrate clinical guidelines with other initiatives to improve the quality of care, such as clinical audit, the involvement of patients in decision making and so on. A further challenge lies in enabling practitioners and patients to use informatics in the implementation of clinical guidelines. If practitioners and patients are to be able to use an informatics-based support system they need to understand clearly what informatics is, its links with more customary forms of communication, the language used, and the processes of using hardware, software and other forms of technology. There are also a number of ethical issues that need to be resolved. The reasons that make informatics helpful in supporting the use of clinical guidelines can also jeopardize the quality of care. Improved access to information means improved access to all kinds of information and there can be uncertainty about its quality, unless strict standards are adhered to and enforced. There are also related questions about the security of integrated systems and the ability to maintain confidentiality of patient records, for example. In turn, people working with informatics can set challenges for those of us working to develop and implement clinical guidelines. These challenges include making available clinical guidelines that can be made accessible via an information network and the provision of sufficient detail within a guideline to enable decision support systems to be developed or quality indicators to be identified for audit. Guideline implementers can assist informatics specialists by highlighting essential data elements that must be represented in records. They can also help define communication pathways and other aspects of care processes that might impact on the use of

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Using informatics to help implement clinical guidelines

informatics. Finally, it is important that guideline implementers and informatics specialists work together collaboratively to increase our understanding of how people make decisions, how this may differ between professional and patient groups, how partnerships can be promoted between patients and professionals, and how informatics can facilitate collaborative decision making. In summary, care provided in accord with contemporary clinical guidelines has the potential to be clinically and cost effective. The truth of this statement depends on many factors that insure the integrity of clinical guidelines, the delivery of relevant guidelines in a timely way to the clinical setting, and evidence of the resulting outcome. Specifically, what is required is (1) access to good quality guidelines; (2) the collection and synthesis of comprehensive, reliable and valid information about patients and their preferences; (3) accurate diagnostic reasoning including strategies that explicitly share decision making between patients and practitioners; (4) clinical actions based on the guidelines; (5) evaluation of the efficacy, acceptability of the guidelines; and (6) mechanisms by which each of these steps feeds back into a research and development agenda. The role of the informatics functions of knowledge browsing, messaging and counting in reaching these six goals is explored below.

THE ROLE OF INFORMATICS IN IMPLEMENTING CLINICAL GUIDELINES There is much confusion about what clinical guidelines are and how they relate to other initiatives in the field of quality improvement. In particular, the role of informatics in supporting the generation and incorporation of patient preferences in clinical guidelines – as a means of promoting their inclusion in evidence-based decision making – remains a challenge. The confusion around clinical guidelines arises from the widespread use of the word ‘guidelines’ to describe different types of information that healthcare professionals, practitioners and patients can use in decision making. A useful working definition describes clinical guidelines as ‘systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances’. [3] Clinical guidelines are made up of statements about different aspects of the client’s/patient’s con-

91

dition and the care to be given [4]. They are thought to provide syntheses of evidence that overcome the difficulties of memory and time for individual practitioners wishing to access evidence for care. The systematic methods by which clinical guidelines are developed ensures they retain the quality of knowledge derived from available research. A key defining attribute of clinical guidelines is that they are based on evidence. Evidence can be usefully thought of as composed of knowledge, accrued from research findings and through experience [5, 6]. ‘Evidence’ in the context of clinical guidelines is usually understood to mean research evidence [7, 8, 9]. When there is no, or only limited, scientific evidence available from which to develop a clinical guideline, professionals will provide ‘expert’ opinion on which to base recommendations made [10]. In the same way, patients can take part in developing clinical guidelines and provide ‘expert patient opinions’ on care options. Patient involvement in guideline development is not easily achieved, however, and further research is required if it is to become the norm. Clinical guidelines are thought to have the potential to influence positively the quality and effectiveness of patient care for three reasons: first, they provide evidence about the efficacy of care options that can be drawn on during decision making; secondly, they outline a course of treatment or intervention which can act as a ‘blueprint for care’; and thirdly, they provide evidence based definitions of care against which practice, and sometimes costs, can be measured. It is believed that the evidence base of a guideline should provide the information needed to identify, question, and possibly stop those practices which waste money and are ineffective or dangerous; those practices for which there is uncertainty about the health outcomes achieved; and those practices in which there is a great deal of unwarranted variation. Henry and colleagues have explained how informatics provides an infrastructure for quality assessment and improvement in nursing [5, 11] and how the use of computerbased decision support mechanisms can enhance the contribution of clinical guidelines to the quality of decision making [6]. The model outlined in Fig. 1 aims to demonstrate how Benson’s [1] description of the three core functions of informatics (browsing, messaging and counting) can integrate clinical guidelines into the processes of quality healthcare, while promoting the involvement of patients throughout [12]. Within the model, informatics achieves

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

92

Health Informatics Journal

Browsing

New Knowledge: reflection, audit, research

KNOWLEDGE personal the 'body' of knowledge CLINICAL PROCESSES judgments decisions actions

Messaging INFORMATION about the patient (the record)

New Information INFORMATION about populations Database/Counting

Fig. 1. The clinical process context model: informatics view.

integration of clinical guidelines with the components of quality healthcare by using them to support three key clinical processes: ● ● ●

Diagnostic reasoning: what is the most likely diagnosis? what else is going on? Clinical decision making: what needs to be done? when? by whom? Clinical actions: investigation, treatment, care, etc.

The model defines the practitioner as a user and generator of knowledge. Knowledge in the context of the model describes personal knowledge, developed through education and experience and the ‘body’ of professional knowledge available as clinical guidelines, research articles and so on. The patient is described as a user and communicator of personal information. During the process of managing the patient, new information is generated and recorded. In addition new personal knowledge may be developed through reflection or audit of the case. From single records or the records of patient populations, new professional knowledge may be generated through audit and research. Adding technology to support the use and generation of knowledge and the use and communication of information completes the picture. Knowledge browsing takes place when professional knowledge in libraries or embedded in decision making support systems is accessed. Messaging takes place when information moves between two points. For example, the electronic patient record (EPR) can be defined as a set of messages and reminders to oneself and to others about what was found and what was done or planned for the patient. Database/counting technology is required for the analysis of patient data for audit, research and management/policy. In addition to the technology, there must

be one structured clinical language to link the patient record and information to the ‘knowledge’ and to the counting function of the computer system. There also needs to be more work on how to represent knowledge to match the many ways that clinicians want to access and use that knowledge. The following sections examine the possible role of these informatics functions in assisting practitioners to implement clinical guidelines.

Improving access to clinical guidelines: browsing For a practitioner or patient to be able to turn to a clinical guideline whenever evidence is needed would require a huge coordinated library of guidelines that could be accessed at any time of the day or night. A number of problems make this difficult to achieve, not least of which is the availability of good quality clinical guidelines and a system by which guidelines can be located and accessed at any point in the process of evaluation of care. It is also important that when no guidelines exist for a topic some other form of evidence can be accessed. Informatics has the potential to improve practitioners’ and patients’ access to clinical guidelines and other forms of evidence. The most common way in which it provides access at present is library ‘look-up’. Some clinical guidelines are available in full-text form via an electronic database, such as the Cumulative Index to Nursing and Allied Health Literature (CINAHL) or via the Internet using World Wide Web technology such as the Agency for Health Care Policy and Research (AHCPR). However, as Zielstorff [13] points out, locating guidelines via such mechanisms only solves the problem of accessing the guideline itself; access to the knowledge embedded within the guideline is another problem. She suggests that one method of overcoming the problem is to provide indexed access to the knowledge embedded within the guideline. Key words can then enable practitioners to find the sections of the guidelines that they need in order to selectively retrieve information. Locally adapted guidelines can also be accessed by menu, by using keyword searching or through linkage to structured fields. Such ‘field specific’ access to guidelines and pathways is becoming more common. For example, in one paediatric EPR system, a single mouse click on the patient diagnosis of ‘RSV bronchiolitis’ presents the user with a list of guidelines for relevant infection precautions, investigations and treatment protocols. These are based on locally adapted profes-

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Using informatics to help implement clinical guidelines

sional guidelines and are available as clinical or parent versions so that the latter can be printed and given to parents for their information [12]. An important factor in implementing knowledge in practice is access to that knowledge at the point of care. Mobile technology that supports the practitioner during the clinical encounter – at the bedside, in the clinic or the patient’s home – is one way of improving access. Beyond a simple link between the patient record and the available knowledge are decision support and expert systems that evaluate clinical data about the patient against parameters in the knowledge base and propose solutions to diagnostic and treatment questions. Decision support systems are discussed in the next section.

IMPLEMENTING CLINICAL GUIDELINES: MESSAGING Once clinical guidelines have been accessed they can be implemented in practice by assisting decision making about care and in providing guidance on the care to be given. Informatics can support guideline implementation through the function of messaging: communicating information in formats that support decision making and care planning. Making healthcare decisions is a complicated process for everyone involved, whether healthcare professional or client/patient. Writers such as Hogarth [14] and Eddy [15] suggest that decision making takes place in phases. These include recognizing that a decision is needed, identifying the available options, assessing their consequences, the desirability of the outcomes, gains and losses, and the freedom and ability to implement a choice. Increasingly we are focusing on the impact our decisions have on the cost of care as well as health outcomes for patients. Decision making is further complicated by the vast quantity of evidence that can be drawn on [4, 16]. However, even simple decisions may not be based on available evidence because of the difficulty we may experience in locating and interpreting, as well as assimilating, new information [16]. As Eddy [15] points out ‘the complexity of modern medicine exceeds the inherent limitations of the unaided human mind’. To keep abreast of new developments professionals may, therefore, benefit from carefully evaluated summaries of available evidence such as clinical guidelines. For patients an additional prerequisite to decision making is confidence and feeling

93

expected to take part in making decisions. Exacerbating this difficulty is patients’ dependence on practitioners for information, and knowledge leads to an asymmetry in the power relationship between them. The power relationship can influence both how well practitioners hear patients’ views and how much choice is actually made available to them. For example, when we measure performance and evaluate the quality of care we usually select outcome measures on the basis of our own values [17, 18]. Lind and colleagues [19], and other researchers [20, 21] have demonstrated that patients who are closely involved in making decisions about their care are more satisfied and more likely to agree to treatment regimes than those who are not. In view of such findings and increasing requests from patients to be more involved in decision making, we need to think of ways to work more collaboratively. One method that has been suggested to increase decision making with patients is to involve them in developing and using national clinical guidelines [22, 23]. Clinical guidelines can help provide patients with information and knowledge about the care options open to them. Informatics can be used to increase patients’ access to clinical guidelines and make the information they contain more relevant, easier to understand, and more appealing. It can also help make information more responsive to patients’ needs rather than the condition-specific information more usually available. A number of initiatives are under way to examine ways of presenting the knowledge contained in a clinical guideline to patients in a manner that promotes their ability to make informed decisions about care [24]. The research evidence on which clinical guidelines are based comes from a variety of sources [10]. However, patient views and opinions may not always be found in studies using traditional scientific designs such as randomized controlled trials. Instead, we may need to look to qualitative studies to find evidence about what patients think and value. Information technology can support the generation of patient-based evidence in a variety of ways. First, through the generation and documenting of information about individual patients, and their preferences. Secondly, with the growth in the use of networks such as the World Wide Web, the sample frame for studies that includes investigating patient preferences can be more easily expanded and located than when traditional means are used. There is scope for networked ‘discussion’ or consensus building on various topics. Even traditional quantitative techniques of collecting

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

94

Health Informatics Journal

information about how patients evaluate care, such as questionnaires, can be administered via a computer [25]. Thirdly, informatics allows for regular collection of epidemiological, audit and patient-based data which can be drawn on when patient-based evidence is sought [5, 26]. Informatics can be used to ensure that the new information generated during the process of managing the patient is generated and recorded. Finally, informatics can be particularly helpful in providing creative ways for us to seek the views, priorities and preferences of people who may have difficulty expressing them through other means, such as the use of Normative Decision Theory [27, 28]. There are a number of methodological and ethical issues to be resolved but the potential is exciting. Decision support systems aim to present the evidence represented in a clinical guideline in a format that can be used to assist practitioners and patients in making decisions [6, 13, 27]. They also have the added advantage of being able to provide prompts for the addition of patient preferences during interview or during the course of care provision. More work is needed, however, to improve the way knowledge is represented to ensure it matches the ways that practitioners and patients want to access and use it. Decision tables [6] are one option, but not the only possibility for replicating and supporting the way human experts think [29, 30]. Perhaps the most successful mechanisms for decision support are pathways or protocols developed from guidelines. The aim of a protocol-based decision support system is to provide a set of tools enabling a practitioner to access up-to-date guidelines and then use them in the management of patients [31]. Studies show, however, that even when clinical guidelines are available, clinicians forget to follow them or deviate from them without clarifying their reasons [32]. To help overcome this difficulty, assessment or investigation protocols and treatment plans can be copied into the patient’s electronic record with individual adaptations if necessary [2]. Such an approach to assisting decision making helps achieve the main aim of guidelines – reducing the variability in standards or practices for managing particular patient conditions. However, support for diagnostic reasoning and clinical decision making is not yet widely available in EPR systems in the UK, although allergy alerts and decision support for prescribing are becoming more common. As well as assisting with decision making, it is believed that clinical guidelines can help us improve the quality of healthcare in other ways [3]. In particular the recommendations

that make up a clinical guideline can be drawn on for care planning. They can also be used as educational tools [33]. Studies indicate that if implemented according to a strategy, clinical guidelines can have a positive impact on the process and outcomes of care [2, 9]. A number of studies show that clinical guidelines are also more effective when they are introduced alongside other initiatives such as reminders, feedback mechanisms, audits, education [2, 33–36], all of which can be supported by computer systems [37, 38]. For example, the interactive nature of information technology enables the educational potential of clinical guidelines to be exploited for practitioners, patients and the general public [39]. The interactive potential of informatics can also be used to facilitate the adaptation of nationally developed clinical guidelines for local use. Studies show that practitioners are often reluctant to change their practice and adopt new recommendations [40–43], particularly when they feel the change threatens their competence and autonomy. If, however, changes are presented in a way that clearly shows how they will benefit patients and gives practitioners a chance to state what they feel about them without presupposing they are necessary, attitudes to the change are more accepting [44–47]. A sense of ownership of clinical guidelines, and greater acceptance of their recommendations, can be engendered in practitioners by asking them to adapt the optional elements of a clinical guideline – those which are included on the basis of inconclusive research or expert opinion. To make local adaptations practitioners and patients need access to further information about both subsequent or related research in the field and about the local context in which a guideline is to be implemented, including the organization of care, skill mix available, patient characteristics, patient preferences and so on [48]. Such information can be accessed via the browsing function of informatics as discussed above. Integrating clinical guidelines with strategies for quality improvement [2, 3] is also found to promote their use in practice. Clinical guidelines can contribute to quality improvement programmes at each stage of the cycle. They can help us with quality improvement because by delineating what quality practice comprises (process) and its impact on care effectiveness (outcomes) [11], they provide us with statements against which to measure care and guidance for how care can be improved. The use of clinical guidelines in evaluating the quality of care is discussed in the next section.

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Using informatics to help implement clinical guidelines

EVALUATING CLINICAL GUIDELINES: COUNTING Clinical guidelines based on strong evidence can be used to define standards for care [9]. Alternatively, where standards or targets for care exist, guidelines provide information about how such a target can be achieved. For example, a clinical guideline on the management of leg ulcers might suggest to nurses that they need to set targets or standards for healing rates of the leg ulcers of patients in their care. The guideline recommendations then help them plan the care to be given to reach these targets. The components of the care recommendations also provide them with statements against which their practice can be measured. In this way the statements and recommendations of a clinical guideline can be used to generate medical review criteria or quality indicators, indicating by their presence or absence whether or not the defined quality of care has been achieved [3, 7]. Having established what we wish to know, we then have to collect data about our performance and other factors that influence the effectiveness of care. Informatics can provide an infrastructure to enable data about our achievement of clinical guidelines to be collected and synthesized. Importantly it also has the potential to link structure, process and outcome variables [5]. These terms describe in turn inputs or the materials, cognitive and behavioural skills, values and beliefs brought to the care situation; assessments, actions, and evaluations of care; and the outcomes of care [11, 49, 50]. Henry [5] illustrates the potential of informatics to link structure, process and outcome variables by explaining how it can support strategies to control or minimize variation in one of the variables while examining relationships among the dimensions. She gives as an example the case of risk-adjusted outcomes models in which structure variables are controlled to enable comparisons of risk-adjusted outcomes across providers, units or organizations. It is important that we understand such links if we are to know whether a clinical guideline has been implemented, what its impact on health and patient outcomes is, and what other factors may have prevented its use or influenced its impact. The first contribution that informatics can make to evaluating the effectiveness of care is to data collection. A vast amount of data are available to us to demonstrate or explain the use, or not, of clinical guidelines. Data comes from a variety of sources covering structure,

95

process and outcome variables [5, 11, 50]. For example, we may need to gather such diverse data as the use of equipment, administration of medicines, their side effects, medication errors, patient knowledge, patient competence and so on to evaluate the use and efficacy of a clinical guideline on self-administered medication in the management of chronic pain. Some data will be accessible through systems – such as those designed to record drug administration – and some will not. Such systems that do exist are likely to have varying degrees of integration [51]. We hope that in future far greater integration of information and knowledge with quality improvement data will be possible. In the meantime, informatics can provide useful ways of collecting data such as by computerized questionnaires and so on. This can be particularly helpful in finding out what practitioners or patients think about care. Care can be evaluated in a number of ways but it generally involves the measurement of outcomes. Two outcomes are usually referred to. The first is the physical condition of patients following an intervention. This may also be referred to as a health status outcome. The second is so-called patient-centred outcomes, such as patient satisfaction and quality of life. Patient-centred outcomes describe patients’ attitudes and emotional responses both to the care provided, including its organization and method of delivery, and to their physical condition following treatment. People judge the care they receive against a wide variety of criteria, only some of which reflect their reason for first contacting a healthcare professional. The judgements they make may have serious consequences both for their health outcomes and for their psychological health when receiving treatment [20, 21]. To identify where we can make improvements to care and ensure that it is effective and the planned outcomes reached, we need to understand all of the judgement criteria used by patients. Finding out what patients think about the quality and outcomes of their care is notoriously hard to achieve [52, 53]. Informatics can enable us to take a more creative approach to involving patients in evaluating care. Normative Decision Models, for example, can be used in which patients’ experiences are evaluated against their stated preferences and compared with population norms [28, 54]. Other examples include the use of automated questionnaires [55], and computerized nursing documentation that enables systematic administration of functional assessment instruments such as health status, pain profiles and quality of life measures to patients

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

96

Health Informatics Journal

during their admission assessment, and patient satisfaction instruments at discharge [56]. Once collected, data require analysis and interpretation if they are to indicate where improvements to the quality of care can be made. An obvious use of informatics is to abstract and aggregate data to provide scores or other forms of quality and utilization data [5]. The use of statistical packages such as Statistical Products, Services and Solutions (SPSS) [57] or NUDIST [58], a qualitative data management program, can make a meaningful analysis of data much easier to achieve. Informatics also has other uses, however. Key to the interpretation of data is information about how they compare with norms. The essence of measurement is variability [11] therefore we need to know what any detected variation means in the context of a whole population. Informatics and its electronic networks enable the collection and analyses of population based data-sets which can be drawn on to establish norms. Drife [26] for example, points out the value in collecting data on obstetric outcomes for use in benchmarking the achievements of local maternity units. Finally, improvement to the quality of care can only be achieved if the results of evaluation are acted on. Where they exist, clinical guidelines provide the knowledge needed to plan improvements to care. However, where no guidelines exist other forms of evidence need to be accessed – ease of access again being promoted by the use of computers and their knowledge browsing function. It is also important that the lack of evidence is recorded, whether from a clinical guideline, a systematic review or other research literature. In an ideal world health service researchers and funders of research would be able to use informatics to find out priority topics for research from databases collating and recording the results of audits and other quality improvement initiatives. The link between clinical guidelines and the research agenda would thus be more easily achieved.

CONCLUSION Clinical guidelines are thought to provide a positive contribution to improving the clinical effectiveness of care. However, there are a number of difficulties to be overcome if this potential is to be realized. Difficulties such as the availability and accessibility of clinical guidelines, inconsistencies in their use in practice, and unresolved inequities in the relationships between patients and practitioners

suggest clinical guidelines need to be integrated with other initiatives designed to improve the effectiveness of care: initiatives such as increased patient involvement in decision making, evidence based healthcare, measurement of outcomes and action to improve the quality of care. Informatics can support and enable the integration of clinical guidelines with other quality improvement initiatives and their use in practice through three key functions: knowledge browsing, messaging and counting. Knowledge browsing ranges from library searches, through field specific access from the electronic record and use of pathways based on guidelines. Messaging includes decision-support and expert systems which use guideline parameters and information about the patient to provide evidence-based support for clinical practice. Recording information in the electronic patient record using a structured language, supports the analysis of data on patients and populations, providing audit and research data to validate and update existing guidelines and develop new ones. Finally, the counting function enables us to evaluate the impact of a guideline on the quality of care and to identify where a guideline might help improve care efficacy and acceptability.

References [1] Benson T. The message is the medium. Health Service Journal 1997; 107; 5538 (IT Update): 4-5. [2] Grimshaw J M, Russell I T. Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations. The Lancet 1993a; 342: 1317-22. [3] Institute for Medicine. Guidelines for Clinical Practice: from Development to Use. Washington DC: National Academy Press, 1992. [4] Eddy D M. Practice policies: where do they come from? Journal of American Medical Association (JMA) 1990a; 263 (9): 126575. [5] Henry S B. Informatics: Essential infrastructure for quality assessment and improvement in nursing. Journal of the American Medical Informatics Association (JAMIA) 1995; 2: 169-82. [6] Shiffman R N. Representation of clinical practice guidelines in conventional and augmented decision tables. J Am Med Informatics Assoc 1997; 4: 382-93. [7] Agency for Health Care Policy and Research. Clinical Practice Guideline Development. Rockville, Maryland: AHCPR, 1993. [8] Eccles M, Clapp Z, Grimshaw J, Adams P C, Higgins B, Purves I, Russell I. Developing valid guidelines: methodological and procedural issues from the North of England Evidence-Based Guideline Development Project. Quality in Health Care 1996; 5: 44-50. [9] Woolf S H. Practice guidelines: A new reality in medicine III: Impact on patient care. Archives of Internal Medicine (AIM) 1993; 153: 2646-55. [10] NHSE Clinical Guidelines. Leeds: Department of Health, 1996. [11] Holzemer W L, Reilly C A. Variables, variability, and variations research: implications for medical informatics. JAMIA 1995; 2: 183-90.

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.

Using informatics to help implement clinical guidelines

[12] Casey A. A Model for the Future of Nursing Education. London: Royal College of Nursing, 1997. [13] Zielstorff R D. Online practice guidelines: issues, obstacles and future prospects. JAMIA 5 (3): 227-36. [14] Hogarth R. Judgement and Choice. New York: Random House, 1987. [15 Eddy D. Clinical decision making: the challenge. JAMA 1990; 263: 287-90. [16] Haynes R B. Loose connections between peer-reviewed clinical journals and clinical practice. Annals of Internal Medicine 1990; 113: 724-28. [17] Howe D. A framework for understanding evaluation research in social work. Research Policy and Planning 1988; 5: 6-12. [18] Smith G, Cantley C. Pluralistic evaluation. In: Evaluation, Research Highlights in Social Work. London: Jessica Knightly, 1988: 8. [19] Lind E A, Tyler T R. The Social Psychology of Procedural Justice. New York: Plenum Press, 1988. [20] Greenfield S, Kaplan S H, Ware J E, Yano E M, Frank H J L. Patients’ participation in medical care: effects on blood sugar control and quality of life in diabetes. Journal of Internal Medicine 1989; 86: 645-48. [21] Greenfield S, Nelson E C. Recent developments and future issues in the use of health status assessment measures in clinical settings. Medical Care 1992; 30 (5 Suppl): MS23-MS41. [22] Duff L, Kelson M, Marriott S, McIntosh A, Brown S, Cape J, Marcus N, Traynor M. Clinical Guidelines: Involving Patients and Service Users. London: Royal College of Nursing, 1996. [23] Bergstrom N. Development of Guidelines: Contribution of Patient’s Preferences. In: Nursing Informatics ’97 Conference, Post-Conference Meeting. Stockholm, Sweden, 2–4 October 1997. [24] Entwistle V A, Watt I S, Herring J E. Information about Health Care Effectiveness. London: Kings Fund, 1996. [25] Lenert L A, Soetikno R M. Automated computer interviews to elicit utilities: potential applications in the treatment of deep venous thrombosis. JAMIA 1997; 4: 49-56. [26] Drife J O. Assessing the consequences of changing childbirth. BMJ 1995, 310: 144. [27] Brennan P F, Casper G R. Modelling for decision support. In: Ball M J, Hannah K J, Newbold S K, Douglas J V (eds.) Nursing Informatics Where Caring and Technology Meet. New York: Springer-Verlag, 1995. [28] Ruland C, Bolton F P. Integrating patient preferences in nurses’ care decisions: clinical research and its implications for nursing informatics. In: Nursing Informatics ’97 Conference, Post-Conference Meeting. Stockholm, Sweden, 2–4 October 1997. [29] Rosch E. Principles of categorisation. In: Rosch E, Lloyd B B (eds.) Readings in Cognitive Science. Los Altos, CA: Morgan Kaufman, 1988; 312-22. [30] Rolfe G. Science, abduction and the fuzzy nurse: an exploration of expertise. Journal of Advanced Nursing 1997; 25: 1070-1075. [31] Coiera E. Medical informatics. BMJ 1995; 310: 1381-87. [32] Renaud-Salis J L. Distributed clinical management-information systems: an enabling technology for future health programmes. In: Barahona P, Christiansen J P (eds.) Knowledge and Decisions in Health Telematics – the Next Decade. Amsterdam: IOS Press, 1994: 139-46. [33] Field M J, Lohr K N. In: Institute of Medicine Guidelines for Clinical Practice: From Development to Use. Washington, DC: National Academy Press, 1992. [34] Cohen S J, Weinberger M, Hui S L, Tierney W M, McDonald C J. The impact of reading on physician’s adherence to recommended standards of care. Social Science and Medicine 1985; 21: 909-14. [35] Davis D A, Thompson M A, Oxman A D, Haynes R B. Evidence for the effectiveness of CME: a review of 50 randomised controlled trials. JAMA 1992; 268: 1111-17. [36] Putnam R W, Curry L. Impact of physician care appraisal on physician behaviour in the office setting. Canadian Medical Association Journal 1985; 132: 1025-29.

97

[37] Grol R. Implementing guidelines in general practice care. Quality in Health Care 1992; 1: 184-91. [38] Lomas J, Haynes B. A taxonomy and critical review of tested strategies for the application of clinical recommendations: from ‘official’ to ‘individual’ clinical policy. American Journal of Preventive Medicine 1988; 4: 77-94. [39] Grimshaw J, Russell I. Achieving health gain through clinical guidelines II: Ensuring guidelines change medical practice. Quality in Health Care 1994; 3: 42-52. [40] Oxman A D, Thomson M A, Davis D A, Haynes R B. No magic bullets: A systematic review of 102 trials of interventions to improve professional practice. Canadian Medical Association Journal 1995; 153: 1423-31. [41] Greer A L. The state of the art versus the state of science: the diffusion of new medical technologies into practice. International Journal of Technological Assessment in Health Care 1993; 4: 5-26. [42] McNicol M, Layton A J, Morgan G. Team working: The key to implementing guidelines (editorial). Quality in Health Care 1993; 2: 215-16. [43] Haynes R B. Some problems in applying evidence in clinical practice. Annals of New York Academy of Sciences 1993; 703: 21025. [44] Berwick D M, Coltin K L. Feedback reduces test use in a health maintenance organisation. JAMA 1986; 255: 1450-54. [45] Harvey G. Nursing quality: An Evaluation of Key Factors in the Implementation Process. PhD thesis: South Bank University, London, 1993. [46] Kitson A, Harvey G, Hyndman S, Sindhu F, Yerrell P. The Impact of a Nursing Quality Assurance Approach, the Dynamic Standard Setting System (DySSSy) on Nursing Practice and Patient Outcomes (the ODySSSy Project). Oxford: National Institute for Nursing, 1994. [47] Klein D. Some notes on the dynamics of resistance to change: the defender role. In: Bennis W G, Benne K D, Chin R, Corey K E (eds.) The Planning of Change. New York: Holt, Reinhart & Winston, 3rd edn, 1976: 117-24. [48] Delamothe T. Wanted: Guidelines that doctors will follow. BMJ 1993; 307: 218. [49] Donabedian A. Evaluating the quality of medical care. Millbank Quarterly 1966; 44: 166-206. [50] Royal College of Nursing Standards of Care Project. A Framework for Quality. London: Scutari RCN Standards of Care Series, 1989. [51] Dick R S, Steen E B. The Computer-Based Patient Record: An Essential Technology for Health Care. Washington DC: National Academy Press, 1994. [52] Fitzpatrick R, White D. Public participation in the evaluation of health care. Health and Social Care in the Community 1997; 5: 3-10. [53] Avis M, Bond M, Arthur A. Questioning patient satisfaction: an empirical investigation in two outpatient clinics. Social Science and Medicine 1997; 44: 85-92. [54] Brennan P F. ‘Patient satisfaction and normative decision theory’. Informatics: the Infrastructure for Quality Assessment and Improvement in Nursing: Proceedings of the Fifth International Nursing Informatics Symposium Post-Conference, Austin Texas, June 24–25. UC California: UC Nursing Press, 1994. [55] Carr R L, Snow C D, Youngson R A J. Design and implementation of an automated questionnaire. In: Gerdin U, Tallberg M, Wainwright P (eds.) Nursing Informatics: The Impact of Nursing Knowledge on Health Care Informatics. Oxford: IOS Press, 1997. [56] McCormick K. Improving nursing documentation to include outcomes of care in computerised information systems. In: Gerdin U, Tallberg M, Wainwright P (eds.) Nursing Informatics: The Impact of Nursing Knowledge on Health Care Informatics. Oxford: IOS Press, 1997. [57] SPSSinc Statistical Package for Social Sciences 1898-1995. [58] Software and Qualitative Solutions and Research Pty Ltd, Melbourne Australia, 1985–1994, NUDIST.

Downloaded from http://jhi.sagepub.com at PENNSYLVANIA STATE UNIV on April 15, 2008 © 1999 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.