Education Adopting and adapting clinical guidelines for local use

Transcription

1 Education 2007;9: /toag The Obstetrician & Gynaecologist Education Adopting and adapting clinical guidelines for local use Author Gillian C Penney Key content: Clinical guidelines help to standardise and improve patient care. Development of a valid clinical guideline requires resources and skills. Local adaptation of a national guideline provides a cost-effective approach to guideline introduction. A formal consensus process allows local stakeholders to adapt a national guideline. The adapted guideline combines the rigour of national development with the acceptability engendered by local consensus. Learning objectives: To gain an overview of the fundamental principles of clinical guideline development. To learn how to adapt national guidelines for use by local NHS organisations or by groups of clinicians in non-uk settings. Ethical issues: Guideline introduction diverts scarce resources from direct clinical care. Evidence indicates that guidelines result in only modest to moderate improvements in patient outcomes. Clinicians wishing to introduce clinical guidelines should adopt an approach that minimises resource use. Keywords clinical guidelines / decision making / evidence base Please cite this article as: Penney GC. Adopting and adapting clinical guidelines for local use. The Obstetrician & Gynaecologist 2007;9: Author details Gillian C Penney MD FRCOG FFFP Director Scottish Programme for Clinical Effectiveness in Reproductive Health, Room 64, Aberdeen Maternity Hospital, Cornhill Road, Aberdeen AB25 2ZD, UK g.c.penney@abdn.ac.uk 48

2 The Obstetrician & Gynaecologist 2007;9:48 52 Education Introduction Clinical research continually generates new findings that can contribute to effective and efficient patient care. However, research findings cannot change patient outcomes unless health services and health care professionals adopt them into practice. Recent years have witnessed much progress in the synthesis of evidence. Secondary research in the form of systematic review and metaanalysis endeavours to collate evidence relating to discrete clinical questions. 1 Clinical guidelines summarise available evidence on broader clinical topics, or patient journeys of care, in the form of practice recommendations. Guidelines have the potential to improve patient care by promoting interventions of proven benefit and discouraging ineffective interventions. The National Guideline Clearinghouse (an initiative of the United States Agency for Healthcare Research and Quality) has collated clinical guidelines that meet defined quality criteria from around the world. 2 Similarly, the Guidelines Finder within the National Electronic Library for Health provides a searchable database of quality-assessed guidelines developed by UK organisations. 3 Developing valid clinical guidelines A widely accepted definition of clinical practice guidelines is systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances. 4 Internationally, numerous agencies have published methodologies for guideline development. These have been collated by the Guidelines International Network. 5 Published methods for development of valid guidelines differ in their detail but all are founded on three essential principles. Firstly, guidelines must be evidencebased, with recommendations based on a systematic review, including critical appraisal, of published literature. Secondly, individual recommendations must be evidence-linked, using a recognised grading scheme that explicitly summarises the type and quality of evidence on which they are based. Thirdly, guideline development must be multidisciplinary, undertaken by a group in which all stakeholders (including patients or service users), for the clinical topic, are represented. The United Kingdom national guideline organisations: the National Institute for Health and Clinical Excellence (NICE) and the Scottish Intercollegiate Guidelines Network (SIGN), have both published comprehensive guides to the development of valid guidelines. NICE has published both an overview of the guideline development process 6 and a detailed web-based technical manual. 7 Similarly, SIGN has published a guideline developers handbook 8 which has a regularly updated web version. 9 The process of guideline development, incorporating the three essential principles (as recommended by SIGN), is summarised in Figure 1. Introducing a clinical guideline In the United Kingdom, clinical guidelines can be developed by NICE and SIGN, by national professional organisations (such as the RCOG), or by teams within local healthcare organisations. Introduction of a clinical guideline into a healthcare setting is a process involving several stages, all of which have resource implications. Moreover, the available body of evidence on the effectiveness of clinical guidelines indicates that improvements in care are only modest to moderate. 10 Therefore,when deciding to introduce a new guideline (whether by de novo guideline development or by adoption of an existing guideline), clinicians and policy makers must give attention to the implications of each stage.advice on all stages of guideline introduction is available from a number of sources, 2,4,7 and is summarised in Box 1. Adopting an existing guideline versus de novo guideline development The literature from the early years of the NHS Clinical Guidelines initiative emphasised the tensions between local and national guideline development. 11,12 Development of a local guideline by a local stakeholder group provides sense of ownership a factor thought to be associated with increased likelihood that the guideline will actually be adopted. Conversely, development by a national expert group provides greater confidence that the content will be valid but less likelihood that the guideline will actually be used in practice. A recent Figure 1 Diagrammatic representation of the development of valid clinical guidelines as recommended by the Scottish Intercollegiate Guidelines Network 49

3 Education 2007;9:48 52 The Obstetrician & Gynaecologist Box 1 Framework to aid decisions by policy makers and clinicians about introducing clinical guidelines Stage of guideline introduction Selection of clinical topic Availability of existing guidelines Guideline adaptation or development Dissemination and implementation Consider resource implications of change Evaluate impact AGREE Appraisal of Guidelines, Research and Evaluation in Europe Activities Consider: local burden of disease availability of effective and efficient healthcare interventions evidence of variation in practice evidence of current suboptimal performance receptivity and preparedness to change as indicated by a network of local stakeholders. Consider the availability, or otherwise, of existing high-quality, cost-conscious guidelines. If such a guideline is available (for example, within the databases of the National Guideline Clearinghouse or the Guidelines International Network), then it may be adapted for local use. Otherwise, the resources required to develop a robust guideline de novo are substantial. Consider using a formal consensus method with a local stakeholder group to endorse or adapt an existing robust guideline. If no existing guideline available and resources permit, develop de novo using a published methodology and self-assessment using the AGREE instrument. Seek resources for implementation: explore sources of funding through partnerships and stakeholders. Identify the most likely barriers to implementation across all levels of healthcare organisation (individual, team, organisational and environmental). Prioritise those which are most important and amenable to change. Identify available intervention options based on available resources and expertise. Seek the best fit between barriers and likely interventions in the light of known evidence of effectiveness and models of behaviour change. Distribute recommendations in paper/electronic format as a minimum. Policy makers and clinicians should be alert to the potential impact on different budgets of changes in practice resulting from adoption of guideline recommendations. Guideline introducers should design a means of evaluating the impact of guideline introduction, for example, by means of related clinical audit. meta-analysis 13 provides further support for externally developed guidelines, showing greater effect sizes for external, rather than local, guidelines after adjusting for other factors. Rigorous guideline development carries significant resource implications and clinicians in local NHS organisations (and those in non-uk settings) often choose to adopt an existing guideline from an authoritative source (for example, an RCOG Clinical Green-top Guideline or one meeting the quality criteria of the National Guideline Clearinghouse) rather than developing their own in-house guideline de novo. The adoption of an existing guideline has advantages: it removes the need for the time and skills-dependent steps of literature search and critical appraisal. However, nationally developed guidelines may not be directly transferable to local NHS organisations (or to non- UK settings): recommendations may refer to services and interventions that are unavailable, or inappropriate, in the target setting. At the Scottish Programme for Clinical Effectiveness in Reproductive Health we have developed an approach for adapting a clinical guideline for setting-specific use. 14 This approach uses a formal consensus development method (the Rand version of the nominal group technique) and is theoretically based. 15 Adaptation using a formal consensus process with a local stakeholder group has advantages over simply adopting a guideline developed nationally. Clinicians, policy makers and service users from the intended setting can bring their local perspectives to the final wording of recommendations, ensuring that these accord with local culture and with available services and interventions. In this approach to guideline adaptation, the benefits of literature search and critical appraisal conducted within the resources of the parent guideline development organisation (for example, the RCOG) are combined with the benefits of input from a multidisciplinary group representing the intended local users. Once a local stakeholder group has agreed that a given topic is appropriate and has identified a rigorously developed guideline that is considered suitable for local adaptation, then the following approach may be used. Consensus method for adapting a national guideline for local use Recruitment of local stakeholder group The local guideline adaptation group should comprise representatives of disciplines similar to those included in the original guideline development group. For a guideline addressing an obstetric topic, the group might include senior and trainee obstetricians, midwives, neonatologists, neonatal nurses, anaesthetists and representatives of a service-user group. An overview of formal consensus methods indicates that a group size of 50

4 The Obstetrician & Gynaecologist 2007;9:48 52 Education between six and 12 participants is optimal. 15 In practice, an appropriate group for guideline adaptation is likely to require nine or 11 participants to ensure that all relevant stakeholders are represented. Consensus questionnaire The Rand consensus method includes private decision making by individual participants using posted questionnaires. As an example, an extract from a consensus questionnaire developed for adaptation of the RCOG Clinical Green-top Guideline No. 21 on the management of tubal pregnancy is shown in Figure 2. Participants are sent the questionnaire in advance of a group meeting and asked to score their level of agreement with each guideline recommendation in terms of suitability for use in their own local setting. Scoring on a ninepoint Likert scale has proved to be practicable. 14 Aggregating the scores The aim of the private decision making step is to achieve consensus agreement with a proportion of guideline recommendations to reduce the number requiring face-to-face discussion at a group meeting. Authorities differ as to what level of agreement constitutes consensus. 15 Purists would argue that consensus has not been achieved unless all participants are in agreement. However, pragmatic consensus rules are acknowledged as acceptable, as long as these rules are pre-agreed at the outset. With a group of 11 participants, consensus may be deemed to have been achieved if nine of the 11 score a recommendation within a three-point band on the nine-point Likert scale. For example, for the recommendation Laparoscopic salpingotomy should be considered as the primary treatment when managing tubal pregnancy in the presence of contralateral tubal disease and the Figure 2 Extract from consensus questionnaire for adaptation of an RCOG Clinical Green-top Guideline for setting-specific use 16 51

5 Education 2007;9:48 52 The Obstetrician & Gynaecologist desire for future fertility, if 10 participants gave it an agreement score of 7, 8 or 9 and the remaining participant gave it a score of 5, then this would be deemed consensus agreement with the recommendation. Conversely, for the recommendation Expectant management is an option for clinically stable women with minimal symptoms and a pregnancy of unknown location, if eight participants gave it an agreement score of 7, 8 or 9 and the remaining participants gave scores of 2 or 3, there is no consensus and the recommendation requires face-to-face discussion. During a guideline adaptation exercise, postal questionnaires are returned to a central point and scores aggregated as described above. We have applied this approach with groups of clinicians in the UK, Iraq and Sri Lanka. Generally, the private decision making exercise results in consensus agreement with most of the recommendations in the original guideline; these can then be adopted for the local setting without modification. A minority of recommendations will remain for which the adaptation group have reached no consensus or a consensus of disagreement (at least nine participants giving a score of 1, 2 or 3): these require face-to-face discussion and local modification. The formal Rand consensus method elicits private decisions using posted questionnaires as described above. However, we have also used this approach successfully in a workshop format with private decision making taking place at the beginning of the meeting, scores being aggregated on site and face-to-face discussion taking place thereafter. Modification of contentious recommendations Following the consensus questionnaire exercise a limited number of recommendations will require group discussion and modification. For each of these the evidence base for the recommendation (as presented in the original guideline) should be reviewed and options for modified wording explored. Often, necessary modifications will be easily resolved (for example, replacement of the term NHS by a term that describes the local healthcare system, or substitution of a drug that is unavailable locally). In modifying recommendations, caution is required to ensure that key points of evidence are not ignored because they are deemed inappropriate locally. Repeat private decision making If time and resources permit, a second consensus questionnaire should be compiled containing all recommendations in the adapted guideline (those which achieved consensus agreement in the first survey and those for which the wording was modified at the face-to-face meeting). Participants should then score their level of agreement with each using the nine-point Likert scale to demonstrate that consensus has been achieved for all recommendations in the final adapted guideline. In practice, the modifications required may be few in number and minor in nature. In this case, agreement on the adapted guideline may be achieved by informal consensus at the end of the face-to-face meeting. The use of an explicit and transparent consensus process permits adaptation of a clinical guideline developed nationally for one healthcare system for use at the level of a local NHS organisation or a different healthcare system. Such adaptation results in a valid guideline that adheres to the three fundamental principles: it is evidence-based; recommendations are evidence-linked (drawing on structured literature review and critical appraisal undertaken by the parent guideline developers); and it incorporates multidisciplinary input from a local stakeholder group. References 1 The Cochrane Collaboration. [ 2 National Guideline Clearinghouse. [ 3 National Electronic Library for Health. [ guidelinesfinder/]. 4 Field MJ, Lohr K, editors. Clinical Practice Guidelines: Directions fora New Program. Washington, DC: National Academy Publishing; Guidelines International Network. [ 6 National Institute for Health and Clinical Excellence. The Guideline Development Process. An Overview forstakeholders, the Public and the NHS. 2nd ed. London: NICE; [ page.aspx?o=guidelinesprocess]. 7 National Institute for Health and Clinical Excellence. The Guidelines Manual London: NICE; [ page.aspx?o=308639]. 8 Scottish Intercollegiate Guidelines Network. Sign 50: a Guideline Developers Handbook. Edinburgh: SIGN; Scottish Intercollegiate Guidelines Network. [ 10 Grimshaw JM, Thomas RE, MacLennan G, Fraser C, Ramsay CR, Vale L, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol Assess 2004;8:iii iv, Mann T. Clinical Guidelines: Using Clinical Guidelines to Improve Patient Care within the NHS. NHS Executive; p Grimshaw J, Eccles M, Russell I. Developing clinically valid practice guidelines. J Eval Clin Pract 1995;1: Dijkstra R, Wensing M, Thomas R, Akkermans R, Braspenning J, Grimshaw J, et al. The relationship between organisational characteristics and the effects of clinical guidelines on medical performance in hospitals, a meta-analysis. BMC Health Serv Res 2006;6:53. doi: / Glasier A, Brechin S, Raine R, Penney G. A consensus process to adapt the World Health Organization selected practice recommendations for UK use. Contraception 2003;68: doi: /j.contraception Murphy MK, Black NA, Lamping DL, McKee CM, Sanderson CFB, Askham J, et al. Consensus development methods, and their use in clinical guideline development. Health Technol Assess 1998:2:i iv, Royal College of Obstetricians and Gynaecologists. The Management of Tubal Pregnancy. Clinical Green-Top Guideline No. 21. London: RCOG Press;