There is little point in developing a high quality guideline if its recommendations cannot be implemented
Watch Professor Sally Green from Cochrane Australia take us through what implementability is and how it can be incorporated into your guideline.
‘Implementation’ is the process of putting guideline recommendations into practice (NICE 2014). The features of the development, content and presentation of a guideline that enable it to be implemented are referred to together as ‘implementability’. Implementability is a well-recognised quality for a guideline.
The lack of adherence to guidelines is a worldwide phenomenon. Despite investments internationally to improve the quality of guidelines, their implementation in practice and policy is still a significant challenge (Gagliardi, Brouwers et al. 2014; Gagliardi, Alhabib et al. 2015). This can lead to beneficial therapies being underused, low value therapies being overused, poor use of resources and sometimes avoidable harm (Gagliardi, Alhabib et al. 2015).
An implementable guideline is one which develops and presents its content in a format that will assist in its uptake, therefore implementability is a function of both the content of a guideline and the communication of that content (Kastner, Bhattacharyya et al. 2015). There is mounting evidence that the degree of adherence to or uptake of a guideline depends on its implementability. This may also be supported by including implementation advice and tools when a guideline is disseminated (see Implementation and Dissemination) (Gagliardi and Brouwers 2012).
The following six principles of implementability should underpin guideline development and extend from the planning phase through to dissemination (adapted from Gagliardi and Brouwers 2012; Kastner, Bhattacharyya et al. 2015):
Involve and engage a range of stakeholders in development representing key audiences for the guideline, including clinicians, policymakers, patients and the public (see Engaging stakeholders and Public consultation).
Follow transparent methods for the inclusion and synthesis of the best available and most up to date evidence for a guideline that can be trusted by its users (see related Develop modules).
Be transparent about judgements used to balance available evidence, benefits and harms, patient variability and values and preferences (see related Develop modules).
Assess the local applicability of the recommendations, including resource considerations, divergence from current practice, workforce issues and community values.
5. Clarity of message
Use simple, clear messages with enough information to facilitate adherence to the recommendations (see Making actionable recommendations).
6. Appropriate formats
Develop multiple formats and ensure appropriate visual presentation to meet the needs of expected users (see Dissemination and Structuring and reporting the guideline).
This module provides practical advice on how you can improve your guideline’s implementability during development and make it easy to find, understand and use.
Although resource and time constraints may make implementing all the recommended actions in this module impractical, these do represent best practice principles. Further information about things you can do as a developer to help implement a completed guideline is provided in the Implementation module.
What to do
1. Plan for implementation
Writing an implementable guideline requires planning for its implementation early and often throughout development.
You should begin by considering the needs and preferences of the target population. This is particularly important when determining how and by whom a guideline will be used and is also useful for identifying barriers and facilitators to its use (Shekelle, Woolf et al. 2012; Kastner, Bhattacharyya et al. 2015). For example:
- Developers of a guideline related to diabetes should consider the specific needs of Aboriginal and Torres Strait Islander peoples given its high incidence and prevalence in these populations.
- Issues related to access to services for people living in rural and remote areas should be considered.
Related advice is provided in the Consumer involvement and Engaging stakeholders modules and it may be helpful to refer to recent examples of guidelines that have done this well. The Australian Government’s Clinical practice guidelines for antenatal care demonstrate how these considerations might shape the final guideline, as they include sections on Indigenous women, adolescent women and women in rural and remote areas, based on their specific needs.
There are many other aspects of a guideline’s development that will affect its implementation and it is important that you consider these early in the development process.
Consider asking the following questions (adapted from Canadian Task Force on Preventive Health Care 2014) at each stage of development to enhance implementability:
- Who are the expected users of the guideline?
- Who will be interested in its recommendations and outcomes?
- What are the key messages for its users?
- Who are the main target audiences, organisations and groups for each of these messages?
- What are the barriers and facilitators to using the guideline in the target audiences?
- What knowledge translation strategy will be useful?
- What will be the impacts of the guideline’s use?
- Will uptake of the guideline have a sustained effect?
- How can uptake be improved?
The Guideline Implementation Planning Checklist developed by Gagliardi, Marshall and colleagues (2015) provides a more detailed list of considerations based on a systematic review of guideline development and implementation advice. Another approach is outlined in the Guideline Implementability for Decision Excellence Model (GUIDE-M).
Any planning or decisions made about implementation should be documented in the guideline or resources that might accompany it (Wang, Norris et al. 2016). Constructing an implementation plan is discussed in further detail in the Implementation module.
2. Scope the guideline
The implementability of your guideline will be affected by the relevance of the questions it seeks to address (see Forming the questions). Other factors related to the need for the guideline can impact its implementability, such as the amount of existing variation in practice and the availability of research evidence (see Identifying the evidence).
The role of the guideline development group
The guideline development group should be multidisciplinary and represent a range of perspectives. Guideline implementability should be an ongoing consideration of the group throughout development. Consumer and stakeholder representation is crucial in this respect so that the group can assess the difference between the recommendations and current practice while the guideline is being developed. In addition, it may be helpful to identify opportunities for group members to act as champions for implementation across the health system or within organisations. Related advice is provided in the Guideline Development Group module.
Consider supplementing the development group with a knowledge translation or implementation working group that includes representatives of those likely to be using and implementing the guideline (Canadian Task Force on Preventive Health Care 2014). This group might then be consulted while recommendations are being developed to comment on their wording, style and format. They can also help identify any other potential issues with contextualising recommendations for a local setting. If resource constraints make it difficult to support such a group, you can instead select two or three members of the development group to provide explicit advice on implementation.
This is the aim of the Translation Research in a Dental Setting (TRiaDS) Programme, which provides a framework for planning knowledge translation by integrating it with guideline development (Clarkson, Ramsay et al. 2010).
Forming the right questions
An implementable guideline should have a specific purpose. The components of the recommended interventions or advice should be clearly described so that it can be reproduced in practice. This will only be possible if guideline questions are carefully written to outline the populations, interventions/exposures, comparisons (if relevant) and outcomes of interest (i.e. use the population, intervention/exposure, comparator, outcome (PI/ECO) structure). Related advice is provided in the Forming the questions module.
If a guideline has a broad scope, consider undertaking a formal priority setting exercise with the involvement of consumers and other stakeholders which can help to refine the questions and ensure they are relevant to the needs of expected users. If feasible it may also be useful to conduct or access an audit of practice or a clinical registry to identify current approaches to treatment or health care (Gagliardi, Marshall et al. 2015).
The Royal Australian College of General Practitioners (RACGP) has developed a useful approach to describing interventions (developed for non-drug interventions but suitable for broader application) in its Handbook of Non-Drug Interventions (HANDI) project, that encourages delineation between intervention, indication, adverse effects and availability. This approach can help improve the relevance and applicability of guideline questions; for example, consider the HANDI description of exercise as an intervention for type 2 diabetes.
When considering the patient population, you should acknowledge the potential for multimorbidity (also referred to as comorbidities) (Shekelle, Woolf et al. 2012). For example, developers of a clinical practice guideline on diabetes should consider whether additional or modified advice should be included for people with cardiovascular disease, since the two are frequently related (see Multimorbidity).
While you are planning the guideline, you can refer to other high quality guidance on conditions and health topics that frequently overlap in scope to ensure usability and consistency with existing advice. Also consider whether your guideline should include explicit links to other relevant advice. For more information on the management of multimorbidity in clinical practice, see the NICE guidance on this topic.
3. Ensure trustworthiness
An implementable guideline should be underpinned by a transparent development process to ensure its recommendations are trustworthy.
In this context, trustworthiness refers to the degree to which an audience finds a resource reliable—this will affect its acceptability and therefore its implementability. Your guideline’s reliability is impacted by:
- the way key decisions are made and documented during guideline development
- the involvement of stakeholders
- the identification, management and reporting of conflicts of interest.
If your guideline development is not transparent and does not follow a robust development process it is likely that users will question its credibility, relevance and applicability. Trustworthiness is also related to reputation, which can be affected by both the organisation developing the guideline and how their guideline development methods compare to best practice standards (e.g. NHMRC approval of clinical practice guidelines).
Organisations that develop guidelines should consider their profile and relationship with stakeholders, as building trust with them might enhance the prominence of the guideline and improve its uptake; consider opportunities to collaborate with or seek the endorsement of reputable stakeholders. The guidelines for the management of absolute cardiovascular disease risk were developed by four condition groups collaborating as the National Vascular Disease Prevention Alliance and were subsequently endorsed by the RACGP.
4. Form actionable recommendations
Implementable guidelines and recommendations should be specific. There is evidence to support the use of specific, concrete statements to modify behaviour and increase recall. Despite this, many guidelines fail to word recommendations in precise terms by considering the ‘what’, ‘who’, ‘when’, ‘where’ and ‘how’ components of behaviour (Michie and Johnston 2004).
You should make sure your guideline recommendations are clearly written and behaviourally focused. You should also avoid giving multiple instructions in a single statement. For example, consider the following recommendation (adapted from Michie and Johnston 2004):
‘Acute day hospitals should be considered as a clinical and cost-effective option for the provision of acute care, both as an alternative to acute admission to inpatient care and to facilitate early discharge from inpatient care.’
This statement could be made clearer by using the active voice instead of the passive voice:
‘Offer acute day hospital care to inpatients and those facing acute admission to inpatient care.’
This can be made even clearer by specifying to whom the advice applies:
‘Service managers responsible for making treatment decisions should offer acute day hospital care to inpatients and those facing acute admission to inpatient care.’
For clinical guidelines, the Template for Intervention Description and Replication (TIDieR) checklist may be useful in helping to describe interventions fully to ensure they can be reproduced in practice (see also training on how to use TIDieR). For example, it is important to provide relevant details about the timing and setting for interventions in relation to specific events (Hoffmann, Glasziou et al. 2014). Consider the following example recommendation (adapted from Morrell, Slade et al. 2009):
‘Trained health professionals should deliver psychologically-informed sessions to postnatal women in the woman’s home.’
This statement could be made clearer with added details on the timing of the intervention:
‘Trained health professionals should deliver psychologically-informed weekly one hour sessions to postnatal women in the woman’s home for up to eight weeks, starting at eight weeks postnatally.’
Consider inviting science or technical writers to assist with the wording of recommendations and messaging elsewhere in the guideline.
For more detailed information, see Making actionable recommendations, Evidence to decision and chapter 6 of the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) Handbook (GRADE 2013).
5. Consult the community
One of the benefits of public consultation prior to publishing a guideline is that it can be used to gather further information about how to facilitate its implementation. For example, public comments may relate to the technology needed, how the recommendations align with industry standards or policies, human resource implications, workforce capacity, workflow considerations and costs (Gagliardi and Brouwers 2012). For this reason it is important to reach a wide range of stakeholders at various levels during public consultation, such as senior managers, multi-disciplinary clinicians and consumer groups (see Consumer involvement).
Public consultation may also reveal barriers and facilitators to implementing a guideline. Even if developers have a limited role in this, comments may point to changes required to enhance the guideline’s implementability. For example, public comments may lead to changes in the messaging of recommendations for different local contexts to tailor guidelines to specific settings. This information may also assist in prioritising some recommendations over others depending on the context, particularly if the guideline includes a large number of recommendations.
For more detailed information, see the Public consultation module.
6. Format the guideline
The presentation of a guideline is an important feature that can enhance implementability. Planning for this may be generic across audiences or may lead to the development of additional or modified resources to target specific populations and expected users.
In finalising the format of your guideline, you should consider:
- who will use the guideline
- how they will access it
- whether more than one format should be developed
- whether you could make use of existing platforms to reach users (e.g. consumer support websites, journals, clinical tools) (Gagliardi and Brouwers 2012).
Plan for how the content should be navigated and pay attention to the table of contents, section headings and online tabs. Consider whether the information should be layered by differentiating content that must be read from content that should be read. Include a list or summary of recommendations and consider using tables and graphics to display information clearly. For example, the Australian Government provides a useful overview at the beginning of its Clinical practice guidelines for antenatal care to summarise each recommendation, its level of evidence and its page number in the main document.
Guidelines will ideally be developed in multiple formats to support wide dissemination, such as downloadable files, web-based platforms and in peer reviewed journals (Gagliardi, Marshall et al. 2015; Kastner, Bhattacharyya et al. 2015). Guidelines should also be paired with supporting resources, such as plain language summaries, policy briefs, podcasts and blogshots (see Derivative products). For example, the Australian Commission for Safety and Quality in Healthcare has developed clinician and consumer fact sheets, infographics, supporting evidence sources and videos to accompany their Acute Stroke Clinical Care Standard.
You may consider more innovative approaches, including interactive functionality or the use of living recommendations that are regularly updated when new evidence becomes available. For example, Elliott and colleagues (2014) have described a process for creating living systematic reviews which could be applied to guideline development.
7. Assess implementability
It may be useful to assess the overall implementability of a guideline when forming recommendations or finalising the guideline thereafter. Consider potential barriers or facilitators to implementing the recommendations in their intended setting (Shiffman, Dixon et al. 2005). You should also identify opportunities to enhance implementability further at this point prior to publication and dissemination.
There are tools available to undertake such an assessment, such as the GuideLine Implementability Appraisal (GLIA) tool (and its recent update); however, these have not been formally evaluated and their use may be resource intensive, particularly if each recommendation must be assessed separately (Shiffman, Dixon et al. 2005).
Developers seeking NHMRC approval of their guideline must meet all procedures and requirements. Many of these are related to aspects of guideline development that, when undertaken according to the advice in this module, can enhance the guideline’s implementability. For example, they cover aspects of the guideline’s governance and stakeholder involvement (including public consultation), scope and purpose, evidence review process, recommendations and structure and style.
The following Standards apply to the Implementability module:
1. To be relevant and useful for decision making guidelines will:
1.1 Address a health issue of importance.
1.2 Clearly state the purpose of the guideline and the context in which it will be applied.
1.3 Be informed by public consultation.
1.4 Be feasible to implement.
7. To make actionable recommendations guidelines will:
7.1 Discuss the options for action
7.2 Clearly articulate what the recommended course of action is, and when it should be taken
7.3 Clearly articulate what the intervention is so it can be implemented
7.4 Clearly link each recommendation to the evidence that supports it
7.5 Grade the strength of each recommendation.
9. To be accessible guidelines will:
9.1 Be easy to find
9.2 Ideally will be free of charge to the end user
9.3 Be clearly structured, easy to navigate and in plain English
9.4 Be available online.
Examples of implementation planning in guidelines
Clinical practice guidelines for antenatal care (DoHA 2012)
Gagliardi, A. R. and M. C. Brouwers (2012). Integrating guideline development and implementation: analysis of guideline development manual instructions for generating implementation advice. Implementation Science 7: 67.
Gagliardi, A. R., M. C. Brouwers, et al. (2014). A framework of the desirable features of guideline implementation tools (GItools): Delphi survey and assessment of GItools. Implementation Science 9: 98.
Gagliardi, A. R., C. Marshall, et al. (2015). Developing a checklist for guideline implementation planning: review and synthesis of guideline development and implementation advice. Implementation Science 10(1): 1-9.
Kastner, M., O. Bhattacharyya, et al. (2015). Guideline uptake is influenced by six implementability domains for creating and communicating guidelines: a realist review. Journal of Clinical Epidemiology 68(5): 498-509.
Morrell, C. J., P. Slade, et al. (2009). Clinical effectiveness of health visitor training in psychologically informed approaches for depression in postnatal women: pragmatic cluster randomised trial in primary care. British Medical Journal 338.
Shekelle, P., S. Woolf, et al. (2012). Developing clinical practice guidelines: reviewing, reporting, and publishing guidelines; updating guidelines; and the emerging issues of enhancing guideline implementability and accounting for comorbid conditions in guideline development. Implementation Science 7: 62.
Shiffman, R. N., J. Dixon, et al. (2005). The GuideLine Implementability Appraisal (GLIA): development of an instrument to identify obstacles to guideline implementation. BMC Medical Informatics and Decision Making 5: 23.
NHMRC would like to acknowledge and thank Professor Sally Green from Cochrane Australia and Dr Rebecca Armstrong from the University of Melbourne for their contribution to development of this module as authors and Professor Dianne O’Connell from Cancer Council NSW for her contribution as editor.
Version 5.1. Last updated 20/11/2018.
Suggested citation: NHMRC Guidelines for Guidelines: Implementability. https://nhmrc.gov.au/guidelinesforguidelines/plan/implementability. Last published 20/11/2018.