Guidelines for Guidelines

An introduction to living approaches to guideline development

Development of living guidelines is similar to traditional guideline updating but involves a structured, high throughput process for identifying and incorporating new evidence and updating recommendations. Searches are conducted at regular intervals to identify new clinical evidence. Associated recommendations and additional information are then updated in consultation with clinicians and consumers, and the new version is published and disseminated. The process is then repeated. Living guideline approaches allow authors to continually seek feedback from stakeholders and iteratively review the methods and processes to ensure the guidelines remain useful and relevant.

The unit of update in a living guideline is the recommendation. A living guideline will include one or more living recommendations, as well as other stable, or static recommendations. There are many important considerations when determining a recommendation should be living. Three key criteria are often applied which help determine whether a specific topic is suitable for a living approach:

1. The topic of interest represents a high priority for clinical decision making
2. There is uncertainty in the strength or direction of recommendation
3. There is a high likelihood of new, relevant information being available in the near future

Additional important considerations when choosing whether to develop a living guideline include factors such as resourcing (living guideline updates have far lower per unit cost than traditional guideline updates but occur more frequently) and the experience and capacity needed to develop and maintain a living guideline.

Developing the baseline guideline

Clinical guidelines are only as good as the evidence syntheses on which they are based, and this remains true for living guidelines. Whether adopting, adapting or developing a de novo synthesis, all living evidence syntheses underpinning living guidelines should be designed in a way to facilitate easy updating of evidence tables and meta-analyses and re-publication of the guidelines. The use of tools such as Covidence, Cochrane RevMan and MAGICapp should be considered to help streamline this process, and attention should be given to how this content is presented to guideline panels for easy interpretation when drafting or updating associated recommendations.

When developing the baseline guideline, the methods and processes used should be designed in a way to facilitate an easier transition to a living guideline. For example:

• Which recommendations will be living?
• How frequently will databases be searched for new clinical evidence?
• Where will the recommendations and supporting information be published?
• What software will you use to streamline the transition of information from study to recommendation?

The living cycle

Living guidelines employ an iterative cycle in which a guideline developer continually searches for new clinical evidence and other relevant information and uses this to build upon the previous version of the guideline. This cycle should evolve over time to ensure that both the evidence on which existing recommendations are based is up to date, but also that the right questions are being addressed and answered in the most appropriate way. When planning a living cycle, some important considerations include: 

Are you still asking the right questions, and in the right way?

Living guidelines are based on living methods; nothing is set in stone. Over the course of a guideline new therapeutic options may become available, priorities can shift based on policy and practice, and outcomes of interest or other PICO criteria might need to be adjusted. Be sure to document any changes in approach and provide justification for your decisions. 

How frequently will you search for new clinical evidence?

It is generally recommended that searches be conducted every three to six months, however the frequency of searching will depend on a few factors, such as priority of the recommendation, how active the research space is, and resourcing. This may vary between recommendations, resulting in a variable search schedule across the guideline. 

Will you apply thresholds for incorporating new data into existing evidence profiles?

Not all evidence necessarily needs to be added to the existing evidence base immediately, and clear decisions need to be made in advance about thresholds for updating to avoid bias. One approach is that when a new study is identified, developers consider whether there it is likely to impact the recommendation. If so, they incorporate the new data into the existing evidence base, convene the guideline panel and update accordingly.

If the study is not considered as impactful (e.g. small studies or studies that provide evidence consistent with the existing recommendation), it is added to a surveillance list and results included into the evidence base when further data becomes available. In this case, it is still important to inform guideline users that a study has been identified and the intention to include the data at a future time. Other approaches, such as time-based approaches, or immediate incorporation regardless of impact on recommendation can be considered but have implications for workload and resource requirements. 

How frequently will you convene your guideline panel?

Living guideline panels are unique in that their job is never complete, but that doesn’t mean that quarterly updates require quarterly panel meetings. Potential changes in scope and new evidence might require an official meeting, however in the absence of these, an email keeping the panel informed of progress may be sufficient. Even if there are no changes to the guideline, it is important to maintain contact with the panel to ensure that members wish to remain involved and potential conflicts of interest remain up to date. 

How can you consistently improve your living guideline?

Although the focus of living guidelines has been the inclusion of new, relevant clinical evidence, it’s important to remember that living guidelines do not exist at a fixed point in time. Over successive iterations of the guideline, living recommendations should evolve to capture changes in clinical practice, policy context (e.g. approval, availability, reimbursement), alternative perspectives and any other information that might impact the applicability and implementability of the recommendation. 

Flexibility in approach

There are many avenues that can be explored to optimise the development of living guidelines. For example, how many recommendations in the guideline need to be living? Can you assign different priorities to each living recommendation? Have you applied thresholds for updating the underlying evidence base? Have you automated your searching? Are you using appropriate technology enablers? Streamlining your approach can help simplify the process of updating living recommendations and produce maximise impact while minimising required resources.

For more information on how to apply a living approach to development of guidelines, see the Australian Living Guidelines Handbook.

Acknowledgements

NHMRC would like to acknowledge and thank Heath White from the Australian Living Evidence Collaboration, Monash University, for his contribution to this module.