NICE process and methods

14 Updating published clinical guidelines and correcting errors

Changes to review process from August 2013

An interim clinical guideline surveillance process and methods guide was published by NICE in August 2013, which replaces sections 14.1 and 14.2 in this guidelines manual. This interim process applies to guidelines being considered for updating from August 2013 onwards. After evaluation over a 12-month period, the process will be used to inform the NICE guidance development project and will be subject to formal consultation.

Clinical guidelines developed by NICE are published with the expectation that they will be reviewed and updated as necessary. Any decision to update a guideline must balance the need to reflect changes in the evidence against the need for stability, because frequent changes to guideline recommendations would make implementation difficult. This chapter describes the process and methods for reviewing the need to update NICE clinical guidelines and for producing an updated guideline.

The responsibility for updating a clinical guideline usually rests with the National Collaborating Centre (NCC) that originally developed it (or the NICE Internal Clinical Guidelines Programme if applicable). In some circumstances, an NCC or the NICE Internal Clinical Guidelines Programme[20] may be asked to update a guideline developed by another NCC. This will occur only after consultation with the relevant NCCs, and clarification of copyright issues.

When scheduling updates of guidelines into its work programme, NICE prioritises topics for updated guidelines and topics for new guidelines according to the need for new guidance. The relative priorities will be communicated to NCCs through the NICE business planning process.

This chapter also describes the process for correcting errors that are identified after publication of a guideline.

14.1 Process and methods for reviewing the need to update a published guideline

After publication of a clinical guideline, NICE collects information relevant to the guideline that might affect the timing or content of a subsequent update. This may include any queries or comments received by NICE or the NCC after publication, and evidence submitted by researchers or other stakeholders. However, NICE and the NCC will not actively seek new evidence, unless it has been identified in the guideline that important new information is likely to emerge before the 3-year scheduled review that may result in the need for an exceptional update or amendment (see section 14.4).

A formal review of the need to update a guideline is usually undertaken by NICE 3 years after its publication. The review includes the following key stages and methods (see figure 14.1 for a summary of the process and methods):

14.1.1 Stage 1

  • Collect information from members of the Guideline Development Group (GDG) that developed the original guideline (including patient, service user and carer members) using a questionnaire, and from the NCC, about new developments in the field.

  • Collate other types of information, including the results of a re-run of the original search used for the scope of the published guideline (see section 2.3.3), post-publication comments and feedback about guideline implementation (for example, feedback gathered by the NICE field team of implementation consultants and local and national audit data).

  • Conduct a broad search for relevant ongoing clinical trials.

  • Conduct a high-level search for evidence for relevant systematic reviews and randomised controlled trials (RCTs) using MEDLINE, MEDLINE In-Process, Embase and Cochrane (CDSR and Central only), with PsycINFO as an optional database for specific mental health topics.

  • Use the review protocol and inclusion and exclusion criteria from the published guideline to assess the relevance of, and summarise the evidence from, the systematic reviews and RCTs identified in the high-level search, based on abstracts only.

  • Collate all information and evidence identified so far to assess the need to conduct further focused searches on specific clinical areas in the guideline and/or new areas that may be important for an update of the guideline.

  • Discuss with the original guideline GDG Chair and/or Clinical Adviser, as well as other GDG members if necessary, to confirm the decisions on further focused searches and to develop PICO (population, intervention, comparator and outcome) frameworks for the focused searches.

14.1.2 Stage 2

  • Conduct further focused searches based on the relevant PICO frameworks using MEDLINE, MEDLINE In-Process, Embase and Cochrane (CDSR and Central only), plus two optional extra databases: PsycINFO (for mental health and psychological interventions questions only) and NHS EED (for health economics and cost effectiveness questions only).

  • Use the review protocol and inclusion and exclusion criteria from the published guideline to assess the relevance of, and summarise the evidence identified from, the focused searches, based on abstracts only.

  • Assess and summarise all information and evidence collected in stages 1 and 2, and develop a draft review decision on the need to update the guideline.

  • Discuss with the original guideline GDG Chair and/or Clinical Adviser, as well as other GDG members if necessary, to confirm that they agree with the draft review decision before consultation with stakeholders.

14.1.3 Stage 3

  • Check the stakeholder list for the original guideline with both the NICE Patient and Public Involvement Programme (PPIP) and other NICE teams to ensure that stakeholder details are up to date.

  • Consult stakeholders on the draft review decision for 2 weeks.

  • Consider (and if necessary revise) the draft review decision, taking into account the stakeholders' comments, and decide on the advice for NICE's Guidance Executive (see section 14.2).

  • NICE Guidance Executive makes the final review decision.

Figure 14.1 Process and methods for the 3-year formal review of the need to update a published clinical guideline

Process and methods for the 3-year formal review of the need to update a published clinical guideline

14.2 Deciding whether to update a clinical guideline

The Centre for Clinical Practice (CCP) at NICE considers the draft review decision in the light of evidence and information identified during the review process and stakeholders' comments.

CCP advises NICE's Guidance Executive on whether an update is needed and, if so, whether it should be a full update of the guideline or whether only particular areas of the guideline need updating. CCP may also advise that no update is needed. Two other options that can be suggested by CCP are transferring the guideline to the 'static list' or withdrawing the guideline.

Guidance Executive will decide which of these options is most appropriate. The decision is based on predefined criteria, as listed in table 14.1. The decision will take into account the competing priorities of other guideline topics and the capacity to schedule the work within the guidelines programme.

Table 14.1 Criteria for deciding whether to update a guideline

Update decision

Criteria

Actions

Update of whole guideline

  • Major sections of the guideline need updating

  • Many of the recommendations are no longer necessary

  • New clinical areas have been identified

  • Prepare a new scope

  • Consult on the scope

Update of part of guideline

  • Some recommendations need updating in the light of new evidence, or because they are unclear

and/or

  • New clinical areas have been identified that need to be covered in the guideline

  • Prepare a new scope

  • Consult on the scope

Exceptional update

  • A small number of recommendations may need updating in the light of new evidence

  • Exceptional updates may also be triggered by the identification of errors in a guideline after publication

  • Use the original scope

  • Do not consult on the scope

  • Inform stakeholders

No update

  • No new evidence has been identified that would overturn any of the recommendations

  • There is no evidence from clinical practice to indicate that any of the recommendations need changing

  • There is no evidence from clinical practice that the original scope need changing

  • The guideline is not updated

  • The guideline is reviewed after a further 3 years to determine its update status

Transfer to the 'static list'

  • The recommendations are unlikely to change in the foreseeable future

  • No further update planned

  • May be reviewed if new evidence emerges

Withdraw the guideline

  • The guideline no longer applies

  • Consult with stakeholders

14.3 Next steps

14.3.1 Conducting an update

If the entire guideline needs to be updated, the NCC prepares a new scope, following the process described in chapter 2.

Recruitment of GDG members follows the usual process (see section 3.1). The NCC should inform members of the original GDG that they are recruiting a new GDG, but the composition of the GDG should be tailored to the requirements defined by the new scope. The time required for development of the guideline is agreed between NICE and the NCC, and depends on the number of review questions. The guideline is developed using the same process as for a new guideline and is subject to the normal 6-week consultation period (see chapter 11). The usual process for finalising and publishing the guideline is also followed (see chapter 12).

If only part of the guideline is being updated, there are two possible scenarios:

  • some recommendations need to be updated and/or

  • new clinical areas have been identified that require new recommendations.

For both of these scenarios, a new scope is prepared and consultation with stakeholders takes place through the usual process (see chapter 2). The scope will make clear exactly which sections of the guideline are and are not being updated. The scope will also make clear that all recommendations in the original guideline, including those that have not been otherwise reviewed, will be checked to ensure that they comply with NICE's equality duties.

The NCC recruits a new GDG to undertake the work, using the usual recruitment process (see section 3.1). The time needed to undertake the update is agreed between NICE and the NCC.

14.3.2 No update

If it is decided that a guideline does not need to be updated, the guideline will be reviewed after a further 3 years. The same process for deciding whether an update is needed will be followed.

14.3.3 The 'static list'

There may be circumstances in which the clinical areas covered in a guideline do not need to be considered for updating. This may be the case, for example, if the evidence base is so poor that it is unlikely that any of the recommendations will change in the foreseeable future. In this case the guideline will be transferred to the 'static list' and no further update will be required. When a guideline has been placed on the static list, this will be made clear on the home page for the guideline on the NICE website. Guidelines on the static list may be transferred back to the 'active list' for further review if new evidence or information from clinical practice becomes available that is likely to mean that changes to the recommendations are required.

14.3.4 Withdrawing the guideline

It may be decided that the recommendations in a guideline no longer apply, but that the guideline is not of sufficiently high priority for updating. In this case the guideline will be withdrawn. This decision will be consulted on with stakeholders.

14.4 Exceptional updates

Exceptionally, significant new evidence may emerge that necessitates an update of a guideline before the formal 3-year review. This might be a single piece of evidence, an accumulation of evidence or other published NICE guidance (such as other clinical guidelines or technical appraisal guidance). This evidence must be sufficiently robust to make it likely that:

  • one or more recommendations in the guideline will need updating in a way that will change practice significantly or

  • patient safety issues need to be addressed

Examples of such evidence include significant data from RCTs, changes in licensing and patents or warnings issued by licensing agencies, or major changes in costs. Exceptional updates may also be triggered by the identification of errors in the guideline after publication (see section 14.7).

14.4.1 Determining the need for an exceptional update

CCP advises NICE's Guidance Executive on the following questions:

  • Is the update necessary?

  • Is there any other evidence (published, unpublished or from ongoing studies) that is relevant to the newly identified evidence?

  • Which recommendations need to be reviewed in the light of the new evidence?

Guidance Executive then decides on the need for an update based on the answers to these questions. If an exceptional update is necessary, CCP commissions an NCC to carry out the work. Stakeholders are informed at this point by NICE.

The aim of an exceptional update is to be responsive to new evidence, so it is imperative that changes to recommendations are published quickly. The process for developing exceptional updates should be the same as that for conducting an update, except that the original scope is used (see section 14.3).

14.5 Presenting updates

When presenting guidelines that have been partly updated, the aim is to ensure that there is a single set of publications that bring together relevant information from all previous versions of the guideline and the updated information. In this way, readers of the updated guideline will be able to easily identify which recommendations were made when. The rest of this section covers general principles to be used when part of a guideline has been updated.

As noted earlier in this chapter, a decision may be made to update the whole guideline. In this case, the process is as for a new guideline, except that the previous version of the guideline should be available for comparison.

The NCC and the CCP at NICE should agree as early as possible how the full guideline will be presented for consultation. Usually, the updated sections will be integrated into a single document with the existing full guideline.

14.5.1 Submitting the consultation draft

Before the NCC submits the draft guideline to NICE, the following should be checked:

  • Sections of the full and NICE guidelines have been updated as agreed at the scoping stage, or in line with any changes to the plan agreed with the Guidelines Commissioning Manager since scoping.

  • The full and NICE guidelines include standard text at the beginning, setting out which sections have been updated, how these are marked in the consultation draft and which sections are open for comment during consultation.

  • In the full guideline, updated sections (including the evidence, evidence to recommendations and recommendations) are clearly marked with paragraph borders, preferably a strip down the right hand side of the relevant pages bearing the word 'updated' and the year of the update. This will allow stakeholders to easily identify what they can comment on. The text that is superseded is placed in an appendix.

  • The recommendations have been marked up as described in box 14.1.

  • Recommendations from sections not being updated have been checked to determine whether any changes are essential (for example, if a drug is no longer available).

  • Changes in recommendations from sections that have not been updated are kept to a minimum (for example, changing from the passive to the active voice) and have been checked with the Guidelines Commissioning Manager and the CCP lead for the guideline.

  • There is an appendix in the full and NICE guidelines containing a table summarising the proposed changes to the original recommendations (see below for more information).

  • The status of any guidance incorporated into the previous version of the guideline has been confirmed with NICE. For example, has the other guidance been updated by the guideline update?

  • All current recommendations (new, updated and unchanged) have been assessed for the purposes of identifying key priorities for implementation.

  • All current recommendations (new, updated and unchanged) have been assessed with respect to NICE's equality duties.

Box 14.1 Labelling and rewording recommendations

In both the consultation and final published versions of the full and NICE guidelines, label all recommendations so that it is clear when the evidence was reviewed and whether the recommendation is new. The example below is of a guideline first published in 2008 with an update published in 2012.

Sections where the evidence has been updated

  • New recommendations, either an additional clinical area for the guideline or changed because of new evidence – add [new2012] to the end of the recommendation.

  • Unchanged recommendations where the evidence has been reviewed for the 2012 update but the recommended action is the same as in the 2008 guideline – add [2012]. Reword these recommendations into the direct style (see section 9.3.1), but check with the GDG that rewording has not changed the meaning.

Sections where the evidence has not been reviewed

  • For the consultation, add a grey background tint to recommendations that are not being updated, to indicate that they are not being consulted on.

  • Unchanged recommendations from 2008, where the evidence has not been reviewed for the 2012 update – add [2008] to the end of the recommendation.

  • Changes to recommendation wording that change the meaning (for example, because of equalities duties or a change in the availability of drugs, or incorporated guidance has been updated) – add [2008,amended2012] to the end of the recommendation, mark the change with yellow highlighting for the consultation and add a footnote explaining the reason for the change. This also applies if part of a recommendation (for example, a bullet point) has been deleted because it has been updated by other NICE guidance.

  • Evidence has not been reviewed, but there have been minor changes in 2012 to the wording of a 2008 recommendation that do not affect the meaning, for specific reasons such as changes in terminology – add [2008]. For the consultation, mark small changes in these recommendations with yellow highlighting. Include a general note about these changes in the appendix table.

  • Recommendation is incorporated from another published guideline use the label to show when that other guideline was published, for example [2006]

Explaining the proposed changes in the consultation version

Standard text at the beginning of the guideline

Refer to the latest full guideline and NICE guideline templates for the standard text.

Appendix explaining the changes

Create a table (which will form an appendix to the full and NICE guidelines) summarising the proposed changes to the original recommendations, including:

  • The text and recommendation number(s) of the recommendations that have been deleted in the update (either because they are being changed significantly in light of new evidence, or because they have become redundant), and the number(s) and text of any replacement recommendations. If there is no replacement for a recommendation, explain the reasons for the deletion.

  • A general note about any small changes made to recommendations that have not been updated, such as terminology changes. (These changes are marked with yellow highlighting for the consultation.)

  • A note about every change to a recommendation that has changed the meaning without an evidence review (labelled 'amended' and marked with yellow highlighting for the consultation). Include the new text of the recommendation.

Keep explanations as short as possible – only brief details are needed.

14.5.2 Submitting the final draft

Before the NCC submits the final draft of the full guideline, the following should be checked:

  • The recommendations are labelled as described in box 14.1.

  • Grey shading and yellow highlighting have been removed from the recommendations.

  • Footnotes explaining changes to recommendations labelled [2008,amended2012] are retained.

  • The appendix table summarising the changes to recommendations has been revised in line with the final recommendations.

  • The appendix with the superseded text is retained.

  • The paragraph borders indicating which sections have been updated are retained.

  • The standard text box at the beginning of the guideline explaining which sections have been updated has been revised.

The following should also be checked in the NICE guideline:

  • The recommendations are labelled as described in box 14.1.

  • Grey shading and yellow highlighting have been removed from the recommendations.

  • Footnotes explaining changes to recommendations labelled [2008,amended2012] are retained.

  • The appendix table summarising the changes has been removed.

  • The standard text box at the beginning of the guideline explaining which sections have been updated has been revised.

14.5.3 The NICE pathway and 'Information for the public'

NICE pathway

The NICE pathway for the existing guideline (if there is one) will be updated in line with the updated guideline. If there is no existing NICE pathway for the topic, one will be produced. If the existing guideline has a quick reference guide, this will be withdrawn.

A general description of what has been updated will be included at the start of the NICE pathway and will point to more detailed information in the NICE guideline about changes to individual recommendations.

'Information for the public'

When the updated guideline is published, the existing 'Information for the public' (or 'Understanding NICE guidance') will be withdrawn and replaced by a new 'Information for the public' in web format. Usually, the new 'Information for the public' will not differentiate between old and new recommendations. Sometimes it may be useful to explain which sections have been updated, particularly if it has been a rapid update, and patients are likely to notice changes in their care (for example, in drug treatment).

14.6 Maintaining records

In accordance with its contract with NICE, the NCC should maintain records throughout the development of an updated clinical guideline to ensure that the following information is readily available:

  • Details of the GDG membership, including declarations of interest.

  • Search strategy details, including when the most recent search was conducted.

  • Copies of the papers used.

  • Data-extraction forms.

  • Evidence tables.

  • Minutes of GDG meetings.

  • Any additional information presented to the GDG.

14.7 Correcting errors in published clinical guidelines

Measures are in place throughout the development of a clinical guideline to ensure that errors in the collection, synthesis, interpretation or presentation of the evidence are avoided as far as possible. However, on rare occasions errors may be found after publication of the guideline. These errors may not always warrant changes to the guideline, in which case they will be logged for consideration when the guideline is reviewed for updating. If an error is found, the following criteria and process will be used by NICE and the NCCs to determine whether changes are necessary.

14.7.1 Criteria and process for a correction

Corrections or changes to a published clinical guideline will be made if an error:

  • puts patients at risk, or impacts on their care or

  • damages NICE's reputation or

  • significantly affects the meaning of the recommendation.

If it is necessary to correct an error in a published guideline, we will follow our internal policy for dealing with errors. The individual or organisation who reported the error will be contacted in writing, and we will explain our rationale for the decisions and actions taken.

If a correction is to be made, a notification is put on the guideline's 'home' page on the NICE website. Depending on the nature and significance of the error and the time since publication of the guideline, stakeholders may also be notified in writing (usually by email). The relevant web-based documentation is corrected, and this is also highlighted on the guideline's home page on the NICE website.

14.8 Further reading

Clark E, Donovan EF, Schoettker P (2006) From outdated to updated, keeping clinical guidelines valid. International Journal for Quality in Health Care 18: 165–6

Eccles M, Rousseau N, Freemantle N (2002) Updating evidence-based clinical guidelines. Journal of Health Services Research and Policy 7: 98–103

Shekelle P, Eccles MP, Grimshaw JM et al. (2001) When should clinical guidelines be updated? British Medical Journal 323: 155–7

Shekelle PG, Ortiz E, Rhodes S et al. (2001) Validity of the Agency for Healthcare Research and Quality clinical practice guidelines: how quickly do guidelines become outdated? Journal of the American Medical Association 286: 1461–7

Shojania et al. (2007) Updating systematic reviews. Technical Review, Number 16, AHRQ Publication No. 07-0087

Turner T, Misso M, Harris C et al. (2008) Development of evidence-based clinical practice guidelines (CPGs): comparing approaches. Implementation Science 3: 45–52



[20] Information throughout this manual relating to the role of the National Collaborating Centres in guideline development also applies to the Internal Clinical Guidelines Programme at NICE.