Decision framework for update proposals

Is the review in a priority area? 

• Clear link or application to policy, systems, practice? 

• Topical (e.g. recent media coverage) 

• Rapidly moving area (high volume of new studies, deepening conflict in the outcome of trials) 

DECISION: Not a priority area, no update / living status ceases and do not reassess for at least two years. 

Has the review been widely cited and/or disseminated and/or accessed and/or included in guidelines or informed policies?  

Contributes to the decision but not a deciding factor. 

Have any previously identified studies changed categorization i.e.  

• those awaiting classification or ongoing now complete and available for full assessment of eligibility for inclusion in the review or  

• included eligible studies now have an associated retraction or there are concerns they are potentially problematic? Provide evidence of these studies 

Yes (with free text for evidence) / No 

Is new research available and likely to meet selection criteria for inclusion in the review? Provide evidence of these studies  

Yes (with free text for evidence) / No 

Note this should be more than a negligible number of new studies, it should comprise enough new data to conceivably alter the main findings of the review, or strengthen its GRADE ratings or conclusions. This is  particularly important for outcomes presented in the Summary of Findings (SoF) table (or for critical outcomes if no SoF), but may also apply to other important outcomes. The certainty of evidence in the existing (published) review will need to be considered when making this decision. 

Please also note that if all of the studies identified in the original review as ongoing or awaiting assessment were subsequently found to be ineligible for the review, the existence of these studies alone would not justify the need for an update. 

Authors must be able to provide evidence of new research, such as results of preliminary scoping searches that identify new studies that are likely to fulfil the inclusion criteria for the review 

DECISION: If an included study now has an associated retraction or concerns its potentially problematic and supported by evidence, continue with the below questions.  

DECISION: If number of new eligible studies comprise enough new data to conceivably alter the main findings of the review, continue with the below questions. If not, no update and do not reassess for at least two years or no new living update until next assessment.  

What other rationale is there for updating this review within the current context and uncertainties for this topic? For example: 

• a guideline that uses a particular review needs the review searches to be up to date in order to be credible. 

• a large, high profile study has become available (or further data has become available) and should be included in the review. 

• much of the research in the original review was of high uncertainty. If so, an update may be likely to change the conclusions of the review with the addition of evidence. 

• the review question is clearly relevant to current policy papers and/or service improvement. 

• the intervention, diagnostic test, prognostic factor or model, or other review component / focus / criteria is still in use. 

• the intervention, diagnostic test, prognostic factor or model, or other review component / focus / criteria has changed or developed over time and an update to the review is needed to reflect these changes. 

• the intervention, diagnostic test, prognostic factor or model, or other review component / focus / criteria is still being used despite no systematically assembled and appraised evidence base. 

• the interventions, diagnostic tests, prognostic factors or models, or other review component / focus / criteria and/or comparisons the review addressed need to be updated to align with changes in practice or policy. 

Free text box 

Each of these options require some monitoring or familiarity with the review area. It is the authors’ responsibility to monitor the review area and to present to Cochrane a clear rationale for the review update which meets at least one of the stated criteria. 

If an author team retires or hands over the review to a new author team, the new team must perform this assessment as part of the preparation to decide whether a new protocol is justified. 

Note. Updates are only possible when there is at least one author from the previous version included in the update proposal. If the author team is entirely new, they must start a new protocol (see editorial policy). 

Some of these may be a deciding factor and some may contribute to the decision, depending on the impact. If the review meets one or more criteria, continue with the questions below. 

DECISION: If none of these criteria are met, no update and do not reassess for at least two years or no new living update until next assessment. 

Have the interventions, diagnostic tests, prognostic factors or models, or other review components / focus / criteria changed so much since original introduction that a later date for the start of the review (i.e. searching start date) would make sense? 

Yes (with free text for details) / No 

Also consider whether the original review now overlaps significantly with other recently published Cochrane reviews? 

If yes to any of the points, consider whether changing the review scope would be useful or not. What issues in the body of the evidence contained in the review might influence decisions regarding the review’s scope? 

Consider whether there are uncertainties or issues in the evidence already contained in the review that might influence the scope of the update. For example, are some interventions no longer in use? Have certain outcomes (e.g. consumer-focussed outcomes) become more established in determining effects? Are there new variants of the intervention or comparison that need to be considered in the update? Does the population under study (or users of the review or intervention) need to be considered differently in this update? 

Scope change might be needed if users of the original review have identified gaps (e.g. exclusions from the original review) that could be incorporated into an update, or additional questions that might be addressed by adding to or modifying the review’s objectives or selection criteria. 

Most likely decision in this case is no change to scope of the review. 

Also consider, if the review is too narrow or focused, consider expanding. Is there a gap that this review could fill if the scope was expanded (e.g. interventions, comparisons, populations) that would improve relevance to practice, policy and/or research? 

If the review is very broad, consider modifying it to a more focussed question: 

Is the review so broad and/or large (n) it will be too big to be practically manageable (e.g. very large search outputs, large numbers of included studies)? 

Is there a strong rationale for splitting the review to focus on only part of the review’s original scope (e.g. splitting to update one or a selection of comparisons with promising results in the original review; splitting to address more closely a question/ outcomes of importance to or specific to policy makers or other decision makers or patients)? 

DECISION: Depending on the above, assess whether the scope of the review needs to change. If scope changes have implications for the review question, review criteria (PICO for intervention, PIT for diagnostic test accuracy, PICOT for prognosis, SPICE, SPIDER or PerSPecTIF for qualitative, etc.), study designs and / or methods, request a new protocol instead of an update and if a living review, the living status ceases.   

DECISION: If assessment shows overlap with other Cochrane reviews consider if there should be adjustment to its scope to justify a new protocol. If not, no update or new living review due to overlap (authors could reach out to the authors of the other reviews to ask to join the author team). 

Has there been feedback or commentary on the original review that needs to be taken into account regarding the scope of the Update? 

Yes (with free text for details) / No 

Would a precision review update be more appropriate at this stage, for example, updating the review for only specific comparisons or a smaller set of outcomes?  

Yes (with free text for details) / No 

Are the study designs eligible for inclusion the same as in last published version of the review? If no, provide details 

Yes / No (with free text for details) 

Do the responses to these questions result in the need to prospectively reconsider methods, justifying a new protocol? 

Are there any (major) discrepancies between the methods used and accepted standards (i.e., those described in the protocol and review templates, corresponding to MECIR standards)? 

The aim of this is to ensure that new methods are adequately considered and/or that compromise on key issues is reached between the authors and editorial team (eg if the review is a priority but resources are limited, changes to methods may need to be prioritised in order for the update to go ahead). 

Authors should clearly highlight additions or changes from original protocol/ review for easy internal editorial assessment. 

For older reviews, the protocol will additionally need to assess whether features such as the previously included study designs should be altered based on changes in standards, or the availability of new (higher quality) research 

Consider logistics of updating the review methods (author team with or without review development support): is there capacity for the author team to complete the update? Team may need to consider:  

• application of new methods to new studies and retrospectively (i.e. to all previously included studies) 

• new synthesis/ analysis methods and outputs may be possible based on the inclusion of additional studies/ data 

• funding and staff allocated to do the update versus volume of work (including retrospective application of new methods) 

• timelines for delivery of the update, search dates in relation to resource allocation etc 

DECISION: If the assessment shows that no substantive changes to methods are needed, approve an update or next living version. 

DECISION: If changes to methods are needed, authors should start a new protocol and if a living review, the living status ceases. Protocol must highlight changes from original protocol/ review methods. 

Are the risk of bias or appraisal tools to be used the same as in the last published version of the review, and do these meet current standards?  If no, provide details 

Yes / No (with free text for details) 

Are the methods to assess the certainty or quality of the evidence the same as in the last published version of the review, and do these meet current standards? If no, provide details 

Yes / No (with free text for details)