Should an Updated Review Reevaluate the Older Included Studies?

Do All the Old Studies Have to Be Assessed Again?

You are updating a systematic review. New studies have been found and will undergo screening, data extraction, and critical appraisal. But what happens to the studies already included in the previous version?

One option is to leave them untouched. They were already assessed once, and repeating every extraction and appraisal can consume substantial time. The opposite approach is to reassess everything so that old and new studies are evaluated using exactly the same current methods.

Neither approach should be automatic. An update can introduce new information about previously included studies, revised review methods, different outcomes, corrections or retractions, and methodological tools that did not exist when the original review was conducted. Current Cochrane guidance therefore treats an update as an opportunity to reconsider methods and explicitly requires attention to newly available information about previously included studies.

Reassess Older Studies When Something Relevant Has Changed

Reevaluation Can Mean Several Different Things

Do not treat reevaluation as one all-or-nothing task

Reevaluating an included study can involve several distinct activities. You might need to confirm that it still satisfies the eligibility criteria, obtain newly available data, extract an outcome that was not considered previously, reassess risk of bias using a revised method, correct previously extracted information, or determine whether the publication has subsequently been corrected or retracted.

Those tasks have different triggers. An update may require one without requiring all of them.

Always check for important new information about older studies

Previously included studies do not remain frozen after the original review is published. Additional follow-up data may appear. Authors may publish secondary reports. Journals can issue corrections, expressions of concern, or retractions.

The Cochrane Handbook specifically advises update authors to search for notifications, corrections, and retractions concerning previously included studies and to incorporate relevant additions or corrections. Where data appear incorrect or potentially fabricated, the implications for the analysis and related studies should be considered and decisions reported transparently.

Changed eligibility criteria can affect old studies as well as new ones

An update provides an opportunity to reconsider the review question, scope, comparisons, outcomes, and eligibility criteria. If those criteria change, the implications do not begin on the date of the new search.

Suppose the update narrows an eligible population or changes how an intervention is defined. A study included previously may no longer satisfy the revised criteria. Conversely, broader criteria could make previously excluded or uncollected evidence relevant.

This is one reason changes should be planned rather than introduced casually. Altering eligibility can require revisiting historical screening and included studies, not merely applying new rules to newly retrieved records.

New outcomes or comparisons may require returning to the original reports

An older study may contain information that was irrelevant to the previous review but important to the update. The Cochrane Handbook explicitly notes that adding a new comparison or outcome can require going back to previously included studies to collect information that was not extracted before.

You therefore cannot assume that an existing extraction spreadsheet contains everything required for the updated synthesis. The study itself may not have changed, but what you need from it has.

Risk-of-bias methods require a deliberate decision

Methodological standards change. The appraisal tool used in the original review may have been revised, replaced, or applied differently from the framework planned for the update. Current Cochrane guidance explicitly identifies updated risk-of-bias assessment as an example of a methodological development that may require revision of review methods.

Whether that requires reassessing every older study depends on the review and the methodological change. There has also been methodological debate over how extensively old studies should be reassessed when risk-of-bias tools change, with some authors arguing for reassessment to keep an updated review methodologically current.

The first can be methodologically defensible. The second is an efficiency decision masquerading as a methodological one.

New studies and old studies should form one interpretable evidence base

Newly included studies should undergo data collection and risk-of-bias assessment, and their findings should be integrated into the review synthesis. Cochrane also recommends completing or revising certainty assessments such as GRADE while considering the complete updated body of evidence.

If old and new studies are evaluated using substantially incompatible approaches, interpretation can become difficult. You need to decide whether methodological differences are acceptable, whether previous assessments can be mapped to the current framework, or whether reassessment is necessary.

A new analysis may require old data to be processed again

Updating a review is not limited to adding studies. New statistical or synthesis methods can change how the same evidence is analyzed. The Cochrane Handbook identifies revised synthesis methods as another reason an update may depart from the methods used previously.

If you change the effect measure, meta-analytic model, handling of missing data, subgroup structure, or another analytical decision, previously included studies may need to be revisited so that the required data are processed consistently.

Reassessment does not imply that the original review was wrong

A previous assessment can have been appropriate under the standards and information available at the time while still requiring revision today. Methodological development is one reason systematic reviews are updated in the first place. Current Cochrane guidance explicitly recognizes new methods, as well as new data, as legitimate components of an update.

The objective is not to retroactively grade the original reviewers. It is to make the current version methodologically defensible.

When New Studies Trigger a Reassessment of the Old Ones

Now imagine instead that the appraisal method had not changed, no relevant new information existed for the older studies, and the original assessments were complete and appropriately documented. Repeating every judgment from zero might offer considerably less methodological value. The decision depends on what has changed and whether that change can affect interpretation.

Common Mistakes When Handling Previously Included Studies

Revisit Older Studies According to What Changed

This decision belongs within the broader process of updating a review without unnecessarily repeating valid work. Reuse should be deliberate, just as reassessment should be deliberate.

Once old and new evidence have been brought onto an appropriate methodological footing, the next question is whether the updated evidence changes the previous conclusion.

Before Carrying Older Studies Into the Updated Review

Questions About Reevaluating Older Included Studies

Do Not Automatically Freeze or Rebuild the Old Evidence

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.