Living Systematic Reviews and Other Approaches for Updating Evidence

Systematic reviews use predetermined criteria to identify studies on a clearly defined issue, then assess their quality, synthesize the evidence, and draw conclusions from the included studies. In some cases, conclusions from systematic reviews are linked to recommendations, such as those made by the US Preventive Services Task Force and the Community Preventive Services Task Force,¹ or to guidelines for clinical practice. For recommendations or guidelines to be accurate and a reliable resource for decision-makers, the evidence must be up to date. A relatively new method for updating systematic reviews is called a “living” systematic review, in which a systematic review is updated as new research becomes available and relevant evidence is incorporated into the review.

WHAT ARE LIVING SYSTEMATIC REVIEWS?

Living systematic review refers to how systematic reviews are updated rather than a methodology for conducting them.² For a living systematic review, literature is searched frequently (e.g., monthly), and newly identified studies are incorporated into the review, metrics (such as meta-analysis or other summary measures) are updated with the additional study results, and review findings are revised accordingly.

Living systematic reviews are recommended under specific circumstances when a significant need exists for continuous monitoring and routine updating: when a particular field is moving quickly, and new evidence is likely to emerge regularly; when new information is likely to change the findings of the review; and when the research question is of high importance to decision-making to merit the ongoing allocation of resources. Not all systematic reviews merit being living systematic reviews, and not all living systematic reviews should maintain that status indefinitely. The balance of effort and payoff for this intensive method of updating systematic reviews may be helpful to consider in determining whether a living systematic review is a good option.

The consideration of living systematic reviews as a viable option for updating systematic reviews is more relevant as new technology is developed and made widely available to enhance the process. More published studies in the scientific literature over the past few decades means that the potential pool of qualifying studies for any given systematic review likely increases over time. Screening of identified studies may be expedited through machine learning techniques that reduce the burden for individual scientists. Platforms enabling the participation of citizen scientists (and algorithms to check their work) can increase capacity and reduce the time necessary to abstract a set of identified studies.³

PUBLISHED LIVING SYSTEMATIC REVIEW EXAMPLES

In December 2019, Cochrane issued updated practical guidance on living systematic reviews.³ Although the number of existing living systematic reviews—conducted by Cochrane or others—is currently small, examples can provide insights into how these continuous update methods can inform public health.

Traumatic Brain Injury in European Countries

A study of the epidemiology of traumatic brain injury in Europe was first published in 2016 and had five updates as of January 2019, adding 21 new studies during this period. The updates produced a wider range of national incidence rates and the ability to produce pooled results from multinational studies. In the original article, the authors noted some evidence of changing patterns of mechanisms of injury over time, which could merit a living systematic review approach to monitor those changes closely and adjust policies and clinical guidelines accordingly.⁴

Nutrition Interventions for Children

The Cochrane Library currently includes six reports from living systematic reviews. Of interest to public health is a living systematic review on interventions to increase fruit and vegetable intake among children aged five years or younger. This study was first published in 2012 and updated in 2017 as a living systematic review. Regular searches have been performed since then. Despite having a large number of studies (78 were included in the 2019 update), the authors noted that, given the low quality of evidence, “future research will likely change estimates and conclusions. Long‐term follow‐up of at least 12 months is required and future research should adopt more rigorous methods to advance the field.”^5(p2) This example underscores the trade-off between frequent searches for new evidence and sufficient time needed for researchers to address the noted evidence gaps.

OTHER METHODS FOR UPDATING SYSTEMATIC REVIEWS

Approaches similar to those used for living systematic reviews are commonly used by systematic review groups to update evidence regularly or following a decision-making framework^1,3,6,7 (A. Aloe, Campbell Collaboration, personal communication, March 27, 2020). Decisions about updating systematic reviews for selected groups conducting reviews or issuing recommendations are described in Table 1, which includes a comparison with similar decision points for living systematic reviews.

TABLE 1—

Indications and Decision Points for Updating Systematic Reviews

Indications for Updating Systematic Reviews	Commonly Used Decision Points	Living Systematic Reviews3
Recent evidence	Are new methods for analysis available?3,7	Is new evidence likely?
	Is there new evidence of harm?6
	Is there a new intervention for comparison or a large new trial with different results from the previous review’s conclusion?6
	Have studies been recently published that would add to the body of evidence?1,7
Priority and relevance of the topic	Does the original research question or topic remain relevant?1,3,7	Is the systematic review a priority for decision-making?
	Have stakeholders used the review?3,7
Likely change to findings or recommendations	Could existing recommendations (insufficient evidence for a recommendation) or the nature of an existing recommendation change?1	Is certainty in existing evidence low, such that new evidence is likely to change the findings of the review?
Time since completion of original review	Has an update been conducted within the past 5 y?1	Are all the above conditions met? If not, it is time to end the living systematic review.

For systematic reviews that include meta-analysis, conducting cumulative meta-analysis (i.e., recalculating the effect size of a body of evidence each time a new study is published) is another method for monitoring changes in the evidence over time. The chronological ordering of studies indicates the point at which no further studies are necessary because the results consistently show a particular conclusion. This is more of a retrospective review, in contrast to the prospective viewpoint of a living systematic review.

Determining whether or when to update a systematic review relies on engagement of the ultimate users of the information—government officials, stakeholders, and other decision-makers. Guidance from Cochrane in conceptualizing, supporting, and conducting living systematic reviews is encouraging for scientists yet must be matched by the demand for, and use of, continually updated evidence by decision-makers. One indication of this demand could be the use of living systematic reviews to inform living guidelines.³ Resource allocation decisions may consider a balance between building and sustaining infrastructure to support living systematic reviews when they are needed in the shorter term and using other approaches to update reviews over the long term because the bodies of evidence grow at a slower pace.

Living systematic reviews may be useful within their designated purpose and with appropriate infrastructure and stakeholder support. This approach, and other approaches to updating systematic reviews, helps to ensure that the best and most recent evidence is available for public health decision-makers.