Managing unmanageable loads of evidence: are living reviews the answer?
Bibliographic record
Abstract
During the COVID-19 pandemic, many researchers and health decision-makers have discovered first-hand how difficult it is to manage and digest the quickly accumulating and ever-changing flood of information, both online and offline. The World Health Organization and other researchers studying this ongoing surge of information (and misinformation) have coined the term “infodemic” to describe the situation.1 Looking more broadly than the pandemic context, researchers and health decision-makers are challenged daily by this constant and rapid accumulation of research evidence in many other clinical and public health areas that are difficult, or impossible, to manage and process. Evidence synthesis, such as systematic reviews, rapid reviews, and scoping reviews, summarize or describe the current scientific knowledge about therapies, procedures, tests, and public health interventions for decision-makers who use the information to inform guidelines, policies, or other high-priority decisions. Evidence synthesis is generally updated when the research question is still relevant, when new studies are available and would make a difference to the results or credibility of the findings, and when an ongoing decision need has been identified.2 When faced with a deluge of studies, evidence synthesis teams face difficulties producing and updating high-quality syntheses in a timely manner, which leads to problems translating evidence into action. The need for timely and up-to-date evidence has seen the emergence of “living” reviews into the evidence ecosystem, aptly named as research teams continuously refresh or update evidence.3 These approaches are underpinned by active evidence surveillance and often facilitated using technologies and processes (eg, artificial intelligence) to support the effort and workflow required to “manage the unmanageable.” Living evidence reviews are not a new approach or idea,4 yet the evidence ecosystem, until lately, has been somewhat tentative about fully embracing this approach. Historically, a few international specialist methodology groups tasked themselves with mapping a path for living reviews in the mainstream evidence synthesis landscape alongside their more static contemporaries. The current COVID-19 pandemic has precipitated a global virtual explosion of living evidence reviews to support clinical and public health goals.5-7 Despite increasing acceptance of the living review approach, several challenges associated with the methodology and process persist, and these limit the sustainable and efficient production and uptake of living reviews. Many of these issues were identified long before the pandemic4 but have been amplified or exacerbated throughout the pandemic response, such as i) silos of research topics that lead to duplication and research waste, and ii) identifiable gaps in the linkages between evidence producers, evidence synthesizers, and decision-makers who are the end-users of the review. This limits communication and continuous improvements across the entire evidence ecosystem. The scientific advancement of methods is critical yet underfunded and not prioritized. Action is needed before a complete paradigm change can be realized. One fundamental change for living evidence reviews that has been realized globally is a move away from online-only or self-publishing towards more traditional peer-reviewed publications. JBI Evidence Synthesis's first living scoping review protocol, with a plan for evaluating global evidence for gender equity in academic health research, was published in October 2020.8 This scoping review is currently ongoing and aims to map the evidence from more than 1000 included studies. The current issue of JBI Evidence Synthesis highlights two additional living review protocols.9,10 The living systematic review protocol by Adjei et al.9 aims to synthesize the available evidence on COVID-19 genomic variations on the African continent using monthly automated updates. This protocol represents an ideal application of a living review approach applied to the current pandemic context in Africa, given the rapidly changing profile of prevalent SARS-CoV-2 variants globally; the vast spatial geography of the continent; and a common, ongoing need for timely evidence to inform practice and policy. Gomes et al.10 present a protocol for a living scoping review to map nursing knowledge on skin ulcer healing, and address the continual need to integrate new knowledge into informatic systems. Authors will also screen the literature monthly, but will implement a threshold for updating when a minimum of 10% new literature is achieved (compared to the current search results). These protocols highlight the importance of living reviews for mapping the evidence and decision-making. It is our hypothesis that living reviews will only become more common in the future. Alongside this increase of living reviews, mechanisms will need to be in place to support the evidence ecosystem. Further work on the methodology of living reviews is encouraged, as well as capacity-building efforts for evidence synthesis teams providing support for decision-makers. Similarly, there is a need to attend to process issues related to supporting patient and citizen engagement in living reviews. Finally, sustainability mechanisms, such as prioritization across a realm of decision-making organizations and continuous funding for evidence synthesis teams responsible for living reviews, will need to be established. Acknowledgments David Moher, Brian Hutton, George Wells, and Sharon Straus for their contributions to discussions on living reviews. Funding ACT receives funding from a Tier 2 Canada Research Chair in Knowledge Synthesis; the funder was not involved with the development of the editorial.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".