Close
Found 0 results for your search. Showing 0 to 0.
No results could be found for your search.
x Close Search
Close

Current issues in evidence synthesis

Issues commonly reported relating to the production of evidence syntheses include:

  • A scope that is too narrow, which makes the review process smoother but less useful to inform practice;
  • The pervasive problem of becoming out of date very quickly (including within the time from submission to publication) and the additional resources needed to update it;
  • The amount of work they entail for the results they produce;
  • The literature being too heterogeneous, which can lead to narrative syntheses that are descriptive or inconclusive;
  • Methods for appraising the risk of bias or quality of the studies, including lack of consistency. Some argue that this is more problematic when the findings are used to inform clinical practice. 

Another issue is the argument over the best methodological approaches and types of evidence to consider in order to synthesise evidence, including different review types (e.g., systematic reviews, meta-analysis, meta-ethnography) and documents that are not published studies such as grey literature, government policies and reports. 

There have been recent calls to ensure that evidence syntheses become more responsive to policy and public needs through greater inclusion and transparency. Yet, the incorporation of different types of evidence brings additional challenges such as questions relating to the suitability and adaptability of existing tools to appraise their risk of bias or quality. There is also recognition that it can be challenging to synthesise the newer forms of evidence using existing methods. 

Further, as with all scientific work, reproducibility is important. There are currently a range of tools and guidance to ensure transparency in reporting, but the extent to which they are suitable for newer evidence forms and advanced technologies, is unclear.

In addition to standard published evidence, there is a desire in the field to engage with stakeholders and people with lived experiences. Doing so can help to identify gaps in the evidence and ensure that the methods, findings and interpretation are accurate and relevant to the people included in the review. However, doing so involves the need to identify and adapt the ways to engage with different types of stakeholders as well as ways to holistically integrate those views and perspectives in the evidence synthesis. 

Artificial Intelligence and Machine Learning in evidence synthesis

There is growing interest in the use of Artificial Intelligence and Machine Learning in evidence synthesis, especially to address some of the common issues mentioned above, namely the potential lengthy endeavor of a review process (especially screening) and reviews becoming out of date quickly. There is potential to increase efficiency with technological advances, but concerns remain on the potential for bias in AI algorithms and the need for human oversight. 

These debates reflect the evolving nature of evidence synthesis as a field and highlight the importance of adapting methodologies to meet the needs of diverse stakeholders while maintaining scientific rigor. 

Interesting papers on some of the topics mentioned above include: