Navigating the expanding body of research literature is an increasing problem. Global research output is growing rapidly, as is the number of systematic reviews being produced. Systematically developed reviews (called ‘systematic reviews’ [SRs]) provide the highest quality evidence that is needed to inform clinical and public health recommendations. A SR is a synthesis of all the medical literature on a given topic. It is estimated that global scientific publications double every nine years, and for SRs the situation is more marked. For example 25,000 SRs are added to the database Epistemonikos annually.
With the explosion of SRs comes an epidemic of multiple SRs published on the same topic. One study found 24 SRs on vitamin D supplements for preventing bone fractures, all with conflicting results owing to different methods. When encountering multiple reviews on the same question, clinicians may be confused and unable to formulate a conclusive answer to their patient’s question.
We aim to develop an automated algorithm which will help select the best SR amongst several on the same question. Our algorithm will have significant impact and application worldwide to every health field. We, as an academic group of methodologists and clinicians (primary contact carole.lunny@ubc.ca), would love to meet with anyone interested in partnering or funding our multi-year project. Our project plan can be found at https://osf.io/nbcta/?view_only=6b06b3f490c04ba0856f4cf95fcfd5ac