Note: OpenAI leads the general AI space, but AI companies are developing deep research tools and experimenting with AI-powered academic searching in support of research. Perhaps you have faculty or students asking you to present these tools to classes.
Also: This open textbook (or wiki channel) is intended to help librarians and other information professionals learn about AI. It is not, in itself, meant to be seen as promotion of AI. If anything, the goal is harms mitigation or harms reduction.
Introduction
Knowledge synthesis (KS) (also, evidence synthesis or research synthesis) — "...is the process of searching, integrating, analyzing, and consolidating information from multiple sources of scientific information to create a coherent understanding or new insights". KS goes beyond simple summarization by critically evaluating and combining data found in primary research studies often with the goal to address complex problems or inform decision-making. KS is used in a range of scientific fields to inform policy development and evidence-based practice.
The Canadian Institutes of Health Research (CIHR) defines knowledge synthesis as: "...the contextualization and integration of research findings of individual research studies within the larger body of knowledge on the topic. A synthesis must be reproducible and transparent in its methods, using quantitative and/or qualitative methods, and will often take the form of a systematic review. Such an investigation will follow the methods developed by organizations such as Cochrane and the Joanna Briggs Institute."
"... current evidence does not support GenAI use in evidence synthesis without human involvement or oversight. However, for most tasks other than searching, GenAI may have a role in assisting humans with evidence synthesis." — Clark et al (2025).
Artificial intelligence (AI), machine learning and natural language processing are increasingly used in screening and extraction processes in evidence synthesis. Several studies (e.g., Dennstadt, 2024) reveal reduced workloads in screening using AI without affecting the quality of decisions made or sensitivity (number of correctly identified relevant studies divided by total number of relevant studies).
Platforms such as EPPI-Reviewer and DistillerSR incorporate machine learning and text mining tools, which semiautomate screening by prioritizing studies according to relevance. These tools rely on prior development of inclusion criteria and humans manually screening a portion of the documents to train the system. Covidence, for example, uses machine learning to text mine the remaining documents and sort/prioritize these according to relevance. Researchers then focus on screening the most relevant studies until a predetermined completion or stopping point is reached. Machine learning and artificial intelligence can also be combined with other innovative approaches described here.
Evidence synthesis is, along with research synthesis, a synonym for knowledge synthesis. It is a broad term referring to different types of reviews that systematically compile and analyze information from multiple documents such as studies or reports on a similar topic to develop an overall understanding of the results. These reviews are the cornerstone of evidence-based decision-making in health care. Since the 1990s, systematic reviews of interventions, which examine the benefits and harms of health interventions such as medical treatments, have been the predominant form of evidence synthesis. As the field has evolved, clinicians and other decision makers need answers to questions that go beyond the benefits and harms of interventions. This has led to the development of diverse types of
evidence syntheses, each tailored to specific health-related questions. A scoping review by Pollock et al (https://osf.io/znjeg/) leading to the Evidence Synthesis Taxonomy Initiative identified over 1,000 different terms used to describe different types of evidence syntheses.
Overview: Provides detailed methodological guidance for conducting scoping reviews, including steps for mapping evidence and identifying research gaps.
Overview: Provides methodological guidance for rapid reviews, a streamlined form of knowledge synthesis, balancing speed and rigour; addresses need for timely evidence synthesis, particularly in public health emergencies.
Overview: Offers guidance on distinguishing between systematic and scoping reviews, detailing their objectives, methods, and reporting standards. Clarifies when to use each method, widely used by researchers planning knowledge synthesis projects.
Overview: Explores how stakeholders can be involved in knowledge synthesis to enhance its relevance to policy and practice; bridges gap between research and policy, emphasizing practical applications of synthesis.
Overview: Analyzes the volume and attributes of knowledge synthesis research in PubMed, highlighting trends and gaps in emerging methods; offers insights into the evolution of knowledge synthesis, guiding future research priorities.
Overview: Reviews advancements in knowledge synthesis methods, including systematic reviews, meta-analyses, qualitative synthesis, and mixed-methods reviews, with examples in cardiopulmonary research. Highlights practical applications and reporting guidelines, enhancing the rigor of synthesis in health sciences.
Overview: outlines a scoping review to identify and compare various knowledge synthesis methods, addressing quantitative, qualitative, and mixed-methods approaches; provides a framework for matching synthesis methods to research questions, particularly in healthcare.
Overview: This paper provides a comprehensive introduction to knowledge synthesis, discussing methods such as systematic reviews, meta-analyses, and realist reviews; emphasizes transparency, reproducibility, and strategies to avoid bias in synthesis processes. This is a foundational guide for researchers, widely cited for its clarity on methodological rigor.
Overview: A classic article that categorizes 14 types of knowledge synthesis reviews (e.g., systematic, scoping, narrative) and outlines their methodologies, strengths, and limitations. Helps with selecting appropriate synthesis methods, influencing research design across disciplines.
Overview:Examines methods for integrating qualitative and quantitative evidence in knowledge synthesis, including narrative and mixed-methods approaches. Key resource for mixed-methods synthesis, influencing interdisciplinary research.
Note: Please use your critical reading skills while reading entries. No warranties, implied or actual, are granted for any health or medical search or AI information obtained while using these pages. Check with your librarian for more contextual, accurate information.
_paper_on_otto-SR.png)
