Skip to Main Content

Evidence Syntheses and Systematic Reviews: Overview

What is Evidence Synthesis?

Evidence Synthesis: general term used to refer to any method of identifying, selecting, and combining results from multiple studies. There are several types of reviews which fall under this term; the main ones are in the table below: 

Types of Reviews

Type of Review Description Search Formal Inclusion Criteria Use of Protocols Results
Systematic Review Comprehensive literature synthesis on a specific research question, typically requires a team Systematic; exhaustive and comprehensive; search of all available evidence Yes Yes Narrative and tables, describes what is known and unknown, recommendations for future research, limitations of findings
Rapid Review A quicker, simplified version of the Systematic Review; to assess what is already known about a policy or practice issue May not involve as many databases as the Systematic Review Yes; determined by time constraints Yes; determined by time constraints Usually narrative summary, tables
Scoping review or systematic map Seeks to identify gaps, trends, themes, and opportunities for evidence synthesis on a broad topic Broad search, exploratory in nature Less rigorous Yes, but not a strict one May critically evaluate existing evidence, summarizes results qualitatively
Umbrella Review Synthesizes evidence from multiple systematic reviews Searches existing systematic reviews and meta-analyses Yes Yes Integrates findings from existing reviews; no statistical pooling; what is known – recommendations from practice; what is unknown – recommendations for future research
Meta-analysis A statistical technique for combining the findings from disparate quantitative studies, may stand alone or be part of a systematic review Searches results from multiple studies on a specific research question; may include unpublished studies Yes Yes Quantitative synthesis; numerical analysis of measures of effect
Narrative literature review Standalone review (not to be confused with a literature review in an empirical study), may be broad or focused, represents a range of levels of comprehensiveness May or may not be comprehensive Varies No Narrative describes what is known and unknown, recommendations for future research, limitations of findings

 

General Steps for Conducting Systematic Reviews

The number of steps for conducting Evidence Synthesis varies a little, depending on the source that one consults. However, the following steps are generally accepted in how Systematic Reviews are done:

Prepare

  1. Develop a Research Question and Apply a Framework
    • Identify a gap in the literature and form a well-developed and answerable research question which will form the basis of your search
    • Select a framework that will help guide the type of study you’re undertaking
  2. Select a Reporting Guideline
    • Different guidelines are used for documenting and reporting the protocols of your systematic review before the review is conducted. The protocol is created following whatever guideline you select.
  3. Select Databases and Grey Literature Sources
  4. Write a Search Strategy
    • For steps 3 and 4, it is advisable to consult a librarian before embarking on this phase of the review process. They can recommend databases and other sources to use and even help design complex searches.
  5. Register a Protocol
    • A protocol is a detailed plan for the project, and after it is written, it should be registered with an appropriate registry.

Search

  1. Search Databases and Other Sources
  2. Translate Search Strategies
    • Not all databases use the same search syntax, so when searching multiple databases, use search syntaxes that would work in individual databases.
  3. Manage Your Citations
    • Use a citation management tool to help store and organize your citations during the review process; great help when de-duplicating your citation results

Analyze and Report

Analyze

  1. Screen the Articles
    • Inclusion and exclusion criteria already developed help you remove articles that are not relevant to your topic. 
  2. Assess the Risk of Bias
    • Assess the quality of your findings to eliminate bias in either the design of the study or in the results/conclusions (generally not done outside of Systematic Reviews).

Extract and Synthesize

  1. Extract the data
    • Extract the data from what's left of the studies that have been analyzed
    • Extraction tools are used to get data from individual studies that will be analyzed or summarized. 
  2. Synthesize the main findings of your research

Report Findings

Report the results using a statistical approach or in a narrative form.

Need More Help?

Librarians can:

  • Provide guidance on which methodology best suits your goals
  • Recommend databases and other information sources for searching
  • Design and implement comprehensive and reproducible database-specific search strategies 
  • Recommend software for article screening
  • Assist with the use of citation management
  • Offer best practices on documentation of searches

Steps of a Systematic Review - Video