Systematic Reviews & Other Syntheses

Other introductory videos and modules:

• Introduction to Systematic Reviews by Queen's University 
• Systematic Reviews series from Yale University
• Introduction to Systematic Reviews from Johns Hopkins University
• Introduction to Conducting Systematic Reviews from Cochrane Interactive Learning (note: this particular module is free but creating an account is required).

Eight stages of a systematic review and meta-analysis* (Uman, 2011):

1. Formulate the review question
2. Define inclusion/exclusion criteria
3. Develop reproducible search strategies and locate studies
4. Select studies
5. Extract data
6. Assess study quality
7. Analyze and interpret results
8. Disseminate findings

*Important: before embarking on a systematic review, make sure that 1) a recent review on the same topic has not already been published, and 2) that a review protocol has not already been registered for the same topic. 

1. To check for systematic reviews that have already been published on your topic, search databases such as Ovid MEDLINE/PubMed, Embase, PsycINFO, CINAHL, and even Google Scholar to catch any journal content that is not covered in these databases. You can also search Epistemonikos for published systematic reviews, a freely available resource that attempts to identify all of the systematic reviews relevant for health decision-making.
    
2. To check for registered (i.e. forthcoming) systematic reviews, search PROSPERO, the international prospective register of systematic reviews. You can also search for registered protocols by systematic review collaborations such as Cochrane, Campbell and the Joanna Briggs Institute. When you are confident that your research will not be duplicating efforts, you are ready to begin your own review.

Cochrane and JBI describe a number of different systematic review question types: 

• Intervention reviews assess the benefits and harms of interventions used in healthcare and health policy
• Diagnostic test accuracy reviews assess how well a diagnostic test performs in diagnosing and detecting a particular disease
• Prognosis reviews address the probable course or future outcome(s) of people with a health problem
• Etiology and risk reviews assess the relationship (association) between certain factors and the development of a disease, condition or other health outcome
• Prevalence and incidence reviews assess proportional data (often percentages) of a population experiencing a particular disease or condition
• Cost/economic evaluation reviews assess intervention costs, costs relative to benefits, intervention resource use/costs and/or cost effectiveness in addition to a range of other questions about the intervention
• Experience reviews investigate perspectives and experiences of an intervention or health condition 

Most often the systematic review design/approach will depend on the review question:

• Systematic reviews of healthcare interventions are often quantitative reviews that include research studies in the form of randomized controlled trials.
• Systematic reviews of patient experiences are generally qualitative reviews that may include qualitative and mixed methods research studies.
• To address the limitations of single method reviews, which are frequently too narrow in scope, mixed methods reviews are utilized: "By including diverse forms of evidence from different types of research, mixed methods reviews attempt to maximize the findings - and the ability of those findings to inform policy and practice" (Peters et al., 2015, p. 5).

Some workload considerations:

• Research topic: The review topic will  impact the amount of time required to conduct the review. For example, a systematic review on the effectiveness of a relatively new drug therapy may require less time to complete if the search yields a low number of results to screen and if the review itself only includes a small number of studies.
• Searching for studies: The amount of time needed to complete this component of the review will depend on: how straight-forward or complex the topic is, the number and type of resources that will be searched (refer to Where to Search), and the searcher's level of expertise. Working with an experienced librarian is a best practice recommendation (refer to Librarian Support).
• Research team: Estimating how long it will take to complete a systematic review will depend on the size of the research team and how available the team members are to work on the review. Researchers often have competing priorities throughout the duration of the systematic review that can delay the process.

Research evidence:

• The average length of time for systematic review projects (using the project start date from the registered protocol in PROSPERO to the review's publication date) was 67.3 weeks (Borah, 2017).
   
• A research study found that the median time spent searching was just under 8 hours, including 1.5 hours spent searching the grey literature, while the average time spent searching was approximately 24 hours, of which 6.5 hours were spent searching the grey literature (Saleh et al., 2014). The average number of resources searched was 9, including grey literature resources. Grant funding influenced the amount of time spent searching and institution type impacted the number of resources searched.
   
• Greenhalgh and Peacock (2005) found that: "Electronic searching, including developing and refining search strategies and adapting these to different databases, took about two weeks of specialist librarian time..." (p1065).