Systematic Reviews and other evidence synthesis projects

What is a protocol?

The PRISMA guidelines for the transparent reporting of systematic reviews and meta-analyses define a protocol as a document that  "describes the rationale, hypothesis, and planned methods of the review. It should be prepared before a review is started and used as a guide to carry out the review."

Why write a protocol?

• Provides a foundation...
  • For the research:
    • Acts as a guard against arbitrary decision-making during review process
    • Addresses methodological issues at the beginning of the process
  • For the project:
    • Acts as road map for process
    • Identifies the project's scope, resources, team members, and timeline
• Improves the quality and rigor of the review
  • Improves transparency of completed review
  • Enables readers to assess for selective reporting or deviation from planned methods.
• May be a requirement for publication
• Complies with guidelines & standards

Why register a protocol?

• Establishes a public document stating that you are working on a topic
• Reduces wasteful duplication of reviews
• Creates opportunity for peer review to improve review methodology
• Potentially increases visibility to potential researchers or editors
• Establish credibility and accountability
• An increasing number of journals require it

A 2018 study identified an association between registration of a protocol in PROSPERO and the overall review quality of its systematic review based on the AMSTAR tool. The authors speculated that the time spent planning and documenting all of the review procedures might indirectly result in more robust methods. But regardless of where you register, completing a protocol is time well invested.

When should I register?

Review the checklist for the PRISMA-P protocol guidelines or the PROSPERO registration form. Can you complete all the applicable items? If so, you may be ready to register. You will be expected to stick to your protocol just as if it were a lab protocol, so make sure you have documented the process you intend to follow.

There are several options for protocol format, but for all of them you should prepare to answer the following:

• Review title & review question
• Anticipated start date & completion date
• Research team members & roles
• Search strategy
• Inclusion & exclusion criteria for studies
• Data extraction strategy
• Risk of bias assessment/approach to critical appraisal
• Tools that will be used
• Dissemination plans

Here are some of the standard protocol templates. If you are conducting a review according to a specific standard, you must use that standard's protocol guidelines. If you are not writing to a specific standard, you may still use a standard's guidelines if they are appropriate to your topic, or use PRISMA-P.

• PRISMA for systematic review protocols (PRISMA-P)
  The Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) was developed to improve the quality of systematic review protocols and facilitate registration by providing authors with a minimum set of items to be included in the protocol. The 17 checklist items pertain to the administrative elements, rationale and objective, and methods of the review, including risk of bias assessment and data synthesis. The accompanying Explanation and Elaboration paper provides more detail about each of the items.
• Cochrane: Standards for the Reporting of Protocols of new Cochrane Intervention Reviews
  Systematic reviews performed according to the Cochrane standards are required to have a protocol that follows the standards described in Methodological Expectations for Cochrane Intervention Reviews (MECIR).
• JBI
  Protocols for different types of systematic reviews written according to JBI standards, and also for scoping reviews.
• Campbell Collaboration
  Campbell reviews and their protocols are conducted using templates in RevMan review support software. See Chapter 5 in the Campbell policies and guidelines for a list and descriptions of the fields.

Where you register your protocol depends partially on your type of review and your topic. Here are some of the options.

• PROSPERO
  For systematic reviews, meta-analyses, rapid reviews and umbrella reviews in health and social sciences. Does not accept scoping reviews.
• Cochrane Library
  For medical/healthcare systematic reviews conducted according to the Cochrane guidelines.
• Campbell Collaboration
  For systematic reviews in a broad range of social sciences, including nutrition and disability. Also accepts protocols for evidence and gap maps.
• BEME (Best Evidence Medical and Health Professional Education)
  For evidence syntheses on medical and health professions education.
• SYREAF (Systematic Reviews for Animals & Food)
  For evidence syntheses in food safety, animal health, and animal welfare.
• OSF, Figshare, Or Zenodo
  Protocols for any type of evidence syntheses in any discipline can be published on these open science platforms as registration. Supplemental materials such as full search strategies can be added to the file later in the process.
• Journals that publish protocols
  • BMC Systematic Reviews: https://systematicreviewsjournal.biomedcentral.com/
  • JMIR Research Protocols: https://www.researchprotocols.org/
  • JBI Evidence Synthesis: https://journals.lww.com/jbisrir/pages/default.aspx
  • BMJ Open: https://bmjopen.bmj.com/