Skip to main content

Systematic Reviews: Plan

Basic information on types of "knowledge syntheses"/systematic reviews to help users plan the best review type, team, methods and have a clear sense of the expectations/responsibilities of performing these comprehensive literature review projects.

Organizing the work

Any complex project flows more smoothly with forethought and careful planning. Researcher experiences indicate that the following are worth serious consideration before beginning:

1. Study team

a. Number:: Minimum of 2 reviewers of abstracts and full-text. Ideal: Input from a 3rd  achieving consensus.

b. Expertise:

  • Searching (bibliographic databases, grey literature) – librarian or experienced searcher
  • Review process – someone who has performed the type of review you propose to do at least once before
  • Evaluation of articles for scientific rigor/methodologic quality and risk of bias
  • Biostatistics – if you’re planning to do a meta-analysis, someone with expertise in the types of missing and/or inconsistent data you’re likely to encounter could make a huge difference in the quality of your review and its generalizability

2. Time

a. Overall - A full systematic review will probably take about 18 months from conception to submission for publication.

b. Individual time commitments:

1) Depend upon the number of studies being reviewed and the tasks.

a) 1st round of reviews (titles and abstracts) UF libraries experience estimates  2-15 hours

b) 2nd round of reviews (full-text) - possibly twice the above estimate

2) Depend on the experience of the reviewers: Horton et al* found that median data extraction time varied from 171 minutes per selected article for experienced data extractors to 246. 

* Horton J, Vandermeer B, Hartling L, Tjosvold L, Klassen TP, Buscemi N. Systematic review data extraction: cross-sectional study showed experience did not increase accuracy. J Clin Epi 63 (2010): 289-298. 

3) Most of the projects with which UF HSCL librarians have been involved have stalled when study team reviewers faced the time-consuming processes of reviewing or extracting data. 

3. Protocol

A written protocol lays out the methods, facilitating writing up that section when the team begins to draft for publication and clarifying to the entire team what the project will entail. In the event new team members are recruited, the written protocol can bring them up to speed quickly. The protocol should address the following issues:

  • Specific research question being addressed (emphasis on specific-this is the best point in the project to prevent "mission creep")
  • Purpose for the project or for addressing the research question
  • Names of team members and role(s) each will play in the project
  • Deadlines for various phases of the project
  • Frequency and means of communication between study team members (recommend regular intervals rather than after completion of phases to avoid the worst stalling of the project)
  • Inclusion/exclusion criteria for studies
  • Method for resolving disagreements (e.g.  on studies to be included)
  • Process for addressing shifts in study team membership, including time commitment shifts
  • Dissemination strategy: target conferences/journals, timeline, responsibilities of team members

GENERAL TIPS

Registering your protocol for a systematic review on PROSPERO , an international registry of proposed systematic reviews, provides documented proof of your intent to write on the research question. This may discourage others from embarking on the same project and possibly getting it published before yours. 

Look closely at the sections of the PRISMA flow chart to see the counts and rationales that may be expected or required at various points in the project.  

Investigate the Equator Network’s reporting standards for criteria you may need to report on when you finally write up the project. 

Form for working with librarians

University of Florida Home Page

This page uses Google Analytics - (Google Privacy Policy)

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.