A librarian helping a health professional to locate a book.A librarian helping a health professional to locate a book.A librarian helping a health professional to locate a book.Clinical tools showcaseClinical tools showcaseClinical tools showcase


Systematic reviews

Thinking of conducting a systematic review?

It's worth considering the following:

  • A systematic review usually takes 12-18 months to complete - do you have time?
  • Do you have a clearly defined question with established inclusion and exclusion criteria?
  • Do you have a team of at least 3 people assembled?
  • Do you have capacity to screen a very large number of search results?
  • Do you have the resources to translate foreign language articles?
  • Do you have access to statistical support to analyse and pool data?

If you answered "No" to any of these, a traditional literature review (often called "narrative review") will be more appropriate.

Before you start

Establish that a systematic review on your topic has not already been published. Two key databases to search for existing systematic reviews are Epistemonikos or the Cochrane Library.

What is a systematic review?

systematic review requires a team – it cannot be completed by an individual. 
You need to work with subject experts to clarify issues related to the topic; librarians to develop comprehensive search strategies and identify appropriate databases; reviewers to screen abstracts and read the full text; and/or a statistician who can assist with meta-analysis.

Cochrane is globally accepted as the standard for systematic reviews (SRs) and most publishers have the expectation that any SR will follow the Cochrane Reviewers handbook.

“A systematic review attempts to collate all empirical evidence that fits pre-specified eligibility criteria in order to answer a specific research question. It uses explicit, systematic methods that are selected with a view to minimizing bias, thus providing more reliable findings from which conclusions can be drawn and decisions made (Antman 1992; Oxman 1993).”

Key stages

A systematic review usually takes 12-18 months and the key steps are:

  1. Define the question and assemble your team
  2. Write a review protocol
  3. Develop and test search strategies
  4. Select studies as per your protocol
  5. Conduct quality assessment of included studies
  6. Extract data
  7. Synthesise data (meta-analysis if applicable)
  8. Write up the review

For a visual overview of the tasks and stages see "The Systematic Review Process" - Yale University

Twelve tips for medical students to conduct a systematic review. Medical teacher 2019; 41(4): 471–475 

Protocol & Methodology

Review protocol:

A protocol is required to outline the methodology, rationale, and eligibility criteria. PRISMA-P was published in 2015 aiming to facilitate the development and reporting of systematic review protocols. There is an explanation and elaboration paper available and an operationalized checklist from BMC Systematic Reviews (Word, 35KB) enabling you to mark completed sections.

It is recommended that protocols are registered with PROSPERO. PROSPERO is the only open access prospective register of systematic reviews with a health related outcome. Registration is web-based, completely free to use and open to all researchers planning to conduct a systematic review. 
PROSPERO can be a useful resource to search at the beginning of your research to establish what research is currently taking place in your field. 

Review methodology:

Search strategies for a systematic review need to include a combination of subject headings and text terms. A detailed search is usually established and tested in one database - often MEDLINE, depending on the nature of the question. Once the author team is satisfied that the search is sufficiently comprehensive and appropriate, the search strategy is then accurately “translated” across other relevant databases.

The number of results from each database search are noted for the PRISMA flow chart and then exported to reference management software such as EndNote. It is recommended that deduplication be done using EndNote.

Grey literature also needs to be searched to assist in removing publication bias. The Austin Librarians can recommend how to search for this material.

Screening occurs in a two-stage process according to pre-established inclusion and exclusion criteria. Austin Health has an organisational account for Covidence, a tool which can assist with these steps.

Quality assessment and data extraction involve transcribing information from the primary studies under review to a standard form or template that has been designed to capture all relevant details. Covidence includes customisable data extraction forms and automatically populates Risk of Bias tables. There are several options for appraising the evidence you will use in a SR: 

Meta-analysis occurs at this stage if the included study data are sufficiently homogenous.

Prior to writing the review the database searches need to be updated to capture any new evidence.


  • Cochrane Handbook and CRD’s guidance for undertaking reviews in healthcare outline the importance of a clearly-formulated question, the need for a review team as well as the steps involved in preparing the protocol and all other aspects of the review.
  • EQUATOR network (Enhancing QUAlity and Transparency Of health Research) provides reporting guidelines for different study types including systematic reviews.
  • IOM Standards for Systematic Reviews provide items for initiating and reporting a systematic review. (PDF download available via free registration)
  • PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) is an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses. PRISMA 2020 focuses on the reporting of reviews evaluating randomized trials, and can also be used as a basis for reporting systematic reviews of other types of research, particularly evaluations of interventions. The recently released PRISMA-S extension provides additional reporting items that specifically relate to the search process. We recommend that you become familiar with the PRISMA 2020 checklist, PRISMA-S checklist and PRISMA 2020 flow diagram as a key section of your methodology for your systematic review.

Next steps

1.       Establish that a systematic review is the best method for answering your question

2.       Search the databases Epistemonikos and Cochrane Library to ensure that a systematic review has not already been conducted on your topic

3.       Check the Prospero database to establish if others have already registered their intention to conduct a review on your topic

4.       Start drafting a protocol for your systematic review

5.       Contact the Library for further assistance once you have a draft protocol written

The Library team will discuss the level of involvement you wish to have from an Austin Librarian and what capacity we currently have to assist you. The options will be to:

a) provide guidance and advice for you to create, run and manage the searches and review process yourself; or

b) we do more of the process on your behalf – namely we would design and manage complex searches in multiple databases ensuring that all best practice requirements for conducting a systematic review are met (aligning with the IOM standards for systematic reviews). 
We provide an EndNote library of de-duplicated results ready for upload into Covidence, search strategies for your manuscript, narratives of the search methodology, and we will review the manuscript -  with the expectation that we are included as a co-author.