Systematic Reviews- Legacy Guide
What Is a systematic review?
"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). The key characteristics of a systematic review are:
a clearly stated set of objectives with pre-defined eligibility criteria for studies;
an explicit, reproducible methodology;
a systematic search that attempts to identify all studies that would meet the eligibility criteria;
an assessment of the validity of the findings of the included studies, for example through the assessment of risk of bias; and
- a systematic presentation, and synthesis, of the characteristics and findings of the included studies"
Higgins, Julian. Cochrane Handbook for Systematic Reviews of Interventions. Hoboken, NJ, USA: John Wiley & Sons, 2009. ProQuest ebrary. Web. 7 October 2014.
What do I need to get started?
Consider the following questions before you begin a systematic review.
1. Is your question suited to a systematic review?
While a systematic review is considered the highest form of evidence, not all questions are suited to a systematic review. A question that is well suited to this methodology is:
- based on a specific question or patient problem and will ensure that your search produces relevant results.
- narrow enough that identifying and reviewing all of the relevant literature is possible. An extremely broad question (e.g. examining pain management and assessment) may be more suited to a general literature review or evidence mapping.
- broad enough that you will be able to find some relevant literature. An extremely narrow topic (e.g. an examination of a very small subset of a population) or one that is examining a very new treatment or concept may not return enough relevant literature for a systematic review to be able to draw any meaningful conclusions.
- defined with some consistency across the literature. A question with a multi-stepped intervention, or one in which different elements of an intervention may or may not be included in each study may be more suited to a decision modeling approach.
2. Are you asking a complete question?
Thinking about your PICO (population, intervention(s), comparator(s), and outcomes) and your Key Question(s) will help you to ensure that you have thought of the important parts of your question and are not leaving out anything that may significantly alter your question. You can use the HSL Systematic Review Development Worksheet to help you refine and develop your PICO and Key Questions.
3. Has a recent systematic review already addressed this exact question?
If you determine that your question is suited to a systematic review, you will then want to search for recent or upcoming reviews on your topic or on similar topics, to ensure that your review will not be duplicating someone else's efforts and that you will be adding something new to the literature. A librarian can help you to search for recently published reviews on your topic, as well as searching systematic review protocol registries such as PROSPERO, which will give you an idea of whether or not a review of your topic is likely to be published soon.
4. Do you have the time and resources needed to conduct a systematic review?
A systematic review usually takes at least 1 year, or an average of 1139 hours, from formulating your question through journal acceptance. Input from at least 3 team members is needed, in addition to advice or input from a librarian or team member who is experienced in searching the literature. One team member handles the processes of the review: creating data abstraction forms, keeping track of inclusion/exclusion decisions, handling team communications, managing databases and forms, etc. Two team members perform inclusion/exclusion decision making and data abstraction. The additional team member acts as a tiebreaker in the case of a disagreement. One to two team members analyze the data and write your publication. Depending on the size of the literature, you may want to add additional team members. A team of up to ten or twelve people is not unusual for a large systematic review.
5. Can you meet all of the requirements for a systematic review detailed in the PRISMA checklist?
The PRISMA checklist contains 27 items that each systematic review should report in order to ensure transparency and completion of reporting. These items include parts to include in your title, abstract, introduction, methods, results, discussion, Using this checklist will help to ensure that your methods are considered adequate by the journal you submit your review to.