Systematic Reviews & Evidence Synthesis Methods

Formulating a strong research question for a systematic review can be a lengthy process. While you may have an idea about the topic you want to explore, your specific research question is what will drive your review and requires some consideration. 

You will want to conduct preliminary or exploratory searches of the literature as you refine your question. In these searches you will want to:

• Determine if a systematic review has already been conducted on your topic and if so, how yours might be different, or how you might shift or narrow your anticipated focus.
• Scope the literature to determine if there is enough literature on your topic to conduct a systematic review.
• Identify key concepts and terminology.
• Identify seminal or landmark studies.
• Identify key studies that you can test your research strategy against (more on that later).
• Begin to identify databases that might be useful to your search question.

A narrow and specific research question is required in order to conduct a systematic review. The goal of a systematic review is to provide an evidence synthesis of ALL research performed on one particular topic. So your research question should be clearly answerable from the data you gather from the studies included in your review. 

Another consideration is whether the question has been answered enough to warrant a systematic review. If there have been very few studies, there won't be enough data to synthesize. You then have to adjust your question... widen the population, etc.

When developing your question, it can be helpful to consider the FINER criteria (Feasible, Interesting, Novel, Ethics, and Relevant). Read more about the FINER criteria on the Elsevier blog.

If you have a broader question or aren't certain that your question has been answered enough in the literature, you may be better served by pursuing a systematic map, also know as a scoping review. 

Try framing your research question. For example, a commonly used framework in health sciences is PICO: Patient/Population, Intervention, Comparison, Outcome.

Try it out in the PICO Linguist tool.

Another common framework is PEO (population, exposure, outcomes).

For other frameworks, you can use:

• This excellent guide from City University of London, on choosing research question frameworks.
• This guide from Cornell University.
• Table 1 in the article How to Do a Systematic Review (pages 140 - 141).
• A chart in Step 1-1 from PIECES; Excel workbook to choose the framework that best fits your research question. Created by Magaret J. Foster at Texas A&M University. CC-BY-3.0 license.