A systematic review aims to answer a clear and focused clinical question. The question guides the rest of the systematic review process. This includes determining inclusion and exclusion criteria, developing the search strategy, collecting data and presenting findings. Therefore, developing a clear, focused and well-formulated question is critical to successfully undertaking a systematic review.
A good review question:
Research topic vs review question
A research topic is the area of study you are researching, and the review question is the straightforward, focused question that your systematic review will attempt to answer.
Developing a suitable review question from a research topic can take some time. You should:
When considering the feasibility of a potential review question, there should be enough evidence to answer the question whilst ensuring that the quantity of information retrieved remains manageable. A scoping search will aid in defining the boundaries of the question and determining feasibility.
For more information on FINER criteria in systematic review questions, read Section 2.1 of the Cochrane Handbook.
Check for existing or prospective systematic reviews
Before finalising your review question, you should determine if any other systematic review is in progress or has been completed on your intended question (i.e. consider if the review is Novel).
To find systematic reviews you might search specialist resources such as the Cochrane Library, Joanna Briggs Institute EBP Database or the Campbell Collaboration. "Systematic review" can also be used as a search term or limit when searching the recommended databases.
You should appraise any systematic reviews you find to assess their quality. An article may include ‘systematic review’ in its title without correctly following the systematic review methodology. Checklists, including those developed by AMSTAR and JBI, are useful tools for appraisal.
You may undertake a review on a similar question if that posed by a previously published review had issues with its methodology such as not having a comprehensive search strategy, for example. You may choose to narrow the parameters of a previously conducted search or to update the review if it was published some years ago.
Searching a register of prospective systematic reviews such as PROSPERO will allow you to check that you are not duplicating research already underway.
Once you have performed scoping searches and checked for other systematic reviews on your topic, you can focus and refine your review question. Any PICO elements identified during the initial development of the review question from the research topic should now be further refined.
The review question should always be:
|Work through the first section of the PICO worksheet to define your review question|
Review questions may be broad or narrow in focus; however, you should consider the FINER criteria when determining the breadth of the PICO elements of your review question.
A question that is too broad may present difficulty with searching, data collection, analysis, and writing, as the number of studies retrieved would be unwieldy. A broad review question could be more suited to another type of review.
A question that is too narrow may not have enough evidence to allow you to answer your review question. Table 2.3.a in the Cochrane Handbook summarises the advantages and disadvantages of broad versus narrow reviews and provides examples of how you could broaden or narrow different PICO elements.
It is essential to formulate your research question with care to avoid missing relevant studies or collecting a potentially biased result set.
A systematic review protocol is a document that describes the rationale, question, and planned methods of a systematic review. Creating a protocol is an essential part of the systematic review process, ensuring careful planning and detailed documentation of what is planned before undertaking the review.
The Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) checklist outlines recommended items to address in a systematic review protocol, including:
The PICO Worksheet has been designed to help you create your systematic review protocol
Systematic reviews must have pre-specified criteria for including and excluding studies in the review. The Cochrane Handbook states that "predefined, unambiguous eligibility criteria are a fundamental prerequisite for a systematic review."
The first step in developing a protocol is determining the PICO elements of the review question and how the intervention produces the expected outcomes in the specified population. You should then specify the types of studies that will provide the evidence to answer your review question. Then outline the inclusion and exclusion criteria based on these PICO elements.
For more information on defining eligibility criteria, see Chapter 3 of the Cochrane Handbook.
A key purpose of a protocol is to make plans to minimise bias in the findings of the review; where possible, changes should not be made to the eligibility criteria of a published protocol. Where such changes are made, they must be justified and documented in the review. Appropriate time and consideration should be given to creating the protocol.
You may wish to register your protocol in a publicly accessible way.
This will help prevent other people from completing a review on your topic.
If you intend to publish a systematic review in the health sciences, it should conform to the IOM Standards for Reporting Systematic Reviews.
If you intend to publish a systematic review in the Cochrane Database of Systematic Reviews, it should conform to the Methodological Expectations in Cochrane Intervention Reviews (MECIR).
A clinical question needs to be directly relevant to the patient or problem and phrased to facilitate the search for an answer. A clear and focused question is more likely to lead to a credible and useful answer, whereas a poorly formulated question can lead to an uncertain answer and create confusion.
The population and intervention should be specific, but if any or both are described too narrowly, it may not be easy to find relevant studies or sufficient data to demonstrate a reliable answer.
Types of clinical questions
|Question type||Explanation||Evidence types required to answer the question|
|Diagnosis||Questions about the ability of a test or procedure to differentiate between those with and without a disease or condition||Randomised controlled trial (RCT) or cohort study|
|Etiology (causation)||Questions about the harmful effect of an intervention or exposure on a patient||Cohort study|
|Meaning||Questions about patients' experiences and concerns||Qualitative study|
Questions about the effectiveness of an intervention or exposure in preventing morbidity and mortality. Questions are similar to treatment questions. When assessing preventive measures, it is essential to evaluate potential harms as well as benefits
|Randomised controlled trial (RCT) or prospective study|
|Prognosis (forecast)||Questions about the probable cause of a patient's disease or the likelihood that they will develop an illness||Cohort study and/or case-control series|
|Therapy (treatment)||Questions about the effectiveness of interventions in improving outcomes in patients suffering from an illness, disease or condition. This is the most frequently asked type of clinical question. Treatments may include medications, surgical procedures, exercise and counselling about lifestyles changes||Randomised controlled trial (RCT)|
PICO is a framework for developing a focused clinical question.
Slightly different versions of this concept are used to search for quantitative and qualitative reviews, examples are given below:
PICO for quantitative studies
|Population/Patient/Problem||Intervention or Exposure||Comparison or Control||Outcome|
What are the characteristics of the Population or Patient?
|How do you wish to Intervene? What do you want to do with this patient - treat, diagnose, observe, etc.?||What is the Comparison or alternative to the intervention - placebo, different drug or therapy, surgery, etc.?||What are the possible Outcomes - morbidity, death, complications, etc.?|
Here is an example of a clinical question that outlines the PICO components:
PICo for qualitative studies
What are the characteristics of the Population or Patient?
|Interest relates to a defined event, activity, experience or process||Context is the setting or distinct characteristics|
Here is an example of a clinical question that outlines the PICo components:
Two other mnemonics may be used to frame questions for qualitative and quantitative studies - SPIDER and SPICE.
SPIDER for qualitative or quantitative studies
SPIDER can be used for both qualitative and quantitative studies:
|Sample||Phenomenon of Interest||Design||Evaluation||Research type|
|Sample size may vary in quantitative and qualitative studies||Phenomena of Interest include behaviours, experiences and interventions||Design influences the strength of the study analysis and findings||Evaluation outcomes may include more subjective outcomes such as views, attitudes, etc.||Research types include qualitative, quantitative, or mixed-method studies|
Within social sciences research, SPICE may be more appropriate for formulating research questions:
|Setting is the context for the question - where||Perspective is the users, potential users or stakeholders of the service - for whom||Intervention is the action taken for the users, potential users or stakeholders - what||Comparison is the alternative actions or outcomes - what else||Evaluation is the result or measurement that will determine the success of the intervention - what result or how well|
More question frameworks
For more question frameworks, see the following: