Systematic reviews: finding and managing the evidence

 How to focus your question

Before you begin searching for literature to support your review, you need to carefully focus and define your research question.

It's important to be as specific as possible when defining your research question in order to develop an effective search strategy to find all relevant papers that match your review question.  This is a very important aspect of creating a systematic review, as your search needs to be as comprehensive as possible in order to create a high quality, unbiased review.

The best way to initially start planning your search is to break up your topic into its main concepts or ideas, and then develop a list of keywords related to each of your concepts in advance of searching your databases.

Planning with P I C O S- quantitative studies

To help you focus your question for quantitative research topics and begin planning your search strategy, use frameworks such as P I C O S. 

P I C O S, a variation of the PICO tool described by Richardson, W et al in 1995,   is explained below:

• P identify the specific characteristics of their Population group and their Problem. For example: ethnicity, age, gender, condition or disease, presentation, healthcare setting
   
• I -specify which Intervention or exposure is being looked at.  For example: treatment (e.g. drug therapy, surgery or non-pharmacological treatments); service intervention (e.g. midwifery-led service), diagnostic test, risk factors (e.g. smoking)
   
• O- outline which, if any, Comparisons are being made with alternative interventions or population groups.  What is it that we are comparing against?
   
• C- target which Outcomes are of importance to the review question. Outcomes could be patient related (such as mortality, return to function, patient satisfaction or complications) or  cost related (such as cost-effectiveness, length of stay or reduction in referrals)
   
• S- state which type of Studies would best provide answers to the question. For example, would randomized controlled trials be the best kind of study design to answer your question?

It is not essential for each of these elements to appear in your search strategy but all research questions should at the minimum have a population group and intervention of interest. In many cases the outcomes are not included as part of the search strategy itself, but rather these will act as a guide when you are reviewing papers to include and exclude to define your final set of papers that will form the basis of your review.

Click here for a Cochrane presentation for more detailed information on PICOS. 

Planning with S P I D E R- qualitative studies

For qualitative studies, the framework S P I D E R , first described by Cook, A, Smith, D and Booth, A (2012) and may work better than P I C O S

• S- Sample - who are the people involved in the study?
• PI-  Phenomena of Interest- What do you hope to understand? Beliefs, attitudes or experiences?
• D- Which methodological Design works best for your research
• E- Evaluation- What outcomes measures will you judge the results by?
• R- What Research type best suits your question: Qualitative, quantitative or mixed methods?

As with P I C O S, it is not essential for each of these  S PI D E R elements to appear in your search strategy.

Why spend time planning?

If you do not allot  time to  focus your question and plan your search strategy, your review may suffer from the following:

• If your question is too broad you may be overwhelmed with results to sift through
   
• If you do not identify the main concepts in your research and draw up a list of alternative words or phrases for these concepts, you may potentially not find all the papers relevant to your research question, thus introducing bias into your review
   
• If you skip this stage, your database searching will lack structure and, as a result, be less effective

For more additional guidance on how to focus your question, read this article by Andrew Booth from ScHARR.

Search plan templates, worked examples and information about other planning frameworks are available at the bottom of this page.

Example of an unfocused question:

"I need information on diabetes and patient education?"

Example of a more focused question:

”Is patient education effective in improving blood sugar control
in teenage diabetics?”

Below is a PICOS framework for the question:

”Is patient education effective in improving blood sugar control
in teenage diabetics?”

Below is a SPIDER framework for the  question:

"What are the attitudes of elderly people with stroke towards the daily use of assistive devices and technologies?"

Alternatives to P I C O S and S P I D E R

P E S T E L :

• Political
• Economic 
• Social 
• Technological
• Environmental 
• Legal Factors 

S P I C E :

• Setting
• Population
• Intervention
• Comparison
• Evaluation 

E C L I P S [management and service related issues] :

• Client Group
• Location
• Impact,
• Professionals Involved
• Service 

M I P [medical ethics review] :

• Methodology
• Issues
• Participants