Skip to Main Content

Systematic reviews and other evidence synthesis projects

To Do during this step

  • Formulate your research question
  • Identify the important concepts
  • Develop your search strategy
  • Identify relevant databases and grey lit sources

Developing your Question

Developing your research question is one of the most important steps in the evidence synthesis process. At this stage in the process, you and your team have identified a knowledge gap in your field and are aiming to answer a specific question:

  • If X is prescribed, then Y will happen to patients?

OR assess an intervention:

  • How does X affect Y?

OR synthesize the existing evidence

  • What is the nature of X? ​

​​Whatever your aim, formulating a clear, well-defined research question of appropriate scope is key to a successful evidence synthesis.

The research question will be the foundation of your synthesis. From it your research team will identify 2-5 possible search concepts. Later, you will use these search concepts to build your search strategy. 

The nature of your question will help you figure out what type of evidence synthesis to write: Do you have a focused research question with narrow parameters? You may want a systematic review, especially if it fits into the PICO question format. Do you have a broad question that looks at answering larger, more complex, exploratory research questions? You may want a scoping review.

Remember that one type of review is not inherently better: they each serve different purposes. But all will benefit from planning and documenting your process.

Equity Considerations

When formulating a review question, think about the objective of your research! Traditionally these reviews are for research with purposes of intervention. Thinking about how the review or research question affects various populations, explicitly reporting the purpose of intervention, as well as considering aspects of life that may impact your information. Consider the following levels of impact:

  • Your objective of intervention
  • The effect on various populations as a result 
  • Outside factors influencing the data that impact information

Make sure that you can start to conceptualize these levels of impact from your question. Continue to think about other modes of impact in your research project to ensure equitable steps and practices. 

Research Question Frameworks

Formulating a research question takes time and your team may go through different versions until settling on the right research question. To help formulate your research question, some research question frameworks are listed below.

A well-formulated research question will make the searching step easier and faster. The concepts you identify now will be used in your concept table.

PICO

The PICO model is probably the most commonly-used framework, especially for clinical research questions:

  • Patient or Population in relation to the Problem
    • What are the characteristics of the patient or population?
    • What is the condition, disease, or situation you are interested in?
  • Intervention or exposure
    • What is the intervention, test, or exposure you want to find out about in relation to the patient/population/problem?
  • Comparison (not always applicable)
    • What is the alternative to the intervention (e.g. placebo, different drug, surgery)?
    • If the alternative is usual care or no intervention, skip Comparison
  • Outcome
    • What are the relevant outcomes (e.g. morbidity, death, complications, quality of life)?

Example: Is gabapentin (intervention), compared to placebo (comparison), effective in decreasing pain symptoms (outcome) in middle aged male amputees suffering phantom limb pain (population)?

Sometimes the model is described as PICO(T), where the T stands for Time or for Type of Study (the study design(s) most appropriate to answer the question).

While PICO is a helpful framework for clinical research questions, it may not be the best choice for other types of research questions, especially outside the health sciences. Here are a few others:

PICo for Qualitative Studies

  • P       Population/Problem
  • I         Phenomenon of Interest
  • Co    Context

Example: What are the experiences (phenomenon of interest) of caregivers providing home based care to patients with Alzheimer's disease(population) in Australia (context)?

 

​​SPICE

  • S    Setting
  • P   Perspective (for whom)
  • I    Intervention/Exposure
  • C   Comparison
  • E   Evaluation

Example: What are the benefits (evaluation) of a doula (intervention) for low income mothers (perspective) in the developed world (setting) compared to no support (comparison)?

 

SPIDER

  • S     Sample
  • PI   Phenomenon of Interest
  • D    Design
  • E     Evaluation
  • R    Study Type

Example: What are the experiences (evaluation) of women (sample) undergoing IVF treatment (phenomenon of interest) as assessed?
Design: questionnaire or survey or interview
Study Type: qualitative or mixed method

 

There are dozen of different types of these frameworks. For a table of other common types, see the University of Maryland Libraries' Developing a Research Question guide. For a comprehensive but concise overview of the almost 40 different types of research question frameworks, see this review from the British Medical Journal: Rapid review of existing question formulation frameworks.¹

¹Published as supplemental material to Booth A, Noyes J, Flemming K, Moore G, Tunçalp Ö, Shakibazadeh E. Formulating questions to explore complex interventions within qualitative evidence synthesis. BMJ Glob Health. 2019;4(Suppl 1):e001107. Published 2019 Jan 25. doi:10.1136/bmjgh-2018-001107.

Boolean Operators

Boolean operators are the way that AND, OR, and NOT are used in databases. You will use them to build your search strategy.

If you search on...

  • Myocardial Infarction AND stroke: All results will contain both myocardial infarction and stroke.
  • Myocardial infarction OR stroke: Results may include just myocardial infarction, just stroke, or both.
  • Myocardial infarction NOT stroke: Any records containing stroke will be excluded. Use very carefully! It is easy to accidentally remove relevant records.

Parentheses enable the creation of multi-part searches. They allow you to group your terms and control the order in which the database interprets them. With no parentheses, the database will process the terms from left to right.

  • asthma AND (children OR pediatric): All results will contain both asthma and one or more of the synonyms.
  • asthma AND children OR pediatric: Results may include asthma and children, just pediatric, or both sets of terms.

Developing your Search Strategy

To formulate your research question, you identified the important concepts. These concepts will be the building blocks for your search.

For now, focus on the concepts. You will develop the terms later.

A simplified version of how this works can be written as:
(Concept_A) AND (Concept_B) AND (Concept_C) AND (Concept_D)

For each concept, you will develop search terms. These will be joined by OR:
(TermA1 OR TermA2 OR...) AND (TermB1 OR TermB2 OR...) AND (TermC1 OR TermC2 OR...) AND (TermD1 OR TermD2 OR...)

A search concept table can help you organize the concepts and terms:
Blank concept table (Word doc) (PDF)
developed by Cushing/Whitney Medical Library, Yale University

 

Example: Concept tables for PubMed and PsycInfo for the research question What is the effectiveness of acupuncture vs. hypnosis for smoking cessation?

 PUBMED concept: smoking cessation concept: hypnosis concept: acupuncture
Database Thesaurus Terms
(MeSH terms)		   smoking cessation
  smoking/prevention and control
  smoking/therapy
  tobacco use disorder/prevention and control
  tobacco use disorder/therapy		   hypnotherapy   acupuncture therapy
Keywords   smoking
  smoker*		   hypnosis
  hypnotherapy		   acupuncture
  acupressure

 

(“smoking cessation”[Mesh] OR “smoking/prevention and control”[Mesh] OR “smoking/therapy”[Mesh] OR “tobacco use disorder/prevention and control”[Mesh] OR “tobacco use disorder/therapy”[Mesh] OR smoking OR smoker*)
AND
(“hypnotherapy”[Mesh] OR hypnosis OR hypnotherapy)
AND
(“acupuncture therapy”[Mesh] OR acupuncture OR acupressure)

 

 PsycInfo (EBSCO) concept: smoking cessation concept: hypnosis concept: acupuncture
Database Thesaurus Terms  smoking cessation
  tobacco smoking
  tobacco use disorder		   hypnosis
  hypnotherapy
  autohypnosis		  acupuncture
Keywords   smoking
  smoker*

  hypnosis
  hypnotherapy

   acupuncture
  acupressure

 

(DE "Smoking Cessation" OR  DE "Tobacco Smoking" OR DE "Tobacco Use Disorder" OR smoking OR smoker*)
AND
(DE "Hypnosis" OR DE "Hypnotherapy" OR DE "Autohypnosis" OR hypnosis OR hypnotherapy)
AND
(DE “Acupuncture” OR acupuncture OR acupressure)

Selecting Databases

Most evidence syntheses in health sciences will include the three core biomedical databases, but the others you include depend on your research question. Some frequently-used ones are described below, but the UW has access to many more.
Not sure what to include? Schedule an appointment with a librarian!

Core biomedical:

More:

  • Campbell Collaboration - Systematic reviews of the effects of social interventions in crime & justice, education, international development, and social welfare
  • CINAHL - Cumulative Index to Nursing and Allied Health Literature
  • ERIC - Education-related literature, from pre-K through adult education
  • Global Health Database - Global public health research and practice
  • PsycInfo - Psychology literature and psychological aspects of other topics
  • Web of Science - Multidisciplinary database covering journals in the sciences, social sciences, and arts and humanities
  • WHO Global Index Medicus - Global health topics from international journals

For additional specialty databases, see:

Grey Literature: beyond journal articles (eg conference proceedings, dissertations, unpublished reports, etc.)
    Read more about grey lit--what is it, why to use it, and ways to find it