Skip to Main Content

Systematic Reviews and Evidence Syntheses

FINER criteria

Following the FINER criteria (Cummings, 2007; Cochrane Handbook § 2.1), the research question should be:

  • Answerable by the author team using the evidence available
  • Manageable quantities of information to synthesize
  • Up-front scoping work will help to define boundaries
  • Systematic reviews take a lot of work; the authors must be motivated to remain committed
  • Must address a genuine gap in knowledge
  • Authors must be aware of related/overlapping reviews to avoid redundancy
    • Pre-existing syntheses in the published litearture
    • Syntheses in progress (check PROSPERO register for systematic review protocols)
  • Research questions are often not value-neutral
  • How a problem is approached can have political/social implications
    • e.g., could result in unintentional widening of health disparities
  • Involve relevant stakeholders in the process of defining the focus and research questions to be addressed
  • Report results so as to facilitate translation of the findings to inform decisions.
    • Use the GRADE framework

Aims

A systematic review can address any question that a primary study can address (Cochrane Handbook § 2.2).

  • Relative effects of two interventions (or one intervention vs. controls)
    • Benefits and harms
  • Relative effects of many different interventions
    • This is achieved by network meta-analysis, which can be broader and more challenging, but is increasingly important in evidence synthesis
  • How a single intervention's effect size varies according to other factors:
    • Characteristics of the study cohort
    • Differences in intervention itself (dosage, route of administration, etc.)
    • Outcome measures
    • Methodology of primary studies (randomized controlled trials, quality improvement studies, retrospective observational studies, etc.)

Reviewers may need to address:

  • Mechanism of action
    • Not always necessary, but can aid the reader in determining how applicable the review is to their situation
  • Economic context
    • Policy and practice decisions are often subject to resource constraints

Formulating the research question

In plain English, a research question concerning the effect of an intervention can typically be formulated as the following, adapted from the Cochrane Handbook § 2.3:

What are the effects of [intervention or comparison] for [health problem] in [types of people, disease, or problem and setting if specified]

This formulation can often be broken down into the PICO (population, intervention, comparator, outcome) framework.

What are the effects of [intervention vs. comparator] for [outcome] in [population]

For more information on PICO and similar frameworks for other study/question type, see:

PICO

Creating PICO Questions

PICO Alternatives

PICO works best for quantitative questions related to interventions. If PICO isn't working for your question, one of these frameworks may be helpful:

 

  • PEO (Population/Problem, Exposure, Outcome/Themes) (for qualitative questions)
  • PFO (Population/Problem, Prognostic Factor, Outcomes/Themes) (for questions related to prognosis)
  • CoCoPop (Condition, Context, Population) (for questions related to prevalence/incidence)
  • CLIP (Client, Location, Improvement/Interest, Provider) (for questions related to cost effectiveness/service improvements)

PICO Help

References

Cummings SR, Browner WS, Hulley SB. Conceiving the research question and developing the study plan. In: Hulley SB, Cummings SR, Browner WS, editors. Designing Clinical Research: An Epidemiological Approach. 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2007. p. 14–22.

Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.4 (updated August 2023). Cochrane, 2023. Available from www.training.cochrane.org/handbook.

Top Bottom