Inclusion and Exclusion Criteria: Systematic Review Examples
Turn a broad review question into usable eligibility rules, with a worked example and a process for handling uncertain studies.
“Relevant articles were included” is not an eligibility criterion. It is the decision you need to explain.
Useful inclusion and exclusion criteria turn a review question into rules that can be applied consistently. They specify which populations, contexts, interventions or phenomena, designs and other features belong in the review. The exact dimensions depend on the review question and type.
In this guide
Write criteria that resolve real decisions
Consider a fictional review of interventions intended to improve appointment attendance among adults attending outpatient clinics. “Studies about attendance” is too broad. Does an inpatient discharge reminder count? What if a study includes both adults and children? Are simulated appointment choices eligible?
Before screening, write a short decision table.
| Dimension | Proposed rule for this example | Decision it resolves |
|---|---|---|
| Population | Adults receiving outpatient care | Excludes inpatient-only populations |
| Intervention | An implemented strategy intended to influence attendance | Excludes hypothetical preference exercises |
| Comparison | Usual practice or another attendance strategy | Clarifies the effect question |
| Outcome | Define attendance and relevant time points in advance | Prevents switching definitions to suit findings |
| Design | Specify eligible comparative designs | Avoids relying on vague design labels |
| Mixed populations | Include only if adult data can be separated, or use a justified prespecified rule | Handles mixed-age samples |
This is not a universal template. A qualitative evidence synthesis would need different criteria, and an intervention review must consider carefully whether outcome reporting should determine study eligibility. Excluding studies simply because they do not report your preferred result can distort the evidence base.
Distinguish absence from uncertainty
An abstract may not mention a comparison group even when the full text contains one. At title and abstract screening, uncertain records often need to move forward for closer assessment. “Not stated” is not automatically “not eligible”.
Create an uncertainty route: retrieve the full report, check related publications, discuss the case and, where appropriate, contact authors. Set a practical process for unresolved cases and report it.
Pilot the rules before the full screen
Have the screeners independently apply the draft rules to a small, varied set of records. Include obvious inclusions, obvious exclusions and difficult boundary cases.
Compare reasons, not just yes/no decisions. Two people can agree to exclude a paper for different—and incompatible—reasons. Revise ambiguous definitions and repeat the exercise where necessary. The purpose is to improve the rules, not reward a high agreement percentage.
The Cochrane MECIR. Setting eligibility criteria for including studies. illustrates the importance of prespecified, question-linked criteria in Cochrane intervention reviews. Apply the underlying discipline while following standards appropriate to your own review design.
Amend transparently
You may discover a genuine gap in the protocol. Record the change, date, rationale and stage of the review. Reassess earlier records if the change could alter their eligibility. Do not quietly narrow a criterion after seeing results you dislike.
At full-text screening, use clear, mutually understandable exclusion reasons. If a paper fails several criteria, use a consistent hierarchy for the primary recorded reason while preserving fuller notes internally.
Keep the flow diagram honest
Records, reports and studies are not interchangeable. One study can produce multiple reports. Use the PRISMA 2020 flow diagram and templates. appropriate to your review and make its counts reconcile with your screening records.
Before screening thousands of titles, test one difficult paper against your rules. If the decision depends mainly on who happens to read it, the criteria need another edit. Find related guidance in the literature review collection.
Continue your research
Three useful next steps for this topic.
Research notes, once a week
Practical research guidance, useful tools and funding opportunities in your inbox.