Inclusion & Exclusion Criteria

Why Criteria Matter

Explicit inclusion and exclusion criteria ensure your systematic review is reproducible and minimizes selection bias. These criteria guide screening decisions and should be defined before searching the literature.

Types of Criteria

Study Design

Specify which study types will be included based on your research question.

  • Inclusion: 'Randomized controlled trials, quasi-experimental studies'
  • Exclusion: 'Case reports, editorials, conference abstracts'
  • Consider hierarchy of evidence for your question type

Population Characteristics

Define specific participant characteristics beyond your PICO/PCC population.

  • Age ranges: 'Adults 18-65 years'
  • Exclusion: 'Pregnant women, institutionalized populations'
  • Disease severity or stage: 'Early-stage disease only'
  • Comorbidities: 'Exclude patients with severe cardiac disease'

Intervention Details

Specify acceptable variations in interventions, doses, durations, or formats.

  • Inclusion: 'Minimum intervention duration of 4 weeks'
  • Exclusion: 'Combined interventions where effects cannot be isolated'
  • Delivery format: 'Individual or group therapy (exclude self-help only)'

Publication Characteristics

  • Language: 'English language publications only' or 'Any language with translation available'
  • Publication status: 'Peer-reviewed journals only' or 'Include gray literature'
  • Date range: 'Published between 2010-2025'
  • Geographic restrictions if relevant to the question

Writing Effective Criteria

Be Specific and Operational

Criteria should be specific enough that two reviewers would make the same decision when applying them.

  • Weak: 'Adults with diabetes'
  • Strong: 'Adults 18+ years with diagnosed type 2 diabetes mellitus (ICD-10: E11)'
  • Weak: 'Recent studies'
  • Strong: 'Studies published between January 2015 and December 2024'

Avoid Redundancy

Don't repeat information already captured in your PICO/PCC framework. Criteria should add specificity, not duplicate existing definitions.

Consider Gray Areas

Think through edge cases and decide in advance how they'll be handled:

  • Mixed populations: 'Include if >80% meet population criteria'
  • Multiple interventions: 'Include if intervention of interest reported separately'
  • Incomplete reporting: 'Attempt to contact authors for missing data'
  • Ongoing studies: 'Exclude if results not yet published'

Common Inclusion/Exclusion Examples

Example: Clinical Trial Review

Inclusion:

  • Randomized controlled trials
  • Adults 18+ with diagnosed major depressive disorder
  • Cognitive behavioral therapy (individual or group)
  • Minimum 6-week intervention duration
  • Depression symptom outcomes using validated scales
  • Published in peer-reviewed journals 2010-2024

Exclusion:

  • Non-randomized studies
  • Participants with bipolar disorder or psychotic features
  • Combined interventions (CBT + medication) without separate reporting
  • Conference abstracts or unpublished data
  • Studies with >30% attrition without intention-to-treat analysis

Applying Criteria During Screening

During the Abstract and Fulltext phases, each study is evaluated against your criteria. Scholara's AI checks every inclusion and exclusion criterion individually and produces an Include / Exclude / Uncertain decision with the reasoning behind it. You can override any decision from the screening drawer or by asking the assistant in chat — the override is then sent back to the AI on your next message so subsequent decisions take it into account. Exclusion reasons feed automatically into your PRISMA flow diagram.

Keep criteria focused on factors that genuinely affect the validity or applicability of your review. Overly restrictive criteria can lead to empty reviews, while overly broad criteria reduce relevance and increase heterogeneity.