Recording exclusion reasons during full-text screening ensures transparency in your systematic review process and helps justify which studies you removed and why.
Use a Standardized Exclusion Coding System
Create a predefined list of exclusion criteria before you begin screening. Common categories include wrong population, wrong intervention, wrong outcome, wrong study design, duplicate publication, or insufficient data. Assign each criterion a short code (e.g., "P1" for population mismatch, "D1" for wrong design) so you can quickly tag papers during review.
Document in a Screening Log or Spreadsheet
Most researchers maintain a dedicated spreadsheet with columns for paper ID, title, first author, year, and exclusion reason. When you reject a paper, enter the corresponding code and optionally add a brief note explaining the specific issue. This creates an audit trail that reviewers or journal editors can verify later.
Leverage Reference Management Tools
If you're managing hundreds of papers, tools like Zotero, Mendeley, or Covidence allow you to tag items with custom labels or notes. Some systematic review platforms have built-in exclusion tracking that automatically generates PRISMA flow diagrams. For researchers working with large literature bases, WisPaper's My Library offers a reference manager with AI chat capabilities, letting you organize papers, extract metadata automatically, and maintain notes on why specific studies were excluded—all synced to the cloud for team collaboration.
Track Inter-Reviewer Agreement
If multiple reviewers are screening, record who made each exclusion decision and whether there were disagreements. Calculate Cohen's kappa or percentage agreement to demonstrate reliability. Resolve conflicts through discussion or a third reviewer, and document the final decision.
Report in Your Methods Section
In your published review, summarize exclusion reasons in a table or PRISMA diagram showing how many papers were removed at each stage and for what reasons. This transparency is essential for reproducibility and helps future researchers understand your selection logic.
