How do I document a search audit when a co-author questions my screening decisions?

Rebuild the search with its dates, sources, and record counts, then list every excluded record with the reason it was excluded so each screening decision is traceable.

A search audit is a written record that lets a co-author follow the path from your search strategy to the final included set. Cochrane guidance covers planning searches, selecting sources, developing strategies, searching reference lists, documenting searches, and managing records. PRISMA 2020 supplies a checklist and flow diagrams for reporting methods and the studies found and included. Use both as the structure for your appendix.

Build the audit appendix

1. Record the search strategy for each source: the database or index, the date the search ran, the full query as executed, and any filters applied. PubMed supports field tags, phrase searching, filters, Advanced Search, and saved searches; Europe PMC supports keyword and fielded searches across literature, grants, preprints, citations, and references. Save or export the executed query so the appendix shows what actually ran.
2. Report the record counts at each stage: records retrieved per source, duplicates removed, records screened at title and abstract, full texts assessed, and studies included. PRISMA 2020 flow diagrams are built for exactly this sequence, so the counts line up with the diagram your co-author already expects.
3. Map each excluded record to one exclusion reason. Give the record identifier, the screening stage where it was excluded, and the specific criterion it failed, such as wrong population, wrong outcome, or wrong study design. One reason per record keeps the count reconcilable.
4. Note any supplementary searching, such as reference-list checks, and the date each was done, since Cochrane guidance treats these as part of the documented search.

Answer the challenge in the revision window

When a co-author disputes a decision, point to the record's row in the appendix: the criterion, the stage, and the evidence from the record that triggered the exclusion. If the row is missing or the reason is ambiguous, re-screen that record against the stated criteria and update the appendix and the flow-diagram count together. Keep the appendix version-dated so the revision history shows what changed and when.

A short methods paragraph in the paper can cite the appendix, and the appendix carries the record-level detail. That split keeps the main text readable while giving your co-author the trace they asked for.