# A recorded literature search: peer-review duration

The previous guide included a real search about peer-review duration. Its counts remain useful as an example of an auditable run: 604 records retrieved across 15 queries, six set aside for missing identifiers, 507 after deduplication, 459 excluded, 11 referred for a decision, and 37 included. Those are historical results, not a heat-and-birthweight search or a new run.

That run's identifier-only rule needs the broader treatment above: an absent identifier should trigger a manual-verification route, not automatic exclusion. Twenty-five included records had retrieved abstracts; twelve had only titles and needed further reading. The limitation field was empty because the abstracts did not supply the information, not because the studies had no limitations. These distinctions are why the saved evidence matters as much as the final prose.
