Abstract
Systematic reviews are foundational to evidence-based practice, but the reproducibility and auditability of study identification and selection remain persistent challenges. Existing reporting standards, such as Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 and PRISMA-Search extension, have improved transparency, yet important gaps remain in documenting search strategies, retrieved records, screening decisions and supplementary identification methods. In this article, we discuss key barriers to reproducibility, including incomplete or inconsistent search documentation, limitations in date filter precision, restricted access to databases and platforms, non-repeatable supplementary search methods and variability in applying eligibility criteria. We suggest a minimum documentation set specifying the search histories, source-specific exports, record-management information, screening decisions and artificial intelligence (AI)-related records that review teams should preserve and share. We also recommend information-specialist involvement and carefully documented human-in-the-loop AI use. These strategies may not guarantee perfect reproducibility, but they can improve auditability, reduce research waste and strengthen trust in systematic reviews.