Prediction models and risk assessment tools for peri-implantitis: a systematic review with meta-analysis of the Implant Disease Risk Assessment (IDRA) - registered protocol (PROSPERO CRD420261540821) and search strategy

Protocol materials for a systematic review with meta-analysis of the predictive performance of the Implant Disease Risk Assessment (IDRA) and other prediction models and risk assessment tools for peri-implantitis. The review is prospectively registered in PROSPERO (CRD420261540821; registered 20 October 2026). Files: (1) PROSPERO_Registered_CRD420261540821_2026-10-20.pdf - the registered PROSPERO record (version 1.0) as published on 20 October 2026; (2) Search_Strategy_Final_v3_2026-10-17.pdf - full search strategy (version 3.1) reported per PRISMA-S, with exact search strings, the search log of the formal searches run on 17 October 2026 (2,520 records identified; 860 duplicates removed; 1,660 records to screen) and the databases not searched. Planned methods: dual independent screening and data extraction (CHARMS), risk of bias with PROBAST/PROBAST+AI, random-effects meta-analysis of discrimination (logit AUC, REML with Hartung-Knapp-Sidik-Jonkman adjustment) and of IDRA high-risk classification (odds ratio, sensitivity and specificity; bivariate model if feasible), and certainty of evidence with GRADE for prognostic studies. Reporting follows PRISMA 2020, PRISMA-S and TRIPOD-SRMA. Any amendments will be recorded in PROSPERO and as new versions of this record.