Early rehabilitation for critically ill adult patients: protocol for an umbrella review.

BACKGROUND: Advances in critical care medicine have improved short-term survival for intensive care unit (ICU) patients, yet a large proportion of survivors experience lasting physical, cognitive, and psychological impairments collectively termed post-intensive care syndrome (PICS). Early rehabilitation (ER) is a core strategy to reduce PICS burden, but available systematic reviews (SRs) vary widely in intervention protocols, population scope, and methodological rigor. Evidence remains inconsistent regarding ER safety and efficacy in high-risk subgroups, and no high-level evidence synthesis has comprehensively appraised the current evidence base. OBJECTIVE: This protocol outlines an umbrella review (overview of systematic reviews) designed to synthesise evidence on the effectiveness and safety of ER in critically ill adult patients, evaluate the methodological quality of existing SRs, and grade the certainty of evidence for key outcomes. METHODS: Nine electronic databases (CINAHL, Cochrane Library, Embase, PubMed, Web of Science, CNKI, CBM, Wanfang, VIP) will be searched from inception to 30 April 2026, supplemented by manual reference screening and PROSPERO registry searches. Two independent reviewers will conduct title/abstract screening and full-text eligibility assessment, with disagreements resolved via consensus or third-party adjudication. Methodological quality of included SRs will be appraised using the AMSTAR 2 tool, and the certainty of evidence for primary outcomes will be graded using the GRADE framework. A narrative evidence synthesis will be performed to map findings across reviews; quantitative meta-analysis will only be conducted at the primary study level where methodologically appropriate, to avoid bias from overlapping study samples. Subgroup and sensitivity analyses will be undertaken to explore sources of heterogeneity. ANTICIPATED CONTRIBUTION: This umbrella review will provide a rigorous, high-level synthesis of the current ER evidence base. If evidence certainty is sufficient, findings may inform evidence-based clinical decision-making, support standardised protocol development, and identify priority evidence gaps for future primary research. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420250651183, Registration No. CRD420250651183.