Background <p>The National Early Warning Score 2 (NEWS2) has been adopted as the standard approach for early detection of deterioration in clinical settings in the UK, and is also used in many non-UK settings. Limitations have been identified, including a reliance on ‘normal’ physiological parameters without accounting for individual variation.</p> Objective <p>This review aimed to map how the NEWS2 has been modified to improve its predictive accuracy while placing minimal additional burden on clinical teams.</p> Methods <p>The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA-ScR) and the Population, Intervention, Comparator, Outcome, and Study (PICOS) frameworks were followed to structure the review. Six databases (CINAHL, PubMed, Embase, ScienceDirect, Cochrane Library and Web of Science) were searched for studies which reported the predictive accuracy of a modified version of NEWS2. The references were screened based on keywords using EndNote 21. Title, abstract and full-text screening were performed by 2 reviewers independently in Rayyan. Data was extracted into a pre-established form and synthesised in a descriptive analysis.</p> Results <p>Twelve studies were included from 12,867 references. In 11 cases, modified versions of NEWS2 demonstrated higher predictive accuracy for at least one outcome. Modifications that incorporated demographic variables, trend data and adjustments to the weighting of the score’s components were found to be particularly conducive to enhancing the predictive accuracy of NEWS2.</p> Conclusions <p>Three key modifications to NEWS2—incorporating age, nuanced treatment of FiO₂ data and trend analysis—have the potential to improve predictive accuracy without adding to clinician burden. Future research should validate these modifications and explore their composite impact to enable substantial improvements to the performance of NEWS2.</p>