<![CDATA[Child stunting remains a critical public health challenge globally, with early detection essential for preventing irreversible developmental consequences. This review uniquely synthesizes predictive and measurement tools for early stunting detection across multiple low- and middle-income countries, addressing the gap in systematic evaluation of their diagnostic performance. Following PRISMA 2020 guidelines, we systematically searched PubMed, ProQuest, Emerald, and Springer databases for studies published between 2019-2024. We used the SPIDER method to define this study's inclusion, and eligible studies included those developing or validating predictive models and diagnostic tools for stunting detection in children aged 0-19 years. The articles included in this assessment use different methodologies conduct a critical appraisal to determine the quality of the articles using the instrument tool from Joana Briggs Institute (JBI). Thirteen studies met the inclusion criteria, identifying four primary tool categories clinical scoring systems, omposite anthropometric indices, diagnostic charts and machine learning models. The MEIRU chart demonstrated superior diagnostic accuracy (sensitivity 97.6%, specificity 96.3%, kappa 0.93) compared to traditional WHO methods. Clinical scores exhibited moderate performance with sensitivity ranging from 61.9% to 90.0%. Common predictors across models included maternal education, birth weight, child age, gender, and socioeconomic status. Predictive models and measurement tools are valuable for the early detection of stunting, enabling timely and targeted interventions to reduce its long-term impact on child health and development. Keywords: Anthropometry, Diagnostic accuracy, Early detection, Prediction models, Risk factors]]>
Copyrights © 2025
