DOI: https://doi.org/10.36347/sasjm.2025.v11i10.005

Pages: 964-969

Introduction: Diagnosing Tuberculosis (TB) in children is a major challenge. This study aims to evaluate the diagnostic performance of the Keith-Edward score compared to the Genexpert in health care facilities (HCF) in Haut-Uélé province, in the northeast of the Democratic Republic of Congo (DRC). Methods: Using a documentary review, we collected sociodemographic, clinical and paraclinical data. retrospectively from 105 children included in our study. We represented the performance of the Keith-Edward score using the Receiver Operating Curve (ROC) curve and also calculated the area under the ROC curve (AUC) as well as all the parameters of this curve. In all cases, the significance threshold was 5% (p-value < 0.5). The rest of the methodological approach is described in our previous article [15]. Results: Keith-Edward score performance analysis showed that the area under the curve was 0.73. Its sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 66.67%, 67.86%, 34.1% and 89.1%, respectively. Conclusion: The Keith-Edward score was found to be less effective and therefore could not be recommended as the sole means of diagnosing tuberculosis in children.