Evaluating the Accuracy of the Current Tuberculosis Scoring System in Screening Suspected Cases: A Quality Improvement Project

Abstract

Background Early identification and isolation of suspected Tuberculosis (TB) cases are crucial for reducing the risk of transmission to healthcare workers and other patients. The current TB scoring system, with a cutoff point of 4, has been widely used but lacks specificity. This quality improvement project aims to evaluate the accuracy of the current TB scoring system in screening suspected cases and explore the potential for improving the system's performance. Methods A retrospective analysis of 309 male patients presenting with respiratory symptoms was conducted at Hazm Mebaireek General Hospital. The sensitivity and specificity of various TB scoring cutoff points were assessed using a Receiver Operating Characteristic (ROC) curve. Additionally, the potential benefits of incorporating symptom duration into the scoring system were explored. Results The current TB scoring system, with a cutoff point of 4, demonstrated a sensitivity of 83% and a specificity of 53%. By increasing the cutoff point to 6, the sensitivity decreased to 79%, while the specificity increased to 67%. Incorporating symptom duration into the scoring system was identified as a potential factor for further enhancing its diagnostic performance. Conclusion The current TB scoring system can be improved by adjusting the cutoff point to achieve a better balance between sensitivity and specificity. A dynamic scoring approach that considers symptom duration may provide additional benefits in terms of diagnostic accuracy. Further research is needed to validate these findings and explore other factors that could refine the scoring system, ultimately enhancing its effectiveness in screening and managing TB cases.