Diagnostic accuracy of a gastroenterology triage score: A sensitivity analysis in a tertiary care center

Background: Efficient triage of patients referred to gastroenterology is essential to prioritize endoscopic evaluation and optimize resource utilization; yet validated quantitative tools remain limited in the local clinical setting.
Objective: To evaluate the diagnostic performance of a continuous triage score in discriminating patients with significant gastrointestinal pathology using receiver operating characteristic (ROC)-based sensitivity analysis in a large referral cohort at King Hussein Medical Center.
Methods: A retrospective analysis was conducted on consecutive patients attending the Gastroenterology Department at King Hussein Medical Center between January 2021 and December 2023. All patients were assigned a numerical triage score prior to investigation. Multiple candidate cutoff values were generated, and for each cutoff, sensitivity, specificity, and corresponding true/false positive and negative counts were calculated to derive the ROC curve and area under the curve (AUC).
Results: The triage score demonstrated high overall discriminative ability, with an AUC of 0.84 (95% CI: 0.81–0.87). At a cutoff of ≥7, sensitivity was 92.3% and specificity was 71.4%. At a higher cutoff of ≥10, sensitivity decreased to 68.5% while specificity increased to 94.2%.
Conclusion: The continuous triage score is a robust tool for stratifying gastroenterology referrals by risk of significant pathology. Sensitivity analysis supports data-driven cutoff selection tailored to institutional capacity and clinical priorities, enabling optimized endoscopic referral triage.