59 Streamlining Oesophago-Gastric Cancer Referrals: Clinical Dysphagia Score Guides Safe GP Triage for Dysphagia Referrals

Abstract Aim To assess the real-world safety and performance of the Cancer Dysphagia Score (CDS) when used by general practitioners (GPs) to triage patients with dysphagia for suspected upper gastrointestinal (GI) cancer. Method In this retrospective validation study, 2,425 adults referred for upper GI endoscopy with dysphagia were analysed over 12 months in a single NHS trust. Referrers completed a modified CDS at the point of referral, using a threshold equivalent to ≥5.5 on the original scale to classify patients as high- or low-risk. Cancer diagnoses were confirmed histologically. Diagnostic accuracy metrics were calculated to evaluate the CDS under routine clinical conditions. Results Among 2,425 patients (median age 64), 1,670 were classified as high risk and 755 as low risk. Seventy-eight cancers were identified (3.2%), 76 of which were in the high-risk group (4.6%). Only two cancers (0.3%) were found in the low-risk group, both oesophageal squamous cell carcinoma. The adjusted CDS showed strong performance: sensitivity 97.44%, negative predictive value 99.74%, and relative risk 17.18 (95% CI 4.23–69.76). These metrics are sufficient to re-route low-risk patients away from urgent pathways without compromising safety. Conclusions When embedded in primary care referral workflows, the CDS reliably separates high- from low-risk dysphagia patients. Its high sensitivity and low false-negative rate make it a powerful triage tool. In the face of rising endoscopy demand, GP-led triage using the CDS offers a practical and safe solution to preserve urgent capacity for those at highest risk of malignancy.