O34 Consensus on upper gastrointestinal endoscopy quality indicators; a modified Delphi process

IntroductionUpper gastrointestinal (UGI) endoscopy lacks established key performance indicators (KPI). We aimed to develop evidence based KPIs to reduce post endoscopy upper gastrointestinal cancer (PEUGIC) and improve quality of endoscopy.MethodsA modified Delphi process was conducted in the form of two consensus workshops, with stakeholder representation from clinicians (including gastroenterologists, UGI surgeons, nurse endoscopists, UGI cancer nurse specialists and trainees), and patients and their carers, moderated by James Lind Alliance facilitators. Potential indicators were identified from three sources: umbrella systematic review; National Cancer Registration and Analysis Service and Hospital Episode Statistics database analysis to identify PEUGIC associations; and differences in endoscopy practices between providers with the highest and lowest PEUGIC rates from the National Endoscopy Database. KPIs were categorised as endoscopy provider related and endoscopist or procedure related and ranked as major and minor based on their importance. Minimum standards were proposed where possible.ResultsFour endoscopy provider related and 12 endoscopist or procedure related KPIs were proposed (table 1).Abstract O34 Table 1 Key performance indicators Standards Endoscopy provider related (major) PEUGIC rate Below 25th centile of national rate Less intense endoscopy lists 7–9 points per list Accreditation by Joint Advisory Group (JAG) in GI endoscopy All providers Dedicated lists by trained endoscopists for surveillance of pre-malignant conditions* >90% Endoscopist or procedure related (major) Examination time Minimum 7 minutes Training in lesion recognition All endoscopists Quality of mucosal view recorded, and mucosal cleaning techniques employed if needed >90% Intravenous sedation Offered to all patients Minimum annual number of UGI endoscopies performed per endoscopist 100 If cancer associated lesion found** recommended biopsies taken and follow up endoscopy organised. >90% Recommended biopsies from pre-malignant conditions* and appropriate surveillance organised.>90% Detection rate of premalignant conditions* and early cancer/dysplasia Not possible to suggest Endoscopist or procedure related (minor) Use of image enhancement techniques for high risk endoscopies or to assess focal lesion >90% Photo-documentation of important anatomical sites >90% Detection rate of ampulla, inlet patch and hiatus hernia.Not possible to suggest Use of artificial intelligence Not