TY - JOUR T1 - OTH-005 Prospective multi-site audit of acute upper gastrointestinal bleeding delivered across multiple regions by trainee-led networks JF - Gut JO - Gut SP - A270 LP - A270 DO - 10.1136/gutjnl-2018-BSGAbstracts.527 VL - 67 IS - Suppl 1 AU - Richard James Michael Ingram AU - Andrew Alan Baxter AU - Keith Siau AU - Monika Maria Widlak AU - Katherine White AU - Kirsty Nixon Y1 - 2018/06/01 UR - http://gut.bmj.com/content/67/Suppl_1/A270.1.abstract N2 - Introduction Trainee-led networks were developed to support improvements in patient care by delivering audit, quality improvement and research projects across multiple sites. The Midlands was the first UK region to establish gastroenterology trainee research networks: WMRIG in 2015 and the GARNet in 2016. North West trainees established GasTRIN NoW in 2017. Here, we report the first project in the UK to be delivered simultaneously by trainee physicians through multiple networks.Methods We audited the care of patients with acute upper GI bleeding (AUGIB) against national quality standards (NICE QS38 and JAG GRS) and collected data on patient and process outcomes. Patients aged ≥16 years admitted with suspected AUGIB who underwent an inpatient OGD were prospectively identified in a consecutive 30 day window from November 2017. Anonymised data and outcomes to 30 d were collected using common proformas and protocols.Results 471 patients were identified across 30 sites (EM: 157, 8 sites, WM: 193, 12 sites, NW: 121, 10 sites). We report the population characteristics and selected audit and outcome measures for 455/471 (97%) with complete data available. The mean age was 65 years (SD 18.1, range 17–100) and 54% were male. 45% presented on weekdays 7am-7pm, 32% on weekdays outside these hours and 23% at weekends. The median time from presentation to OGD was 22.5 hour (IQR 12.7–43.9 hour), with 54% performed within 24 hour (range at sites 22%–88%;>50%: 18 sites,>75%: 5 sites). Glasgow Blatchford scores were recorded pre-OGD in 47% (range at sites 8%–100%), with median score 7 (IQR 1–12;≥8: 46%,≥12: 26%). Endoscopic findings were: 63% non-variceal, 11% variceal, 8% other, 18% no lesion identified. 116 patients required endotherapy (25%; 80 non-variceal, 34 oesophageal and 4 gastric varices). A rebleed plan was documented in 36% (range at sites 0%–81%). Rebleeding within 8 d occurred in 36 patients (7.9%). 29 died (6.3%) within 30 d of OGD. Median length of stay was 5 d (IQR 3–9 d).Conclusions Collaboration between gastroenterology trainee-led research networks can successfully deliver projects across multiple sites and regions. This audit shows wide variations in practice. Local action plans are still needed to meet JAG quality measures. However, we propose to harness the enthusiasm of trainees to drive improvements in the quality of care received by patients with GI bleeding – regionally, through the BSG national QI agenda, and in partnership with patients via Core. We encourage our peers to establish their own trainee-led research networks. ER -