Introduction National guidelines for the management of upper gastrointestinal (GI) bleeding exist and are based on conclusive evidence for effective clinical practice. A mortality rate in acute admissions of 7% was reported in a national audit of upper GI bleeding. This is an area of high volume, high risk and high cost where improvements can be made.
Methods Three prospective audits of all acute admissions with upper GI bleed were undertaken for 4 week periods in 2009 (Audit 1), 2011 (Audit 2) and 2013 (Audit 3). After Audit 1, a new GI bleed proforma was introduced,a rolling,targeted educational programme for Accident and Emergency (AandE) and Medical Admissions Unit (AMAU) trainees was started,mandatory fields for risk scoring were included in the electronic requests and additional evening inpatient endoscopy lists were started. After Audit 2, Saturday and Sunday inpatient endoscopy lists were introduced and a dedicated endoscopy co-ordinator supervised triaging of patients to appropriate lists.
Results A total of 115 patients were included in the three audits. 88% were admitted through AandE. There were no deaths and no patients underwent surgery in each of the three audit periods. 13% of all patients had lesions at endoscopy requiring therapy (6% band ligation for variceal bleeding, 7% endotherapy for peptic ulcer bleeding). The proportion of patients in whom a risk score was calculated in the 2009, 2011 and 2013 audits improved with each audit period with completion rates of 0%. 39% and 94%, respectively. (P < 0.001 for comparison of 2009 to 2011, and 2011 to 2013). However, the risk scores were inaccurately calculated by the admitting doctors in 46% and 33% of cases in Audit 1 and Audit 2. The improvement in accuracy between the audit periods was not statistically significant (p = 0.64). There was a statistically significant improvement in the time from admission to endoscopy between the audit periods 2009 and 2013 (median 33.5 h (range 15 to 214 h) versus 23.25 h (range 1.5 to 92 h) (p = 0.0017). The proportion of patients having endoscopy within 24 h of admission improved between audit 1 and Audit 3 (23% and 46%, respectively (P = 0.04)).
Conclusion Targeted interventions have been associated with incremental improvements in the quality of care for patients admitted acutely with acute GI bleeding in the last 4 years. Mortality rates have been consistently well below the national average. Further interventions will include targeted education to improve the accuracy of risk stratification of patients admitted with upper GI blood loss and changes to the mechanism of triage to inpatient endoscopy lists to improve the time from admission to endoscopy.
References 1 NICE(Clinical guideline 141.) 2012
2 Hearnshaw SA, et al. Gut 2010;59:1022–1029
Disclosure of Interest None Declared.