Author: James Sallis
Sallis, James, 2025 The Cost of a Bleed – A Retrospective Analysis of Non-Variceal Upper Gastrointestinal Bleeding in Hospital-Inpatients, Flinders University, College of Medicine and Public Health
Terms of Use: This electronic version is (or will be) made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. You may use this material for uses permitted under the Copyright Act 1968. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact copyright@flinders.edu.au with the details.
Despite improvements in medical and surgical management in recent years, non-variceal upper gastrointestinal bleeding (NVUGIB) remains an expensive medical emergency with a high mortality rate. In this clinical landscape, antithrombotics (antiplatelets and anticoagulants) present unique and consequential challenges both prior and following a NVUGIB. These medications are essential for the treatment and prevention of a variety of serious and debilitating cardiovascular conditions. However, these agents undoubtedly increase the risk of recurrent gastrointestinal bleeding.
This thesis focuses on the costs associated with NVUGIB management and the real-world management of antithrombotics following acute NVUGIB, at two hospital networks in South Australia, with a comparison to contemporaneous recommendations in the literature. Variation between real and recommended management may signify an opportunity to reduce the future incidence, clinical severity and financial burden of NVUGIB.
The literature surrounding NVUGIB and antithrombotics indicated that the reinitiation of this therapy following a NVUGIB is largely beneficial, decreasing mortality and thrombosis, despite increasing bleeding risk. Primary prevention, especially in the elderly, should be ceased following a bleed and secondary prevention should be restarted when clinically appropriate. If indication allows, anticoagulation should be swapped to apixaban, given its superior safety profile (especially when compared to rivaroxaban and warfarin). Proton pump inhibitors (PPIs) should be initiated and continued upon discharge – with duration largely dictated by the duration of antithrombotics therapy.
When examining 358 eligible cases of NVUGIB, it was clear that real-world practice differed from literature in several important ways. Only 27 (60%) of primary prevention therapy cases were ceased upon discharge. Secondary prevention was ceased in 16 (13.9%) cases. Despite adequate CHA2DS2-VASc scoring, 19 (15.1%) AF cases had their anticoagulation ceased upon discharge. Of the 62 AF cases that had anticoagulation continue upon discharge, there were 19 (30.1%) not treated with apixaban. PPI therapy was not prescribed in 23 (6.4%) cases where an antithrombotic was prescribed at point of discharge.
Following NVUGIB, the treating team’s evaluation as to the ongoing benefits and risks of antithrombotics is essential. Documentation of these insights in the electronic medical record and discharge summary is essential to guide ongoing treatment and to inform future discussions between the patient and their families and health professionals providing care in the community. In 66.9% of anticoagulant and 73.4%% of antiplatelet cases, there were no such statements made in the patient’s medical discharge summary.
Of the eligible 358 cases, costing analysis was undertaken on 85 cases, with a median total costing of $11,227.01 (IQR $6,434.71 – $19,637.09) per presentation. In those that presented to hospital with a NVUGIB, a strong and positive Pearson Correlation was found between the Total Cost and a patient’s length of stay (0.663, p = <0.001). The estimated HAC costing had a wide variance, with a mean value of $2,099.31.
An analysis of case costing buckets revealed that staff labour, ward medical costs and ward nursing costs were consistently among the highest costing areas of NVUGIB cases. Drivers of cost appeared largely consistent and well portioned between high and low costing groups, as revealed by a K-means cluster analysis. The overall impression from this costing analysis was that unless length of stay can be meaningfully reduced, cost savings will be difficult without preventing the NVUGIB itself.
The reported research findings suggest there is scope to improve real-world patient care and to reduce the costing burden of NVUGIB by altering antithrombotic and PPI prescribing at the point of patient discharge. Communication within medical discharge summaries can be improved. The research methods used could be applied to other real-world settings to inform improvement actions in other local contexts.
Keywords: Anticoagulants, Antithrombotic Agents, Gastrointestinal Hemorrhage, Health Care Economics and Organizations, Hospitalization
Subject: Medical Science thesis
Thesis type: Masters
Completed: 2025
School: College of Medicine and Public Health
Supervisor: Professor Jonathan Karnon