A team from Glasgow Royal Infirmary tested the feasibility of using the Glasgow-Blatchford bleeding score, which uses clinical and laboratory test results as a risk scoring system – a score of zero identifies low-risk patients who might be suitable for outpatient management.
A group of 676 patients with stomach bleeds were assessed using the bleeding score, and another method called the Rockall score, which assessed patients at admission pre-endoscopy and post-endoscopy.
The team found that the Glasgow-Blatchford score identified 16% of patients as scoring zero. For prediction of need for intervention or death, it was better than Rockall scores.
When the Glasgow-Blatchford score was introduced into clinical practice in two of these hospitals, the proportion of individuals with stomach bleeds admitted to hospital also fell from 96% to 71%.
The study is published online in the Lancet.