Oakland-Jairath Score Validation

Oakland and Jairath developed a clinical prediction rule for safe discharge among patients with LGIB using data from their UK National Audit. They defined safe discharge as the absence of rebleeding, blood transfusion, need for endoscopic/radiologic/surgical intervention for hemostasis, readmission with LGIB, and death and developed a seven variable risk score: age, sex, previous history of LGIB admission, presence of blood on digital rectal exam, heart rate, systolic blood pressure, and hemoglobin. Using a cut-off score ≤8, they reported excellent discrimination (AUC 0.84, 95% CI 0.82-0.86), good calibration, and a 95% probability of safe discharge.