Abstract
Objective: To examine the factors which identify patients with gastrointestinal bleeding (GIB) who benefit from intensive care unit (ICU) admission and the impact of ICU triage decisions on cost and outcome. Design: Retrospective cohort study. Setting: Tertiary care university hospital, medical-surgical ICU. Patients: Adult patients with GIB. Interventions: None Measurements and main results: There were 124 patients with a diagnosis of gastrointestinal bleeding admitted to the hospital. Patients were grouped based on presence of the following risk factors: active bleeding, haemodynamic instability, and massive transfusion requirement (greater than five units of RBC) and admission to the ICU. Group 1 included patients who had one or no risk factor and were not admitted to the ICU. Group 2 included 30 patients who had one or no risk factor and were admitted to the ICU. There was no significant difference in mortality between these two groups of low risk patients. The average total hospital cost per patient for Group 2, who experienced ICU admission, was 43% ($5466) greater than group 1. Groups 3 and 4 included 46 patients with two or more risk factors. Eighty five percent of deaths in this study occurred among these patients. The total cost of hospital care for all patients in the study population was 2.9 million dollars, and 76% of this amount was consumed by Groups 3 and 4, which comprised only 37% of the study population. Conclusion: Appropriate triage of GIB patients can result in significant cost savings without any negative effect on clinical outcome.