Background and Aims: The aim of this study was to investigate the influence of shock index, and hematocrit levels on the course of disease and mortality in patients with gastrointestinal bleeding. Material and Methods: This study was conducted as a retrospective analysis of laboratory results and medical records of patients referred to the emergency room from January 1, 2013, to September 30, 2015, who presented with gastrointestinal bleeding symptoms and diagnosed by endoscopic methods. Vital signs on emergency room referral, results of endoscopic interventions and initial blood tests, and the number of hospitalizations and discharges were noted after chart reviews. A total of 709 patients were identified, of whom 266 were found to be eligible. A total of 150 patients whose medical records were accessible were included in this study after the chart surveys. Results: Of the study patients, 94 were males, and patients aged >65 years constituted half of the study population. The most frequent complaints were hematemesis and black stool. The mortality rate among the patients was calculated as 25%. After evaluating the patients’ initial vital signs, we found that there was no effect of shock index on mortality (p > 0.05). The initial blood sample results showed that the hematocrit levels were inadequate in predicting mortality (p > 0.05). Conclusion: Predicting mortality using only the shock index and hematocrit levels was found to be inadequate; therefore, these parameters should be supported by additional data.