BACKGROUND/AIMS: Endoscopic hemoclip application is an effective and safe method of endoscopic hemostasis. We conducted a multicenter retrospective study on hemoclip and hemoclip combination therapy based on prospective cohort database in terms of hemostatic efficacy not in clinical trial but in real clinical practice. METHODS: Data on endoscopic hemostasis for non-variceal upper gastrointestinal bleeding (NVUGIB) were prospectively collected from February 2011 to December 2013. Among 1,584 patients with NVUGIB, 186 patients treated with hemoclip were enrolled in this study. Subjects were divided into three groups: Group 1 (n=62), hemoclipping only; group 2 (n=88), hemoclipping plus epinephrine injection; and group 3 (n=36), hemocliping and epinephrine injection plus other endoscopic hemostatic modalities. Primary outcomes included rebleeding, other therapeutic management, hospitalization period, fasting period and mortality. Secondary outcomes were bleeding associated mortality and overall mortality. RESULTS: Active bleeding and peptic ulcer bleeding were more common in group 3 than in group 1 and in group 2 (p<0.001). However, primary outcomes (rebleeding, other management, morbidity, hospitalization period, fasting period and mortality) and secondary outcomes (bleeding associated mortality and total mortality) were not different among groups. CONCLUSIONS: Combination therapy of epinephrine injection and other modalities with hemoclips did not show advantage over hemoclipping alone in this prospective cohort study. However, there is a tendency to perform combination therapy in active bleeding which resulted in equivalent hemostatic success rate, and this reflects the role of combination therapy in clinical practice.