Purpose: This study shows the change in emergency room use behavior by homeless patients after implementation of the ‘Seoul Type Citizen Sympathy Emergency Room Project’ in July 2015. Methods: A retrospective study was conducted in a public hospital between January 2014 and December 2014 and January 2016 and December 2016. Homeless patients who visited the emergency room in 2014 and 2016 were compared based on age, gender, mode of insurance, admission, revisit within 48 hours, length of stay (LOS) in the emergency department (ED), total cost, and major diagnostic category. Results: A total of 3,642 homeless patients were enrolled during the study period, of which 1,876 visited in 2014 and 1,766 in 2016. Fewer homeless patients in 2016 revisited within 48 hours (p=0.046). Homeless in 2016 had a shorter ED LOS (p<0.001) and lower total cost (p=0.040). More homeless patients who visited due to alcohol revisited within 48 hours in 2016 (p=0.036). Moreover, these patients did not have a different ED LOS (p=0.060) or total cost (p=0.475). Medicaid homeless patients were less connected compared health insurance by screening, brief intervention, referral to treatment program. Conclusion: Comparison of homeless patients who visited the emergency room in 2014 and 2016 revealed fewer total homeless patients in 2016, as well as less revisits within 48 hours, shorter ED LOS and lower total cost. These findings indicate that the ‘Seoul Type Citizen Sympathy Emergency Room Project’ was effective in the emergency room, but improvements for alcohol and medicaid homeless patients are needed.