Introduction: Emergency department (ED) mortality is a critical issue that has a major impact on individuals, society, and the entire health system and is one of the indicators of emergency department service performance. This research aims to provide input to hospital management regarding the impact of non-medical factors (number of visits, length of stay/LOS, and ED crowding) on patient mortality, related to improvement policy-making in ED patient services. Method: A retrospective study on 130 ED shifts from medical record data during 2022-2023. The independent variables were the number of visits, LOS, ED crowding measured by CEDOCS, and ED mortality as the dependent variable. Results: There were deaths in 38 out of 130 shifts in the ED, with a total mortality of 1.6%. Statistical tests showed a significant effect between crowding and ED mortality (p = 0.01), a non-significant effect between visits and ED mortality (p = 0.76), and a non-significant effect between LOS and ED mortality (p = 0.62). The results of the regression analysis were able to explain the data (p = 0.91; Hosmer-Lameshow test). The area under the curve (AUC) for prognosis in hospital treatment is 0.641 with a 95% confidence interval of 0.539-0.744 (medium category). Conclusion: The significant effect between crowding and patient mortality in the ED indicates that strategic management must be carried out to reduce ED crowding.