PURPOSE: Lengthy hospitalization places a burden on patients and healthcare resources. However, the factors affecting the length of hospital stay (LHoS) and length of emergency room stay (LERS) in non-fatal bicycle accidents are currently unclear. We investigated these factors to inform efforts to minimize hospitalization. METHODS: We performed a retrospective analysis of data from non-fatal injured bicyclists admitted to the Emergency and Critical Care Center at Kyoto Medical Center between January 2012 and December 2016. We measured LHoS, LERS, mechanism of injury, head injury prevalence, polytrauma, operations performed, injury severity score (ISS), abbreviated injury scale (AIS) score, maximum AIS score, and trauma and injury severity score probability of survival. We conducted multiple regression analysis to determine predictors of LHoS and LERS. RESULTS: Within the study period, 82 victims met the inclusion and exclusion criteria and were included. Mean age was (46.0 ± 24.7) years. Overall mean LHoS was (16.8 ± 25.2) days, mean LERS was (10.6 ± 14.7) days, median ISS was 9 (interquartile range (IQR): 3-16), median maximum AIS was 3 (IQR: 1-4), and median trauma and injury severity score probability of survival was 98.0% (IQR: 95.5%-99.6%). Age, maximum AIS, ISS, and prevalence of surgery were significantly greater in long LHoS and LERS group compared with short LHoS and LERS group (p < 0.05). Performance of surgery independently explained LHoS (p = 0.0003) and ISS independently explained LERS (p = 0.0009). CONCLUSION Surgery was associated with long hospital stays and ISS was associated with long emergency room stays. To improve the quality life of the bicyclists, preventive measures for reducing injury severity or avoiding injuries needing operation are required.