Objective. Extended length of stay (eLOS) has been contributing to the reduction of hospital resources and patient satisfaction in inpatient psychiatry. With the mean LOS in psychiatry being almost twice the length of other medical conditions, our study aims to identify factors that are associated with eLOS in order to make quality improvements. Methods. This was a retrospective, observational study. A total of 17,257 patients admitted into the inpatient psychiatry unit at Queen’s Medical Center between January 2019 and December 2023 in Honolulu, Hawaii were included in this analysis. Descriptive analysis was used to identify associations of eLOS between various clinical and demographic variables. Results. The strongest sociodemographic predictors of LOS came from: sex (p=0.005), age (p=<0.0001), employment (p=<0.0001), race (p=<0.0001), marital status (p=0.03), homelessness (p=0.006), and having access to Medicaid and/or Medicare (p=<0.0001). The following clinical predictors revealed a strong relationship with LOS: establishment of a PCP (p=<0.0001), diagnosis of intellectual disability (p=0.002), low CSSRS (p=<0.0001), the non-use of illicit drugs, cannabis, ethanol, and the rising number of prescribed psychotherapeutics per patient (p=<0.0001). Conclusion. Identifying strong predictors of eLOS in the United States still remains under-investigated. Enhancing our understanding about factors associated with eLOS will lead to improved treatment planning and resource allocation for inpatient psychiatric hospitalization.