1Mohammed Audu and 2Usman Muhammad Godowoli
1,2Department of Statistics, Yobe State College of Agricultural Science and Technology, Gujba
*Corresponding author’s Email: jajere306@gmail.com, doi.org/10.55639/607.02010011
ABSTRACT
This study aims to predict the Hospital Length of Stay (LOS) for patients in the Emergency Departments (ED) of General Hospital Damaturu using the Weibull Regression Model. LOS is a critical metric for hospital performance, and accurate prediction can optimize resource allocation, reduce costs, and enhance healthcare efficiency. A retrospective analysis of patients admitted to the ED of General Hospital Damatu from January 2022 to December 2023 was conducted, considering variables such as Age (Chi-Square = 30.8531, df = 1, p < 0.0001), Gender (Chi-Square = 36.7738, df = 1, p < 0.0001), Time of Arrival (Chi-Square = 11.4658, df = 1, p = 0.0007), Previous Emergency Unit (Chi-Square = 42.9153, df = 1, p < 0.0001), Availability of Inpatient Bed (Chi-Square = 19.6701, df = 1, p < 0.0001), and Emergency Department Crowding (Chi-Square = 22.5123, df = 1, p < 0.0001) were found to be statistically significant predictors of length of stay (LOS) while abnormal vital signs, previous medical history, and laboratory tests were not significant. Cox Snell residual was used to investigate whether the model fits the data and the result showed that, it generally support the model fit, with most values close to zero.
Keywords:
Weibull Regression,
Hospital Length of
Stay,
Emergency
Department