Document Type : Original Article

Authors

1 MSc Student, Department. of Medical Informatics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran

2 Assistant Professor of Medical Informatics, Department. of Medical Informatics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran

3 Assistant Professor of Computer, Department. of Medical Informatics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran

Abstract

Introduction: Hospital beds and resources are extremely limited for patients. A few factors such as Length of Stay or LOS affected Availability of Hospital Resources. Identifying factors associated with this benchmark index can be valuable in planning to optimize the utilization of the existing resources. Aiming to identify factors associated with length of stay based on admission data, the present study has been conducted.Methods: This survey is a descriptive-retrospective study. We extracted 449,678 patient data from computerize patient records in Emam Reza and Ghaem Hospital in Mashhad, Iran. After data cleaning and preprocessing, univariate analysis was conducted using t-test and one-way ANOVA test. Multiple linear regression was also used to determine the factors associated with LOS. All analyses were conducted with SPSS statistical software version 19.Results: The mean LOS in Emam-Reza and Ghaem hospital were 5.5 and 6.5 days. Age, ward, admission reason, referral status, admission month and day on month, insurance type, residence type, marital status and patient job were associated with length of stay. No significant differences in LOS were found by payment type, admission day on week, gender, doctor and patient Companionship.Conclusion: LOS can be predicted by socio-demographic and clinical factors using data mining models on hospital admission data. The procedure can be a useful tool for planning and optimal resource allocation in hospitals.

Keywords