Page 32 - The-5th-MCAIT2021-eProceeding

Descriptive Statistics of Length of Stay in Hospital Wards:
A Case Study at HCTM

a*
Syazwan Md Yid , Rosmina Jaafar , Seri Mastura Mustaza
a,b
a
a Dept. Electrical, Electronic & Systems Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia,
Bangi 43600, Malaysia
b Medical Engineering Technology, Universiti Kuala Lumpur British Malaysian Institute, 53100 Gombak, Malaysia
*Email: rosmina@ukm.edu.my

Abstract

Information on hospital length of stay (LOS) of patients in the hospital ward is an important factor for planning and
managing the resource utilization of a hospital. There has been considerable interest in controlling hospital cost and
increasing service efficiency, particularly in surgical units where the resources are severely limited. The main objective of
this paper is to analyze statistically patients’ data for future LOS prediction models. A cross-sectional study was conducted
on patients’ data from 2015 to 2020 requiring intensive care unit (ICU) admissions in Hospital Canselor Tuanku Muhriz
(HCTM), which is a teaching hospital in Malaysia. Factors that determined the LOS at the ICU were also explored by using
multivariate regression analysis. It was found that general intensive care unit (GICU) patients were staying in the hospital
at an average of 6 days across all age. However, patients of younger age (less than 12 years) requires longer ICU stay.
Multivariate regression model shows only the category of the primary team handling the patients is the determinant for LOS.
The diagnosis remarks in the database can be transformed into a code forms to allow better LOS prediction model.

Keywords: Length of stay; statistics; cross-sectional study

1. Introduction

Patient length of stay is one of important performance indicators of a hospital which gives a better
understanding of the resource consumption and patients’ flow through healthcare system (Panchami & Radhika,
2014). Researchers have found that there is a strong correlation between medical cost and length of stay (LOS)
by studying the factors influencing medical cost (Luo, Lian, Feng, Huang, & Zhang, 2017). The productivity of
hospitals drop significantly in two situations: First, if the hospital is in short supply for required resources such
as facilities and manpower. Second, if the hospital is over equipped and the demand is less than the supply. Both
of these situations occur due to significant fluctuations in hospital occupancy, which seriously restricts the
efficient scheduling for resource allocation and management (Azari, Janeja, & Mohseni, 2012).
LOS in the ICU is one of the most important and influential factors in health financial management. Some
studies considered LOS as a surrogate for hospital cost (Zhang & Liu, 2011). Analysis of determinants that can
influence the length of ICU stay is of interest for both medical quality assurance and health economic aspects.
Total LOS at the hospital and at the ICU varied with different care policies, different hospitals, and different
countries (Gruenberg & D. A., 2006). A literature review mentioned that patients stayed on average 3.3 days at
the ICU (Hunter, Johnson, & Coustasse, 2020). Nevertheless, the LOS could be different based on the condition
of the patient and the location of the study.
In this article, we do a statistical analysis for possible determinants for predicting LOS and assess their
suitability for planning resources, identifying unexpectedly long LOS, and benchmarking. This study used the
general intensive care unit (GICU) census database recorded at Hospital Canselor Tuanku Muhriz (HCTM),
Kuala Lumpur, Malaysia. This database is a system that records patients based on age, sex, race, diseases,

E- Proceedings of The 5th International Multi-Conference on Artificial Intelligence Technology (MCAIT 2021) [20]
Artificial Intelligence in the 4th Industrial Revolution