ORAHS2024
Abstract Submission

214. A predictive model for patient length of stay and hospital pathway for specialized respiratory diseases hospital.

Invited abstract in session TB-2: Advances in Health Economics and Healthcare Management, stream Regular talks.

Tuesday, 11:00-12:30
Room: Room S1

Authors (first author is the speaker)

1. Paolo Landa
Département d’opérations et systèmes de décision, Université Laval
2. Marzia Angela Cremona
Opérations et Systèmes de Décision, Université Laval
3. Luca Murazzano
Département d'Opérations et systèmes de décision, Université Laval
4. Jean-Baptiste Gartner
Management, Universite Laval
5. Andre Cote
Management, Universite Laval

Abstract

The analysis of hospital and department performances is a key element for understanding and enhancing the quality of services given by healthcare providers, enabling continuous improvements and benefits for the users, third payers, and community. Over the past decades, there has been increased interest in employing advanced data mining methods such as machine learning to improve hospital performance. The length of stay is often used as a surrogate for other outcomes, and it can define healthcare resource utilization. In this study, we present a predictive model for patient length of stay and hospital pathway definition of four clinical diseases (chronic obstructive pulmonary disease, pneumonia, lung cancer, and pulmonary fibrosis). Data were extracted from the electronic patient record system of the Pulmonary and Cardiology University Institute of Quebec for Respiratory and Cardiovascular diseases for hospital admissions and discharges. A total of 22,194 episodes, which contain 4,714 patient records from January 2018 to December 2022 at one Canadian hospital were analyzed by using parametric and nonparametric statistical methods. A regression analysis was carried out to model the length of stay and hospital pathway definition as a function of several independent variables such as lab tests, diagnostic imaging, admission type, and patient origin. The preliminary results show the possible economic and clinical implications for both patients and hospital management.

Keywords

Status: accepted

Back to the list of papers