ORAHS2024
Abstract Submission

253. Patients who leave the Emergency Department without being seen: a dynamic approach based on survival regression

Contributed abstract in session HC-5: Emergency Department /2, stream Regular talks.

Thursday, 14:00-15:30
Room: Room S6

Authors (first author is the speaker)

1. Davide Duma
Dipartimento di Matematica "Felice Casorati", Università degli Studi di Pavia
2. Roberto Aringhieri
Dipartimento di Informatica, Università degli Studi di Torino
3. Vittorio Meini
Dipartimento di Matematica "Felice Casorati", Università degli Studi di Pavia

Abstract

The rate of patients who Leave Without Being Seen (LWBS) is an important performance indicator for Emergency Departments (EDs), reflecting operational efficiency. High LWBS rates are common in situations of overcrowding and signal critical issues in ED management. This leads to a risk for patients who did not receive the necessary medical attention, since their health condition could worsen due to limited or delayed access to care. While descriptive analytics are commonly employed to address this phenomenon, little attention is given to predictive and prescriptive analytics. Previous Machine Learning studies proposed predictive methods capable of identifying categories of ED patients at risk of abandonment, without capturing their behavior over time. At the same time, Operation Research studies focus on other metrics, such as the Door-To-Doctor Time (DTDT) and the Length of Stay.

This talk introduces a dynamic physician-patient assignment approach informed by predictive models. Survival analysis on censored data is conducted to learn patients' behavior according to their characteristics as the waiting time increases. In addition to enhancing decision-making in the patient admission process, survival regression models allow the generalization of scenarios with an increased DTDT with respect to historical records. A computational analysis assesses the effectiveness of the proposed approach, investigating the trade-off between the DTDT and the LWBS rate.

Keywords

Status: accepted

Back to the list of papers