Simulating wait time in healthcare: Accounting for transition process variability using survival analyses

Wait or queuing time is a principal performance measure for many discrete-event simulation (DES) models in healthcare. However, variation in wait time is often caused by both occupied downstream servers (e.g., beds) and organizational and human transition processes. DES models that attribute wait so...

Full description

Saved in:
Bibliographic Details
Main Authors: Levin, Scott, Garifullin, Maxim
Format: Conference Proceeding
Language:English
Subjects:
Accounting
Boarding
Computer simulation
Health care
Hospitals
Queues
Servers
Survival
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Wait or queuing time is a principal performance measure for many discrete-event simulation (DES) models in healthcare. However, variation in wait time is often caused by both occupied downstream servers (e.g., beds) and organizational and human transition processes. DES models that attribute wait solely to occupied servers, ignoring transition process variability, face challenges in adequate baseline validation. Embedding regression models for survival data in DES to estimate patient wait times is a method capable of integrating the effects of transition processes with queuing. Developing these models as a sub-component is further valuable in understanding the socio-technical system factors that drive prolonged waits. These general methods are exhibited in a DES for a large urban hospital with a primary output of wait time in the emergency department (ED) for transfer to an inpatient bed (boarding time). Simulated boarding time is compared before and after accounting for transition processes using survival analysis.
ISSN:1558-4305
DOI:10.1109/WSC.2015.7408250