Calibrate to Operate : Aligning Patient-Level Predictions with System-Wide Forecasts
We explore the challenges of re-using machine learning models originally developed for patient-level predictions to forecast system-wide quantities in hospitals. As a running example, we examine the use of patient-level predictions of length of stay to predict short-term occupancy in an emergency department. We first provide analytical evidence for the potential consequences of the misalignment between the patient-level and system-level prediction tasks: the emergence of bias and the failure to detect non-stationarity. Using patient-level data from an emergency department, we demonstrate the existence of such challenges in our data. We show that the simple calibration procedures we propose produce unbiased and more accurate system-wide predictions. We also show that dynamic re-training of the calibration over time further improves the accuracy in our non-stationary setting. Overall, we provide a practical roadmap for hospitals considering making sophisticated patient-level predictions to improve system operations
Year of publication: |
[2023]
|
---|---|
Authors: | Kim, Song-Hee ; Overman, William ; Pauphilet, Jean ; Cha, Won Chul |
Publisher: |
[S.l.] : SSRN |
Saved in:
freely available
Extent: | 1 Online-Ressource (32 p) |
---|---|
Type of publication: | Book / Working Paper |
Language: | English |
Notes: | Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments July 27, 2023 erstellt |
Other identifiers: | 10.2139/ssrn.4523116 [DOI] |
Source: | ECONIS - Online Catalogue of the ZBW |
Persistent link: https://www.econbiz.de/10014359767
Saved in favorites
Similar items by person
-
Kim, Song-hee, (2015)
-
A data-driven model of an appointment-generated arrival process at an outpatient clinic
Kim, Song-Hee, (2018)
-
The effect of active participation on performance and understanding on the Iowa gambling task
Vanhille, Sean, (2018)
- More ...