Prospective comparison of econometric, machine learning, and foundation models for forecasting emergency department boarding patients
Lily Poursoltan, Jie Cao, Brian Clay, Beth Trimble, Leah Adrid, Jeffrey Pan, Andrew Chua, John Bell, Christopher A. Longhurst, Kevin Zhu, Karandeep Singh · 2025
Read Prospective comparison of econometric, machine learning, and foundation models for forecasting emergency department boarding patients by Lily Poursoltan, Jie Cao, Brian Clay, Beth Trimble, Leah Adrid, Jeffrey Pan, Andrew Chua, John Bell, Christopher A. Longhurst, Kevin Zhu, Karandeep Singh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13354169.txt.
Description
Lily Poursoltan and co-authors present a 2025 study comparing ways to forecast emergency department boarding. Boarded patients wait in the emergency department for a hospital bed, crowding the unit and worsening care. Good forecasts up to four days ahead let managers act early.
The team used two years of data from two San Diego emergency departments. They tested six approaches: a classic time-series model, a boosted tree model, a large foundation model, two hybrids, and a simple moving average. Over four validation months, the hybrid of the classic model and boosted trees cut error most, sixteen to forty-two percent at one site. The classic model alone was nearly as good and has been deployed in the system's Mission Control center.
Catalog record
| CREATOR | Lily Poursoltan, Jie Cao, Brian Clay, Beth Trimble, Leah Adrid, Jeffrey Pan, Andrew Chua, John Bell, Christopher A. Longhurst, Kevin Zhu, Karandeep Singh |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |