Prediction of prolonged length of stay from first 2 days of hospitalization data: a SHAP value-based variable selection method for a simplified model
While it is possible to predict prolonged LOS with a large complex set of variables, most models are difficult to use in clinical practice. SHAP-value-based variable selection allowed a reduction in the number of variables for equivalent predictive performance, making optimum prediction of prolonged LOS easier to implement in routine clinical practice by prioritizing certain predictive factors, allowing preventive measures to be taken for identified patients.
Authors
- Daniel Stoessel (ORCID: https://orcid.org/0000-0002-2140-7390)
- Hadiseh Nowparast Rostami
- Caroline Landelle (ORCID: https://orcid.org/0000-0002-3592-1948)
- Rui Fa (ORCID: https://orcid.org/0000-0003-4588-7971)
- Svetlana Artemova (ORCID: https://orcid.org/0009-0006-8374-8197)
- Ursula von Schenck
- Jean‐Luc Bosson (ORCID: https://orcid.org/0000-0003-0967-6026)
- Pierre-Ephrem Madiot
Institutions
- Centre National de la Recherche Scientifique (FR)
- RELX Group (Netherlands) (NL)
- Centre Hospitalier Universitaire de Grenoble (FR)
- RELX Group (United Kingdom) (GB)
- Université Grenoble Alpes (FR)
Publication Details
- Journal
- International Journal for Quality in Health Care
- Published
- 2025-10-01
- DOI
- https://doi.org/10.1093/intqhc/mzaf116
- Citations
- 1
- Primary Topic
- Sepsis Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 1.35