EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Continuous prediction of prolonged intensive care unit stay from longitudinal electronic health record data: development and external validation

Mateen Jangda iD, Christopher Capone, Armando Villegas, Ashwin Sawant iD, Ahmed Shaikh, Bojana Milekic, Eugenia Alleva iD, Evan Leibner iD, Dhruv Patel, Rohit Gupta, Sophia Kaplan, Cappi Lay und 16 weitere

JAMIA Open · 06.11.2026 · 9 (6)

DOI: 10.1093/jamiaopen/ooag216 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JAMIA Open, online erschienen am 13.11.2026. Verfasst von Mateen Jangda und 27 weiteren Autor:innen. Beteiligte Einrichtung: The Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, New York, NY 10029, United States. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: NIDDK NIH HHS, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

To develop and externally validate Prognostic Estimator of RiSk for prolonged ICU STay (PERSIST), a model that continuously updates the probability of an intensive care unit (ICU) stay longer than 7 days during the first week of ICU admission. Materials and Methods We developed PERSIST, an XGBoost model, on 49 255 adult ICU stays from the Mount Sinai Health System (MSHS; 2021-2025) and externally validated it in 72 156 Medical Information Mart for Intensive Care–IV (MIMIC-IV) stays (2008-2019). Sixty routinely available electronic health record (EHR) features were aggregated hourly, with predictions every 4 hours from preceding data, benchmarked against L2-regularized logistic regression. We assessed discrimination, calibration, net benefit, SHapley Additive exPlanations attributions, and subgroup performance.

Results

Prolonged stay occurred in 16.1% of MSHS and 11.1% of MIMIC-IV admissions. PERSIST achieved macroaveraged areas under the receiver operating characteristic curve of 0.80 (95% confidence interval [CI], 0.79-0.81) and 0.78 (0.77-0.78), respectively, compared with 0.75 (0.73-0.76) and 0.74 (0.73-0.74) for logistic regression. Corresponding areas under the precision-recall curve were 0.73 and 0.68. Brier scores were 0.14 and 0.13. Across modeled thresholds, PERSIST provided greater net benefit than treat-all and treat-none strategies. Discrimination was similar across reported age, sex, and racial groups.

Discussion

Prognostic Estimator of RiSk for prolonged ICU STay transported across 2 health systems differing in era, case mix, and outcome prevalence, and outperformed logistic regression under identical conditions. Net benefit was greatest during the early ICU days, when few patients have declared a prolonged trajectory and a risk estimate can still inform planning.

Conclusion

Continuously updated, EHR-based prediction of prolonged ICU stay is feasible and generalizable. Prospective evaluation should determine whether it improves ICU care coordination and capacity planning.

Schlagwörter

Stichwörter der Autor:innen

clinical prediction modelselectronic health recordsintensive care unitslength of staymachine learning

Autor:innen und Institutionen

  • The Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, New York, NY 10029, United States.
  • University of Miami Miller School of Medicine, Miami, FL 33136, United States.
  • The Mount Sinai AI Assurance Lab, Mount Sinai Health System, New York, NY 10029, United States.
  • Institute for Critical Care Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, United States.
  • The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, NY 10029, United States.
  • Division of Hospital Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, United States.

ORCID-Profile: Mateen Jangda, Ashwin Sawant, Eugenia Alleva, Evan Leibner

Förderung

  • NIDDK NIH HHS
  • National Institutes of Health

Zitieren

APA

Jangda, M., Capone, C., Villegas, A., Sawant, A., Shaikh, A., Milekic, B., Alleva, E., Leibner, E., Patel, D., Gupta, R., Kaplan, S., Lay, C., Ahmed, S., Oropello, J., Gidwani, U., Manasia, A., Kohli-Seth, R., Patel, D., Timsina, P., …, & Sakhuja, A. (2026). Continuous prediction of prolonged intensive care unit stay from longitudinal electronic health record data: development and external validation. JAMIA Open, 9(6). https://doi.org/10.1093/jamiaopen/ooag216

Vancouver

Jangda M, Capone C, Villegas A, Sawant A, Shaikh A, Milekic B, et al. Continuous prediction of prolonged intensive care unit stay from longitudinal electronic health record data: development and external validation. JAMIA Open. 2026;9(6). doi:10.1093/jamiaopen/ooag216

BibTeX
@article{Jangda2026,
  author = {Jangda, Mateen and Capone, Christopher and Villegas, Armando and Sawant, Ashwin and Shaikh, Ahmed and Milekic, Bojana and Alleva, Eugenia and Leibner, Evan and Patel, Dhruv and Gupta, Rohit and Kaplan, Sophia and Lay, Cappi and Ahmed, Sanam and Oropello, John and Gidwani, Umesh and Manasia, Anthony and Kohli-Seth, Roopa and Patel, Dhaval and Timsina, Prem and Escalon, Miguel and Vakil, Aditi and Kovatch, Patricia and McGreevy, Jolion and Stump, Lisa and Reich, David and Charney, Alexander W and Nadkarni, Girish N and Sakhuja, Ankit},
  title = {Continuous prediction of prolonged intensive care unit stay from longitudinal electronic health record data: development and external validation},
  journal = {JAMIA Open},
  year = {2026},
  volume = {9},
  number = {6},
  doi = {10.1093/jamiaopen/ooag216}
}
Kostenlos starten →