Faculty, Staff and Student Publications

Publication Date

6-1-2025

Journal

OTA International

DOI

10.1097/OI9.0000000000000364

PMID

40061868

PMCID

PMC11888973

PubMedCentral® Posted Date

3-7-2025

PubMedCentral® Full Text Version

Post-print

Abstract

Introduction: Prediction of nonhome discharge after open reduction internal fixation (ORIF) of distal femur fractures may facilitate earlier discharge planning, potentially decreasing costs and improving outcomes. We aim to develop algorithms predicting nonhome discharge and time to discharge after distal femur ORIF and identify features important for model performance.

Methods: This is a retrospective cohort study of adults in the American College of Surgeons National Surgical Quality Improvement Program database who underwent distal femur ORIF between 2010 and 2019. The primary outcome was nonhome discharge, and the secondary outcome was time to nonhome discharge. We developed logistic regression and machine learning models for prediction of nonhome discharge. We developed an ensemble machine learning-driven survival model to predict discharge within 3, 5, and 7 days.

Results: Of the 5330 patients included, 3772 patients were discharged to either a skilled nursing facility or rehabilitation hospital after index ORIF. Of all tested models, the logistic regression algorithm was the best-performing model and well calibrated. The ensemble model predicts discharge within 3, 5, and 7 days with fair discrimination. The following features were the most important for model performance: inpatient status, American Society of Anesthesiology classification, preoperative functional status, wound status, medical comorbidities, age, body mass index, and preoperative laboratory values.

Conclusion: We report a well-calibrated algorithm that accurately predicts nonhome discharge after distal femur ORIF. In addition, we report an ensemble survival algorithm predicting time to nonhome discharge. Accurate preoperative prediction of discharge destination may facilitate earlier discharge, reducing the costs and complications associated with prolonged hospitalization.

Keywords

distal femur fractures, geriatric trauma, discharge destination, machine learning, risk calculator

Published Open-Access

yes

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.