A machine learning model for predicting hospitalist consult appropriateness among elective joint arthroplasty patients: A single‐center experience with hospitalist comanagement screening

INTRODUCTION: Hospitalist comanagement improves outcomes for total joint arthroplasty (TJA) patients, but is a limited resource, and data on which patients are most likely to benefit are sparse. METHODS: A total of 400 unique TJA patients and their post-operative trajectories were analyzed retrospectively to determine which patients warranted comanagement, defined by hospitalists making meaningful contributions to post-operative care. From this data, a random forest machine learning (ML) model was developed to predict hospitalist consult appropriateness. The performance of this model was compared to our current approach to screening patients for hospitalist comanagement, which relies on physician discretion and a set of criteria known as the risk stratification tool (RST). RESULTS: A total of 303 of the 400 TJA patients received comanagement, but only 134 were ultimately considered to have warranted comanagement. Our current approach to screening achieved a sensitivity of 0.918 and a specificity of 0.378. Among the 400 cases, only one missed opportunity for consult was identified. The ML model achieved a sensitivity of 0.970, a specificity of 0.463, and an area under the receiver operating characteristic curve of 0.783, suggesting good discriminatory capacity. Several features, including functional status, albumin, medical optimization, and cardiac comorbidities, were considered important for model prediction. DISCUSSION: The random forest model performed comparably to our manual physician discretion-led approach to screening for comanagement. ML approaches have the potential to often facilitate time-intensive, manual identification of patients who may warrant closer post-operative management. Furthermore, patients deemed low risk can proceed safely without comanagement per both ML and physician screening approaches.

Authors

Institutions

Publication Details

Journal
Journal of Hospital Medicine
Published
2026-11-17
DOI
https://doi.org/10.1002/jhm.70554
Primary Topic
Total Knee Arthroplasty Outcomes
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A machine learning model for predicting hospitalist consult appropriateness among elective joint arthroplasty patients: A single‐center experience with hospitalist comanagement screening

Daniel Montes, Michele M. Fang, Emmanuel Gibon, Jeffrey Huang et al.
Journal of Hospital Medicine
Total Knee Arthroplasty Outcomes
article

A machine learning model for predicting hospitalist consult appropriateness among elective joint arthroplasty patients: A single‐center experience with hospitalist comanagement screening

Daniel Montes, Michele M. Fang, Emmanuel Gibon, Jeffrey Huang, Laura Kosseim
article en

Abstract

INTRODUCTION: Hospitalist comanagement improves outcomes for total joint arthroplasty (TJA) patients, but is a limited resource, and data on which patients are most likely to benefit are sparse. METHODS: A total of 400 unique TJA patients and their post-operative trajectories were analyzed retrospectively to determine which patients warranted comanagement, defined by hospitalists making meaningful contributions to post-operative care. From this data, a random forest machine learning (ML) model was developed to predict hospitalist consult appropriateness. The performance of this model was compared to our current approach to screening patients for hospitalist comanagement, which relies on physician discretion and a set of criteria known as the risk stratification tool (RST). RESULTS: A total of 303 of the 400 TJA patients received comanagement, but only 134 were ultimately considered to have warranted comanagement. Our current approach to screening achieved a sensitivity of 0.918 and a specificity of 0.378. Among the 400 cases, only one missed opportunity for consult was identified. The ML model achieved a sensitivity of 0.970, a specificity of 0.463, and an area under the receiver operating characteristic curve of 0.783, suggesting good discriminatory capacity. Several features, including functional status, albumin, medical optimization, and cardiac comorbidities, were considered important for model prediction. DISCUSSION: The random forest model performed comparably to our manual physician discretion-led approach to screening for comanagement. ML approaches have the potential to often facilitate time-intensive, manual identification of patients who may warrant closer post-operative management. Furthermore, patients deemed low risk can proceed safely without comanagement per both ML and physician screening approaches.

Journal of Hospital Medicine
Hospital of the University of Pennsylvania (US), University of Pennsylvania (US)
Total Knee Arthroplasty Outcomes
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon