Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models

Abstract Background Central line-associated bloodstream infections (CLABSI) are preventable hospital-acquired infections. Predicting CLABSI helps improve early intervention strategies and enhance patient safety. Aim To develop and temporally evaluate dynamic prediction models for continuous CLABSI r...

Full description

Saved in:
Bibliographic Details
Main Authors: Elena Albu, Shan Gao, Pieter Stijnen, Frank E. Rademakers, Christel Janssens, Veerle Cossey, Yves Debaveye, Laure Wynants, Ben Van Calster
Format: Article
Language:English
Published: BMC 2025-04-01
Series:BMC Infectious Diseases
Subjects:
Online Access:https://doi.org/10.1186/s12879-025-10854-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850172882927222784
author Elena Albu
Shan Gao
Pieter Stijnen
Frank E. Rademakers
Christel Janssens
Veerle Cossey
Yves Debaveye
Laure Wynants
Ben Van Calster
author_facet Elena Albu
Shan Gao
Pieter Stijnen
Frank E. Rademakers
Christel Janssens
Veerle Cossey
Yves Debaveye
Laure Wynants
Ben Van Calster
author_sort Elena Albu
collection DOAJ
description Abstract Background Central line-associated bloodstream infections (CLABSI) are preventable hospital-acquired infections. Predicting CLABSI helps improve early intervention strategies and enhance patient safety. Aim To develop and temporally evaluate dynamic prediction models for continuous CLABSI risk monitoring. Methods Data from hospitalized patients with central catheter(s) admitted to University Hospitals Leuven between 2014 and 2017 were used to develop five dynamic models (a landmark cause-specific model, two random forest models, and two XGBoost models) to predict 7-day CLABSI risk, accounting for competing events (death, discharge, and catheter removal). The models’ predictions were then combined using a superlearner model. All models were temporally evaluated on data from the same hospital from 2018 to 2020 using performance metrics for discrimination, calibration, and clinical utility. Findings Among 61629 catheter episodes in the training set, 1930 (3.1%) resulted in CLABSI, while in the test set of 44544 catheter episodes, 1059 (2.4%) experienced CLABSI. Among individual models, one XGBoost model achieved the highest AUROC of 0.748. Calibration was good for predicted risks up to 5%, while the cause-specific and XGBoost models overestimated higher predicted risks. The superlearner displayed a modest improvement in discrimination (AUROC up to 0.751) and better calibration than the cause-specific and XGBoost models, but worse than the random forest models. The models showed clinical utility to support standard care interventions (at risk thresholds between 0.5–4%), but not to support advanced interventions (at thresholds 15–25%). Conclusion Hospital-wide CLABSI prediction models offer clinical utility based on medium-risk thresholds. Clinical utility at present may be limited as the model performance deteriorated over time.
format Article
id doaj-art-72fab0ae421f48458e56624d98f3571a
institution OA Journals
issn 1471-2334
language English
publishDate 2025-04-01
publisher BMC
record_format Article
series BMC Infectious Diseases
spelling doaj-art-72fab0ae421f48458e56624d98f3571a2025-08-20T02:19:58ZengBMCBMC Infectious Diseases1471-23342025-04-0125111210.1186/s12879-025-10854-1Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction modelsElena Albu0Shan Gao1Pieter Stijnen2Frank E. Rademakers3Christel Janssens4Veerle Cossey5Yves Debaveye6Laure Wynants7Ben Van Calster8Department of Development & Regeneration, KU LeuvenDepartment of Development & Regeneration, KU LeuvenManagement Information Reporting Department, University Hospitals LeuvenFaculty of Medicine, KU LeuvenVascular Access Specialty Team, University Hospitals LeuvenDepartment of Development & Regeneration, KU LeuvenDepartment of Cellular and Molecular Medicine, University Hospitals LeuvenDepartment of Development & Regeneration, KU LeuvenDepartment of Development & Regeneration, KU LeuvenAbstract Background Central line-associated bloodstream infections (CLABSI) are preventable hospital-acquired infections. Predicting CLABSI helps improve early intervention strategies and enhance patient safety. Aim To develop and temporally evaluate dynamic prediction models for continuous CLABSI risk monitoring. Methods Data from hospitalized patients with central catheter(s) admitted to University Hospitals Leuven between 2014 and 2017 were used to develop five dynamic models (a landmark cause-specific model, two random forest models, and two XGBoost models) to predict 7-day CLABSI risk, accounting for competing events (death, discharge, and catheter removal). The models’ predictions were then combined using a superlearner model. All models were temporally evaluated on data from the same hospital from 2018 to 2020 using performance metrics for discrimination, calibration, and clinical utility. Findings Among 61629 catheter episodes in the training set, 1930 (3.1%) resulted in CLABSI, while in the test set of 44544 catheter episodes, 1059 (2.4%) experienced CLABSI. Among individual models, one XGBoost model achieved the highest AUROC of 0.748. Calibration was good for predicted risks up to 5%, while the cause-specific and XGBoost models overestimated higher predicted risks. The superlearner displayed a modest improvement in discrimination (AUROC up to 0.751) and better calibration than the cause-specific and XGBoost models, but worse than the random forest models. The models showed clinical utility to support standard care interventions (at risk thresholds between 0.5–4%), but not to support advanced interventions (at thresholds 15–25%). Conclusion Hospital-wide CLABSI prediction models offer clinical utility based on medium-risk thresholds. Clinical utility at present may be limited as the model performance deteriorated over time.https://doi.org/10.1186/s12879-025-10854-1Bloodstream infectionRisk predictionCLABSIDynamic model
spellingShingle Elena Albu
Shan Gao
Pieter Stijnen
Frank E. Rademakers
Christel Janssens
Veerle Cossey
Yves Debaveye
Laure Wynants
Ben Van Calster
Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
BMC Infectious Diseases
Bloodstream infection
Risk prediction
CLABSI
Dynamic model
title Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
title_full Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
title_fullStr Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
title_full_unstemmed Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
title_short Hospital-wide, dynamic, individualized prediction of central line-associated bloodstream infections-development and temporal evaluation of six prediction models
title_sort hospital wide dynamic individualized prediction of central line associated bloodstream infections development and temporal evaluation of six prediction models
topic Bloodstream infection
Risk prediction
CLABSI
Dynamic model
url https://doi.org/10.1186/s12879-025-10854-1
work_keys_str_mv AT elenaalbu hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT shangao hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT pieterstijnen hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT frankerademakers hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT christeljanssens hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT veerlecossey hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT yvesdebaveye hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT laurewynants hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels
AT benvancalster hospitalwidedynamicindividualizedpredictionofcentrallineassociatedbloodstreaminfectionsdevelopmentandtemporalevaluationofsixpredictionmodels