Risk prediction models for venous thromboembolism in the intensive care unit: a systematic review protocol assessing model performance, risk of bias, and clinical applicability.
7 affiliations
- 1College of Nursing, Wenzhou Medical University, Wenzhou, Zhejiang, China.
- 2Cixi Biomedical Research Institute, Wenzhou Medical University, Ningbo, Zhejiang, China.
- 3The First School of Medicine, School of Information and Engineering, Wenzhou Medical University, Wenzhou, Zhejiang, China.
- 4Ward 271, Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China.
- 5Ward J705, Department of Critical Care Medicine, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China.
- 6Nursing Department, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China. [email protected].
- 7Ward 271, Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China. [email protected].
Abstract
Background
Venous thromboembolism (VTE) is a prevalent and serious complication in intensive care unit (ICU) patients, often leading to long-term issues, increased healthcare costs, and disrupted care. Given its preventable nature, evaluating the risk of VTE in ICU patients and identifying high-risk groups is crucial. Several VTE risk prediction models for ICU patients have been developed, but their performance, risk of bias, and clinical applicability remain unclear. Therefore, this systematic review protocol aims to evaluate all eligible multivariable prediction models and compare their effectiveness in predicting the risk of VTE in ICU patients, rather than studies that focus on a single risk factor or univariate risk scores.Methods
A literature search will be conducted across multiple databases, including PubMed, Cochrane Library, Web of Science, CINAHL, Embase, CNKI, Wanfang Database, VIP, and Sinomed, from inception to April 2025. The language of publications will not be restricted. The search aims to identify studies reporting the development and/or external validation of original multivariable prediction models, including both traditional and machine learning models, to evaluate emerging methods for predicting VTE risk in ICU patients. Two reviewers will independently perform literature screening, data extraction, and assessment of quality and risk of bias, with discrepancies resolved through consensus or adjudication by a third reviewer. Risk of bias and certainty of evidence will be assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. If at least five external validation studies are available, a random-effects meta-analysis will be employed to evaluate model performance across these studies. Specifically, the Hartung-Knapp-Sidik-Jonkman method will be used to pool concordance (c)-statistics and observed-to-expected (O:E) ratios, providing a comprehensive assessment of each model's performance. Subgroup analyses and meta-regression analyses will be performed to explore potential sources of heterogeneity where feasible.Discussion
This study aims to address a significant research gap by comprehensively evaluating existing VTE risk prediction models in ICU patients. The findings will inform clinical decision-making, enabling healthcare professionals to select appropriate risk prediction models and develop individualized VTE prevention strategies. Anticipated challenges in this review include substantial heterogeneity among included studies in terms of population characteristics, outcome definitions, and model methodologies; a limited number of external validation studies for existing prediction models; and potential incomplete or non-transparent reporting of model development and validation processes, which may hinder the assessment of methodological quality and risk of bias.Systematic review registration
PROSPERO CRD420251020721.
Related
- Artificial intelligence and machine learning for prediction of postoperative complications: A systematic review focused on anesthesiology, perioperative risk stratification, and clinical applicability.
- Prediction models for venous thromboembolism in pediatric patients: A protocol for systematic review and critical appraisal
- Comparative Performance of Oxygenation Indices for Mortality Prediction in Patients With Acute Respiratory Distress Syndrome at a Tertiary Care Intensive Care Unit in Bangladesh.
- Pulmonary thromboembolism with prolonged pulmonary infarction during intensive decongestion for severe heart failure : A case report
- Methodological quality and clinical applicability of long COVID guidelines and scientific statements: a systematic review using AGREE II and AGREE-REX
BibTeX
@article{zhu2026risk,
title = {{Risk prediction models for venous thromboembolism in the intensive care unit: a systematic review protocol assessing model performance, risk of bias, and clinical applicability.}},
author = {Furong Zhu and Jiahong Chen and Wenyu Tong and Peng Qiu and Jingjing Mu and Xiaoqun Xu and Qixia Zhang},
journal = {Systematic reviews},
year = {2026},
month = jul,
doi = {10.1186/s13643-026-03259-x},
url = {https://europepmc.org/article/MED/42902437},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Risk prediction models for venous thromboembolism in the intensive care unit: a systematic review protocol assessing model performance, risk of bias, and clinical applicability.
AU - Furong Zhu
AU - Jiahong Chen
AU - Wenyu Tong
AU - Peng Qiu
AU - Jingjing Mu
AU - Xiaoqun Xu
AU - Qixia Zhang
PY - 2026
DA - 2026/07/01
JO - Systematic reviews
DO - 10.1186/s13643-026-03259-x
UR - https://europepmc.org/article/MED/42902437
AB - Background Venous thromboembolism (VTE) is a prevalent and serious complication in intensive care unit (ICU) patients, often leading to long-term issues, increased healthcare costs, and disrupted care. Given its preventable nature, evaluating the risk of VTE in ICU patients and identifying high-risk groups is crucial. Several VTE risk prediction models for ICU patients have been developed, but their performance, risk of bias, and clinical applicability remain unclear. Therefore, this systematic review protocol aims to evaluate all eligible multivariable prediction models and compare their effectiveness in predicting the risk of VTE in ICU patients, rather than studies that focus on a single risk factor or univariate risk scores.Methods A literature search will be conducted across multiple databases, including PubMed, Cochrane Library, Web of Science, CINAHL, Embase, CNKI, Wanfang Database, VIP, and Sinomed, from inception to April 2025. The language of publications will not be restricted. The search aims to identify studies reporting the development and/or external validation of original multivariable prediction models, including both traditional and machine learning models, to evaluate emerging methods for predicting VTE risk in ICU patients. Two reviewers will independently perform literature screening, data extraction, and assessment of quality and risk of bias, with discrepancies resolved through consensus or adjudication by a third reviewer. Risk of bias and certainty of evidence will be assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. If at least five external validation studies are available, a random-effects meta-analysis will be employed to evaluate model performance across these studies. Specifically, the Hartung-Knapp-Sidik-Jonkman method will be used to pool concordance (c)-statistics and observed-to-expected (O:E) ratios, providing a comprehensive assessment of each model's performance. Subgroup analyses and meta-regression analyses will be performed to explore potential sources of heterogeneity where feasible.Discussion This study aims to address a significant research gap by comprehensively evaluating existing VTE risk prediction models in ICU patients. The findings will inform clinical decision-making, enabling healthcare professionals to select appropriate risk prediction models and develop individualized VTE prevention strategies. Anticipated challenges in this review include substantial heterogeneity among included studies in terms of population characteristics, outcome definitions, and model methodologies; a limited number of external validation studies for existing prediction models; and potential incomplete or non-transparent reporting of model development and validation processes, which may hinder the assessment of methodological quality and risk of bias.Systematic review registration PROSPERO CRD420251020721.
SN - 2046-4053
ER -