Scholarlink

Aim <p>Central venous catheters (CVCs) are essential for long-term therapies but carry a high risk of central line-associated bloodstream infections (CLABSIs), which significantly impact patient outcomes and healthcare costs. This study aimed to develop a causal model for CLABSI using expert knowledge to guide future clinical trials and prevention strategies.</p> Methods <p>We constructed a directed acyclic graph (DAG) informed by literature and expert knowledge elicitation. A multidisciplinary team of clinicians, including infectious disease and vascular access experts, participated in interviews and workshops to refine the DAG, resulting in a final model with 30 variables representing CLABSI development.</p> Findings <p>The expert-elicited DAG identified two main pathways, patient-related and CVC-related, each contributing to CLABSI risk. Variables and relationships in the DAG highlighted key patient characteristics, CVC management practices, and overlapping factors influencing infection. This model serves as a novel framework to understand CLABSI causation and supports trial design by identifying confounding factors, causal pathways, and meaningful endpoints.</p> Conclusions/implications <p>Our causal DAG provides a structured representation of CLABSI risk factors, which may support the design of clinical trials examining interventions to reduce CVC-related infections. By clarifying causal mechanisms, the DAG can enhance the specificity of endpoints and improve the rigor of prevention strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Central venous catheter infections: building a causal model with expert domain knowledge to inform future clinical trials

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月8日

查看全文
  • Jessica A. Schults,
  • Yue Wu,
  • Thomas Snelling,
  • Gladymar Pérez Chacón,
  • Daner Ball,
  • Karina Charles,
  • Julie Marsh,
  • Charlie McLeod,
  • Hideto Yasuda,
  • Claire M. Rickard

摘要

Aim

Central venous catheters (CVCs) are essential for long-term therapies but carry a high risk of central line-associated bloodstream infections (CLABSIs), which significantly impact patient outcomes and healthcare costs. This study aimed to develop a causal model for CLABSI using expert knowledge to guide future clinical trials and prevention strategies.

Methods

We constructed a directed acyclic graph (DAG) informed by literature and expert knowledge elicitation. A multidisciplinary team of clinicians, including infectious disease and vascular access experts, participated in interviews and workshops to refine the DAG, resulting in a final model with 30 variables representing CLABSI development.

Findings

The expert-elicited DAG identified two main pathways, patient-related and CVC-related, each contributing to CLABSI risk. Variables and relationships in the DAG highlighted key patient characteristics, CVC management practices, and overlapping factors influencing infection. This model serves as a novel framework to understand CLABSI causation and supports trial design by identifying confounding factors, causal pathways, and meaningful endpoints.

Conclusions/implications

Our causal DAG provides a structured representation of CLABSI risk factors, which may support the design of clinical trials examining interventions to reduce CVC-related infections. By clarifying causal mechanisms, the DAG can enhance the specificity of endpoints and improve the rigor of prevention strategies.