Predicting the Absence of Serious Bacterial Infection in the PICU

预测 PICU 中不存在严重细菌感染

• 批准号: 10806039
• 负责人: Blake Martin
• 金额: $ 16.28万
• 依托单位: UNIVERSITY OF COLORADO DENVER
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-21 至 2028-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10806039
• 关键词: Acceleration; Acute Renal Failure with Renal Papillary Necrosis; Admission activity; Adolescent; Adverse effects; Antibiotic Resistance; Antibiotic Therapy; Antibiotics; Automobile Driving; Award; Bacterial Infections; Biological; Biometry; Calibration; Child; Childhood; Childhood Injury; Clinical; Clinical Data; Clinical Decision Support Systems; Critical Illness; Critically ill children; Data; Data Set; Decision Making; Development; Education; Electronic Health Record; Equity; Ethics; Evaluation; Exclusion; Functional disorder; Funding; Goals; Harm Reduction; High Prevalence; Hospitals; Hour; Infection; Institution; Interview; Knowledge; Laboratories; Left; Length of Stay; Life; Machine Learning; Mentors; Methods; Mission; Modeling; Multicenter Trials; National Institute of Child Health and Human Development; Organ; Outcome; Patient-Focused Outcomes; Patients; Pattern; Pediatric Hospitals; Pediatric Intensive Care Units; Performance; Predictive Value; Probability; Provider; Pseudomembranous Colitis; Public Health; Qualifying; Research; Research Design; Research Personnel; Retrospective cohort; Risk; Sepsis; Site; System; Technical Expertise; Testing; Time; Transcript; Trust; United States National Institutes of Health; Work; career development; clinical decision support; cohort; computer human interaction; design; disability; experience; improved; improved outcome; innovation; machine learning method; machine learning model; machine learning prediction; model design; mortality; multidisciplinary; novel; pediatric sepsis; predictive modeling; predictive tools; prospective; severe injury; skills; support tools; tool; tool development; treatment duration; user centered design

Proposal Summary There are no validated systems for identifying children without serious bacterial infection (SBI) upon admission to a pediatric ICU (PICU). Given the high prevalence of SBI among critically ill children (up to 46%) and risks associated with delayed antibiotic administration, nearly 50% of children without SBI receive antibiotics while microbiologic studies are pending. However, antibiotics can have adverse effects including acute kidney injury, clostridium difficile colitis, and development of antibiotic resistance. The long-term goal of this research is to validate and disseminate machine learning (ML)-based clinical decision support (CDS) tools able to improve PICU antibiotic decision-making thereby reducing antibiotic associated harm among critically ill children. In prior work, Dr. Martin developed ML-based predictive models, which use electronic health record (EHR) inputs (vital sign, laboratory, and other clinical data), to accurately identify children without SBI upon PICU admission in a single center retrospective cohort. The central hypothesis is that these models will demonstrate similar robust performance during prospective and multicenter evaluations, and that an antibiotic decisional needs analysis of PICU clinicians will inform the optimal design of model-based CDS tools. The central hypothesis will be tested via three aims: 1) prospectively evaluate two SBI predictive models within a single center EHR and determine the potential effect on antibiotic-days per child; 2) evaluate ML model generalizability by testing them in a multicenter EHR cohort; and 3) perform a multicenter, multidisciplinary antibiotic decisional needs analysis of PICU clinicians to facilitate user-centered design of equitable model-based CDS tools. In Aim 1, two SBI predictive models will be prospectively evaluated in silent fashion (predictions not revealed to clinicians) at a single center over two years. Model predictions will be compared to patient SBI outcomes to determine their negative predictive value and potential to reduce unnecessary antibiotics. In Aim 2, the same models will be applied to a retrospective dataset of six US children's hospital PICUs (~178,000 encounters over 8+ years) to assess generalizability by determining each model's negative predictive value and potential to reduce unnecessary antibiotics. In Aim 3, a rigorous qualitative content analysis of PICU clinician interviews from five institutions will identify the values driving antibiotic decision-making and enable user-centered design of model- based CDS tools. The research is innovative because it involves development of the first clinically validated system for excluding SBI at PICU admission and uses a ML approach to do so. The research is significant as it accelerates development of generalizable antibiotic decision-making tools to assist PICU clinicians in safely minimizing unnecessary antibiotics and associated harm. The educational component of this application will allow Dr. Martin to attain expertise in biostatistics, probability, ML bias, and study design, as well as technical skills in programming, ML, and CDS. This will allow him to transition to independence and make him uniquely qualified to develop, validate, and implement CDS tools able to improve the outcomes of critically ill children.

提案摘要 没有经过验证的系统可以识别入院时没有严重细菌感染 (SBI) 的儿童 转至儿科重症监护室 (PICU)。鉴于危重儿童中 SBI 的高患病率（高达 46%）和风险 与延迟使用抗生素有关，近 50% 没有 SBI 的儿童在接受抗生素治疗时接受了抗生素治疗 微生物学研究正在进行中。然而，抗生素可能会产生副作用，包括急性肾损伤、 艰难梭菌结肠炎和抗生素耐药性的发展。这项研究的长期目标是 验证和传播基于机器学习 (ML) 的临床决策支持 (CDS) 工具，能够改进 PICU 抗生素决策从而减少危重儿童中抗生素相关的伤害。在 在之前的工作中，Martin 博士开发了基于 ML 的预测模型，该模型使用电子健康记录 (EHR) 输入 （生命体征、实验室和其他临床数据），以便在入院 PICU 时准确识别无 SBI 的儿童 在单中心回顾性队列中。中心假设是这些模型将展示类似的 在前瞻性和多中心评估中表现强劲，并且抗生素决策需要 对 PICU 临床医生的分析将为基于模型的 CDS 工具的优化设计提供信息。中心假设将 通过三个目标进行测试：1）前瞻性评估单中心 EHR 内的两个 SBI 预测模型，以及 确定对每个儿童抗生素使用天数的潜在影响； 2）通过测试评估ML模型的泛化性 他们在多中心 EHR 队列中； 3) 执行多中心、多学科抗生素决策需求 对 PICU 临床医生进行分析，以促进以用户为中心设计基于模型的公平 CDS 工具。在目标 1 中，两个 SBI 预测模型将以静默方式进行前瞻性评估（预测不会向临床医生透露） 一个中心超过两年。模型预测将与患者 SBI 结果进行比较，以确定他们的 阴性预测值和减少不必要抗生素的潜力。在目标 2 中，相同的模型将 应用到美国六家儿童医院 PICU 的回顾性数据集（8 年以上约 178,000 次接触） 通过确定每个模型的负面预测值和减少的潜力来评估普遍性 不必要的抗生素。在目标 3 中，对来自 5 名 PICU 临床医生的访谈进行了严格的定性内容分析 机构将确定推动抗生素决策的价值观，并实现以用户为中心的模型设计 基于 CDS 的工具。该研究具有创新性，因为它涉及第一个经过临床验证的药物的开发 系统在 PICU 入院时排除 SBI，并使用机器学习方法来实现这一点。这项研究意义重大，因为它 加速通用抗生素决策工具的开发，以协助 PICU 临床医生安全地进行抗生素治疗 最大限度地减少不必要的抗生素和相关危害。该应用程序的教育部分将 让 Martin 博士获得生物统计学、概率、机器学习偏差、研究设计以及技术方面的专业知识 编程、机器学习和 CDS 方面的技能。这将使他能够过渡到独立并使他变得独一无二 有资格开发、验证和实施能够改善危重儿童治疗结果的 CDS 工具。