Early Mobilization: Operationalizing Big Data & Implementation Science to Lead Expansion to ICUs (E-MOBILE-ICU)

项目总结 每年有近80万危重患者需要机械通气，四分之三的幸存者 患有持续性残疾，随着重症监护越来越广泛，这构成了一个重大的公共卫生问题 已利用且可用。虽然早期活动，使患者在机械过程中进行体力活动 通风，是一种有希望的循证干预措施，可以预防残疾，不到10%的患者 蒂蒂斯从来不会下床。这项提案旨在应用精准医学来识别最有可能 从早期动员中受益并阐明如何成功实施以扩大 及早动员长期残疾风险最大的重症护理幸存者。我假设这是- 源密集型干预可以更精确地应用于最有可能受益的患者子集- 可以制定实施科学战略，以成功地推动采用这种干预措施-- 超越了临床试验环境。我将在三个目标中检验我的假设：目标1)我将确定最佳关键 使用尖端机器学习方法实施早期动员的疾病表型；目的 2)我将确定早期动员对长期功能性残疾的影响，以激励采用这一方法 实践；目标3)我将确定在五个方面实施早期动员的障碍和促进者 各机构确定与成功实施相关的背景特征，以便为战略提供信息 这可以弥合证据基础和临床实践之间的差距。我的长期目标是缓解 危重疾病的应用与临床试验，使用基于精度的方法来识别有风险但又易于... 受益人群与实施科学方法相结合，说明如何实现这些干预- 送到床边去。为了实现这一目标，我组建了一支杰出的跨学科导师团队(Dr。 Vineet Arora、Matthew Churpek和John Kress)和顾问(Shyam Prabhakaran博士、Donald Hedeker博士、 Laura Damschroder和Matthias Eikermann)，他们有NIH资助和成功的导师记录- 博士后考生的船。我打算在我作为一名有成就的临床试验人员的基础上再接再厉 制定了深入的职业发展计划，以获得机器学习方法方面的专业知识，以确定不同- 治疗效果(Churpek和Prabhakaran)，纵向数据分析(Arora和Hedeker)，以及实施- 心理科学方法(Arora，Prabhakaran，和Damschroder)，以制定战略，带来复杂的多- 从临床试验(Kress和Eikermann)到日常ICU护理的纪律干预。完成这项工作 提案将训练我满足国家医学科学院最近出版的一份出版物所定义的未得到满足的需求 指出，区别对待效果的确定必须与严格执行相结合，以帮助 将证据库过渡到常规临床护理。具备先进的统计技能和实施 科学的方法，我将能够设计混合有效性-实施试验，以有针对性地实施 在未来R01级别的应用中对弱势人群进行复杂的多学科干预。

PROJECT SUMMARY Almost 800,000 critically ill patients require mechanical ventilation every year and three quarters of the survivors suffer from persistent disability, which poses a major public health problem as critical care becomes more widely utilized and available. Although early mobilization, which engages patients in physical activity during mechanical ventilation, is a promising evidence-based intervention that may prevent disability, less than ten percent of pa- tients ever get out of bed. This proposal aims to apply precision medicine to identify patients who are most likely to benefit from early mobilization and elucidate how it can be implemented successfully to extend the benefits of early mobilization to critical care survivors at greatest risk for long-term disability. I hypothesize that this re- source-intensive intervention can be applied with greater precision to a subset of patients most likely to bene- fit, and that implementation science strategies can be devised to successfully drive adoption of this interven- tion beyond a clinical trial setting. I will test my hypothesis in three aims: Aim 1) I will identify the optimal critical illness phenotype for implementation of early mobilization by using cutting-edge machine learning methods; Aim 2) I will determine the effect of early mobilization on long-term functional disability to incentivize adoption of this practice; Aim 3) I will determine the barriers and facilitators of implementation of early mobilization across five institutions to identify the contextual features associated with successful implementation to inform strategies that can bridge the gap between evidence base and clinical practice. My long-term goal is to mitigate the com- plications of critical illness with clinical trials using precision-based methods to identify at-risk and yet apt-to- benefit populations paired with implementation science methodologies to illuminate how to bring these interven- tions to the bedside. To accomplish this, I have assembled an exceptional interdisciplinary team of mentors (Drs. Vineet Arora, Matthew Churpek, and John Kress) and advisors (Drs. Shyam Prabhakaran, Donald Hedeker, Laura Damschroder, and Matthias Eikermann) who have a track record of NIH-funding and successful mentor- ship of post-doctoral candidates. I intend to build on my foundation as an accomplished clinical trialist and have formulated an in-depth career development plan to gain expertise in machine learning methods to identify differ- ential treatment effects (Churpek and Prabhakaran), longitudinal data analysis, (Arora and Hedeker), and imple- mentation science methods (Arora, Prabhakaran, and Damschroder) to craft strategies that bring complex mul- tidisciplinary interventions from clinical trials (Kress and Eikermann) to everyday ICU care. Completion of this proposal will train me to fill an unmet need defined by a recent National Academy of Medicine publication which indicated that identification of differential treatment effects must be paired with rigorous implementation to help transition evidence base to routine clinical care. Equipped with advanced statistical skills and implementation science approaches, I will be able to design hybrid effectiveness-implementation trials to target and implement complex multidisciplinary interventions to vulnerable populations in future R01 level applications.