Strengthening implementation science in Acute Respiratory Failure using multilevel analysis of existing data

利用现有数据的多级分析加强急性呼吸衰竭的实施科学

• 批准号: 10731311
• 负责人: Alison Turnbull
• 金额: $ 13.65万
• 依托单位: JOHNS HOPKINS UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-07-15 至 2025-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10731311
• 关键词: Acute respiratory failure; Data; implementation science; multilevel analysis

PROJECT SUMMARY Up to 1 million Americans experience acute respiratory failure (ARF) and require mechanical ventilation in an intensive care unit annually. Studies repeatedly revealed incomplete penetration of proven-effective, sometimes life-saving, evidence-based practices (EBP) for these patients, and it is unclear how to select optimal implementation strategies that can bridge the gap between evidence and practice. Common approaches to selection have inherent limitations. For example, concept mapping and implementation mapping rely heavily on stakeholder perspectives, are labor-intensive, and may focus on stakeholder preferences instead of strategies with the greatest potential impact. Quantitative approaches are also challenging because important determinants of practice - such as individual motivation and organizational culture - are difficult to measure at scale. One important goal of implementation is to reduce variability in the uptake of EBPs attributable to clinicians and the environmental setting. While clinical practice should vary in response to patient factors and preferences, implementation programs try to overcome clinician and environmental factors (e.g. insufficient knowledge or resources) that limit EBP uptake. Applying the Consolidated Framework for Implementation Research (CFIR) to this conceptual model, the domains of Individuals and Inner Setting should have minimal influence on adherence to EBPs after a successful critical care implementation program. We hypothesize that variability attributable to the CFIR domains of Individuals and Inner Setting is lower among patients when a treatment is supported by high-quality evidence compared to patients for whom the existing evidence for a treatment is weaker. Our overall objective is to demonstrate 1) how established multilevel modeling techniques can be used to estimate the proportion of variation in the use of EBPs that is attributable to the CFIR domains of Inner Setting and Characteristics of Individuals, and 2) how the resulting information can inform selection of implementation strategies and evaluate their effectiveness. As a proof of concept, we will study two proveneffective interventions - low tidal volume ventilation for acute respiratory distress syndrome and bag mask ventilation during intubation. We will use existing multicenter datasets from the Low Tidal Volume Universal Support: Feasibility of Recruitment for lnterventional Trial (LOTUS-FRUIT) cohort study and from 3 randomized trials that collected data on the use of bag-mask ventilation.

项目总结