Clinical Decision Support Tool in PARDS Pilot Study

Overview

Previous clinical trials in adults with acute respiratory distress syndrome (ARDS) have demonstrated that ventilator management choices can improve Intensive Care Unit (ICU) mortality and shorten time on mechanical ventilation. This study seeks to scale an established Clinical Decision Support (CDS) tool to facilitate dissemination and implementation of evidence-based research in mechanical ventilation of infants and children with pediatric ARDS (PARDS). This will be accomplished by using CDS tools developed and deployed in Children's Hospital Los Angeles (CHLA) which are based on the best available pediatric evidence, and are currently being used in an NHLBI funded single center randomized controlled trial (NCT03266016, PI: Khemani). Without CDS, there is significant variability in ventilator management of PARDS patients both between and within Pediatric ICUs (PICUs), but clinicians are willing to accept CDS recommendations. The CDS tool will be deployed in multiple PICUs, targeting enrollment of up to 180 children with PARDS. Study hypotheses: 1. The CDS tool in will be implementable in nearly all participating sites 2. There will be \> 80% compliance with CDS recommendations and 3. The investigators can implement automatic data capture and entry in many of the ICUs Once feasibility of this CDS tool is demonstrated, a multi-center validation study will be designed, which seeks to determine whether the CDS can result in a significant reduction in length of mechanical ventilation (LMV).

The central hypothesis is that CDS will help standardize ventilator management consistent with evidence-based recommendations leading to shorter LMV by limiting VILI (Ventilator Induced Lung Injury), preventing VIDD (Ventilator Induced Diaphragm Dysfunction) and allowing earlier recognition that patients are ready for liberation from the ventilator. However, key questions must be addressed prior to wide dissemination of this CDS tool: Specific Aim 1: To assess the feasibility of implementing a web-based, de-identified CDS tool for MV in pediatric ARDS in multiple PICUs. Hypothesis: this CDS tool will be implementable in all PICUs to function consistent with each hospital's specific Information Technology (IT) capabilities. Specific Aim 2: To assess the acceptability and compliance with recommendations from the CDS tool related to oxygenation, ventilation, weaning and extubation readiness testing in PARDS patients at each of the participating PICUs (anticipated 20 patients enrolled per site). Hypothesis: over time, adherence with recommendations in each of these domains will exceed 80% in all PICUs. Specific Aim 3: To implement methods for automated data capture within CDS to provide the right information, to the right person, using the right format, in the right channel and at the right time during workflow ("CDS Five Rights") framework, which can be adapted to the individual IT capabilities at each hospital. Hypothesis: Over 90% of necessary data can be pulled into the CDS in an automated fashion in sites which have access to electronic data capture of ventilator settings and blood gases. Patients will be managed on the eVentilator protocol (the CDS tool), through the acute, stable and weaning phases of mechanical ventilation, including Spontaneous Breathing tests (SBTs) and Extubation Readiness tests. Data will be qualitatively assessed for implementation barriers. Acceptance and rejection and mode stratification will be examined. Data needed for the CDS protocol will be available electronically or can be interfaced through the bedside monitor. Patients will be on the CDS protocol, as intent to treat, while patient is on invasive mechanical ventilation, capped at 28 days, limitation of care, or death, whichever comes first.