Project Details - Ongoing
Grant Number:R18 HS026188
- Funding Mechanism(s):
AHRQ Funded Amount:$1,193,033
- Principal Investigator(s):
- Project Dates:9/1/2018 to 6/30/2022
- Care Setting:
- Medical Condition:
- Type of Care:
- Health Care Theme:
Acute Respiratory Distress Syndrome (ARDS) contributes to the morbidity and mortality seen in many acute conditions such as pneumonia, influenza, sepsis, trauma, and aspiration. ARDS is under-diagnosed in up to 40 percent of patients and proven treatments are significantly underutilized due to the lack of diagnosis. Appropriate management of ARDS requires timely diagnosis and evidence-based best practices that are aligned with clinical decision making. However, the cognitive overload resulting from the large amount of changing data in electronic medical records (EMRs), and the involvement of multiple care providers at different times with different expertise, makes it particularly challenging to diagnose ARDS.
This project will develop and evaluate the effectiveness of a clinical decision support system (CDSS) for identification and management of ARDS called TRanslate Evidence into AcTion (TREAT). As part of TREAT, an EMR-based tool, ARDS Sniffer 2.0, will identify patients with ARDS. Management of ARDS will be supported by ECARDS, Electronic Clinical decision support in ARDS. A dashboard will allow for auditing and feedback. This project aims to harness the power of big data in a hospital information system to identify patients with ARDS and prompt clinicians with the evidence-based best management of this condition.
The specific aims of the project are as follows:
- To develop and validate an automated, EMR-based tool to identify patients with ARDS.
- To develop an evidence-based, context-appropriate system, ECARDS, for the management of ARDS and an ARDS Dashboard for audit and feedback.
- To evaluate the effectiveness of the ARDS Sniffer 2.0 and ECARDS in a real-world clinical setting.
- To promote the dissemination of ARDS Sniffer 2.0 and ECARDS through partner professional organizations.
Researchers will develop and validate a model to identify patients with ARDS using real-time EMR data, natural language processing, deep phenotyping, and machine learning algorithms. They will extract clinical data from the EMR on all inpatients at the three major hospitals in the Montefiore Healthcare System. A multidisciplinary group of content experts, health informaticians, data scientists, and qualitative researchers will develop a framework for the CDSS. The team hypothesizes that the new CDSS will result in a more patient-centered approach to identification, management, and treatment of ARDS for the patients’ multi-providers care team.