Northwell Health's Center for Health Innovation and Outcomes Research (CHIOR) has spent a decade conducting clinical decision support system (CDSS) related research. Our work began with the science of deriving and validating clinical prediction rules (CPRs). As health information technology improved, our research focus shifted to the efficient use of the electronic heath record (EHR) as a tool for the dissemination and implementation of these evidenced-based tools. CPRs are the engines driving complex CDSSs, which utilize multiple forms of data to calculate patient specific probabilities to inform decision making at the point of care (we refer to them as integrated CPRs or iCPRs). Our extensive experience in this area sparked an ultimate vision, to build an EHR agnostic platform from which to implement these tools. This proposal focuses on the development, testing, and measurement of the impact of a novel, complex, evidence- based CDSS, built using a service oriented architecture (SOA), that will function across multiple EHR platforms. The key innovation of this approach is building a CDSS using a SOA employing a modern, platform- independent methodology (adhering to HL7 and SMART on FHIR standards where possible) that allows it to integrate with other open health information technology (HIT) solutions. For the purposes of this project, the CDSS we develop will operate within Northwell's health information exchange (HIE) environment and EHRs. The HIE is a data warehouse and exchange which aggregates information across Northwell's HIT continuum (i.e. EHRs, billing systems, registration systems) to gather multiple types of data (i.e. radiology, medications, diagnoses, appointments, discharge notes). Building the CDSS using a SOA integrated with the HIE, as well as with individual EHRs, allows the CDSS to widely and effectively communicate with every clinical care environment while allowing for customization specific to the needs of each particular micro- environment with a minimal amount of effort. Our CDSS will implement two key evidence based CPRs ? Wells' Criteria to determine the risk for pulmonary embolism and IMPROVE Risk Model for venous thromboembolism ? and assess their ability to impact care in two healthcare environments (emergency department care and inpatient medical care) using two EHRs (Allscripts Sunrise Clinical Manager and Cerner). Specific sets of clinical features will trigger useful and usable EHR enabled provider alerts, iCPRs, and recommendations. With the ability to continuously monitor patient records and trigger CDS to the right provider at the most appropriate time, we can infuse and measure the impact of evidence-based clinical care throughout the health system.

Public Health Relevance

This proposal focuses on building a clinical decision support system (CDSS) using a modern independent way that allows it to interact with any health information technology solution. The CDSS will target two key disease states ? 1) blood clot in the lungs and 2) bleeding risk for hospitalized patients ? and assess their ability to impact care in two healthcare environments (emergency medicine care and inpatient medical care) using two different electronic health record systems. With the ability to continuously monitor patient records and trigger clinical decision support to the right clinical provider at the most appropriate time, we can infuse and measure the impact of evidence-based clinical care throughout the health system.

Agency
National Institute of Health (NIH)
Institute
Agency for Healthcare Research and Quality (AHRQ)
Type
Research Demonstration and Dissemination Projects (R18)
Project #
5R18HS026196-02
Application #
9882226
Study Section
Healthcare Information Technology Research (HITR)
Program Officer
Lomotan, Edwin A
Project Start
2019-03-01
Project End
2022-02-28
Budget Start
2020-03-01
Budget End
2021-02-28
Support Year
2
Fiscal Year
2020
Total Cost
Indirect Cost
Name
Feinstein Institute for Medical Research
Department
Type
DUNS #
110565913
City
Manhasset
State
NY
Country
United States
Zip Code
11030