Leveraging Big Data in Healthcare

By Nancy Beale, VP-Clinical Systems and Integration, NYU Langone Medical Center And Sheryl A.Bushman, Chief Medical Information Officer, NYU Langone Medical Center

Nancy Beale, VP-Clinical Systems and Integration, NYU Langone Medical Center

In recent years, healthcare organizations across the country have raced to the finish line to achieve “Meaningful Use” of the Electronic Health Record (EHR). Stimulated by the Health Information Technology for Economic and Clinical Health (HITECH) Act,meaningful use is centered on improving health outcomes for populations of patients and the general public at large (U.S. Office of the National Coordinator, 2014). These efforts to modernize the national healthcare infrastructure are beginning to pay dividends beyond financial incentive payments. Healthcare organizations are nowable to leverage the rich data created by the EHR through the use of analytics. NYU Langone Medical Center is among organizations that are pioneering waysto empower clinicians to improve clinical outcomes at the point of care and beyond. NYU Langone is leveraging the rich data from within the EHR to shift the healthcare paradigm from one of reactive treatment to proactive prevention.

A well-implemented EHR includes ongoing governance, change management, engaged key stakeholdersalongside the technology, and training. This is essentialto create a solid foundation for leveraging data to present the right information to the right care provider at the right time. For instance, EHRs provide basic actionable data to clinicians in real-time,such as identifying when a patient’s laboratory results are abnormal, or when a medication is overdue. More advanced functions take advantage of data from multiple sources to introduce predictive analytics that can then be translated into improved clinical outcomes. Examples below illustrate a continuum from basic to advanced functions utilized at NYU Langone including:

1.Patient lists with icons, data, and decision support in the form of alerts to assist the clinician in real-time as they provide direct patient care.
2.Ad-hoc reporting tools, allowing search and aggregation of the data to allow the provider to analyze a group of patients or a particular intervention.
3.More advanced reporting tools, powered to search the patient database to cull information for quality metric reporting
4.Advanced analytics, introducing algorithmic logic to analyze or identify a cohort of patients such as real-time risk of sepsis or readmission.
5.Combining EHR data with other technology and outside data sources to create information and reporting synergy.

Most organizations hover along the continuum between #2 and #3, but NYU Langone has combined EHR and external data to create new information for population health management and prevention—advancing along this continuum and continually searching for new and meaningful ways to utilize the wealth of data available to us. As part of NYU Langone’s Clinically Integrated Network, a network of voluntary and faculty physicians practices in the tri-state area that have business relationships with NYU Langone and share patient data, the Medical Center is combining the use of claims data; data obtained from a health information exchange (HIE) and EHR data to proactively manage patients with specific conditions. NYU Langone has a well-established Enterprise Data Warehouse (EDW) which integrates data from multiple enterprise software systems and was expanded to incorporate payor claims data alongside EHR patient data. Using the EDW along with EHR reporting tools, condition specific registries are created and managed by care coordinators. The coordinators are then able to follow specific prescribed interventions and outreach to patients to prevent complications and promote health among those patients who have been identified most at risk.

Sheryl A.Bushman, Chief Medical Information Officer, NYU Langone Medical Center

With the use of predictive modeling, NYULangone is leveraging data from multiple sources to identify patients who are at risk for readmission. This risk stratification algorithm leverages the data in the EHR to calculate a score,which is reviewed daily by clinical care coordinators. The care coordinators then have the ability to schedule additional interventions to mitigate the risk of readmission.

“EHRs provide basic actionable data to clinicians in real-time, such as identifying when a patient’s laboratory results are abnormal, or when a medication is overdue”

Organizational transparency is created when real-time data is shared at all levels, from the front line to executives. NYULangone is leveraging EHR data and increasing transparency within the organization through the use of a real-time dashboard that provides access to patient throughput information. Using data available in the EHR, real-time dashboards create organizational visibility into how long a patient has been waiting for a bed along with census and other throughput data. This real-time data is used across the organization to remove obstacles to patient throughput and is visualized at all levels, from the CEO to the staff in Patient Placement and Progression.

Mobility and access to information on-the-go is another key success factor for NYU Langone. Using some of the same statistical data available in the EHR, NYU Langone has created a mobile application to view real-time information about wait times for patients in the emergency department. This increased transparency has heightened awareness among executive leaders in the organization, enabling more rapid responses to organizational obstacles that might be impeding patient flow.

Effective change management is an essential component to the transparency created by the broad visibility into real-time data. Data alone does not create change; however, visibility into real-time data can create fertile ground for action given the right environment, support and tools afforded at the point of care and beyond.  

Beyond these current applications, institutions are poised to leverage patient data to prescribe interventions and educational content in real-time. The patient’ personal health record provides a rich store of data that affords important opportunities for advanced analytics and patient-level decision support. With growth in systems integration, data availability, and innovation, we can expect to see a future in which systems automatically suggest or extend to patients specific support based upon their health data. Advances in analytics, decision support, and open Application Programming Interfaces (APIs), along with partnerships among vendors, will move this future toward reality for an increasingly educated and demanding consumer population seeking not only better interaction with their healthcare providers, but support for and integration with the emerging constellation of their personal fitness devices.