Using Interventional Analytics for Accurate Risk Stratification and Integrated Care

Managing the health of patient populations through effective risk stratification is vital to successful value-based care programs. To raise the bar on patient risk stratification, many hospitals and health systems have turned to Interventional Analytics to identify which patients are at the highest risk for adverse events, hospital admissions, and readmissions.

Using Interventional Analytics for Accurate Risk Stratification and Integrated Care
Using Interventional Analytics for Accurate Risk Stratification and Integrated Care

In this issue brief, learn how Interventional Analytics is connecting hospitals and health systems with LTPAC facilities, to improve clinical performance by reducing avoidable hospital readmissions, managing care coordination efforts, and detecting early warning signs of infectious disease.

Content Summary

Introduction
Going beyond predictive analytics
Customizing care for individual patients
Reducing readmissions
3 Ways interventional analytics reduce hospital readmissions

Introduction

As value-based care models gain dominance over traditional fee-for-service care delivery, health care organizations are realizing how much their long-term success and financial viability are tied to meeting the requirements of those models. Value-based reimbursement requires health systems to set up care pathways that standardize processes while maintaining quality. Effectively structuring those pathways demands rigorous, accurate patient risk stratification — a core component of population health management and the central tenet of value-based care.

Managing the health of patient populations through effective risk stratification is vital to successful value-based care programs. To raise the bar on patient risk stratification, many hospitals and health systems have turned to predictive analytics to predetermine which patients are likely to be at the highest risk for adverse events, hospital admissions, and readmissions. But the bar can go higher.

Going beyond predictive analytics

“Hospitals have been hearing about predictive analytics for a long time and, until recently, it was considered the cutting edge of health care statistical analysis,” says Phyllis Wojtusik, R.N., executive vice president of health systems, Real Time Medical Systems, Linthicum Heights, Md. “But because predictive analytics models forecast likely outcomes and assess risk for patient groups with certain conditions or symptoms, they rely on dated, lagging data that provide generalized predictions of what could happen to a future patient who presents with a similar condition or symptoms.” But that method is no longer timely nor targeted enough, she argues. Today’s cutting edge is Interventional Analytics.

“Interventional Analytics goes beyond predictive analytics,” Wojtusik explains. “Employing comprehensive algorithms, Interventional Analytics uses software that ‘sits on top’ of the electronic health record (EHR) and can be used remotely by multiple providers along the care continuum to constantly pull data on the cloud from the EHR.” The software sifts through vast amounts of patient data, finds correlations, and identifies interventional moments — in real-time. Those distinctions make all the difference, Wojtusik says.

“Interventional Analytics uses constant real-time data to study what is happening to the patient at the moment to affect outcomes — data points such as blood pressure, pulse rates and other indicators — and, if necessary, pushes clinical alerts to providers that allow them to intervene before adverse events occur,” she says. Interventional Analytics also minimizes manual data entry and data stratification, she adds.

“Interventional Analytics goes beyond predictive analytics. Employing comprehensive algorithms, Interventional Analytics uses software that ‘sits on top’ of the electronic health record (EHR) and can be used remotely by multiple providers along the care continuum to constantly pull live data from the EHR.” – Phyllis Wojtusik, R.N., executive vice president of health systems, Real Time Medical Systems

Wojtusik gives a hypothetical example of a patient with congestive heart failure (CHF) who gains three pounds over three days. “That may not be significant in and of itself, but if the patient also experiences shortness of breath on the second day and develops edema in his or her ankles on the third day, there are now three factors affecting that patient [status],” she says. “Providers can see those changes in real-time and determine when to step in to improve an outcome and avoid urgent hospital admission or readmission.”

Customizing care for individual patients

Shane Dearing, executive vice president of growth, Real Time Medical Systems, explains, “Interventional Analytics customizes care for the individual patient as well as analyzes trends for groups of patients.” In addition to CHF, medical conditions that particularly benefit from Interventional Analytics monitoring include head injuries, joint replacements, sepsis, chronic obstructive pulmonary disease (COPD), and diabetes, Wojtusik says.

Real Time’s Interventional Analytics platforms are currently utilized in over 1,000 long-term post-acute care (LTPAC) facilities nationwide — a crucial care site for avoiding adverse patient outcomes, managing length of stay, and reducing hospital readmissions in particular. “We are discharging patients from the hospital sicker than we ever have before, whether sending them home or to nursing facilities,” Wojtusik says. “Interventional Analytics helps us get patients to the right level of post-acute care at the right time by tracking their functional status while they’re still in the nursing facility.”

Over the past two years, Real-Time has started bringing its LTPAC capabilities into the hospital setting, says Keri DeSalvo, marketing director, Real Time Medical Systems, with goals to improve care transitions and maintain continuity of care. “We give hospitals a direct line of sight into how their patient is progressing in the nursing facility — they didn’t have that visibility before,” she says. “And for the LTPAC, [the platform] increases their referrals because they can better manage lengths of stay and improve quality and patient care.” Wojtusik adds, “If an LTPAC manager calls a hospital case manager with an Interventional Analytics clinical alert, that case manager will contact the patient’s physician right away — it gets his or her immediate attention.”

“Interventional Analytics customizes care for the individual patient as well as analyzes trends for groups of patients.” – Shane Dearing, executive vice president of growth, Real Time Medical Systems

Reducing readmissions

To further streamline the identification of high-risk patients, particularly those who have just been discharged from the hospital to a nursing facility, Real-Time features its patent-pending CARD readmission risk scoring system. Using live risk stratification, patients’ CARD scores are determined by four principal factors: the number of clinical alerts in their charts such as weight changes or urine production; how recently they have been admitted to the nursing facility; the number of readmissions they have had over the past 180 days; and their diagnosis scores, which are based on their types of illness. Higher-risk conditions such as CHF or COPD generate higher diagnosis scores. The higher a patient’s overall CARD score, the greater his or her risk for complications and hospital readmissions.

“The first 72 hours after hospital discharge present the highest risk for readmission,” Wojtusik says. “The receiving facility may not yet have all the information they need about the patient. The CARD score provides outcomes-based data that help [LTPAC staff] quickly recognize patients’ post-acute trends, and it helps the hospital properly place patients in the right post-acute facility for the best care.” Their mutual goal: lower costs, reduced readmissions, a better-managed length of stay, and greater patient satisfaction.

“Value-based care involves every stage of treatment and looks at how patients should be cared for as they move through the care continuum,” Wojtusik says. “With 73 cents of every post-acute health care dollar spent on skilled nursing facilities, all providers in the continuum need this real-time data and the direct line of sight it provides into patient care. If both nursing facilities and hospitals used an interventional analytics system, they not only could provide customized patient care, they also could reduce readmissions by up to 52% while managing down the length of stay.”

“If both nursing facilities and hospitals used an Interventional Analytics system, they not only could provide customized patient care, they also could reduce readmissions by up to 52% while managing down length of stay.” – Phyllis Wojtusik, R.N., executive vice president of health systems, Real Time Medical Systems

3 Ways interventional analytics reduce hospital readmissions

  1. Live clinical alerts. Using live data within the electronic health record (EHR), Interventional Analytics detects a change in a patient’s condition as it happens. Based on predefined parameters, when the data indicates a potentially negative outcome, a clinical alert — including diagnosis and specific intervention recommendations — is immediately sent to clinical teams.
  2. Optimized nurse time. Live clinical insights delivered from the EHR system through Interventional Analytics help LTPAC nurses quickly prioritize and focus on patients at the highest risk for readmission.
  3. Enterprise dashboards. Interventional Analytics generates enterprise-level dashboards that utilize live data within the EHR, searching nursing notes and flagging keywords (e.g., dehydration)

Source: Real Time Medical Systems

Thomas Apel Published by Thomas Apel

, a dynamic and self-motivated information technology architect, with a thorough knowledge of all facets pertaining to system and network infrastructure design, implementation and administration. I enjoy the technical writing process and answering readers' comments included.