Reducing hospital admissions: The new goal for HIT

December 3, 2015

In today’s world of value-based care, having a high volume of inpatients means something is broken in the healthcare system. Here's how technology is helping fix it.

One of the most basic goals of any business is to get customers to use more of services so it can make more money. For a hospital, especially under the fee-for-service model, that typically means having more “heads in beds.”

That’s why readers of The New York Times may have been surprised to see a recent ad from Mount Sinai Hospital with the headline, “If our beds are filled, it means we’ve failed.” Aren’t hospital beds supposed to be filled? Shouldn’t Mount Sinai Hospital want to treat as many patients as possible?

In the past, perhaps that was the objective. But the Mount Sinai ad demonstrates that in today’s world of value-based care and community-based population health management, having a high volume of inpatients means something is broken in the healthcare system.

A shifting mindset

HandmakerThe goal for health systems in the era of value-based reimbursement is to treat patients in the appropriate care setting with the lowest cost. While sometimes that will necessitate admittance to the hospital, increasingly healthcare providers are beginning to focus on providing the bulk of care, whenever possible, in ambulatory clinics, the patient’s own home, or other non-hospital settings. These options not only cost less, they are less disruptive to the patient and often safer.

Yet it’s one thing to talk about shifting away from an acute care-centric model. It’s another to put the mechanisms in place to make it happen, especially when the financial incentives haven’t caught up to the advancing care transformation changes that progressive organizations are implementing.

Strategies that are working

Keeping patients out of the hospital means staying on top of the health status of everyone in a health system’s care. That requires developing new models for delivering care, a commitment that crosses traditional organizational boundaries and requires unprecedented insight into each patient’s health and timely, proactive action.

The limitations of the human brain make it impossible for physicians to comprehend every piece of information needed to manage an entire patient panel proactively. That is why health information technology (HIT) even beyond the electronic health record (EHR) has become so essential to delivery of value-based care that keeps patients out of the hospital. HIT applications must evolve to monitor patients so that clinicians can anticipate their care needs rather than react to them.

Health systems such as Orlando Health in Orlando, Florida, and Bon Secours Medical Center in Virginia, are retooling their IT systems, putting functionality in place to aggregate and analyze huge volumes of data, helping them proactively manage patient populations.

Orlando Health, an eight-hospital integrated delivery system, is using HIT to identify patients that are or will soon be sick but don’t know it yet. By using technology to aggregate multiple sources of data, including claims, clinical, administrative and even patient-reported information, algorithms can identify patients at risk for serious illness or avoidable admissions and ER visits. With this information “teed up”, organizations and care teams can intervene earlier.  Using this approach, Orlando Health has improved the quality of care, earned millions of dollars in value-based savings incentives, and is successfully participating in four accountable care organization (ACO) contracts.

Bon Secours Health System, with 19 hospitals in six states, has rolled out a program that combines population health management, patient engagement and care management technologies for a variety of functions, including ensuring that discharged patients get the care and attention they need to prevent unnecessary hospital readmissions.

Discharges trigger an automated communications system that contacts patients within 72 hours of leaving the hospital or ED and asks them to complete a basic health assessment confirming their understanding of discharge instructions and medications. Patient responses are analyzed and are automatically escalated to nurses, social workers and care managers for one-on-one counseling as indicated.

Next: Cognitive computing

 

 

Cognitive computing

Yet even with the help they’re receiving from computing systems, physicians still don’t have all of the information they need or the capacity to parse it at the point of care. Much of the deficiency is the result of the high volume of unstructured data entered into the notes section of the digital chart.

While this is the easiest place for providers to record insights about a patient’s condition, most computers are programmed to look only for data in structured fields. As a result, it’s far too easy for physicians to miss this important, rich information.

The answer to this issue lies in a new technology called cognitive computing. This game-changing development means not only can computers “see” unstructured data; they can also interpret it the way a human would.

Cognitive computing systems can evaluate what is important to the immediate need and what should be ignored in delivering a diagnosis and identifying the best treatment recommendations.

These systems can then make decisions on how information is presented to physicians, with the highest-probability solutions listed first.

Just as important, cognitive computing systems can learn from their successes and their errors, just as people do. Only they learn millions of times faster. With such powerful, data-driven insights at clinicians’ fingertips, we can look forward to a future where all patients receive the targeted care they need to keep them healthy and out of the hospital.

Mount Sinai has challenged conventional thinking with its ad campaign. But this strategy is far more than a marketing gimmick. The health system is joining many other leading organizations across the U.S. in going all-in on innovations that will help it profile its patient population, engage individuals and families, empower physicians and care teams to support patients between and outside of face-to-face encounters, and deliver formerly high-cost services more efficiently in lower-cost settings, all while improving quality and the patient experience. 

Karen Handmaker, MPP, is vice president of population health strategies for Phytel, an IBM Company that is part of Watson Health.