Skip to main content



jStart portfolio

learn about how we bring value to our clients



UNC Healthcare: How big data was leveraged to reduce
Medicaid re-admissions

Continuing a long history of collaboration, jStart and UNC Healthcare recently revealed a project in which IBM Big Data technologies (and specifically text analytics technologies) were used to create a system to reduce an inefficient--and costly--challenge facing medical providers: Medicaid re-admissions.

The goal: reduce expenses for the state of North Carolina, and at the same time, enhance patient care by reducing paperwork and bureaucratic overhead--something no patient (or their families) should have to deal with during a time when they are faced with medical issues.

Medical Records image

Broader applications for text analytics

Text analytics solutions can not only help medical providers reduce the number of Medicaid re-admissions, but can also be leveraged to help in numerous other ways. One of the largest private healthcare providers in the Midwest is using IBM text analytics software to help accelerate it's medical research efforts; another teaching hospital is leveraging text analytics to better understand how to make patient care more efficient--reducing patient waiting times while making operations more efficient for the hospital. Healthcare provides a rich environment for text analytics since so much data is produced. The challenge is understanding what insights might be "hidden" within the vast amounts of information a medical provider faces.

Image courtesy of and (C) UNC Healthcare

How jStart helped UNC Healthcare

Leveraging jStart's big data and text analytics expertise, the team worked with UNC Hospitals to deploy a solution built around IBM Content Analytics. A major barrier to reducing hospital readmissions is ensuring that patients follow-up with their healthcare providers in a timely manner and receive needed care after hospital discharge. These follow-up appointments and tests are often only documented as free-text in patients’ hospital discharge summaries. The IBM text analytics technology enabled UNC Healthcare to automatically extract discharge instructions from patients’ discharge summaries and use the structured data to generate alerts and reports to care managers who then ensure that patients follow-up appointments and tests are completed. This has allowed the hospital to pro-actively engage patients to reduce the chance that they will need to be re-admitted--and thus reduce the cost that the state of North Carolina must pay for costly readmissions of Medicaid patients.

"Working with the jStart team, we were able to rapidly create a solution which we believe will help UNC Healthcare improve patient care by reducing hospital readmissions--and at the same time reduce the risk of Medicare penalities for the state of North Carolina."

Dr. Carlton Moore
Associate Professor of Medicine
UNC Healthcare

About the jStart Team

IBM jStart is the company’s emerging technologies client engagement team. Focused on creating solutions built on the latest technologies, the team has been working with companies since 1997 to address their business challenges while providing business value, today. jStart’s latest initiative is the commercialization of IBM’s Watson system as well as the technologies which are a part of it, including DeepQA and Machine Learning.

An official Smarter Planet case reference

InfoSphere Guardium Data Redaction in Action

In addition to improving customer care and lowering readmission rates, UNC is a research hospital focused on a collaborative effort involving its’ 5 Health Affairs Schools, plus the College of Arts & Sciences, the Business School, the School of Social Work and the Law School. This requires sharing data across different systems. To ensure this data is secured and protected, UNC leverages InfoSphere Guardium Data Redaction. Read this whitepaper to understand how InfoSphere Guardium Data Redaction protects sensitive information in documents and forms.

Resources and Links

Want more information?

Want to know more about this jStart engagement?