Electronic Health Records

Policy & Regulation News

IBM, Carilion Clinic detect patients with heart failure risk

By Nicole Freeman

Carilion Clinic, a Virginia health system, used a pilot program with IBM and Epic to identify 8,500 patients at risk for heart failure. The program utilized predictive modeling of Carilion Clinic’s electronic medical record (EMR) data, along with discharge and physician notes, and IBM’s natural language processing technology to identify patients with an 85 percent accuracy rate.

Over 5 million adults in the US are affected by heart failure, and about half of those with the condition will die within five years of diagnosis. Caused by conditions like diabetes or hypertension, heart failure is a common hospitalization cause for patients over 65. Early detection is often difficult without the use of advanced analytics, and the methodology used in the pilot program may provide an opportunity for early intervention and better patient care.

Predictors analyzed by IBM’s Advanced Care Insights included use of alpha blockers, beta blockers, beta agonists, and other medications; obesity; previous diagnoses; and lifestyle and environmental factors. IBM Advanced Care Insights combines healthcare-specific content analysis and predictive modeling.

“Many predictive factors are included in structured data within today’s EMR systems, but a lot of it is hidden in doctors’ notes, discharge papers, and other sources of unstructured data,” said Sean Hogan, vice president of global healthcare, IBM.  “By tapping into the unstructured data, our clients have more complete and accurate information that allows them to make targeted interventions when appropriate that can help prevent more severe and costly medical complications.”




Sign up to continue reading and gain Free Access to all our resources.

Sign up for our free newsletter and join 60,000 of your peers to stay up to date with tips and advice on:

EHR Optimization
EHR Interoperability

White Papers, Webcasts, Featured Articles and Exclusive Interviews

Our privacy policy

no, thanks

Continue to site...