$2 Million Awarded to Sutter Health, IBM and Geisinger Health System to Study Heart Failure Prediction

SACRAMENTO, Calif., DANVILLE, Pa. and YORKTOWN HEIGHTS, N.Y., October 9th, 2013—Today, Sutter Health, IBM Research (NYSE: IBM) and Geisinger Health System announce the award of a $2 million joint research grant from the National Institutes of Health (NIH) to develop new and sophisticated big data analytics and application methods that could help doctors detect heart failure years sooner than is now possible.

Heart disease is the leading cause of death, disability and costly hospitalizations in the United States. One in five adults will develop heart failure, a type of heart disease that remains nearly impossible to detect early. About half of people who have heart failure die within five years of diagnosis.

The Research Project

Sutter Health, IBM and Geisinger Health System will use the NIH funding to develop practical and cost-effective early-detection methods for application in primary care practices with an electronic health record (EHR) system. The research aims to:

  • Create a deeper understanding of how to use the data contained within EHRs and advanced analytics to help detect heart failure earlier.
  • Identify best practices that help health systems nationwide integrate big data analytics into primary care. This "Smarter Care" approach will help doctors and caregivers use evidence-based insights to better partner with patients and identify more tailored treatment options and holistic approaches to disease management that are personalized for each individual.

EHR data provides a rich, expansive view of a patient's health history that may include a variety of big data, including demographics, medical history, medication and allergies, laboratory test results, and more. Sophisticated analysis of this data could help doctors identify patient's risk of heart failure and reveal signals and patterns that are indicative of such outcome. Once patients are identified as high-risk for heart failure, physicians can better monitor their status, help motivate a patient to make potentially life-saving lifestyle changes and test clinical interventions to potentially slow or possibly reverse heart failure progression.

"Heart failure will remain among our nation's most deadly and costly diseases unless we discover new methods to detect the illness much earlier," said Walter "Buzz" Stewart, Ph.D., MPH, chief research and development officer for Sutter Health and principal investigator for the project. "Sophisticated analysis of EHR data could reveal the unique presentation of these symptoms at earlier stages and allow doctors and patients to work together sooner to do something about it. Through this research we could transform how heart failure is managed in the future."

"IBM is applying advanced tools for analyzing medical data, including text, and reviewing a patient's health records for new insight," said Shahram Ebadollahi, Ph.D., program director, Health Informatics Research for IBM T.J. Watson Research Center and co-principal investigator for the project. "By pairing IBM's expertise in Big Data analytics with the domain knowledge and data of our healthcare partners this project will result in the development of new analytic algorithms for more accurate detection of the early onset of heart failure. Ultimately, we hope to advance a smarter approach to care for patients with heart failure."

Steve Steinhubl, M.D., a cardiologist member of the research team from Geisinger, added, "Our earlier research showed that signs and symptoms of heart failure in patients are often documented years before a diagnosis and that the pattern of documentation can offer clinically useful signals for early detection of this deadly disease. Now we have the technology to enable earlier diagnosis and intervention of serious conditions like heart failure, leading to better outcomes for patients."

The three parties began their initial research in 2009 and published a series of findings in medical journals and conferences.

  • Prediction modeling using EHR data: challenges, strategies, and a comparison of machine learning approaches
  • Automatic identification of heart failure diagnostic criteria, using text analysis of clinical notes from electronic health records
  • Combining knowledge and data driven insights for identifying risk factors using electronic health records

The NIH funding allows the team to look deeper into the progression of factors that are predictors of heart failure so clinicians can implement timely care-management plans to improve health outcomes. They will begin testing predictive methods for heart failure in clinical practice over the next several years. Their findings may also provide insights for providers to use EHR data to improve health outcomes for other chronic conditions.

Heart Failure

Despite major improvements in the treatment of most cardiac disorders, heart failure remains the leading cause of hospitalization for Americans older than 65. The combined costs of heart failure in the United States were estimated at $34.4 billion in 2011, and experts predict the number of people diagnosed with heart failure to double by 2030.

Today, doctors typically diagnose heart failure during later stages of the disease's progression, after irreversible organ damage.

##