How predictive analytics can help clinical care

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Health systems need to start using analytics to reduce spending and improve routine care for patients, according to  an article in the  Harvard Business Review.

Several health systems that have begun to implement analytics to help clinical decisions were reviewed by the article’s authors: Ravi Parkih, MD, resident in internal medicine at Brigham and Women’s Hospital; Ziad Obermeyer, MD, assistant professor of emergency medicine at Harvard Medical School; and David Westfall Bates, MD, a senior vice president and chief innovation officer at Brigham and Women’s.

“Big data remains largely a buzzword, not a reality, in the routine delivery of health care,” the authors said. “Health systems are still learning how to broadly apply such analytics, outside of case examples, to improve patient outcomes while reducing spending.”

The trio recommended several steps for health systems to make more effective use of that data, beginning with identifying which clinical decisions can be answered by analytics.

Parkland Health and Hospital System in Dallas was used as an example for creating an algorithm to determine the risk of heart failure patients of being readmitted to the hospital. Those labeled as higher risk receive extra attention, such as a follow-up call to within two days to make sure the patient is taking their medication and scheduling an outpatient appointment within seven days.

“In a prospective study, the algorithm-based intervention reduced readmissions by 26 percent. Parkland’s success stems from focusing its algorithm on a specific population and tying it to discrete clinical interventions,” Parikh, Obermeyer, and Bates said.

The other recommended steps were utilizing the greatest amount of data possible to inform algorithms, applying it to high-cost treatments, and gently integrating analytics, warning that otherwise, it “may just be another number that physicians ignore.”