Machine learning identifies preop risks linked to postop Medicare super-users

Editor's Note

In this study of more than 1 million Medicare patients, 4.8% were super-users of healthcare and incurred 31.7% of Medicare expenditures after surgery.

A machine learning approach identified the following as the most significant risk factors linked to super-utilization of healthcare in the year following surgery:

• hemiplegia/paraplegia
• weight loss
• congestive heart failure with chronic kidney disease stages I to IV.

By proactively identifying super-users of healthcare after surgery with machine learning, targeted efforts may decrease the cost burden on the healthcare system and improve quality of care and outcomes for those patients, the researchers say.