A Wearable Device for Predicting Heart Failure Readmissions

A predictive algorithm using data from a noninvasive sensor demonstrated promising specificity and sensitivity.

Predicting readmissions for heart failure (HF) is notoriously difficult. Statistical models employing traditional clinical variables discriminate poorly and thus fail to identify high-risk individuals. Digital wearable technologies measuring previously inaccessible factors might improve risk prediction. Researchers in the single-arm LINK-HF study (NCT03037710) tested the accuracy of an algorithm using data from an adhesive sensor worn on the chest to identify impending hospitalizations in 100 veterans discharged after HF hospitalization (mean age, 68; 98% men; 26% with preserved systolic function).

The sensor collected electrocardiographic data and measurements of skin impedance, temperature, activity, and posture and uploaded them to a clou…

Reviewing Author

Frederick A. Masoudi, MD, MSPH, FACC, FAHA

Disclosures

Consultant/Advisory BoardBristol Myers Squibb; CPC Clinical Research

Grant/Research SupportNational Heart, Lung, and Blood Institute

Editorial BoardsUpToDate; American College of Cardiology Self-Assessment Program (SAP)

Leadership Positions in Professional SocietiesAmerican College of Cardiology (Chair, Innovations Committee)