Project Grant R00HL169940

The National Heart, Lung, and Blood Institute awarded the University of Pennsylvania $237,153 on August 20, 2026, under the Cardiovascular Diseases Research program (CFDA 93.837) to advance predictive modeling of hospitalizations and emergency department visits for heart failure patients in managed long-term care settings through analysis of audio-recorded telephone communications between patients and care coordinators.

The project, HEAR-HEARTFELT (Identifying the Risk of Hospitalizations or Emergency Department Visits for Patients with Heart Failure in Managed Long-Term Care Through Verbal Communication), builds on prior K99-phase work that identified clinically relevant risk factors embedded in patient-coordinator conversations using three complementary approaches: conversational characteristics analysis, linguistic indicators based on a curated lexicon of heart failure symptoms and psychosocial stressors, and acoustic feature extraction from voice signals. The R00 phase advances to predictive modeling incorporating these multimodal features and expands the data collection period. Heart failure remains a leading cause of unplanned hospitalizations and ED visits despite community-based care coordination programs designed to reduce such events; the project addresses an untapped source of temporally sensitive patient-status information in existing risk stratification models.

Performance takes place in Philadelphia, Pennsylvania. The award period runs through June 30, 2029. This is a project grant, the standard NIH investigator-initiated research mechanism.