The study proposal is to deploy a wearable solution that predicts physiological perturbation comparable to invasive devices and to perform continuous remote patient monitoring; this will be connected to a structured, cascading, escalation pathway involving home health nurses, advanced practitioner providers, and heart failure specialists, and has the potential to transform heart failure management in the post-discharge period, where patients are the most vulnerable for readmission. This feasibility study will contribute to the understanding of post-discharge heart failure continuous remote patient monitoring, promote patient self-care, and has the potential of improving patient outcomes.

Heart failure is a leading cause of hospital readmission. It results in significant mortality, morbidity, and health care utilization. Effective continuous remote patient monitoring (CRPM) can reduce readmissions, but it has only been realized via invasive monitoring. The study will focus on non-invasive heart failure CRPM through a structured cascading and escalating alert system. In this feasibility study, the study team will use a wearable biosensor and collect ambulatory physiological data that are analyzed by machine learning algorithms, potentially identifying physiological perturbation in heart failure patients. Alerts from this algorithm may be cascaded with other patient status data to inform management by the home health team via a structured protocol. The escalation pathway will engage home health, advanced practitioner providers, and heart failure specialists. In the first aim, the study team will perform a soft launch on five patients with an extensive evaluation to assess feasibility for the pilot trial. In aim 2, the study team will implement the feasibility pilot study. In aim 2a, the study team will conduct surveys and semi-structured interviews with both providers and patients. The surveys and interviews will be applied at three time points (initiation, maintenance, and post-study) to evaluate perceptions, acceptance, and experience of this CRPM solution. In aim 2b, the investigators will leverage temporal data mining, feature extraction, and patient clustering methods to identify valid patterns associated with the pathophysiological events of interest, using continuous physiological data, patient reports, and electronic health record data. The study team will also compare outcome and process measures from our pilot study to a retrospective cohort matched for key demographics and disease severity. This feasibility study will provide key learning for a larger efficacy clinical trial to evaluate if this non-invasive telemonitoring solution tied to structured patient management via cascading and escalating alert pathways can improve outcomes and reduce heart failure readmission.

Inclusion Criteria:

• Patient is an inpatient at Evanston Hospital
• Patient is followed by the heart failure service team after discharge
• Patient has a history of heart failure
• Patient received at least one dose of IV diuretic at index hospitalization
• Symptoms corresponding to NYHA function class II-IV
• Patient has heart failure with reduced left ventricular ejection fraction (LVEF)<40%, or HF with mid-ranged ejection fraction (LVEF 40-50%), or HF with preserved ejection fraction (LVEF≥50%)
• Patient is in the top 50% risk of readmission across NorthShore University HealthSystem's CAPE 30-day readmission model
• Patient is at least 18 years of age
• Patient is fluent in English
• Patient agrees to protocol-required procedures

Exclusion Criteria:

• Patient has cognitive or physical limitations that, in the investigator's opinion, limit the patient's ability to maintain patch device and phone
• Patient has visual impairments
• Patient has an allergy to hydrocolloid adhesives
• Patient has present skin damage preventing them from wearing a study device
• Patient has renal dysfunction requiring dialysis
• Patient has CardioMEMS
• Pregnancy
• Patient receiving hospice care

Baseline characteristics and/or adverse events were not collected from clinical team staff.

Adverse event data was collected for 30 days after discharge for the enrolled HF patients. HF providers were not monitored/assessed for adverse events.