Inconspicuous Daily Monitoring to Reduce Heart Failure Hospitalizations

项目总结/文摘

Project Summary/Abstract In-home monitoring technologies have the potential to transform the healthcare system by enabling the transition from reactive care to proactive and preventive care. This is especially important for cardiovascular disease (CVD); the leading cause of death worldwide. Heart failure (HF), a type of CVD characterized by a weakened heart muscle, impacts approximately 6.5 million Americans with over 960,000 new cases each year. HF costs the US an estimated $30.7 billion annually and is expected to increase 127% to $69.7 billion by 2030. With approximately 80% of the total cost associated with HF due to hospitalization, there is an opportunity to reduce the cost of HF by lowering hospitalization rates through remote patient monitoring. Since patient awareness of symptomology often lags deterioration, successfully tracking physiologic changes in the home is a critical component of an early intervention strategy. Classical approaches to in-home monitoring, such as blood pressure and weight monitoring have had limited success, with patient adherence cited as major barrier to reducing hospitalizations. The central hypothesis of this research is that hospitalization rates and duration of stay for heart failure patients can be significantly reduced through inconspicuous in-home monitoring and early intervention. This research will leverage a highly innovative technology for cardiovascular in-home daily monitoring; the fully integrated toilet seat (FIT). The FIT seat automatically captures a comprehensive cardiovascular assessment in the home, while ensuring long-term patient adherence. A multidisciplinary research team, comprised of engineers, physicians, advanced practice providers (APP), data scientists, biostatisticians, designers, and software developers, will advance an automated system that provides health care providers with early warning of patient deterioration using the FIT system measurements captured in the home. The success of this system will be evaluated through an in-home clinical trial of heart failure patients. Specific Aim 1 seeks to create a learning dataset and data visualization architecture from HF patient in-home physiologic data, perceived wellness, and adverse events. The FIT seat will be deployed for a 90-day in-home study of 200 HF patients, with patient perceived wellness and activity captured through a custom application. This physiologic, wellness, and activity data will be combined with adverse events from the electronic medical record to create an integrated dataset for retrospective analysis and alert model development. In Aim 2, an automated prediction model for early alert of all-cause hospitalizations will be created using novel machine learning techniques. The objective of Aim 3 is to demonstrate that inconspicuous in-home monitoring and early intervention can reduce hospitalizations in a second cohort of 200 HF patients. We hypothesize that the integrated FIT-based alert system will reduce the burden of all-cause hospitalization and will improve the quality of life for patients.