The goal of this Roybal Center is to develop, test, and implement novel interventions that bridge the dramatic gap between the supply of and need for palliative care services among persons with dementia (PWD) living in long-term service and support (LTSS) facilities, as well as their family caregivers. Several innovative aspects of this proposed Center position us to succeed in this challenging but increasingly important mission. We have assembled a team comprising more than 50 investigators and staff from five schools within the University of Pennsylvania, and the most accomplished investigators of palliative approaches to care for PWD from 11 peer institutions. These investigators possess content knowledge and tremendous expertise in leading all stages of behavioral intervention development as defined by the NIH Stage Model. We will support these behavioral scientists by linking them with Center faculty who have expertise in several key methodologic areas including behavioral economics, data science, statistics, qualitative methods, health policy, health economics, and implementation science.

 

Our focus on translating behavioral economic insights to change the behaviors of clinicians and caregivers involved in the care of PWD builds on our recent success using similar approaches to improve palliative care delivery in outpatient and hospital settings. Our focus on using machine learning, natural language processing, and network methods will enable us to (a) identify mechanisms of clinician behavior that account for aggressive care for PWD, (b) identify PWD with the greatest unmet palliative care needs so as to sustainably target intervention delivery, and (c) more efficiently measure person-centered outcomes in the context of large, pragmatic trials of the most promising interventions.

 

This Center is being launched in partnership with Genesis HealthCare, the nation's largest owner of LTSS facilities, with 401 nursing homes and assisted living facilities across the U.S.. Genesis Healthcare provides a tremendous network within which to develop and test interventions that will be effective and sustainable beyond research trials. Further, by developing and testing the interventions in these facilities, while continually increasing our collaborations with other owners of LTSS facilities, we will be ideally positioned to implement effective interventions in the places in which the majority of Americans with dementia are cared for near the end of life. We propose two pilots for funding in the first year: a Stage I study that seeks to adapt a web-based advance care planning platform for use among PWD and their caregivers in LTSS settings, and a Stage 0 study that will use data science methods to identify clinician behaviors associated with hospitalization risk among PWD, yielding targets for future interventions.