Many patients in intensive care units (ICUs) don't receive high-quality palliative care. The small palliative care specialist workforce can't meet the demand of patients perceived to be at high risk for poor outcomes, ICU teams often lack the full range of palliative medicine skills, and current care often focuses on clinician-centered proxies of patient/family need to activate the delivery of palliative care. A better solution would be to develop a patient-centered, needs-targeted collaborative model of ICU- based palliative care delivery—the overarching goal of this R21. Over 300 physicians and nurses in our preliminary studies stated that they desired a care model that used collaborative practices (i.e., ICU team plus palliative care team) to target family-reported needs. However, such a replicable, scalable, and theory-driven model doesn't exist. To address this gap, we propose a 2-year prospective cohort study involving 250 patients, 275 family members, and 170 clinicians at both academic and community sites. Our multi-disciplinary team's specific aims are (1) to characterize palliative care needs (types, severity, association with outcomes) as reported by ICU family members to derive clinically relevant `need typologies,' and then, (2) using the need typologies, to develop a provisional need-targeted collaborative model of ICU-based palliative care delivery. Exploratory aims will rigorously evaluate the association between needs and popular clinical markers of poor outcomes (i.e., palliative care `triggers') as well as the impact of unmet needs on long-term family member distress. This R21 could provide a paradigm-changing care delivery model that could reduce clinical variability, an easily disseminatable mobile app tool for measuring needs and guiding the choice of clinicians most able to address them, as well as a novel intervention target that is linked to distress. Additionally, this R21 would enable our team to expediently conduct a number of `next step' trials designed to test the novel patient- and family-centered care model and its components. This proposal is innovative in its conceptual orientation, focus on scalability, mixed methods, and use of innovative technologies to personalize care. Furthermore, this R21 addresses care delivery research priorities noted by the NINR's `Innovative Questions in End of Life Research,' the 2016 NIH-NPCRC Workgroup for Surgery and Medical Sub-Specialty Palliative Care, and the National Academy of Medicine. This project is feasible because of the investigators' multidisciplinary expertise, the rich critical care and palliative care environment, and the support of the institutions involved.

Project narrative (relevance statement) The quality of intensive care unit (ICU)-based palliative care is highly variable, in part because it is difficult to reliably deliver the right care to the right patient at the right time. We propose a study in which we will first describe patients' and families' needs, in terms of type and severity. Next, we will use this rich knowledge (plus the insights of physicians, nurses, social workers, and chaplains) to design a personalizable care delivery model that can align ICU and palliative care clinicians' skills with the needs of patients and families.