Population-Based Data Core

University Of Colorado Denver, Aurora CO

Investigators

Abstract

ASCENT: POPULATION-BASED DATA CORE PROJECT SUMMARY/ABSTRACT Existing population-based data offer an opportunity to expand our understanding of palliative care (PC) across the lifespan. Research involving primary data collection from persons with serious illness (PWSI) is challenging due to high symptom burden, caregiver strain, and disease-related functional and cognitive decline. Results from individuals well enough to participate in clinical trials may not be generalizable to other PWSI. Nationally representative rich datasets can be used to strengthen the evidence base for caring for PWSI. To understand the best ways to provide and finance PC, there is a need to identify existing data sources that can be used to analyze populations for whom PC is relevant, measure PC interventions, patient-centered outcomes and caregiver experiences, and estimate associated costs. To efficiently leverage existing survey, claims, electronic health record (EHR), registry, provider, and neighborhood data requires expertise in data management, merging, cleaning, and linkage; cohort identification; maintenance of secure analytic files; and accounting for endogenous treatment selection in observational data, where treatments and policies cannot be randomly assigned. As part of the Advancing the Science of Palliative Care Research across the Lifespan (ASCENT) Consortium methods cluster, the Population-Based Data Core (Pop Core) will provide training and resources to support research on existing data sources that will advance our understanding of PC delivery, financing and outcomes across the lifespan. The Pop Core’s aims are to 1) Support the design and conduct of rigorous PC research across the lifespan using existing datasets, including administrative and contextual data; 2) Enhance capacity of PC researchers to design and conduct rigorous PC research using existing datasets through development of guidelines on procedures to access and use EHR and claims data for ASCENT Pilot, Research Scholar and developmental projects and by sharing best practices for accounting for confounding in observational studies and working with PC-relevant cost data, and 3) Generate new knowledge and disseminate research findings and best practices in PC research using existing datasets by characterizing the presence of potential time-varying confounders in panel studies, developing a directory of secondary data sources relevant to PC, and disseminating code to operationalize key PC concepts in EHR and claims data and to identify appropriate samples for PC research that can be leveraged in future studies. Working collaboratively across ASCENT, the Pop Core will advance the rigorous study of PC for PWSI across the lifespan by educating the next generation of PC researchers, developing new knowledge to facilitate analyses of observational data in PC, and facilitating the use of population-based data in PC research.