The Statistical Core (Stats Core) is an essential component of the National Person-Centered Assessment Resource (PCAR), forming the scientific infrastructure necessary to maintain the high quality and rigorous standards associated with PROMIS, NIH Toolbox, Neuro-QOL, and ASCQ-Me. The Stats Core has two specific aims: 1) To curate the PCAR portfolio;and 2) To equip users with resources they need to conduct high quality patient-centered outcomes research.
Aim 1 will be accomplished by developing, documenting, and disseminating high quality PCAR measurement standards based largely on our prior work, and adding new developments in the field. We will also promote collaborator testing and evaluating of existing and new PCAR items and instruments for inclusion in an expanding PCAR instrument portfolio. Working with the Research Resource, we will conduct dynamic upgrades and enhancements of current and new item banks and performance-based measures. To foster PCAR score interpretability, we will link the metrics of PCAR measures and legacy measures of the same or similar constructs, and update and apply state-of-the-art norming methods (e.g., norm and sub-norm comparisons) for all four PCAR instrument metrics. We will also ensure psychometrically sound and culturally relevant translations of items and instruments in multiple languages to support cross-cultural applicability and global reach of the PCAR measures. In support of Aim 2, we will equip clinicians and researchers with resources that maximize their abilities and confidence in independent and proper use of PCAR instruments and methods. We will develop freeware programs that meet user needs and reduce recurring demands on PCAR statistical personnel. We will also support the Outreach Core to develop condition-specific manuals to guide selection and use of PCAR instruments. We will develop tutorials that increase PCAR user independence in effective, informed, and appropriate use of PCAR measures and methods. We will support and encourage use of the PCAR Dataverse repository of datasets collected by our collaborators and by other researchers willing to share their de-identified data. This repository will enable secondary psychometric analyses and clinical benchmarking in specific diseases and conditions. We will develop a self-sustaining PCAR statistical support team that will maintain the highest PCAR standards long after the conclusion of this award. To do this, we will work with the Admin Core to set up an efficient system for tracking and processing incoming requests, transitioning statistical support activity into a costrecovery recharge model that includes cost recovery associated with maintaining PCAR quality. Requests requiring specialized review of literature, data analysis, or the development of new resources will trigger the development of a contract with the requestor. Though these aims, the Stats Core will build the scientific resources and organize the personnel necessary sustain the National PCAR beyond the funded period.