4020 — GECDAC Core Files: A Data Infrastructure Resource

Lead/Presenter: Sharon Dally,  geriatrics and extended care data analysis center
All Authors: Dally s (Palo Alto VAMC), Intrator, O (Canandaigua VAMC) Phibbs, C. (Palo Alto) Kinosian, B (Philadelphia VAMC)

Objectives:
The Office of Geriatrics and Extended Care (GEC) is responsible for providing services along a continuum of care extending from Ambulatory Care and Home and Community Based Long-Term Services and Supports to Nursing Home care and End-of-Life care. The Geriatrics and Extended Care Data Analysis Center (GECDAC) collects and analyzes population-based data about GEC programs and services and facilitates and provides support to affiliated researchers studying healthcare and services for aging Veterans. The GECDAC Core Files (GCF) were created to consolidate information from multiple sources to facilitate a variety of analyses investigating utilization patterns, costs, and outcomes of GEC services.

Methods:
A variety of VHA data sources, including the Corporate Data Warehouse (CDW), Managerial Cost Accounting System National Data Extracts, Community Care claims, the Home-Based Primary Care (HBPC) Master File, the Vital Status File and Planning Systems Support Group (PSSG) Geocoded Enrollment Files as well as Medicare claims files, beneficiary summary files, and Minimum Data Set assessment files are accessed to create annual datasets containing over 1300 patient-level variables specific to a given fiscal year.

Results:
GCF fiscal-year files have been created for FY2010-FY2021. These SAS (Statistical Analysis System) files include all VHA users in the given fiscal year, regardless of utilization of GEC services. Variable categories include: demographics; enrollment (VA and Medicare); over 100 different health condition indicators (medical and mental health); utilization and costs of GEC services; utilization of acute care, long-term institutionalization, and end-of-life care; risk indicators such as the JEN Frailty Index (JFI), Care Assessment Need (CAN) scores, Hierarchical Condition Categories (HCC) scores, Nosos scores, High-Need High Risk, and Predictive Long-Term Institutionalization; and total annual VHA and Medicare costs. Many of these variables are calculated two ways, using only VHA data and using blended VHA and Medicare data.

Implications:
The GCF datasets are available for sharing with VA researchers through a Data Access Request Tracker (DART) request. The shared data can allow researchers to more easily analyze GEC program utilization and costs. A guidebook describing how each variable was derived is also available. Inclusion of Medicare data is particularly valuable in providing a more complete picture of conditions and utilization patterns for older veterans, since most veterans over age 65 are eligible for Medicare.

Impacts:
The GCF datasets provide a tool for researchers to more easily include GEC services data in VA research and hopefully improve the quality of GEC programs and services