Use of VA Benefits and Services: 2021


This page serves as the Appendix for a three-part report on Use of VA Benefits and Services: 2021

Administration of Services

Data Sources

The United States Veterans Eligibility Trends and Statistics (USVETS) is an integrated database of all living and deceased Veterans, including those that do not use any VA benefits. It is comprised of administrative records from the Veterans Benefits Administration (VBA), the Veterans Health Administration (VHA), the National Cemetery Administration (NCA), and the Department of Defense’s (DoD) Defense Manpower Data Center (DMDC). This integrated dataset is also augmented by commercial data sources that include additional geographic and socio-economic information on the Veteran population. VA program data are included in USVETS based on the following criteria:
  • Health Care: All Veterans who receive either VA inpatient care, VA outpatient care, purchased (fee basis) care, VA long-term services and support, or VA pharmacy care were included. VA Health Care enrollees who did not seek care from VA during the current year were not included. Veterans who only sought care from a VHA Vet Center were not included.
  • Compensation & Pension: All Veterans who receive monthly VA Disability Compensation or Pension payments were included. Veterans with pending or denied claims were not included.
  • EducationAll Veterans who received benefits for Chapter 30, 32, 33, 1606, and 1607 education programs were included.
  • Home Loan Guaranty : All Veterans who have an active, new or re-financed VA-guaranteed home loan were included. Veterans who received Special Adaptive Housing benefits were also included in the analysis.
  • Life Insurance: All Veterans who have an active VA life insurance policy or were in receipt of a benefit from a policy that was administered or supervised by VA were included. VA insurance programs included in the analysis were National Service Life Insurance (NSLI), United States Government Life Insurance (USGLI), Veterans’ Special Life Insurance (VSLI), Veterans’ Reopened Insurance (VRI), Service-Disabled Veterans Insurance (S-DVI), Veterans’ Mortgage Life Insurance (VMLI), Traumatic Injury Protection (TSGLI), and Veterans’ Group Life Insurance (VGLI). The analysis does not include active Service-members and spouses/dependents who use Service-members’ Group Life Insurance (SGLI) and Family Service-members’ Group Life Insurance (FSGLI).
Data on headstones and markers from the National Cemetery Administration sourced the information on memorial benefits presented in this report. This data contains details on headstones and markers provided for Veterans in cemeteries around the world, including: 
  • VA national cemeteries,
  • Veterans Cemetery Grants Program funded cemeteries,
  • Private cemeteries,
  • Arlington National Cemetery and other federal cemeteries. 

Methodology and Assumptions

  • A Veteran user is defined as any Veteran who received or used at least one VA benefit or service during the fiscal year. Veteran spouses, Veteran dependents, and active military service members who used VA benefits and services were not included in the analysis. Each Veteran is only counted once in the overall total even if he/she used multiple programs.
  • VA uses a number of methods to uniquely identify Veterans in its data systems, including social security number (SSN), scrambled SSN (SCRSSN) in VHA, claim or file numbers in VBA, and Electronic data interchange personal identifier (EDIPI) in DoD. Because each input source included SSN, records from each source were combined using the Veteran’s SSN as the primary identifier.  Additional information on data integration can be found in USVETS Technical Documentation.
  • Veteran records from each input source were verified against official Social Security Administration data to validate SSN, name, date of birth, and gender.
  • Missing SSN or SSNs determined to be incorrect by SSA were omitted from the analysis.
  • Estimates for the number Veterans who did not use any VA benefits or services were defined as the difference between total population estimates from the Veteran Population Projection Model (VetPop2020) and the number of VA users identified in USVETS. Non-users counts were adjusted to ensure total population estimates were consistent with VetPop.
  • Some data elements were not available for all Veteran records. A minimum of 70% coverage for any particular data element was assumed to be an appropriate sample of all Veteran users to create frequency distributions.  However, at the time of this release, we have not completed an assessment of any potential bias effects as a result of missing data.

Contact Information and Credits

For questions, please email us at
Title photo by Christopher Neugebauer licensed under Creative Commons CC BY-SA 2.0.