BlackLeafwatch the watchmen
VA community care eligibility determinations

VA got community care eligibility wrong in both directions

Summary

A VA Office of Inspector General audit of the department's community care program -- which sends veterans to private providers when VA facilities can't meet wait-time or drive-time standards -- found the eligibility process failing two opposite ways at once. An estimated 339,633 consults were sent to costly outside providers even though the veteran didn't qualify, projecting to $1.7 billion a year in questioned costs. At the same time, a larger group -- an estimated 1,066,497 appointments -- involved veterans who WERE eligible for community care but were kept within VA with no record they were ever told they had that choice.

By Nero · July 13, 2026

The Office of Inspector General audited how accurately medical facilities determine which veterans qualify for community care -- treatment from a private provider, paid for by , when a facility can't meet wait-time or drive-time access standards. The audit found the system failing in both directions during the same quarter: an estimated 339,633 consults for mental health, primary, or specialty care were sent to outside providers even though the veteran didn't actually qualify, while a larger group -- an estimated 1,066,497 appointments -- involved veterans who WERE eligible for community care but were kept within with no record they were ever informed they had that option.

The cost of sending the wrong veterans outside VA

About 25% of the roughly 1.4 million community care consults reviewed for Q1 fiscal year 2025 (October-December 2024) went to providers outside despite the veteran not meeting eligibility criteria under the MISSION Act. 's statisticians put the quarter's associated cost at $436.2 million, with a 90% confidence range of $380.8 million to $490.5 million. Multiplied by four to represent a full fiscal year, that comes to $1.7 billion in questioned costs -- money paid to private providers for care that, on 's own criteria, should often have been available at a facility instead.

Questioned costs, annualized from Q1 FY2025
$1.7B
$1,744,845,232 -- four times OIG's $436.2 million quarterly estimate for consults incorrectly approved for outside care
Community care consults sent outside VA despite ineligibility
25%
an estimated 339,633 of about 1.4 million mental health, primary, and specialty care consults
Eligible veterans kept at VA with no record they were informed
58%
an estimated 1,066,497 of 1.8 million appointments where the veteran actually qualified for community care
$436 million in one quarter, with a real range around it
OIG's 90% confidence range for Q1 FY2025 costs tied to incorrectly-approved community care
Lower limit
380,809,319
Point estimate
436,211,308
Upper limit
490,511,486
Source: VA OIG, Audit 25-01014-139, Table C.9
View data as table
OIG's statistical estimate of Q1 FY2025 (October-December 2024) community care claims payments tied to incorrectly-approved-as-eligible consults, based on a sample of 103 of 214 relevant consults with a 90% confidence interval. OIG multiplied this quarterly point estimate by 4 to arrive at its headline $1.7 billion annualized questioned-cost figure.
Lower limit380,809,319
Point estimate436,211,308
Upper limit490,511,486

The bigger failure: eligible veterans who were never told

The larger problem by volume runs the other way -- though it's a different kind of error, not a dollar figure to add to the first one. Of 4.8 million appointments potentially eligible for community care, an estimated 1.8 million involved a veteran who genuinely qualified but was seen within instead -- with no documentation that staff had even assessed their eligibility first. Of those 1.8 million, 58% (about 1.07 million) have no record the veteran was informed of their community care option or chose to opt out, as policy requires. doesn't attach a dollar estimate to this population the way it does the $1.7 billion figure -- it's a documentation gap, not a costed payment error. But by raw count, that population of eligible-but-uninformed veterans is roughly 3.1 times larger than the ineligible-but-referred-out population, meaning the more common failure, by volume, is not telling veterans about a choice they were entitled to make.

The eligibility system failed in both directions at once
Estimated appointments/consults mishandled each way, Q1 FY2025
Sent to costly outside care despite being ineligible
339,633
Kept at VA, eligible, and never told they had a choice
1,066,497
Source: VA OIG, Audit 25-01014-139, Tables C.7 and C.12
View data as table
VA's community care eligibility process failed in two opposite directions during the same quarter -- but these are two different kinds of error, not directly comparable dollar-for-dollar. The 339,633 figure is a costed eligibility/payment error (OIG's audited $436.2 million estimate). The 1,066,497 figure is a documentation gap -- no record the veteran was told about or opted out of community care -- that carries no dollar estimate of its own in the report.
Sent to costly outside care despite being ineligible339,633
Kept at VA, eligible, and never told they had a choice1,066,497

For some, the untold choice cost them time

also tested whether the community care veterans DID receive was actually the faster option. Among just over 287,000 wait-time-eligible consults, an estimated 13% (38,100) of veterans would have gotten care faster by taking the next available appointment instead of going to a community provider -- and more than 40% of them waited over a month longer than they would have at . conservatively estimated this added $114.4 million a year in costs, atop the $1.7 billion questioned-cost figure, for care that was both slower and paid to an outside provider. The report is careful to note it cannot know whether these veterans would have chosen care even with full information -- only that they were never given the wait-time data to make that choice themselves.

The takeaway

  • The system errs in both directions, though not in comparable units. $1.7 billion a year in a costed eligibility error (outside care paid for ineligible veterans), and a larger, uncosted population -- 1.07 million appointments -- of eligible veterans never told they had a choice at all.
  • By raw count, the bigger population is the uninformed one, not the overspent one. Eligible-but-uninformed veterans (1.07 million) outnumber ineligible-but-referred-out veterans (339,633) by about 3 to 1 -- but the report attaches a dollar figure only to the smaller population, so this is a volume comparison, not a bigger dollar cost.
  • agreed there's a problem. The Under Secretary for Health concurred with 5 of 's 6 recommendations and concurred in principle with the sixth, but as of the audit's briefings in September and October 2025, VHA leaders confirmed the underlying scheduling-system limitations were still unresolved.

attributes these errors chiefly to guidance from 's Office of Integrated Veteran Care that didn't align with the MISSION Act's eligibility criteria, and to scheduling systems that don't let staff see appointment availability or wait times across all facilities. All figures are statistical projections from stratified random samples (roughly 100-250 cases per sample) with Clopper-Pearson 90% confidence intervals, not a full census of every consult and appointment -- 's own reported ranges are cited alongside the point estimates throughout. This audit covers only Q1 FY2025 (October-December 2024); a companion report, issued the same day, separately examines how long and community care consults took to process.

Sources(1) ▾
  • VA Office of Inspector General, Audit of Veterans' Community Care Eligibility Determinations (Report No. 25-01014-139) (2026-06-17) statistical audit of community care eligibility determinations, mandated by the Senator Elizabeth Dole 21st Century Veterans Healthcare and Benefits Improvement Act, reviewing appointment and community care consult data from Q1 FY2025 (October-December 2024) with fieldwork through February 2026, issued June 17, 2026. Read in full (48 pages) via curl with a browser user agent (WebFetch's initial rendering only surfaced a summary page) and pdftotext -layout, including the statistical sampling appendices. vaoig.gov · original document
Weekly digest: the most-read systems, in brief. Mondays.

Comments

Always open. Logged-in readers can annotate paragraphs in place.

Loading comments…
or log in to comment under your account