VA Disability Benefits: The Federal Data Behind 5.5 Million Compensation Recipients and $130 Billion in Annual Spending

The Department of Veterans Affairs administers one of the largest benefit systems in the federal government. As of 2024, approximately 5.5 million veterans received monthly disability compensation — up from 3.5 million in 2010 — and total VA benefits spending exceeded $130 billion per year. Behind those numbers sits a layered data infrastructure spanning rating schedules, claims management systems, education benefit tracking, and loan guaranty records, much of it publicly accessible through the VA Open Data portal.

The Department of Veterans Affairs: scale and structure

The VA serves approximately 22 million living veterans, of whom roughly 9 million are enrolled in VA healthcare. The department is organized into three main administrations with distinct missions. The Veterans Benefits Administration (VBA) adjudicates and administers disability compensation claims, pension programs, education benefits, vocational rehabilitation, and the home loan guaranty program. The Veterans Health Administration (VHA) operates the largest integrated health care system in the United States — 170 medical centers and more than 1,000 outpatient sites — serving about 6 million patients annually. The National Cemetery Administration (NCA) maintains 155 national cemeteries.

The three main benefit programs administered by VBA are disability compensation, pension, and education and training. Disability compensation provides tax-free monthly payments to veterans with service-connected injuries or illnesses. Pension provides income support to wartime veterans with low income and limited assets who are permanently and totally disabled or aged 65 and older. Education and training benefits — primarily the GI Bill programs — fund post-service education and vocational training. Additional programs include the VA Home Loan Guaranty, Vocational Rehabilitation and Employment (Chapter 31), Survivors and Dependents' Educational Assistance (Chapter 35), and life insurance products.

Total VA budget authority in FY2024 was approximately $370 billion, of which mandatory spending for benefits (compensation, pension, education) accounted for more than $130 billion. VHA medical care discretionary appropriations added roughly $110 billion, with the remainder covering construction, information technology, and administration. The VA is consistently among the three largest federal agencies by budget.

The disability rating system

Disability compensation eligibility and payment amount are determined by a disability rating — a percentage from 0 to 100, assigned in 10-percentage-point increments — that reflects the severity of service-connected conditions. The rating is based on the VA's Schedule for Rating Disabilities (VASRD), codified at 38 CFR Part 4, which specifies diagnostic codes and rating criteria for hundreds of physical and mental health conditions.

When a veteran has multiple service-connected conditions, the ratings are not simply added together. The VA uses the “whole-person” combined ratings formula: the most severe disability is applied first, reducing the hypothetical whole person from 100 to the remaining percentage, and each subsequent disability is then rated against that remaining healthy percentage. A veteran rated 60% for one condition and 40% for another is not rated at 100%: the combined rating is 60% plus 40% of the remaining 40% (16 percentage points), yielding 76%, which rounds to 80% under VA rounding rules. This formula means that reaching a combined 100% rating requires either a single condition rated at 100% or an accumulation of conditions that mathematically exhaust the remaining whole person.

Monthly compensation rates are set by statute and adjusted annually for cost-of-living increases tied to Social Security COLA. For 2024 the base rates for a single veteran with no dependents were approximately: 10% — $171/month; 20% — $338/month; 30% — $524/month; 40% — $755/month; 50% — $1,075/month; 60% — $1,362/month; 70% — $1,716/month; 80% — $1,995/month; 90% — $2,241/month; 100% — $3,737/month. Veterans with dependents (spouses, children, or parents receiving aid and attendance) receive additional amounts on top of the base rate.

Special Monthly Compensation (SMC) provides payments above the 100% rate for veterans with severe disabilities not fully captured by the standard rating schedule. SMC categories cover loss or loss of use of a limb (SMC-K through SMC-N), blindness, the need for regular aid and attendance from another person, and combinations thereof. SMC rates can substantially exceed the standard 100% rate — SMC-T for veterans who would require hospitalization without in-home care reaches over $9,000 per month for some configurations of dependents and ratings.

The total number of veterans receiving disability compensation has grown steadily. In FY2010, approximately 3.5 million veterans received compensation; by FY2024 that figure reached roughly 5.5 million, representing about 25% of all living veterans. The growth reflects the post-9/11 veteran cohort aging into the claims system, expanded presumptive conditions, and the claims surge triggered by the PACT Act of 2022.

Claims processing and the historic backlog

VBA adjudicates disability rating claims through a process that begins with the veteran (or an accredited representative) filing a claim, proceeds through evidence gathering and a Compensation and Pension (C&P) examination, and concludes with a rating decision. C&P exams are ordered from VA medical centers or from contract examination vendors — QTC Medical Group (owned by Leidos), LHI (OptumServe), and Veterans Evaluation Services (VES) are the largest contractors — who conduct the physical or mental health examination and provide an opinion on the nexus between the claimed condition and military service.

Claims processing speed became a national crisis in the early 2010s. The backlog of claims pending more than 125 days peaked at 884,000 in March 2013, driven by the surge of Iraq and Afghanistan veterans filing claims, a 2010 Agent Orange presumptive expansion, and outdated paper-based processing. VBA responded with the Veterans Benefits Management System (VBMS), a digital claims processing platform that replaced paper folders with electronic records, and with the Fully Developed Claims program that incentivizes veterans to submit all evidence upfront to accelerate processing.

The backlog fell below 100,000 by 2015, but claims volume increased again after the PACT Act of 2022 opened eligibility to millions of additional veterans. VA publishes weekly claims pending and timeliness data on its Benefits Dashboard at va.gov/data, including average days to complete a rating decision by claim type. As of mid-2024, VBA was processing roughly 1.5 million claims per year and the pending inventory again approached 300,000.

The Veterans Appeals Improvement and Modernization Act of 2017 (Appeals Reform Act) restructured the appeals process, which had been a major driver of delays. Before 2017, all appeals went to the Board of Veterans' Appeals (BVA) in Washington, creating a multi-year backlog. The 2017 law created three review lanes: a Supplemental Claim lane for new evidence; a Higher-Level Review lane for a senior VBA adjudicator to review the original decision without new evidence; and direct appeal to the BVA, which itself offers three tracks (direct review, evidence submission, and hearing request). The Rapid Appeals Modernization Program (RAMP) piloted the new lanes before full implementation. BVA hearing wait times and appeal outcomes are published in annual BVA reports and on the VA Open Data portal.

The PACT Act of 2022

The Sergeant First Class Heath Robinson Honoring our Promise to Address Comprehensive Toxics (PACT) Act was signed into law in August 2022. It represents the most significant expansion of veteran benefits eligibility since the Agent Orange Act of 1991. The central mechanism is the addition of presumptive service connection for toxic exposure conditions — meaning veterans do not need to prove their condition was caused by military service if it falls within a designated presumptive category.

The PACT Act addressed four primary exposure categories. Burn pit and airborne hazard exposure for veterans who served in Southwest Asia (Iraq, Afghanistan, Kuwait, Bahrain, Gulf War theater broadly) after August 1990 or in any combat zone after September 11, 2001: the law added 23 presumptive conditions including constrictive bronchiolitis, glioblastoma, bladder cancer, and other rare cancers associated with particulate matter exposure from open-air waste burn pits. Agent Orange exposure was expanded to include veterans who served in additional locations (Thailand, Laos, Cambodia, Guam, American Samoa) not previously covered by the presumption. Radiation exposure presumptives were extended to additional military occupational specialties. Camp Lejeune water contamination from 1953 to 1987 became a separate basis for disability claims covering 15 designated conditions.

The Congressional Budget Office estimated the PACT Act would cost approximately $280 billion over 10 years, primarily from newly eligible compensation recipients. The VA received more than 2 million PACT Act-related claims in the first year after enactment. The act also expanded VA health care eligibility to an estimated 3.5 million additional veterans who had not previously qualified for VA enrollment, specifically post-9/11 combat veterans who had been turned away because they lacked a service- connected disability. The data effects are visible in VA's Benefits Utilization data: average monthly applications for compensation roughly doubled in 2022–2023 compared to pre-PACT baselines.

GI Bill and education benefits

The Post-9/11 GI Bill (Chapter 33), enacted in 2008, is the primary education benefit for veterans who served after September 10, 2001. Eligible veterans can receive up to 36 months of education benefits, covering: actual tuition and fees for public in-state schools (or up to a private school cap — $27,120.05/year for FY2024) for private institutions; a monthly housing allowance (MHA) equal to the Basic Allowance for Housing (BAH) at the E-5 with dependents rate for the ZIP code of the school; and an annual books and supplies stipend of $1,000 (prorated by enrollment rate). Benefits are prorated based on length of service, from 40% at 90 days to 100% at 36 months or more.

Veterans can transfer unused Post-9/11 GI Bill entitlement to spouses or dependent children after completing six years of service with a commitment to serve four additional years. The transfer-of-entitlement program has become a significant military retention tool and means a growing share of Post-9/11 GI Bill users are dependents rather than veterans themselves.

The Yellow Ribbon Program fills gaps between the private school cap and actual tuition at participating private and out-of-state institutions. Schools agree to contribute a portion of tuition above the VA cap, and VA matches the school's contribution dollar-for-dollar, potentially covering full tuition. Not all schools participate, and schools cap the number of Yellow Ribbon slots. The VA's GI Bill Comparison Tool at va.gov/education/gi-bill-comparison-tool allows users to search participating schools, Yellow Ribbon amounts, and graduation and retention rates.

The Montgomery GI Bill (Chapter 30) is the older analog, requiring veterans to contribute $100 per month for 12 months during service in exchange for a flat monthly stipend rather than tuition payment and housing allowance. Chapter 30 rates are substantially lower than Post-9/11 GI Bill benefits for most beneficiaries, and most eligible veterans elect the Post-9/11 program when given the option.

Vocational Rehabilitation and Employment (VR&E, Chapter 31) serves veterans with service-connected disabilities who need vocational training, independent living services, or employment assistance. Unlike Chapter 33, which the veteran controls, Chapter 31 is administered through a rehabilitation plan developed with a VA case manager. Total expenditure across all GI Bill and education programs exceeds $10 billion per year. VA publishes quarterly GI Bill enrollment and expenditure statistics disaggregated by program, school, and state.

VA Home Loan Guaranty

The VA Home Loan Guaranty program, established by the Servicemen's Readjustment Act of 1944, enables eligible veterans, active-duty servicemembers, and surviving spouses to obtain mortgages without a down payment and without private mortgage insurance (PMI). The VA does not directly lend money but instead guarantees a portion of the loan to private lenders, reducing lender risk and enabling more favorable terms. In FY2022 the program guaranteed more than 4 million home loans cumulatively, with annual new guaranties running around 750,000 to 900,000 loans per year.

Loan limits for VA-backed loans without a down payment were tied to conforming loan limits prior to 2020; the Blue Water Navy Vietnam Veterans Act of 2019 eliminated loan limits for veterans with full entitlement, meaning veterans can borrow above conforming limits without a down payment if the lender is willing. The 2024 conforming loan limit of $766,550 ($1,149,825 in high-cost areas) still functions as a practical ceiling for most transactions. Veterans with partial entitlement (due to existing VA loans or prior foreclosures) are subject to county-level limits.

In lieu of PMI, VA loans carry a funding fee — a one-time upfront charge rolled into the loan or paid at closing — that helps offset program costs. For purchase loans with no down payment, the funding fee is 2.15% for first-time use and 3.3% for subsequent use for regular military; it drops to 1.25% if the veteran makes a down payment of 10% or more. Veterans receiving disability compensation at any rating are exempt from the funding fee, as are surviving spouses of veterans who died in service or from service-connected causes.

The Native American Direct Loan (NADL) program allows eligible Native American veterans to use their VA home loan benefit to purchase, construct, or improve homes on federal trust land. Unlike the standard guaranty program, NADL involves a direct loan from the VA rather than a guaranty to a private lender.

VA loan data is accessible in two places. The VA Loan Policy Manual documents program rules. For large-scale analysis, the Home Mortgage Disclosure Act (HMDA) dataset reported by financial institutions to the CFPB includes a loan type field where loan_type = 2 identifies VA-guaranteed loans, enabling analysis of VA loan origination volume, denial rates, interest rates, and demographics by lender, geography, and borrower characteristics across the entire mortgage market.

Open data: VA.gov/data and the VA Open Data portal

The VA maintains several public data resources relevant to benefits analysis. The VA Open Data portal at data.va.gov, built on the Socrata platform, hosts datasets covering VBA benefits utilization by state (compensation recipients, average disability rating, total expenditure), VHA utilization by Veterans Integrated Service Network (VISN), NCA cemetery data, and appeal disposition statistics. Most datasets are accessible via the Socrata REST API and OData endpoints without authentication.

The VBA Benefits Utilization data cube is particularly useful for policy research. It provides state-level annual counts of compensation recipients broken down by disability rating band (0–10%, 10–30%, 30–50%, 50–70%, 70–100%), average monthly compensation payment, and total annual expenditure. Cross-referencing these figures with Census Bureau veteran population estimates from the American Community Survey (Table B21001) yields a utilization rate — the share of veterans in each state receiving compensation — that varies substantially across states due to differences in veteran age distribution, discharge era, and awareness of VA services.

VA suicide prevention data are published in the National Veteran Suicide Prevention Annual Report, which draws on mortality data from the Centers for Disease Control and Prevention's National Death Index linked to VA and DoD records. The report provides age-adjusted suicide rates for veterans compared to the general population and trends over time. VA's MMWR (Morbidity and Mortality Weekly Report) supplements also cover veteran mortality by cause. Veterans are at approximately 1.5 times the age-adjusted civilian suicide rate, a figure that has driven substantial VA and congressional investment in mental health and crisis services.

Community care referral data under the MISSION Act became publicly available through VA's Open Data portal beginning in 2020. The MISSION Act of 2018 expanded veterans' ability to access non-VA providers when VA care is unavailable or inconvenient; the referral data documents the volume of authorizations by VISN, specialty, and community care program, enabling analysis of how much care is being purchased outside the VA system.

Veterans Service Organizations and claims representation

Filing a VA disability claim without representation is technically straightforward but practically difficult. The rating schedule, nexus requirements, claims development obligations, and appeals process are complex enough that the VA accredits representatives to assist veterans. VA-accredited claims agents and attorneys are listed in the VA's Office of General Counsel Accreditation Search. Veterans Service Organizations (VSOs) — including the Disabled American Veterans (DAV), the American Legion, the Veterans of Foreign Wars (VFW), and Vietnam Veterans of America — employ accredited service officers who provide free claims assistance to veteran members and in many cases to any veteran.

A common strategy for complex claims is obtaining a private nexus letter — an opinion from an independent physician who has reviewed the veteran's service records and medical history and opines that the claimed condition is “at least as likely as not” caused or aggravated by military service. This standard of proof is the preponderance of evidence threshold (50% probability or greater) required for service connection. A well-documented nexus letter from a credentialed physician specializing in VA disability evaluations can significantly increase the probability of a favorable rating decision.

Two specialized rating outcomes deserve particular attention. Total Disability based on Individual Unemployability (TDIU) allows veterans rated below 100% — typically at 60% for a single condition or 70% combined — to receive compensation at the 100% pay rate if their service-connected conditions prevent them from maintaining substantially gainful employment. TDIU recipients numbered approximately 370,000 in FY2024. Service connection versus non-service-connected pension is a distinct determination: pension provides income support to wartime veterans with low income and limited assets regardless of disability etiology, while compensation requires proof that the disability arose from or was aggravated by military service.

Mental health and Military Sexual Trauma data

VA mental health programs treated more than 2 million veterans annually as of 2023, covering outpatient therapy, inpatient psychiatric care, substance use disorder treatment, and community-based mental health services. Post-traumatic stress disorder (PTSD) is the most prevalent mental health diagnosis in the VA system. VA and DoD studies estimate PTSD prevalence among Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF) veterans at 11–20%, compared to roughly 7% lifetime prevalence in the general adult population.

Military Sexual Trauma (MST) — defined by VA as sexual assault or sexual harassment experienced during military service — is an independent basis for a PTSD or other mental health condition disability claim without requiring documented in-service records of the incident. VA policy allows claims based on MST to be substantiated through a broader range of indirect evidence (behavioral changes noted in service records, performance evaluations, buddy statements, civilian medical records) because the underreporting of MST during service is well-documented. VA screens all enrolled veterans for MST history at initial enrollment and annually thereafter. The VA publishes annual MST summary reports covering the proportion of MST-positive veterans by gender, era of service, and utilization of MST-related mental health services.

The MISSION Act community care expansion has affected mental health access significantly: veterans who face long wait times for VA mental health appointments can now be referred to community providers under community care authorizations. VA publishes community care authorization volumes by specialty, with mental health representing one of the largest community care categories by referral count.

Python: state-level disability compensation utilization analysis

The script below pulls the VA Benefits Utilization State Data from the VA Open Data portal via Socrata, merges it with Census ACS veteran population estimates, computes disability compensation recipients per 10,000 veterans, ranks states, and reports the weighted average monthly payment nationally. The VA Socrata API requires no authentication for public datasets.

import requests
import pandas as pd

# -------------------------------------------------------
# VA Benefits Utilization: State-Level Disability Analysis
# Downloads the VA Benefits Utilization State Data from the
# VA Open Data portal (Socrata), computes disability
# compensation recipients per 10,000 veterans (using Census
# ACS veteran population estimates), ranks states, and
# compares average monthly payment across states.
# -------------------------------------------------------

# VA Open Data portal Socrata dataset IDs
# "VBA Benefits Utilization by State" (compensation recipients, avg rating)
VA_BENEFITS_DATASET = "jubf-tpaw"
VA_BASE_URL = f"https://data.va.gov/api/odata/v4/{VA_BENEFITS_DATASET}"

# Census ACS 1-Year estimates for veteran population by state
# Table B21001 from the Census Bureau API (no key required for small queries)
CENSUS_API = "https://api.census.gov/data/2022/acs/acs1"


def fetch_va_benefits(limit=200):
    """Fetch VA benefits utilization data from the VA Open Data Socrata API."""
    url = f"https://data.va.gov/resource/{VA_BENEFITS_DATASET}.json"
    params = {"$limit": limit, "$order": "fiscal_year DESC"}
    resp = requests.get(url, params=params, timeout=60)
    resp.raise_for_status()
    return pd.DataFrame(resp.json())


def fetch_census_veteran_population():
    """Fetch state veteran population from Census ACS via the Census Bureau API."""
    params = {
        "get": "NAME,B21001_002E",   # B21001_002E = civilian veterans 18+
        "for": "state:*",
    }
    resp = requests.get(CENSUS_API, params=params, timeout=60)
    resp.raise_for_status()
    rows = resp.json()
    headers = rows[0]
    data = rows[1:]
    df = pd.DataFrame(data, columns=headers)
    df = df.rename(columns={"NAME": "state_name", "B21001_002E": "veteran_pop"})
    df["veteran_pop"] = pd.to_numeric(df["veteran_pop"], errors="coerce")
    return df[["state_name", "veteran_pop"]]


# -------------------------------------------------------
# Step 1: Pull VA benefits data
# -------------------------------------------------------
print("Downloading VA benefits utilization data ...")
va = fetch_va_benefits()
va.columns = [c.strip().lower().replace(" ", "_") for c in va.columns]
print(f"Rows fetched: {len(va)}")
print("Columns:", list(va.columns))

# Identify key columns (names vary by dataset vintage)
state_col = next((c for c in va.columns if "state" in c and "code" not in c), None)
recipients_col = next((c for c in va.columns if "recipient" in c or "veteran" in c), None)
payment_col = next((c for c in va.columns if "payment" in c or "compensation" in c or "amount" in c), None)
rating_col = next((c for c in va.columns if "rating" in c or "average" in c), None)
year_col = next((c for c in va.columns if "year" in c or "fiscal" in c), None)

print(f"Using columns -> state: {state_col}, recipients: {recipients_col}, "
      f"avg_payment: {payment_col}, avg_rating: {rating_col}, year: {year_col}")

if not all([state_col, recipients_col]):
    raise ValueError("Cannot identify required columns; inspect columns above and adjust.")

# Keep most recent fiscal year
if year_col:
    latest_year = va[year_col].max()
    va = va[va[year_col] == latest_year].copy()
    print(f"Filtered to fiscal year: {latest_year}")

va_state = va.rename(columns={
    state_col: "state_name",
    recipients_col: "comp_recipients",
})[["state_name", "comp_recipients"]].copy()

if payment_col:
    va_state["avg_monthly_payment"] = pd.to_numeric(va[payment_col], errors="coerce")
if rating_col:
    va_state["avg_rating"] = pd.to_numeric(va[rating_col], errors="coerce")

va_state["comp_recipients"] = pd.to_numeric(va_state["comp_recipients"], errors="coerce")

# -------------------------------------------------------
# Step 2: Pull Census veteran population estimates
# -------------------------------------------------------
print("\nDownloading Census ACS veteran population estimates ...")
census = fetch_census_veteran_population()
print(f"States from Census: {len(census)}")

# -------------------------------------------------------
# Step 3: Merge and compute utilization rate
# -------------------------------------------------------
merged = va_state.merge(census, on="state_name", how="inner")
merged = merged[merged["veteran_pop"] > 0].copy()

# Recipients per 10,000 veterans (enrollment/utilization rate proxy)
merged["recipients_per_10k"] = (
    merged["comp_recipients"] / merged["veteran_pop"] * 10000
).round(1)

# -------------------------------------------------------
# Step 4: Rank states by utilization rate
# -------------------------------------------------------
ranked = merged.sort_values("recipients_per_10k", ascending=False).reset_index(drop=True)
ranked.index += 1

print("\nTop 15 states: disability compensation recipients per 10,000 veterans")
display_cols = ["state_name", "comp_recipients", "veteran_pop", "recipients_per_10k"]
if "avg_monthly_payment" in ranked.columns:
    display_cols.append("avg_monthly_payment")
if "avg_rating" in ranked.columns:
    display_cols.append("avg_rating")
print(ranked[display_cols].head(15).to_string())

print("\nBottom 10 states: disability compensation recipients per 10,000 veterans")
print(ranked[display_cols].tail(10).to_string())

# -------------------------------------------------------
# Step 5: National summary
# -------------------------------------------------------
total_recipients = int(merged["comp_recipients"].sum())
total_vets = int(merged["veteran_pop"].sum())
national_rate = round(total_recipients / total_vets * 10000, 1)

print(f"\nNational totals (from merged dataset):")
print(f"  Compensation recipients: {total_recipients:,}")
print(f"  Veteran population (ACS): {total_vets:,}")
print(f"  National rate: {national_rate} per 10,000 veterans")

if "avg_monthly_payment" in ranked.columns:
    wtd_avg = (
        (merged["comp_recipients"] * merged["avg_monthly_payment"]).sum()
        / merged["comp_recipients"].sum()
    )
    print("  Weighted avg monthly payment: $" + str(round(wtd_avg, 2)))

print("\nDone.")

The output surfaces two distinct patterns. States with large post-9/11 veteran populations (Virginia, Colorado, Washington) tend to have higher utilization rates because younger veterans file at higher rates than older cohorts accustomed to avoiding the VA. States with older veteran populations dominated by World War II and Korea-era veterans show lower rates because those generations filed at lower rates historically. The average monthly payment correlates with the proportion of higher-rated veterans, which itself correlates with the era of service — combat veterans from Iraq and Afghanistan accumulate more rated conditions from blast injuries, musculoskeletal trauma, and PTSD than peacetime-era veterans.

Cross-referencing with HMDA data (loan_type = 2) adds a housing dimension: states with high VA loan origination volumes relative to veteran population reveal where VA home loan awareness and usage are highest, which often correlates with the presence of large active-duty installations driving veteran settlement patterns post-separation.

The data landscape going forward

Several trends will reshape VA benefits data in the coming years. The PACT Act claims wave is still working through the system: tens of thousands of claims filed in 2022 and 2023 remain in adjudication or appeal. As those resolve, the compensation recipient count and total expenditure will increase substantially, with VA projecting the total compensation caseload could reach 7 million by 2030.

VBA's modernization of its IT infrastructure — replacing the legacy Benefits Delivery Network with VBMS and expanding the Veteran-Facing Services platform — is producing richer real-time data on claims status that is increasingly exposed through the va.gov/data portal. The direct submission API used by VSOs to file claims electronically (the Benefits Claims API, part of VA's Lighthouse platform at developer.va.gov) generates machine-readable claims submission records that will eventually become a new data source for claims volume and processing time research.

The intersection of VA disability compensation data with Social Security Administration data (SSI/SSDI recipients who are also veterans), Medicare data (dual users of VA and Medicare), and Census demographic data remains an active area of federal research conducted primarily through the VA's Office of Mental Health and Suicide Prevention and the National Center for Veterans Analysis and Statistics (NCVAS). NCVAS publishes annual state summaries with veteran population projections through 2045, demographic breakdowns by age, gender, and period of service, and estimates of VA-enrolled versus unmet-need veteran populations — all downloadable from va.gov/vetdata.