BETRAYED BY THE VA Thomas Alan Bowman Sr. asked his country for help. He got silence. Sign the demand →
Dear Doug Collins

The Gulf War Illness Claimants: They Won the Shortest War, Then Spent Thirty-Five Years Proving They Were Sick

Their war lasted six weeks. The ground war lasted one hundred hours. Then somewhere between 175,000 and 250,000 of the nearly 697,000 Americans who served in Southwest Asia in 1990 and 1991 came home with an illness that had no name, no test, and no diagnosis — crushing fatigue, widespread pain, failing memory, burning skin, ruined digestion — and spent the next thirty-five years being told, in examination rooms and in denial letters, that there was nothing measurably wrong with them. Congress wrote them a presumption in 1994: if you served there and you are sick with what cannot be explained, the law will presume your service did it. The Department of Veterans Affairs then denied their claims anyway, at rates the Government Accountability Office measured at three times the norm. This page is the accounting of what was done to them, what has been done since, and what — a working presumption, a diagnostic test, a Khamisiyah reckoning — is still owed.

The service and the promise

On August 2, 1990, Iraq invaded Kuwait. The United States built a coalition and a force in the Saudi desert — Operations Desert Shield and Desert Storm — and between August 1990 and June 1991, by VA’s count, about 697,000 men and women served in the theater. The air war opened on January 17, 1991; the ground war ran from February 24 to February 28, and it was, by the standards of American wars, astonishingly brief and astonishingly one-sided. Most of the troops came home to parades.

The promise made to them was the promise made to every American service member: serve, and if service breaks you, the country will carry the cost. For this generation the promise had a specific recent history behind it. Vietnam veterans had spent two decades fighting for recognition that Agent Orange had made them sick, and had finally won presumptions — legal rules under which VA concedes the connection between service and illness instead of making each veteran prove causation one file at a time. The Gulf War generation had watched that fight. What they could not have known is that they were about to live a version of it themselves, against an illness that was harder to pin down than any herbicide: an illness with no single lesion, no laboratory value, and — for its first thirty-four years — no code in the book of diagnoses at all.

They had also been exposed, in the ordinary course of that short war, to an unusual concentration of things whose long-term effects were not well understood: pyridostigmine bromide (PB) pills, issued as a pretreatment against nerve-agent attack; pesticides applied to skin, uniforms, and living areas on a scale the Research Advisory Committee would later call heavy and often excessive; smoke from hundreds of burning Kuwaiti oil wells; and, at one ammunition depot in southern Iraq, the nerve agents sarin and cyclosarin — released not by the enemy, but by American engineers blowing up captured Iraqi stocks after the cease-fire, without knowing what was in them.

What happened

The illness announced itself within a few years of the homecoming, in base towns and VA waiting rooms across the country: veterans in their twenties and thirties who could no longer run, remember, sleep, or digest normally. The symptoms, as GAO would later summarize them, can include fatigue, headaches, joint pain, indigestion, insomnia, respiratory disorders, skin problems, and memory impairment. Deployed veterans reported them at rates far above veterans of the same era who had not deployed. The estimates of how many are affected have been built by three successive official efforts. The congressionally mandated Research Advisory Committee on Gulf War Veterans’ Illnesses, in its November 17, 2008 report, put the figure at 175,000 to 210,000 of the nearly 700,000 deployed, and concluded — in the sentence the veterans had waited seventeen years to hear from an official body — that the extensive body of scientific research “consistently indicates that Gulf War illness is real, that it is the result of neurotoxic exposures during Gulf War deployment, and that few veterans have recovered or substantially improved with time.” The committee found the evidence “strongly and consistently” tied the illness to two exposures in particular: the PB pills and pesticide use. The Institute of Medicine, in its Gulf War and Health series, estimated 250,000 veterans affected by chronic multi-symptom illness — while concluding, in the same body of work, that the evidence was inadequate to tie the illness to any specific exposure. VA’s own research office, citing a 2020 Department of Defense report, puts the range at 175,000 to 250,000. However the count is drawn, between one in four and one in three of everyone who went came home sick with something that would not resolve.

One exposure, at least, could be dated to the day. On March 4 and March 10, 1991 — after the cease-fire — U.S. Army engineers demolished Iraqi ammunition at the Khamisiyah storage site in southern Iraq. Some of the munitions, unmarked, contained the nerve agents sarin and cyclosarin. The Pentagon’s first position was that no troops had been exposed to chemical weapons at all. In June 1996, five years after the demolitions, the Pentagon acknowledged that American soldiers had blown up a bunker at Khamisiyah containing chemical-filled rockets. Computer modeling followed, and in 1997 the Department of Defense sent notification letters to approximately 99,000 veterans telling them they may have been exposed to low levels of nerve agent; a remodeled analysis completed in 2000 identified 100,923 veterans in the potential hazard areas. Men and women who had been sick for years learned from a form letter, half a decade late, that their own side had gassed them — at levels the government characterized as too low to have caused symptoms.

The government then built its exposure science on those models, and the models could not carry the weight. In June 2004, the Government Accountability Office reported (GAO-04-159) that the Defense Department’s conclusions about the extent of troops’ exposure “cannot be adequately supported”: the models were “neither sufficiently certain nor sufficiently precise to draw definitive conclusions about the size or path of the plumes.” GAO gave five reasons — the models were not fully developed for long-range dispersion of chemical agents; the source-term assumptions about the quantity and purity of the agent were inaccurate; plume heights were significantly underestimated; the field testing used to estimate the release did not realistically simulate Khamisiyah; and the individual models diverged widely from one another. GAO’s conclusion was blunt: Defense and VA “cannot know which troops were and which troops were not exposed,” and it recommended that the plume-modeling data not be used as the basis for future epidemiological studies. At a June 2003 hearing, GAO’s chief technologist had already suggested the remedy that advocates — among them Steve Robinson, executive director of the National Gulf War Resource Center — were urging: that VA presume, from the outset, that ill Gulf War veterans in the theater were exposed. No Khamisiyah-specific presumption was ever adopted. The modeling was faulted; the individual burden of proof stayed where it had always been, on the veteran.

What the institution did afterward

Congress moved early — earlier, in fact, than the institutions that would have to carry its intent out. On November 2, 1994, the Persian Gulf War Veterans’ Benefits Act, Title I of the Veterans’ Benefits Improvements Act of 1994 (Public Law 103-446), added 38 U.S.C. § 1117, authorizing VA to pay compensation to any Persian Gulf veteran suffering from a chronic disability resulting from an undiagnosed illness that became manifest during service in the Southwest Asia theater or within a presumptive period afterward. VA implemented the statute at 38 C.F.R. § 3.317. In 1998, the Persian Gulf War Veterans Act (Public Law 105-277) added § 1118, a framework for presumptions based on positive association with toxic agents, environmental hazards, and preventive medications of Gulf War service. In 2010, VA added nine infectious diseases to the list of Gulf War-related conditions. On paper, this was the Agent Orange lesson learned and applied in advance: presume, do not make each sick veteran litigate causation against his own government.

In practice, the presumption was denied at the counter. The first GAO examination of the new law found that, as of July 1995, VA had denied nearly 95 percent of the 4,144 undiagnosed-illness claims it had processed. Twenty years later the numbers had improved only from catastrophic to disgraceful. The oversight pressure was sustained and bipartisan: Representative Christopher Shays of Connecticut ran a long series of subcommittee hearings through the 1990s — at a December 1996 hearing he charged that reports of chemical exposures had been “dismissed, discounted, discredited or denied” — and the House Veterans’ Affairs Committee returned to the claims process again and again, in 2016 hearing VA’s own data put at a 16 percent approval rate and an 84 percent denial rate for undiagnosed-illness and chronic multi-symptom-illness claims.

The definitive audit is GAO-17-511 (July 2017), Gulf War Illness: Improvements Needed for VA to Better Understand, Process, and Communicate Decisions on Claims. Its scope is precise, and this page states it as GAO defined it: GAO analyzed VA’s decisions on Gulf War illness medical issues — the individual medical issues within claims, grouped by VA into undiagnosed illnesses, medically unexplained chronic multi-symptom illnesses, and the nine presumptive infectious diseases — for fiscal years 2010 through 2015. Of about 102,000 such medical issues, VA approved 17 percent. Over the same years, VA approved 57 percent of all other medical issues — about 14 million of 24.7 million. The approval rate for Gulf War illness issues was, in GAO’s words, about three times lower, and it was lower in every year reviewed. By category, as GAO’s analysis reported, undiagnosed-illness issues were approved at 13 percent, chronic multi-symptom-illness issues at 29 percent, and infectious-disease issues at 14 percent. The claims also took longer — on average about a year to complete, some four months longer than other claims, because a Gulf War illness claim carries roughly twice as many medical issues. GAO found the machinery itself unprepared: VA medical examiners told GAO that the range of symptoms made the Gulf War general medical exam challenging, VA offered the examiners only an elective 90-minute web-based training, and only 10 percent of examiners had taken it as of February 2017. Decision letters for denied claims, GAO found, did not always tell the veteran what the law requires them to be told — why. And VA, GAO found, had no plan to develop the single, uniformly used case definition of Gulf War illness that its own 2015 research strategic plan had promised — the missing definition on which consistent adjudication, research, and treatment all depend. GAO also found some 57,000 chronic multi-symptom-illness medical issues in VA’s data that VA had not identified as Gulf War illness-related at all. A later analysis by Veterans for Common Sense and Vietnam Veterans of America of VA decisions from 2002 through the first quarter of 2018 found the same pattern from the veterans’ side of the counter: VA rejected about 90 percent of undiagnosed-illness disability claims.

The auditors kept finding a process that could not reliably decide in either direction. In September 2023, VA’s Office of Inspector General reported that the Veterans Benefits Administration had prematurely approved an estimated 23 percent — about a quarter — of the roughly 13,800 Gulf War undiagnosed-illness and multi-symptom claims decided between October 1, 2021 and March 31, 2022, granting them without the assessments needed to determine whether the requirements were met, with more than $5 million in payments the report questioned — while noting that many of those veterans may well have been eligible. Read together, the GAO and OIG records describe not a gate that swings one way but an adjudication culture that never learned the file: presumptive claims denied for want of proof the law says is unnecessary, other claims granted without proof at all, and the veteran unable to tell, from the letter in his hand, which kind of error he was holding.

Where it stands now

Congress returned to the statute in the PACT Act — the Sergeant First Class Heath Robinson Honoring our PACT Act of 2022, Public Law 117-168, signed August 10, 2022. Section 405 amended § 1117 directly: a qualifying undiagnosed illness or medically unexplained chronic multi-symptom illness may now become manifest to any degree at any time — the old requirements that the disability reach 10 percent within a presumptive period ending December 31, 2026 are gone from the statute, and the definition of Persian Gulf veteran was expanded to additional locations including Afghanistan, Israel, Egypt, Turkey, Syria, and Jordan. VA’s claims guidance implements § 405 on those terms and states that the end date “is no longer applicable,” while the text of § 3.317 has awaited conforming revision. Section 406 of the PACT Act did something different from what it is sometimes described as doing: it enacted 38 U.S.C. § 1120, a presumption of service connection for a list of diseases — respiratory conditions and cancers, asthma among them — associated with exposure to burn pits and other toxins. The nine infectious-disease presumptions were not a PACT Act creation; they sit in § 3.317(c) and date to VA’s 2010 rulemaking. Gulf War veterans can benefit under both frameworks, but they are separate frameworks, and this page keeps them separate.

On October 1, 2025, thirty-four years after the war, Gulf War illness received its own diagnostic code. The code is ICD-10-CM T75.830, “Gulf war illness,” effective with the fiscal year 2026 code set, under the new parent category T75.83, “Effects of war theater” (T75.838 covers effects of other war theaters); like other codes in its family it takes a seventh character for the encounter. The submission was led by Dr. Beatrice Golomb of the University of California San Diego and carried by a consortium of researchers and clinicians; in the words of the university releases announcing it, the code is “long-overdue validation for the suffering of the quarter-million afflicted veterans — and a formal acknowledgment that their illness is real, physical and service-related.” Veterans for Common Sense’s fact sheet on the code explains what it practically does: clinicians and researchers, inside and outside VA, can now identify which veterans have Gulf War illness in their records — for care, for tracking co-occurring conditions, and for testing treatments. What the code does not do, as of this writing in October 2026, is decide claims. VA has issued no adjudication guidance tying T75.830 to § 3.317, and VA has no dedicated disability rating for Gulf War illness itself; claims are still rated symptom by symptom under the undiagnosed-illness and multi-symptom framework. In 2026, advocates — Veterans for Common Sense and Vietnam Veterans of America among them — have pressed VA to close that gap and recognize Gulf War illness as a distinct condition in the rating schedule. That campaign, reported by Stars and Stripes in August 2026, is the current form of a thirty-five-year-old request: believe us on the record you already have.

The science, meanwhile, continues to move toward the objectivity the claims process has always demanded and never had. In 2022, researchers at UT Southwestern led by Dr. Robert Haley published a gene-environment study of 1,116 Gulf War veterans in Environmental Health Perspectives: veterans who reported hearing chemical-warfare alarms in theater — the study’s proxy for low-level sarin exposure — were more likely to have Gulf War illness, and the risk rose stepwise with variants of the PON1 gene that govern how efficiently the body breaks sarin down, reaching roughly nine-fold for exposed veterans with the weakest variant. The researchers stated that the pattern gives a high degree of confidence that sarin is a causative agent; the study does not rule out other exposures in a share of cases, and it is research, not a test VA accepts. There is still no biomarker and no objective diagnostic test for Gulf War illness that VA uses — no blood value a veteran can hand a rater. The illness that a diagnostic code finally names in 2025 is still proven, in 2026, the way it was in 1995: by service in a place, a pattern of symptoms, and a file that someone at VA has to be trained — and willing — to read as the presumption directs.

The ledger

The accounting, in three lines — what the record shows has been done, what it shows has not, and what is still being sought.

What has been done

What has not been done

What is being sought

The record, on video

The Gulf War illness record was built in hearing rooms, in veterans’ own testimony, and in research laboratories — and much of the early hearing video has not survived online. Four films that have: a wartime commander telling a Senate committee in 1997 that no report of chemical exposure ever reached him; a local-news profile of one veteran living with the illness twenty years on; a scholar’s account of what the illness did to her own family and her country’s obligations; and Boston University’s researchers building the stem-cell library the search for a biomarker now runs on.

The commander's testimony. January 29, 1997: retired General Norman Schwarzkopf tells the Senate Veterans' Affairs Committee he never received any report of Iraqi chemical-weapons use or discovery — the institutional position, on the record, months before the Khamisiyah notification letters went out. (AP Archive.)

One veteran, twenty years on. WBIR's "Service & Sacrifice" profile of a Gulf War veteran living with joint pain, memory loss, and chronic fatigue — and with the VA acknowledgment that his symptoms may trace to sarin exposure in theater. (WBIR Channel 10.)

A casualty of war. Dr. Jennifer Gkourlias's TEDxSUNYGeneseo talk on Gulf War illness — what the condition is, what it cost the veterans who carry it, and why recognition took a generation. (TEDx Talks.)

The search for a marker. Boston University's Gulf War Illness Consortium on its stem-cell biorepository — blood cells from ill and well veterans, banked so researchers can hunt the objective measures the claims system has never had. (Boston University School of Public Health.)

Sources

Part of Betrayed, Not Forgotten — this site's accounting for the service members and veterans their own institutions failed. Videos are embedded from their publishers; nothing here is re-hosted.