A Correction, a Deeper Look, and a More Damning Conclusion
A NOTE ON OUR PRIOR REPORTING
In our April 2, 2026 article “The Lasix Question: What The Science Actually Says” we cited the Gross 1999 study as the most powerful evidence of furosemide’s performance-enhancing effect: 22,589 horses, race times 0.56 to 1.09 seconds faster, roughly three to five and a half lengths over six furlongs. We presented HISA’s pending vote as likely to ignore that literature. We were partially wrong on both counts. HISA’s Furosemide Advisory Committee did engage Gross 1999 and offered a legitimate methodological critique that we did not. The science in this story is more complicated than we presented. What follows is a corrected and more complete account. The governance conclusion, however, is more damning than before.
On May 5, 2026, the Horseracing Integrity and Safety Authority announced that its board voted unanimously on April 24 to permit the continued race-day use of furosemide for all covered horses except 2-year-olds and stakes horses. The announcement landed, as predicted, in the direction of the status quo.
What was not predicted by PTW or by most observers, was the scope of what the Furosemide Advisory Committee actually produced. Having now reviewed the full 158-page FAC Report and the Board’s formal Determination, we owe our readers something the racing press rarely offers: a genuine reckoning with what we got right, what we got wrong, and why the full picture is actually worse than our original framing suggested.
WHAT WE GOT WRONG: THE SCIENCE IS MORE SERIOUS THAN WE CREDITED
Our April piece leaned heavily on the Gross 1999 study, the largest observational study conducted at the time, showing meaningful performance differences between horses racing with and without furosemide. We presented that study as a live inconvenient truth that the industry was suppressing.
The FAC engaged it directly, and their critique is legitimate. Gross 1999 was conducted when furosemide use was voluntary — meaning trainers essentially self-selected which horses received the drug. Horses with known EIPH, the most likely candidates for performance problems, were also the most likely to be put on furosemide. That creates severe selection bias: the comparison group was not random. Additionally, at the time of that study, neither dose nor administration timing was strictly regulated. You cannot draw clean causal conclusions from data polluted by that much confounding.
The FAC funded four original research studies at the University of Florida (DeNotta), Washington State (Bayly), Texas A&M (White-Springer and Johnson), and Nationwide Children’s Hospital (Waller) and commissioned additional internal analysis. The Waller study (2025), the most methodologically rigorous examination of actual racing performance, examined North American graded stakes races before and after regulated furosemide bans took effect, using official speed figures as the outcome variable. It found no significant association between furosemide and winning speed figures, finishing position, or percentage of the field beaten at either the group level or in a tightly matched case-crossover analysis of individual horses racing on and off the drug in the same conditions.
The DeNotta placebo-controlled treadmill study found no sustained weight differences, no meaningful electrolyte disruption, no cardiac changes, and no bone density loss though a non-statistically significant directional signal in reduced bone strength improvement warrants the further research the FAC recommended. The Johnson gene expression study found no differences in muscle transcription between furosemide and control horses.
This is real science. The FAC was not a rubber stamp. It funded five studies across multiple institutions, ran a competitive grant process, and produced a document that engages the literature honestly. When we wrote our April piece, the Waller study had not been released publicly. We cited the best available evidence at the time. That evidence has since been updated, and intellectual honesty requires saying so.
The FAC was not a rubber stamp. It produced real science. The question is what the board did with it and what it chose not to ask.
WHAT WE GOT RIGHT: THE ANSWER WAS WRITTEN BEFORE THE QUESTION WAS ASKED
None of that changes the central indictment. In fact, the full document makes it sharper.
Consider what HISA’s board was required by statute to find in order to permit continued Lasix use. Finding #3: “That furosemide has no performance-enhancing effect on individual horses.” That finding was adopted unanimously. But here is the mechanism by which it was reached: the FAC and the board defined “performance-enhancing” narrowly as a drug that builds muscle, increases strength, enhances endurance, acts as a stimulant, or speeds recovery. Under that definition, furosemide is not a PED. The weight-loss mechanism, a horse shedding 20 to 30 pounds of fluid before post was acknowledged in the document but categorized as “small” and “inconsistent across studies,” and therefore excluded from the PED analysis.
There is a Carnegie Mellon study presented at the New England Symposium on Statistics in Sports in September 2025, based on 28,617 matched horses that found furosemide increases a two-year-old’s odds of finishing in the money by 55%. The FAC declined to fund it due to design concerns. They discussed it with the authors. They included it in the report as a non-funded study. And then the board found that furosemide has no performance-enhancing effect.
We are not suggesting fraud. We are observing that when you define your terms narrowly enough, and when you discount the studies that point the wrong direction, and when you rely on the industries you regulate to fund your operating budget you tend to arrive at the answer that those industries require.
“The U.S. Thoroughbred racing industry prioritize animal welfare over political or commercial interests in imposing regulatory policy governing race day use of furosemide.” — FAC Additional Recommendation #5, March 2026. The board then voted to maintain the status quo.
THE 76% NUMBER AND WHAT IT REVEALS
The Board’s Determination contains a statistic that deserves to stand alone for a moment: in 2025, 76% of all covered horses (excluding 2-year-olds) were administered furosemide on race day. That is 25,578 horses.
The board’s reasoning for maintaining the status quo explicitly included this: that a ban would lead to mass retirements, reduced field sizes, cancelled races, “reduced wagering handle,” and “diminished fan engagement.”
Read that again. The regulatory body charged by Congress with the safety and integrity of Thoroughbred racing — the body that exists because the sport’s self-regulatory history failed horses and the public weighed the welfare of 25,578 horses against wagering handle and fan engagement. And wagering handle made the brief.
This is not a minor editorial observation. It is the structural corruption at the heart of the debate. When 76% of your active racing population requires a powerful diuretic four hours before competing in order to safely participate, the question is not whether to permit the drug. The question is what you have bred, trained, and normalized and whether any institution funded by that system is capable of asking it honestly.
The FAC itself knew this. Their Additional Recommendation #5, buried at page 47 of a 158-page document reads: “The U.S. Thoroughbred racing industry prioritize animal welfare over political or commercial interests in imposing regulatory policy governing race day use of furosemide.” They wrote that. They submitted it to the board. The board adopted the findings, ratified the status quo, and cited wagering handle as a relevant consideration.
When 76% of your racing population needs a diuretic to compete, the question isn’t whether to permit the drug. The question is what you’ve bred — and whether the institution built on that system can ask it honestly.
THE INTERNATIONAL RECORD: 158 PAGES AND STILL NO ANSWER
Our April piece asked what we called the question the pro-Lasix establishment has spent fifty years carefully avoiding: if furosemide is a medical necessity, how does the rest of the world race without it?
The FAC Report, for all its genuine rigor, does not answer this question. It acknowledges it in two sentences: furosemide administration on race day is unique to the United States, and there is a “marked bias against US Thoroughbreds” internationally because of it. That is the complete engagement with the international record in 158 pages of scientific analysis funded at $2 million.
The UK does not use race-day furosemide. Australia does not. Japan does not. Germany developed a breeding protocol to reduce the genetic prevalence of EIPH rather than masking it with medication. The FAC report does note, in the environmental section, that “jurisdictions in other countries place a far greater emphasis on the reduction of environmental allergens in stabling facilities” to reduce airway inflammation associated with EIPH. This is a direct validation of the Germany point we raised in April: the rest of the world chose to address the underlying condition. North America chose the syringe.
Not one funded study examined why the EIPH prevalence rate in countries without race-day furosemide is not producing the welfare catastrophe that the horsemen’s advisory group predicted would follow any ban. If EIPH “may exceed 60-70% in racing populations” — the FAC’s own figure — then the UK, Australia, and Japan are racing a population with similar underlying biology and producing no epidemic of bleeding horses retiring en masse. Either their management practices are dramatically superior, or their breeding programs have diverged materially, or the welfare argument for furosemide dependency is less compelling than its financial argument. None of those possibilities were studied.
THE BOARD DEVIATED FROM ITS OWN COMMITTEE — IN THE WRONG DIRECTION
One detail in the Determination has received almost no attention and deserves it. The FAC did not recommend maintaining the current prohibition across all stakes races. It recommended maintaining it only for Graded Stakes races with commercial breeding implications, and explicitly stated that “there is no scientific justification for prohibiting the administration of furosemide in horses competing in Named and Graded stakes races” that the Named Stakes prohibition is “grounded in historic concerns about the marketability of elite North American horses in the international markets.”
In other words: the FAC told the board that the Named Stakes ban is not about science or welfare. It is about sales prices. The board’s response was to keep the broader prohibition anyway, on the grounds of administrative simplicity, clear rules, reduced confusion, no inadvertent noncompliance. They acknowledged no scientific basis for the Named Stakes ban and maintained it for organizational convenience.
This is not the board being more cautious than its committee on welfare grounds. It is the board acknowledging that one of its current rules has no scientific justification and choosing to keep it because simplicity is easier than accuracy. That is not what “horse comes first” looks like as policy.
DOES THE HORSE COME FIRST?
HISA’s stated mission repeated in every press release, every town hall, every announcement, is the safety and wellbeing of horse and rider. It is the justification for the entire federal regulatory apparatus. Let us measure the May 5 announcement against that standard, using the board’s own document.
The board acknowledged that the DeNotta study found a non-statistically-significant but directionally concerning signal in bone strength. Bone strength matters when catastrophic musculoskeletal injury is the leading cause of racing fatalities. The board recommended further research. It set no timeline. It required nothing.
The board cited wagering handle and fan engagement as relevant considerations in a decision about whether a drug is appropriate for 25,578 horses. These things do not belong in the same sentence. One is a financial metric. The other is an animal welfare question.
The board’s FAC told it in writing that political and commercial interests may be overriding animal welfare in Lasix policy. The board ratified the status quo.
The board noted that every state racing commission applied for and received a three-year exemption from the furosemide prohibition — meaning zero jurisdictions chose to move toward the drug-free international standard even when given the option. The board treated this unanimous opt-in as evidence that the status quo is working. It is equally evidence that no regulator in the United States is willing to absorb the short-term economic pain of doing anything differently.
The 2-year-old ban is the one place where the board showed genuine restraint, citing preliminary signals about bone development and career starts. That is commendable and consistent with the science. The board also acknowledged, based on Waller’s data, that 2-year-old racing metrics actually improved after the furosemide ban with more starters relative to the foal crop, not fewer. The catastrophe the horsemen predicted did not occur. The board noted this. It did not draw the obvious conclusion about what that implies for the broader population.
Zero jurisdictions chose to move toward the drug-free international standard even when given the option. The board treated unanimous opt-in as evidence the status quo is working. It is equally evidence that no one is willing to absorb the cost of doing anything differently.
THE HONEST SUMMARY
The FAC produced more rigorous science than we anticipated. Gross 1999 was a legitimate target for methodological critique. The Waller study, conducted under regulated ban conditions with matched controls, is the best current evidence on performance effects and it found none. We should have had access to that study when we wrote in April. We did not. We reported what we had. What we had was incomplete.
At the same time: the board used a statutory definition of “performance-enhancing” narrow enough to exclude the weight-loss mechanism its own FAC acknowledged. It discounted an unfunded study showing a 55% improvement in money-finish odds. It cited wagering handle as a relevant welfare consideration. It deviated from its own committee’s recommendation on Named Stakes without a scientific basis. It set no enforceable timeline for the further research its own FAC said was warranted. And it produced a document in which its own advisory committee warned against prioritizing commercial interests over animal welfare and then did exactly that.
The race really is between fifty years of institutional inertia and the sport’s ability to ask itself an honest question. The HISA board, built from the ranks of the institutions that built that inertia, held a vote, commissioned real science, and arrived at the answer those institutions required.
That is not a conspiracy. It is something more ordinary and more corrosive: a system that has fully internalized its own rationalization, funded the research most likely to validate it, and built a regulatory architecture whose independence is real enough to provide cover and insufficient to produce change.
A Final Word
Belief is a full-time choice. You’re all in, or you fold. There is no halfway.
The biggest charade in horse racing is the one that has outlasted every reform, every committee, every federal statute is the idea that it’s all about the horse. It isn’t. It was never meant to be. The horse is a participant in a business, and a remarkably durable one, but a business nonetheless. Racing is far from the only industry built on that arrangement. It’s just the one we’re talking about today.
What the HISA board produced on April 24, 2026 is a document that says the right things in the right order and arrives at the answer the industry required. The FAC’s own scientists warned against letting commercial interests override animal welfare. The board thanked them, filed the report, and cited wagering handle.
I’m pretty sure when God gave us dominion over the animals, this isn’t what He had in mind. Genesis doesn’t say tend the horse until it’s no longer profitable, then retire it to aftercare and call it stewardship. But here we are, 76% of a racing population on a race-day diuretic, and the central argument for keeping it there is what happens to the handle if we don’t.
All in, or fold. The industry has made its choice for years running.
The horse didn’t get a vote.
One final thought, because the timing demands it.
Six times since Justify won the Triple Crown in 2018, the Kentucky Derby winner has not run in the Preakness. Golden Tempo became the latest, his connections announcing Wednesday that the Derby winner will bypass Laurel Park. The conversation now turning is whether to move the Preakness to three weeks after the Derby spreading the series out, easing the scheduling pressure, saving the television product from its own attrition. Viewership for last year’s Preakness dropped 16%, and whoever wins the broadcast rights battle may yet jockey the calendar.
This is where the industry’s sideways logic completes the circle. Racing spent this week unanimously finding that a race-day diuretic causing horses to shed 20 to 30 pounds of fluid has no performance-enhancing effect and that the central argument for keeping it is what happens to the handle without it. Meanwhile, the same industry is now contemplating restructuring its most storied series because horses can no longer reliably run back in two weeks. Spend a Buck skipped the Preakness in 1985 for a $2 million bonus. That was about the money, and everyone knew it. What’s happening now isn’t greed. It’s biology. Years of breeding for pure speed, racing on Lasix, managing horses as depreciating assets rather than athletes and the industry’s answer to the inevitable consequence is to move the furniture.
The Triple Crown isn’t being threatened by greedy owners or short-sighted trainers. It’s being threatened by the accumulation of every choice the industry made when it decided the handle came first. They built this. The calendar change won’t fix it. It’ll just make the Preakness a better television product for a few more years while the underlying question goes, as always, unanswered.
The horse didn’t get a vote on that, either.
2019 Country House Developed a cough and minor illness shortly after the Derby. 2020 Authentic N/A* (The Preakness was run after the Derby, but Authentic did compete).2021 Mandaloun Did not run. (Elevated to winner later after Medina Spirit’s DQ). Medina Spirit did run. 2022 Rich Strike Trainer Eric Reed opted for extra rest to target the Belmont Stakes. 2025 Sovereignty Trainer Bill Mott chose to skip the Preakness to focus on the Belmont. 2026 Golden Tempo Trainer Cherie DeVaux opted for rest after a “race of a lifetime” effort.
How many remember or are aware Citation ran in Chicago between his Preakness and Belmont wins in his Triple Crown season? He is not the only one.
HISA: DETERMINATION OF THE BOARD OF DIRECTORS OF THE HORSERACING INTEGRITY AND SAFETY AUTHORITY, INC. RELATING TO FUROSEMIDE
We Don’t Get Fooled Again:
