The Ruti Aga File: How a Forged Prescription Turned a Two-Year Case Into a Career's End

September 10 appears three times in my ledger. The first time, on the day the Athletics Integrity Unit stamped a provisional suspension against Ruti Aga's name. The second time, on the day the six-year ban was counted from that same date. And the third time, in the margin, where I wrote: "Return age, thirty-eight."

The story begins at a half marathon in March, in China, which the original statement does not even name. Ruti Aga won that race. There was a number on the results sheet, and under the number, a sample. The sample returned methylprednisolone. What followed was not about the race. It was about a piece of paper that never existed.

The headlines called it a doping case with a forged prescription and moved on. Under the headline there is a smaller story, and it is usually the true one: a thirty-two-year-old marathon runner, a sore knee, a relative who called himself a doctor, a clinic that was never there, and an old prescription pad that the clinic explicitly disowned. The six-year ban comes out of that whole structure.

In this case the arithmetic of the sanction speaks loudest, and the arithmetic is about a document, not a drug. Two years for presence, four for tampering. The real hinge arrived where the athlete tried to manufacture her own defence.

I am filing this because I hold notes on three cases of the same family, and all three show the same thing: the first violation is the substance, and the second, larger violation is the attempt to explain it.

Context: the athlete who was in the file before the race

Ruti Aga is not a new story. The name of the athlete who won the Tokyo Marathon on 17 October 2026 was in my ledger before that race, because at Berlin 2026 she had run 2:18:34 for second place. That figure comes from my old notebook and I would want to check it again against the official results page, but the point holds: she was a 2:18–2:20 marathoner, which means world class, and which means one name inside the depth of Ethiopian women's marathon running of that period.

Her record includes three second-place finishes at World Marathon Majors. That number sounds large and is large, but it needs to sit in the right box: a second place at a Major is a commercial-tier result, not a championship conversion. She holds no global championship medal. That distinction will matter later.

The Ethiopian women's marathon landscape is such that a 2:18 performer is not automatically selected for a national team. At the top there is a world-record class; below it a Kenyan sub-2:17 cohort; and below that the band where runners like Aga sit. The removal of one name from that band changes nothing in national capacity. The real subject of this case is governance, not competition.

Now to that March race. The original statement says the positive sample was collected after a half marathon won in China in March. No name, no year, no time, no field depth. I have three blank spaces on this page and I will not fill them. What can be inferred is that if an AIU test exists, this was a World Athletics Label-tier road race, with an elite women's field and a testing framework in place.

On the road, the testing window opens at 23:59 the day before the race. That is a technical detail at the centre of the whole case, and I will come back to it.

Core: the drug class, the arithmetic, and the prescription that was never written

The first correction

The original report called methylprednisolone a "prohibited steroid" and said WADA classes it as performance enhancing. That classification is wrong, and the error distorts every comparison this case invites.

Methylprednisolone is a glucocorticoid, class S9 on the WADA Prohibited List. It is not an anabolic steroid, which sits in class S1. The difference is not cosmetic: anabolic steroids act towards muscle building and are prohibited at all times; glucocorticoids are anti-inflammatory drugs used to reduce pain and swelling, and under the current List they are prohibited in-competition only — and only by injection, oral or rectal routes.

The consequence is direct: the case does not rest on "steroid use"; it rests on route of administration and timing relative to the competition. A therapeutic knee injection out of competition is entirely legal. It becomes a violation only if residues are present in-competition.

One more date matters here, and it is absent from the report: in 2026 the List changed, and in-competition glucocorticoid injection became a strict violation. The framework before that change left more room. Aga's case sits inside the changed framework.

The sum

Where the six years come from needs to be understood. Two years for the presence violation — the standard sanction for a substance the Code calls "specified", absent proven intent. Four years for tampering — the Code's standard tampering sanction. Two plus four, six.

But there is a silence inside the sum, and the silence is the loudest part. Had the panel accepted a no-significant-fault mitigation, the presence sanction could have fallen below two years. It did not. The arithmetic tells us the panel found presence proven, intent not proven, and tampering fully culpable.

There is a near-paradox I want on the record: a forged document is itself circumstantial evidence of concealment intent. Yet it did not escalate the presence sanction. The panel kept the two boxes separate, which is procedurally correct and humanly strange.

Time served under provisional suspension counts against the ban, so counting from September 10 is not a concession. It is a procedural benefit, not a reduction. The athlete gains nothing real; the clock simply starts where it stopped.

The paper

Now to the document that decided the case. According to the AIU statement, Aga cited a knee injury and submitted a prescription as evidence of treatment. Three things broke in verification, and they broke in three different ways.

First: the clinic named on the prescription did not exist. Second: the person who wrote it was a family member with no medical registration. Third: that person claimed to have used "an old prescription pad from a former workplace" — and the clinic named as that workplace explicitly disowned the claim.

Three independent verification failures. Had even one held, the case might have gone another way. The forgery here is not a circumstance; it is the architecture.

What stings me is not that someone lied. What stings me is that someone believed it would work. A thirty-two-year-old elite marathoner with a Tokyo title and three Major second places had nobody around her saying, do not take this route. Or nobody who knew what the correct route was.

What was missing: the TUE structure

A structured pathway exists for a glucocorticoid injection — a Therapeutic Use Exemption. An athlete with documented knee pathology either obtains one, or manages route and timing so that no residue remains in-competition.

Neither appears here. And that silence is itself information. When a valid pathway exists and an athlete chooses a fabricated one instead, the conclusion is that nobody around her knew the valid pathway existed.

This is bigger than an individual lapse. Chronic knee pathology is a known load-bearing failure mode in high-volume marathoners. The pressure to mask pain is real, because missing an appearance-fee race is direct income loss. Glucocorticoids are the cheapest, most available tool for that masking. And when the surrounding medical network is informal, the forged prescription becomes the natural next step of that network.

I am not calling this a defence. I am saying there is an untreated governance gap behind the forgery, and that gap belongs to more than one athlete.

The four years for tampering: a charge built from an interview

A technical point most readers miss. The tampering charge here did not come from a document alone. It came from an interview — from statements made under questioning.

The distinction matters. Proving a document false is straightforward; proving a statement false requires looking at the conditions under which it was given. Language and translation conditions in AIU interviews with Ethiopian athletes have been a recurring point of contention, and I am filing this as an open page, not a closed finding.

The Chinese label circuit and the testing economy

Back to the unnamed race. Label-tier road races in China are a distinct economic tier for East African marathoners — the gap between Majors, decent appearance fees, technically low risk. For a 2:18–2:20 runner this is the ideal model: few races, high fees.

But that same tier is now among the AIU's highest-yield testing grounds. The appearance-fee economy and the testing economy are colliding in the same races.

This collision is not accidental. Where the budget is large, the elite field is deep; where the field is deep, the testing value is highest. The result is that results from this tier will increasingly carry integrity information, not only performance information.

The empty page

I have to write down what is not in this file, because if I do not, someone else will colour it in.

The year of the March race is absent. Its name is absent. Its time is absent. Field depth is absent. The scope of the disqualification — whether the March result and prize money are voided — is not explicitly stated. Standard AIU practice disqualifies the competition at which the sample was collected, but I will put that in the inference box, not the fact box.

And most importantly: does the 2026 Tokyo title survive? The violation is anchored to a March race after that date. Absent a finding of earlier use, results before that date are not automatically annulled. That is what survives in this file — a Major title, won by the athlete when she was still only an athlete.

The contrarian angle: what the headline says and what the file says

The headline's reading is simple: a cheat was caught, six years, justice done. That reading is not wrong. It is incomplete, and an incomplete reading often does more damage than a wrong one.

The first counter-fact: the substance at the centre of the case has no proven endurance benefit. A glucocorticoid reduces inflammation and pain. Whether pain reduction is performance enhancement is a separate argument — and a legitimate one, because making it possible to race injured is a kind of benefit. But placing this alongside anabolic steroid cases is factually wrong. The media did exactly that.

And from there the second counter-fact follows: this case was built by the weakness of a document, not the strength of a drug. Had Aga held genuine records, she had a valid defence — route, timing, medical necessity. She destroyed the one defence that might have worked. Six years instead of two did not happen because the drug was more dangerous; it happened because the proof of innocence was fake.

The third counter-fact is the least comfortable, and I will place it carefully. A six-year ban on a substance with no proven endurance benefit is a striking outcome. It will be read two ways. One reading: how strict the AIU is. The other: how unevenly enforcement falls on African road runners.

I want both readings held together, because each holds part of the truth and each is incomplete alone. The strictness reading ignores that the bulk of the sanction came from a document, not a drug — this is an example of document enforcement, not substance enforcement. The disparity reading ignores that a forged prescription is not a small thing; it breaks trust in the whole system.

My ledger holds more than one case of this family. In East African road-running cases I have recorded over the years, one pattern repeats: the first violation is the substance, and the second, larger violation is the attempt to explain it. The enforcement frontier has shifted from detection to document verification. Aga's case is not the outlier of that shift; it is the sample.

And the fourth counter-fact, likely the least written about. The athlete who received six years holds no global championship medal. Having won what she won — three Major second places, a Tokyo title — she now becomes commercially near-invisible in a sport where distance-running sponsorship is already thin. This does not reduce the seriousness of the case. It only shows where the sanction will actually land.

Forward

One number I kept returning to in this file is thirty-two. Add six, and thirty-eight appears. Finding a route back into the depth of Ethiopian women's marathon running at thirty-eight is a theoretical possibility, not a practical plan. The ban is technically six years and functionally a career.

Writing this, I remember the sixty calls that went out into an empty year; not one of them could say who a runner with a sore knee should ask. Where a TUE comes from, which route is safe, how many days before the race to stop — this is all information, and information is also a resource that is not distributed evenly.

The question I am left with: next time a runner has a sore knee and no registered doctor around her, which route will she take? The answer depends on whether this six-year ban is read as a warning or as a diagnosis. A warning changes nobody. A diagnosis might.

The page for September 10 is still open in my ledger. I have not put a full stop on it yet.