Do Performance Enhancing Drugs Actually Work in a 100 Mile Ultra?

We know EPO can improve aerobic physiology. We know some stimulants can change performance and the way fatigue feels. We know certain hormones can increase muscle mass. What we don't know is how much, if any, of that translates into a faster 100 mile ultramarathon.

For all the research into drugs in sport, our targeted review did not identify a controlled performance trial of the prohibited substances we examined across a continuous ultramarathon lasting 24 to 48 hours.

That is a substantial gap. Much of the evidence comes from cycling tests, short running efforts, clinical studies or experiments measuring physiological changes. Research involving actual ultrarunners mainly tells us what substances have been reported or detected, not whether they improved race performance.

So rather than assume the effect of a laboratory study transfers directly to a 100 mile race, we will examine what was tested, what the results actually show, and where the evidence ends.

This is Article 2 in Track3D's research series on medications and performance. It is an evidence review, not medical advice, a recommendation to use prohibited substances or a substitute for the sport's anti doping rules.

What have researchers actually tested?

The research becomes clearer when we separate studies involving actual ultramarathons from studies that use shorter exercise tests or clinical measurements.

Why a 100 mile race changes the question

Picture a runner 130 kilometres into a mountainous race. Their legs have already absorbed thousands of steps downhill, they have been eating and drinking while moving for a day, their stomach is unreliable, and darkness and lack of sleep are affecting their judgement.

That runner's performance limitations are unlikely to be captured by an isolated measurement of VO₂ max or 30 minutes of cycling power.

• Muscle damage and terrain: Eccentric loading on descents and technical foot placement introduce demands absent from many cycling experiments.
• Fueling: Appetite, gastrointestinal tolerance and carbohydrate availability can limit output regardless of laboratory aerobic capacity.
• Sleep: Staying awake does not necessarily preserve balance, judgement or safe navigation.
• Temperature: Heat, cold and changes in fluid balance can alter cardiovascular and metabolic strain.
• Time: Single short exposures and continuous exertion over 24–48 hours are different research questions.
• Individual differences: Training, sex, age, medical conditions, altitude and race conditions can all change how relevant an experimental result is.

This is why results from a short exercise test cannot be converted into a reliable prediction for a 100 mile race. The physiological advantage might persist, become less important or interact with the demands of the event, but the direct performance evidence is missing.

What do we mean when we say a drug works?

Imagine three studies: one finds a higher VO₂ max, another finds greater cycling power and a third finds participants can continue exercising for longer before exhaustion.

All three might be reported as an improvement in performance. They do not measure the same thing, and none independently tells us how someone will perform over a mountainous 100 mile course.

Time to exhaustion is especially easy to misread. Lasting 15% longer at a laboratory workload does not mean finishing a race 15% faster. The same goes for a percentage improvement in haemoglobin or muscle mass.

Banned does not automatically mean proven effective

WADA does not require a controlled ultramarathon trial before prohibiting a substance. Its inclusion framework considers potential or actual performance enhancement, potential or actual health risk, and the spirit of sport, along with the potential to mask other prohibited substances or methods.

The 2026 Prohibited List distinguishes substances prohibited at all times from those prohibited in competition. EPO, anabolic agents and blood manipulation are examples of the former. Dexamphetamine is a stimulant prohibited in competition, while tramadol and glucocorticoids administered through certain routes are restricted under separate categories.

Several of these drugs have legitimate medical uses. An EPO prescription for a patient with anaemia is not the same thing as using EPO to enhance competition, and lawful medical treatment does not automatically create a competition exemption.

The 2027 List has already been published but does not take effect until 1 January 2027. This article uses the 2026 rules applicable at the October 2026 research date. WADA's 2027 announcement explains the transition.

What the available evidence actually supports

The point of this comparison is not to rank ways of cheating. It is to prevent a real effect in one experiment being turned into a promise about a completely different event.

EPO: more oxygen is not the whole story

Erythropoietin, or EPO, stimulates production of red blood cells. Those cells carry haemoglobin, which transports oxygen. It is a genuine medical treatment for certain forms of anaemia and a prohibited performance enhancing agent when used outside the rules of sport.

In a 2017 randomised trial of 48 trained cyclists, EPO improved maximal cycling power and VO₂ max compared with placebo. The estimated difference in maximal power was about 10 watts but the study did not demonstrate a statistically significant advantage in its submaximal test lasting 45 minutes or its Mont Ventoux road climb endpoint.

That does not mean EPO never helps in competition. The result from the road climb was imprecise and influenced by conditions during an actual race such as pacing, terrain and weather.

A later 2023 controlled study involving 48 men and women found a positive effect on cycling time trial performance. The treated group improved mean power by about 4.1% from its own baseline. Crucially, that number is a change within the treated group, not a verified advantage over placebo of exactly 4.1%.

EPO also carries recognised medical risks, including hypertension and thrombotic complications. Research involving people with anaemia and chronic disease documents serious clinical hazards, but those patient event rates cannot simply be transferred to healthy ultrarunners. Small athlete trials are not large enough to reliably measure rare serious adverse events.

The research gap is not just how much oxygen can be carried. It is how much that matters when muscle damage, energy intake and hours without sleep begin to dominate the race.

Blood transfusions: a measurable effect, a tiny evidence base

Increasing circulating red blood cell mass through blood manipulation can influence oxygen delivery. Whole blood and red blood cell manipulation for performance are prohibited methods under WADA rules, legitimate emergency transfusions have a separate medical and regulatory context.

A 2019 controlled study of autologous blood transfusion enrolled nine men, but only seven usable cycling time trial datasets were available. The study reported improved power in the short time trial. It did not establish a benefit in a different protocol involving repeated sprints, and it did not test an ultramarathon.

That is evidence of a particular athletic effect, but seven results cannot establish a universal race advantage, let alone a reliable safety profile over 30 or 40 hours.

Other techniques involving oxygen transport provide further examples of the distinction between a mechanism and a result. A small carbon monoxide experiment changed haemoglobin mass without demonstrating an improvement in measured swimming performance. Some experimental products designed to deliver oxygen carry serious clinical concerns, but studies involving hospital patients cannot establish specific risks or benefits in healthy ultrarunners.

WADA prohibits relevant carbon monoxide manipulation outside permitted diagnostic use, while ordinary supplemental oxygen by inhalation is treated differently under its method rules. This distinction is regulatory, not proof of an endurance advantage.

Stimulants: feeling less tired does not mean being less tired

This category is especially relevant to ultrarunning, where wakefulness and judgement can become as important as leg strength. Prescription stimulants such as dexamphetamine, methylphenidate and modafinil have legitimate clinical uses, but are prohibited in competition under the relevant WADA stimulant classes.

One widely cited study tested methylphenidate in eight trained men. In cool conditions, performance was not significantly different. In the heat, the stimulated group completed a cycling task roughly 16% faster, despite greater physiological strain. Core temperatures exceeded 40°C, while perceived effort and heat stress did not rise correspondingly.

That is a striking result. But it is a small, short cycling experiment, not evidence of a 16% ultramarathon improvement. Nor does an elevated temperature measurement, by itself, tell us how frequently heatstroke would occur.

Modafinil increased time to exhaustion in one short exercise experiment, while another study found no significant benefit for physical endurance following mental exertion. Different populations and tests help explain why one experiment should not be taken as a universal rule.

Potential cardiovascular stimulation, disturbed sleep, appetite suppression and temperature related effects also matter. Their interactions with heat, poor nutrition and repeated exertion over two nights have not been quantified well enough to produce a reliable ultramarathon risk estimate.

Tramadol: did it improve performance by reducing pain?

Tramadol is an opioid analgesic prohibited in competition. You might assume any potential endurance benefit comes from masking pain, but controlled research reveals a more complicated picture.

A 2023 trial found that trained cyclists completed a time trial covering 25 miles about 51 seconds faster than with placebo, a difference of roughly 1.3%. 27 participants completed the procedures, and 25 were included in the main analysis following two exclusions.

The interesting detail is that perceived exertion during the preceding exercise was lower, while measured pain was not significantly lower. That means the study does not establish pain relief as the actual mechanism behind the performance difference.

Other controlled cycling research did not reproduce a clear performance advantage under its particular protocols. Some trials also reported nausea, dizziness, drowsiness or vomiting.

Opioids have recognised medical risks, including sedation, dependence and potentially dangerous interactions. Those known clinical concerns are separate from the question of how often an adverse event would occur during an ultra. The studies cannot tell us whether masking discomfort would enable a better race or simply encourage someone to continue through an injury.

Corticosteroids are not steroids used to build muscle

Prednisone and prednisolone are glucocorticoids used to treat inflammatory conditions and disorders of the immune system. They are different from anabolic steroids such as testosterone, even though both are often casually referred to as steroids.

A 2023 systematic review of 15 trials involving 175 participants found a small overall performance signal. However, sensitivity analysis excluding particularly small comparisons weakened that finding, and the evidence did not establish reliable differences between administration routes or treatment durations.

Some small experiments reported improved endurance capacity; others, studying different exercise tasks, reported no significant effect. An improvement in cycling time to exhaustion does not demonstrate improved muscle healing, or suggest a predictable advantage in a mountain ultra.

There are genuine clinical effects too, including possible changes in blood glucose, mood, sleep and immune function. Under WADA rules, glucocorticoids are prohibited in competition when administered by certain routes, including oral and injectable routes. Other therapeutic routes can have different permissions. The rules for different administration routes should not be confused with a proof tied to a particular administration route of performance enhancement.

Testosterone, growth hormone and SARMs: bigger is not always faster

Testosterone has established anabolic effects, including increased muscle mass and strength. But carrying more muscle up a mountain is not the same as improving the ability to run for two days.

In a controlled study of 48 active young women, testosterone improved a short running measure of time to exhaustion. That is a measured result, but the test lasted minutes, not hours. It cannot be converted into a percentage reduction in a 100 mile finish time.

Another controlled experiment during severe exercise and energy deficit found greater preservation of lean mass with testosterone without preventing a decline in measured muscle function in the lower body.

Human growth hormone provides a related example. A randomised trial in recreational athletes found a small improvement in a short measure of sprint capacity, without corresponding evidence of improved VO₂ max or other results from longer endurance tests. Some apparent effects on lean mass reflect retained fluid rather than additional functional muscle.

Selective androgen receptor modulators, or SARMs, are often marketed using the language of muscle growth and recovery. Some pharmaceutical research documents changes in lean mass or clinical function in particular populations. It does not establish that marketed products make healthy ultrarunners faster, improve injury healing or are safe for use over the long term.

These substances also have recognised or emerging endocrine, cardiovascular and concerns involving the liver. Clinical replacement of a hormone deficiency is a different situation from deliberate performance enhancement.

Experimental drugs: promising mechanisms are not race results

Experimental compounds are frequently discussed online as if laboratory findings already prove endurance or recovery benefits. In several cases, the strongest performance evidence is still in animals or uncontrolled clinical reports.

GW501516 is especially revealing. A mouse study investigated a mechanism that preserves glucose and reported longer running capacity. Separate human studies examined metabolic endpoints, not verified athlete performance. You cannot take an improvement in mouse treadmill endurance and apply it to a 100 mile finisher.

For BPC-157, a 2026 review of the available human literature identified three small uncontrolled studies involving 30 participant records, and rated the evidence very low certainty. That is not the same as saying no human studies exist, but it is a long way from demonstrating healing, athletic recovery or safety.

Meldonium illustrates a different trap: improvements in clinical exercise tolerance among patients with heart disease are not established performance improvements in healthy athletes. The available athlete study was also limited by nonrandom allocation and baseline differences.

For all these compounds, the quality, identity and legality of products marketed online create questions separate from their claimed physiological effects. No sourcing or use instructions are appropriate from this evidence.

Do we know how many ultrarunners use drugs to enhance performance?

There is research, but it does not support a universal percentage of ultrarunners who deliberately dope.

A 2018 survey of 609 ultramarathon runners found that 8.4% reported using drugs used to enhance performance during training or competition. It was a voluntary survey, not a representative audit of everyone who races ultras. Importantly, the study did not find a difference in athlete rankings between runners who reported using the drugs and those who did not.

A different 2024 investigation analysed 412 anonymous urine samples collected before the race from male ultraendurance participants. About 16.3% contained substances classified within prohibited categories under the study's assessment.

Survey responses, chemical screening and formal decisions on doping violations measure different things. They should not be combined to suggest that a known proportion of all ultrarunners are cheating.

A legal substance can have better endurance evidence than a banned one

Caffeine is a useful counterexample to the assumption that WADA permission indicates a substance does not enhance performance. It is permitted, yet the Australian Institute of Sport includes caffeine among its performance supplements supported by evidence for specific contexts.

Dietary nitrate, sodium bicarbonate, creatine and some other legal aids also have performance evidence from specific tasks. Those effects are not uniform across sports, and this does not establish that any particular supplement will help over 100 miles.

Meanwhile, experimental compounds may be prohibited despite lacking convincing human evidence of athletic improvement. WADA rules reflect a broader framework for regulating doping than a simple ranking of which drugs produce the largest measured benefit.

The useful conclusion for runners is to keep three questions separate: what a substance does to the body, whether it changes actual race performance, and whether the rules permit it.

Common questions

Related Track3D Guide: The Hidden Risks of Everyday Medication in Ultrarunning. The next article in this research series will examine accidental doping through ordinary prescriptions, supplement contamination and medical procedures.

Research notes and primary sources

Research snapshot: 13 October 2026. This educational research article draws on a structured review and targeted independent verification, not a formal systematic review or external clinical peer review. Study designs, sample sizes, conflicting findings and limits on access to original sources matter. This is not medical advice, permission to use a substance, a claim of safe performance enhancement or an individual ruling under doping rules. Specialist medical and regulatory review is required before publication, and rules should be rechecked for the applicable event and year.