Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Wednesday, November 07, 2007

The effect of EPO on performance: Who wouldn't want to use it?

It's been a busy time for running posts here at the Science of Sport, with first the marathon world record in Berlin, the heat of the Chicago marathon and then the drama of the New York Marathon dominating our recent postings. And as offshoots of these events, we've covered a variety of interesting, relevant (and somewhat controversial) topics in our "series" over the last 6 weeks. We'll return to that theme next week, when we'll look at muscle cramps and electrolytes in a little more detail.

But until then, we thought we'd do a more general week, posting on a variety of interesting articles related to sports. And to begin with, we thought we'd return to another controversial topic - drug use in sport (read cycling!) That's harsh on the sport, I admit, but I saw an interview with Pat McQuaide on BBC recently (HardTalk), where he denied that cycling had a problem with doping. Quite frankly, if that is their attitude, then they deserve to have all the sponsors withdraw and media coverage of their events pulled.

Around the time of the Tour de France this year, we featured a number of posts on drug use. We looked at the history of doping in the sport, evidence of the effectiveness of drugs, and then of course, the Tour came along and gave us plenty to write about! Rasmussen, Vinokourov, etc! Cycling seems to deliver on controversy without fail! (You can find the links to these articles below)

A clean winner vs. a dirty 'loser': Can an athlete win without doping?

It was pretty clear around that time that one of the main problems with the argument that doping was rife in the sport was the fact that very few studies have even evaluated the effect of drugs on performance. There is a remarkable scarcity of data around the effects of doping on performance. It has often been argued that if "one cyclist is using, then they must all be using". There are some flaws in that thinking, for sure, but it does raise the interesting question that if the guys who are finishing middle of the pack, and losing by minutes on a climb or time-trial, are found to be using drugs, then how does the athelte who WINS the race do so clean?

The answer must be that:

  • Either the drugs are not that effective, and so the riders using them are gaining little or no benefit anyway (if this was true, then one has to wonder why drug use is still prevalent - you just KNOW it's not the case), or;
  • The winner is so vastly physiologically superior without drugs that he overcomes whatever benefit the drug provides to his rivals. If this is the case, then the natural, physiological differences between a "clean winner" and "dirty loser" must be greater than the improvement provided by doping
Either way, we need to know just how much drugs improve performance. And that is what today's post is about...

Evidence for the effect of drugs on performance

In previous posts, we've looked at the data from the East German doping "machine" published by Franke & Berendonk in 1997 in Clinical Chemistry, looked at the effect of a steroid programme on performance. That paper showed a great example of how a 'mediocre' shot-put athlete became a world record holder, driven on by a 17% improvement in ONE season

But, that was steroids, not the most widely used drugs in cycling, though they are certainly used, if the tests and testimonies are believed. There are benefits to using steroids, for sure, in that they will assist in recovery. There is ample evidence of how testosterone levels fall progressively during a period of hard training or racing and so the correct use of steroids will improve recovery and thus performance in the longer races.

But we're more interested in EPO. There is plenty of evidence that shows indirect benefits, including Marco Pantani's remarkble hematocrit graph, which correlated precisely with his performance over the course of three seasons. But it's direct evidence we're after, and that's where we turn to a study published in August this year.

EPO improves performance by 54% in a laboratory trial

This great study, published in the European Journal of Applied Physiology earlier this year, evaluated the effects of EPO use on performance during cycling. We'll try to break the study down as simply and clearly as possible:

Who was tested?

They had 16 reasonably fit cyclists take part in the study. The pre-testing VO2max tests showed an average VO2 of about 3.90L/min and a Peak Power Output of 325 W. By no means world-class cyclists, but fit athletes. This does have some implications for the application of the data, which we'll get to later.

How were they tested?

The testing involved an 13-week period, where the 16 athletes were split into two groups. The control group received placebo injection, whereas the 8 cyclists in the EPO group received a dosage of EPO on a schedule worked out over the 13-week period. One potential problem with the study was that the EPO group could not be blinded that they were receiving EPO, for ethical reasons. What this means is that everyone receiving EPO KNEW that they were, and there's good reason to believe that simply knowing you're receiving a drug improves performance as well! The control subjects were blinded, so they did not know whether they were on EPO or not, which does partly offset this problem.

Measures of performance?

All the athletes were tested BEFORE and AFTER the injections doing two performance-trials:
  • Peak Power Output testing - here, the subjects start off riding at a low power output and the workload increases every 90seconds until exhaustion. Basically, the cyclist has to go harder and harder until they cannot push anymore! The test is used to measure VO2max and also a Peak Power Output
  • This was followed by a Trial to Exhaustion at 80% of the previously determined Peak Power Output. In this trial, the cyclist rides at ONE power output - 80% of their maximum, and they ride until exhaustion. This test is used as a measure of endurance performance. This trial was done after 4 weeks and again after 11 weeks of the trial.
The results: A 54% improvement in performance

We don't wish to go into all the blood analysis and DEXA work done - they measured all kinds of things, but this is a post about performance. And the main finding was that EPO use improved time to exhaustion by an enormous 54% within 4 weeks! Peak Power Output improved by 13% in the first four weeks of the trial. The graph below shows the results:

What are the implications?

The graph above clearly shows the benefits that can be gained through the use of EPO. At the intensity of cycling tested in this study, the improvement in time before exhaustion was roughly 11 minutes on a 22 minute trial. Of course, there are some (including us) who would debate the merits of a "ride to exhaustion" as a measure of performance. Normally, we prefer to see a Time-trial as a measure of performance, because the concept of riding to fatigue is not really an accurate or repeatable measure of performance.

However, when you consider the sport of cycling, the final climb of a big mountain day in the Tour is effectively just this - a ride to exhaustion. A group of riders will arrive at the bottom of a 10km climb together, and by the time they reach the summit, the winners are the ones who have managed to sustain a high power output without reaching exhaustion! Now, if you can improve your time to fatigue (that is, the point at which you crack and get dropped) by 11 minutes, that's clearly a massive improvement - it is the difference between making the summit and being dropped with 5km to go!

Another potential problem with the study is the extrapolation of the data to the elite. These subjects were fit, but clearly not elite. It's likely that in the elite, the improvement would be smaller. For example, you could hardly take an Alberto Contador, who might have a Peak Power Output of 500W and bump it up to 565 (13% increase, see graph above) in 4 weeks! Having said that, if you could take this figure and get it to 515W, that would be a very significant increase at the elite level. Similarly, if you could help an elite cyclist improve his average power output by only 5%, that would represent a major step forward. Whether or not EPO would do this is debatable, but given this study, it would seem that 5% is a pretty conservative guess for how much EPO would improve performance...

So can you win a cycling race clean?

So while the results still don't fully answer the question of "how much does EPO improve performance?", they do go a long way to showing us that the effect is potentially massive. Even a quarter of this improvement - 15%, would be the difference between a yellow jersey and the autobus during the Tour de France.

There is the slight problem that EPO is probably not the most widely used drug AT THE MOMENT - in the 1990's, certainly, its use was widespread, by admission and testing result. But in the current generation, it seems that more advanced chemical compounds, as well as blood doping are the choice of dopers. Is their effect the same? I'd lean towards saying, yes, they are, in which case you still have this potentially enormous increase in performance.

So can you win clean? As much as I'd like to think so, when you have this situation where a guy finishing in the top 10 is using drugs and being beaten by minutes on a mountain climb, I find it difficult to believe that physiologically, the margins can be that large. I believe that the NATURAL, physiological difference between riders is tiny - maybe 1% separates a champion from tenth place. So take a drug that improves performance by, let's be conservative and say 5%, and that mid-packer still can't win the race, then you have to wonder about the guy who is winning...?

This study clearly shows that EPO works. I'd extend that to say that any practice that increases the body's ability to carry O2 will work - so the same goes for blood doping. If they work, and work by the sort of margins we seem to be talking here - tens of percent, then can one gifted, unique individual dominate the sport? I think not.

Unless of course, he's genetically altered like a certain mouse was reported to have been last week...but yes, that's ridiculous, as we pointed out in our post on the "Mighty Mouse", and so I'm highly sceptical....


More doping posts from our archives:

Culture of doping in cycling
Basic overview of EPO and testosterone
The effect of steroids on performance, and indirect evidence for EPO from Marco Pantani


Meg & Dave said...

I have a little different take on the EPO performance advantage at the super professional level of the Tour De France riders.

You elude to a system of diminishing returns, a mediocre athelete can gain greatly via quicker muscle recovery and better blood, but a professional can gain only a percent or two. The super elite atheletes are those that push long term physical training to the craziest level. In the long term, it is very much repitition to infinity that defines professional atheletes. The return diminishes until you are just as good as the next guy who is or maybe isn't taking drugs. But you are as good as the other guy, still not as good as you can be, drugs or not.

I have too much faith in human performance to write off the top five in the world as cheaters.

Anonymous said...

Would it be possible to get the exact reference of the European Journal of Applied Physiology article?



Ross Tucker and Jonathan Dugas said...

Hi Francois,

Thanks for visiting us here at The Science of Sport.

The reference for the study in this post is:

Thomsen JJ et al. EJAP, Volume 101, pages 481-486.

The Pubmed id is:


We hope that helps!

Kind Regards,

Anonymous said...

@ energetich20,

1% difference in the Tour de France is about 30 minutes behind the winners time.

Think about that,


Anonymous said...

There was a study published in the Journal of Sports Sciences in 2006that showed endurance athletes could accurately perceive the effects of EPO.


Anonymous said...

Hi there
This is pretty sad, I ran a 1:44 800m and a 3:53 mile clean in my career. I have a moral problem with drug taking but often wondered what could have been or where I would have been in the pack, no so much if I had taken drugs but rather if nobody else did. Anyway, I will never know but when racing in europe I become well aware that this was going on on quite a large scale turning potentially middle of the pack guys in champs and great runners into possibly more average runners. anyway very interesting stuff

DBrower said...

I think the subheading "improves performance by 54% in a laboratory trial" is correct.

What went up was the TTE value, which isn't necessarily the sole judge of performance. The less inflated value was the peak power output, and that might be a better measurement, depending on what one is looking at.

TTE measures typically aren't done at PPO levels, but something lower. So the question arises, in some event, is the relevant measure of performance PPO or TTE?


Ross Tucker and Jonathan Dugas said...

Hi tbv, and thanks for your comment here.

You are correct that a "time-to-exhaustion" or TTE is not necessarily the best predictor of performance. In addition, these "open-ended" tests tend to have high variability. In fact it puzzled us when we read this study why they chose a TTE and not a time trial. it is certainly a limitation to this study.

Second, TTE tests can certainly be done at 100% of peak power. If you check in the literature you will see these tests are completed over a range of intensities from 70% to 100%. The intensity will vary depending on the aims and preferences of the investigators.

Finally, although these tests might not be the best measure of performance and are worse at measuring changes in performance (due to so much variability) the fact that the mean improvement in this trial was 54% is an important point, for this is such a large change that even if it overestimates, there is still likely an improvement---although we cannot be sure if it is 1%, 5%, 10%, or something else.

A change of 54% exceeds even the high variability of these open-ended tests.

Thanks again for your comment here and for visiting The Science of Sport.

Kind Regards,

Unknown said...

So, as experts on doping and cycling - what is your take on the 2008 tour? What are the odds that the top riders were blood doping, using EPO or steroids, or whatever?

Ross Tucker and Jonathan Dugas said...

Hi Mike

Thanks for the mail.

We gave a few comments on it in our series during the course of the Tour de France...

Those articles are a little "scattered", but if you go to the home page of the site, and then do a search for Tour de France, then you will find those articles. Or perhaps more convenient, you can just pull up the archives from July, and you'll find a whole series of articles, where we gave some opinion and insight on the Tour as it happened.

In particular, an article I wrote on July 30 gives some opinion.

I wouldn't commit a number to it, so I don't want to say it was likely to be clean. I think the sport as a whole is much cleaner, but there is of course a chance of the occasional doping. I think that the systematic doping of the 90's and early 2000's has been greatly reduced, so it's a step in the right direction. But clean...doubtful...


Anonymous said...

I have some doubts with these conclusions. Does EPO works, off course it does. Does it work for cycling, that's a total different question. I'm not an expert but I guess it would be safe to say that the time to exhaustion for a clean elite cyclist would be much longer then 22 minutes, probably at least 30 minutes. The time it takes to ride a climb of 10 km is about 25 minutes, 30 at most, depending of the difficulty of the climb. And usually the favorites won't start attacking eachother before reaching the final 5 km of the stage . So what good would 11 extra minutes do if you are riding at 80% for only about 10 miuntes? Of course the higher the power output the shorter the exhaustion time, but normally a rider would not ride close to full power when one or more riders are using his slipstream. He would slow down and wait for a another moment to attack, which gives the other rider(s) time to recover. Unlike sports as running, speed-skating or other endurance-sports, cycling is not a sport of a couple of men racing and the fastest wins. That why I don't think you can say that easy that the study-results proves that doping works for cycling.

Furthermore I think you underestimate the physiologically factor in cycling. There are many examples of riders who buckled under pressure, especially in the Tour de France. Or riders who did great things but failed when given a chance as a leader, or when being in lead. I think there is much more to cycling then just train hard, race and the one who is the fittest wins.

Anonymous said...

hello guys
fantastic article

could we use it for our resources page on


we are a not for profit anti doping campaign.

many thanks


Unknown said...

Hi. I trully believe that the administration of EPO in only 4 weeks can increase theperformance of an athlete, however i am quite curious about the dose of EPO administrated in this study. The gain of performance trully depends on that right???


Anonymous said...

"So while the results still don't fully answer the question of "how much does EPO improve performance?", they do go a long way to showing us that the effect is potentially massive. Even a quarter of this improvement - 15%, would be the difference between a yellow jersey and the autobus during the Tour de France."

Are you having a laugh? How would a 15% increase in TTE when riding at 80% of VO2 max be the difference between the yellow jersey and the autobus? That is a ridiculous statement.