So Asbel Kiprop was handed a four-year ban over the weekend, for his EPO-positive from November 2017. I shared a few thoughts on it over the Easter Weekend, but wanted to expand them a little here in my latest #shortthoughtonsport.
If you’re interested, you can read the entire Final Decision at this link. Below are some things that struck me, a highlights of that decision, along with some “between the lines” thoughts that I think may be of interest:
[ribbon toplink=true]The Kiprop Kitchen sink[/ribbon]
Kiprop’s strategy was to throw pretty much every single procedural error accusation at the testing authorities. You might recall that when the news of the failed test broke, Kiprop alleged a pay-off, a tip-off, bribery and a set-up, and they all made their appearance, along with a few new ones.
Here’s the list, from the Final Decision:
So, we have some conceptual/theoretical matters at the start – EPO at altitude and medicines. Then we move on to to the procedural stuff, which includes the sample collection issues, spiking or substitution, chain of custody and analytical errors.
According to the tribunal, Kiprop’s lawyer presented the Panel with “an a la carte menu of reasons why the charges should be dismissed”. He also argued with a certain style, which I found amusingly mentioned in the Final Decision:
Just as the panel did, let’s consider each of those menu options in turn.
[ribbon toplink=true]EPO at altitude and medical triggers[/ribbon]
The first two issues – EPO triggered at altitude, or by medicine he was taking, are dealt with very swiftly. Easy. Swatting away mosquitos.
Basically, Kiprop tries to argue that while at altitude, 2400m and 2700m, he trains intensely, and this combination of altitude plus intense exercise causes his body to produce more EPO.
The second explanation was somewhat similar – he says that he was taking a few medicines a week before the test, and that these, in a way similar to exercise plus altitude, caused an increase in EPO.
Both these suggestions are very easily dealt with. The test he failed is not looking for the EPO produced naturally by the body, even in circumstances where altitude and exercise combine, or where medicine somehow triggers its production. Natural EPO differs from the EPO that is injected, so even though it is plausible that the levels may increase as a result of exercise at altitude, this is not what he’s ‘on trial’ for. This argument was not even challenged, and so the panel was able to move swiftly along.
[ribbon toplink=true]Procedural errors[/ribbon]
The next batch of issues proved a bit more complex, and even in the Final Decision, you get the sense that they didn’t quite reach the point of certainty about what transpired at that collection of the sample that day in November. However, they did decide that it didn’t matter, because no link could be drawn between what Kiprop was alleging and the fact that his samples were found to contain EPO
First, relating to Kiprop’s claim that he was warned in advance of the testing, the Assistant Doping Control officer admitted giving Kiprop advance warning of testing. Apparently, 1 day before samples were collected on 22 & 27 Nov, & again on 24 Jan the following year, Kiprop was told to expect a test.
For obvious reasons, this is a major problem in the anti-doping process. The odd thing, and which the Panel recognized, is that usually this prevents a positive test from occurring, because the athlete would either avoid that test, or take some steps to pass the test despite having doped (use of a diuretic, urine swap etc).
In this instance, you have a positive test despite the warning. This comes up again later in the Final Decision, so we’ll see the Panel’s thoughts again, but basically, what Kiprop is positioning here is the “Would I be so stupid” defense. That is, he is going to say that if he were doping, and if he knew that a test was coming up, then he would not be so brazen or foolish as to allow the test to be done without some actions to pass it. And thus, he must not be doping. It’s circular, and impossible to evaluate, but I think invites some interesting insights, which we’ll come to later, when we see the panel’s evaluation of this position.
Next, though, the money. This is where the Tribunal probably didn’t quite get to the bottom of the allegation, but decided that ultimately it did not matter.
The highlighted bit above is the key, I think. Kiprop needed to draw a clear link between that payment and the positive test, but could not. One argument would be that Kiprop didn’t pay enough, and in retaliation, the officer spiked the sample. But there are so many inconsistencies and leaps of faith one has to take to make that work that ultimately, it’s just a conspiracy theory without basis.
The Panel arrives at this point through a fairly lengthy credibility assessment of the major witnesses, ultimately concluding the following:
The other procedural issues – spiking or substituting the sample, and the chain of custody issues – are dealt with similarly. The Panel concludes that there’s no evidence to support the assertion, and that if anything, all seems to be in order. “integrity intact” is the two-word summary.
Specifically, the Officers are found to have had no real opportunity or reasons to make any swaps or to contaminate the sample (“means” and “motive”), and nor would he have any capacity to know how much to use (implausbility). The seals were intact, the documents filled out correctly, and ultimately, Kiprop’s assertion has no support at all. You can read all this from Point 73 to Point 85, and it’s best summed up as “Any contrary hypothesis rests on a fragile basis of implausibility too extreme to require rebuttal”.
[ribbon toplink=true]Between the lines – some interesting concepts emerge[/ribbon]
And then finally, there are some quite interesting points that exist between the lines, where I want to explore some anti-doping concepts using the Final Decision.
First, there was the issue of why Kiprop would use EPO at all? It’s November, he’s done for the year, coming off a six-week rest period after the 2017 Track season, and will not run in any consequential races for months.
The reason this matters, by the way, is that one of the things that ‘informs’ or contextualizes the Athlete Biological Passport is the competition schedule of the athlete. In other words, an ABP profile might be more likely to be interpreted as indicating doping when it exists during a season, a week or two out from a major event, or after the rest day of a Grand Tour in cycling, for instance, as opposed to outside of all those windows.
It would appear that Kiprop introduced a similar argument here. He was basically saying that using EPO is implausible at that time of the year, and thus some other explanation for the positive test should be found.
But this is where I think the doping paradigm has shifted. Perhaps in a previous generation, the use of EPO would be limited to competitions, because a) it was expensive, b) it was undetectable. Therefore, you’d use it when it would give you the best outcomes for performance or training, because you could, without risk.
But now, I think the improved sophistication of testing, plus the longitudinal monitoring of blood in the ABP has to some extent constrained the freedom of use of EPO. The response? Smaller doses, spread out, enabled by the fact that if you’re using say one fifth of what you did before, it’s costing so much less per use, and thus can be used over longer periods for the same total cost.
Plus, you’re using it for training benefits, and possibly also to mask the future changes in the ABP (year long doping might be required, now that it’s being monitored for change over a year), so that you can adapt more rapidly to your onset of higher volume training (especially when at altitude), and so there are in fact many plausible reasons, rational and theoretical, for why EPO would be used that far out from competition.
The Panel, of course, don’t need to evaluate all these arguments. Their conclusion on the matter is a lot more brief:
[ribbon toplink=true]”Would I be that stupid?” and implications for anti-doping[/ribbon]
Then we come back to the “Would I be that stupid” defense that I raised earlier. Remember that Kiprop is asserting that he received a tip-off about the test (this was admitted as a procedural error), and so therefore, he must be innocent of doping, because knowing that he was about to be tested, he surely would not have given a sample if was doping.
It is, as mentioned, circular, and the Panel was unconvinced by it.
It’s the next statement that is especially interesting, because it raises a couple of the fundamental challenges in anti-doping:
So, three things:
First, I would suggest that Kiprop was being target tested. There’d be no reason for these tests, so densely packed, during this time of year, unless there was some reason to suspect they’d find something.
This is where, as I’ve written before, his biological passport would have been really interesting – perhaps it hinted very strongly at doping, but was just not sufficient to actually build the legal case (unlike happened recently for Cyrus Rutto). It would be really intriguing to see that, even if only as an illustration of a “false negative” of the ABP, but also a feather in the cap of the ABP (it can be both at the same time, when the system works).
Second, one of Kiprop’s arguments was that the test must be flawed because he was tested on 21 November, and 29 November, either side of the 27 November sample that was positive for EPO. Kiprop maintained that it’s impossible for the sample in the middle of that sequence to be positive, but not the one after. However, this is actually pretty easily dealt with, because it’s a moot point – the samples that ‘bookend’ the positive one were not tested for EPO. They were merely used for the ABP, which measures the blood variables hemoglobin, hematocrit and reticulocyte percentage. So that turned out to be a non-starter argument anyway.
[ribbon toplink=true]A flash of light in a dark tunnel – the challenge for doping detection[/ribbon].
Third, and more important, is the concept raised by the expert during testimony – note the highlighted bit above: “There is no scientific reason why a dose administered on a particular date should not give rise to an AAF three days later, but none five days later”.
This is really illuminating if you follow anti-doping. What it highlights is that the ‘window of opportunity’ to catch a doped athlete is so narrow that it’s like trying to catch a glimpse of a flash of light in a night full of darkness. You can visualize a long, dark tunnel, and at some point along its length, a light is going to shine for what is, relatively speaking, a split second. And anti-doping authorities have to be right there at that very moment in order to actually catch the athlete.
This is a huge issue, and is a significant driver of any poor efficiency of anti-doping. It’s also the driver for the biological passport concept, which at least allows authorities to estimate that the light had shone, even if they can’t see it directly.
Globally speaking, this represents a recognition that an athlete may be doping every time they are tested, but will never be caught because the timing just doesn’t work out, in combination with the smaller doses they would be using. This is why people rightly make a big deal about whereabouts failures – the athlete clearly doesn’t have to “hide” for long, just long enough to test negative despite doping. And here, in this Final Decision on the Kiprop case, you have an expert stating just how short that period can be.
Now, we come back to the previous point – “Would I be so stupid?”. Well, given that many athletes will appreciate the lack of sensitivity in anti-doping tests, provided they are intelligent with the doses used, the answer is that they would definitely be satisfied to provide a sample even though they’re “glowing”. Their action says much more about their confidence in getting away with doping (or their lack of confidence in anti-doping’s ability to catch them) than it does about their innocence or guilt.
After a cyclist was caught doping with EPO recently, someone on Twitter asked “Why would anyone still use EPO?”. I’d offer “Why not?”. When you discover that there’s no good scientific reason for a sample to be negative that close to one that is positive, it should give pause for the challenge facing antidoping. The reward way exceeds the risk, as long the doper is clever with timing, dosage, and avoidance of testers (the doorbell must fail at just the right time).
[ribbon toplink=true]Kiprop as barometer for Kenya’s problem[/ribbon]
And so, on the basis of all we discussed above, Kiprop is handed a four year ban. Despite “tenacity, ingenuity and charm”, he utterly failed to provide any reasonable basis for that Adverse Analytical Finding, and if anything, provided a bunch of theories that make it even more likely he was doping.
As for Kenya, I think it’s pretty clear there are major problems there. I read that Kiprop was the 138th athlete who tested positive since 2004, and I suspect the majority of those have come in the last five years. Kenya was named and placed under intensive surveillance by WADA because of suspicions, and so that would certainly account for some of the numbers (arguably, many countries would see a dramatic leap in doper numbers with increased scrutiny), but the reality is that Kenyan performances must now carry a high degree of suspicion.
From an anti-doping perspective, it is also time they really went after the entourages, and not just the athletes. Who doped Kiprop? Unlikely it was just Kiprop? Look at the support group, starting from the agent? Is there history?
Looking at the entourage is nominally in the WADA code, which is good. But it rarely happens, except when it’s so obvious that it almost can’t be avoided. I’d have thought that by now, there is a pretty good idea of which groups in Kenya are doping, and who might be facilitating that doping.
The Athletics Integrity Unit spoke at a recent London conference about the importance of intelligence. I’m generally impressed with the way the AIU is run, and its success is no doubt responsible for many of those 138 cases. But it seems to me that Kenya is now one of the first places where the coaches, agents and managers need to and can be targeted, because the drip-feed of dopers from certain groups is becoming a steady stream, it’s concentrated enough that it is no longer possible to attribute to “lone wolf” isolated doping by the athlete, and Kiprop is a big enough fish in that stream that maybe it triggers a next phase in this particular anti-doping frontier.