Rush Job

Have you ever been in a situation where something had to be done, and it had to be done quickly because of a looming deadline? If you have, did you find yourself making mistakes because you were stressed, working too quickly, or not focusing completely on the work?

Now imagine that you’re a lab tech at LNDD in France, on Thursday, August 3rd, 2006. Two days before the lab is going to shut down for a three week holiday. And someone hands you about the hottest property in the place — Floyd Landis’ B sample from Stage 17 of the Tour, to be analyzed for the presence (or not) of synthetic testosterone by close of business the following day.

That would be a little bit of stress, wouldn’t it?

But no matter, you’ve done this before. So often, in fact, that you feel confident that you’ll be done with time to spare. And you’ve got a couple of other lab techs helping out. Techs who worked on a similar test less than three weeks ago. So you go about setting up the tests. You get out the written procedures, but you’ve done these tests so many times before that you only glance over the steps, just to make sure you haven’t skipped anything major.

Friday comes. You’ve run the initial tests and determined those results, you just need to do the carbon isotope test. You get some help from one of the other techs as you ready the CIR/IRMS apparatus for the work to be done. As it settles into its operating state, you go outside and grab a quick puff of a Gaulois. The system has been so reliable, you know it will be operating perfectly when you get back. No need to check it, even. After a few minutes you head back inside and get back to work.

The tests run, the data comes out, you hand it off to your colleague who is going to interpret the results. Your job is done, with a few hours to spare. Off you go to your vacation. No worries.

Your colleague, the poor sap who’s going to have to write up all the results and get that information over to the cycling authorities ASAP, starts going through your data. The graphs all seem to be lining up and confirming what had been done before. But he notices something that might be a problem. One of the gauge readings is higher than it’s supposed to be. That makes him nervous, so he goes back and looks for the written procedure.

Crap. The procedure says something he doesn’t want to see. “You must have the pressure reading between 2E-6 and 4E-6 mbar before proceeding.” This is bad. The reading is 5.2E-6 mbar. Perhaps the manufacturer’s documentation can clarify the matter. So he goes looking. Uh-oh. No OEM documentation to be found. Better not let that get out. Bit of a violation of standards not to have those books. But what to do? Is there any way to determine whether the data is OK? Hmmm.

Finally, he checks the gauge reading for the initial tests. It’s the same. Well, if the original tests were run at the same pressure, it must be all right. Phwew! There’s a nasty bullet dodged.

After extensive review, including an all-nighter, your colleague sends off the results early on Saturday morning, August 5th. He goes home to get some much-needed sleep. A few hours later the news hits the wire services. The damage is done. Before he’s even had breakfast, Floyd Landis is dumped by Phonak and the anti-doping scandal that has consumed the whole cycling world shifts into overdrive: Stories flash far and wide. The lab result shows that the 2006 Tour winner has tested positive for synthetic testosterone.

Or does it? A few weeks later, after everyone has returned from their holiday nice and refreshed, other staff look at the test results and reach a disturbing conclusion. There’s a huge problem with the data — the conclusions might even be wrong. Now that the information has been released to the world, how can the story be pulled back?

They go to their bosses and tell them what they’ve discovered. The bosses consult with others, including various members of the cycling world and WADA. What do we do? No one knows for sure. However we proceed, someone says, we have to be very careful. These officials punt the decision making to the highest authorities, hoping they’ll do the right thing.

And so, the highest muckety-mucks get on a conference call and hear the news. LNDD’s data has problems. Big problems. The biggest problems. One could even call them “the worst case scenario” of problems. Errors made during the testing mean that the data itself could be ruled invalid, similar to errors in other cases involving LNDD.

Two areas of concern are discussed: One, because the lab was short-staffed, two of the technicians who worked on Landis’ A sample also worked on the B sample. The Landaluze case is in full swing, but there’s been no ruling. Someone says, “Hey, the WADA code allows us to use some of the same staff to verify the data, and the CAS hasn’t ruled. I don’t see a problem here.” So that issue is put to rest.

One down, one to go. The other problem, something to do with how the CIR/IRMS apparatus was run. Some pressure thing, but someone from the lab says, “Not to worry, it’s a minor glitch that ultimately doesn’t affect the results.” There’s much disagreement, however, among the more technically savvy and the issue is hotly debated for more than an hour. This is bad. Very bad. If word gets out, LNDD is going to look even worse than they have up to now.

Again, someone decrees that nothing be done and the case should move forward. “Look, it’s just a minor technicality that the data isn’t perfect. We all know there’s not a single clean cyclist out there. We can bully this guy into accepting the sanction. We’ve got more resources. We can bleed him dry, if necessary. We’ll get the conviction.”

However, Floyd Landis isn’t quite so pliable as the muckety-mucks expect. He chooses to fight, rather than admit to something he didn’t do. And with that, the real fun begins …