How does steroid testing actually work?


With steroid testing back in the news, I can completely understand if a lot of the details seem overwhelming. It’s such a fiery topic, prone to quick reactions based on little fact, that I find that it’s really useful to understand what actually happens in the testing process.

After androstenedione was found in Mark McGwire’s locker during his record-breaking 1998 home run chase, Major League Baseball felt the need to start testing for steroid use. In 2002, MLB started testing for performance enhancing drugs for the first time, and in 2004, test administration was moved into the care of the World Anti-Doping Agency (WADA), an international multi-sport organization dedicated to stopping the rise of performance enhancing drugs. Major League Baseball has a Joint Drug Prevention and Treatment Program, governed by the Joint Drug Agreement (JDA), which outlines the specific rules agreed to by MLB and the Player’s Association regarding substance abuse in baseball. Collection of test samples is carried out by Comprehensive Drug Testing, Inc. (CDT), who then ships the samples to the WADA laboratory in Montreal, Canada, for testing. That last part was crucial in the Ryan Braun fiasco, as CDT didn’t get Braun’s sample to Montreal for two whole days, breaking strict rules regarding chain of custody.

MLB has an exhaustive list of every banned substance, which includes every substance classified in Schedules I or II under the Controlled Substances Act, a federal US law. MLB also bans 70 performance enhancing steroids, hormones, and masking agents, as well as 56 stimulants, such as amphetamines.

Every player on an MLB 40-man roster is tested at least twice per year for performance enhancing drugs or stimulants. On the first day of spring training, every player has to submit a urine sample. After that, CDT will collect a total of 1,400 additional random unannounced urine samples throughout the season. The Joint Drug Agreement also allows for more than 200 urine sample collections on unannounced random dates in the offseason.

There are extra tests for previous violators of the JDA, but there’s also a clause that says if the league or the Player’s Association has evidence to suspect steroid use, they can submit a notification of “reasonable cause”, at which point the player has to submit a test sample within 48 hours.

Melky Cabrera and Bartolo Colon were nailed for the presence of exogenous (non-natural) testosterone in their urine samples. The WADA laboratory in Montreal performs two tests for testosterone, but to explain why, I’m going to have to dip into biochemistry.

The body produces testosterone and epitestosterone, two very chemically-similar hormones, in roughly equal quantities. In healthy adult males, the ratio of the two hormones in the body tends to sit somewhere around 1:1. The relevant difference between the two is that while testosterone has the usual litany of performance enhancing benefits, epitestosterone doesn’t. So as you might expect, urine samples of athletes who have taken illicit testosterone supplements (or steroids that are metabolized into testosterone, like McGwire’s androstenedione) often show large amounts of testosterone with relatively normal amounts of epitestosterone.

The first urine test that WADA will perform is a testosterone/epitestosterone ratio screening test, where a urine sample is chemically analyzed to find the ratio of the concentrations of the two hormones. WADA sets the test limit at 4:1. If an athlete’s urine sample contains four times as much testosterone as epitestosterone, the sample is marked as a failed test. Failed steroid tests carry an enormous public stigma, so it’s important to note that somewhere around 5 percent of the population of adult male professional athletes have a natural T/E ratio of 4:1 or greater. Since CDT collects a urine sample from everybody during spring training as well as 1,400 samples over the course of a season, this means that even if steroid use was completely eradicated from organized baseball, more than a hundred urine samples would fail the T/E test every year. This does not mean the player was taking any banned substances. WADA uses the T/E test simply as a culling process, not as a damning confirmation of steroid use.

If a player fails the T/E test, WADA moves onto what they call the confirmation test, an isotope ratio test. I’m going to have to make one more detour into biochemistry here, but I’ll try to keep it quick.

All carbon on the planet exists as a mixture of isotopes, different versions of carbon. Almost all of the carbon on the planet has six protons and six neutrons inside of the atom, which we call carbon-12. However, one percent of the carbon on Earth has an extra neutron, seven in total. We call this carbon-13, and it’s ever so slightly heavier than your usual garden-variety carbon-12. All organic molecules have a significant quantity of carbon, so naturally, the carbon in your body will have a small amount of carbon-13 mixed in. For the most part, carbon-13 walks and talks just like carbon-12, but due to its relative heaviness, plants have been very, very slowly preferring carbon-12 over carbon-13 over several million years. Because of this tiny selection process, the carbon in living things is slightly higher in carbon-12 content than the rest of the carbon in non-living things.

The testosterone naturally produced by the body is made from material that was once a carbon-12 selecting plant (which was eaten by a cow which turned into your steak, for example), so it has a similar isotopic carbon-13/carbon-12 ratio. Testosterone produced outside the body comes from the world of pharmaceuticals, where the carbon hasn’t been through the same million-year-old selecting cycle. Because of this, synthetic testosterone tends to be slightly richer in carbon-13 than natural testosterone. WADA uses an isotope ratio mass spectrometer to check the isotopic ratio of testosterone in a urine sample. If it deviates from natural testosterone by three carbon-13 atoms per thousand or more, it’s marked as a failed test, which indicates the presence of exogenous testosterone.

It’s this isotope ratio confirmation test that Cabrera and Colon failed. And unlike a simple T/E test, an isotope ratio test is essentially impossible to fake. But what about Ryan Braun? His T/E test was a reported 20:1, and his isotope ratio test also reportedly came back positive for exogenous testosterone. He won his appeal due to a chain of custody breach, and received a full acquittal. Officially, Braun hasn’t failed a test, so the mandatory three extra tests over the next calendar year won’t kick in. But that positive test is more than enough reason for the league to submit a notification of reasonable cause, and it’s a virtual certainty that Braun’s been tested as frequently as allowable. For the sake of baseball, no news is great news.

References & Resources

  • MLB’s Joint Drug Agreement can be found in pdf form here.
  • WADA’s testosterone test procedure is here, if you wanted to know the gory details.
  • A great article from 2006 on the Floyd Landis cycling steroid case is here. It also has a (deprecated) link to a published study on the testosterone/epitestosterone ratio. The link is dead, but a nice graph showing the distribution of the T/E ratio among healthy male athletes is still up.
Print Friendly
 Share on Facebook1Tweet about this on Twitter1Share on Google+0Share on Reddit0Email this to someone
« Previous: ‘Football scores’ in baseball
Next: Matt Moore is a fantasy stud »


  1. Dan Lependorf said...


    Hmm, you know, that’s a great question, and I don’t think I really have a good answer for you. You’re absolutely right in that the incentive to make a “natural” testosterone product is enormous. I see two possibilities:

    1. It’s been done, and the reason we haven’t heard about it is because it’s still a closely guarded secret.

    2. It can’t be done, for either biochemical reasons or financial cost reasons.

    Most industrial chemicals are made from precursors that come from raw materials like petroleum, natural gas, etc, but as far as synthetic biological compounds, my understanding is that chemical manufacturers will start with plant extracts wherever feasible. The more steps there are in a chemical synthesis, the smaller the yield, so chemical manufacturers will choose the simplest and easiest way to go. I don’t know if testosterone is produced that way.

    That said, I’m a chemist, but I sit on the non-biological side of things. If there’s a biochemist or an industrial production chemist in the house who can shed some light on this, that’d be fantastic.

  2. Paul G. said...

    Interesting!  Though I do find it amusing that the article about drug testing is written by a guy who welcomes offers for cheap male enhancement pills. grin

  3. aweb said...

    So if a clever drug maker can get the C13:C12 ratio correct somehow, they can’t fail a test? I imagine that with millions of dollars and mega-rich clients, someone has managed to pull that off…is there a reason it is impossible? For instance, can you harvest testosterone from organic donors (blood samples, animals, etc), or just the raw materials to make it?

  4. Nuke LaLoosh said...

    It costs a few million to setup, I was talking to an agriculture biologist a year ago about his plans to add C-14 via CO2 into plants in order to trace the effects of global warming.  It was a much fancier setup than required for the testosterone ratio, yet only ran $8M/month for tons of CO2.  In this case, you just want to slightly suppress C-13, so a single pass thru a mass spectrometer should be sufficient – and that is a lot cheaper to run.  The result would be more expensive, and that would separate the successful well-paid player from the rest.

    I’d sad about Colon, he was an outstanding example of what stem cell body repair is capable of, and old guys love that stuff.  The amazing thing is that the medical work on Colon would have been legal in the USA if he had been a race horse.

  5. rubesandbabes said...

    Good links, good article – I am not sure if I believe you about the marriage of WADA and MLB.

    Furthermore, you are optimistic about the Frequency of Braun’s current testing, but without providing any data about this. 


    Using the phrase ‘two whole days” to explain why Braun got off is misleading, which is compounded by the lengthier science spiel following.

    Braun got off on a rules violation. The handling of the sample was never compromised, even after “two whole days.” People attacking the Braun bust on a science level are simply true believers.

    I would also mention you are providing no specific data of any kind for Melky and Bartolo, yet still a lot of numbers.

  6. rubesandbabes said...

    Following the link, it looks like Braun now has to eat 3 additional random tests a year, so maybe he get tested 5 times in 2012, and also an HGH test once in spring.

    (not claiming too much expertise – just making an internet post, here, kids.)

    Different than the article – I read that every player gets tested twice during the season, and then an additional random 1,400 tests are done annually, with the vast majority done in season.

  7. rubesandbabes said...

    1) Okay, sorry got this wrong in previous comment – it looks like because Braun got off, he would absolutely not be subject to more testing as the article implies. Offenders get an additional 3 tests per year, but since Braun is not an offender, does that mean he would not receive more than 2 additional tests maximum, total 4 annually?That’s what it looks like.

    We presume the stuff that Braun was taking goes out of the system overnight. Plus, his episode created a safe time of Saturday after lunch for steroid taking, since the “two whole days” crowd will believe a household fridge isn’t as good as a lab fridge, or whatever they can conveniently grab onto.

    2) Of the 1400 additonal tests, it looks like about 200 tests are done in the offseason. Well, it_is_not_possible_to_have_random_drug_testing_in_the_offseason_without_a_100%_whereabouts_program.

    Baseball does not have this, and therefore no offseason testing is random, more precisely offseason testing is at the player’s convenience.

    3) I would like the author to explain/restate the two test types in terms of “A sample” and “B Sample.” I think every test is done on the first “A sample,” but after reading this, now not sure.

    4) Can baseball really be acting in full concert with WADA when the players do blood doping, AKA platelet rich plasma transfusion, exactly the same as one of the main charges leveled by the USADA at Lance Armstrong? The answer is no, MLB has an in-house drug testing program, and is only using some of the WADA methods.

  8. rubesandbabes said...

    Okay, Hardballtimes one last –

    The Floyd link is a true believer “steroids don’t help” link from 2006, when even Floyd himself was proclaiming innocent.

    Article reads:

    “Floyd Landis tested positive on day 17 of the Tour de France — right before he made a miraculous comeback to retake the lead. Some people may speculate that testosterone may have helped him do this.”

    And then the very next sentence in bold font:

    “This is unlikely to be the case.”

    (Floyd Landis claims this was a bad bust, and that his Tour win is clean, but he also admits to a dirty PED using career, and he has done some other bad things, so you decide.)


    To have this extra testosterone is performance enhancing. PEDs work and the steroids are very helpful for recovery. Tom Verducci’s story gets it right.


    For the “steroids don’t help on the field” crowd, hey I don’t agree with you, but I have a task for you:

    How about starting to connect the dots from steroids don’t help to why Igor Gonzalez got pushed off the Hall-of-Fame ballot way too quickly?

    Not only does Igor Gonzalez not make the hall-of-fame, he doesn’t even get a candidacy. Jim Rice is in, Alan Trammell’s gonna get a long look, but not Igor.  And steroids don’t help, right? You guys didn’t stick up for Igor!

    So, yes I humbly and kindly call on the steroids are placebos crowd to start rooting for Rafeal Palmeiro, Sammy Sosa, Igor Gonzalez, and Miguel Tejada’s candidacies for the Hall of Fame. Come on, guys – it’s only fair, and you all haven’t spoken up about this so far. Thanks for your consideration and support down the road.

  9. Hank G. said...

    If the ratio test is not above the 4:1 level, is the sample ever tested further?

    If not, is there some reason that someone couldn’t take additional epitestosterone along with testosterone (or testosterone precursors) to keep the ratio down to a more normal level?

  10. Mike F said...

    “plants have been very, very slowly preferring carbon-12 over carbon-13 over several million years.”

    Not quite how it works.

    Living organism have always had a preference to using carbon-12 over carbon-14, because the lighter isotope appears to be easier to capture.  In fact, the oldest evidence of life is from carbon deposits in Greenland dating back to 3.8 BILLION years ago.  They have the same light carbon ratio that would be expected from living material.

  11. nuke 2 said...

    I cannot believe someone here would cite the HOF voters as having a clue.  Steroids do not help you see better, they simply help you repair better.  Worried about performance enhancement??  Explain how a Tommy John repair is not a performance enhancement.

  12. Dale said...

    Is it possible to take testosterone and epitestosterone together to get the enhancing effects of testosterone, but to avoid exceeding the 4:1 ratio?

  13. WaddellCanseco said...

    These people wrote a paper on how to detect a person taking synthetic epitestosterone to equalize the ratio.  The conclusion:

    The mean delta(13)C values for urine samples obtained from 43 healthy males was -23.8 per thousand (SD, 0.93 per thousand). Nine of 10 athletes’ urine samples with epitestosterone concentrations >180 microg/L (624 nmol/L) had delta(13)C values within +/- 3 SD of the control group. The delta(13)C value of epitestosterone in one sample was -32.6 per thousand (z-score, 9.4), suggesting that epitestosterone was administered. In addition, the likelihood of simultaneous testosterone administration was supported by low delta(13)C values for androsterone and etiocholanolone.

    I don’t know whether this is actually used by MLB

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>