Talking about Things You Don’t Understand

Some clarifications in response to some very confused statements over the past few days.

1.  Amphetamine is a particular molecule: It isn’t a category.  Stimulant is the category.  Amphetamine has nine carbon atoms, 13 hydrogen atoms and a nitrogen atom that are arranged in a very specific way.  There aren’t ‘types’ of amphetamine.  Here’s its picture.  If the atoms are different, or the positions are different, it’s not amphetamine.

There are, however, different products that contain amphetamine.  Adderall contains amphetamine and some other stuff.  Dexadrine contains amphetamine and some other stuff.  Over-the-counter products legally cannot contain amphetamine.

What Tara Ragan, AJ Allmendinger’s business manager, said on SiriusXM Speedway Thursday afternoon what that they hadn’t been told the substance.  I think she’s confusing the drug and the product in which the drug could have been.   The toxicology testing cannot tell you where the drug came from, only that the drug is there.

2a.  “nanograms” is not the appropriate unit to use in discussing toxicology tests.  The unit, as I described in detail, is nanograms per milliliter of fluid.  If the allowed limit is 250 ng/mL, then you would find 250 ng in one mL, 500 ng in 2 mL, 750 ng in 3 mL, etc.  A number in nanograms is meaningless without reference to the volume of fluid being tested.  If you don’t understand what you’re talking about, find someone who does or don’t talk about it.

2b.  Again, on SiriusXM Speedway, Ragan said that her comment on ‘nanograms over’ was made without knowing what the NASCAR limit is for amphetamine.  First, I’m incredulous that the testing information would be presented without the threshold.  Second, I’m incredulous that AJ’s camp would have made that important a statement based on a guess.  She cited a figure of 1000 ng/mL,  As I mentioned above, the federal guidelines are 250 ng/nL, which is four times smaller than the figure she used.  The DOT limit is 500 ng/mL.  “A few nanograms” means totally different things depending on the cutoff.

3.  It doesn’t matter how much he’s over – if he’s at all over, he’s guilty.  It doesn’t matter in terms of the rule.  Yes – you’re over, you’re guilty.  It does matter in terms of public perception.  If you hear that someone was picked up for drunk driving with a blood alcohol concentration of 0.09 or with a BAC of .22, most of us think two different things.  Most of us can conceive of misjudging having one drink more than we should have – but getting a BAC of .22 means you are totally, out-of-the-park plastered and I don’t think most of us could conceive of getting behind the wheel under those conditions.

4.  “I’m not a (doctor, toxicology, pharmacist, chemist) and I don’t understand everything”.  Perfectly fair.  Then find someone who does understand it and get them to explain it.

I suspect everyone in AJ’s camp is wandering around in that daze you get when something terrible happens.  When someone dies or your spouse leaves you, everything becomes confusing.  Angie Skinner suggested that they should get a crisis management specialist.  I think that’s the right idea because you need someone who is not emotionally caught up in the situation.

ADDED:   A couple people asked whether there is really no way to tell where the amphetamine came from.  Here’s the long answer:  The test only tells you that the molecule was there, but not where it comes from.  Caveat:  Adderall has a specific combination of amphetamine, dextroamphetamine, etc.  If you do the analysis and find that ALL the components of that prescription drug or other substance are there, that’s a pretty clear suggestion that it came from a particular drug that has a well-characterized composition.

The only analogy I’ve been able to think of is imagine that you were examining the contents of  dead person’s stomach because they had a peanut allergy and died.  You look in the stomach and you find peanuts.  There’s no way to tell where the peanuts came from if that’s all you found.  However, let’s say you find bread and jelly, too.  Then you can conclude that the person ate a peanut butter and jelly sandwich.  But with only the peanuts and not the bread and jelly, you can’t say.  Let me know if that’s not clear.

What is Does it Mean to be “Nanograms Over”?

As a nanomaterials research, I usually love seeing the word ‘nano’ in the news, but not in reference to finding out that AJ Allmendinger’s ‘B’ sample tested positive “within nanograms”.  I would have been very surprised if the B sample had come out any different than the A sample – a testing lab would not have gone public if they weren’t 99.99% positive their results were correct.  AJ has a lot of respect from people in the business who know him well.  We don’t know the specifics and shouldn’t speculate, but we can agree that, whatever the truth, it’s a very sad situation for everyone involved and the sport in general.

I’ve covered a number of issues about drug testing in general and answered some of your questions in previous posts.  I’ll reiterate that the limits at which a drug test is considered ‘positive’ are determined to take into account possible measurement limits of the equipment.  I wanted to address the statement by AJ’s manager specifically.

“This was not the news we wanted to hear and we will work to get to the source of what may have caused this. To that end, we have secured the services of an independent lab to conduct thorough testing on every product within AJ’s home and motor coach to find what might collaborate with his test, which created results that were within nanograms of accepted standards. We are working closely with NASCAR and Penske Racing to identify the next action steps in this process.”  (via The Daly Planet)

“Nano” is a metric prefix meaning “a billionth”.  A nanogram is one billionth of a gram.  A gram of sugar is about 1/4 teaspoon.  To understand a nanogram, go pour out a 1/4 teaspoon of sugar on a piece of waxed paper.

Divide the sugar into ten parts.  Each part is a decigram, or one-tenth of a gram.

Throw away nine parts and divide the tenth part into ten more parts.  Each part is a centigram or one-hundredth of a gram.

Throw away nine parts and divide the tenth part into ten more parts.  Each part is a milligram, or one-thousandth of a gram.

Do that six more times and one pile will be a nanogram.  We’re talking about really, really small quantities.

The problem is that “nanograms” is the wrong unit to use when talking about drug tests.  The quantity we’re interested in is nanograms per milliliter or ng/mL.  The limit for initial testing for amphetamine in federal drug testing is 500 ng/mL.  A secondary screen is considered to be positive if it exceeds 250 ng/mL.  We do not know whether NASCAR uses the federal guidelines or if they have stricter limits (which would be completely within their right to have), but at least the federal limit gives us an order of magnitude with which to compare.

Assuming the statement about the test being “nanograms over” actually means “nanograms per mL over”, let’s assume it was over by 5 ng/mL. If you assume the lower federal limit of 250 ng/mL, you’re talking about being over by 2%. That’s a very different situation that if the limit was 5 ng/mL and you were “a few nanograms” over because then you’re over by 20-100%.

When your science teachers annoyed you because you didn’t include the units in your answer, they weren’t doing it to be annoying.  This is exactly the situation they were trying to help you avoid.  The information provided is meaningless without knowing what the limit is and the correct units.  I realize that AJ’s people are probably reeling from this, but if you’re trying to explain how close the test was, you can’t do it with the information they provided.

If you were 2% over for a blood alcohol level of 0.08, your BAC would be 0.0816. Is your driving significantly more impaired at 0.o79 than 0.08?  Probably not, but that is the nature of having to define a line.  Drug testing is much like the requirements on the car that are tested during inspection.  There is a hard line and if you’re over, you’re over.  The lines are somewhat arbitrary, but they are well defined and known in advance.

Despite what many fans seem to think, it’s not our ‘right’ to know the specifics, but Allmendinger is in a no-win situation.  As I stated before, he almost has to disclose exactly what he tested positive for, what the limit was and what he tested if he is to redeem himself in the eyes of the fans.  Otherwise, he will be dogged by speculation that is probably a lot worse than the facts.

Right now, the Allmendinger camp is likely getting all the products they can find and having them tested to try to correlate the exact molecule found with the tests; however, given the lack of quality control of supplements (see this really well-written article by David Newton) and the fact that he may have been taking supplements from a bottle that he finished off and threw away, there is a low probability that he is going to be able to clear himself completely.






More Drug Testing FAQs

I cleaned out my office a couple of weeks ago and threw away all the papers I’d been saving from the Mayfield situation.  Figures.

Q:  I thought NASCAR had a zero tolerance drug policy.  What did it mean that Almendinger’s test was ‘over the threshold’?

A:  A threshold can refer to two things.

First, there are some chemicals that we all have present in varying amounts, like testosterone.  There’s a certain level that is acceptable and that level is different for men and for women.  If someone is above threshold, it means that they have more than would be normal.  The Olympics is struggling with this right now because some women have levels of testosterone that are much higher than the average woman.  This may give them a significant advantage in some sports.  Likewise, there’s a certain range of caffeine that might reasonably be expected to be found in your body.  If  you have more than that level, it indicates that you’re abusing caffeine.

The second use can refer to a test threshold.  If there are five amphetamine molecules in 10 mL of urine, you’re not going to find them with any test.  Likewise, identifying the precise molecule is difficult and it is sometimes possible that one molecule looks like another to the testing apparatus.  Different types of testing equipment  have different sensitivities. If you’re down around the limits of the equipment that’s being used, there’s another possibility for what ‘threshold’ means.

Q:  Phenylalanine?

Phenylalanine is a common molecule found in energy drinks and a lot of other places:  the breast milk of mammals, in protein-rich foods (dairy, seeds, nuts, poultry and fish) and in higher amounts in things like diet sodas, artificial sweeteners and energy drinks.  Phenylalanine is a by-product of how your body breaks down the artificial sweetener aspartame.

Phenylalanine differs from amphetamine by a few atoms, as shown in the figure at right.  Chemists use a shorthand – everywhere two lines join and there are no other atoms indicated, it means there’s a carbon atom and two hydrogen atoms there.  It’s just such a common configuration that everyone got bored writing it all the time.  It’s like using “$” instead of writing “dollars” all the time.

You may have heard that phenylalanine had the same backbone as common stimulants.  I’ve highlighted the backbone in red on the top molecule.  You can see that the carbon chain (the six-membered ring, plus the three horizontal lines) are the same.  Phenylalanine has a C-O-O-H group where amphetamine has a C-H-H-H group).

You may have heard that phenylalanine can show up as a positive on a drug test for stimulants.  It depends on the type of measurement used. A broad test like an immunoassay might confused the two, but the molecules have fundamentally very different weights.  Phenylalanine has a molecular weight of 165.19 g/mole.  Amphetamine has a molecular weight of 135.2084 g/mole.  Methamphetamine has a molecular weight of 149.233 g/mole.  I realize that sound s like a very small difference, but in chemical analysis terms, it’s a huge difference.

Q:  Does a product label really tell you what’s in the product?

A:  Great question.  How much Vitamin C is in an orange?  Depends on the size of the orange, how it was grown, how ripe it was when it was picked, etc.  Natural things have inherent variations.  If you have a prescription for a medicine, there is a certain level of precision required in manufacturing that medicine and the medicine goes through quality control tests to ensure that the right dosage is present and the medicine is not contaminated with some other medicine.

That only happens if the item is something that is regulated by the FDA (Food and Drug Administration).  Vitamins, nutritional supplements and energy drinks are not regulated by the FDA.  Take a look at, which tests supplements and vitamins and see how much variation they found in very common supplements like fish oils. You might be especially interested in their comments on energy drinks.  Supplements and other unregulated materials are often indicated by some disclaimer like “supports heart health” because the one area that the government does regulate these items is in how they can advertise.

In addition to natural variations, you don’t know what the factory making the product you’re taking was making before they started making your product.  Some products will have a warning that they are made in a facility that also processes tree nuts.  There’s no guarantee that every little last bit of nut was cleaned out, so if you’re allergic to nuts, you’re safer to just not chance it.

Q:  What are the chances of the ‘B’ sample coming back negative?

A:  Hard to say because we do not know the precise testing protocol.  A wide screen immunoassay indicates that there is a possibility of a problem.  If there is a positive on that test, a more sensitive test is performed.  I can’t believe that Aegis (the drug testing lab) would make a public accusation without having done the GC-MS test (which is described in my previous blog).  If the B-sample measurement is the same GC-MS test, I would be extremely surprised if they get different results.

Q:  What will the toxicologist do when the  ‘B’ sample is tested?

A:  I talked to a a toxicologist who served as an athlete advocate.  Their role is to ensure that all standard protocols are followed and that nothing has been overlooked.  They will look at the raw data produced by the measurement and provide their own interpretation.  “Interpretation” in the case of scientific measurements is much less fuzzy than in something like law.  Toxicology communities have agreed-upon standards, so both the testing company and the Allmendinger toxicologist are working on the same page.

Q:  Is an energy drink a ‘stimulant’?

A:  Yes.  So is a cup of coffee and so is an energy drink.  There’s a big difference between enjoying a cup of coffee in the morning and not being able to start functioning without the cup of coffee.  There’s even more difference when you’re talking about a college student who uses coffee and energy drinks to stay up all night studying and is so wired the next morning that they have a car accident.  Anytime you talk about drugs, you’re talking about matters of degree.

Q:  What will NASCAR do if the B sample comes back positive, but they are pretty sure it’s due to an energy drink or supplement?

A:  Please stop asking this.  We have no idea what NASCAR will do and right now, I’m pretty glad I’m not NASCAR because this is a tough situation.  NASCAR makes a big deal with car inspections that they don’t deal with intent, they just penalize the situation.  Even if you make an honest mistake, you’re going to pay the penalty.  We have tens if not hundreds of prior cases for tech inspection failures.  When it comes to drivers, we have a handful and the overwhelming majority of those were people who admitted using illegal drugs.


Infographic: Energy Drink Label Comparison

With all the talk about phenyalanine being a potential problem in energy drinks, thought I’d point out that fuel in a bottle isn’t the only energy drink that includes it.

Drug Testing FAQ

Updated 7/10/12.

Some clarifications to clean up the incorrect information circulating around in the wake of A.J. Allmendinger’s failed drugs test.  More questions – send them to admin(at)buildingspeed(dot)org and I’ll add them.

What Drugs are Prohibited?

NASCAR has a non-exhaustive list that includes:

  • Seven different amphetamines, including methamphetamine and PMA, a synthetic psychostimulant and hallucinogen.
  • Three drugs classified under ephedrine.
  • 13 different narcotics, including codeine and morphine.
  • Ten different benzodiazepines and barbituates.
  • Marijuana, cocaine, zolpidem, nitrites, chromates and drugs that can increase the specific gravity of urine.

Amphetamines are ‘uppers’, which decrease fatigue by increasing levels of the stress hormones norepinephrine (attention and responding/fight-or-flight reactions), and the neurotransmitters serotonin (modulates anger, aggression, mood, metabolism, etc.), and dopamine (increases heart rate and blood pressure).

Ephedrines are stimulants, appetite suppressants, decongestants. The molecule Ephedrine looks very similar to amphetamine. Psuedoephedrine is like the left-handed version of ephedrine and that’s what’s in over-the-counter decongestants like Sudafed.

Narcotics is an imprecise term that usually refers to anything that deadens the senses. Codeine, morphine, heroin, etc. fall in this category.

Benzodiazepines and barbituates are downers, which do the opposite of amphetamines and decrease the action of the central nervous system.

Zolpidem is a sleep aid found in drugs like Ambien).  Nitrites and chromates can be used to adulterate a urine sample and these substances have been used in the past to to try to cover drug use.   For example, the soccer player Hope Solo recently tested positive for a diuretic.  Diuretics would have not impact on performance; however , they can mask the use of other drugs and are therefore not allowed.  In her case, she took a pre-menstrual medication and was unaware that it included a diuretic.  She was let off with a warning.

“Drugs that can increase specific gravity of urine” is an interesting phrase. One routine check for adulterated samples is measuring the pH, temperature and specific gravity (the density of the sample relative to water) of the urine sample.  If the values are outside of the expected range, the sample is considered suspect.  This part of the rule isn’t about illegal substances, it’s about things that could hide the use of illegal substances. For example, if you drink large quantities of water (1-2 liters), most of it comes out in the urine and that dilutes anything else in there.

In typical NASCAR style, the list is non-exhaustive and those who are subject to testing are encouraged to check with NASCAR before they take anything.  reserves the right to test for anything they think could potentially impair a driver on the track. has a list from the 2010 rulebook.

What are ‘A’ and ‘B’ Samples?

Bob Pockrass has a very good summary of the process of collecting the urine sample, so I won’t repeat it here.   Splitting a sample into multiple parts is a standard testing protocol.  The sample is immediately split into an ‘A’ sample and the ‘B’ sample. The A sample is tested with a chemical test called an immunoassay. Immunoassay (IA) tests, as I’ll describe in a moment, are more general tests. The B sample is stored and there are specific protocols for how the sample must be stored. If the IA test shows one or more positives, then the A sample is tested using a more sophisticated test called Gas Chromatography-Mass Spectroscopy.

While companies guard against contamination zealously, there is always the possibility that someone grabs a pipette they think is new, but was already used for something else, or mislabels samples.  There are tons and tons of checks and balances in the testing system, but let’s say that this is the case.  That’s the reason for the ‘B’ sample.  The ‘B’ sample and the ‘A’ sample are identical:  the chances of someone making the same mistake twice are very small.  Allmendinger is allowed to have a representative present during the testing of the ‘B’ sample so that he has assurances that the procedure follows the standard protocols.  A trained toxicologist who is hired by the person being tested will be extremely picky and on the lookout for anything that might produce a false negative.

The GC-MS test is repeated on the B sample.  Often, only the test for the substance that was found is done because time is of the essence.  If the B sample tests negative, then the results from the ‘A’ test are voided and the driver is OK to proceed.   If the B sample tests positive, the driver has no recourse – he is out until he completes a substance abuse program NASCAR has designed for him.

How do the Tests Work?

Immunoassays work sort of like a jigsaw puzzle. There are antibodies (detectors) and antigens (targets). The two types of molecules bind to each other.

The molecule in the upper left of the picture above is our detector molecule. The circle represents something that can be detected – a fluorescent molecule, a magnetic bead or something similar. The molecules on the right-hand side are the molecules in the sample to be analyzed. Our detector molecule has a binding site – the fork pointing to the right, and that binding site will allow it to only link to specific shapes of molecules. (In reality, the binding is chemical, not by shapes, but I find the analogy to shapes easier to understand.) Our detector will only bind to triangle shapes, but there’s a problem here in that there are blue triangles and green triangles. Immunoassays are very sensitive – they’re good at detecting things, but they are not always as specific as we might like. The green triangles might represent a banned drug, while the blue ones are harmless – but both bind to our detector molecule.

Gas chromatography is a more sensitive test used to separate molecules by weight. The sample is run through a column so that all the molecules with the same weight as the target substance are collected. We demonstrate chromatography to kids in elementary school.  The analogy we use is to make two rows of kids the same length. Two kids are chosen to be ’sample molecules’. The first kids is allowed to just walk down the row. The second kid must shake hands with everyone in the row as he walks. We start both kids at the same time and, not surprisingly, the student who has to shake hands takes a lot longer to travel the same lengths as the student who doesn’t. In GC-MS, sorting is done by weight, so all the molecules with the same weight as the suspicious molecule are separated.

The molecules are then heated. All molecules break into fragments in rather predictable manner when heated, so the mass spectrometer detects fragments and you can infer what the molecule was before it was broken up, and how much of it there was.

Why not use GC-MS immediately?  It is more expensive and slower.

What are they Actually Detecting?

When drugs enter the body, they are metabolized.  Metabolized means that the drugs are chemically changed – broken down into other molecules or added to in order to form other molecules.  The rates and types of metabolize depend sensitively on the type of drug.  Amphetamine passes through the body into the renal system (the urine) mostly unadulterated by the body; however, other drugs are transformed by the body and what you test for are the products of metabolizing the drug. Some drug tests are therefore very indirect.

Could this be a False Positive?

A false positive is when your measurement indicates that the molecule you’re testing for is there, but either a) it isn’t, b) you’re detecting a molecule that looks like the molecule you’re testing for, but isn’t; or c) the molecule is there, but was not produced by anything on the list.  A lot of substances can cause false positives. For example, buproprion (an anti-depressant) can cause a false positive for amphetamine. So can an awful lot of other things.

However, the drug company asks drivers to provide lists of medication and the driver is given a chance to explain how the drug that was found might have gotten there through legal means.

Couldn’t A.J. Ask for a More Sensitive Test?

When we talk sensitivity, we’re talking about how much of a chemical can be detected.  You can’t detect a single molecule.  Different types of drugs (and their metabolites) can be tested for at different levels.

Blood tests are more sensitive for some drugs than others.  Hair tests have the ability to give you a longer term picture:  Drugs go out through the body in urine, but some are are embedded in the hair and remain there for a long time.

There’s two problems with asking for a more sensitive test.  First, they didn’t collect blood at the time of the urine test, which a week plus ago.  A test of current blood is an entirely new test and might not detect something that was there a week ago.  Hair doesn’t retain all substances, so depending on what drug they are concerned with, hair might not be an option.

Finally, some of these tests are much more sensitive, so if the drug was there in a less sensitive test, it’s definitely going to show up in the more sensitive test.

What is the Likelihood that the Lab Made a Mistake?

Regrettably, slim to none.  There are many checks and balances in the process – testing companies know that people’s livelihoods are on the line.  Given the high profile of this case, I suspect they ran multiple test to make sure.  If the B sample comes back negative, the testing company is going to have a major problem.  They must have had a very high degree of confidence in their results to release them and have NASCAR publicly take action.

Why Doesn’t NASCAR Identify the Drug that was Found?

Jenna Fryer makes some compelling arguments as to why NASCAR should identify the drug  publicly, but I’m a little more leary about crossing the line.  I believe it would be irresponsible for any details to be disclosed until the B sample is tested on the off-chance that the lab made a mistake.  The ‘A’ sample says “we have a reasonable enough suspicion to act”.  It’s like the proof you need to arrest someone.  For me, the ‘B’ sample is “proven guilty’.  Releasing any information until the ‘B’ sample is confirmed is wrong in my opinion.

Then there’s a privacy issue.  What if the drugs are legitimately prescribed and a doctor can make a convincing argument that a combination of drugs produced the positive test?  If a driver has mental health or impotence issues, should those become public because the treatment he is using caused a positive drugs test?

I suspect we will learn what the test results showed from Allmendinger’s camp because I can’t see how he restores his public image without being entirely transparent.  Compare the cases of Randy LaJoie and Jeremey Mayfield.  Randy ‘fessed up and said “I screwed up”, did what NASCAR asked and he’s back in their good graces.  Mayfield denied the allegations and tied up the matter in lawsuits.  He is in the midst of financial and legal troubles. Allmendinger can’t return to a racing career in any series without specifying exactly what happened, taking responsibility for it and completing the mandated rehab.  But I don’t think it is NASCAR’s place to release the information.

NASCAR has a very fair policy in terms of remediation and reinstatement.  The only time they shut the door completely is if you have multiple violations. If this was an innocent mistake (akin to the Hope Solo issue mentioned above), then it’s a simple matter of admitting you did something inadvertently and going through a bunch of counseling.  If someone has a substance abuse problem, the most important thing is helping them recover their health and their life.

NOTE:  Some of the information in this post was originally published by me on 5/15/09 on