Skip to Content

How Not to Do It: Dosing Volunteers

Well, here I post about the ethical problems of using normal volunteers in Phase I studies, and this story comes along. It’s not exactly an investigational drug trial – two students (in “Sports Science”) at Northumbria University in England were being given caffeine to measure its effects on exercise. But there was a bit of violation, of ethics, medicine, and common sense. The volunteers were supposed to get 300mg each, which should certainly be enough to start your engine a bit (three cups of coffee simultaneously). But they’d apparently switched over from caffeine tablets to powder, and the person mixing up the dose was new and inexperienced, and. . .well, you see where this is going.

But you may not see just how far it’s going: instead of 300mg, each student got thirty grams. Yeah, I’d say that’s a sign of inexperience – no alarm bells went off when they weighed out a big fluffy white pile of the stuff? Or when they tried to dose it? It was pointed out on my Twitter feed that caffeine goes into room temperature water at about 2g/100mL, although if they warmed it up, it probably dissolved pretty readily. But that must have been a bitter draught to get down.

It’s not clear just when someone realized that things might have gone a tiny bit wrong, but the story does mention “violent side effects”, which you’d have to think were neurological seizures, and those must have kicked in pretty shortly. The students were hospitalized immediately and put on dialysis, and appear to have recovered with no permanent damage, which is pretty remarkable (one of them has some short-term memory loss). Each of them apparently lost over 20 pounds during the whole thing, which makes me hope that this doesn’t catch on as some sort of idiotic emergency weight loss plan. Being hospitalized near death will take off the pounds, but it’s not recommended. There’s a literature report of a fatality with a 12 gram dose, so I think we can conclude that (1) taking multigram amounts of caffeine is extremely dangerous, and (2) these two students were fortunate to have survived, and without immediate medical care they might well not have.

In such cases, there are several things to worry about. Caffeine hits the adenosine receptors, and a dose of that magnitude is going to saturate all of them but good. So you have those immediate effects, but the blood levels will be such that whatever other targets the compound is capable of binding to are also going to get a solid jolt. Moreover, the liver is going to be trying to deal with the onslaught, too – caffeine is metabolized by various N-demethylations, and the CYP1A2 enzyme in particular will be slammed trying to deal with things. That will lead to unexpected changes in any other things it’s normally processing, for one, and it could also lead to some unusual caffeine metabolites as it gets shunted over to enzymes that normally don’t get a crack at it. That can be a real problem – it’s what happens with acetaminophen/paracetamol overdose, which sends that compound past the usual glucuronidation pathway and into oxidative metabolism, producing a toxic metabolite that then causes liver damage. Meanwhile, the kidneys are having to mop up many of these metabolic byproducts from the bloodstream, and they’ll be coming on in large amounts (thus the dialysis treatment).

So no, this is not recommended. Neither is giving human volunteers drug substances when you’re too inexperienced to notice that you’ve jacked two decimal places, either (involuntary shiver).

72 comments on “How Not to Do It: Dosing Volunteers”

  1. Big deck says:

    Sports “science” strikes again.

    1. Snooty chemist says:

      This is correct. It is relevant here that those calculating and weighing the dose and those taking were both sports scientists. You’d have to be educationally sub-normal to imagine that taking 30 g of any white powder is safe. For heaven’s sake, you could choke on it.

      1. SW says:

        Watch me eating 30 g of sugar!

        But yeah, WTF. I mean, did no one realize that the pile of powder was several times the size of the tablets they were using previously?

  2. mallam says:

    At such super norma and known to be safe doses, there are many conceivable reasons for the resulting dysfunction of kidney function; which could include saturation of all possible elimination pathways to the point where normal biological processes cannot occur (eg no normal kidney eliminations). If this is the case, normal kidney function could resume once the bulk of the caffeine “bleeds” through the system. Is there any observed evidence for a possible toxic metabolite? I know this is a commonly proposed reason for unwanted physiological effect, but most often is not found to be the underlying cause, and the proposal in absence of data certainly is not needed for the message.

  3. Fox Chaser says:

    Our new weight loss supplement is clinically tested!

  4. NC says:

    I note your scepticism at ‘sports science’ – FYI for those unfamiliar, this is a popular subject in UK universities, usually for people with less-than-stellar grades. I’m not sure how long it’s actually been around, but about a decade ago, every young male I know (and no females) seemed to be doing it. It’s not well-regarded, but it is a degree.

    I wonder what position those administering the terrifying dose held?

    1. Synthon says:

      My nephew did a Sports Science degree at Oxford Brookes University starting 17 years ago.I was a bit sniffy but it seemed to be a rigorous course, well taught. His father, a Med Stats prof, even thought Stats was well taught. I think the quality of the courses did vary a lot around the country, some emphasising sport rather than science. They seemed to generate a lot of self-styled nutritionist/ lifestyle guru types.

      1. Me says:

        Had a buddy doing Sport Science – ended up doing a fairly hardcore lab-based PhD in oncology in a quite high-profile group.

    2. anon says:

      Subject of sports seems to be some sort of antidote to science. On one hand there’s ‘sports science’ of, well, varying quality. On the other, sometimes scientists in other fields with no knowledge of sports want to get easy publicity by writing a ‘sporty’ article and then we get things like ‘in 2030, the men’s 100 metre sprint world record will be under eight seconds’.

  5. Mad Chemist says:

    Clearly someone thought, “It’s fine, caffeine’s in tea. It won’t be a problem.” Given that caffeine in plants functions as a pesticide, that’s not a smart approach.

    1. Andrew says:

      I mean, caffeine really is one of the safer drugs that’s ever been discovered. It’s got miles and miles between the effective dose and LD50, side-effects at reasonable doses are mild and temporary, and long-term tolerance and dependence basically don’t exist. Of all the things to screw up with, it’s one of the better choices.

  6. David says:

    Just reading that dose gave me a sharp twitch…

    1. DW says:

      I got jittery just from the number hitting my eyeballs. XD

  7. anonymous says:

    Caffeine for the breakfast?

  8. R/S says:

    Just think, the world nearly lost two future PE teachers!

    1. Dave says:

      More information on this here:


      P.S. Is there any truth to the rumor that the two former students are now multi-millionaires, after having successfully opened a high profit coffeehouse? Something about the high profits result from having zero expenses? 😉

    2. Anonymous says:

      That remark is inappropriate- they could have lost their lives. Who are you to judge other people’s educational and career choices?

  9. Joy says:

    Zeroes matter.

  10. Barry says:

    If memory serves, oxidation-mediated paracetamol toxicity kicks in when our sulfation (not our glucuronidation) path is saturated.
    But yes, the modes of metabolism and therefore toxicity can change drastically with dose.

  11. John Campbell says:

    The ethics protocol for this sort of thing should always ensure all dosing is double checked as nurses do. It is incredible that no one involved stopped to think- five tablespoons each FGS!

    1. NJBiologist says:

      Yeah, I was wondering if IRB approvals specified dosing forms….

    2. Nick K says:

      Unfortunately, a substantial proportion of British school-leavers are functionally illiterate and innumerate as well as lacking in basic common sense, and some of them end up in Universities.

  12. CMCguy says:

    Does the UK use a decimal period or a comma like others in EU? I know I have had to retrain my self when reviewing batch records from that part of the world about this difference in conventions. I saw error due to calculation on a cell phone where decimal not clear? Typically see clinical and lab dosages written out or on forms that get double checked to avoid such mistakes and assume is in GCP.

    As Derek has often pointed out old adage its can be the dose that makes the poison and this clearly demonstrates that caffeine which has know and generally mild effects can cross that line.

    1. John Dallman says:

      The UK uses periods for decimal points, like the USA.

      1. Some idiot says:

        Yep… it is continental Europe that tends to use a comma instead.

        1. dearieme says:

          We once had a Chancellor of the Exchequer (Finance Minister) who didn’t understand decimals, and complained about “those damned dots”.

  13. bacillus says:

    “The staff were inexperienced”. In what? Reading, weighing or just plain common sense? I sure hope they weren’t real lab techs.

  14. Anon says:

    30 grams???!!

    What, were they taking it Scarface style?

    1. Anon says:

      No, rectally.

  15. Paracelsus says:

    “What differentiates a poison from a medicine is its dose”

  16. Diver dude says:

    None of the reports mention a protocol, IRB review or observance of any GCP guidelines. If none of this was done, then this procedure constitutes assault, in my view.

    I spent 30 years designing, running and participating in dozens of First in Man and Phase I studies. The regulations in the UK are clear and the legal penalties for non-compliance are severe. If I were one of these students, I would be consulting my lawyers.

    And to Derek question yesterday? It has become my view that First Time in Man studies in healthy subjects ARE unethical. I have designed and run 10s of these studies on small molecules and biologics and the only upside for the volunteers is the money and the potential downside is huge. A big problem is that, with biologics, we don’t even know what that downside might be. I might be persuaded that small molecules are more acceptable, but not biologics.

    1. Isidore says:

      The other (serious) issues notwithstanding, a caffeine ingestion study hardly qualifies as “first in man”.

      1. Diver dude says:

        Absolutely true. And under the current UK legislation it is arguable about whether this constitutes a “clinical trial” under the meaning of the act (as caffeine is neither a “prescription only” medicine or an “investigational product”). However, there are established norms for this type of procedure and a clear duty of care for the institution. At the very least , what they are described as having done consititutes pharmaceutical “compounding” which makes this a new formulation of caffeine and there is a large and scary body of UK law around that. Its why we have pharmacists.

    2. Mark Thorson says:

      Back in the old days, wasn’t the n=1 first-in-man study done by the inventor in himself.

      1. Patrick says:

        Alexander Shulgin did this for several hundred compounds, although they were related chemically (phenethylamines and tryptamines), without even testing them in animals first.
        He had a protocol that involved starting with very small doses and gradually increasing them while looking for certain early warning signs.
        No idea of how effective this actually was, but he survived the experiments and lived to be 88.

        1. milkshaken says:

          though in poor health in his last years, which could have been age related. Giving the world MDMA, DOM and 2C-B, he managed not to kill himself but others were more idiotic. By becoming a guru of garage synthesis enthusiasts, he stimulated lots of bad behavior (all that notoriety actually hindered research of useful drugs like MDMA, predictably, as they became Schedule 1.)

      2. Ken says:

        Go back far enough, and the first-in-man studies involved figuring out which berries were poisonous. I wonder how many fatalities, and how much desperation, were involved in the discovery of how to safely eat cassava, rhubarb, and taro.

        1. Anonymous Researcher snaw says:

          To say nothing of fugu!

        2. Jamil Gregory says:

          There is a good section in the SAS survival guide with a protocol to test plants/fungi for survival in the wilderness. Foraging is not as dangerous as people make out.

          1. fajensen says:

            No, not *as* dangerous … but … happy-go-lucky Foraging will absolutely kill someone, especially with the “Nordic Kitchen” gaining public popularity and odd herbs garnishing every “viking-style” dish.

            I “do” wild mushrooms myself, I very deliberately stick to a few species where the “screwup-potential” is very limited. Because they are easy to identify or because most are not really that poisonous (the “sponge” mushrooms).

            Apart from the usual suspects, mushrooms, we have the Aethusa Cynapium – which looks like parsley – and Conium Maculátum (hemlock), Cicuta Virosa (another kind of more poisonous hemlock or maybe the same, experts differ) – which can be mistaken with – Angelica Archangelica rsp. Litoralis which are edible.

            Those plants are not very forgiving. A few grams will put a crimp in ones retirement plans!

            The mushrooms kills Chinese every year because they pick something they think is a Chinese mushroom and is not. – wierd structure, will kill the liver, nasty stuff.

          2. Derek Lowe says:

            Just mention of the genus Amanita is enough to send me in other direction!

          3. Curt F. says:

            Derek, Would Amantia be fall on your TIWWW list? I’d avidly read the natural products branch of your TIWWW list, if it exists… 🙂

          4. Mark Thorson says:

            I’ve seen cooking shows that showed oxalis and sorrel gathered from the wild being used in salad. Not lethal, but I wouldn’t do it. Their tart flavor is oxalic acid, which will give you kidney stones. That’s why you shouldn’t eat the green part of rhubarb. Even spinach can give you a stone, as happened to me after a few months of eating a big spinach salad every day.

  17. daisyj says:

    Reminds me of a (far less terrifying) story from my misty past as a Bio 1a TA. One of the lab projects the students were supposed to do involved making serial dilutions of saliva and using a colorimetric assay to determine the amount of amylase. The assay was done in 5 ml tubes, and you were supposed to collect a small amount of saliva from one member of the team and dilute it into the first, and then on down the line. But at least one of the groups failed the reading comprehension portion of the task, and spent a good ten minutes diligently spitting into the tube, until they had collected the full 5 ml of saliva. Needless to say, they didn’t end up generating a very useful curve from their data.

  18. Li Zhi says:

    So, either the solution was created by someone who didn’t believe it would be ingested undiluted (hence a communication/protocol failure) or by someone unqualified (or rather incompetent) to dispense potentially toxic medicines. I wonder what the basis is for the claim that they “appear to have recovered with no permanent damage…” Almost certainly there’s been permanent damage, depending on how precise (or sloppy) you want to be about defining “permanent”. How many concussions are required to cause “permanent”damage? Let’s see, first compare their physiologic capabilities to their baseline (what? no baseline?) then biopsy their livers, kidneys, and heck, why not?, brains and compare those slides to pre-incident biopsies…And then autopsy them…not a problem. This should be filed in the “things that should never happen” category. The PI should be held accountable, if it was the system which enabled such incompetence. (There’s always a chance of down-stream negligence or malice to consider.)

    1. Some idiot says:

      I can recommend the article that Dave linked to (above). It has a bit more detail. Apart from other things, it happened in 2015, and a court has just fined the university £400 000 (for, to paraphrase, negligence and extreme stupidity). The article reported (presumably from the court proceedings) that the investigator calculated the dose with a calculator on a smart phone, and forgot (or misplaced) the decimal point.

      My father (who was not a chemist) always impressed on me the importance of having a feeling as to what was real or not. This guiding principle has served me well over the years. I feel very strongly that if the investigator here had had more common sense, this would not have happened. Or, as the quote says, “the thing about common sense is that is not so common anymore…”

      And I agree 100% with Diver Dude that this is something that just should not have happened Forsikring many different reasons. I am just relieved that no one died…!

      1. Some idiot says:


        “should not have happened for so many different reasons”

      2. PV=nRT says:

        reminds me of the story where the aerosol manufacturer was calculating how much propellant to put into the cans and plugged the ºC temperature instead of absolute temperature into the ideal gas equation…when the cans started exploding on the assembly line they realized that there might have been a mistake…

  19. Wavefunction says:

    It’s very easy to get misled by pure caffeine powder because of its potency. Even a teaspoon or two of the pure stuff amounts to several grams, and just a few years ago there was a report of a fatality resulting from an undergraduate stirring a few teaspoons into his coffee. Since then I believe the FDA has clamped down on sales of pure caffeine powder. The subjects in the report above were extremely lucky to have survived.

  20. Xiaomiao says:

    30 grams of caffeine would be quite a task to get through, I think I’d prefer the seizure part.

  21. biologist says:

    Adenosine receptors? Not too much worry any more at this dose. 30 g caffeine is 2 MILLIMOLAR. At 2 mM, caffeine hits the ryanodine receptor, and calcium starts flowing out of the endoplasmic/sarcoplasmic reticulum. Nearly every aspect of cell physiology has some connection to calcium. The ultimate cause of muscle contraction? Calcium. Learning and memory? Calcium. Etc.

  22. Anonymous says:

    1. You can still get caffeine pills as a nutritional supplement on the internet, e.g., 200 x 200 mg tablets for $14. (Pure powder is $37 per kilo, not sold to individuals, but I’m not sure how they authenticate whether you are a qualified business.)

    2. Calculation errors. As pencils and slide rules transitioned to calculators, educators lamented that students would lose their understanding or intuition for powers of 10 and just trust the calculator. “And the diameter of the earth must be 12.7 cm. The calculator says so.” (I still have my 6″ circular slide rule with scales that spiraled around to be roughly equivalent to a 3 foot linear slide rule.)

    3. I used to post “sports science” and “nutrition science” papers on the bulletin boards at the gym in the hope that undergrads would read them and see where the latest Oprah or People Magazine headline came from. Some of those studies are pretty bad but that doesn’t stop the popular press from generalizing. Many such studies from sports sci, nutrition, even psych departments are (a) small – 10-100 subjects (b) college students – not exactly indicative of your typical couch potato in terms of age, general health, education level, … yet the results are freely extrapolated to the population at large. (c) … other problems. I don’t have my old PDFs at hand, but some of the popularly reported studies were pretty bad. Some were pretty good, even though conducted at small departments, presumably on minimal budgets.

    Time for a snack and a Jolt Cola: “All the sugar and twice the caffeine,”

    1. Other Anonymous says:

      “(Pure powder is $37 per kilo, not sold to individuals, but I’m not sure how they authenticate whether you are a qualified business.)”

      I think the answer is that they don’t – either they don’t ask, or they take whatever story you tell them at face value. A kid in my high school class (this was around 2003) was expelled for dealing it – a dollar or two for a teaspoon in your water bottle. At university some students apparently used it also, this time less as a study aid than as something to put in liquor. Unless there’s been some dramatic crackdown, I reckon it’s still probably a substance someone can just go and buy online.

  23. Anonymous Researcher snaw says:

    The first question that occurs to me is, why were these idiots using pure caffeine in ANY form in the first place? In the US there are plenty of commercial products, already formulated for human consumption, with stated amounts of caffeine (one well-known brand being NoDoz). Don’t such also exist in the UK?

  24. David Cockburn says:

    Many years ago I was involved in a case when an ‘inexperienced’ person gave a 1g bolus of lidocaine to a cardiac patient rather than the intended 100mg. The 1g pack was intended to be added to a litre of D5W and given as a drip. The experience taught me that packaging cannot be too fool-proof.

    1. oldnuke says:

      The problem is that schools keep producing better fools.

      Innumeracy seems to have grown since the advent of handheld calculators. I grew up with slide rules, log tables, and back-of-the-envelope math and I still have a habit of desk-checking the results of new software programs.

      You should see the looks on the new comp sci students when I whip out the trusty glucose-powered analog computer (slide rule) in class. 🙂

      1. M. Welinder says:

        > Innumeracy seems to have grown since the advent of handheld calculators

        Are you sure? [Insert quip about correlation and causation here.]

        There used to bit a lot of jobs with no more math requirements than
        counting to 20. You just did not *observe* innumeracy in people with
        those jobs regardless of whether it was present.

  25. dearieme says:

    “a staff researcher had calculated the dosages on a mobile phone”: just think, if this had happened in the US they could have sued Apple.

  26. Chris says:

    In the school district where my kids went to school, calculators were commonplace (required in some cases), so many early math and science classes had sections on “estimation”. Here you would estimate a result before calculating the actual result. I think it was a very good idea.

    1. loupgarous says:

      Great essay/cartoon on “Fermi estimation”:

    2. tangent says:

      I am all in favor of this combination, calculators combined with one-sig-fig estimation. Or even zero sig figs, order-of-magnitude estimation, and then you don’t even have to know your times tables.

      But none of this is proof against the grade of fool who would proceed with thirty goddamn grams of material. What do you even.

  27. milkshaken says:

    Caffeine is pretty fluffy, we had a 500g widemouth bottle of the stuff from Fisher, and the bottle was huge, like a kitchen jar. And the taste is fairly disgusting – you need something pretty strong-tasting, like tea or coffee, to mask 100mg of caffeine in a cup, and there is a good reason energy drinks are so sweet and over-flavored. 30g is going to be a nice pile of fluff.

    I am amazed that no-one noticed anything strange, up to the point two people ate 30g of pure caffeine – I think one would have a strong reaction even after consuming 30 g of table salt.

  28. Kaleberg says:

    I’m impressed this happened given the levels of approval needed for doing an experiment with human subjects. I was just reading a paper comparing users’ efficiency and attitudes towards Microsoft Word as opposed to Latex in producing scientific publications, and it had a long, careful explanation of the protocols and selection procedures involved. There was no way anyone was going to get a 3 gram dose of Latex or a 4 gram does of Word. (Those would probably be fatal.)

    [You can insert your own joke about having people use Microsoft Word and/or Latex here.]

  29. AndrewD says:

    Of Topic (and I apologise) but has our resident critic of Dementia theories seen this

    1. Anonymous says:

      Play nice.

      1. Mark Thorson says:

        Do not tease the bear.

    2. fajensen says:

      There is much more to be said on this subject, surely. The latest CERN Courier linked this:

      “Gamma frequency entrainment attenuates amyloid load and modifies microglia”

      So, if mice were watching old-style CRT TV they would be safer with respect to Alzheimer.

  30. Crocodile Chuck says:


  31. BK says:

    A few years ago, I read an instance where some woman tried to commit suicide by using caffeine. The exact dose she used is mixed up in my head for some reason; it was either 50 grams, or 200 grams. I obviously forget. But it was published in a medical journal somewhere. I cannot seem to find it but I cannot imagine how awful that experience was, or even these poor souls. Back in my heaviest stim-using days, I only made it to 1g daily but that was before every meal.

  32. Bla says:

    Sports science is the only two words you need to read to know why this happened.

  33. Chris Phoenix says:

    Here’s a fun “first in humans” story, told by a researcher who self-dosed – at the time (during WWII), he was looking for a better malaria drug:

    “Having already ingested 33 grams of assorted naphthoquinones without discomfort, I became the second subject (Case 29). At the same time that I went on drug regime Dr. Astwood started administering the propionate to mice. We joked about the concurrent tests that evening at a conference of cooperating groups at the Boston Psychopathic Hospital, after which I boarded the Federal Express bound for Edgewood Arsenal. My boys did not know where to reach me the next morning to warn me of a call from Astwood stating that his mice were dying off rapidly. I needed no warning. Never in my life have I felt so wretched as I did on day 2. I concluded my business at the Arsenal as soon as possible and spent the rest of the day stretched out on a cot in the officers’ quarters. Drug regime for just one day had brought on a condition of malaise which lasted for about 36 hours. Edgewood doctors persuaded me to stop taking propionate pills and drew blood samples which were subsequently analyzed for the record. The blood levels achieved (13%) were twice those previously attained, but the price paid was excessive. The clinical trial was stopped.”

    Earlier in the chapter are fun snippets like:
    “a group of conscientious objectors who had volunteered for experiments in medical research. ”
    “The work on drug metabolism eventually included 75 trials in human subjects and provided data on metabolic oxidation products from 19 naphthoquinones.”
    “For one thing, he lined up as subjects a number of polycythemic patients who reported regularly to the hospital to be bled for relief from excess red blood cells;”

    and last but not least, “When given at maximum level of food intake, hydrolapachol had killed 30% of the mice tested but still was rated safe enough for human consumption.”

    From _The Scientific Method_ by Fieser, pdf searchable by Google.

Leave a Reply

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

Time limit is exhausted. Please reload CAPTCHA.