Start Here TDG Five-Test Programme Resilient Gut System Stress Programme Initial Consultation — £145 Movement & Training AIdan Concierge Products Blog Learn Clinical Diagrams Resources Beyond the Mainstream Book a Call

What’s In Your Cup · Supplements · Evidence graded

Your Coffee Isn’t a
Delivery System.

Most of what social media tells you to stir into your coffee is either too small a dose to matter, poorly absorbed, or wrong for someone reading it. A few additions are genuine upgrades. None of them is right for everyone.

Stephen DuncanFDN-P MSc BSc
PublishedSeptember 2026
Reading time13 minutes

I have spent nearly forty years working with people, and a great deal of time and money on training along the way. Almost all of it comes down to one lesson: the right answer depends on the person. Your hormones, your gut, your liver, your medications, your blood results.

Then I open my phone and someone with a ring light tells a million strangers to add two drops of CBD oil to their morning coffee. No questions asked. No tests. No mention of who should not do it.

This post takes the most popular coffee add-ins and asks the questions the videos skip.

How coffee became a delivery system

Bulletproof Coffee turned a cup of coffee into a product platform, and the template it set is still running.

Dave Asprey began selling "low-mould" coffee beans, oils and supplements online in 2011 and left his tech job to run the Bulletproof companies by 2013 (Wikipedia). The recipe was coffee blended with butter and a branded MCT oil. The pitch was that ordinary coffee carried performance-robbing mould toxins, and his beans did not.

It worked commercially. Bulletproof raised at least $69 million, went nationwide and opened at least two cafés in California (Daily Coffee News). It never became a café chain: both cafés closed, and Asprey stepped down as CEO in 2019. By then the range covered coffee, protein bars, supplements and even a $1,495 vibrating plate (Seattle Times).

The telling detail came in 2025. The brand, now part of a larger food group, relaunched with new packaging and without the mould-and-toxin message it was built on (Daily Coffee News). Roasted coffee sold in the UK and EU already has legal maximum levels for the main coffee mycotoxin, ochratoxin A (Regulation (EU) 2023/915). The fear sold the product; the product outlived the fear.

That is the pattern to watch for: a real-sounding problem, a proprietary fix, and a daily habit to hang it on. Coffee is the perfect hook because you were going to drink it anyway.

Four questions for anything you add to your cup

Before you add anything, it should pass four checks. Most social posts answer none of them.

  1. Is the dose meaningful? Compare what goes in the cup with the dose used in actual studies.
  2. Does it get absorbed? Oils float, powders clump, and some compounds barely cross the gut wall.
  3. What is the evidence, and in whom? A mouse study, a single small trial, or repeated human results?
  4. Who should not take it? Medications, pregnancy, liver or kidney health, thyroid, iron status.

The fourth question is where personalisation lives. It is also the one a 30-second video never has time for.

CBD in coffee: too little to help, enough to matter for some

One or two drops of CBD oil in coffee is almost certainly too little to do anything. For people on certain medications, though, CBD still deserves care.

Dose. The UK Food Standards Agency advises healthy adults to keep CBD from food to 10 mg a day, which it puts at about 4 to 5 drops of a 5% oil (FSA). So one or two drops is roughly 2 to 5 mg. Clinical trials use hundreds of milligrams. Two drops is a flavour, not a treatment.

Absorption. Swallowed CBD is poorly absorbed, and fat changes that a lot. In one trial a high-fat meal raised blood levels of CBD roughly four- to five-fold (Taylor et al., 2018; the authors worked for GW Research, maker of the prescription CBD medicine Epidyolex). Oil floating on black coffee is about the worst delivery you could design. Hot coffee is not the main problem; dose and dispersion are.

Interactions. CBD slows several liver enzymes that clear medicines. Caffeine is one of the drugs it affects. At 750 mg twice daily for weeks, CBD roughly doubled total caffeine exposure in healthy volunteers (Thai et al., 2021; GW Research authors again). In an independent university study, a single brownie containing 640 mg of CBD (plus 20 mg of THC) raised blood levels of omeprazole by 207%, losartan by 77%, midazolam by 56% and caffeine by 39% (Bansal et al., 2023).

The liver. In a small trial of healthy adults taking 1,500 mg of CBD a day, 5 of 16 developed liver enzyme rises that met the international criteria for drug-induced liver injury, all within two to four weeks (Watkins et al., 2020; two authors worked for GW's US arm). That is a medicinal dose, far above a coffee drop, but it is why the regulator's limit is low.

At two drops, none of that is likely to matter for your coffee. The mechanism does matter for anyone taking higher-dose CBD alongside medicines cleared the same way, such as blood thinners, some anti-seizure drugs, some antidepressants and acid-reducers.

Who should avoid it. The FSA says under-18s, anyone pregnant, breastfeeding or trying to conceive, and people taking medication should avoid CBD, because safety evidence is lacking (FSA). It cut its limit from 70 mg to 10 mg in 2023, citing possible long-term liver and thyroid effects (CIEH).

Verdict: mostly pointless at social-media doses. Not a daily habit for anyone on medication without checking first.

Genuine upgrades

A few additions stand up to the four questions, and they are cheap.

L-theanine, or switching some cups to tea. This is the best-evidenced pairing with caffeine. A meta-analysis of 11 placebo-controlled trials found moderate improvements in alertness and attention-switching in the first two hours when caffeine and L-theanine were taken together, though the size of the effect tracked the caffeine dose more than the theanine dose (Camfield et al., 2014). Tea already contains both. If coffee makes you wired and edgy, tea for the second cup is a simpler fix than any powder.

Protein. Adding a scoop of protein to coffee ("proffee") is useful for people who otherwise run on coffee until lunch. The benefit is the protein, not a special effect of mixing it with coffee. Heat does not ruin its nutritional value, but whey can clump in very hot coffee, so blend it or let the coffee cool a little.

Matcha: an upgrade instead of coffee, not in it. Matcha is powdered whole tea leaf, so you swallow the leaf rather than an infusion. That gives you caffeine plus L-theanine, the pairing above. Europe's food safety authority considers catechins from traditionally prepared green tea drinks generally safe. The liver concern is concentrated extracts: taken as a supplement, 800 mg of EGCG a day or more raised liver enzymes in trials, and the authority could not name a dose of extract it considered safe (EFSA, 2018). Stirring matcha into coffee, as in a "dirty matcha", mostly stacks two doses of caffeine. Swap a cup; don't combine them. Like other tea, it also reduces iron absorption with meals.

One more upgrade costs nothing: drink coffee with food, not instead of it. A coffee that replaces breakfast removes protein, fibre and micronutrients from your morning, whatever you stir into it.

Mostly pointless or overpriced

These are rarely dangerous for healthy people. The problem is that you pay a premium for a dose or an effect that is not there.

Mushroom coffee. Lion's mane is the headline ingredient. The most cited human trial gave 30 older adults with mild cognitive impairment 3 g of powder a day for 16 weeks; scores improved, then fell again within four weeks of stopping (Mori et al., 2009). It was small, and its lead author worked for a mushroom producer. More in Fungi of the Month: Lion's Mane. Many blends list a "proprietary" amount, so you cannot tell whether you get anywhere near that. What you reliably get is less coffee per spoon, so less caffeine, at a higher price.

Butter and MCT oil ("bulletproof"). A full cup can carry several hundred calories, most of them saturated fat, with no protein, fibre or micronutrients. MCT oil commonly causes stomach cramps and loose stools until people adapt. Some people see a marked rise in LDL on this pattern; others do not. That is a question for a blood test, not a guess (what the LDL and statin evidence actually says).

Collagen. It is a protein, so it counts towards intake. It is an incomplete one: it lacks tryptophan, an essential amino acid (Paul et al., 2019, written by authors from the collagen industry), so it is a poorer choice than whey or a complete plant blend if protein is the aim.

Maca. Maca is sold for energy, libido, hormones and fertility. The human evidence is small and mixed. A 2022 meta-analysis of randomised trials found no clear improvement in sperm concentration versus placebo, and called the trials too few and too small for firm conclusions (Lee et al., 2022). A 2024 review of its antioxidant effects noted that human data are scarce (Huerta Ojeda et al., 2024). It is a food with a long traditional use, not a proven supplement. If you like the taste, fine; if you are paying for hormone effects, test the hormones first.

Risky for some people

These four are sold as natural, daily and harmless. For the wrong person, taken daily, they are none of those.

Chaga. Chaga is very high in oxalate (Fungi of the Month covers it in October). A 69-year-old man taking 10 to 15 g of chaga powder a day, plus vitamin C, developed acute kidney injury within three months, needed dialysis, and recovered after treatment (Kwon et al., 2022). Another man reached end-stage kidney disease after years of daily chaga (Lee et al., 2020). These are case reports, not proof of common harm. They are a clear reason for anyone with kidney stones or reduced kidney function to avoid it.

Cassia cinnamon. Most supermarket cinnamon is cassia, which can contain up to 1% coumarin, a compound that can harm the liver. Coumarin from cinnamon is absorbed almost as well as the pure compound (Abraham et al., 2010). Germany's risk assessment institute puts the tolerable daily intake at 0.1 mg of coumarin per kg of body weight, and says a 60 kg adult reaches it with about 2 g of cassia a day, less than a level teaspoon; a 15 kg child with about 0.5 g (BfR). A pinch is fine; a daily heaped spoon is not, and Ceylon cinnamon is the safer habit.

Ashwagandha. Liver injury from ashwagandha is now documented on several continents. A US and Icelandic series described jaundice starting 2 to 12 weeks after people began taking it (Björnsson et al., 2020). In an Indian series, three people with existing liver disease developed liver failure and died (Philips et al., 2023). Denmark has made products containing it illegal to sell, on the strength of a Danish risk assessment that raised concerns about effects on sex hormones and reproduction (Danish Veterinary and Food Administration). It can also raise thyroid hormone levels. More in Supplement of the Month: Ashwagandha. It is not a coffee-shop topping.

Turmeric. Turmeric has two problems pulling in opposite directions. First, curcumin is barely absorbed: in volunteers given 2 g of curcumin, blood levels were undetectable or very low. Adding 20 mg of piperine from black pepper raised absorption about 2,000% (Shoba et al., 1998) — in a small single-dose study, which is where that famous figure comes from. So a spoon of turmeric in a latte delivers very little curcumin.

Second, the products that fix absorption are the ones that raise the stakes. Piperine works by blocking the same liver and gut pathways that clear many medicines. Turmeric is now among the main supplements linked to herbal liver injury in the US Drug-Induced Liver Injury Network (Halegoua-DeMarzio & Navarro, 2024). As a spice in food, turmeric is fine. As a daily high-strength or piperine-enhanced supplement, anyone with liver issues or on regular medication, including blood thinners, should check first.

The bit nobody posts: timing your coffee

The most clinically useful coffee advice has nothing to do with add-ins. It is about when you drink it.

Iron. A cup of coffee with a meal cut iron absorption by 39%; tea cut it by 64% (Morck et al., 1983). Coffee an hour before the meal made no difference, but coffee an hour after reduced absorption just as much. If your ferritin is low, drink coffee well before meals, not with or straight after them.

Thyroid medication. Swallowing levothyroxine with coffee reduced its absorption by roughly a quarter to a third. Coffee taken 60 minutes after the tablet had no effect (Benvenga et al., 2008). Anyone on thyroid medication whose TSH will not settle should check this before changing the dose.

These are exactly the kind of findings a blood panel picks up and a social post never mentions.

Quick reference

Add-in Verdict Main issue Be careful if you…
L-theanine / tea Upgrade Well supported for calm focus Are sensitive to caffeine overall
Matcha (instead of coffee) Upgrade Caffeine plus theanine; don't stack with coffee Have low iron or are caffeine-sensitive
Protein Upgrade Benefit is the protein itself Have a milk allergy (whey)
CBD, 1–2 drops Mostly pointless Dose too low, oil floats Take medication, are pregnant or trying to conceive
Mushroom blends Overpriced Doses rarely match trials Pay by the claim, not the label
Maca Overpriced Small, mixed human evidence Are buying it for hormone effects untested
Butter + MCT Situational Calories, no nutrients, GI upset Have raised LDL
Collagen Situational Incomplete protein Rely on it as your main protein
Chaga Risky for some Very high oxalate Have kidney stones or kidney disease
Cassia cinnamon Risky in large amounts Coumarin, liver Use it heavily every day
Ashwagandha Risky for some Liver injury, hormones Have liver or thyroid issues, are pregnant
Turmeric supplements Risky for some Poorly absorbed alone; enhanced forms raise liver and drug risks Have liver issues or take medication
Coffee itself Timing matters Blocks iron and thyroxine Have low ferritin or take levothyroxine

Bottom line: test, don't guess

Your coffee is not a delivery system, and a stranger's morning routine is not a protocol.

The same two drops that do nothing for one person matter for someone on a blood thinner. The same cinnamon habit that is harmless in a pinch becomes a liver question by the spoonful. The same coffee that perks you up may be quietly draining your iron.

None of that shows up in a video. It shows up in your history, your medications and your results. That is where personalisation starts, and it is why I test before I recommend anything, including what goes in your cup.

If a post tells everyone to do the same thing, ask the four questions. If it cannot answer the fourth, scroll on.

Your next step

Free: If You Can't Check It, It Isn't Evidence: how to tell a claim you can check from one you can't.

Test / talk: Blood testing options, if you want to know your iron, liver or thyroid numbers before adding anything; or a free 20-minute call to work out whether testing is worth it for you.

I sell the testing above. The article's grades apply to it too.

References

  1. Taylor L, et al. CNS Drugs 2018. PMID 30374683. doi:10.1007/s40263-018-0578-5 (GW Research authors)
  2. Thai C, et al. Clin Pharmacol Drug Dev 2021. PMID 33951339. doi:10.1002/cpdd.950 (GW Research authors)
  3. Bansal S, et al. Clin Pharmacol Ther 2023;114(3):693–703. PMID 37313955. doi:10.1002/cpt.2973
  4. Watkins PB, et al. Clin Pharmacol Ther 2020;109(5):1224–1231. PMID 33022751. doi:10.1002/cpt.2071 (two Greenwich Biosciences authors)
  5. Camfield DA, et al. Nutr Rev 2014. PMID 24946991. doi:10.1111/nure.12120
  6. Mori K, et al. Phytother Res 2009. PMID 18844328. doi:10.1002/ptr.2634 (first author, Hokuto Corporation)
  7. Lee HW, et al. Front Pharmacol 2022. PMID 36110519. doi:10.3389/fphar.2022.934740
  8. Huerta Ojeda Á, et al. Antioxidants 2024. PMID 39334705. doi:10.3390/antiox13091046
  9. Kwon Y, et al. Medicine 2022. PMID 35451393. doi:10.1097/MD.0000000000028997
  10. Lee S, et al. J Korean Med Sci 2020. PMID 32419395. doi:10.3346/jkms.2020.35.e122
  11. Abraham K, et al. Mol Nutr Food Res 2010. PMID 21462332. doi:10.1002/mnfr.201000394
  12. Björnsson HK, et al. Liver Int 2020. PMID 31991029. doi:10.1111/liv.14393
  13. Philips CA, et al. Hepatol Commun 2023. PMID 37756041. doi:10.1097/HC9.0000000000000270
  14. Shoba G, et al. Planta Med 1998. PMID 9619120. doi:10.1055/s-2006-957450
  15. Halegoua-DeMarzio D, Navarro V. Liver Int 2024. PMID 39136211. doi:10.1111/liv.16071
  16. Morck TA, et al. Am J Clin Nutr 1983. PMID 6402915. doi:10.1093/ajcn/37.3.416
  17. Benvenga S, et al. Thyroid 2008. PMID 18341376. doi:10.1089/thy.2007.0222
  18. EFSA ANS Panel. Green tea catechins. EFSA J 2018. doi:10.2903/j.efsa.2018.5239
  19. Paul C, Leser S, Oesser S. Nutrients 2019;11(5):1079. PMID 31096622. doi:10.3390/nu11051079 (authors from GELITA AG and the Collagen Research Institute)
  20. BfR, FAQ on coumarin in cinnamon (read 29 Sep 2026). FSA, Food Fact Check: CBD safety and health claims (22 Jun 2026; read 29 Sep 2026). Regulation (EU) 2023/915.

Related reading