I have a coffee solution. I do not have a caffeine solution.
The coffee part is settled: one cup of Danger Coffee medium roast, brewed in a French press, at least three hours after I wake up, with bone broth ahead of it. The rest of my caffeine arrives from other directions. Tea, Huel Energy+, and sometimes a Bloom. None of it is a crisis. All of it is data. This post is me reading the research on caffeine dependence, running the numbers on my own sources, and asking an operator's question about a new kind of leash: what does energy cost in freedom per effort?
Disclosure: Danger Coffee runs an affiliate program, and links to Danger Coffee on this page may become affiliate links that earn me a commission at no extra cost to you. I will update this note when that changes. Prices, lab details and caffeine figures were checked on October 28, 2026 and can change. None of this is medical advice.
Am I addicted to caffeine?
For most people who use caffeine daily, the honest answer is dependent, yes, and addicted, it depends on what you mean. The two words get blurred, so here they are apart.
Dependence is physical. Use caffeine every day, stop, and the body complains. The DSM-5, the American Psychiatric Association's diagnostic manual, recognizes caffeine withdrawal as a diagnosis. Addiction, in the sense of losing control and using despite harm, has a research label called caffeine use disorder, and the DSM-5 lists it only as a condition for further study. In one survey of U.S. adults, 8 percent met the proposed criteria. An estimated 89 percent of U.S. adults consume caffeine every day. Most of us use it. Far fewer meet even the proposed criteria for a disorder.
The withdrawal timeline is well mapped. A 2004 review from Johns Hopkins and American University gathered 57 experimental studies and 9 surveys. Symptoms typically begin 12 to 24 hours after the last dose, peak between 20 and 51 hours, and last 2 to 9 days. They fall into five groups: headache, fatigue or drowsiness, low or irritable mood, trouble concentrating, and flu-like aches. About half of the people in the experiments got a headache.
Dependence says nothing about character. It is what a nervous system does when a daily dose goes missing.
What caffeine does to a tired brain
Tiredness has a chemical signal, and the signal is adenosine. It builds up while you are awake and binds to receptors that make you sleepy. Caffeine takes the signal's seat. It binds to the same receptors and blocks adenosine from landing, so the adenosine is still there and you stop hearing from it.
Tolerance has a mechanism worth knowing. A recent review describes chronic caffeine use as appearing to increase the number and activity of adenosine receptors, with much of the evidence from animals. Caffeine then has to compete with more adenosine each day, and its effect shrinks. If the dose stops, the larger receptor pool may amplify adenosine's effects, which fits the withdrawal picture above. The system adapts to the input, and the adaptation sets the price of staying level.
The clock on all this is slower than most people assume. A 2024 review in the Journal of the International Society of Sports Nutrition puts caffeine's half-life in healthy people at about five hours, with a range of 1.5 to 9.5 hours from person to person. Nearly all of it is absorbed within 45 minutes. Genetics, pregnancy, smoking, oral contraceptives and altitude can all move your number. For the average person, a cup at 10 a.m. is still half-present at 3 p.m.
Genetics deserves its own line. The liver enzyme CYP1A2 clears most caffeine, and people carry faster or slower versions. In a 2006 JAMA study of 2,014 people with a first nonfatal heart attack and 2,014 matched controls in Costa Rica, coffee was linked to higher heart attack risk only in slow metabolizers. Among carriers of the slow allele, four or more cups a day carried an odds ratio of 1.64. Rapid metabolizers showed no increase, at 0.99. It is observational data from one country, so read it as a reason to know your own response, not a verdict. Genotyping can show which version you carry.
The slow clock is also where the cost hides. In a 2013 study of 12 healthy sleepers, 400 milligrams of caffeine taken six hours before bedtime cut total sleep time by more than an hour. In their own reports, they hadn't noticed. I keep coming back to that one, because energy you can't feel the price of is hard to budget.
Does caffeine give you energy, or just your baseline back?
This is the finding that reframes the morning ritual. In 2010, a University of Bristol team led by Peter Rogers asked 379 people to go 16 hours without caffeine, then gave them either caffeine or a placebo. About half were non or low consumers, and half were medium or heavy consumers. Medium and heavy consumers who got the placebo reported lower alertness and more headache. Medium and heavy consumers who got caffeine ended up no more alert than the non-consumers who got nothing. Rogers' reading: what frequent users feel as a lift is mostly the end of withdrawal.
A 2013 follow-up from the same group landed in similar territory. Caffeine helped motor performance, but tolerance to its effect on sleepiness meant that frequent consumption failed to enhance mental alertness.
If you drink it every day, you may be paying a daily fee to feel like yourself.
That is one research group's conclusion, and it is about mental alertness, not about everything people like in caffeine. It does not say caffeine is useless. It says a standing fee for baseline is worth examining in daylight.
Is coffee itself the problem?
Probably not, and the evidence is worth holding next to the withdrawal data. A 2017 BMJ umbrella review pulled together 201 meta-analyses of observational studies and 17 of trials. For several outcomes, the largest benefit showed up at three to four cups a day versus none: roughly 17 percent lower all-cause mortality, 19 percent lower cardiovascular mortality and 15 percent lower cardiovascular disease, as summarized by MDedge. The authors were careful. These are observational findings and cannot prove causation. Pregnancy was the clear exception, with higher odds of low birth weight, preterm birth and pregnancy loss at high intakes, and the review also flagged higher fracture risk in women.
One more caution on that one. It studied coffee, not energy cans, so it says nothing about a 200-milligram drink with added vitamins. The research on dependence and the research on coffee can both be true. A daily habit can carry a withdrawal fee and still sit in a pattern that looks fine at the population level.
Why Danger Coffee is my current coffee solution
The coffee was the easy part to settle. I drink medium roast, only medium, and I drink it plain. I landed on Danger Coffee because its founder, Dave Asprey, got me with two things: mold-free and minerals. Asprey is often called the father of biohacking, and I like the instinct even where I doubt the claims: treat the body as a system you can measure, change one input, and watch the output. I will be straight about both claims, because the research is less flattering than the pitch.
I looked at the loud end of the shelf first. Death Wish blends arabica and robusta, is USDA Organic and Fair Trade certified, and its listing puts a 6-ounce cup near 180 milligrams of caffeine. Black Insomnia claims 702 milligrams in 12 ounces, which is 302 over the FDA's 400-milligram daily line in a single cup. Robusta is how they get there. A study cited by Medical News Today put arabica at roughly 34 to 39 grams of caffeine per kilogram of dry coffee and robusta at 69 to 82. On a question about dependence, a coffee built to maximize the dose is the wrong tool.
Then the tested-coffee field. Lifeboost sells single-origin arabica from one region in Central America, shade-grown and USDA Certified Organic at the farm level, and says it screens the finished coffee for more than 450 pesticides, mycotoxins and heavy metals. Purity Coffee describes organic, specialty-grade beans with third-party testing for mold, mycotoxins and contaminants. Both are real options. On the pages I read, neither posts certificates the way Danger does.
Asprey founded Bulletproof and stepped down as its CEO in 2019. He launched Danger Coffee in 2022, according to Daily Coffee News, which also reports that Bulletproof's June 2025 rebrand left mycotoxin messaging off its homepage and product pages. Danger still carries it.
Here is the pitch: much of the world's coffee carries mold toxins that drain energy and cause brain fog. The evidence is less on its side. The concern traces to studies of green, unroasted beans. ConsumerLab's review found that roasting cuts ochratoxin A by 30 to 90 percent and that retail coffee tests well under European limits, and a 2021 Polish study found every sample under the legal limit. The Daily Coffee News piece carries an editor's note that the scientific consensus appears to be that coffee does not contain mold at levels toxic to humans. Detection is common, though. A 2022 meta-analysis found ochratoxin A in about 58 percent of coffee products worldwide.
Finding mold toxins in coffee and being harmed by them are separate claims.
So why pay for tested coffee? At about $1.30 a brew, I will pay for a company that posts its paperwork. Here is what is posted. Danger's lab page lists certificates for 14 mycotoxins, including aflatoxins, ochratoxin A, fumonisins and zearalenone; more than 500 pesticide residues; glyphosate; and arsenic, cadmium, mercury and lead. The lab is described as independent and IAS-accredited but is not named. Every mycotoxin certificate on the page covers green beans before roasting. The company's protocol says testing happens before and after roasting, and the page shows no post-roast results.
The minerals get the same treatment. Each 310-gram bag is listed with 255 milligrams of fulvic acid and 1,035 milligrams of humic acid, part of what Danger calls 50-plus trace minerals. At a 15-gram brew, my math puts that near 12 milligrams of fulvic and 50 of humic per cup. The research on those compounds is early. A review of fulvic acid centers on animals and animal cells, and the RxList supplement monograph rates the evidence for its uses as insufficient. I found no large human trial showing a benefit at doses like these. I count the minerals as a tiebreaker, not a reason.
One more line from the lab page, because I would want to know it. Danger also posts a certificate for the mineral powders themselves, and the humic powder shows 4.37 ppm lead and 3.40 ppm arsenic. If that powder is what goes in the bag, one brew carries about 0.2 micrograms of lead by my math, roughly 2.5 percent of the FDA's 8.8-microgram daily reference level for women of childbearing age, set in 2022. Small. Not zero.
The three-hour rule
My mornings run on a rule I picked up during a stretch of diet research, mostly around homeostasis. The idea that stuck: wake up, don't eat, let your glycogen deplete. I can't prove it is right. It is my routine. In practice I start with bone broth, and caffeine waits at least three hours.
The wait is the part with the thinnest science behind it, so I looked. The 2024 sports nutrition review mentioned above took on the claim that waiting an hour and a half to two hours after waking prevents the afternoon crash, and the authors were skeptical. Their argument is that caffeine raises cortisol whether you delay or not, and that regular drinkers show a blunted response.
Behind that argument sits a 2005 double-blind crossover trial of 96 adults. Daily caffeine blunted the cortisol response to a morning dose, and at 300 or 600 milligrams a day it disappeared for the first dose of the day. After five days with no caffeine, the response came back strong. That is why total intake matters more than the one cup I am careful about. Tolerance is built by all of your caffeine. As for a three-hour wait specifically, I did not find a trial that tests it.
Food has some data too. A University of Bath trial of 29 healthy adults, average age 21, gave strong black coffee, about 300 milligrams of caffeine, 30 minutes before a sugary drink after a night of broken sleep. Their blood glucose response ran about 50 percent higher than after the same night without the coffee. It is a fair reason to put something in your stomach before caffeine. It is also a small study built on bad sleep and a sugar drink, and it says nothing about bone broth in particular.
So the three hours is a habit with a reasonable rationale and no proof behind it. If you drink coffee daily and feel fine, the research I read gives you no reason to change.
How I brew it, and the French press catch
Danger publishes its own French press recipe, which makes a good baseline. The ratio is 1 gram of coffee to 15 of water, so 15 grams for 225 milliliters. Grind medium. Water between 195 and 205 degrees Fahrenheit. Pour, stir, stir again at two minutes, start plunging slowly at 3:30, and decant at about four minutes. To me it is smooth with little bitterness, even with nothing in it, which is why I stay on the medium roast.
On cost, a 310-gram bag starts at $26.95. At 15 grams a brew that is about 20 cups, roughly $1.30 each, or about $1.17 on the 10 percent subscription discount.
There is a catch. With no paper filter, cafestol and kahweol, oils in coffee that raise LDL cholesterol, end up in the cup. A randomized trial of 46 people drinking 0.9 liters a day of French press or filtered coffee for 24 weeks found French press raised LDL, and a protein that helps move cholesterol around rose first. In a Dutch trial summarized by the Center for Science in the Public Interest, people drinking nearly four cups a day of unfiltered coffee had LDL 16 points higher than filtered drinkers after 11 weeks. The advice for people with elevated LDL was to avoid unfiltered coffee daily. Decaf does not get you out of it, since the oils are still in the bean. One occasional cup is nowhere near those doses. If your LDL is high, ask your doctor before the press becomes a daily habit. Switching to a paper filter cuts more than 90 percent of the cafestol, per NutritionFacts.