I take a strong coffee to bed with me. Not a milky one, not a decaf, an actual coffee, drunk sitting up in bed at about eleven o'clock at night. Then I put the cup down and I go to sleep.
People find this hilarious. Or alarming. I have been told for years that I am doing it wrong, and I understand why, because everything anybody has ever been told about caffeine says the opposite. But if I skip it, I do not drift off half an hour later than usual. I lie there. Sometimes for hours, with my brain working through a list of everything I have ever said wrong since about 1994.
I have ADHD. And it turns out I am very far from the only person with ADHD who has quietly worked out that a coffee at bedtime helps them sleep, and then quietly stopped mentioning it, because of the look people give you.
So. Is it real, or are we all imagining it?
What caffeine is actually doing
All day long a chemical called adenosine builds up in your brain. It is the thing that makes you feel steadily more tired the longer you have been awake. Caffeine works by sitting in adenosine's parking space, so the tiredness signal cannot get through.
That much most people know. The interesting part is what happens next. When caffeine blocks those receptors, it indirectly nudges up dopamine, particularly in the parts of the brain involved in attention and motivation.
Dopamine is exactly what an ADHD brain handles differently. It is the reason the medications that treat ADHD are stimulants: they work by raising dopamine and noradrenaline.
So caffeine is not doing something random and unrelated over in a different part of the building. It is a weak, short-acting, far less precise version of what ADHD medication does deliberately.
The bit that sounds backwards, and is not
Anyone who has been prescribed a stimulant for ADHD, or watched a child be prescribed one, has met the thing that makes no sense to everybody else. A stimulant can calm someone with ADHD down.
This was first written up in 1937 and it still surprises people. The explanation is fairly ordinary once you hear it.
The reason I cannot fall asleep is not that I am not tired. It is that I cannot get my brain to stop.
That distinction is the whole thing. For a lot of people with ADHD the obstacle at bedtime is not a shortage of tiredness. It is racing thought, restlessness, and a strange inability to stop doing whatever you are doing and go to bed, even when you very much want to. Take the edge off that and you settle. Not because you have been sedated, but because the thing that was in the way has moved.
There is a second piece to this. Research using sleep diaries, movement monitors and melatonin testing has found that a large proportion of adults with ADHD, by some measures around three quarters, have a body clock that genuinely runs late. The biological night starts later. So somebody lying awake at midnight may not be failing to sleep at all. They may be exactly on time, for them.
What about the coffee nap?
This one comes up constantly, so it is worth explaining. Caffeine takes roughly twenty minutes to get going. So the idea is that you drink a coffee, lie down immediately, and wake up twenty minutes later just as it starts working, with the benefit of both.
Plenty of people with ADHD swear by it and the timing does work the way it is described. Two honest caveats. It is not an ADHD-specific finding, it works like that for most people. And it is not a substitute for sleeping at night. If you are needing one every single afternoon, the useful question is what is happening to your nights, not how to get better at napping.
Why it does not work for everybody
Here I have to be careful, because the honest answer is that people differ enormously, and a good deal of that difference is written into your genes.
Two separate things vary. How quickly you clear caffeine out of your system, which is largely down to one liver enzyme. And how sensitive your receptors are to it in the first place, which is a different gene entirely. You can be a fast clearer with sensitive receptors, or a slow clearer who barely notices it, and every combination in between.
Which means two people with ADHD can drink the same coffee at the same time and have completely different nights. If you have tried this and it left you staring at the ceiling until four in the morning, you were not doing it wrong. You are simply not one of the people it works for.
The part I am not going to skip
I could stop there and it would make a much better story. But you should know what the evidence does and does not say, because a great deal of what circulates about ADHD online is one person's experience presented as a finding.
Nobody has run the study. There is no trial anywhere examining caffeine timing and sleep in adults with ADHD. That is not a negative result. It is an absence.
The trials that do exist are not encouraging. The controlled trials of caffeine for ADHD symptoms were done in children, they are small, and taken together they found no meaningful benefit over placebo. A survey of more than two thousand adults found no relationship between ADHD symptoms and how much caffeine people drink, which argues against the popular idea that we are all unconsciously self-medicating.
And the sleep evidence points the other way. One well-known study gave people a large dose of caffeine at bedtime, three hours before bed, and six hours before bed, and measured their sleep properly rather than asking them about it. All three disrupted it, including the dose six hours before.
There is a genuinely awkward possibility for those of us who do this, as well. If you take caffeine every day your brain adapts to it, and what feels like "it does not affect me" may partly be that adaptation. Your sleep can be measurably disturbed while you sleep through it quite happily and report no problem whatsoever.
So what I have described is a mechanism that makes sense, and a lot of people independently reporting the same thing. That is not nothing. It is also not proof, and I am not going to dress it up as proof.
What I would actually say if you asked me
Do not start drinking coffee at eleven at night because a woman on the internet says it works for her.
But if you are somebody who has always done this, and been told for years that you must be imagining it, you are probably not imagining it. There is a plausible reason it might work for you specifically, your response to caffeine is partly genetic, and your own long experience of your own body is real evidence about you, even where the general advice says otherwise.
If you want to know properly, do it the boring way. Two weeks with, two weeks without. Write down what time you actually fell asleep and how you felt the next morning. That will tell you more about you than any study can, because studies describe averages and you are not an average.
And if you take medication, have a heart condition, are pregnant, or are already being treated for insomnia, ask a doctor rather than me. That is not me covering myself. Caffeine genuinely interacts with things.
Why I care about this beyond my own bedtime
Because of what happens to people who can no longer explain themselves.
Most care homes take caffeine off the drinks trolley somewhere in the middle of the afternoon. The reasoning is sensible enough. Caffeine is a stimulant, older people often sleep badly, and a disturbed night is hard on everybody, particularly a small night staff team. As a general rule it is perfectly reasonable.
But somewhere in this country there is a woman in her eighties who has had a strong coffee before bed every night since she was twenty five. She was never assessed for ADHD, because nobody assessed women of her generation. She is now offered a milky drink at seven o'clock, and she is recorded as unsettled at night.
Nobody has done anything wrong. A rule that is right for most people is being applied to somebody it was never right for, and she has lost the words to say so.
That is my whole job in one example. Not that the rules are bad, but that a lifelong history is information, and a person who can no longer tell you theirs deserves to have somebody ask on their behalf.
If you want the fully referenced version of all this, with the studies named and the counter-evidence set out properly, I have written it up as a research paper on caffeine and the ADHD brain.
And if you are reading this at midnight with a coffee in your hand: same.