Sleep and Migraines: Why 'Sleep More' Is Useless Advice

The relationship between sleep and migraines runs in both directions: poor sleep makes attacks more likely, attacks disrupt sleep, and the risk sits at both ends of the night, because too little sleep and too much are both commonly reported triggers. That last part is why “sleep more” is useless advice. It assumes the problem only lives at one end.

The useful question isn’t whether sleep matters in general. It’s whether it matters for you, and by how much. Answering that takes a specific comparison, and almost nobody records the half of it that happens on ordinary days.

Reported at both ends of the night

When 1,207 patients at a headache clinic were asked what set off their attacks, half named sleep disturbance: 49.8%, the fifth most common trigger after stress, hormones, skipped meals, and weather. That’s the short end. The deadline night, the newborn, the 6:10 flight.

The long end is stranger. The Migraine Trust notes that some people wake with an attack after more sleep than usual — the Saturday lie-in following a week of 7am alarms is its textbook case — and its explanation is that the migraine brain reacts badly to a change in routine, in either direction.

So the honest shape of the evidence is not “more sleep, fewer attacks.” It’s a U: risk at both ends, your usual amount in the middle. The advice that actually follows from that is “keep your sleep boring,” which is harder to sell than eight hours.

The weekend attack has a second suspect

Before you write “oversleeping” in your diary as a confirmed trigger, notice what else changes on a morning you sleep until ten: your usual coffee arrives three hours late.

A 1992 study of weekend migraine attacks found the risk concentrated in people who did both things at once, drinking a lot of caffeine daily and lying in late at weekends. In that group, 69% had weekend attacks. Among people who did only one of the two, 4%.

A diary can’t fully untangle this, but it can stop you attributing with false confidence. If your weekend attacks vanish on the weekends you keep your weekday wake time, you still don’t know whether the fix was the sleep or the on-schedule coffee. You know something more useful: what to keep doing.

The Saturday attack has more suspects than these two, and they stack — why migraines arrive at the weekend works through the rest of them.

Attacks wreck sleep right back

The American Migraine Foundation describes the connection as a two-way street: poor sleep raises the likelihood and severity of attacks, and attacks make it harder to fall asleep and stay asleep.

That bidirectionality has a practical consequence for anyone keeping a diary. A single bad night followed by a single attack proves nothing, because the attack may already have been under way while you tossed; the warning phase of a migraine can start hours before the pain does. One night is never the unit of analysis. The average over months is.

The disruption also spreads sideways. An attack that runs past midnight costs the other person in the bed a night’s sleep too, which is one reason the room-by-room changes that make a home easier to have migraines in start with the bedroom.

Log hours, roughly, and log the quiet nights

Record one number: hours slept, to the nearest half. Not sleep stages, not a wearable’s readiness score, not minutes of REM. Precision theatre is what kills diaries; a field you can answer while the kettle boils is a field still being answered in month three. Sleep is already one of the six fields worth putting in a minimal trigger diary, and it’s the easiest of the six to record honestly.

The trap is when people record it. Nearly everyone logs sleep after the terrible night, the one that preceded the attack, and skips the ordinary Tuesdays. That produces a diary saying “I slept badly before some attacks,” which is true of every human alive and tells a clinician nothing. The quiet nights are the denominator. Without them there is no comparison, and the comparison is the entire analysis.

On paper, the method is one number every morning, same page, attack or no attack. If you log in an app instead: Migraine Journal asks for the previous night’s sleep in four rough bands (under 4 hours, 4–6, 6–8, 8+) on every entry, because a band you’ll actually answer beats a decimal you’ll abandon by Thursday.

The sentence a neurologist can use

Three months of that produces something like: “Attacks averaged 5.9 hours of sleep the night before, against 7.1 on quiet nights.” Fourteen words a clinician can act on inside a ten-minute appointment, which is more than “my sleep is terrible” has ever achieved.

The null result is worth just as much. If the two averages come out at 6.8 and 6.9, sleep drops off your suspect list, and you can stop treating every late night as a loaded gun. Ruling a trigger out is the diary outcome nobody celebrates and everyone should.

One boundary worth drawing: if your bad sleep has a sound to it, snoring, gasping, breathing that a partner says stops and starts, take that to your GP rather than your diary. A sleep log can show a pattern; it cannot rule out a sleep disorder, and it shouldn’t try.

The rest of the method, including what to record on attack days and what on the quiet ones, is on the tracking page.