Insomnia

Insomnia is a complaint with three parts: difficulty getting to sleep, difficulty staying asleep, and a cost the following day. All three are load-bearing. A short night that costs nothing the next day is not insomnia, and neither is a bad night in a week that never offered a chance to sleep.

That is the definition in both of the manuals clinicians use, and the surprise inside it is what the word does not mean. It is not a number of hours. Nobody is diagnosed by sleeping less than some figure, because the figure that leaves one person wrecked leaves another fine.

The manuals list them separately because they behave separately, and because what a person tells a clinician is usually one of them rather than all three.

Onset. Sleep will not start. The convention in the research is more than about half an hour of lying there, which is not an arbitrary line: when Lichstein's group went looking for a quantitative cut in two decades of trial data, half an hour was roughly where ordinary bad nights stopped and clinical patterns began.

Maintenance. Sleep starts and does not hold. What matters here is the total time spent awake rather than the number of awakenings, because brief arousals are a normal feature of every night and nobody counts them in a bed they are sleeping well in.

Early waking. Sleep ends before it was supposed to, with no route back into it. This one is most often misread, because the same shape is produced by a body clock that has simply drifted earlier with age, and an earlier clock with a satisfying night in it is not a disorder.

Most people have a mix, and mixtures are the commonest presentation rather than the complicated case. The profile is still worth knowing, because an onset problem with an intact next day points somewhere different from all three together. If sleep arrives readily when nothing is scheduled in the morning, the question is timing rather than insomnia, and that fork is worked through in insomnia or a late chronotype.

This is the clause people drop, and dropping it is how the word gets used wrongly in both directions.

A complaint about the night with no consequence in the day does not meet the definition. Natural short sleepers exist. They sleep six hours or fewer, wake unprompted, feel well, and have nothing to treat. Meanwhile somebody who logs seven and a half hours in broken pieces, wakes unrestored, and is dropping threads by mid-afternoon meets it comfortably. The second person is the one with a disorder, and the first person is the one the internet keeps alarming.

The daytime half is also the part that explains why insomnia is described as a round-the-clock condition rather than a nighttime one. Fatigue, irritability, a mind that will not hold a thread, and a particular dread about the coming night. Notice what is missing from that list: being sleepy. People with chronic insomnia are often measurably harder to put to sleep in the daytime than good sleepers are, which is the opposite of what exhaustion predicts, and it is the single best clue that the problem is arousal rather than a shortage. The entry on tired but wired is about that state specifically.

Both produce a tired person, and almost everything else about them differs.

The definition contains a condition that is easy to skim past: the trouble has to happen despite an adequate opportunity for sleep. That clause is not a technicality. It is what separates insomnia from sleep deprivation, and it excludes a large population of exhausted people who do not have it: the parent of a newborn, the night nurse, the student who allots five hours, the person on a commute that makes seven impossible. Nothing is wrong with their sleep. They are not being given enough of it, which is a different problem with a different fix, described in the entry on sleep debt.

Then the counterintuitive part. Insomnia often arrives with more time in bed than average rather than less. Somebody sleeping badly reasons that the way to get more sleep is to offer more opportunity, goes to bed at ten, lies there until midnight, and stays until eight to make up what was lost. The opportunity is now nine or ten hours wide and the sleep inside it has not increased, so the extra hours are spent awake in a bed, which is the specific condition under which a bed becomes a cue for being awake.

That is why the first-line treatment narrows time in bed rather than widening it, a move nobody arrives at unaided. The entry on CBT for insomnia sets out the reasoning.

The thresholds look bureaucratic and each one is doing a job.

Three nights a week. Two ruined nights in a month is a life. A pattern that shows up most weeks is a pattern.

Three months. Short runs of bad sleep after a bereavement, a diagnosis, a move or a deadline are extremely common and mostly resolve on their own; the manuals keep a separate short-term category for exactly that, so the three-month line marks the point where a reaction has become a condition rather than the point where someone is allowed to mind.

What makes the three months more than an administrative wait is the best idea in this literature, and it is thirty years old. Spielman's three-factor model splits the causes into predisposing traits, a precipitating event, and perpetuating behaviours, and its argument is that the three take turns. Something starts a bad run. What the bad run then provokes, entirely reasonably, is the set of responses that keep it going after the original cause has gone: more time in bed, naps, an earlier bedtime, cancelled mornings, and attention trained on sleep all day. By three months the trigger is frequently no longer present and is no longer the point, which is why "but the stressful thing is over" is such a common and such a confusing sentence in a consultation.

It is also the reason insomnia is treated by changing current behaviour rather than by excavating what started it.

Most chronic insomnia sits alongside another condition, and leaving the other thing unexamined is the usual reason a year of effort produces nothing.

The ordinary company it keeps: anxiety and depression, where the relationship runs both ways, since insomnia also predicts the later onset of depression in people who are not depressed yet; pain of any kind, which delays onset and fragments the middle; obstructive sleep apnoea, which coexists with insomnia far more often than either gets diagnosed next to the other; restless legs, with its signature of an urge to move that moving relieves; menopause and the hormonal shift around it; thyroid and other endocrine problems; medication, including steroids, stimulants, some antidepressants, beta blockers and ordinary decongestants; alcohol, which sedates the first half of the night and breaks up the second; and caffeine at a dose or hour the person has stopped noticing.

A note on the word secondary, which used to be applied to all of this and no longer is. The field moved to calling it comorbid insomnia because treating the other condition routinely fails to clear the sleep problem. Depression lifts and the insomnia stays. The pain is managed and the two in the morning waking continues. By then the perpetuating machinery from the section above has its own momentum, and it needs treating in its own right.

Which is why an assessment worth having asks about more than the night. Snoring with witnessed pauses in breathing, or sleepiness heavy enough to be dangerous at the wheel, points at apnoea and should be looked at before anything is done about the schedule.

• Separate the three components before anything else. Which part of the night fails, and whether the day pays for it. That is most of the information.
• Answer the free-morning question honestly. With no alarm and nothing to get up for, does sleep come and does it hold. Yes points at timing. No points at insomnia.
• Count how wide the opportunity is. Ten hours in bed and six hours of sleep is a different situation from six in bed and six of sleep, and only the first is being made worse by the obvious remedy.
• Write it down for a fortnight. Bedtime, rise time, roughly how long sleep took, time awake in the night, naps, alcohol, one line on the day. Every diagnosis in this area is made from that document, and it is the one thing worth doing before any appointment exists.
• Do not start with the hygiene list. It is the weakest component of the real treatment and is frequently mistaken for all of it. The entry on sleep hygiene explains what it is actually good for.

MoodyWonder's Insomnia Check scores onset, maintenance and daytime cost separately rather than adding them into one number, because the shape is what carries the meaning. It is a screening-style read and not a diagnosis.

The scope note, which matters more here than on most pages. A bad fortnight is not a disorder. Three or more difficult nights a week for three months or more, with the days visibly paying for it, is the recognised shape of chronic insomnia, and it has a first-line treatment that is not a tablet and works better over the long run than one does. Insomnia also carries real medical weight: it is an independent risk factor for depression, and persistent sleeplessness in someone who has stopped wanting to be here is a reason to get help now rather than to wait and see. The sentence that shortens that appointment by ten minutes is that you have already done the hygiene list and want to be referred for the treatment by name.