Insomnia tracked death risk only when sleep was measured

Insomnia is diagnosed by asking. Its definition rests on what a person reports about their nights, and so does almost all the work on whether insomnia shortens life.

Twenty years of it has never agreed with itself. Some cohorts find a raised risk of dying, others find none.

An analysis of the Wisconsin Sleep Cohort, published in the journal SLEEP, points at a reason. Once you measure how long people actually slept, the ones who said they slept badly and the ones who genuinely slept little turn out to be different groups, with different outcomes.

What a sleep recording captures that a questionnaire misses

Polysomnography, PSG for short, is a sleep study. It records certain body functions as you sleep, or try to sleep: brain waves, eye movement, heart rate, breathing effort and rate, and the level of oxygen in your blood. It can be done either at a sleep center or in your home.

The output is a total sleep time that owes nothing to anyone’s recollection of the night.

Recall is the weak point here, and it is weakest in exactly the people being studied. Individuals with insomnia tend to overestimate time to fall asleep or time of being awake after sleep onset, and to underestimate total sleep time in comparison to PSG measures.

So a person certain they managed three hours may have managed six and a half. The reverse happens too.

The insomnia group that carried the risk

The team followed 373 participants with insomnia symptoms and 394 participants without insomnia symptoms, with a median follow-up duration of 8 years.

Insomnia symptoms meant difficulty in falling sleep, maintaining sleep or early awakening more than five times a month. Objective short sleep duration was defined as total sleep time under 6 hours based on polysomnography. Self-reported short sleep duration was defined as under 7 hours.

That splits people four ways, and only one of the four stood out. Insomnia symptoms with objective short sleep duration were associated with a 2.6-fold higher risk of all-cause mortality against people with neither problem.

Everything else was flat. Participants with insomnia symptoms and normal sleep duration, or with objective short sleep duration but without insomnia symptoms, were not associated with increased risk of all-cause mortality. Short sleep on its own did nothing. Insomnia on its own did nothing.

Then the result that reframes two decades of work: insomnia symptoms with self-reported short sleep duration were not associated with risk of all-cause mortality. The group most studies would have flagged as high risk showed no rise at all.

Why measured short sleep may mark a more severe insomnia

The idea that measured sleep length identifies a distinct and worse form of insomnia is not new, and this study was designed to test it.

The proposal is that insomnia with objective short sleep duration is the most biologically severe phenotype of the disorder, a version driven by physical arousal rather than by worry about sleeping. It has been associated with cognitive-emotional and cortical arousal, activation of both limbs of the stress system, and a higher risk for hypertension, impaired heart rate variability, diabetes, neurocognitive impairment. Cortical arousal means the outer brain staying switched on when it should be winding down.

A dose relationship is what makes that credible. The activity of the stress system is directly proportional to the degree of objective sleep disturbance: the more of it a recording picks up, the larger the stress response.

What 767 adults cannot settle about insomnia

This is an observational cohort, so it can show that a pattern travels with dying sooner. It cannot show that short sleep did the killing.

The size of the risk is also loosely pinned. The range the data allow runs from barely raised to almost six times, which is a wide window around a number reported as 2.6-fold. Direction is clearer than magnitude here.

Sleep was measured once, on one night, in a laboratory. People sleep differently in laboratories, and a single night is a thin sample of a chronic condition. The paper is paywalled, so the authors’ own limitations section was not available for this piece.

What to do about insomnia you cannot measure at home

Almost nobody reading this will have had an overnight recording, and the study offers no way to work out which side of six hours you fall on. Self-report is precisely what it found wanting.

The authors state their conclusion narrowly: objective short sleep duration appears to be a valid marker of the biological severity of insomnia. That is a claim about a research tool. It says the field has been mixing two conditions and calling them one, which would explain twenty years of contradictory results.

For a person, the useful part is smaller and duller. Persistent insomnia is worth taking to a doctor rather than waiting out, and this is one more reason to treat it rather than tolerate it. Cognitive behavioral therapy for insomnia remains the first-line treatment, and it does not require knowing your phenotype.

The measurement problem was never a technicality. It was hiding the group that mattered.