What changes about sleep as you age is real — but most of what keeps you up is treatable

Older people often worry that their sleep has fallen apart. Some of what they notice is a normal, measurable feature of ageing; some of it is insomnia, which is not. Telling the two apart matters, because the normal changes need no treatment while insomnia responds well to one — and the first-line treatment is not a sleeping pill [s1][s2]. The useful distinction is between sleep that has simply changed and sleep that is a disorder.

This is general information, not medical advice. Persistent poor sleep, loud snoring or daytime sleepiness in an older adult can signal a treatable condition and is worth raising with a clinician.

What genuinely changes

The clearest map of normal sleep across life comes from a meta-analysis of 65 studies covering 3,577 healthy people aged 5 to 102, all measured objectively [s1]. In adults, several parameters declined steadily with age: total sleep time, sleep efficiency (the share of time in bed actually spent asleep), the percentage of deep slow-wave sleep, the percentage of REM sleep, and REM latency [s1]. Others rose: the time taken to fall asleep, the proportion of light stage-1 and stage-2 sleep, and time spent awake after first falling asleep [s1]. In plain terms, sleep becomes lighter, shorter and more fragmented, with less deep sleep — which is why older people wake more easily and more often.

One detail reframes the worry. Most of these changes occur through mid-life and then largely plateau; of all the parameters, only sleep efficiency continued to decline significantly after age 60 [s1]. So the sense that sleep collapses in old age is partly misremembered — much of the change happened decades earlier, and the later years add comparatively little beyond a continued slippage in efficiency. The review also cautioned that effect sizes were strongly shaped by how well studies screened out illness and sleep disorders: in poorly screened samples, age associations were diminished or even masked, meaning some "age effects" are really the effects of undiagnosed conditions such as sleep apnoea [s1]. That is exactly why waking, un-refreshing sleep should not be dismissed as age alone — the very act of blaming age can bury a treatable disorder.

What is not normal — and what to do about it

Chronic insomnia — persistent difficulty falling or staying asleep with daytime consequences — is a disorder, not a stage of life, and it is common in older adults. The American College of Physicians guideline is unambiguous about the first move: it recommends that all adults receive cognitive behavioural therapy for insomnia (CBT-I) as the initial treatment (a strong recommendation) [s2]. Only if CBT-I alone fails does the guideline turn to a shared-decision discussion about short-term medication, and even then as a weak recommendation on low-quality evidence [s2].

That ordering is deliberate, and it matters most in older people, in whom sedative-hypnotic drugs carry heightened risks of falls, confusion and next-day impairment. CBT-I is a structured course addressing sleep scheduling, time restricted in bed, stimulus control and the thoughts and habits that maintain insomnia, and the guideline places it first for every adult with chronic insomnia, turning to medication only where it has been tried and failed [s2]. The strength-of-evidence comparison in CBT-I versus hypnotics favours the therapy, which is also why digital CBT-I is being tested specifically in older adults to widen access to a treatment in chronically short supply.

How to read this

Two things are true. Lighter, more broken sleep with less deep sleep is a real part of ageing and, on its own, not a problem to be medicated. And insomnia is a distinct, treatable disorder whose best first-line treatment is behavioural, not pharmacological. Conflating the two leads to both errors at once — worrying about normal changes and reaching for pills for a disorder that has a better answer. How much sleep is actually needed is covered in the adult sleep-duration consensus, and what night-waking looks like across the population in the surveys on waking in the night.

What to watch

Access is the bottleneck: CBT-I is recommended for everyone with chronic insomnia but is in short supply, which is why trials of medication such as newer agents in older adults continue. The evidence on what to try first, though, is settled.