Sleep in Older Adults: Aging, Insomnia, CBT-I, Apnea and Medication Risk

Bottom line: Sleep changes with age, but “you are older, so bad sleep is normal” is a poor evidence-based conclusion. Older adults often have lighter, more fragmented sleep and changes in circadian timing, yet clinically important insomnia, sleep apnea, restless legs and medication-related sleep disruption remain identifiable problems with their own treatment pathways. CBT-I still works, and medication safety deserves more—not less—attention with age.[1-9]

What actually changes with normal aging?

Normal aging changes sleep physiology.

Recent reviews describe a pattern that commonly includes:

• less slow-wave sleep;
• somewhat less REM sleep in some populations;
• more nighttime awakenings;
• lower sleep efficiency;
• greater sleep fragmentation;
• weaker circadian amplitude; and
• a tendency toward earlier sleep and wake timing in many older adults.[1,2]

Those are population-level tendencies, not a rule that every older adult should expect severe insomnia.

The key distinction is between age-related change and a treatable sleep disorder.

Someone who sleeps a little lighter than at age 30 is answering a different question from someone who spends hours awake every night, repeatedly cannot return to sleep, snores with witnessed pauses, or develops dangerous daytime sleepiness.

Does aging mean people simply need less sleep?

Not necessarily.

Age-related changes in sleep architecture and sleep timing are sometimes interpreted as proof that older adults biologically need very little sleep. That conclusion goes beyond what the physiology establishes.

A person may spend less time in deep sleep, wake more often, or shift sleep earlier without those changes proving that severe sleep restriction is harmless.

This distinction matters because persistent daytime sleepiness, cognitive difficulty, falls, mood symptoms or inability to function should not be dismissed with “older people just sleep less.”

Insomnia is common—but it is still insomnia

Insomnia becomes more common with age, partly because aging brings more opportunities for sleep disruption:

• chronic pain;
• nocturia;
• cardiopulmonary disease;
• mood disorders;
• caregiving stress;
• retirement-related schedule changes;
• medication effects;
• sleep apnea;
• restless legs; and
• reduced daytime activity or light exposure.[1,2,6,8]

The result is often a layered problem rather than one isolated “sleep chemical” deficiency.

That is why adding a sedating product without checking the rest of the system can produce a poor tradeoff.

CBT-I still works in older adults

Cognitive behavioral therapy for insomnia is not only a young-adult intervention.

A systematic review and meta-analysis of 14 studies in older adults found significant improvements in multiple sleep-log outcomes.[4]

Pooled estimates included approximately:

• 8.36 percentage points higher sleep efficiency;
• 9.29 minutes shorter sleep-onset latency; and
• 23.44 minutes less wake after sleep onset.[4]

These are clinically useful directions of change.

A 2025 systematic review of nine recent clinical trials in community-dwelling adults over 60 likewise concluded that cognitive interventions generally improved sleep quality and insomnia symptoms and were feasible and safe.[3]

A newer network meta-analysis of 34 randomized trials and 3,078 older participants also ranked CBT-I among the strongest non-drug interventions for sleep quality.[5]

The exact effect sizes vary substantially by intervention, outcome and study design, so these findings should not be turned into a guarantee that every older adult will respond identically.

A weird but important CBT-I result: total sleep time may not rise

The older-adult CBT-I meta-analysis found that total sleep time did not increase in the same way as sleep efficiency; the pooled sleep-log estimate actually favored slightly less total sleep time.[4]

At first glance, that sounds like failure.

It is not necessarily.

CBT-I often uses sleep restriction or sleep compression to reduce long periods of wakefulness in bed and strengthen sleep drive. Early treatment can therefore produce:

• less time in bed;
• more consolidated sleep;
• higher sleep efficiency; and
• less wakefulness,

without immediately producing more total minutes asleep.

That is a perfect example of why “did total sleep time go up?” is not the only insomnia endpoint that matters.

Subjective and objective sleep can still disagree

A 2023 systematic review of behavioral components of CBT-I in older adults found stronger improvements in subjective sleep measures than in actigraphy or polysomnography outcomes.[9]

That does not make the subjective improvement fake.

Insomnia is partly defined by the lived experience of difficulty sleeping and daytime impairment. But it does mean the site should preserve the distinction between:

• feeling that sleep is better;
• spending less time awake in bed; and
• measurable changes in sleep architecture.

See Subjective vs Objective Sleep for the broader measurement problem.

Sleep apnea becomes too important to ignore

Obstructive sleep apnea can easily masquerade as “bad sleep,” frequent waking, or unexplained daytime fatigue.

A systematic review and meta-analysis of 39 studies involving 33,353 older adults estimated a pooled OSA prevalence of 35.9%.[7]

That number needs a giant caution label: heterogeneity was extremely high (I² about 98.8%), meaning prevalence varied substantially across populations and study methods.[7]

So 35.9% should not be treated as a universal prevalence for every older community.

The defensible conclusion is simpler: OSA is common enough in older adults that persistent sleep complaints should not automatically be treated as a sedation problem.

Snoring, witnessed breathing pauses, choking or gasping, morning headaches, resistant hypertension, and unexplained daytime sleepiness change the decision tree.

See Sleep Apnea vs Insomnia for that distinction.

Restless legs can also look like insomnia

Older adults can also have restless legs syndrome, periodic limb movements, neuropathic symptoms, or iron-related RLS risk.

Someone who repeatedly says “I cannot settle down at night” may not be describing generalized insomnia at all.

That matters because sedating supplements do not correct the underlying RLS pathway, and blind iron supplementation is not the evidence-based way to evaluate it.

See Restless Legs, Iron and Sleep for the dedicated evidence framework.

Medication risk changes with age

Medication is sometimes attractive because it can be simpler and faster than behavioral treatment.

But the risk-benefit equation changes with age.

A 2025 review of insomnia treatment in older adults emphasizes CBT-I as the gold-standard treatment while noting that adverse effects from some hypnotics become more important in older patients.[6]

A 2025 review of common management pitfalls highlights another problem: polypharmacy. Older adults are more likely to use multiple medications, increasing opportunities for:

• drug-drug interactions;
• additive sedation;
• dizziness;
• falls;
• confusion;
• cognitive adverse effects; and
• sleep disruption caused by another medication.[8]

This is why an OTC label does not automatically mean “low risk for nightly use.”

The site's OTC Antihistamines for Sleep review is especially relevant because first-generation antihistamines add anticholinergic burden as well as sedation.

“PM” products deserve extra scrutiny

Combination nighttime products can be particularly awkward in older adults.

A person seeking sleep may unintentionally take:

• diphenhydramine or doxylamine;
• acetaminophen or an NSAID they do not actually need; or
• multiple sedating ingredients across different products.

The important question is not whether a package says “nighttime.” It is which active ingredients are present, why each is being used, and what else the person is taking.

That is a medication-reconciliation problem, not a sleep-hack problem.

Naps are not automatically bad

Napping becomes more common with age, but the meaning depends on context.

A short planned nap after a poor night is different from spending large portions of the day asleep because nighttime sleep is fragmented, depression is present, medications are sedating, or sleep apnea remains untreated.

Long or late naps can also reduce nighttime sleep pressure in someone with sleep-onset insomnia.

So the useful question is not “are naps good or bad?” It is:

What problem is the nap solving, and what is it doing to the next sleep period?

See Naps and Nighttime Sleep for that framework.

Sleep duration and frailty: important, but mostly observational

A 2025 systematic review reported associations between poor sleep quality, atypical sleep duration and frailty in older adults.[10]

Those findings are clinically important but should not be flattened into a causal rule.

Frailty can worsen sleep through pain, inactivity, disease burden and medication use. Poor sleep may also contribute to inflammation, reduced activity, cognitive problems and vulnerability.

The relationship is plausibly bidirectional.

Observational associations therefore justify taking sleep seriously, but they do not prove that forcing every older adult into one exact sleep-duration target prevents frailty.

What the evidence does not support

Current evidence does not justify claims that:

• severe insomnia is an unavoidable part of aging;
• older adults simply do not need meaningful sleep;
• every nighttime awakening is pathological;
• every sleep complaint should be treated with medication;
• a sedating OTC antihistamine is harmless because it is nonprescription;
• CBT-I is too late to matter after age 65;
• an OSA prevalence estimate from pooled studies applies identically to every older population; or
• one supplement can fix sleep disruption driven by pain, apnea, nocturia, RLS, depression or polypharmacy.

A more useful decision tree

For persistent sleep problems in an older adult, ask:

1. Is there enough sleep opportunity?
2. Is the problem falling asleep, staying asleep, waking too early, or feeling unrefreshed?
3. Is breathing abnormal during sleep?
4. Are pain, nocturia or restless legs repeatedly waking the person?
5. Could a medication or combination of medications be contributing?
6. Is the schedule or circadian timing drifting earlier?
7. Is chronic insomnia present even after those contributors are addressed?

Only then does “what sleep aid should we add?” become the right question.

Bottom line

Older-adult sleep deserves its own evidence framework.

Normal aging can make sleep lighter, more fragmented and differently timed, but that should not become an excuse to ignore persistent insomnia or another sleep disorder. CBT-I remains well supported. OSA and restless legs are common enough to stay high on the differential. Medication and OTC-sedative risks become more important as fall risk, cognitive vulnerability and polypharmacy increase.[1-9]

The most defensible principle is simple:

Do not treat age itself as the diagnosis. Identify the actual sleep bottleneck.

Related reading

• CBT-I vs Sleep Supplements
• Sleep Apnea vs Insomnia
• Restless Legs, Iron and Sleep
• OTC Antihistamines for Sleep
• Naps and Nighttime Sleep
• Sleep Onset vs Sleep Maintenance