Sleep in Older Adults: A Clinical Guide for Aging, Menopause, Pregnancy, Shift Work, and Jet Lag

Poor sleep after midlife is rarely one problem. Aging changes sleep architecture on its own, but the biggest disruptions in practice come from five distinct, overlapping mechanisms: normal circadian and slow-wave sleep decline with age, vasomotor symptoms of menopause, the mechanical and hormonal load of pregnancy, misalignment between work schedule and biological clock, and rapid time-zone travel. Cognitive behavioral therapy for insomnia (CBT-I) is the first-line treatment recommended by the American College of Physicians for chronic insomnia in adults generally, but the pharmacologic and hormonal add-ons that actually help differ by which of these five mechanisms is driving the problem. Treating a postmenopausal hot-flash-driven insomnia the same way as jet lag, or treating an older adult's sleep complaint without first screening for obstructive sleep apnea (OSA), is a common and correctable failure mode.

This guide separates general population-level evidence from claims specific to one subgroup, and flags where a precise number in circulation needs verification against the primary literature before it should guide a clinical decision.

What is established, what is plausible, and what is not established

Established: Slow-wave sleep and sleep continuity decline with normal aging. CBT-I is an effective, guideline-recommended first-line treatment for chronic insomnia across adult age groups, with effects that tend to persist after treatment ends better than medication effects do. OSA prevalence rises with age and also rises after menopause. Benzodiazepines and Z-drugs (zolpidem, eszopiclone, zaleplon) carry recognized risks of falls, cognitive impairment, and next-day impairment in older adults, which is why geriatric prescribing guidance advises caution with this drug class in people 65 and older. Menopausal hot flashes disrupt sleep for many women, and treating vasomotor symptoms with hormone therapy or an approved non-hormonal agent can improve associated sleep complaints.

Plausible but not fully settled at the population level cited here: The exact size of effects (for example, precise percentage reductions in hot flash frequency, precise minutes of sleep gained from a given melatonin dose, or precise re-entrainment rates for jet lag) varies across trials and populations. Numbers like "23 minutes of lost sleep" or "54% reduction in sleep-onset latency" appear in the sleep medicine literature in some form, but the exact figures depend on the specific study, dose, and population, and should be verified against the primary paper before being quoted as a fixed clinical fact.

Not established from the material available here: Melatonin's safety profile in human pregnancy has not been established through controlled human trials. Universal numeric prevalence figures (such as a single fixed percentage of shift workers who develop shift work sleep disorder, or a single fixed percentage of pregnant women reporting sleep disturbance) vary substantially by study definition and should be treated as approximate ranges, not precise facts.

How sleep changes with normal aging, and when it is not "just aging"

Older adults typically spend less time in deep, slow-wave sleep and shift toward an earlier circadian phase, so many people feel sleepy earlier in the evening and wake earlier in the morning than they did at 40. This is a normal shift tied to age-related changes in the suprachiasmatic nucleus, the brain's master circadian clock, and it is not itself a disease.

What turns normal age-related change into a treatable problem is usually a comorbidity layered on top: nocturia, chronic pain, OSA, restless legs syndrome, or depression. Each becomes more common after 60 and each independently fragments sleep. This matters clinically because "I don't sleep well anymore" in a 70-year-old is a symptom that needs a cause identified, not a diagnosis on its own.

OSA is common in older adults and is under-recognized because the classic presentation (loud snoring, witnessed apneas) is less reliable in this group, especially in women. A STOP-BANG screening score of 3 or higher, or an Epworth Sleepiness Scale score above 10, is a reasonable threshold for referring a patient for polysomnography rather than starting a sleep medication empirically. Restless legs syndrome is also more common with age; when pharmacotherapy is warranted, sleep medicine guidelines have historically favored low-dose dopamine agonists or alpha-2-delta calcium channel ligands, though current guidance should be checked directly, since RLS treatment recommendations have shifted over time due to concerns about dopamine agonist augmentation.

CBT-I: the first-line treatment most patients never receive

CBT-I combines sleep restriction, stimulus control, cognitive restructuring, and sleep hygiene education, typically delivered over six to eight sessions. Randomized evidence, including trials of digital CBT-I programs, has generally found meaningful improvements in sleep-onset latency and total sleep time, with benefits that tend to hold up better over the following year than medication benefits do once the medication stops. This is the basis for guideline bodies recommending CBT-I ahead of medication as the initial treatment for chronic insomnia disorder in adults.

For older adults specifically, the sleep restriction component (deliberately limiting time in bed to consolidate sleep) needs a cautious starting point. A common clinical approach is to avoid restricting time in bed below roughly six hours at the outset in older or medically frail patients, and to advance the window gradually based on measured sleep efficiency, rather than applying a single fixed protocol regardless of age or cardiovascular status. This is a judgment call for the treating clinician, not a fixed rule with strong trial support at every age cutoff.

Insomnia in postmenopausal women

Vasomotor symptoms, hot flashes and night sweats, are a well-established driver of sleep fragmentation in the menopause transition. Each episode is brief but produces a cortical arousal, and women with frequent moderate-to-severe hot flashes report measurably worse sleep than asymptomatic peers, though the exact minutes-per-night figure varies by study and should not be treated as a fixed number.

Menopausal hormone therapy (MHT) addresses the underlying vasomotor trigger for many women and has been studied for sleep-specific outcomes, generally showing improvement in nocturnal waking when vasomotor symptoms are the driver. Current guidance from menopause specialty societies supports MHT as a reasonable option for healthy women under 60 or within about 10 years of menopause onset, with the decision individualized based on personal risk factors; this is guideline-level judgment, not a claim that MHT is appropriate for every postmenopausal woman with insomnia.

For women in whom MHT is contraindicated (active breast cancer, unexplained vaginal bleeding, active thromboembolic disease), non-hormonal options with FDA approval specifically for vasomotor symptoms include low-dose paroxetine and fezolinetant, a neurokinin-3 receptor antagonist approved by the FDA in 2023. Both reduce hot flash frequency in trial populations; exact effect sizes should be checked against the current FDA label and the pivotal trial reports rather than quoted from memory. CBT-I remains a reasonable insomnia treatment regardless of whether vasomotor symptoms are separately treated.

Postmenopausal women should also be screened for OSA. Loss of progesterone, which appears to help maintain upper-airway muscle tone, is a proposed mechanism for why OSA prevalence rises after menopause and approaches rates seen in men of similar age. This is a plausible mechanistic explanation more than a fully quantified, settled figure.

Sleep disorders in pregnancy

Sleep disturbance is common throughout pregnancy, but the driver shifts by trimester. Early pregnancy brings progesterone-driven daytime sleepiness and nocturia; many women sleep more total hours but report worse sleep quality due to fragmentation. The second trimester often brings relative improvement. The third trimester introduces mechanical discomfort, reflux, fetal movement, and leg cramps, which reduce sleep efficiency for most women.

OSA can emerge or worsen during pregnancy as gestational weight gain and airway edema increase upper-airway resistance. Sleep-disordered breathing in pregnancy has been associated with adverse outcomes including gestational hypertension and glucose intolerance in observational cohorts; the strength and precise magnitude of this association across different populations needs verification against the primary studies before being cited as a fixed relative risk.

Medication options for sleep in pregnancy are limited:

• Benzodiazepines and Z-drugs are generally avoided in pregnancy given limited safety data and a plausible risk of neonatal withdrawal effects with late-pregnancy exposure.
• Low-dose doxylamine combined with pyridoxine is FDA-approved for pregnancy-related nausea and has a longer pregnancy safety record than most alternatives, though its use specifically for sleep is off-label.
• Diphenhydramine has been used short-term in pregnancy without a strong observational signal for teratogenicity, but tolerance to its sedating effect develops within a few nights, limiting its usefulness.
• Melatonin does not have an established human pregnancy safety record and is not FDA-approved for sleep disorders in pregnancy or in general. Animal data suggest fetal melatonin receptors respond to exogenous melatonin; whether this matters clinically in humans has not been established. Caution and discussion with the treating obstetric provider is warranted before use.

Positional therapy (avoiding the supine position, favoring left lateral sleep) is a reasonable, low-risk first step for pregnant patients with sleep-disordered breathing, and CPAP is considered safe and appropriate for pregnant women with confirmed moderate-to-severe OSA. Behavioral sleep approaches, including CBT-I techniques, are considered safe throughout pregnancy and do not carry known fetal risk.

Shift work sleep disorder

Shift work sleep disorder is a recognized circadian rhythm disorder defined by insomnia or excessive sleepiness that coincides with a work schedule overlapping the person's normal sleep period, present for at least three months. Reported prevalence among shift workers varies widely across studies depending on definition and population; treat any single fixed percentage in this space as approximate.

Rotating and night shift work has been associated with elevated cardiovascular and metabolic risk in large cohort studies, and older shift workers may be more vulnerable because age-related circadian changes compound the misalignment from a rotating schedule. A reasonable, guideline-informed management approach combines:

• Timed bright light exposure after a night shift, paired with light-blocking sunglasses during the commute home, to help shift the circadian clock in the intended direction.
• Low-dose melatonin (commonly in the 0.5 mg range rather than higher doses) taken before the desired daytime sleep period, since higher doses do not clearly improve outcomes and increase next-day grogginess in some studies.
• Wakefulness-promoting medication during shifts when needed. Modafinil and armodafinil are FDA-approved for shift work disorder specifically to promote wakefulness during night shifts. Both have been associated with modest increases in blood pressure in trials, so a cardiovascular risk review, and cardiology input for patients with existing cardiovascular disease, is reasonable before starting either drug, particularly in older adults.

Jet lag

Jet lag results from the mismatch between the body's internal clock and the new local time after crossing multiple time zones. The circadian system re-entrains gradually rather than instantly, and most sleep medicine sources describe eastward travel as harder to adjust to than westward travel, because phase advancement (shifting the internal clock earlier) is physiologically more difficult than phase delay. The precise number of days to full re-entrainment depends on the number of zones crossed and the individual, and should be treated as a rough estimate rather than a fixed formula.

Practical, evidence-informed strategies include:

• Melatonin at destination bedtime. Trials in this area have generally found melatonin useful for reducing jet lag severity when crossing five or more time zones, with low doses (around 0.5 mg) performing comparably to higher doses (up to 5 mg) but with fewer side effects such as morning grogginess.
• Pre-flight circadian shifting. Gradually shifting sleep and wake times by an hour or so per day in the direction of travel, starting a few days before departure, is a commonly recommended strategy, though the size of the benefit varies by individual.
• Timed light exposure at the destination. Eastward travelers are generally advised to seek morning light and avoid late-night light for the first couple of nights; westward travelers benefit more from evening light exposure.
• Short-acting hypnotics for acute symptoms. Low-dose zolpidem is sometimes used for a few nights to manage sleep-onset difficulty after travel, but the same boxed-warning cautions that apply to Z-drugs in general (impaired next-day function, complex sleep behaviors) apply here too, and older travelers are more vulnerable to these effects. Anyone using a sedative-hypnotic for jet lag should ensure a full night of available sleep time before any driving.

Medication caution in older adults: what geriatric prescribing guidance emphasizes

Geriatric prescribing guidance, including the American Geriatrics Society's periodically updated Beers Criteria, has consistently flagged benzodiazepines, Z-drugs, and first-generation antihistamines (diphenhydramine, doxylamine) as medications to use with caution or avoid in adults 65 and older, because of increased sensitivity to central nervous system effects and associated risks of falls, fractures, and cognitive impairment. The FDA has also issued safety communications regarding sedative-hypnotic risks, including impaired next-day function and rare complex sleep behaviors; current guidance should be checked directly at fda.gov, since safety communications and label warnings are updated over time.

When medication is genuinely needed after CBT-I has been tried and failed, options that are generally viewed as lower-risk in older adults, subject to individual contraindications and clinician judgment, include:

• Low-dose doxepin, FDA-approved specifically for sleep-maintenance insomnia at doses far below its antidepressant dose range, with a mechanism limited mainly to histamine blockade.
• Orexin receptor antagonists (suvorexant, lemborexant), a newer drug class with trial data in older adults showing improved sleep measures without some of the impairment concerns seen with older sedative-hypnotics, though a lower starting dose is typically used in this age group.
• Ramelteon, a melatonin receptor agonist not flagged by the Beers Criteria, with a favorable safety profile but a modest effect size on sleep-onset latency compared with placebo, meaning it is unlikely to be sufficient alone for more significant insomnia.

None of this is dosing guidance for an individual reader. Choice of agent, starting dose, and duration depend on kidney and liver function, other medications, fall risk, and comorbidities, and should be decided with the prescribing clinician.

A decision framework: which sleep problem is this, and what should happen first

This framework is designed to help a reader (or a clinician doing initial triage) identify which of the five mechanisms above is most likely driving a given sleep complaint, and what the reasonable first step is before jumping to medication.

Step 1: Rule out sleep apnea first, regardless of age or sex. Trigger: STOP-BANG score of 3 or higher, Epworth Sleepiness Scale above 10, witnessed apneas, or new/worsening insomnia in a postmenopausal or pregnant patient. Action: Refer for polysomnography or a validated home sleep apnea test before starting any sedative-hypnotic. Starting a sleep medication in an undiagnosed OSA patient can worsen outcomes.

Step 2: Match the complaint to its likely driver before choosing a treatment.

Step 3: Try CBT-I, or at minimum its core components, before a hypnotic prescription, unless there is an urgent safety issue (for example, dangerous sleep deprivation affecting a patient's ability to function safely, in which case short-term medication alongside behavioral treatment may be reasonable).

Step 4: If medication is needed, match the choice to the patient's age and risk profile, favoring agents with lower fall, cognitive, and dependence risk in adults 65 and older, and avoiding melatonin, benzodiazepines, and Z-drugs by default in pregnancy.

When to seek care urgently rather than following this framework at home: sudden severe daytime sleepiness with near-miss driving events, chest pain or irregular heartbeat associated with a wakefulness agent, suspected complex sleep behaviors (sleepwalking, sleep-driving, or unusual eating during sleep on a Z-drug), or any new sleep complaint in pregnancy accompanied by high blood pressure, swelling, or reduced fetal movement, which needs same-day obstetric evaluation rather than a sleep-focused workaround.

Common questions

How much sleep do adults over 65 need? Most sleep medicine guidance recommends roughly 7 to 8 hours per night for older adults, similar to younger adults, though individual need varies. Persistent short sleep or a large recent change in sleep duration is worth discussing with a clinician rather than assumed to be a normal part of aging.

Is it normal to wake up several times a night with age? Brief awakenings become more common with age as slow-wave sleep decreases. Frequent awakenings with substantial cumulative time lying awake, or daytime impairment from poor sleep, meet criteria for an insomnia disorder that deserves evaluation rather than being written off as "just aging."

Does melatonin work for older adults? Low-dose melatonin (commonly well under 5 mg) may help with circadian-related complaints such as an unusually early or unusually late sleep phase, or with jet lag. Its effect on sleep-onset latency in general insomnia is modest, and it is not generally considered adequate as the sole treatment for moderate to severe chronic insomnia.

Can pregnant women take melatonin? Melatonin is not FDA-approved for use in pregnancy and human safety data are limited. This should be discussed with the treating obstetric provider rather than started independently.

Is eastward or westward jet lag worse? Most descriptions of jet lag physiology treat eastward travel as more difficult because the body's clock adjusts more easily to a delay than to an advance. The exact number of recovery days depends on the number of zones crossed and the individual.

Evidence sources and verification notes

The clinical statements above draw on general sleep medicine and geriatric prescribing literature, including CBT-I trial evidence, menopause society guidance on hormone therapy, FDA drug safety communications on sedative-hypnotics, and shift work and jet lag circadian research. Many precise numeric figures that circulate in secondary sleep medicine writing (exact percentage reductions, exact minutes gained, exact prevalence rates) vary across the underlying primary studies. Where a specific number appears above, treat it as an approximate, literature-informed figure rather than a verified constant, and confirm against the current primary source or guideline before using it in a clinical or patient-facing context.

For current, dated regulatory information on any specific medication mentioned here (approval status, boxed warnings, or label changes), consult the FDA's drug safety pages directly at fda.gov/drugs/drug-safety-and-availability.

This article is intended for general education and does not provide individualized diagnosis, dosing, or treatment recommendations. Sleep complaints that are new, severe, or accompanied by other symptoms should be evaluated by a qualified clinician.