What the evidence shows for treating chronic insomnia after 50

Chronic insomnia means trouble falling or staying asleep at least three nights a week for three months or longer, with daytime consequences such as fatigue, low mood, or trouble concentrating. It is common after 50, and the questions people ask are practical ones: does therapy really beat a pill, and are the common sleeping drugs safe as we age? Two large U.S. guideline bodies have answered the first question, and the evidence on the drugs is more sobering than the advertising suggests. This is a report on what the trials and guidelines found, not medical advice.

The guideline hierarchy puts therapy first

In 2016 the American College of Physicians (ACP) recommended that all adults with chronic insomnia receive cognitive behavioral therapy for insomnia (CBT-I) as the initial treatment, reserving medication for a shared decision when CBT-I alone does not work (Annals of Internal Medicine, 2016). The American Academy of Sleep Medicine (AASM) reached the same conclusion in its 2021 behavioral-treatment guideline, giving CBT-I a “strong” recommendation, its highest tier, while several component techniques used alone earned only “conditional” recommendations (Journal of Clinical Sleep Medicine, 2021). A strong recommendation in this system means the benefits clearly outweigh the harms for most patients.

The reasoning is not that CBT-I is dramatically more powerful than drugs in head-to-head studies. The ACP noted there was insufficient direct comparison. The case rests on durability and safety: CBT-I improves sleep with few or no reported harms, and its benefits tend to persist after treatment ends, which is not true of hypnotics.

What CBT-I involves

CBT-I is not one technique but a package delivered over roughly four to eight sessions (AASM, 2021). It combines behavioral and cognitive work:

• Sleep restriction temporarily limits time in bed to match the hours a person is truly sleeping, then expands it gradually. This raises the drive to sleep and consolidates fragmented nights.
• Stimulus control re-links the bed with sleep by directing the person to use the bed only for sleep, to leave the bedroom when unable to sleep, and to keep a fixed wake time.
• Cognitive work targets the anxious, catastrophic thoughts about sleep loss that keep the nervous system aroused at night.

Education about sleep habits (sleep hygiene) is usually folded in, but as a supporting element, not the treatment itself.

The trial evidence

A meta-analysis of 20 randomized trials with 1,162 patients found that CBT-I, compared with inactive controls, shortened the time to fall asleep by about 19 minutes, cut time awake after sleep onset by about 26 minutes, and raised sleep efficiency by nearly 10 percentage points, with no adverse outcomes reported and gains sustained at later follow-up (Trauer et al., Annals of Internal Medicine, 2015). These are moderate improvements on paper, but the persistence is the distinguishing feature.

Digital and app-delivered CBT-I

Access to a trained CBT-I therapist is a real bottleneck, and digital programs aim to close it. A 2025 randomized trial of an internet-delivered CBT-I program (SHUTi OASIS) in 311 adults aged 55 to 95 found that, at 12-month follow-up, roughly two to three times as many older adults reached meaningful improvement or remission compared with an online education control (npj Digital Medicine, 2025). Digital CBT-I is not identical to seeing a specialist, and effects vary by program and by how much support is attached, but the randomized evidence for the format is real.

Why sleep hygiene alone is generally not enough

Sleep hygiene, the familiar list about caffeine, screens, and a cool dark room, is the intervention many people try first. Guideline reviewers found it wanting as a standalone treatment. The AASM 2021 guideline concluded that the evidence did not support sleep hygiene used by itself and pointed instead to the multicomponent CBT-I package. Sleep hygiene is best understood as necessary background, not a substitute for the behavioral core of treatment.

Prescription hypnotics: modest and short-term

The AASM issued a separate 2017 guideline for insomnia drugs. Every recommendation in it, including those for zolpidem, eszopiclone, and temazepam, was “weak” (Sateia et al., Journal of Clinical Sleep Medicine, 2017), meaning the evidence of net benefit was uncertain. The size of the drug effect helps explain that caution. In an analysis of z-drug data submitted to the FDA, the drugs shortened measured sleep latency by about 22 minutes on sleep studies but by far less on patients’ own reports, and the authors judged the effect small and close to the placebo response, which accounted for roughly half of the total (Huedo-Medina et al., BMJ, 2012). Benzodiazepines and z-drugs also raise concerns about tolerance and dependence with continued use, which is why guidelines frame them as short-term options.

The specific risks after 50

For older adults the risk side of the ledger grows. The 2023 American Geriatrics Society Beers Criteria advise avoiding benzodiazepines in older adults because they increase cognitive impairment, delirium, falls, fractures, and motor vehicle crashes, and the criteria extend the same avoidance to the z-drugs, noting their adverse events are similar (AGS, 2023). Next-day impairment is not hypothetical: in 2013 the FDA lowered the recommended zolpidem dose for women after data showed about 15% of women still had blood levels high enough to impair driving eight hours after a 10 mg dose (FDA, 2013).

Newer agents: dual orexin receptor antagonists

The dual orexin receptor antagonists (DORAs) block the brain’s wake-promoting orexin signaling rather than broadly sedating. Suvorexant, lemborexant, and daridorexant are FDA-approved in this class. In two phase 3 trials, daridorexant 50 mg reduced wake after sleep onset by about 23 minutes versus placebo at one month and improved daytime functioning, with a favorable safety profile (Mignot et al., The Lancet Neurology, 2022). These are promising results, but suvorexant also received only a weak recommendation in the 2017 AASM drug guideline, and the class does not displace CBT-I in the hierarchy. Whether DORAs carry less fall and cognitive risk than older hypnotics in adults over 50 is an active question rather than a settled one.

This article is evidence reporting, not medical advice. Treatment decisions, including whether to start, change, or stop any medication, should be made with a qualified clinician who knows your full history.

Common questions

Is CBT-I really more effective than sleeping pills?

Guidelines recommend CBT-I first mainly because its benefits last and it carries few harms, not because it dwarfs drugs in direct comparisons. The ACP noted there was insufficient head-to-head evidence to declare one more potent than the other. Where the two differ clearly is durability: CBT-I gains tend to persist after treatment ends, while hypnotic benefits stop when the drug does and can bring tolerance and next-day effects (Annals of Internal Medicine, 2016).

Why isn’t good sleep hygiene enough on its own?

The AASM 2021 guideline reviewed sleep hygiene as a standalone treatment and found the evidence did not support it by itself. Habits like limiting caffeine and screens are useful background, but the behavioral core of treatment (sleep restriction and stimulus control) is what carries the effect in trials.

Are z-drugs like zolpidem dangerous for older adults?

The 2023 AGS Beers Criteria list z-drugs among medications older adults should avoid, citing falls, fractures, delirium, and next-day impairment, similar to benzodiazepines. The FDA also lowered the recommended zolpidem dose for women in 2013 over morning driving-impairment risk. These are population-level cautions; decisions about any specific person’s medication belong with their clinician.

Do the newer orexin drugs solve the safety problem?

Dual orexin receptor antagonists such as daridorexant reduced nighttime wakefulness in phase 3 trials with a favorable safety profile (Lancet Neurology, 2022), and they work differently from older sedatives. But they still carry only weak guideline recommendations, and it is not yet established that they are safer than older hypnotics specifically for falls and cognition in adults over 50.

Does digital or app-based CBT-I work if I can’t find a therapist?

Randomized trials of internet-delivered CBT-I, including one in adults aged 55 to 95, found meaningful improvements in insomnia severity that held up at follow-up (npj Digital Medicine, 2025). Effects vary by program and by how much human support is attached, and digital delivery is not identical to seeing a specialist, but the format has genuine randomized evidence behind it.