UK insomnia medications - the honest picture

Medication has a legitimate place in insomnia management. It is not the first-line treatment (that is CBT-I, covered in its own article), it is not usually a long-term plan, and the choice of agent matters more than most GP prescribing patterns suggest. This article walks through the main UK options honestly, including what they do well and what they do badly.

The short-term role of medication

For acute insomnia triggered by a clear stressor (bereavement, major surgery, overseas travel, short-term work crisis), two to four weeks of a sleeping tablet can prevent the episode becoming chronic. Three weeks is roughly the timeline at which acute sleep disruption consolidates into long-term hyperarousal; bridging that period with medication while addressing the stressor is legitimate.

For chronic insomnia, medication is usually an adjunct to behavioural treatment rather than a primary solution. Used alone, it provides nightly symptom relief without treating the underlying problem, and long-term use comes with dependence risk and cumulative side effects.

Z-drugs (zopiclone, zolpidem)

The most commonly prescribed sleeping tablets in UK general practice. Zopiclone at 3.75 or 7.5mg and zolpidem at 5 or 10mg are the standard preparations.

What they do. Non-benzodiazepine GABA-receptor modulators. Reduce sleep latency by 15 to 20 minutes; reduce middle-of-night awakenings; modestly increase total sleep time. Subjectively, most users report "real" sleep rather than drugged unconsciousness.

Pros. Reasonable short-term effectiveness. Faster onset than benzodiazepines. Shorter half-life means less next-day hangover with zolpidem and modest with zopiclone.

Cons. Taste disturbance (zopiclone famously leaves a metallic taste). Tolerance with nightly use. Dependence risk if continued beyond 4 to 6 weeks. Rebound insomnia on stopping. Parasomnia risk (sleep walking, sleep eating, sleep driving) - rare but real.

UK guidance. NICE recommends no longer than 4 weeks of regular use. In practice, GP prescribing often runs much longer. Patients come to us on zopiclone for 2, 5, 10 years and struggle to stop.

Our approach. Short-term bridging only. We do not run long-term zopiclone repeat-prescribing. Patients already on long-term Z-drugs are offered structured reduction with concurrent CBT-I.

Benzodiazepines (diazepam, temazepam, lorazepam)

Older class. Largely superseded for sleep by Z-drugs but still used in specific circumstances.

What they do. Longer-acting GABA-receptor modulators. Produce sleep induction, anxiolysis, muscle relaxation.

Pros. Well-established. Useful in mixed anxiety-insomnia presentations. Long half-life of diazepam provides smoother daytime anxiolysis for short periods.

Cons. Significant dependence risk. Prominent daytime sedation with longer-acting agents. Cognitive impairment with long-term use, particularly in older adults. Withdrawal syndrome on stopping, sometimes protracted. Associated with falls, road traffic collisions, and impaired driving.

Our approach. Not routinely prescribed for sleep. In rare short-term use for mixed anxiety-insomnia, lorazepam 0.5 to 1mg short course. Not a repeat-prescribing service for benzodiazepines.

Trazodone

An old antidepressant widely used off-label for insomnia, particularly in the US. Available in UK on prescription.

What it does. Sedating antidepressant via histamine and serotonin receptor effects. Sleep-inducing effect is usually within 30 to 60 minutes of a 50 to 100mg evening dose.

Pros. Non-addictive. Cheap. Useful in patients with concurrent depression and insomnia. Generally well tolerated.

Cons. Morning grogginess at higher doses. Orthostatic hypotension in some older adults. Priapism (very rare but serious). Dose required for sleep (50 to 100mg) is much lower than for depression (150 to 400mg), which confuses clinicians and patients.

Our approach. A reasonable option for chronic insomnia where medication is being used alongside CBT-I, particularly in patients who also have low mood component. Not first-line but a reasonable second-line for the right patient.

Mirtazapine

Another sedating antidepressant. 15 or 30mg evening dose has sleep-inducing effect via histamine antagonism.

What it does. Improves sleep onset, increases slow-wave sleep, reduces middle-of-night awakenings. Anxiolytic and antidepressant effects useful in mixed presentations.

Pros. Non-addictive. Genuinely useful in depressive-insomnia. Improves appetite (useful in some patients, unwanted in others). Cheap.

Cons. Weight gain, often substantial. Morning sedation, particularly at 15mg (paradoxically more sedating than 30mg in some patients). Restless legs worsening in susceptible patients.

Our approach. A reasonable option for patients with co-existing depression and chronic insomnia. Not suitable for patients already carrying excess weight. Useful in thin older adults with depression and poor sleep.

Amitriptyline (and doxepin)

Old sedating antidepressants used off-label at low doses for sleep. Amitriptyline in particular is prescribed a great deal.

The honest evidence. For amitriptyline it is weak. A 2025 UK general-practice trial (DREAMING) found low-dose amitriptyline gave a statistically detectable but not clinically meaningful improvement at six weeks, and insomnia guidelines do not recommend it. It also carries an anticholinergic load that matters in older adults.

Doxepin is the better-evidenced cousin: at very low doses (3 to 6mg) it has solid trial support for helping people stay asleep and is well tolerated, though in the UK it usually comes as higher-dose capsules used off-label.

Our approach. We will prescribe a low-dose sedating antidepressant where it genuinely fits, particularly alongside low mood, but we are straight that amitriptyline is more popular than its evidence justifies.

Antihistamines (diphenhydramine, promethazine)

Over the counter and on prescription. Old, cheap, widely used, and sold as Nytol and similar.

What they do. Block H1 histamine receptors, which produces drowsiness.

Promethazine (Phenergan). Licensed in the UK for short-term insomnia, but the evidence is poor, it can blunt the very behavioural techniques that fix sleep for good, and it carries more dependence and misuse potential than people assume.

Cons. Anticholinergic load that matters in older adults (long-term anticholinergic use is linked to higher dementia risk), rapid tolerance, and next-day grogginess.

Our approach. Fine for the occasional bad night, a poor choice for regular long-term use.

Melatonin

Covered in its own article. Useful for circadian problems, not primarily for insomnia.

Daridorexant and the orexin antagonists

The newest evidence-based class, and the most interesting development in insomnia drugs for years. Daridorexant (brand name Quviviq) blocks orexin, the brain signal that keeps you awake, rather than sedating you like the older drugs.

What it does. Helps you fall asleep faster, stay asleep longer, and function better the next day, with trial data out to 12 months.

Where it fits. NICE recommends it for long-term insomnia (symptoms at least three nights a week for three months or more), specifically once CBT-I has been tried and not worked, or is unavailable or unsuitable. Lower dependence potential than the older hypnotics, though courses are still kept as short as the situation allows.

Our approach. A genuine option to consider after CBT-I for the right person, and we will tell you honestly where it sits for you.

Gabapentinoids (gabapentin, pregabalin)

Sometimes used for insomnia in patients with co-existing pain or restless legs syndrome.

What they do. Alpha-2-delta calcium channel modulators. Improve sleep onset and continuity, particularly in RLS-associated insomnia.

Pros. Useful for specific indications (RLS, neuropathic pain with sleep disturbance). Non-addictive in most patients.

Cons. Controlled drug status for pregabalin. Dependence risk in susceptible individuals. Dizziness, fluid retention, weight gain.

Our approach. A reasonable option for patients with RLS-related insomnia or pain-related insomnia. Not first-line for primary insomnia.

Cannabis and CBD

Increasingly asked about. Legal status in UK: prescription-only cannabis-based medicines for specific indications; CBD is available widely but not regulated for insomnia.

Evidence. Mixed. THC-containing products reduce sleep latency in some patients but disrupt REM and produce tolerance. CBD-only products have limited evidence for sleep but appear generally safe.

Our approach. We do not routinely prescribe cannabis medicines for primary insomnia (outside our medical cannabis service for specific indications). CBD is patient choice; we advise realistic expectations.

What we are cautious about

Repeat Z-drugs without review. If you come to us on long-term zopiclone or zolpidem, we will not simply continue the prescription. We will offer structured reduction with concurrent CBT-I, at a pace that suits you.

Quetiapine and olanzapine off-label. These antipsychotics are sometimes used for insomnia. The risk-benefit for simple insomnia is poor, so we keep them to their licensed psychiatric uses.

Paracetamol and night-time combination tablets. Worth clearing up: paracetamol is not a sleep aid and has no effect on sleep. The night versions of painkillers work because of an added antihistamine, not the paracetamol, so they carry the antihistamine drawbacks above.

How medication fits into our sleep care

Medication alone is not the plan for chronic insomnia. Our approach: structured CBT-I with medication only as short-term bridge if needed. For acute insomnia with a clear trigger, a 2 to 4 week course of an appropriate medication plus addressing the trigger.

For chronic insomnia where medication is being used long-term, we work with you on structured reduction while building the skills that make reduction possible.

The honest bottom line

Sleep medications have a short-term role, not a long-term one. CBT-I is first-line for chronic insomnia. Medication selection matters more than most prescribing suggests, and "just give them something" is poor medicine. The right tool for the right problem produces better outcomes than defaulting to whatever is easiest.

Ready to start?

If this article has made you think our assessment might help, the next step is a short consultation with one of our sleep-medicine doctors.

Begin your assessment at this link. Online with a WatchPAT One home study, or in person at Westfield London.