Mirtazapine for Sleep

Mirtazapine is approved by the FDA for one indication: major depressive disorder in adults. Sleep is not on the label. StatPearls lists insomnia first among its off-label uses, alongside panic disorder, PTSD, social anxiety disorder and fibromyalgia. That framing matters more than it sounds. When you take mirtazapine for sleep, you are borrowing a side effect of an antidepressant rather than taking something designed and tested as a hypnotic.

It blocks central presynaptic alpha-2 adrenergic receptors and acts as a potent antagonist at histamine H1 receptors and at 5-HT2A, 5-HT2C and 5-HT3 serotonin receptors. The H1 blockade is where the drowsiness lives, and it is unusually heavy. In the six-week US depression trials the label reports somnolence in 54% of the mirtazapine arm against 18% on placebo.

Very few prescription drugs sedate more than half of a trial arm.

It is also why the label tells prescribers to give the dose in the evening prior to sleep even when the target is depression. The sedation is not incidental to the drug, it is one of its loudest effects, and it arrives on the first night rather than after the four to six weeks a mood response usually needs. That gap is the whole reason mirtazapine ended up being used as a sleep drug at all.

You will read almost everywhere that mirtazapine is more sedating at 7.5 mg or 15 mg than at 30 mg or 45 mg, because noradrenergic activity recruited at higher doses offsets the antihistamine effect. A British Columbia Provincial Academic Detailing review traced that claim to tertiary references such as UpToDate and the Clinical Handbook of Psychotropic Drugs, and to the speculation that the drug is more noradrenergic at higher doses. It is a pharmacology argument, repeated until it hardened into a fact.

That same review concluded the relationship between mirtazapine dose and sedation is unclear, and recorded that the FDA identified somnolence as the most common adverse event causing discontinuation during the 1996 regulatory review and requested a postmarketing dose-response trial, which its literature search could not find completed. Three decades on, the single most repeated claim about this drug still rests on reasoning rather than a trial designed to answer it.

The one analysis that went at it directly, a review of 308 reports of anxiety, agitation, delusion, hallucination, hypertension, insomnia, nightmare or tachycardia in the FDA Adverse Event Reporting System between 1995 and 2015, failed to support the hypothesis that mirtazapine is more activating at higher doses. None of that makes low dosing wrong. It means the defensible reason to start low is tolerability. There is no proven sedation curve to lean on. Your clinician is the one who weighs the dose.

DREAMING, a pragmatic double-blind placebo-controlled trial run in general practices around Amsterdam, randomized 80 adults aged 18 to 85 with insomnia disorder and sleep maintenance problems to mirtazapine at 7.5 to 15 mg a day, amitriptyline at 10 to 20 mg a day, or placebo for 16 weeks. At six weeks mirtazapine lowered Insomnia Severity Index scores by 6.0 points more than placebo, with a 95% confidence interval of -9.0 to -3.0 on a 0 to 28 scale.

Improvement rates were 52% on mirtazapine against 14% on placebo, and recovery rates 56% against 14%, while amitriptyline reached statistical significance without a clinically relevant size. Those are strong numbers for a primary-care sleep intervention, and until this trial published in the British Journal of General Practice in 2025 the off-label use had essentially no placebo-controlled evidence behind it at all. Worth holding onto the scale, though: 80 randomized participants is a small trial, which is why the confidence interval on that 6.0-point difference is as wide as it is.

Then the window closes. From 12 weeks onward the trial found no statistically significant differences in ISI scores, and the authors concluded the results do not support prescribing low-dose mirtazapine for several months, suggesting instead a course of roughly six weeks where non-drug treatment has not been enough.

If you take one thing from this page, take that shape: a real early effect, and no demonstrated advantage once you are past it.

The cleanest case for mirtazapine is insomnia sitting on top of depression, because then the one indication it actually holds is doing real work and the sedation is a bonus rather than the entire point. There is a caution attached even there: an item-level meta-analysis of 37 trials and 5,974 participants traced mirtazapine's edge over SSRIs on the HDRS-17 total to items reflecting sleep, appetite and gastrointestinal dysfunction, with SSRIs or venlafaxine more effective on depressed mood, suicidality and psychic anxiety.

The evidence also points at a specific sleep complaint. DREAMING enrolled only adults with sleep maintenance problems, so the trial support is strongest for waking at 3am and not getting back down, rather than for lying awake at bedtime. And with appetite increase reported in 17% against 2% on placebo, the same effect that pushes one person off the drug is sometimes the reason a clinician chooses it for someone underweight or eating poorly.

The 2017 American Academy of Sleep Medicine guideline on the pharmacologic treatment of chronic insomnia issued weak recommendations for suvorexant, eszopiclone, zaleplon, zolpidem, triazolam, temazepam, ramelteon and doxepin, and weak recommendations against trazodone, tiagabine, diphenhydramine, melatonin, tryptophan and valerian. Mirtazapine appears nowhere in it. The task force picked its 14 interventions from FDA approval status and common off-label use, and mirtazapine did not make the list. For trazodone the guideline suggests clinicians not use it for sleep onset or sleep maintenance insomnia in adults, a weak recommendation drawn from trials of the 50 mg dose. So the honest comparison is not one winner and one loser. It is a drug with a guideline recommendation against it versus a drug the guideline never evaluated. Our trazodone page works through that evidence separately.

One real difference is worth knowing. A Mayo Clinic systematic review of 18 prospective studies found that new or worsened restless legs syndrome and rises in periodic limb movements are uncommon after starting an antidepressant, but that mirtazapine may carry higher rates of both, while trazodone, nefazodone and doxepin do not appear to aggravate periodic limb movements. If your legs already bother you at night, raise it before you start.

The label sets a starting dose of 15 mg once daily taken orally, preferably in the evening prior to sleep, with increases to a maximum of 45 mg a day and no dose change at intervals shorter than one to two weeks. Sleep prescribing sits below that floor: DREAMING used 7.5 to 15 mg a day and a randomized crossover trial in 19 healthy men used a single 7.5 mg dose taken 30 minutes before lights out.

In that crossover trial, under all-night traffic noise used as a model of transient insomnia, 7.5 mg mirtazapine and 50 mg quetiapine each raised total sleep time by about half an hour and cut the number of awakenings by 35 to 40% against placebo, with mirtazapine mainly increasing deep stage N3 sleep. Both also produced daytime sleepiness and measurably worse sustained attention the following day. The benefit and the hangover came from the same dose.

Elimination half-life runs roughly 20 to 40 hours, averaging 37 hours in females against 26 hours in males. A drug that long-acting does not clear by your alarm. That is the arithmetic behind next-day grogginess, and it is why how early you take it and how much sleep runway you leave yourself matter far more here than with a short-acting hypnotic. Dose escalation is a separate conversation: efficacy in depression climbs to about 30 mg and then falls, while dropouts from adverse effects keep rising steeply with dose.

Weight gets the headlines, but oversedation is what actually ends courses. In the six-week US depression trials somnolence led 10.4% to discontinue mirtazapine against 2.2% on placebo, the most common reason for stopping, while 8% discontinued for weight gain across a pool of premarketing studies that included long-term open-label treatment. Both matter, and they pull in opposite directions when a clinician is choosing a dose.

The metabolic side is well quantified. Appetite increase was reported in 17% against 2% on placebo, weight gain in 12% against 2%, and a gain of at least 7% of body weight in 7.5% against 0%. Dry mouth ran 25% against 15%, constipation 13% against 7%, and nonfasting cholesterol rises of 20% or more above the upper limit of normal appeared in 15% against 7% on placebo. None of that is subtle at six months.

The label carries a boxed warning that antidepressants increased the risk of suicidal thoughts and behaviors in children and young adults in short-term studies, with pooled trials showing 14 additional cases per 1000 treated under age 18 and 5 additional per 1000 at ages 18 to 24, against one fewer per 1000 at ages 25 to 64. If you are under 25, that warning is the reason for close monitoring in the first months rather than a reason the drug is off limits. Agranulocytosis is rare but documented, so a sore throat, fever or mouth sores during treatment warrants a same-day call.

The American College of Physicians recommends that all adults receive cognitive behavioral therapy for insomnia as the initial treatment for chronic insomnia disorder, a strong recommendation on moderate-quality evidence, with shared decision-making about short-term medication reserved for when CBT-I alone has not worked. Mirtazapine does not replace that. Given that the trial benefit stops being significant after six weeks, running CBT-I alongside the tablets is what leaves you with something once the tablets stop.

The label contraindicates mirtazapine with MAOIs and requires at least 14 days between stopping one and starting the other in either direction, and warns about serotonin syndrome when it is combined with other serotonergic drugs. It also flags that somnolence may impair judgment, thinking and motor skills, so alcohol, antihistamines and other sedatives compound an effect that is already strong on its own. List everything you take, including over-the-counter sleep aids and supplements.

Mirtazapine is a generic oral tablet with no controlled-substance scheduling, so a telehealth service that treats depression and insomnia can prescribe it after an intake and a clinician review. A licensed clinician in your state decides whether it is appropriate for you, and that decision should turn on whether low mood sits underneath the sleep problem, whether you are waking through the night or struggling to fall asleep, and what a weight change would mean for you. None of this is medical advice.

Judge a service on whether it will follow you rather than on how quickly it says yes. Look for a named prescriber licensed where you live, an intake that actually asks about mood, weight, other sedatives and suicidal thoughts, a stated starting dose with a reason behind it, and a review booked at around six weeks. Since the measured advantage over placebo does not persist past that point, a service with no plan for week six is selling you a refill loop rather than treatment.

The CMS National Average Drug Acquisition Cost file, published on 2 September 2026, prices a generic mirtazapine 15 mg tablet at about $0.065, a 30 mg tablet at about $0.087 and a 45 mg tablet at about $0.105, while the 7.5 mg tablet runs about $0.353. The lowest strength, the one most often reached for at bedtime, costs pharmacies more than five times the 15 mg tablet. The orally disintegrating forms sit higher again at roughly $0.332 for 15 mg.

That inversion is worth raising with your prescriber, because 15 mg tablets are scored and a clinician may be able to reach the same low dose from a splittable tablet. NADAC is what pharmacies pay rather than your counter price, but it sets the floor, and at these numbers the drug itself is never the expensive line. Thirty 15 mg tablets is under two dollars at that benchmark. Your monthly cost is the consultation, the dispensing markup and the follow-ups.