- Mirtazapine is licensed for one thing: major depression in adults. In the largest independent analysis of antidepressants (522 trials, 116,477 participants), it was one of the more effective drugs against placebo (odds ratio 1.89 for response), and people dropped out of treatment about as often as on placebo (Cipriani et al., Lancet 2018).
- Sleepiness and weight gain are expected effects, not rare surprises. In US trials, 54% of people taking it reported somnolence (18% on placebo), 17% reported increased appetite (2% on placebo) and 7.5% gained at least 7% of their body weight (0% on placebo) (FDA Remeron label).
- Compared with SSRIs, a Cochrane review found mirtazapine worked a little faster and caused less nausea and sexual dysfunction, but more weight gain and drowsiness. All but two of the 29 trials in that review were funded by the manufacturer or run with its advice (Watanabe et al., Cochrane 2011).
- Adding mirtazapine to an SSRI or SNRI when depression had not responded did not give a clinically important benefit in the NIHR-funded MIR trial (480 UK primary-care patients) (Kessler et al., BMJ 2018).
- Using it for insomnia is off-label, and the evidence is thin. The 2018 Cochrane review of antidepressants for insomnia had no mirtazapine trials to analyse. The first placebo-controlled trial in older adults with insomnia (60 people, 28 days) found a benefit, but 6 of 30 on mirtazapine stopped because of side effects (Everitt et al., Cochrane 2018, Nguyen et al., Age Ageing 2025).
- In people with Alzheimer's disease, two NIHR-funded trials found no benefit over placebo: one for depression and one for agitation. The agitation trial recorded 7 deaths on mirtazapine against 1 on placebo by week 16 (HTA-SADD, Lancet 2011, SYMBAD, Lancet 2021).
- Evidence grade (depression): Strong · Evidence grade (insomnia without depression): Weak
Independent evidence review · Prescription medicine
Mirtazapine is an effective, well-studied antidepressant with an unusual side-effect profile. It makes most people sleepy and hungry, it rarely causes the nausea and sexual problems common with SSRIs, and it is widely prescribed off-label for sleep even though almost no trials have tested that use. Its licence in the UK and the US covers major depression in adults only (UK SmPC, FDA label). This review separates what independent evidence shows from what rests mainly on trials paid for by its original maker, Organon.
Mirtazapine is a prescription-only medicine. Do not start it, stop it or change the dose without your prescriber. Like all antidepressants, it carries an FDA boxed warning: antidepressants increased suicidal thoughts and behaviours in children, adolescents and young adults in short-term studies. Anyone taking an antidepressant should be watched for worsening mood, especially in the first months and after any dose change (FDA label). If you have thoughts of suicide or self-harm, seek urgent help now: call 999 or go to A&E in the UK, call or text 988 in the US, or contact your local emergency number or crisis line. Mirtazapine can make you very drowsy. Do not drive, cycle or use machinery until you know how it affects you, and avoid alcohol and benzodiazepines unless your prescriber has agreed (NHS). Taking it with opioids, including buprenorphine, adds to the sedation and can cause serotonin syndrome (UK SmPC).
Table of contents
- Evidence summary
- What mirtazapine is
- How it works
- What it is prescribed for
- What works and what does not
- Benefits by claim
- Risks and all side effects
- All interactions
- Who should avoid mirtazapine
- Dosage and how to take it
- Follow the money: who makes it and who funded the evidence
- Related research
- Frequently asked questions
- Sources and funding notes
Evidence summary
The table below grades each main claim about mirtazapine by the strength of the human evidence and shows who paid for that evidence.
| Claim | Evidence | Source | Funding / conflict | Strength |
|---|---|---|---|---|
| More effective than placebo for acute major depression | Network meta-analysis of 522 double-blind trials. Response odds ratio vs placebo 1.89 (95% CrI 1.64–2.20); GRADE certainty moderate for most mirtazapine comparisons. | Cipriani et al., Lancet 2018 | Funded by the NIHR Oxford Health Biomedical Research Centre and the Japan Society for the Promotion of Science; funders had no role. Some authors reported fees from other antidepressant makers. Many of the underlying trials were industry-funded. | Strong |
| Dropout rate similar to placebo | Acceptability (all-cause dropout) odds ratio vs placebo 0.99 (0.85–1.15). | Cipriani et al., Lancet 2018 | As above. | Moderate |
| Works faster than SSRIs; small edge at the end of treatment | 29 trials (n = 4,974). Against SSRIs: OR 1.57 for response at 2 weeks and OR 1.19 at 6–12 weeks. | Watanabe et al., Cochrane 2011 | Academic Cochrane review. The authors state that all but two included trials were funded by, or run with the advice of, a mirtazapine manufacturer, so sponsorship bias is possible. | Moderate (industry-funded trial base) |
| Adding it to an SSRI/SNRI helps treatment-resistant depression | MIR trial, 480 UK primary-care adults. Adjusted BDI-II difference −1.83 (95% CI −3.92 to 0.27; P = 0.09) at 12 weeks; no clinically important benefit. | Kessler et al., BMJ 2018 | Funded by the NIHR HTA programme; funder had no role in the study. | Not supported |
| Helps chronic insomnia (off-label) | No mirtazapine trials were analysed in the Cochrane insomnia review. One RCT in adults aged 65 and over (n = 60, 7.5 mg, 28 days) improved Insomnia Severity Index scores; 6 vs 1 participants stopped because of adverse events. | Everitt et al., Cochrane 2018; Nguyen et al., Age Ageing 2025 | Cochrane review is academic. MIRAGE was sponsored by an academic hospital (CHUM, Montreal) according to ClinicalTrials.gov. | Weak |
| Improves sleep when insomnia is part of depression | A meta-analysis of 30 studies (16 RCTs) found mirtazapine increased total sleep time, slow-wave sleep and sleep efficiency in depressed patients. | Zhang et al., J Affect Disord 2026 | Academic authors (China) declared no competing interests; many included studies were small or non-randomised. | Moderate |
| Treats depression or agitation in Alzheimer's disease | HTA-SADD: no difference vs placebo for depression at 13 weeks. SYMBAD: no difference for agitation at 12 weeks; 7 deaths on mirtazapine vs 1 on placebo by week 16 (p = 0.065, post hoc). | Banerjee et al., Lancet 2011; Banerjee et al., Lancet 2021 | Both funded by the NIHR HTA programme. | Not supported; possible harm |
| Causes sedation and weight gain | Somnolence 54% vs 18%; increased appetite 17% vs 2%; weight gain ≥7% in 7.5% vs 0%. Cochrane: weight gain or increased appetite OR 4.23 vs SSRIs. | FDA label; Watanabe 2011 | Label trial data come from Organon's registration trials and were reviewed by the FDA. An unfavourable finding reported by the maker itself is less likely to be overstated. | Established risk |
Independent evidence and credibility scorecard
Tier 1 = public regulator or government body; Tier 2 = academic, publicly funded; Tier 3 = academic with some industry-linked authors or an industry-funded evidence base; Tier 4 = manufacturer-funded. Credibility A–D reflects methods, size and transparency.
| Source | Who funds it | Country | Independence tier (1–4) | Credibility (A–D) | Why it would tell the truth / residual bias |
|---|---|---|---|---|---|
| MIR trial (Kessler 2018) | NIHR Health Technology Assessment programme (project 11/129/76) | UK | 2 | A | The NHS needs to know whether a cheap generic add-on is worth using, and no company stood to gain. One author reported fees from other antidepressant makers, outside this work. |
| SYMBAD (Banerjee 2021) and HTA-SADD (Banerjee 2011) | NIHR HTA programme | UK | 2 | A | Public funders need honest answers about common off-label use in dementia. Several SYMBAD authors reported industry fees unrelated to mirtazapine. |
| Cipriani 2018 network meta-analysis | NIHR Oxford Health BRC; Japan Society for the Promotion of Science | UK / Japan / international | 3 | A | Methods were public and funders had no role. Some authors had fees from other antidepressant companies. The underlying trials were mostly industry-funded, and the authors note a "novelty effect" and incomplete unpublished data. |
| Cochrane: mirtazapine vs other antidepressants (Watanabe 2011) | Academic Cochrane review; lead authors reported Japanese government grants | Japan / Italy / UK | 3 | B | Cochrane methods include sponsorship-bias ratings. Two authors reported past funds from Schering-Plough, which owned Organon from 2007. Almost all included trials were manufacturer-linked. |
| NICE NG222 | UK government (Department of Health and Social Care) | UK | 1 | A | Has to balance cost and effectiveness for the NHS. Relies on the same trial base as the reviews above. |
| FDA Remeron label / UK SmPC | Written by the licence holder and approved by the regulator | US / UK | 1 (regulator-approved) | A for harms | Legal duty to list known risks. The efficacy data come from the maker's own registration trials. |
| Cochrane: antidepressants for insomnia (Everitt 2018) | Academic Cochrane review | UK | 2 | A | Independent assessment of widespread off-label prescribing. One author disclosed past research support from Organon among other companies. |
| MIRAGE insomnia trial (Nguyen 2025) | Academic hospital sponsor (CHUM) and a Quebec research network on ageing | Canada | 2 | C | No company stands to gain from a generic. The trial was small (n = 60) and short (28 days), at one site. |
| Esmirtazapine insomnia trial (Ivgy-May 2020) | Organon (then Merck); several authors were company employees | US / Europe | 4 | C (for mirtazapine) | A developer testing its own product. It studied esmirtazapine, a related single-enantiomer drug that was never marketed, not mirtazapine itself. |
What mirtazapine is
Mirtazapine is an antidepressant usually grouped as a noradrenergic and specific serotonergic antidepressant, or NaSSA (Everitt et al., Cochrane 2018). It is not an SSRI, an SNRI or a tricyclic. NHS guidance describes it as an antidepressant used to treat depression that is available only on prescription (NHS).
- Brand names: Remeron (film-coated tablets) and RemeronSolTab (orally disintegrating tablets) in the US (FDA label). In the UK it is mainly prescribed as generic mirtazapine. The electronic Medicines Compendium lists many generic tablets, orodispersible tablets and a 15 mg/ml oral solution (UK SmPC example).
- US approval: the FDA approved Remeron (NDA 020415) on 14 June 1996. The sponsor was Organon (Drugs@FDA). RemeronSolTab (NDA 021208) followed on 12 January 2001 (openFDA Drugs@FDA record).
- Originator: Organon, whose pharmaceutical business was owned by Akzo Nobel N.V. and had a major research and manufacturing site in Oss, the Netherlands (Schering-Plough press release, 2007). Ownership since then is traced in the follow-the-money section.
- Current US brand holder: Organon LLC, a subsidiary of Organon & Co. (FDA label). Organon & Co. has its head office in Jersey City, New Jersey (Organon 2025 Form 10-K).
- Generic status: widely available as a generic. FDA application records list generic holders including Aurobindo, Mylan, Sun Pharmaceutical and others (openFDA Drugs@FDA).
- Prescription status: prescription-only in both the UK and the US (NHS, FDA label).
How it works
The FDA label says plainly that how mirtazapine treats depression is unclear. The leading explanation is that it blocks presynaptic α2-adrenergic autoreceptors and heteroreceptors in the brain. These receptors normally act as brakes, so blocking them increases noradrenaline and serotonin activity. Mirtazapine also blocks 5-HT2 and 5-HT3 serotonin receptors and has no significant affinity for 5-HT1A or 5-HT1B receptors (FDA label, sections 12.1–12.2).
Its best-known side effects follow from other receptors it blocks. The FDA label links its strong sedating effect to histamine H1 blockade, and its tendency to cause dizziness on standing (orthostatic hypotension) to peripheral α1-adrenergic blockade. It also blocks muscarinic receptors (FDA label). The UK product information describes its anticholinergic activity as "very weak" (UK SmPC).
What the body does with it
- About 50% of an oral dose reaches the bloodstream, and levels peak about 2 hours after a dose. Food makes little difference (FDA label).
- The half-life is about 20–40 hours, with a mean of 37 hours in women and 26 hours in men. Steady state is reached within 5 days, which is why it can be taken once a day (FDA label).
- It is broken down in the liver by CYP2D6, CYP1A2 and CYP3A, and about 75% is excreted in urine (FDA label).
- Clearance was 40% lower in older men than in younger men, and about 30% (GFR 11–39) to 50% (GFR <10) lower in kidney impairment (FDA label).
What it is prescribed for
Licensed uses
- UK: "treatment of episodes of major depression" in adults (UK SmPC, section 4.1).
- US: major depressive disorder in adults. It is not approved for anyone under 18 (FDA label).
No UK or US licence covers insomnia, anxiety disorders, appetite stimulation or agitation in dementia. Any use for these is off-label, which means the prescriber takes on more responsibility for weighing benefit against risk.
Where guidelines place it
- NICE NG222 (depression in adults, 2022): NICE names SSRIs as "the first choice for most people" when medication is used for more severe depression, and allows an SNRI or "other antidepressant if indicated based on previous clinical and treatment history". Mirtazapine is named specifically as a further-line combination option: for people whose depression has not responded and who accept a higher side-effect burden, NICE lists "adding mirtazapine or trazodone to an SSRI" among the treatment options, ideally with specialist input (NICE NG222). The NIHR-funded MIR trial, covered below, did not show a clinically important benefit from exactly this combination in primary care, so this recommendation rests on mixed evidence.
- Insomnia: the European Insomnia Guideline 2023 recommends CBT for insomnia as the first-line treatment. It says "low-dose sedating antidepressants" (grade B) can be used short-term, for 4 weeks or less, when CBT-I is not effective enough, and that longer use may be started "in some cases" after weighing pros and cons (Riemann et al., J Sleep Res 2023). The abstract does not single out mirtazapine, and the Cochrane review behind this class of advice had no mirtazapine trials to analyse (Everitt et al., Cochrane 2018).
- Older adults: the AGS Beers Criteria 2023 list mirtazapine among drugs to "use with caution" because it may cause or worsen SIADH (a hormone problem) or low sodium. It advises monitoring sodium closely when starting or changing the dose (AGS Beers Criteria 2023).
What works and what does not
| Use | Verdict | Evidence | Key caveat |
|---|---|---|---|
| Acute major depression in adults | WORKS | OR 1.89 vs placebo for response; moderate-certainty evidence (Cipriani 2018). | Average benefit over placebo is modest for all antidepressants; evidence is for acute treatment. |
| Preventing relapse after response | WORKS | In a relapse-prevention trial, people who kept taking Remeron had significantly lower relapse rates over 40 weeks than those switched to placebo (FDA label). | A registration trial; exact relapse percentages are not given in the label. |
| Faster early response than SSRIs | MIXED | OR 1.57 at 2 weeks vs SSRIs (Watanabe 2011). | Trial base almost entirely manufacturer-linked; Cochrane authors flag possible sponsorship bias and low comparator doses. |
| Add-on for treatment-resistant depression (with SSRI/SNRI) | MIXED | NIHR MIR trial: no clinically important benefit at 12 weeks; more adverse effects (Kessler 2018). NICE still lists it as an option (NG222). | The best independent trial was negative on its main measure. |
| Insomnia that comes with depression | WORKS | Increased total sleep time, slow-wave sleep and sleep efficiency (Zhang 2026). US registration trials also showed superiority on the HDRS sleep-disturbance factor (FDA label). | Sedation is a side effect as well as a benefit; next-day drowsiness is common. |
| Chronic insomnia without depression (off-label) | INSUFFICIENT EVIDENCE | One small 28-day RCT in adults 65+ showed a benefit (MIRAGE 2025); no mirtazapine trials were analysed in the Cochrane review (Everitt 2018). | No long-term data; CBT-I remains first-line (European guideline). |
| Sleep problems in Alzheimer's disease | INSUFFICIENT EVIDENCE | Pilot RCT (n = 24): no improvement in night-time sleep and more daytime sleepiness (Scoralick 2017). | Very small study, but the signal points towards harm, not benefit. |
| Depression in Alzheimer's disease | NOT SUPPORTED | No benefit over placebo; more adverse reactions (41% vs 26%) (HTA-SADD 2011). | Independent NIHR-funded trial. |
| Agitation in dementia | NOT SUPPORTED | No benefit; possible higher mortality (SYMBAD 2021). | Authors conclude the data "do not support" this use. |
| Children and adolescents with depression | NOT SUPPORTED | Two placebo-controlled trials in 258 young people were insufficient to establish effectiveness (FDA label). | 49% of young people on mirtazapine gained at least 7% of body weight in 8 weeks. |
Benefits by claim
1. Depression: how much better than placebo?
The Cipriani 2018 network meta-analysis is the most comprehensive independent comparison of antidepressants. It covered 522 double-blind trials and 116,477 participants. All 21 antidepressants beat placebo, with response odds ratios ranging from 2.13 (amitriptyline) to 1.37 (reboxetine). Mirtazapine's odds ratio was 1.89 (95% credible interval 1.64–2.20), the second-highest point estimate among the 21 drugs. Its odds ratio for all-cause dropout was 0.99 (0.85–1.15), meaning people stopped treatment about as often as on placebo (Cipriani et al., Lancet 2018, figure 3).
In head-to-head trials, mirtazapine was one of seven antidepressants that were more effective than other antidepressants (odds ratios between 1.19 and 1.96). The certainty of evidence for mirtazapine comparisons was rated moderate for most comparisons. The authors concluded that escitalopram, mirtazapine, paroxetine, agomelatine and sertraline had "a relatively higher response and lower dropout rate than the other antidepressants" (Cipriani 2018).
What that means in practice. An odds ratio compares the odds of responding, not the percentage of people who respond. The same paper gives a pooled standardised mean difference of 0.30 across all antidepressants, and describes the summary effects as "mostly modest" (Cipriani 2018). Many people improve on placebo in depression trials, so the extra benefit a drug adds is smaller than the total improvement people notice. Differences between individual antidepressants were smaller once placebo-controlled trials were included, and the credible intervals on most comparisons were wide. Mirtazapine ranking near the top is an average across trials, not a guarantee for any one person.
Funding context. The authors report that industry funding "was not associated with substantial differences" in response or dropout. They also note that non-industry trials were few, many trials did not report their funding, and drugs tended to look better when they were new (a "novelty effect"). Unpublished data were obtained for some drugs but not all (Cipriani 2018).
2. Depression: compared with SSRIs
The Cochrane review of 29 randomised trials (4,974 participants), mostly lasting six weeks, found:
- vs SSRIs (12 trials, n = 2,626): mirtazapine was more effective at 2 weeks (OR 1.57, 95% CI 1.30–1.88) and at the end of acute treatment (OR 1.19, 1.01–1.39);
- vs venlafaxine (2 trials, n = 415): OR 2.29 at 2 weeks and OR 1.53 at the end of treatment;
- vs tricyclics (10 trials): no robust difference;
- dropouts: no robust difference from other antidepressants (Watanabe et al., Cochrane 2011).
The authors say mirtazapine "is likely to have a faster onset of action than SSRIs". They add a significant caution: all but two included trials were funded by, or run with the advice of, a mirtazapine manufacturer, and comparator doses sometimes "seemed lower than the usual doses in clinical practice". They call for future trials "funded by non-profit organizations" (Watanabe 2011). Pure City Research therefore rates the faster-onset claim as plausible but not independently confirmed.
3. Add-on for depression that has not responded
Combining mirtazapine with an SSRI or SNRI is a common strategy, and NICE lists it as an option (NICE NG222). The MIR trial tested it directly in 106 UK general practices. It randomised 480 adults who were still depressed after at least six weeks on an SSRI or SNRI to add either mirtazapine or placebo. At 12 weeks, Beck Depression Inventory II scores were 18.0 on mirtazapine and 19.7 on placebo. The adjusted difference was −1.83 (95% CI −3.92 to 0.27; P = 0.09). Adverse effects were more common with mirtazapine and led people to stop the trial drug. The authors concluded the study "did not find evidence of a clinically important benefit" (Kessler et al., BMJ 2018). (MIR was published in the BMJ, not the Lancet Psychiatry.)
4. Sleep when you are depressed
Mirtazapine is often chosen for depressed people who sleep badly. In the US registration trials it outperformed placebo on the sleep-disturbance factor of the Hamilton scale (FDA label). A 2026 meta-analysis of 30 studies found mirtazapine increased total sleep time, slow-wave sleep and sleep efficiency and reduced wake time after sleep onset on polysomnography, with weight gain and sedation as the common drawbacks (Zhang et al., J Affect Disord 2026). Against SSRIs, the Cochrane review found mirtazapine was less likely to cause sleep disturbance (OR 0.52) but more likely to cause somnolence (OR 1.81) (Watanabe 2011).
5. Off-label use for insomnia: what the evidence actually shows
Mirtazapine is widely prescribed at low doses for sleep, but that practice has run well ahead of the evidence:
- The 2018 Cochrane review of antidepressants for insomnia included 23 trials (2,806 participants). Its analyses covered SSRIs, doxepin, trimipramine, mianserin and trazodone, with no mirtazapine trials. Esmirtazapine studies found in later searches were still "awaiting classification". The review found "no evidence ... for long-term antidepressant use for insomnia" (Everitt et al., Cochrane 2018).
- The MIRAGE trial (2025) describes itself as the first randomised, double-blind, placebo-controlled trial of mirtazapine for chronic insomnia in older adults. Sixty people aged 65 and over took 7.5 mg or placebo for 28 days. Insomnia Severity Index scores fell by 6.5 points on mirtazapine and 2.9 on placebo (p = 0.003), and wake time, total sleep time and sleep efficiency also improved. There were no severe adverse events, but 6 people on mirtazapine and 1 on placebo stopped because of side effects. The authors say use "may be limited by mild but clinically relevant adverse events" (Nguyen et al., Age Ageing 2025). The same journal ran an accompanying commentary titled "weighing the risks and potential benefits" (Inglis & Mangoni, Age Ageing 2025).
- A small pilot trial in Alzheimer's disease (15 mg, 2 weeks) found no improvement in night-time sleep and more daytime sleepiness (Scoralick et al., Psychogeriatrics 2017).
- Organon, and later Merck, developed esmirtazapine, a single-enantiomer relative of mirtazapine, specifically for insomnia. A company-funded 6-month trial reported 48.7 extra minutes of self-reported sleep compared with placebo (Ivgy-May et al., J Clin Sleep Med 2020). Merck then ended the programme "for strategic reasons" (Merck 2009 Form 10-K), and the drug was never marketed. These results cannot simply be applied to mirtazapine.
The UK product information also undercuts a popular idea about "sleep doses": it notes that "dose reduction generally does not lead to less somnolence/sedation but can jeopardize antidepressant efficacy" (UK SmPC, section 4.8). For insomnia without depression, the European guideline's first-line treatment remains CBT-I (Riemann et al., 2023).
6. Dementia: two independent trials, no benefit
Mirtazapine is often prescribed to older people with dementia. Two NIHR-funded trials tested it:
- HTA-SADD (depression in Alzheimer's disease): depression score changes at 13 weeks did not differ between mirtazapine and placebo (mean difference 0.01; p = 0.99), and this persisted to 39 weeks. Adverse reactions occurred in 41% on mirtazapine vs 26% on placebo (Banerjee et al., Lancet 2011).
- SYMBAD (agitation in dementia): 204 participants across 26 UK centres, with mirtazapine titrated to 45 mg. There was no difference at 12 weeks (adjusted mean difference −1.74; p = 0.53). There were 7 deaths on mirtazapine and 1 on placebo by week 16; in a post hoc analysis this difference was of marginal statistical significance (p = 0.065). The authors concluded the data "do not support using mirtazapine as a treatment for agitation in dementia" (Banerjee et al., Lancet 2021).
Risks and all side effects
FDA boxed warning
The Remeron label carries the class-wide boxed warning "Suicidal thoughts and behaviors". Antidepressants increased the risk of suicidal thoughts and behaviours in paediatric and young adult patients in short-term studies. All antidepressant-treated patients should be closely monitored for clinical worsening and new suicidal thoughts or behaviours, and Remeron is not approved for paediatric use. In pooled trials, the drug-placebo difference per 1,000 patients was 14 additional cases in under-18s and 5 additional cases in 18–24-year-olds. There was 1 fewer case in 25–64-year-olds and 6 fewer in people aged 65 and over (FDA label). UK product information advises close supervision early in treatment and after dose changes, and notes that the suicide risk "may increase in the early stages of recovery" (UK SmPC).
Side effects
| Side effect / concern | Frequency | What the source says | Practical note | Source |
|---|---|---|---|---|
| Somnolence / sedation | 54% vs 18% placebo | Led 10.4% to stop (vs 2.2%). Unclear whether tolerance develops. Lowering the dose generally does not reduce sedation. | Do not drive or use machinery until you know how it affects you. | FDA, SmPC |
| Increased appetite | 17% vs 2% | Very common in UK product information. | Discuss diet and weight monitoring early. | FDA |
| Weight gain | 12% vs 2% reported; ≥7% body-weight gain in 7.5% vs 0% | 8% stopped because of weight gain in pooled premarketing studies. Cochrane: OR 4.23 vs SSRIs. | More relevant with diabetes, obesity or raised lipids. | FDA, Cochrane |
| Dry mouth | 25% vs 15% | More common than with SSRIs (OR 1.80). | Usually mild. | FDA, Cochrane |
| Constipation | 13% vs 7% | Listed in the US trial table. | Fluids and fibre; tell a pharmacist if persistent. | FDA |
| Dizziness; orthostatic hypotension | Dizziness 7% vs 3% | Significant orthostatic hypotension was seen in early volunteer studies. | Get up slowly; falls risk in older adults. | FDA |
| Raised cholesterol / triglycerides | Cholesterol ≥20% above the upper limit of normal: 15% vs 7%; triglycerides ≥500 mg/dL: 6% vs 3% | Non-fasting values in US controlled studies. | Relevant for cardiovascular risk. | FDA |
| Other listed effects | Common to rare | Abnormal dreams, nightmares, oedema, fatigue, lethargy, tremor, restless legs, paraesthesia, arthralgia, raised prolactin, sleepwalking, amnesia and others. | Report anything new or distressing. | SmPC |
| Advantages vs SSRIs | Comparative | Less nausea/vomiting (OR 0.33), sexual dysfunction (OR 0.31), sweating, diarrhoea, headache and tremor. | A reason it may suit people who could not tolerate an SSRI. | Cochrane |
| Agranulocytosis / bone marrow depression | Rare: 2 of 2,796 in premarketing trials, plus 1 severe neutropenia | Mostly reversible, but fatal cases have been reported, mostly in people over 65. | Fever, sore throat, mouth ulcers or other signs of infection: contact a doctor urgently for a blood count. | FDA, SmPC, NHS |
| Serotonin syndrome | Very rare alone; higher risk in combinations | Risk rises with other serotonergic drugs, opioids including buprenorphine, and MAOIs (contraindicated). | Agitation, fever, sweating, stiffness or jerking: seek urgent care. | FDA, SmPC |
| Severe skin reactions (DRESS, SJS, TEN) | Rare / post-marketing | Can be life-threatening. Stop immediately and never restart. | Rash with fever, swollen glands or peeling skin: urgent care. | FDA, SmPC |
| Hyponatraemia (low sodium), SIADH | Very rare in UK product information | Sodium below 110 mmol/L has been reported. Older adults, people on diuretics and people who are volume-depleted are at higher risk. | Confusion, headache, unsteadiness or falls in an older person: check sodium. | FDA, Beers 2023 |
| QT prolongation / torsades de pointes | Post-marketing reports | Mostly in overdose or with other risk factors, such as other QT-prolonging drugs. | Caution with heart disease or a family history of long QT. | FDA |
| Mania / hypomania | 0.2% in trials | Screen for bipolar disorder before starting. | Unusually high mood or energy: contact your prescriber. | FDA |
| Liver enzyme rises; jaundice | ALT ≥3× upper limit: 2.0% vs 0.3% | Stop if jaundice occurs. | Yellow eyes or skin: seek advice. | FDA, SmPC |
| Seizures; angle-closure glaucoma | 1 seizure in 2,796 trial patients | Caution with epilepsy; pupil dilation can trigger angle closure in eyes with narrow angles. | Sudden eye pain or blurred vision: urgent care. | FDA |
| Discontinuation (withdrawal) symptoms | Reported, particularly after abrupt stopping | Dizziness, abnormal dreams, electric-shock sensations, agitation, anxiety, headache, nausea, sweating. The UK SmPC says most withdrawal reactions are mild and self-limiting. | Reduce gradually with your prescriber. | FDA, SmPC |
| Overdose | — | Disorientation, drowsiness, memory impairment and fast heart rate. Fatalities have been reported, especially with mixed overdoses, and there is no specific antidote. | In the UK call NHS 111, or 999 in an emergency; in the US call Poison Control. | FDA, NHS |
Dependence. UK product information states that mirtazapine "is not addictive". Withdrawal symptoms after abrupt stopping are a separate issue from addiction, and NICE says antidepressant withdrawal can take "weeks or months" to complete in some people (UK SmPC, NICE NG222). Sexual side effects are less common than with SSRIs in trials. The Cochrane authors cite a primary-care survey reporting sexual dysfunction in 41% of mirtazapine users, similar to paroxetine, so the advantage may be smaller in everyday practice than in trials (Watanabe 2011).
All interactions
| Interacting drug / substance | Examples | Risk level | What happens | What the label advises |
|---|---|---|---|---|
| MAO inhibitors | Phenelzine, tranylcypromine, isocarboxazid, selegiline, linezolid, IV methylene blue | Contraindicated | Serotonin syndrome | Allow at least 14 days between stopping one and starting the other. |
| Other serotonergic drugs | SSRIs, SNRIs, triptans, tricyclics, lithium, tramadol, fentanyl, tryptophan, buspirone, amphetamines | High caution | Serotonin syndrome | Monitor, especially when starting or raising doses. |
| St John's wort | Herbal Hypericum perforatum products | Avoid | Serotonergic effects | NHS: do not use with mirtazapine. See our St John's wort review. |
| Opioids including buprenorphine | Morphine, buprenorphine, codeine-type medicines | High caution | Added sedation; serotonin syndrome reported with buprenorphine/opioids | Careful observation, especially at initiation and dose increases. |
| Alcohol | Any alcoholic drink | Avoid | Increased CNS depression and impairment | FDA and UK labels advise avoiding alcohol. |
| Benzodiazepines and other sedatives | Diazepam, alprazolam, Z-drugs, sedating antihistamines, most antipsychotics | Avoid / high caution | Worse impairment of thinking and movement | FDA: avoid combining with benzodiazepines. |
| Strong CYP3A inducers | Carbamazepine, phenytoin, rifampicin | Moderate | Carbamazepine and phenytoin cut mirtazapine levels by 60% and 45% | The dose may need increasing, then reducing if the inducer stops. |
| Strong CYP3A inhibitors | Ketoconazole, itraconazole, ritonavir and other HIV protease inhibitors, clarithromycin, erythromycin, nefazodone | Moderate | Ketoconazole raised peak levels by about 40% and exposure (AUC) by about 50% | The dose may need reducing. |
| Cimetidine | — | Moderate | Mirtazapine levels may rise by more than 50% | The dose may need reducing. |
| Warfarin | — | Monitor | Small but significant INR rise at 30 mg; a larger effect at higher doses cannot be excluded | Monitor INR. |
| QT-prolonging medicines | Some antipsychotics and antibiotics | Caution | Higher risk of QT prolongation and arrhythmia | Use with caution. |
| NSAID painkillers | Ibuprofen | Check with pharmacist | Listed by the NHS as a medicine that may not mix well | Ask a pharmacist before combining. |
| Diabetes medicines | Insulin, oral hypoglycaemics | Monitor | Antidepressants may alter glucose control | Doses may need adjusting. |
Interaction sources: FDA Remeron label, section 7, UK SmPC, sections 4.4–4.5, NHS. No interaction list is complete; always tell your pharmacist about every medicine, supplement and herbal product you take.
Who should avoid mirtazapine
- Must not take: anyone allergic to mirtazapine or its ingredients, and anyone taking an MAOI or within 14 days of stopping one (FDA label, UK SmPC). Anyone who has had DRESS, Stevens-Johnson syndrome or toxic epidermal necrolysis on mirtazapine must never restart it (UK SmPC).
- Under 18s: not recommended. Efficacy was not shown in two trials, and young people on it gained weight markedly (UK SmPC, FDA label).
- Needs extra caution and monitoring: liver, kidney or heart disease; epilepsy; diabetes; schizophrenia or other psychosis; bipolar disorder; low blood pressure; prostate enlargement or difficulty passing urine; narrow-angle glaucoma; and a personal or family history of QT prolongation (NHS, UK SmPC). The orally disintegrating tablets contain phenylalanine, which matters for people with phenylketonuria (FDA label). The oral solution contains maltitol, so people with hereditary fructose intolerance should not take it (UK SmPC).
- Older adults: clearance is lower. Sedating drugs can cause confusion and over-sedation, and older adults are at greater risk of low sodium (FDA label). The Beers Criteria say "use with caution" and monitor sodium (AGS Beers 2023). In people with dementia, independent trials found no benefit for depression or agitation (HTA-SADD, SYMBAD).
- Liver and kidney impairment: clearance falls, and blood levels rose by about 55% in mild-to-moderate liver impairment and by up to 115% in severe kidney impairment, so lower doses may be needed (UK SmPC).
- Pregnancy: the NHS says mirtazapine can be used in pregnancy if needed, after discussing risks and benefits, and you may be advised to give birth in hospital so your baby can be monitored (NHS). The FDA label says published data "has not reliably identified a drug-associated risk" of major birth defects or miscarriage, and notes the risks of untreated depression (FDA label). The UK SmPC says a risk of persistent pulmonary hypertension of the newborn "cannot be ruled out" (UK SmPC). Do not stop suddenly if you find you are pregnant; speak to your prescriber.
- Breastfeeding: mirtazapine passes into breast milk only in very small amounts. The FDA label reports relative infant doses of 0.6–2.8% of the maternal weight-adjusted dose, with no adverse effects reported in the infants of a pooled group of 8 mother-infant pairs (NHS, FDA label).
Dosage and how to take it
Your prescriber sets your dose. The ranges below are the official licensed adult ranges for depression, not advice for any individual.
| Item | UK (SmPC) | US (FDA label) |
|---|---|---|
| Licensed adult dose range | Effective dose usually 15–45 mg a day; starting dose 15 mg or 30 mg | Start at 15 mg once daily; maximum 45 mg a day |
| Timing | Preferably a single dose at night before bed; may be split, with the larger dose at night | Once daily, preferably in the evening before sleep |
| Dose changes | If response is insufficient the dose can be raised to the maximum; if still no response after a further 2–4 weeks, stop | No sooner than every 1–2 weeks |
| Forms | Tablets, orodispersible tablets, 15 mg/ml oral solution | 15 mg and 30 mg tablets; 15, 30 and 45 mg orally disintegrating tablets |
| Older adults / organ impairment | Same dose range; increase under close supervision; take lower clearance into account | Start at the low end; a lower dose may be needed in moderate-to-severe kidney or liver impairment |
Sources: UK SmPC, section 4.2; FDA label, section 2. Low doses used off-label for sleep (for example the 7.5 mg dose tested in the MIRAGE trial) are outside the licence and are a decision for the prescriber.
How long it takes to work
The NHS says mirtazapine usually starts working after 1–2 weeks, and you should tell your doctor if you do not feel better after 2–4 weeks. People usually take it for at least 4–6 months (NHS). The UK SmPC recommends treating for at least 6 months so that people are free of symptoms (UK SmPC). The orodispersible tablet should be placed on the tongue straight after it is removed from the blister, and should not be split or crushed (FDA label).
Missed doses
The NHS advice depends on how you take it. If you take it once a day, skip the missed dose and take the next one at the usual time; never take two doses to make up. If you take two doses a day and forget the morning dose, take it with the evening dose; if you miss the evening dose, skip it (NHS).
How to stop
Do not stop suddenly. Your doctor will reduce the dose gradually "over a few weeks or months" (NHS). NICE advises reducing step by step, with each step a proportion of the previous dose (for example 50%), and smaller steps (for example 25%) at lower doses. It suggests liquid preparations when very small doses are needed, and leaving 1–2 weeks to judge each reduction. The pace should be agreed with the person taking the medicine (NICE NG222). If withdrawal symptoms are severe, NICE suggests going back to the previous dose and then reducing more slowly (NICE NG222).
Follow the money: who makes it and who funded the evidence
Ownership chain
- Originator: Organon (Netherlands). Organon was the pharmaceutical business of Akzo Nobel N.V., with a research and manufacturing facility in Oss, the Netherlands, and neuroscience research in Newhouse, Scotland (Schering-Plough press release, March 2007). Organon held the original US approval for Remeron in 1996 (Drugs@FDA).
- 2007: Schering-Plough (US). Schering-Plough agreed to buy Organon BioSciences from Akzo Nobel for about €11 billion ($14.4 billion). Esmirtazapine for insomnia was listed as one of five Phase III compounds in the deal (Schering-Plough, 2007).
- 2009: Merck & Co. (US). Merck and Schering-Plough completed their merger on 3 November 2009. Merck's 2009 annual report lists Remeron (mirtazapine) among its neuroscience products and records that Merck "terminated the internal clinical development program for esmirtazapine ... for hot flashes and insomnia for strategic reasons" (Merck 2009 Form 10-K).
- 2021: Organon & Co. (US). Merck spun off a new company, Organon & Co., effective 2 June 2021. It is headquartered in Jersey City, New Jersey (Organon 2025 Form 10-K). The current US Remeron label names Organon LLC, a subsidiary of Organon & Co. (FDA label).
- Generics: mirtazapine has long been generic. US generic application holders include Aurobindo (India), Mylan and Sun Pharmaceutical (openFDA). In the UK, many generic products are available, for example a UK-licensed oral solution held by Rosemont Pharmaceuticals, Leeds (UK SmPC).
Revenue
Merck reported worldwide Remeron sales of $232 million in 2012, $206 million in 2013 and $193 million in 2014 (Merck 2014 Form 10-K). Organon's 2025 annual report does not name Remeron or break out its sales. It reports $6.2 billion in total 2025 revenue, including $3,691 million from "established brands", many of which "lost exclusivity years ago" (Organon 2025 Form 10-K). Remeron-specific revenue since 2014 is therefore not documented in the sources we reviewed.
Who funded the evidence
- Registration trials (manufacturer-funded): the FDA approved Remeron on the basis of four 6-week placebo-controlled trials and a relapse-prevention study, as described in the label (FDA label).
- Comparison trials (mostly manufacturer-linked): in the Cochrane review, all but two of the included trials (Amini 2005 and Fava 2006, the latter funded by the US National Institute of Mental Health) were funded by or run with the advice of a mirtazapine manufacturer. The review team also asked Organon for unpublished data (Watanabe 2011). Two review authors disclosed past funding from Schering-Plough, which owned Organon from 2007 (Watanabe 2011, Schering-Plough 2007).
- Independent syntheses and trials (public funders): Cipriani 2018 (NIHR and the Japan Society for the Promotion of Science); MIR, HTA-SADD and SYMBAD (NIHR HTA programme); MIRAGE (academic hospital sponsor). These are the sources that tested the drug in the settings where it is most often used off-label or as an add-on, and none of them found a clear benefit except the small MIRAGE insomnia trial.
- Insomnia development (manufacturer-funded): the esmirtazapine insomnia trials were funded by Organon (under Akzo Nobel, then Merck), with company-employed authors and medical-writing support paid by Merck Sharp & Dohme (Ivgy-May 2020).
What this means. The core finding that mirtazapine beats placebo for depression holds up in the independently funded Cipriani analysis, but that analysis pools mainly industry-funded trials. Claims that it is better than other antidepressants rest heavily on manufacturer-linked head-to-head trials. The publicly funded trials of popular extended uses (add-on therapy, dementia) were negative or showed possible harm. We found no documented research-integrity event specific to mirtazapine in the sources reviewed.
Related research
- Stress, anxiety and depression: prevention and management guide
- Supplements for stress, anxiety and depression: the evidence
- Sleep: prevention and management guide
- Sleep supplements: the evidence
- St John's wort (dangerous combination with mirtazapine)
- Melatonin
- Saffron for depression
- Related medicines: sertraline, escitalopram, venlafaxine, trazodone, amitriptyline, zolpidem, daridorexant
Frequently asked questions
How long does mirtazapine take to work?
The NHS says it usually starts working after 1–2 weeks. Tell your doctor if you are not starting to feel better after 2–4 weeks (NHS). A Cochrane review found higher response rates than SSRIs at 2 weeks, although that evidence is mostly manufacturer-funded (Watanabe 2011).
Does mirtazapine cause weight gain?
Often, yes. In US trials, 17% reported increased appetite (vs 2% on placebo) and 7.5% gained at least 7% of their body weight in 6 weeks (vs 0%). In longer premarketing studies, 8% stopped because of weight gain (FDA label). Weight gain or increased appetite was about four times as likely (OR 4.23) as with SSRIs (Watanabe 2011). If weight gain worries you, raise it with your prescriber rather than stopping suddenly.
Can you drink alcohol on mirtazapine?
Both the FDA and the UK product information advise avoiding alcohol, because it adds to mirtazapine's effects on thinking and co-ordination. The NHS says it is best not to drink while taking it because you may feel very drowsy (FDA label, UK SmPC, NHS).
Is mirtazapine a good sleeping pill?
It is not licensed as one. It clearly causes sleepiness and improves sleep in people with depression (Zhang 2026). For insomnia without depression, only one small, 28-day placebo-controlled trial exists, in adults aged 65 and over, and more people stopped because of side effects on mirtazapine than on placebo (MIRAGE 2025). The European guideline recommends CBT for insomnia first and limits sedating antidepressants to short-term use in most cases (Riemann 2023).
Does a low dose make you sleepier than a high dose?
This is widely repeated, but the sources we reviewed do not support it. UK product information states that dose reduction "generally does not lead to less somnolence/sedation but can jeopardize antidepressant efficacy" (UK SmPC). Any dose change should be made with your prescriber.
How do I stop mirtazapine safely?
Do not stop suddenly. Your prescriber will usually reduce the dose gradually over weeks or months (NHS). NICE advises stepwise reductions (for example 50% of the previous dose, then smaller 25% steps at lower doses), allowing 1–2 weeks between steps, at a pace you agree with your prescriber (NICE NG222). Symptoms reported after abrupt stopping include dizziness, abnormal dreams, electric-shock sensations, anxiety, nausea and sweating (FDA label).
Is mirtazapine better than sertraline or other SSRIs?
It depends on what matters to you. In the Cipriani analysis, mirtazapine and sertraline were both among the antidepressants with relatively higher response and lower dropout (Cipriani 2018). Mirtazapine causes less nausea and sexual dysfunction but more weight gain and sleepiness than SSRIs (Watanabe 2011). NICE suggests SSRIs as the first choice for most people (NICE NG222).
Can I take mirtazapine while pregnant or breastfeeding?
The NHS says it can be used in pregnancy if needed, after a risk-benefit discussion with your doctor. It passes into breast milk only in very small amounts, and side effects in breastfed babies are rare (NHS). Do not stop suddenly if you become pregnant; talk to your prescriber first.
Sources and funding notes
- NHS. Mirtazapine (page last reviewed 18 August 2026): UK public health service; no industry funding.
- FDA-approved prescribing information: Remeron/RemeronSolTab (Organon LLC), via DailyMed: label text submitted by the manufacturer and approved by the US regulator; efficacy data come from manufacturer trials.
- Drugs@FDA: NDA 020415 (Remeron), approved 14 June 1996, sponsor Organon: US regulator record.
- openFDA Drugs@FDA: NDA 021208 (RemeronSolTab) and mirtazapine application holders: US regulator data.
- Summary of Product Characteristics: Mirtazapine 15 mg/ml oral solution (Rosemont Pharmaceuticals, UK; revised 16 January 2025): UK-licensed product information written by the licence holder, a generic manufacturer.
- Cipriani A, et al. Comparative efficacy and acceptability of 21 antidepressant drugs. Lancet 2018: funded by the NIHR Oxford Health Biomedical Research Centre and the Japan Society for the Promotion of Science, with no funder role. Some authors disclosed fees from Eli Lilly, Janssen, MSD, Pfizer, Lundbeck, Otsuka and others. UK/Japan/international.
- Watanabe N, et al. Mirtazapine versus other antidepressive agents for depression. Cochrane Database Syst Rev 2011: academic (Japan/Italy/UK). Lead authors disclosed Japanese government grants and fees from several companies, including Schering-Plough. The authors report that nearly all included trials were manufacturer-linked.
- Kessler DS, et al. Mirtazapine added to SSRIs or SNRIs for treatment resistant depression in primary care (MIR). BMJ 2018: funded by the NIHR HTA programme (11/129/76), with no funder role. One author reported fees from Lundbeck-Otsuka and Takeda outside the work. UK.
- NICE. Depression in adults: treatment and management (NG222), 2022: UK government-funded guideline body.
- Everitt H, et al. Antidepressants for insomnia in adults. Cochrane Database Syst Rev 2018: academic (UK). Some authors disclosed past industry grants or fees, including one who reported past research support from Organon on behalf of an employer.
- Riemann D, et al. The European Insomnia Guideline: 2023 update. J Sleep Res 2023: European Sleep Research Society guideline authors (multinational). Europe PMC lists NIHR grant support; individual author disclosures were not reviewed.
- Nguyen PV, et al. Mirtazapine for chronic insomnia in older adults: the MIRAGE study. Age Ageing 2025: sponsored by Centre hospitalier de l'Université de Montréal with a Quebec research network on ageing (per ClinicalTrials.gov NCT05247697); academic, Canada.
- Inglis JM, Mangoni AA. Mirtazapine for chronic insomnia in older adults: weighing the risks and potential benefits. Age Ageing 2025: academic commentary (Australia); cited for its existence only.
- Zhang X, et al. Insomnia symptoms in depressed patients treated with agomelatine, mirtazapine and trazodone: systematic review and meta-analysis. J Affect Disord 2026: academic (China); authors declared no competing interests.
- Scoralick FM, et al. Mirtazapine does not improve sleep disorders in Alzheimer's disease. Psychogeriatrics 2017: academic (University of Brasília, Brazil); funding not stated in the abstract.
- Banerjee S, et al. Sertraline or mirtazapine for depression in dementia (HTA-SADD). Lancet 2011: funded by the UK NIHR HTA programme.
- Banerjee S, et al. Study of mirtazapine for agitated behaviours in dementia (SYMBAD). Lancet 2021: funded by the UK NIHR HTA programme; several authors disclosed unrelated industry fees.
- Ivgy-May N, et al. Efficacy and safety of esmirtazapine in chronic primary insomnia. J Clin Sleep Med 2020: funded by Organon (Akzo Nobel, then Merck); several authors were employees; writing support funded by Merck Sharp & Dohme. Manufacturer-funded.
- American Geriatrics Society 2023 updated AGS Beers Criteria: professional society (US); "no sponsor for this paper".
- Schering-Plough. Press release: Schering-Plough to acquire Organon BioSciences (SEC filing, March 2007): company statement; used only for corporate facts.
- Merck & Co. Form 10-K for 2009 (SEC): company regulatory filing; merger date, product list, esmirtazapine termination.
- Merck & Co. Form 10-K for 2014 (SEC): company regulatory filing; Remeron sales 2012–2014.
- Organon & Co. Form 10-K for 2025 (SEC): company regulatory filing; spin-off date, headquarters, revenue by product group.
Last reviewed: 26 September 2026. Educational information from Pure City Research, not medical advice. Medicine decisions belong with a qualified prescriber.
Have a question — or want us to cover something?
Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.
One daily research roundup
Get the topics, key findings and links from our new articles in one email. At most one digest a day, only when there is something new.
