Zolpidem (Ambien): Independent Evidence on Insomnia, Side Effects & Withdrawal

Table of contents

• Evidence summary
• What zolpidem 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 zolpidem
• 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

Independent evidence and credibility scorecard

Independence tier: 1 = no money from anyone selling the product; 4 = funded by the seller. Credibility: A (highest) to D.

What zolpidem is

Zolpidem is a sleeping pill (hypnotic) from the "Z-drug" group, alongside zopiclone, eszopiclone and zaleplon. Chemically it is an imidazopyridine, not a benzodiazepine, but it acts at the same receptor site as benzodiazepines (UK SmPC). Brand names include Ambien and Ambien CR (US), Stilnox (France; marketed in over 100 countries) and Myslee (Japan). In the UK it is mostly dispensed as generic zolpidem tartrate. Other US zolpidem products include Edluar, Zolpimist and the low-dose Intermezzo (Sanofi 20-F 2012; FDA 2013).

Zolpidem was developed by the French company Synthélabo, which launched it in France in 1988 (Inpharma, March 1988). The FDA approved Ambien as a new molecular entity on 16 December 1992 under NDA 019908 (Drugs@FDA). Synthélabo merged with Sanofi in 1999, and the group is now Sanofi, headquartered in Paris (Sanofi 20-F 2012; Sanofi 20-F 2025).

Zolpidem is now a cheap generic medicine. Generics have been sold in Europe since 2004 and in the US since 2007 (immediate-release), and US generics of the extended-release form arrived in October 2010 (Sanofi 20-F 2010).

Legal status. In the UK zolpidem is "only available on prescription" (NHS). In the US it is a prescription medicine and a Schedule IV controlled substance under federal law (FDA label). France tightened its rules in April 2017 so that zolpidem must be prescribed on a secure prescription pad, as some narcotics are (Aquizerate et al. 2023).

How it works

Zolpidem boosts GABA, the brain's main calming chemical messenger. It is a positive modulator of the GABA_A receptor and binds to the benzodiazepine site on receptors that contain the α1 subunit. This increases how often the receptor's chloride channel opens, which dampens nerve activity (FDA label).

Benzodiazepines bind several receptor subtypes. Zolpidem binds mainly the "omega-1" subtype. In animals this may explain why it has little of the muscle-relaxing and anticonvulsant effect that benzodiazepines have. All its effects are reversed by the benzodiazepine antagonist flumazenil (UK SmPC). This selectivity has been used to present Z-drugs as a safer alternative to benzodiazepines. In older adults, however, independent reviewers found adverse events similar to those of benzodiazepines (Beers 2023; Treves et al. 2018).

Timing in the body. Standard tablets reach peak blood levels in about 1.6 hours on average, with an average elimination half-life of about 2.5 hours (range 1.4 to 4.5) in healthy adults. Taking the tablet with or just after a meal lowers and delays the peak, so the label advises against doing so for faster sleep onset. Half-life is longer in older people. In liver cirrhosis the mean half-life was 9.9 hours, compared with 2.2 hours in healthy people (FDA label). The NHS says zolpidem takes around 30 minutes to work and "does not stay in your system for more than about 12 hours", although some people still feel sleepy the next morning (NHS).

What it is prescribed for

UK licence. Zolpidem is licensed for "the short-term treatment of insomnia in adults in situations where the insomnia is debilitating or is causing severe distress for the patient" (UK SmPC).

US licence. Ambien (immediate-release) is indicated for short-term treatment of insomnia with difficulty falling asleep. It was shown to shorten sleep latency for up to 35 days in controlled studies (FDA label). Ambien CR (extended-release) is indicated for short-term treatment of difficulty falling asleep and/or staying asleep (FDA Ambien CR label).

Where the guidelines place it:

• NICE (England, TA77, 2004): use hypnotics only after non-drug measures have been considered, and "for short periods of time only". Because there is no compelling evidence to separate zolpidem, zopiclone and short-acting benzodiazepines, prescribe the one with the lowest purchase cost. If one does not work, do not try another (NICE TA77).
• NHS: "GPs now rarely prescribe sleeping pills to treat insomnia". Sleeping pills are prescribed "for a few days, or weeks at the most" when insomnia is very bad and other treatments have not worked. CBT is offered first (NHS insomnia).
• European Insomnia Guideline 2023: CBT-I is first-line for adults of any age (grade A). If CBT-I is not effective enough, benzodiazepine receptor agonists including zolpidem can be used for short-term treatment of 4 weeks or less (A). Longer use "may be initiated in some cases" after weighing the pros and cons (B) (Riemann et al. 2023).
• AASM 2017 (US): "We suggest that clinicians use zolpidem as a treatment for sleep onset and sleep maintenance insomnia (versus no treatment) in adults." This is a WEAK recommendation, based on 12 trials whose overall quality was downgraded to very low (Sateia et al. 2017).
• AGS Beers Criteria 2023 (adults 65+): avoid (moderate-quality evidence, strong recommendation). Also avoid in people with delirium, dementia or a history of falls or fractures (Beers 2023).

In short, zolpidem is a second-line, short-term option in every guideline we reviewed, and is not recommended for most older adults.

What works and what does not

Benefits by claim

1. Falling asleep: how many minutes?

The cleanest estimate comes from Huedo-Medina and colleagues. They analysed the data the manufacturers submitted to the FDA (13 studies, 4,378 participants, covering eszopiclone, zaleplon and zolpidem). Compared with placebo, Z-drugs reduced sleep latency measured in a sleep laboratory by 22 minutes (95% CI 11 to 33). The standardised effect on self-reported sleep latency was small (−0.33). Sleep latency was more likely to be reduced in earlier studies, with larger doses, with longer treatment, in younger and/or female patients, and with zolpidem. The authors concluded that the drug effect and the placebo response were "rather small and of questionable clinical importance", but that the two together produced a reasonably large clinical response (Huedo-Medina et al., BMJ 2012).

The UK product information gives trial-specific figures. In 462 healthy volunteers with transient insomnia, 10 mg reduced the average time to fall asleep by 10 minutes compared with placebo, and 5 mg by 3 minutes. In 114 patients with chronic insomnia, 10 mg cut it by 30 minutes and 5 mg by 15 minutes (UK SmPC). These are figures from individual sponsor trials, and pooled analyses are more reliable.

2. The placebo question

Placebo response in insomnia drug trials is large and shows up even on objective sleep-lab measures. A German meta-analysis of 32 trials (3,969 participants) estimated that 63.56% of the drug response was also achieved in placebo groups (Winkler & Rief, Sleep 2015). In practice, much of the improvement a person notices on zolpidem would probably have happened with a dummy pill and the expectation of help. The drug adds a modest amount on top.

3. Comparison with other drugs

The Lancet network meta-analysis covered 154 double-blind trials and 44,089 participants. For acute treatment it found that benzodiazepines, doxylamine, eszopiclone, lemborexant, seltorexant, zolpidem and zopiclone were all more effective than placebo (SMD range 0.36 to 0.83; high to moderate certainty). Benzodiazepines, eszopiclone, zolpidem and zopiclone were more effective than melatonin, ramelteon and zaleplon. The trade-off was tolerability: zolpidem caused more dropouts due to adverse events than placebo (OR 1.79, 95% CI 1.25 to 2.50; moderate certainty). Along with benzodiazepines, eszopiclone and zopiclone, it was also worse than placebo on the number of people with side effects (De Crescenzo et al., Lancet 2022). If you are weighing melatonin as an alternative, see our melatonin evidence review.

4. Long-term benefit

For long-term treatment, only eszopiclone and lemborexant beat placebo, both with "very low" certainty. Eszopiclone was more effective than zolpidem (SMD 0.60, 95% CI 0.00 to 1.20). Zolpidem had more dropouts due to side effects than placebo (OR 2.00, 95% CI 1.11 to 3.70; very low certainty) (Lancet 2022). One trial in the AASM review followed 91 people for 8 months (AASM 2017), but overall there is little good evidence that zolpidem keeps working safely for months or years.

5. Zolpidem with CBT-I

In a Canadian trial of 160 adults, CBT alone and CBT plus zolpidem 10 mg produced similar response rates after 6 weeks (60% vs 61%). Combined therapy gave a larger increase in sleep time. The best long-term outcome came from starting with the combination and then stopping zolpidem while continuing CBT. At 6-month follow-up, remission was 68% in that group compared with 42% in people who kept taking zolpidem (Morin et al., JAMA 2009). CBT-I on its own improved sleep latency by about 19 minutes and wake time after sleep onset by about 26 minutes in a meta-analysis of 20 trials, and the effects seemed to last (Trauer et al., Ann Intern Med 2015). This is similar to the benefit zolpidem shows over placebo, without the dependence risk. Our sleep prevention guide covers CBT-I techniques and sleep habits in detail.

6. What "clinically meaningful" means here

The AASM task force judged that most well-informed patients would choose zolpidem over no treatment, because of improvements in sleep latency, total sleep time, wake time, efficiency and quality and a "relatively low potential for adverse events". It still rated the evidence very low and the recommendation weak (AASM 2017). In 2023, the geriatrics panel reached the opposite judgement for people aged 65 and over, citing "minimal improvement in sleep latency and duration" alongside falls, fractures, delirium and crashes (Beers 2023). Both bodies read broadly the same evidence. The difference lies in whose risk they weigh most heavily.

Risks and all side effects

Boxed warning: complex sleep behaviours (FDA, 2019)

In April 2019 the FDA reported "rare but serious injuries" with eszopiclone, zaleplon and zolpidem caused by sleep behaviours including sleepwalking, sleep-driving and engaging in other activities while not fully awake. It said these "have also resulted in deaths" (FDA 2019). The Ambien label now carries a boxed warning: these behaviours may occur after the first or any later dose, and "some of these events may result in serious injuries, including death". Other reported behaviours include preparing and eating food, making phone calls and having sex, usually with no memory of the event. They can occur at recommended doses, with or without alcohol or other sedatives. Zolpidem must be stopped immediately if one happens, and it is contraindicated in anyone who has had such an episode (FDA label). The UK SmPC notes that alcohol, other sedatives and doses above the maximum appear to increase the risk (UK SmPC).

Next-morning impairment and driving

The FDA's January 2013 safety announcement was based on driving-simulation and laboratory studies. They showed that zolpidem blood levels above about 50 ng/mL "appear capable of impairing driving to a degree that increases the risk of a motor vehicle accident". About 8 hours after a 10 mg dose, about 15% of women and 3% of men exceeded that level. After extended-release 12.5 mg, about 33% of women and 25% of men did. The FDA warned that patients can be impaired "even if they feel fully awake" (FDA 2013).

Real-world data point the same way. In a Washington State cohort of 409,171 licensed drivers, new users of zolpidem had a higher crash risk than non-users (HR 2.20, 95% CI 1.64 to 2.95). The authors reported that the risk estimates across the three sedatives studied were equivalent to blood alcohol levels of 0.06% to 0.11% (Hansen et al., AJPH 2015). This is observational evidence, so people with insomnia may differ in other ways from people without it.

The dispute over sex differences is worth knowing about. A 2019 reanalysis confirmed that women clear zolpidem about 35% more slowly than men. It found that active drug was not distinguishable from placebo at 8 hours in laboratory studies, and that on-road driving studies showed no impairment at 8 hours after 10 mg in either sex. The authors argued the women's dose reduction "is not supported by available scientific evidence" and noted that no other regulator had copied it (Greenblatt et al. 2019). The FDA label still recommends 5 mg as the starting dose for women (FDA label). Both sides agree that the 8-hour window matters: the risk rises sharply if zolpidem is taken with less than a full night (7–8 hours) left, at higher doses, or with alcohol or other sedatives.

Dependence, tolerance and withdrawal

Both labels warn that zolpidem can cause physical and psychological dependence, and that the risk increases with dose and duration and is higher in people with a history of alcohol or drug misuse (FDA label; UK SmPC). Reported withdrawal symptoms include rebound insomnia, anxiety, tremor, sweating, agitation, confusion, headache, palpitations, nightmares, hallucinations and panic attacks. In severe cases they include hypersensitivity to light, noise and touch, and in very rare cases seizures. Symptoms can appear between doses with short-acting drugs (UK SmPC). In France, rising abuse and forged prescriptions led regulators to impose secure prescription pads in 2017. Reimbursed zolpidem use then fell by about 57%, but the share of dependence reports in people for whom zolpidem was the main substance rose from 43.6% to 59.3% (Aquizerate et al. 2023).

Falls, fractures and older adults

The evidence on falls and fractures is consistent across studies, but it is all observational:

• A meta-analysis of 14 studies found Z-drugs associated with fractures (OR 1.63, 95% CI 1.42 to 1.87, with high heterogeneity) and zolpidem associated with injuries (OR 2.05, 95% CI 1.95 to 2.15). The increase in falls (OR 2.40) was not statistically significant (Treves et al., Age Ageing 2018).
• In 15,528 US nursing home residents with hip fracture, Z-drug use was linked to a higher risk (OR 1.66), and the risk was highest in new users (OR 2.20) (Berry et al., JAMA Intern Med 2013).
• In people with dementia in England, new Z-drug use was associated with more fractures (HR 1.40) and hip fractures (HR 1.59), with risk rising with cumulative dose. The mortality association (HR 1.34) was judged "unlikely to be causal" (Richardson et al., NIHR HTA 2021).
• The FDA label reports that in non-US trials, 30 of 1,959 zolpidem patients had falls. Of these, 28 were aged 70 or over, and 23 of those 28 were taking doses above 10 mg (FDA label).

Mortality: an association, not a proof

A US matched cohort found that people prescribed any hypnotic, including zolpidem, had more than three times the hazard of death, even at fewer than 18 pills a year (Kripke et al., BMJ Open 2012). This study cannot prove cause and effect. People who are prescribed sleeping pills often have other health problems, and the lead author declared a long-standing public position against hypnotics. The UK dementia study also judged its mortality finding unlikely to be causal (Richardson et al. 2021). We grade the mortality signal Insufficient for causation.

Full side-effect table

Suspected side effects can be reported through the MHRA Yellow Card scheme in the UK (NHS) or through FDA MedWatch in the US (FDA).

All interactions

Who should avoid zolpidem

• Absolute contraindications (UK): allergy to zolpidem, obstructive sleep apnoea, myasthenia gravis, severe liver insufficiency, and acute or severe respiratory depression. It should not be prescribed to children or to people with psychotic illness (UK SmPC).
• Anyone who has had a complex sleep behaviour on zolpidem, eszopiclone or zaleplon (FDA 2019).
• People with a history of alcohol or drug misuse, or mental health problems: tell the doctor, because these raise dependence risk and the need for caution (NHS; UK SmPC).
• People with depression: sedative-hypnotics can unmask or worsen depression, and intentional overdose is more common in this group (FDA label). If insomnia is part of depression or anxiety, see our stress, anxiety and depression guide.
• Older adults (65+): the Beers Criteria say avoid, with a strong recommendation, and also avoid in delirium, dementia, or a history of falls or fractures (Beers 2023). Where it is still used, UK and US labels set 5 mg as the dose.
• Liver disease: start at 5 mg for mild to moderate impairment; do not use in severe impairment. Kidney disease: the US label found no significant change in kinetics and no adjustment needed. The NHS notes a doctor may start at 5 mg in kidney problems (FDA label; NHS).
• Pregnancy: not usually recommended. Use near the end of pregnancy can cause drowsiness, floppiness, feeding problems, breathing depression or withdrawal in the newborn. Published data have not shown a clear link with major birth defects (NHS; FDA label).
• Breastfeeding: passes into milk in very small amounts. The NHS says occasional short-term doses may be acceptable if the baby is healthy. Do not share a bed with your baby while taking it (NHS).
• People who must be fully alert early the next morning (for example, professional drivers or shift workers): the FDA says extended-release zolpidem "may not be the right medication choice" for anyone who needs to drive the next morning (FDA 2013).

Dosage and how to take it

Your prescriber sets the dose. The ranges below are the official licensed ranges for adults, given for information only.

How long it takes to work: around 30 minutes (NHS). Take it in bed or just before bed, never earlier in the evening. The US label says to re-evaluate if insomnia continues after 7–10 days, and the UK SmPC says after a 7–14-day course, because persistent insomnia can signal another medical or psychiatric problem (FDA label; UK SmPC).

Missed dose: skip it; never take two doses together or take a second dose in the same night (NHS).

How to stop: the UK product information asks prescribers to agree a plan for ending treatment before starting, to reduce the risk of dependence (UK SmPC). If zolpidem has been taken for less than 4 weeks, stopping is unlikely to cause problems. After longer use, do not stop suddenly. A doctor may reduce the dose gradually over days or weeks to prevent withdrawal and rebound insomnia (NHS). Starting CBT-I while tapering is supported by trial evidence (Morin et al. 2009).

Follow the money: who makes it and who funded the evidence

Who made it and who sells it

• Originator: Synthélabo (France), which launched zolpidem in France in 1988 (Inpharma 1988). Synthélabo merged with Sanofi in 1999 to form Sanofi-Synthélabo, which combined with Aventis in 2004. Today's company, Sanofi, is registered at 46 avenue de la Grande Armée, Paris (Sanofi 20-F 2012; SEC EDGAR, Sanofi 20-F 2025).
• Revenue: Sanofi reported Stilnox/Ambien/Myslee sales of €822 million (2008), €873 million (2009) and €819 million (2010). In 2010, €375 million of this came from Ambien CR in the US. After US generics of Ambien CR arrived in October 2010, sales fell to €490 million (2011) and €497 million (2012). By then Japan (Myslee, co-promoted with Astellas) was a major market (Sanofi 20-F 2010; Sanofi 20-F 2012).
• US brand today: on 10 July 2024, Cosette Pharmaceuticals (Bridgewater, New Jersey; backed by the private equity firm Avista Capital Partners) completed its purchase of the US rights to Ambien and Ambien CR from Sanofi US. It cited IQVIA figures of US$39 million in US sales for the 12 months to April 2024 (Cosette press release). Drugs@FDA now lists Cosette as the sponsor of NDA 019908 (Drugs@FDA), although a Sanofi-Aventis U.S. label was still listed on DailyMed when we checked (DailyMed).
• Generics: zolpidem is off-patent almost everywhere. openFDA lists 12 US generic (ANDA) applications. In the UK, generic licences include Zentiva Pharma UK (licensed 2002) and Novumgen (orodispersible Zalzo, 2024) (UK SmPC). Because the product is cheap and generic, no company today has a large commercial stake in promoting it. Most of the money was made between 1992 and 2010.

Who funded the evidence

• The approval trials were run by the manufacturer. The label's efficacy studies lasted 4–5 weeks in total (FDA label). The AASM guideline downgraded its evidence for zolpidem and other hypnotics partly because "very few clinical trials with adequate sample size have been sponsored by agencies outside of industry", which creates a risk of publication bias (AASM 2017).
• The independent reanalyses are the most useful checks. The BMJ 2012 analysis had no funding and no conflicts, and it used FDA-held data, including results that may never have been published (BMJ 2012). The Lancet 2022 network meta-analysis was paid for with UK public money. Its declared author ties are to other companies (Boehringer Ingelheim, Angelini, Janssen), not to zolpidem's makers (Lancet 2022).
• The safety research is mostly public. Funders include the US National Institute on Aging (Berry 2013), the UK NIHR (Richardson 2021) and French academic addictovigilance centres (Aquizerate 2023). The key CBT-I comparison trial was funded by the US National Institute of Mental Health, although its lead author had consulted for Sanofi-Aventis (Morin 2009).
• Allegiance can run both ways. The strongest mortality claims come from an author who publicly campaigns against hypnotics (Kripke 2012). The strongest challenge to the FDA's women's-dose rule comes from long-standing pharmacology researchers (Greenblatt 2019). We could not check their funding from the abstract. Neither position should be accepted or dismissed because of who holds it.
• Regulators changed course against commercial interest. The FDA's 2013 dose cut and 2019 boxed warning were both safety actions that could only reduce sales. They are therefore unlikely to have been shaped by the manufacturers' interests, even though industry user fees fund much of FDA drug review (FDA PDUFA report FY2025).

We found no documented research-misconduct finding specific to zolpidem trials in the sources we reviewed.

Related research

• Sleep prevention guide: CBT-I, habits and when to see a doctor
• Sleep supplements: independent evidence review
• Melatonin: evidence, dosing and safety
• Magnesium: evidence review
• L-theanine: evidence review
• St John's wort (reduces zolpidem levels)
• Stress, anxiety and depression prevention guide
• Sibling medicine reviews: zopiclone, eszopiclone, temazepam, daridorexant, trazodone, mirtazapine

Frequently asked questions

Sources and funding notes

• NHS: About zolpidem and linked pages (how to take, side effects, who can take it, pregnancy, interactions, common questions). UK public body (Crown copyright). Last reviewed 10 January 2023; the review due in January 2026 had not yet appeared when we checked.
• NHS: Insomnia. UK public body; reviewed 19 March 2024.
• Zolpidem Tartrate 10 mg Tablets, UK Summary of Product Characteristics (Zentiva Pharma UK), text revised 21/11/2025. Written by the licence holder and authorised by the MHRA. UK.
• Ambien prescribing information (Sanofi-Aventis U.S.), DailyMed and Ambien CR prescribing information (Cosette), DailyMed. Manufacturer-written, FDA-approved labels. USA.
• Drugs@FDA, NDA 019908. US regulator record (original approval 16 Dec 1992; current sponsor Cosette, per openFDA).
• FDA, 30 April 2019: boxed warning for complex sleep behaviours. US regulator.
• FDA, 10 January 2013: next-morning impairment and lower zolpidem doses. US regulator; prescription figures from IMS data.
• FDA PDUFA financial report FY2025. US regulator; documents the industry user-fee share of drug review costs.
• NICE TA77 (2004): zaleplon, zolpidem and zopiclone for short-term insomnia. UK public body. NICE is mainly funded by government grant, and companies pay fees for technology appraisals.
• Sateia et al., AASM clinical practice guideline, J Clin Sleep Med 2017. Funded by the AASM. Some panel members declared industry consulting; Krystal recused himself from the suvorexant recommendation. USA.
• Riemann et al., European Insomnia Guideline 2023, J Sleep Res. European Sleep Research Society; author conflicts not verified (publisher page blocked). Pan-European.
• American Geriatrics Society 2023 Beers Criteria, J Am Geriatr Soc. "No sponsor for this paper"; some panel members consult for insurers or data companies. USA.
• De Crescenzo et al., Lancet 2022 (network meta-analysis, 154 trials). Funded by UK NIHR. Declared ties: the first author was a Boehringer Ingelheim employee; the senior author has Angelini fees and runs Janssen seltorexant trials. UK/Italy.
• Huedo-Medina et al., BMJ 2012 (FDA dataset meta-analysis). No funding, no conflicts. USA/UK.
• Winkler & Rief, Sleep 2015 (placebo response). University of Marburg, Germany; funding not shown in the abstract.
• Morin et al., JAMA 2009 (CBT with or without zolpidem). Funded by NIMH; "not an industry-supported study"; the lead author has consulted for Sanofi-Aventis and others. Canada.
• Trauer et al., Ann Intern Med 2015 (CBT-I meta-analysis). Primary funding source: none. Australia.
• Hansen et al., AJPH 2015 (sedatives and car crashes). University of Washington and Group Health; no funding statement found in the fetched text. USA.
• Greenblatt, Harmatz & Roth, J Clin Psychopharmacol 2019 (zolpidem and gender). Academic; funding and conflicts not visible in the abstract. USA.
• Treves et al., Age Ageing 2018 (Z-drugs, falls and fractures). Academic (Israel); funding not shown in the abstract.
• Berry et al., JAMA Intern Med 2013 (hip fracture in nursing homes). Funded by the US National Institute on Aging and Friends of Hebrew SeniorLife; one author consults for OptumInsight on unrelated work. USA.
• Richardson et al., NIHR Health Technology Assessment 2021 (Z-drugs in dementia). Funded by UK NIHR. UK.
• Aquizerate et al., Eur J Public Health 2023 (French addictovigilance). No funding; no conflicts; data from ANSM. France.
• Kripke et al., BMJ Open 2012 (hypnotics and mortality). The lead author declared public criticism of hypnotics and small family holdings in Sanofi-Aventis stock. USA.
• "Zolpidem: world first launch of Synthelabo's hypnotic in France", Inpharma, March 1988. Trade news report (bibliographic record verified via Crossref).
• Sanofi Form 20-F for 2010, Form 20-F for 2012 and Form 20-F for 2025. Company filings with the US SEC (tier 4 for claims, reliable for sales figures). France.
• Cosette Pharmaceuticals press release, 10 July 2024. Company statement. USA.

Last reviewed: 26 September 2026. Educational information from Pure City Research, not medical advice. Medicine decisions belong with a qualified prescriber.