Lithium: Independent Evidence on Bipolar, Suicide & Kidney Risk

Table of contents

• Evidence summary
• What lithium is
• How it works
• What it is prescribed for
• What works and what does not
• Benefits by claim
• Risks and all side effects
• Blood-level monitoring: guideline target ranges
• All interactions
• Who should avoid lithium
• 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 tiers: 1 = regulator or government body; 2 = academic or Cochrane team with no declared commercial funding for the work; 3 = independent authors analysing mostly manufacturer-run trials, or minor declared industry ties; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D.

Most of these sources come from the UK and US, with Oxford researchers involved in several of the key reviews and trials. That clustering is worth knowing, though their findings were broadly matched by work from Denmark, Germany and the US.

What lithium is

Lithium is an alkali metal. In medicine it is used as lithium carbonate (tablets) or lithium citrate (liquid), and it is classed as a mood-stabilising agent, ATC code N05AN01 (Priadel SmPC 5.1). Its use in mania was first reported by John Cade in the Medical Journal of Australia in 1949 (Cade 1949).

• UK brands: The NHS lists Priadel, Camcolit, Liskonum and Li-Liquid (NHS). Priadel comes as 200 mg and 400 mg prolonged-release tablets and a 520 mg/5 ml liquid; Camcolit as 400 mg controlled-release tablets; Liskonum as 450 mg tablets; and Li-Liquid as 509 mg/5 ml and 1,018 mg/5 ml syrups (emc listing).
• US: Sold as generic lithium carbonate capsules and tablets and as Lithobid extended-release tablets (DailyMed: Lithobid).
• Status: Prescription-only (POM) in the UK (Priadel SmPC) and prescription-only in the US, where it carries a boxed warning about toxicity (FDA label).

Why the brand matters

Lithium products are not interchangeable. The Priadel SmPC says that "as bioavailability varies from product to product", a change of product "should be regarded as initiation of new treatment", with blood levels checked first (Priadel SmPC 4.2). The NHS Specialist Pharmacy Service lists lithium among medicines to prescribe by brand because it "has a narrow therapeutic index and preparations vary widely in bioavailability" (SPS, 2026). The NHS also warns not to switch between lithium carbonate and lithium citrate unless a doctor recommends it (NHS).

How it works

Nobody knows exactly. The UK label says "the exact mechanism of action of lithium in the treatment of bipolar disorders is not known". It notes that lithium changes the production and turnover of some neurotransmitters, especially serotonin, may block dopamine receptors, alters calcium and magnesium levels, and may reduce thyroid activity (Priadel SmPC 5.1). The NHS describes a theory that it may protect neurons and help create new ones, but says the mechanism is not fully understood (NHS). These mechanisms come largely from laboratory and animal work. What is proven in humans is the clinical effect, not the pathway.

Pharmacokinetics that drive the safety rules

• Cleared by the kidneys, not the liver. Lithium is cleared by the kidneys. Its elimination half-life is about 24 hours (US label), so steady levels take several days (Priadel SmPC 5.2).
• It behaves like sodium. When the body is short of sodium, the kidney holds on to more lithium, so levels rise (SmPC 4.4). This is why vomiting, diarrhoea, sweating, low-salt diets and some blood pressure drugs matter so much.
• Removable by dialysis. Haemodialysis is the treatment of choice for severe poisoning, and there is no antidote (SmPC 4.9).

What it is prescribed for

Where guidelines place it

• Bipolar disorder (NICE CG185): NICE recommendation 1.7.7 names lithium as first-line for long-term treatment (read directly on the NICE page). If lithium is ineffective, poorly tolerated or unsuitable (for example, because the person does not agree to routine blood tests), consider an antipsychotic.

What works and what does not

Benefits by claim

Does lithium prevent bipolar relapse?

Yes, especially manic relapse. The 2014 update of the main independent meta-analysis pooled seven placebo-controlled trials with 1,580 participants. Lithium reduced overall mood episodes (RR 0.66, 95% CI 0.53 to 0.82) and manic episodes (RR 0.52, 0.38 to 0.71). For depressive episodes, the random-effects estimate (RR 0.78, 0.59 to 1.03) crossed 1, while the fixed-effect estimate (RR 0.73, 0.60 to 0.88) did not (Severus et al. 2014; full text). In plain terms, about a third fewer people relapsed on lithium over the trial periods. The authors concluded that lithium remains the most valuable option for this indication.

The trade-off showed up too: more people stopped lithium for reasons other than a mood episode (RR 1.33, 1.07 to 1.65), usually side effects, although more completed the studies overall (RR 1.69) because fewer relapsed. Against anticonvulsants (seven trials, 1,305 people), lithium was better at preventing mania (RR 0.66, 0.44 to 1.00) but not clearly better on overall episodes.

The BALANCE trial randomised 330 people with bipolar disorder type I in the UK, France, the US and Italy to lithium, valproate or both, for up to 2 years. A relapse needing new treatment occurred in 59% on lithium, 69% on valproate and 54% on the combination. Lithium had a lower relapse rate than valproate; the combination's relapse rate (54%) was lower than valproate alone (69%), but the trial could not reliably confirm or rule out a benefit of the combination over lithium alone (Geddes et al., Lancet 2010). It was open-label and funded by the Stanley Medical Research Institute, with drugs donated by Sanofi-Aventis, a valproate maker.

Does lithium work for acute mania?

Yes, but it is not the fastest option. A network meta-analysis of 68 trials (16,073 adults), with a European Research Council grant listed in PubMed (260559), found these effects against placebo on mania scales at 3 weeks (Cipriani et al., Lancet 2011):

An SMD of 0.37 is a small-to-moderate effect. Haloperidol was significantly more effective than lithium (SMD −0.19). Olanzapine, risperidone and quetiapine also led to fewer dropouts than lithium. This is why lithium is often started alongside or after an antipsychotic in acute mania, then kept for prevention.

Does lithium prevent suicide?

This is the most debated claim. The 2013 BMJ meta-analysis included 48 randomised trials (6,674 people). Compared with placebo, lithium was linked to fewer suicides (OR 0.13, 95% CI 0.03 to 0.66) and fewer deaths from any cause (OR 0.38, 0.15 to 0.95), but no clear reduction in self-harm (OR 0.60, 0.27 to 1.32). In unipolar depression, the suicide OR was 0.36 (0.13 to 0.98) (Cipriani et al., BMJ 2013). The wide confidence intervals reflect how few suicides occurred in the trials.

The US Veterans Affairs Li+ trial then tested the idea directly. It randomised 519 veterans with depression or bipolar disorder who had a recent suicide-related event to extended-release lithium or placebo, added to usual care. It was stopped for futility. Suicide-related events occurred in 65 people on lithium and 62 on placebo (HR 1.10, 0.77 to 1.55). There was one death in the lithium group and three in the placebo group (Katz et al., JAMA Psychiatry 2022; full text). The authors concluded that adding lithium to usual care did not reduce suicide-related events.

How to read both together: The trials differ. Li+ enrolled veterans after a recent suicide-related event. The older trials compared long-term lithium with placebo in mood disorders generally. Our reading is that lithium probably lowers suicide risk partly by preventing relapse in people it suits, but it should not be seen as a stand-alone suicide-prevention drug.

Does lithium help treatment-resistant depression?

It can, as an add-on. A meta-analysis of 10 placebo-controlled augmentation trials (269 people) found lithium tripled the odds of response (OR 3.11, 1.80 to 5.37). Evidence that lithium speeds up response to antidepressants was weaker (SMD −0.43, −0.93 to 0.07) (Crossley & Bauer 2007). A later review of 39 trials found lithium "much more effective than placebo" as an adjunct, but data for lithium alone in acute depression were "few and inconclusive" (Undurraga et al. 2019). Europe PMC lists Chilean public research grants, a Bruce J. Anderson Foundation grant and a Clinica Alemana research grant for Undurraga et al.; Crossley and Bauer's record lists no grant.

The only long head-to-head trial is the NIHR-funded LQD study in six NHS trusts. It randomised 212 adults who had not responded to at least two antidepressants to add lithium or quetiapine. Over 12 months, the quetiapine group had a lower overall burden of depressive symptoms, time to stopping treatment did not differ, and quetiapine was probably more cost-effective (Cleare et al., Lancet Psychiatry 2025; HTA report). It was open-label, and it compared two active drugs, so it does not show that lithium fails. It shows that lithium is not clearly the best add-on.

Does lithium in drinking water reduce suicide?

All evidence here is observational, and much of it is ecological (comparing areas, not people). A meta-analysis of 15 ecological studies found areas with more natural lithium in tap water had lower suicide rates (pooled β −0.27, 95% CI −0.47 to −0.08), but the studies varied greatly (I² 83%). The authors note that ecological studies are open to the "ecological fallacy" and bias (Memon et al. 2020). It lists an NIHR grant (NF-SI-0616-10091); we could not retrieve its competing-interest statements (PubMed record).

A Danish study that followed 3.7 million individual adults for 22 years found "no significant indication" of an association between drinking-water lithium (up to about 31 μg/L) and suicide (Knudsen et al. 2017). None of this supports taking over-the-counter lithium supplements.

Risks and all side effects

Kidneys

Lithium commonly reduces the kidney's ability to concentrate urine. This causes thirst and passing large amounts of urine, and in some people nephrogenic diabetes insipidus (SmPC 4.8). A systematic review of 385 studies found urinary concentrating ability fell by about 15% of normal maximum. Average kidney filtration fell by 6.22 mL/min, which was not statistically significant, (McKnight et al., Lancet 2012; funded by an NIHR programme grant). Oxford laboratory data later linked lithium to stage 3 chronic kidney disease (HR 1.93, 1.76 to 2.12), with the highest risk in women under 60 and in people with above-median levels (Shine et al. 2015). The label also lists rare kidney tumours (microcysts, oncocytomas and collecting duct carcinoma) in people with severe kidney impairment after more than 10 years' use.

Thyroid and calcium

Hypothyroidism was more common on lithium than placebo, and lithium raised blood calcium and parathyroid hormone (McKnight 2012). In the Oxford data, hypothyroidism (HR 2.31) and raised total calcium (HR 1.43) were reported (Shine 2015). The UK label lists hypercalcaemia as "very frequent" and notes hyperparathyroidism and parathyroid adenoma. Lithium-induced hypothyroidism "may be managed successfully" with thyroxine (SmPC 4.8).

Weight, tremor and thinking

• Weight: People on lithium gained more weight than on placebo (OR 1.89, 1.27 to 2.82) but less than on olanzapine (McKnight 2012).
• Tremor: Fine hand tremor, thirst and passing more urine may persist, while early nausea usually settles (SmPC 4.8). A coarse tremor is a toxicity warning sign (NICE NG222).
• Neurotoxicity: NICE says to check at every appointment for tingling, unsteadiness, tremor and cognitive problems, "which can occur at therapeutic levels" (NICE CG185, 1.10.24). The NHS says some people feel their thinking is slowed or they feel "numb" (NHS).

Heart and other effects

The UK label lists slow heart rate, sinus node problems, QT prolongation and T-wave changes, and warns that lithium can unmask Brugada syndrome, an inherited rhythm disorder linked to sudden death. Other listed effects include skin problems (acne, psoriasis flares, hair loss), benign intracranial hypertension, seizures, sexual dysfunction, peripheral oedema and, rarely, DRESS, a serious drug reaction (SmPC 4.4 and 4.8). The US label adds serotonin syndrome and hyponatraemia warnings (FDA label).

Pregnancy

Lithium crosses the placenta. A US Medicaid cohort of more than 1.3 million pregnancies found cardiac malformations in 2.41% of 663 infants exposed in the first trimester vs 1.15% unexposed. (Patorno et al., NEJM 2017; funded by US NIMH grants K01 MH099141 and R01 MH100216 per Europe PMC).

The UK label says lithium should not be used in pregnancy "unless considered essential", recommends ultrasound and ECG screening of the baby, and warns that levels shift during pregnancy and suddenly at delivery (SmPC 4.6). The NHS says lithium can be used in pregnancy when needed after specialist review (NHS). Stopping suddenly carries its own relapse risk, so decisions belong with a specialist.

Breastfeeding

Sources disagree. The Priadel SmPC lists breastfeeding as a contraindication because neonatal toxicity has been reported (SmPC 4.3 and 4.6). The US label says breastfeeding "is not recommended", and advises close infant monitoring if a mother chooses to breastfeed (FDA label 8.2). The NHS says that if the baby is healthy, lithium can be taken while breastfeeding, "usually after specialist advice", with monitoring of the baby (NHS).

Blood-level monitoring: guideline target ranges

Lithium should not be started "unless adequate facilities for routine monitoring" are available (Priadel SmPC 4.2). The ranges below are guideline and label targets for context only. The right level for any person is set by their specialist.

For lithium, mmol/L and mEq/L give the same number, because the lithium ion carries a single charge. For a first course of lithium for bipolar disorder, NICE aims for a plasma level of 0.6 to 0.8 mmol/L, and considers 0.8 to 1.0 mmol/L for a trial of at least 6 months after a relapse on lithium or with persisting subthreshold symptoms ( NICE CG185, 1.10.15–1.10.16 ). For lithium added to an antidepressant, NG222 says levels should not exceed 1.0 mmol/L.

How often tests are done (NICE)

• Levels 1 week after starting and 1 week after every dose change, then weekly until stable (CG185, 1.10.15).
• Every 3 months for the first year, then every 6 months. Testing stays at every 3 months for older people, people on interacting drugs, people at risk of kidney or thyroid problems, people with poor symptom control or adherence, and anyone whose last level was 0.8 mmol/L or higher (1.10.19 and 1.10.20).
• Weight, urea and electrolytes including calcium, eGFR and thyroid function every 6 months, and more often if problems appear (1.10.21). An ECG at the start for people with heart disease or risk factors (1.10.14).
• NG222 adds: do not start repeat prescriptions until lithium levels and kidney function are stable, and check for toxicity signs at every review (NICE NG222).

The 2009 national patient safety alert

In December 2009 the National Patient Safety Agency issued alert NPSA/2009/PSA005, "Safer lithium therapy" (North Bristol NHS Trust statement citing the alert). The North Bristol document, which cites the alert, sets out monitoring in line with NICE, a written shared-care arrangement between psychiatrist and GP, and a lithium treatment pack for every patient: an information booklet, an alert card and a record book for blood results. The Pharmaceutical Journal says pharmacy professionals should make these resources available and repeat the safety information at regular intervals (Pharmaceutical Journal, 2024). NICE NG222 still tells clinicians to provide "the NHS lithium treatment pack" (NICE NG222).

All interactions

From the UK label (Priadel SmPC 4.5) unless otherwise linked:

The NHS says there is not enough information to say herbal remedies and supplements are safe with lithium (NHS).

Who should avoid lithium

The UK label lists these contraindications (Priadel SmPC 4.3):

• Heart disease or heart failure, and Brugada syndrome or a family history of it.
• Severe kidney impairment.
• Untreated hypothyroidism.
• Breastfeeding (but see the different NHS advice above).
• Low body sodium, including dehydration or a low-sodium diet.
• Addison's disease.

Use with extra care: mild or moderate kidney impairment, epilepsy, congenital long QT syndrome, after bariatric surgery (a lower dose may be needed), and in older people, who are "particularly liable to lithium toxicity" (SmPC 4.4). Women who could become pregnant are advised to use effective contraception (SmPC 4.6).

Children: Use in children is "not recommended" in the UK label. The US label approves it for bipolar disorder type I from age 7 (FDA label).

Driving: Lithium may slow reaction time (SmPC 4.7). The NHS notes that people diagnosed with bipolar disorder must tell the DVLA (NHS).

Dosage and how to take it

Your specialist sets the dose by blood level, not by a fixed number of milligrams. The figures below are label doses for context only.

How it is taken

The NHS says lithium is usually taken once a day at night, at the same time each day, so the blood test can be done 12 hours after the dose. It can be taken with or without food, and a missed dose should not be doubled (NHS; SmPC 4.2).

How long it takes

Acutely ill patients may respond within 3 to 7 days, but "the full prophylactic effect may not occur for 6 to 12 months" (SmPC 4.2).

Stopping

Stopping must be planned with a specialist. NICE advises reducing the dose gradually over at least 4 weeks, and preferably up to 3 months, with close monitoring for relapse for 3 months afterwards (CG185, 1.10.25–1.10.26), and NG222 says to stop lithium only in, or with advice from, specialist mental health services (NG222). The exception is suspected toxicity, when the NHS says to stop and get urgent help (NHS).

Product quality

The main quality issue is not contamination but switching. Because products differ in how much lithium reaches the blood, a brand or formulation change is treated as starting again, with new level checks (SmPC 4.2). We found no regulator-reviewed evidence for low-dose lithium supplements sold outside prescription.

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

Who makes and sells it

• Essential Pharma: Holds the UK licences for Priadel (licence holder Essential Pharma Limited, Birkirkara, Malta) (Priadel SmPC), Camcolit 400 mg (Essential Pharma, Egham, Surrey) (Camcolit SmPC) and Lithium Carbonate Essential Pharma 250 mg (SmPC). Sources differ on the ownership timeline: one says the private equity firm Gyrus Capital bought it in 2019, another says a 2020 management buyout backed by Gyrus, and in December 2024 announced a €900 million recapitalisation through a Gyrus continuation vehicle led by AlpInvest, with financing from Sixth Street (Sixth Street announcement, 2024).
• Teofarma S.r.l. (Italy): Liskonum 450 mg (Liskonum SmPC).
• Rosemont Pharmaceuticals (Leeds, UK): Li-Liquid syrups (Li-Liquid SmPC).
• US: Many generic makers, including Hikma (DailyMed), and ANI Pharmaceuticals for Lithobid (DailyMed).

Documented integrity event: the Priadel withdrawal case

In October 2020 the Competition and Markets Authority (CMA) opened an investigation into whether Essential Pharma had abused a dominant position by proposing to withdraw Priadel from UK patients. The CMA said "the vast majority" of UK patients on lithium relied on Priadel, and that withdrawal would push thousands onto more expensive alternatives such as Camcolit, also owned by Essential Pharma (CMA, 6 October 2020). Essential Pharma agreed supply terms with the Department of Health and Social Care (CMA, November 2020).

On 18 December 2020 the CMA accepted legally binding commitments lasting 5 years: Priadel would continue to be supplied on terms agreed with the government. The decision does not give new pack prices; it lists earlier prices of £2.76 and £4.02 per 100 tablets (200 mg and 400 mg, June 2019 to May 2020). The December 2020 release does not say whether a breach was found; the CMA had said it had "reasonable grounds for suspecting" one (CMA, 18 December 2020). The 5-year period ran to about December 2025; we could not find any public record of what happened when it ended.

Who funded the evidence

• Mostly public money. The Li+ suicide trial was funded by VA Cooperative Studies Program grant CSP590, with the sponsor reporting no role (Katz 2022). LQD was funded by the NIHR Health Technology Assessment programme (Cleare 2025). The pregnancy cohort was funded by US NIMH grants (Europe PMC) (Patorno 2017). The toxicity review had an NIHR programme grant (McKnight 2012).
• BALANCE was funded by the Stanley Medical Research Institute, a US non-profit that funds schizophrenia and bipolar research (SMRI), and by Sanofi-Aventis (Geddes 2010). Sanofi sold a competing drug, valproate, yet the trial favoured lithium.

What this means: Lithium is unusual among psychiatric medicines. Its best evidence comes mainly from public funders and academic groups, not from companies selling it. The commercial pressure in this story ran the other way: a dominant supplier tried to raise revenue by threatening to withdraw a cheap product. Where study authors have declared company ties, we report them; where we could not retrieve the declarations, we say so rather than assume them.

Related research

For wider, evidence-graded context, see these Pure City Research guides:

• Bipolar depression: assessment, treatment evidence and safety
• Treatment-resistant depression
• Depression relapse prevention
• Depression with psychotic features
• Postpartum psychosis
• Brain stimulation for depression
• Esketamine (Spravato)
• Chronic kidney disease and the heart
• Ramipril (an ACE inhibitor that raises lithium levels)
• Losartan (an ARB that raises lithium levels)
• Electrolytes, water and oral rehydration
• Sertraline

Frequently asked questions

Sources and funding notes

What we could not verify in this check: This pass re-opened the central sources through PubMed abstracts, Europe PMC, the full text of Katz 2022 and the CMA decision PDF. Confirmed: NICE CG185 (September 2025 update) and NG222 recommendations quoted here, read directly on the NICE site; Severus 2014 (all figures in the abstract); Cipriani 2011 (SMDs and haloperidol comparison; PubMed lists an ERC grant only); Cipriani 2013 (48 trials, suicide OR 0.13, deaths, self-harm and unipolar figures; MRC grant G0700477); Crossley and Bauer 2007 (augmentation OR 3.11 and acceleration SMD); Undurraga 2019 (quoted phrases; 39 trials); Katz 2022 (519 veterans, HR 1.10, 65 vs 62 events, VA CSP590 funding with no sponsor role, author pharma ties); Cleare 2025 (212 participants, Lancet Psychiatry, NIHR HTA version also checked); Shine 2015 (HR 1.93 for stage 3 CKD, 2.31 for hypothyroidism, 1.43 for total calcium; abstract declares no competing interests); Memon 2020 (15 studies, pooled beta -0.27, I2 83%); Knudsen 2017 (3.7 million adults, 22 years, no significant association below 31 μg/L; no conflict declared); BALANCE rates (59% lithium, 69% valproate, 54% combination; funders Stanley Medical Research Institute and Sanofi-Aventis); the CMA decision of 18 December 2020 (case 50951, five-year term from acceptance, reasonable grounds for suspecting abuse); the December 2024 Essential Pharma recapitalisation (AlpInvest and Sixth Street); the FDA-hosted 2020 label wording on 1.5 mEq/L; Patorno 2017 crude rates (16 of 663 exposed vs 15,251 unexposed, 2.41% and 1.15%). Removed because we could not confirm them: the Priadel prices of £7.50 and £8.50 per pack; the exact Patorno denominator of 1,325,563; the claim that the Patorno authors called the effect smaller than suspected; the claim that Essential Pharma paused its withdrawal; the claim that Sanofi donated BALANCE drugs; and the exact NICE 1.7.7 wording (the article now paraphrases it). The Gyrus Capital ownership timeline is reported as disputed (2019 acquisition versus a 2020 management buyout). Still unverified: most SmPC pharmacokinetic and dosing details beyond the half-life; competing-interest statements for Cipriani 2011, Cipriani 2013, Severus 2014, Patorno 2017, Memon 2020, Crossley and Bauer 2007, Cleare 2025 and Undurraga 2019; Patorno funding (NIMH, per Europe PMC only); Undurraga's Chilean grants; the 2009 NPSA alert; and any Priadel pricing after the commitments. Funding is not assumed where marked.

1. NICE CG185: Bipolar disorder: assessment and management (published 2014, updated September 2025) — UK guideline; funded by the Department of Health and Social Care.
2. NICE NG222: Depression in adults: treatment and management (2022) — UK guideline; DHSC.
3. Priadel 400 mg prolonged-release tablets SmPC (revised September 2024) — MHRA-approved; licence holder Essential Pharma Ltd (Malta).
4. Camcolit 400 mg SmPC (revised November 2024) — MHRA-approved; Essential Pharma (UK).
5. Lithium Carbonate Essential Pharma 250 mg SmPC — MHRA-approved; Essential Pharma.
6. Liskonum 450 mg SmPC (revised August 2024) — MHRA-approved; Teofarma S.r.l. (Italy).
7. Li-Liquid 1018 mg/5 ml SmPC (revised September 2024) — MHRA-approved; Rosemont Pharmaceuticals (UK).
8. emc lithium product listing — UK product database.
9. US lithium carbonate prescribing information (Hikma; label revised November 2025; DailyMed version updated August 2026) — FDA-approved label; US NLM.
10. Lithobid label (ANI Pharmaceuticals; DailyMed) — FDA-approved label.
11. NHS: Lithium (pages last reviewed 9 August 2023; next review was due August 2026), including side effects, how to take it, pregnancy and breastfeeding, interactions and common questions — UK government.
12. NHS Specialist Pharmacy Service: Example medicines to prescribe by brand name in primary care (2026) — NHS.
13. North Bristol NHS Trust: Lithium monitoring statement of good practice — NHS trust document citing NPSA/2009/PSA005.
14. Godden H. Lithium monitoring and toxicity management. Pharmaceutical Journal, February 2024 — Royal Pharmaceutical Society journal.
15. CMA: CMA to investigate the supply of bipolar drug (6 October 2020) — UK government.
16. CMA: CMA takes swift action in bipolar drug investigation (November 2020) — UK government.
17. CMA: CMA secures affordable supply of key bipolar drug (18 December 2020) — UK government.
18. Sixth Street: Essential Pharma announces €900 million recapitalisation (12 December 2024) — company financing announcement.
19. Severus E et al. Lithium for prevention of mood episodes in bipolar disorders. Int J Bipolar Disord 2014 (full text) — Germany/UK; no grant record in Europe PMC; competing-interest statement not verified.
20. Geddes JR et al. BALANCE. Lancet 2010 — Oxford-led; Stanley Medical Research Institute and Sanofi-Aventis.
21. Cipriani A et al. Antimanic drugs in acute mania: multiple-treatments meta-analysis. Lancet 2011 — ERC grant 260559 listed in PubMed; NIHR not confirmed.
22. Cipriani A et al. Lithium in the prevention of suicide in mood disorders. BMJ 2013 — Italy/UK; UK MRC grant listed in PubMed.
23. Katz IR et al. Lithium in the prevention of repeat suicide-related outcomes in veterans. JAMA Psychiatry 2022 (full text) — VA CSP590 per the full-text funding statement, sponsor no role; authors report pharma ties.
24. Crossley NA, Bauer M. Lithium acceleration and augmentation meta-analyses. J Clin Psychiatry 2007 — Germany; no grant record in Europe PMC; competing-interest statement not verified.
25. Undurraga J et al. Lithium treatment for unipolar major depressive disorder. J Psychopharmacol 2019 — funded by Chilean public research grants (CONICYT/FONDECYT), the Bruce J. Anderson Foundation and a Clinica Alemana research grant (Europe PMC grant record).
26. Cleare AJ et al. Lithium versus quetiapine augmentation (LQD). Lancet Psychiatry 2025 (HTA report) — UK; NIHR HTA; author competing-interest statements not retrieved in this check.
27. Memon A et al. Lithium in drinking water and suicide rates. Br J Psychiatry 2020 (journal page) — NIHR grant NF-SI-0616-10091 listed; competing-interest statements not retrieved.
28. Knudsen NN et al. Lithium in drinking water and incidence of suicide (Denmark). Int J Environ Res Public Health 2017 — the paper declares no conflict of interest; funding not retrieved.
29. McKnight RF et al. Lithium toxicity profile. Lancet 2012 — UK; NIHR Programme Grant for Applied Research.
30. Shine B et al. Long-term effects of lithium on renal, thyroid and parathyroid function. Lancet 2015 — UK; abstract declares no competing interests; funder not confirmed in the PubMed record.
31. Patorno E et al. Lithium use in pregnancy and the risk of cardiac malformations. NEJM 2017 (full text) — US; funder not verified in this check.
32. Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust 1949 — historical first report.
33. Stanley Medical Research Institute — US non-profit research funder (BALANCE co-funder).

Educational research reviewed 9 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.