Low-Dose Lithium — Body Handbook

Supplements · §519

Low-Dose Lithium

Lithium is in your tap water at trace levels — micrograms per litre, a thousandth of the dose used for bipolar disorder. Towns with more of it have lower suicide rates, replicated across nine countries. The brains of people with early Alzheimer's are missing it, and amyloid plaques actively pull it out of circulation. The story is real and the mechanism is one of the cleanest in nutritional neuroscience. The catch is that no proper trial has tested the supplement dose, so a confident "everyone should take it" hasn't been earned. Here is what you can say for sure, and what the honest bet looks like.

Decide · Once Evidence Mixed Chapter Supplements

The headline finding is that a 2025 Nature paper identified lithium as the only metal significantly depleted in the cortex of people with mild cognitive impairment, and showed that dietary lithium depletion produces Alzheimer-type pathology in mice that low-dose lithium reverses. The supplement form, lithium orotate, costs about a dollar a month and the daily effort is one small capsule. What's missing is a large human trial at the supplement dose — so this is a decide, not a do. If you live in the U.S. mountain West your tap water may already cover it; if you filter heavily or live in the Northeast it probably doesn't.

Start with the dose ladder. The lithium used for bipolar disorder runs at 600 to 1,200 mg a day of lithium carbonate — enough to put a measurable level in your blood and require quarterly monitoring of kidneys and thyroid. The lithium in a tap-water glass is roughly a thousand times less. The lithium in a daily orotate capsule sits in between, but much closer to the water — five to twenty milligrams of elemental lithium, about a sixtieth of the psychiatric dose. When this entry says "low-dose," it means the bottom two rungs.

The mechanism most worked-out is that lithium puts the brakes on an enzyme called GSK-3β. That enzyme is one of the central villains in Alzheimer-type damage: it phosphorylates tau (the protein that tangles inside dying neurons), it ramps up the production line for amyloid (the protein that clumps in plaques between neurons), and it nudges the brain's immune cells into a pro-inflammatory state. Lithium slows all three. Cell culture work shows it; mouse studies show it; the effect holds down the dose curve well below the bipolar range Aron et al. 2025.

The new piece — and this is what makes the topic interesting in 2025 rather than 2005 — is the deficiency story.

That reframes the question. It is no longer just "could a lithium pill be a drug for Alzheimer's?" It is "is dietary lithium deficiency one of the things that lets Alzheimer's start?" Those are different claims with different stakes.

What we actually have

Three streams of evidence point the same direction; none of them by itself is conclusive.

One: the population suicide signal. Across roughly 1,300 localities in nine countries — Japan, Austria, the United States, England, Greece, Italy, Lithuania, and others — towns with more lithium in their tap water have lower suicide rates.

The catch on this whole stream is that it is ecological — town-level, not person-level. Lithium-rich groundwater tends to come with mineral-rich geology, rural geography, and population-density patterns that can themselves bend suicide rates. The signal has held across countries with very different demographics, which makes pure-confounding harder to swallow but not impossible.

Two: the prescription-dose suicide trials. When lithium is used at full psychiatric dose in people with mood disorders, it cuts the suicide rate hard.

This is high-grade trial evidence, but at the full dose, in people who already have a mood disorder. The reasoning that drops it down to the supplement dose in the general population is mechanism-based, not trial-based.

Three: the cognitive-decline trials. One Brazilian research group ran two small trials of lower-than-psychiatric lithium in older adults with mild cognitive impairment — the at-risk-for-Alzheimer's group.

The trials are small, the final analysed sample after attrition was around 34 people, and the dose used was still well above what you can buy over the counter. A larger U.S. multicenter trial (LATTICE) is running but had not reported results as of this writing. The signal is real and the direction is consistent across the two trials; it has not yet been put on the floor of a thousand-person study.

If you stack the three: a robust population suicide signal at trace dose, a high-evidence suicide-prevention finding at full dose, and a small but pointed cognitive-protection finding at half dose. Pulled together by a clean mechanism that operates down the curve. That is the case at its strongest.

What you're already getting

Most adults take in 0.6 to 3 milligrams of elemental lithium a day just from food and tap water, with grains, vegetables, dairy, mineral water, and eggs doing the bulk of the work Schrauzer 2002. The problem is that "average" hides a huge geographic range. The U.S. Geological Survey mapped lithium across U.S. groundwater in 2024 and split it into four bins: under 4, 4–10, 10–30, and over 30 micrograms per litre. Over 30 — the threshold where the population suicide signal appears — covers much of the mountain West and Southwest: Montana, Wyoming, the Dakotas, Colorado, Utah, Nevada, Arizona, New Mexico, Texas USGS 2024. Most of the Northeast, the Pacific Northwest, and the upper Midwest sit below it. Reverse osmosis and most carbon-block filters strip lithium out along with everything else, so a heavy-filter household in any region is effectively in the low-bin. Bottled water varies wildly by source.

That is the first practical question for most readers: are you already in the high-bin or not. Your municipal water-quality report sometimes lists lithium; often it doesn't. If you can find the number, you have your answer; if not, the rough geographic map covers most cases.

If you decide to supplement

The over-the-counter form is lithium orotate, sold at health-food stores and online for roughly $10–30 a year at standard supplement doses. Capsules are usually labelled as 100–150 mg of lithium orotate, of which 5 to 20 milligrams is the actual elemental lithium — orotate is the mass carrier Pacholko & Bekar 2021. Most discussion in the supplement literature lands at 1–5 mg elemental per day for the nutritional-deficiency framing, with the higher end (5–20 mg) reserved for users targeting the neuroprotection story more aggressively.

Two honest framings to hold while you decide. First, this is a long-game bet. The cognitive-protection trials needed two to four years of dosing before signals appeared; nobody walks away from a week of lithium orotate feeling different. If you want a supplement with felt effects in the first month, this isn't it. Second, the dose you can buy is well below the dose that has been trialled for cognitive endpoints. The reasoning that connects them is mechanism plus the Aron 2025 finding that physiological-range dietary lithium is enough to matter — solid, but not the same thing as a positive trial at the dose on your kitchen counter.

When not to do this

The safety profile at supplement doses is good but not zero. The risks come from the same biology that makes psychiatric-dose lithium effective.

Pregnancy and breastfeeding. At psychiatric doses lithium raises the risk of cardiac malformations in the developing fetus, including the rare Ebstein's anomaly — adjusted relative risk 1.65 for first-trimester exposure, dose-dependent, with no detectable excess below 600 mg/day of carbonate Patorno et al. 2017. Supplement doses are far below that, but the data at supplement doses doesn't exist. Default to skipping it during pregnancy and while breastfeeding.
Existing kidney disease. Lithium is cleared by the kidneys; impaired clearance can raise blood levels even from a small dose. If you have known chronic kidney disease, don't self-supplement.
Thyroid disease or thyroid medication. Lithium suppresses thyroid hormone release; about a quarter of patients on full-dose lithium develop hypothyroidism. The risk at supplement doses is presumed small but real enough that a TSH check is worth doing if you're already on thyroid management.
Drug interactions. Thiazide diuretics, loop diuretics, NSAIDs taken daily, and ACE inhibitors all reduce kidney clearance of lithium. Even a supplement dose can stack to meaningful blood levels if you're on these long-term.
It's a supplement, not a regulated drug. The FDA does not pre-market test supplements. A 2007 case report described a teenager who took 18 lithium orotate tablets and developed severe tremor and vomiting at blood levels below the bipolar therapeutic range — meaning either she was unusually sensitive or the orotate carrier moved lithium into the brain more efficiently than the blood number suggests. Don't double up because you forgot yesterday's capsule.

Three things often gotten wrong

"Low-dose lithium is just psychiatric lithium, less of it." The ratio is roughly one to sixty. A bipolar prescription is six hundred or more milligrams of lithium carbonate a day with quarterly bloodwork; a supplement capsule is five to ten milligrams of elemental lithium with no monitoring. The risk story scales nonlinearly — most of the long-term kidney and thyroid harm of therapeutic lithium appears in people with measurable blood levels held there for years. At the supplement dose blood levels usually stay below detection. They are not the same intervention.

"Drinking-water studies prove that more lithium prevents suicide." They show a consistent association across nine countries. That is much stronger than nothing, but ecological studies look at towns, not people; what you cannot rule out is that something else about lithium-rich regions — geography, demography, density — is the actual driver. The signal is robust enough that pure coincidence is hard to defend; it is not the same as a person-level proof.

"Lithium orotate is proven to prevent dementia." No randomised trial of lithium orotate, specifically, has been completed for any cognitive endpoint. The trials that showed cognitive protection in mild cognitive impairment used lithium carbonate at roughly half psychiatric dose — about thirty times what you can buy over the counter. The supplement story rests on mechanism plus the 2025 finding that physiological-range dietary lithium matters in mice. That's interesting; it is not the same as a positive trial.

What this isn't competing with

Low-dose lithium is not a replacement for the things with the strongest evidence for delaying dementia. Those are exercise (especially aerobic plus resistance), sleep adequacy and apnea correction, hearing-loss correction in midlife, blood-pressure control, and a Mediterranean-pattern diet. Each of those has trial evidence and effect-size estimates that are at least as strong, and most are also doing other useful work in the body. If you're going to do one new thing for your brain in your fifties, do one of those.

For suicide prevention at the individual level, the comparison points are addressing untreated depression directly — therapy, SSRIs, ketamine where appropriate — and removing means in the home. Low-dose lithium does not sit at the front of that list either.

Where low-dose lithium earns its place is as an additive bet: low cost, low effort, mechanism-plausible, on top of those higher-evidence moves rather than instead of them.

The long arc

Lithium has a longer history with the human body than most psychiatric drugs. In the 19th century, "lithia springs" at Lithia Springs (Georgia), Bath, and a string of European spa towns were widely held to calm the nerves; the early-20th-century soda 7-Up was originally branded "Bib-Label Lithiated Lemon-Lime Soda" and contained lithium citrate until 1948. The psychiatric story starts in 1949 when John Cade in Melbourne, treating manic patients with lithium as an experimental control, noticed they became calmer. Lithium for bipolar disorder was standard care by the mid-1970s.

The supplement form, lithium orotate, was developed and advocated by the German doctor Hans Nieper in the 1970s on the theory that orotate could carry lithium more efficiently into cells. The trace-nutritional framing was revived academically by Gerhard Schrauzer's 1990 Texas-counties paper, dormant for two decades, and then renewed by the Japanese drinking-water replications in the late 2000s and the Forlenza cognitive-decline trials starting in 2011. The 2025 Aron Nature paper is the latest turn — and reframes lithium not as a candidate drug but as a candidate dietary essential.

Related threads

This entry leaves several adjacent topics for their own pages: full-dose lithium for bipolar disorder, the broader dementia-prevention checklist (exercise, sleep, hearing, blood pressure, Mediterranean diet), water filtration choices and what they strip out, and the public-health debate over fortifying tap water with lithium the way some jurisdictions fortify with fluoride. If your interest here is mood specifically rather than long-term brain health, the higher-evidence moves — sleep, daylight exposure, exercise, social connection — come first.

Substance and claimed effects

Lithium is a monovalent alkali metal present at trace concentrations in drinking water, soil, and plant foods. It enters the human body involuntarily through diet — grains, vegetables, dairy, mineral water — with U.S. daily intake estimated at 0.6-3 mg elemental lithium Schrauzer 2002. Schrauzer proposed a provisional minimum requirement of 1 mg/day for a 70-kg adult on biological-trace-element grounds, though lithium is not formally recognised as essential by major nutrition bodies. Beyond background dietary intake, lithium orotate is sold over the counter in the United States at 5-20 mg elemental lithium per capsule — roughly 1/20 to 1/100 of the psychiatric dose used for bipolar disorder. This entry covers the substance at the nutritional / supplemental scale: (a) lithium present in drinking water and food, (b) lithium orotate at OTC doses. It covers the claimed effects: population-level suicide rate reduction in higher-lithium areas, individual mood stabilization, and brain aging / dementia prevention. It does not cover therapeutic-dose lithium carbonate for bipolar disorder, which is a prescription drug with a different risk profile.

Evidence by addressing question

mechanism

Lithium ion inhibits glycogen synthase kinase-3 beta (GSK-3β) directly and indirectly. GSK-3β phosphorylates tau, drives amyloidogenic processing of amyloid precursor protein through up-regulation of BACE1, and promotes pro-inflammatory microglial activation — all three are central nodes of Alzheimer's pathology. Subtherapeutic lithium decreases GSK-3β activity in hippocampus and prefrontal cortex in rodent models, reducing tau hyperphosphorylation and attenuating Aβ generation Aron et al. 2025. Lithium also activates Wnt/β-catenin signaling, upregulates BDNF, enhances autophagy, and promotes neurogenesis in dentate gyrus. These mechanisms are dose-dependent down the curve: GSK-3 inhibition has been demonstrated at serum levels well below the bipolar therapeutic range (0.25-0.5 mmol/L), and recent murine data show pathology-relevant effects at concentrations consistent with high-end dietary intake.

For suicide-specific effects, the mechanism is less well characterised. Candidate pathways include serotonergic modulation (lithium enhances serotonergic neurotransmission), impulse-control effects through inositol monophosphatase inhibition, and circadian / sleep-architecture stabilisation. Cipriani's meta-analytic finding that lithium reduces suicide independent of mood-stabilization effect suggests a specific anti-suicidal mechanism, but its molecular basis is unresolved Cipriani et al. 2013.

The 2025 Aron et al. Nature paper added a new mechanistic frame: lithium is the only metal significantly depleted in cortical tissue of patients with mild cognitive impairment, and amyloid plaques actively sequester lithium, creating a deficiency state that further accelerates pathology. Reducing endogenous cortical lithium by ~50% in mice — bringing it to the low end of the human range — produced amyloid-β deposition, phospho-tau accumulation, microglial activation, synapse and myelin loss, and accelerated cognitive decline Aron et al. 2025. This reframes the question from "does pharmacologic lithium help?" to "is lithium-deficient diet a contributing cause?"

evidence

Population-level suicide signal (ecological). The strongest empirical signal for low-dose lithium is the cross-country correlation between drinking-water lithium and suicide mortality. Schrauzer's 1990 Texas study compared 27 counties stratified by municipal-water lithium content (range 0–170 μg/L) and found inverse correlations with suicide, homicide, and arrests for drug-related crimes Schrauzer & Shrestha 1990. Ohgami et al. 2009 replicated the suicide finding in 18 Japanese municipalities (lithium 0.7–59 μg/L), with the inverse association significant after adjustment Ohgami et al. 2009. The Memon et al. 2020 meta-analysis pooled 15 ecological studies across Japan, Austria, the U.S., England, Greece, Italy, and Lithuania, comprising 1,286 localities, and reported a pooled β = −0.27 (95% CI −0.47 to −0.08) for total suicide mortality and a separate sex-stratified effect for females (β = −0.13, 95% CI −0.24 to −0.02) Memon et al. 2020. The dose ranges where signals appear are roughly 30–170 μg/L; below 30 μg/L the protective association weakens or disappears. The studies are ecological — the fundamental confounder is that lithium-rich water is often associated with mineral-rich regions, rural geography, and population-density patterns that correlate with suicide independently.

RCT data for low-dose lithium in cognitive decline. The Forlenza group at the University of São Paulo conducted the most-cited low-dose trials. The 2011 study randomised 45 patients with amnestic MCI to lithium (titrated to serum 0.25–0.5 mmol/L — roughly one-quarter to one-half of bipolar dosing) or placebo for one year; lithium reduced CSF phospho-tau and stabilised cognitive scores Forlenza et al. 2011. The 2019 extension followed 61 MCI patients for two years on the same regimen; cognitive scores remained stable in the lithium arm while the placebo arm showed continuous decline (p < 0.05 on Alzheimer's Disease Assessment Scale–cognitive subscale at 24 months), and at four-year open-label extension five of 31 lithium subjects vs nine of 30 placebo subjects had progressed to dementia Forlenza et al. 2019. The trial was small (final analysed n ≈ 34) and the dose (serum 0.25–0.5 mmol/L) sits above OTC lithium-orotate dosing but well below bipolar dosing. Larger replications are pending — the multicenter LATTICE trial (NIA-funded, U.S.) is ongoing as of 2025.

RCT data for therapeutic-dose lithium and suicide. Cipriani et al.'s BMJ 2013 meta-analysis of 48 RCTs (n = 6,674) in unipolar and bipolar mood disorders found lithium reduced completed suicide vs placebo with OR 0.13 (95% CI 0.03–0.66) and all-cause mortality OR 0.62 Cipriani et al. 2013. This is high-evidence but at therapeutic doses (serum > 0.6 mmol/L); the extrapolation to OTC lithium orotate is mechanism-based, not trial-based.

Lithium orotate specifically. No randomized controlled trial of lithium orotate has been published. The two source studies cited in supplement literature are open-label: Nieper's 1973 observational series in mood patients and Sartori's 1986 open-label trial of 5.7 mg/day elemental lithium in 42 alcoholic patients claiming reduced cravings and depression relief Sartori 1986. Both had n < 50, no placebo, and methodological weaknesses. Pharmacokinetics work in mice suggests orotate as the carrier may achieve higher brain-tissue lithium for a given serum level versus carbonate, but human pharmacokinetic data is sparse Pacholko & Bekar 2021.

protocol

Nutritional baseline. Most readers already ingest 0.6–3 mg/day from diet and tap water, varying with geography. U.S. tap-water concentrations span less than 1 μg/L to nearly 400 μg/L in public-supply wells; the USGS estimates 30+ μg/L (the threshold associated with population-level suicide effects) in much of the West and Southwest — Montana, Wyoming, the Dakotas, Colorado, Utah, Nevada, Arizona, New Mexico, Texas — and lower across the East and Midwest USGS 2024. Reverse-osmosis and most carbon-block filters remove lithium; bottled water varies widely.

OTC supplementation. Lithium orotate is sold at 1–20 mg elemental lithium per capsule (often labelled as 100–150 mg lithium orotate, where the orotate moiety accounts for the bulk of mass). The Forlenza MCI trial dose (serum 0.25–0.5 mmol/L) requires 150–300 mg/day lithium carbonate — far above OTC supplementation. Schrauzer's nutritional-minimum recommendation is 1 mg elemental/day. Common supplement guidance for the dementia-prevention rationale targets 1–5 mg elemental/day, well above background diet but several orders below psychiatric dosing.

Monitoring. At OTC doses (< 20 mg/day elemental), serum monitoring is not customary and serum levels stay well below detectable therapeutic-range thresholds. The 2007 case report of toxicity from lithium orotate at 18 tablets demonstrated symptoms (severe tremor, vomiting) at serum 0.3–0.4 mEq/L, which is below the bipolar therapeutic floor — suggesting either idiosyncratic susceptibility or carrier-dependent brain accumulation not reflected in serum.

contraindications

The contraindications are inherited from therapeutic-dose lithium pharmacology, with reduced — but not zero — applicability at OTC doses.

Pregnancy. Therapeutic lithium is associated with a dose-dependent increase in cardiac malformations including Ebstein's anomaly. The Patorno et al. 2017 NEJM cohort (1,325,563 pregnancies) reported adjusted RR 1.65 for cardiac malformations in first-trimester lithium exposure vs unexposed; the effect was dose-dependent, with no excess at <600 mg/day lithium carbonate Patorno et al. 2017. At OTC lithium-orotate doses (5–20 mg elemental ≈ 27–107 mg lithium carbonate equivalent) the risk is presumably very low but unstudied. Default to avoid.
Renal impairment. Long-term therapeutic lithium causes nephrogenic diabetes insipidus and accelerates eGFR decline (~−2.9 mL/min/year). The threshold serum level associated with stage-3 CKD development is around 0.59 mEq/L. At OTC doses serum lithium typically remains below detectable, but patients with existing CKD have impaired lithium clearance and should not self-supplement.
Thyroid disease. Lithium inhibits thyroid hormone release; therapeutic dosing causes hypothyroidism in roughly one-quarter of patients. Threshold serum for hypothyroidism is around 0.5 mEq/L. Risk at OTC dose is presumably small but TSH monitoring is reasonable for sustained supplementation.
Medication interactions. Diuretics (thiazide, loop), NSAIDs, and ACE inhibitors reduce renal lithium clearance and can elevate serum levels even at OTC dosing.

misconceptions

"Trace lithium is psychiatric lithium at low dose." The dosing ratio is roughly 1:60 — OTC supplementation provides 5–10 mg elemental lithium against 600+ mg in bipolar care. The two should not be discussed interchangeably.
"Lithium orotate is proven to prevent dementia." No RCT of lithium orotate for cognitive endpoints has been completed. The Forlenza trials used lithium carbonate titrated to half-therapeutic serum levels, and the Aron mouse data is preclinical. Lithium orotate's evidence base is mechanistic and analogical, not trial-based.
"Drinking-water studies prove population-level effect." They show association; the ecological-study design cannot distinguish lithium-content from co-occurring geographical, socioeconomic, and demographic confounders. The signal is robust across countries — which makes it harder to dismiss — but is not causal evidence.
"OTC lithium is risk-free because it's a supplement." The FDA does not pre-market test supplements; a 2007 case report documented toxicity from a lithium-orotate overdose at serum levels below the bipolar therapeutic range.

population-variability

Drinking-water studies suggest the suicide-prevention effect may be larger in females in some cohorts (Memon 2020 reports a sex-stratified female-specific signal), though other studies find the opposite or no sex difference. Population-baseline lithium intake varies dramatically by geography; supplementation effect is presumably larger in low-baseline populations. Genetic variability in lithium response is well-documented in psychiatric care (only ~30% of bipolar patients are "lithium responders") and likely extends to nutritional doses, though not characterised.

stakes

The stakes argument turns on two endpoints: late-life dementia and suicide. Alzheimer's disease affects roughly 6 million Americans; lifetime risk of suicide death is ~1.5%. A small absolute risk reduction across both, applied to a large addressable population, gives the topic its public-health interest — but the absolute risk reduction is unknown at OTC dosing, and the population-level signal does not straightforwardly translate to individual benefit. Stakes are real but rate-limited by evidence quality.

history

Lithium's mood effect was discovered by John Cade in 1949 in Melbourne — he gave lithium to manic patients as a control after observing calming effects in guinea pigs. Psychiatric lithium became standard care for bipolar disorder by the 1970s. The trace-nutritional thread is older still: spa waters at Lithia Springs (Georgia), Bath, and the lithium-named municipalities had a 19th-century reputation for mood-calming effects, and "lithia waters" were commercial through the early 20th century. 7-Up was originally marketed as "Bib-Label Lithiated Lemon-Lime Soda" (1929) and contained lithium citrate until 1948. Hans Nieper in 1970s Germany proposed lithium orotate as a more brain-bioavailable carrier, beginning the OTC supplement story. Schrauzer's 1990 ecological work revived the trace-nutritional framing in academic discourse.

alternatives

For suicide-prevention at the individual level, the comparator with strong RCT evidence is clozapine (FDA-approved for suicidality in schizophrenia) and, in mood disorders, therapeutic-dose lithium itself. For dementia delay, the comparators with stronger evidence are exercise, sleep adequacy, hearing-loss correction, blood pressure control, and the Mediterranean / MIND diet pattern — each with effect sizes at least as well documented as low-dose lithium. Low-dose lithium is not a replacement for these; it is a low-cost, low-effort additive bet on a mechanism that may complement them.

practicalities

Lithium orotate is widely available at health-food stores and online, $10–30/year at common doses. No prescription required in the U.S. Some other countries (Canada, parts of Europe) regulate it as a drug. Tap-water lithium content is usually disclosed in annual water-quality reports but is often not measured; bottled-water content varies by source. A reader with access to mountain-source mineral water in a high-lithium region may already be at the supplement-equivalent intake.

The credibility range

Optimist case

The optimist points to convergent signals across independent lines of evidence: a robust ecological association between drinking-water lithium and suicide mortality replicated in nine countries (Memon 2020), high-evidence RCT data on therapeutic-dose lithium reducing suicide by ~80% in mood disorders (Cipriani 2013), small but consistent positive RCT data on low-dose lithium delaying cognitive decline in MCI (Forlenza 2011, 2019), a coherent molecular mechanism (GSK-3β inhibition, tau / amyloid attenuation, microglial regulation) demonstrated across cell-culture, mouse-model, and human-tissue work, and as of 2025 a Nature paper identifying lithium as the only metal depleted in MCI cortex and demonstrating that dietary lithium depletion in mice produces Alzheimer-like pathology that supplementation reverses (Aron 2025). Lithium is cheap, widely available, geologically natural in the human food chain, and has a long pre-pharmacological history of mood-calming use. For an intervention with this mechanism strength and a benign expected risk profile at OTC doses, the precautionary bet favours modest supplementation as an additive to standard dementia-risk management.

Skeptic case

The skeptic notes: every population-level suicide finding is ecological and confounded by geography, socioeconomics, and population density that correlate with lithium-rich groundwater independently. The Forlenza MCI trials are small (n ≈ 34 analysed) and from a single research group; they have not been replicated in a large multicenter trial. The dose used in Forlenza (serum 0.25–0.5 mmol/L, achieved with 150–300 mg/day lithium carbonate) is roughly 30× the OTC lithium-orotate dose, so the extrapolation to supplement-grade lithium is conjectural. No RCT of lithium orotate for any endpoint has been completed. The Aron 2025 Nature paper is exciting but a single mouse-and-tissue study with the standard preclinical-to-human translation discount. Therapeutic-dose lithium has a well-characterised long-term harm profile (renal, thyroid, parathyroid); the assumption that risk drops to zero at OTC dose is plausible but not directly tested in long-term human cohorts. A 2007 case report documented toxicity from lithium orotate at unusually low serum levels, hinting at the carrier-pharmacokinetics question. The honest read is "interesting hypothesis, decent mechanism, suggestive epidemiology, no definitive human trial at the supplement dose."

Author's call

The entry lands between optimist and skeptic, closer to the optimist side on plausibility but closer to the skeptic side on confidence. The mechanism is one of the most coherent in nutritional neuroscience and the Aron 2025 Nature paper materially strengthens it. The population-level suicide signal is robust enough across independent ecological studies that pure-chance explanation is implausible, even if the per-individual effect at observed water concentrations is small. The Forlenza MCI trials are the strongest small-RCT signal in this space and survived a four-year extension. But the dose gap between trial and supplement is real, no RCT has tested OTC dosing, and lithium orotate specifically rests on mechanism and analogical extrapolation. The entry's posture: a low-cost, low-risk additive bet that is worth knowing about and worth a reader's deliberate choice. Not a replacement for the high-evidence dementia-prevention interventions (exercise, sleep, hearing, blood pressure) and not a treatment claim. The 2-year+ horizon of any meaningful benefit means it is a long-term bet, not a short-term felt-effect supplement.

Stakeholder and incentive map

Pro-lithium supplement camp: small supplement manufacturers (NCI/Advanced Research, Nutricology, others), naturopathic and integrative-medicine practitioners, longevity-influencer ecosystem (Stanfield, others). Direct commercial incentive at $10–30/year per customer, plus content monetisation. Some of this overlaps with serious researchers (Schrauzer, Forlenza, Aron) who have no apparent commercial incentive.
Mainstream psychiatry and neurology: Cautious by default — lithium-as-supplement collides with the field's hard-won protocols for serum monitoring and renal risk in psychiatric prescribing. Editorial position of major guidelines (APA, NICE) is silent on OTC lithium, neither endorsing nor opposing.
Public-health policy thinkers: A small group has advocated for water-supply lithium fortification (analogous to fluoride). The 2022 PMC review on lithium-as-public-policy-suicide-prevention argues the ecological evidence justifies a feasibility trial of water fortification; no jurisdiction has implemented.
Regulatory: FDA does not approve lithium orotate; it ships as a supplement under DSHEA. EPA includes lithium on the unregulated-contaminants monitoring list but sets no MCL.
Skeptics: Academic skeptics of nutritional-essentiality claims (sparse OTC literature), psychiatric community wary of supplement-displaced psychiatric care.

Population variability

Baseline dietary intake varies ~5× across U.S. regions based on water source (USGS 2024). Bipolar lithium-response genetics include variants in GADL1, SESTD1, and circadian genes; whether these mediate nutritional-dose effects is unknown. Female-specific suicide-protective signals appear in some ecological studies but not all. Genetic and dietary baselines suggest a deficient-baseline population may benefit more than a high-baseline population, but specific cutoffs are unestablished. Elderly readers carry both the largest dementia risk and the largest susceptibility to therapeutic-lithium adverse effects — the cost / benefit reasoning is more delicate at age 70 than at age 40.

Knowledge gaps

No RCT of lithium orotate. The most-discussed supplement form has zero placebo-controlled trials for any endpoint. A two-year RCT in MCI populations is the obvious experiment.
Optimal serum level for neuroprotection. Forlenza used 0.25–0.5 mmol/L; Aron 2025 suggests sub-therapeutic dietary range may be protective. The dose-response below psychiatric range is unmapped in humans.
Long-term safety at OTC dose. No multi-decade cohort of OTC lithium-orotate users exists. Whether the renal, thyroid, and parathyroid risks of therapeutic lithium scale proportionally to dose or have threshold effects is unknown.
Mechanism of the suicide-specific effect. Cipriani's data suggest a specific anti-suicidal effect at therapeutic dose; the same effect at trace dose is hypothesised but unproven, and the molecular mediator is not identified.
Water-supply fortification feasibility. No jurisdiction has run a public-health trial. Ethics, dosing, and population-level effect estimation remain open questions.
Lithium orotate pharmacokinetics in humans. Mouse data hints at carrier-dependent brain accumulation; human pharmacokinetics has not been characterised.

Scoping

The topic brief named three consequences: mood, suicide risk at the population level, and brain aging. All three are covered in the body. The mood and suicide threads are treated as one story — the population-level ecological signal sits inside the evidence section alongside the prescription-dose Cipriani meta-analysis, and the mood pitch carries the combined call. Brain aging is the longevity thread and gets the most weight given the 2025 Aron paper.

Deliberately out of scope:

Full-dose psychiatric lithium for bipolar disorder. Different substance for editorial purposes — different risk profile, different action verb (decide vs respond), different audience.
Public-policy water-supply lithium fortification. Mentioned once in the stakeholder map in the research dossier; not pulled into the article body. Warrants a separate entry under medical or water if the catalogue grows in that direction.
The deeper supplement-quality / sourcing question (which brands, third-party testing, contamination). Treated as one-line practicality; a comparative supplement-vetting entry is a candidate.

Rating difficulties

Mood at 3. The hardest call. The population suicide signal is robust across nine countries (Memon 2020). The Cipriani therapeutic-dose meta is unambiguous. But the per-person felt-mood effect at supplement dose has no trial. Landed on 3 (meaningful, named effect) rather than 4 because the effect at OTC dose is mechanism-extrapolated, not measured.
Longevity at 2. The Forlenza MCI trials and the 2025 Aron Nature paper are the case for going higher; the dose gap between trial and supplement, plus single-research-group provenance for the MCI trials, are the case for staying lower. 2 (real but small contribution) reflects "plausible, suggestive, not yet demonstrated at the dose the reader can buy."
Focus, energy, sleep, beauty, short-term health at 0. No felt-effect data and no mechanism story below the years-time-horizon. Honest zeros — would have inflated to make the page denser, refused.
Evidence at 2. Mechanism + ecological + small-RCT + preclinical Nature paper is convergent but not definitive. The honest ceiling is "well below a 5 (multiple large RCTs at the supplement dose) and above a 1 (mechanism only)."
Controversy at 3. Mainstream psychiatry / neurology silent or cautious; a growing research camp (Schrauzer, Forlenza, Aron) actively advocates. Genuine active debate.
Applicability at 4. Both endpoints (Alzheimer's, suicide) are large-population. The decision audience is broader than the user audience.

Hard calls during the write

Action = decide, not do. The evidence does not support a confident "everyone should take this" recommendation; nor is the evidence weak enough to recommend against. decide is the honest verb.
Cadence = once. The decision is the action. If a reader decides yes, the resulting daily-pill habit is downstream of the decision; the entry's primary recurrence is the choice itself.
No dream-narrative crank on the dek and tagline. Score lands at roughly 23 — below the 40 threshold where the dream cranks the dek. Wrote a brief dream narrative because the relief/clarity lever is real, but the dek and tagline are written straight. The lift the entry earns is specificity, not bold language.
Felt-experience leads, even in a low-felt-effect entry. Article §1 wants felt-experience anchoring, but lithium at OTC dose has essentially no felt effect on a weekly horizon. Worked around by anchoring on the reader's situated decision ("are you in the high-water-lithium bin or not"), not on a felt body change.

Future links

An eventual therapeutic-lithium-bipolar entry under mental, cross-link from contraindications.
An eventual water-filtration-tradeoffs entry under home or water — reverse osmosis strips lithium, plus many other trace minerals.
An eventual dementia-prevention-stack entry under medical or mental — the alternatives section should hand off there.
If gsk3-inhibition or amyloid-pathway mechanism-page entries ever exist, this entry's mechanism section should link to them.

Separate-entry candidates

Lithia waters and the public-water-supply lithium fortification debate is a substantial topic in its own right (public-policy lens). Flag for the backlog.
The 2025 Aron Nature paper has implications beyond supplementation — it suggests dietary lithium status may be a modifiable risk factor for Alzheimer's. If the LATTICE trial reads out positively in 2026–2027, this entry will likely need a substantial rewrite and a possible split.