Celeste McGovern investigates new research asking whether a deficiency in this mineral could leave us vulnerable to everything from mood disorders to Alzheimer’s
When most of us think of lithium, the soft, silvery metal that powers our phones and electric car batteries may come to mind. Some might also know it as a potent psychiatric drug given in high doses to control the extreme mood swings of severe bipolar disorder.
Now, in the wake of recent groundbreaking research on lithium that occurs naturally in the human brain, an extraordinary clinical trial is launching to see if it’s a key to the onset of Alzheimer’s dementia. Researchers at Johns Hopkins are preparing to give tiny doses to people with early Alzheimer’s to see whether it reaches the central nervous system—and what happens to biomarkers of the disease when it does.1
The trial follows a landmark study that took 10 years to complete. Harvard researchers measured 27 different metals in postmortem brain tissue from people who had been cognitively healthy, had mild cognitive impairment (MCI, often a precursor to Alzheimer’s) or had died with the disease.
Of those metals, only one stood out: lithium. Levels were already lower in the brains of people with MCI and lower still in those with Alzheimer’s. What’s more, the amyloid plaques that accumulate in some Alzheimer’s brains appeared to act like a sponge, trapping lithium and making it unavailable to brain cells.
It suggested a whole new way of looking at Alzheimer’s disease. What if lithium depletion isn’t a consequence of the disease but a driver?
The Harvard team put that question to the test in mice. When they experimentally depleted naturally occurring lithium from the animals’ brains, inflammation worsened, and the amyloid plaques and tau tangles often associated with Alzheimer’s increased. Synapses were lost. Memory spiraled.
Then came the bombshell. When they gave the mice tiny doses of lithium orotate, a lithium salt dismissed by the bulk of mainstream practitioners and strictly prohibited by some health regulators, it prevented pathological changes and memory loss in animals at risk of developing the disease. What’s more, in mice with well-entrenched Alzheimer’s-like pathology, lithium reversed much of that pathology and restored memory.2
“The idea that lithium deficiency could be a cause of Alzheimer’s disease is new and suggests a different therapeutic approach,” said Bruce Yankner, professor of genetics and neurology at Harvard Medical School and senior author of the study.
Perhaps even more radical is the implication about lithium itself. “Lithium turns out to be like other nutrients we get from the environment, such as iron and vitamin C,” Yankner said. “It’s the first time anyone’s shown it exists at a natural level that’s biologically meaningful without giving it as a drug.”
No health authority recognizes lithium as an essential human nutrient. Yet the Harvard research has shown lithium isn’t just a drug that changes the brain. It’s already there, naturally, doing important work. That’s made clues uncovered over decades much harder to dismiss.
Traces of lithium in drinking water have been linked to lower rates of suicide and dementia, while research has implicated lithium in brain-cell survival, inflammation, neuroplasticity, gene expression, mitochondrial energy, and even the birth and death of neurons.
Studies have explored its effects in cognitive decline and Parkinson’s disease as well as mood, aggression, impulsivity, addiction, anxiety and depression. But these signals that it was a critical player in brain health were overlooked for decades, eclipsed by its other identity as a toxic metal.
For clinicians like functional psychiatrist James Greenblatt, the concept is hardly new. He has been prescribing tiny doses of lithium orotate for more than 35 years to treat problems ranging from irritability, impulsivity and depression to cognitive decline. For him, the Harvard findings offered some remarkable scientific backup for an idea he has pursued for decades.
“I don’t say, ‘I told you so,’” Dr Greenblatt says, although he clearly appreciates the vindication.
Greenblatt wasn’t alone. Long before Harvard scientists went looking for lithium in the brain, researchers and physicians wondered whether the tiny amounts in food and water affect the mind. Indeed, humans had been consuming lithium medicinally for centuries before anyone even knew what it was (see below).
What Greenblatt means by “low-dose lithium,” however, is very different from what most psychiatrists prescribe today. Lithium has been one of psychiatry’s most potent—and unusual—drugs for more than 75 years. And psychiatry stumbled on it by accident.
In the 1940s, Australian psychiatrist John Cade was investigating whether a toxin related to uric acid might cause mania. He injected guinea pigs with lithium urate, using lithium to make the uric acid soluble. Unexpectedly, the animals became quiet and lethargic.
He used the lithium salts alone and realized that lithium—not uric acid—was behind the calming effect. After testing lithium carbonate on himself, Cade gave lithium citrate to a 51-year-old man who had been chronically manic for five years and was considered the “most troublesome patient in the ward.”
Within days, the patient began settling; after three months, he was well enough to leave the hospital and return to his former job. When he later stopped lithium, his mania returned. He resumed it and recovered again.
Cade treated nine more patients with mania, reporting improvement in all of them, and published his findings in 1949.3
Lithium carbonate is still a gold-standard treatment for bipolar disorder today. It’s also one of the few psychiatric drugs shown to substantially reduce the risk of suicide.
But conventional lithium therapy uses very high prescription doses, typically 600–1,200 mg a day, equivalent to 113–225 mg of elemental lithium, dosed according to concentrations in the blood. That’s hundreds of times above typical amounts in food and water.
Common side effects include tremor, thirst, frequent urination and weight gain. If lithium levels rise too far, the consequences can be much more serious.
Hundreds of case reports in the medical literature describe lithium toxicity involving confusion and severe neurological symptoms, kidney injury, seizures, coma and, in extreme cases, death. That’s why patients taking conventional lithium require regular blood tests and monitoring of kidney and thyroid function.
It’s the main reason lithium has largely been avoided—even far below prescription levels. Doctors and regulators are afraid of it, which Greenblatt says is like fearing sodium or potassium. “Even sodium is toxic at high levels,” he says.
But it’s also essential to life, and the brain and nervous system can’t function without it. The dose makes all the difference.
For Greenblatt, the question isn’t whether lithium can be toxic; 75 years of psychiatric experience have established that it can. The question is what happens at the other end of the dose spectrum.
Can an element be toxic at pharmaceutical doses yet beneficial—even necessary—at nutritional ones? There’s now serious scientific debate about whether lithium should be considered a micronutrient, a trace mineral required for optimal health, of which many of us might have suboptimal or even deficient levels.
In The Conspiracy Against Lithium: The Suppressed Essential Nutrient and Its Benefits for Mental Health (Skyhorse, 2025), German physician and molecular geneticist Michael Nehls argues that lithium deficiency is widespread. It may be an overlooked contributor to depression, anxiety, rising crime rates, suicide and neurodegenerative disease.
It’s a radical claim considering lithium’s near-complete neglect by public health agencies. There is no established lithium deficiency disease or recommended dietary amount.
But Nehls didn’t invent the idea. More than two decades ago, Gerhard Schrauzer, a chemist and professor at the University of California, San Diego, reviewed an already substantial body of animal and environmental evidence and argued that lithium should be considered an essential trace element. In 2002, he proposed a provisional daily requirement of 1 mg for an average 154 lb (70 kg) adult.
Some of the strangest evidence came from research on animals beginning in the 1970s. In one study, rats fed a lithium-deficient diet died younger and had reproductive problems and behavior changes. They became noticeably lethargic and lost their normal daily rhythms, running on their exercise wheels at odd times.
Goats fed a lithium-deficient diet had lower conception rates, more miscarriages and offspring with lower birth weights. They also had smaller spleens and signs of inflammation and blood vessel damage, which sometimes reversed when tiny amounts of lithium were restored to their diets.4
This didn’t prove that lithium was an essential nutrient for humans, but it did raise questions about the effects of differing amounts.
As it happened, nature had already provided an enormous human experiment. Lithium occurs naturally in rocks and soil and leaches into groundwater, so amounts in drinking water vary enormously.
In 1990, Schrauzer and colleague Krishna Shrestha compared drinking-water lithium levels of 27 Texas counties with a decade’s worth of mortality, mental health and crime statistics. What they found was extraordinary.
Counties with more lithium in the water had significantly lower rates of suicide, homicide and rape than those with the lowest levels. Robbery and burglary rates were also lower, as were arrests for drug possession and drunk driving.
These populations weren’t being “medicated” with lithium. The concentrations in the higher-lithium counties were just 0.07–0.17 mg/L, hundreds of times below the psychiatric dose.
“These results suggest that lithium at low dosage levels has a generally beneficial effect on human behavior, which may be associated with the functions of lithium as a nutritionally essential trace element,” Schrauzer and Shrestha concluded.5
The study compared populations, not individuals, and couldn’t account for how much water any particular resident drank or other differences. But the correlation was strong enough for other researchers to follow. Studies in Japan, Austria, Greece, the US and elsewhere ensued.
In 2020, two independent systematic reviews and a meta-analysis reached broadly the same conclusion: Higher naturally occurring lithium in drinking water, lower suicide mortality. Once again, the quantities involved were measured in mere micrograms per liter.
The findings were intriguing enough for researchers at the University of Sussex and Brighton and Sussex Medical School, authors of one of the reviews, to propose randomized community trials adding trace lithium to water supplies in places with high rates of mental illness, violent crime, drug abuse and suicide.6
Meanwhile, other researchers had begun asking whether lithium’s apparent association with brain health extended beyond mood and behavior.
In 2017, Danish investigators used national health records to compare long-term lithium exposure in drinking water among more than 73,000 people diagnosed with dementia and some 733,000 controls. The relationship wasn’t linear; intermediate exposure was actually associated with a slightly higher dementia rate. But people with the highest exposure—more than 15 mcg/L (0.015 mg/L)—had a lower incidence of dementia than those with the lowest.7
And the associations may not stop at the brain. In 2025, a major study looked at more than 250,000 people in the US, matching where they lived with estimates of lithium in groundwater. The result was striking.
People in the highest fifth of estimated lithium exposure had a 71 percent lower risk of cancer than those in the lowest fifth. The association persisted in men and women, among long-term residents, in both eastern and western states, and across all the cancer types the researchers examined.8
A peculiar pattern was emerging. From violence and suicide to dementia and now even cancer, quantities of lithium hundreds of thousands of times below conventional psychiatric doses kept turning up in population studies relating to human health.
Why? The water studies couldn’t answer that question. What was needed was evidence that giving small amounts of lithium directly to people could change the course of disease.
Researchers in Brazil were already on it. In 2011, a team at the University of São Paulo reported the results of a randomized, double-blind, placebo-controlled trial in 45 older adults with memory-affecting MCI—the same early stage of cognitive decline in which the Harvard researchers would later find lithium already depleted in the brain.
For a year, participants received either lithium or placebo. But rather than using the higher blood levels conventionally targeted in bipolar disorder (0.6–1.0 mmol/L and higher for maintenance treatment), the researchers aimed for much lower levels of 0.25–0.5 mmol/L.
After 12 months, the lithium group showed better memory and attention and a significant reduction in an abnormal form of tau tangles.9 It was a small trial, but the findings pointed to lithium’s measurable effects on cognition and on a biological process linked to Alzheimer’s.
And the Brazilian researchers followed up with a longer trial, in which 61 older adults with MCI received either lithium or placebo for two years. Lithium was adjusted to maintain the same relatively low blood concentrations as before.
The differences became more evident the longer people took lithium. Those receiving it remained more cognitively and functionally stable, while those taking placebo declined, and fewer lithium-treated participants progressed to Alzheimer’s. The researchers concluded that long-term, low-dose lithium might have disease-modifying potential—slowing the underlying process rather than simply improving symptoms.10
Another Brazilian team, however, took the low-dose idea much further. They tested a microdose of just 0.3 mg of lithium daily on Alzheimer’s patients for 15 months.
The results were remarkable. People receiving the lithium showed no decline on the Mini-Mental State Examination (MMSE), a standard test of cognitive function. Those in the control group, meanwhile, declined significantly, and differences between the groups started within just three months and increased progressively.
“This data suggests the efficacy of a microdose lithium treatment in preventing cognitive loss, reinforcing its therapeutic potential to treat Alzheimer’s disease using very low doses,” the researchers concluded.11
It was a small trial, and no large clinical trial has been conducted to confirm its extraordinary result. But after decades and billions spent by the drug industry to prevent the relentless cognitive decline in Alzheimer’s, it raises promising questions.
Unlike a drug designed to hit a single molecular target, lithium appears to influence a range of basic processes involved in keeping neurons healthy:
“These are not drug effects,” says Dr Greenblatt. “They’re nutrient effects—the same kind of basic biological support we see with magnesium, zinc, omega-3 fatty acids and B vitamins.”
Rather than acting on a particular disease, lithium is acting as a basic nutrient. It’s helping the brain protect itself and repair damage. That explains why old research on lithium (most of it at high psychiatric doses) sees its impact on so many conditions and it has current researchers looking beyond Alzheimer’s and bipolar disease.
One of them is neurologist Thomas Guttuso at the University at Buffalo, author of The Promise of Lithium (Lioncrest Publishing, 2023). His interest in whether low-dose lithium might slow the progression of Parkinson’s disease began almost by accident.
In 2014, one of his patients with Parkinson’s developed bipolar disorder, and his psychiatrist prescribed a relatively low dose of lithium. To Guttuso’s surprise, the patient’s Parkinson’s symptoms improved. He had a blood lithium level of just 0.37 mmol/L, well below the usual 0.6–1.0 mmol/L target for treating bipolar disorder.
Guttuso started giving lithium to other patients. “It was right around there that I was finding just anecdotally that it was helping with patient symptoms,” Guttuso recalls. Then he realized there was a bigger possibility.
“A much more important potential use was disease modification, to protect brain cells from dying and slow the progression of the disease,” he explains. “That’s the big unmet need in Parkinson’s, to improve people’s long-term prognosis. And we don’t have anything that’s been proven to do that.”
Guttuso started focusing on biomarkers of brain cell damage and degeneration to see if lithium could have positive effects and, if so, what dose worked best. He and his colleagues measured neurofilament light (NfL), a protein released into the blood when neurons are damaged or die, which is increasingly regarded as a marker of disease progression in Parkinson’s.
After 24 weeks of treatment, patients with the highest blood lithium levels had a median 12.8 percent fall in NfL. Those with intermediate levels had a 2 percent reduction, while NfL increased 11.2 percent in those with the lowest levels.
The study involved just 28 patients from two trials, and the groups were divided according to the lithium levels they achieved rather than randomly assigned, so it can’t prove that lithium slowed Parkinson’s. But Guttuso says it’s the first report of any therapy associated with a significant reduction in serum NfL in Parkinson’s.12
An earlier study by Guttuso’s team had randomized 16 Parkinson’s patients among lithium carbonate and two doses of lithium aspartate. The 45 mg/day lithium aspartate group showed the largest and most consistent improvements in the biomarkers studied, although only four patients received that dose—a very small sample.
Guttuso draws a comparison to bipolar disorder treatment. Just as psychiatric lithium doesn’t work below certain blood levels, the neuroprotective effects may depend on getting enough lithium into circulation. Yet his apparent sweet spot is substantially lower: His extension study is targeting 0.25–0.50 mmol/L, aiming for 0.45.
Levels above 5.0 mmol/L produce side effects, including nausea and confusion, which disappear when the dose is lowered. “So, the dosing is so important,” says Guttuso. “It may be why some trials fail.”
And one of the newest and most rigorous human trials might be a case in point. The LATTICE trial followed 80 adults over age 60 with MCI for up to two years, randomly assigning them to lithium carbonate averaging 195 mg a day—about 37 mg of elemental lithium—or placebo.
On the study’s six primary measures of cognition, brain imaging and Alzheimer’s biomarkers, lithium failed to produce a statistically significant benefit, but the dosing produced an average blood level of just 0.17 mmol/L, well below the Parkinson’s study target of 0.45 mmol/L.
On the upside, verbal memory declined at roughly half the rate in the lithium group as in the placebo, but the researchers said a larger trial was needed to confirm the effect.13
Dose may be only part of the puzzle. Timing may matter just as much.
“A lot of brain damage has already occurred, and there’s an aggressive disease process in full force in the background,” says Guttuso. “You’ll probably need a higher amount of lithium to slow that locomotive down after it’s already in motion.”
In the water studies, people living in higher-lithium areas didn’t just begin drinking it at age 75, after receiving an Alzheimer’s diagnosis. “You’re being exposed to that from infancy,” Guttuso says. “That’s a completely different story.”
For now, no one knows how much lithium a healthy brain needs or whether lithium deficiency will eventually become something doctors test for and treat. But after decades of research pointing in that direction, Harvard’s discovery has transformed lithium deficiency from a fringe idea into one mainstream science can no longer easily ignore.
Psychiatrist James Greenblatt, author of The Lithium Revolution (FriesenPress, 2026) has prescribed low-dose lithium orotate for more than 35 years and now trains other practitioners in its use through Psychiatry Redefined (psychiatryredefined.org), the integrative psychiatry education organization he founded. Here’s what he recommends.
For healthy adults: 1–2 mg elemental lithium daily for general health and emotional well-being.
After age 40: 2–10 mg daily, which may support brain health with aging.
For symptoms: 5–40 mg daily, sometimes divided through the day. Irritability is one of the strongest clinical indications. Low-dose lithium may also ease mood instability, impulsivity and addiction.
Family history matters: Repeated suicide, addiction, violence, depression or severe mood problems among close relatives make Greenblatt more likely to consider lithium and may influence his starting dose.
Dr Greenblatt offers the following guidelines for lithium supplementation.
Start low and watch the response. Some people notice changes on only 1–2 mg; others require more. If you become excessively calm, “flat” or sleepy, reduce the dose or move it to bedtime. Medical supervision is recommended above 10 mg/day.
Who should be cautious? Check thyroid function before supplementation and monitor it when appropriate. People who have advanced kidney disease and women who are pregnant or breastfeeding should not take lithium.
What about testing? There is no accepted clinical test for nutritional lithium deficiency. Greenblatt primarily judges need and response from symptoms and family history. Hair tissue mineral analysis is a starting point because lithium concentrates in hair. In as many as 75 percent of the tests Greenblatt orders today, hair shows undetectable levels, indicating a deficiency.
Where can you get personalized help? Greenblatt is co-founder and chief medical officer of Finally Living Now (finallylivingnow.com), a six-month program offering individualized assessment and treatment. It can include nutritional and metabolic assessment, laboratory testing, supplements and conventional medication management where appropriate.
The Romans may have been taking lithium for mental health nearly 2,000 years before anyone knew what lithium was. Second-century physician Soranus of Ephesus recommended certain alkaline spring waters for people suffering from mania and melancholy, and centuries later they were found to contain high levels of lithium.
By the 19th century, people were retreating to lithium-rich spa waters in England, Germany and elsewhere, drinking and bathing in it for its promised health benefits. Bottled “lithia water” became a popular tonic.
When the soft drink 7Up launched in 1929, it contained lithium citrate and was marketed as an uplifting drink and hangover remedy. “It takes the ‘ouch’ out of ‘grouch’” one 7Up marketing slogan said before lithium was taken out of the soda in 1948. Decades of research since then suggest there may have been a grain of truth to the claim.1
What do you think? Start a conversation over on the... WDDTY Community
Great news! Discounts are available based on your cart total. Check the details below:
