Scientists Confirm Inverse Correlation between Lithium and Alzheimers

Recent Harvard Medical School research, published on August 6, 2025, in the journal Nature, has identified lithium deficiency in the brain as a potential early trigger for Alzheimer’s disease (AD). The study, led by Bruce Yankner and colleagues, analyzed human brain tissue from individuals with varying cognitive health, including those with mild cognitive impairment (MCI, a precursor to AD) and full AD. Among 27 metals examined, lithium was the only one significantly reduced in MCI and AD brains. Researchers found that amyloid-beta plaques—one of the hallmarks of AD—bind to lithium, reducing its bioavailability and accelerating neurodegenerative changes.

To explore this further, the team created lithium-deficient diets for wild-type mice and AD mouse models, reducing cortical lithium levels by about 50%. This led to:

• Increased amyloid-beta deposition and phospho-tau accumulation (another key AD pathology).
• Pro-inflammatory microglial activation.
• Loss of synapses, axons, and myelin.
• Accelerated cognitive decline, including impaired learning and memory.

These effects were partly mediated by activation of the kinase GSK3β, which is implicated in neurodegeneration. Notably, the study showed that lithium plays an essential role in normal brain function, shielding against neurodegeneration and supporting all major brain cell types.

For treatment potential, the researchers tested a novel lithium compound (lithium orotate) designed to avoid sequestration by amyloid plaques. In AD mouse models:

• It restored physiological lithium levels at low, non-toxic doses (far below those used for bipolar disorder).
• Reduced amyloid plaque burden by up to 70%.
• Reversed pathology, including tau hyperphosphorylation and inflammation.
• Restored memory and cognitive function, even in advanced stages.

This suggests lithium could address multiple AD pathways (amyloid, tau, inflammation, and gene expression) rather than targeting just one, as current drugs like anti-amyloid antibodies do. The findings build on prior observations, such as a 2017 Danish study linking higher lithium in drinking water to lower dementia rates.

While promising, this is primarily based on mouse models and post-mortem human analyses—human clinical trials are needed to confirm safety and efficacy. Routine blood tests could potentially screen for low lithium to identify at-risk individuals for early intervention. If validated, low-dose lithium supplements might offer a cheap, accessible way to prevent or delay AD onset, affecting over 55 million people worldwide. No other Harvard-specific studies on lithium for AD from August 2025 were identified in the searches.

Natural Sources

Lithium is found in trace amounts in various natural sources, primarily in water, soil, and certain foods. Based on available data, here are the main natural sources of lithium:

• Drinking Water: Lithium occurs naturally in groundwater and surface water, with concentrations varying by region due to geological factors. For example, studies have noted lithium levels in drinking water ranging from 0.1 to 70 µg/L in various global locations, with higher concentrations in areas like northern Argentina, Chile, or the western United States due to lithium-rich mineral deposits. A 2017 Danish study linked higher lithium levels in drinking water (above 10 µg/L) to lower dementia rates, suggesting potential health benefits.
• Plant-Based Foods: Certain plants absorb lithium from soil, though levels are generally low. Foods with detectable lithium include:
• Grains and Cereals: Wheat, oats, and rice may contain 0.4–1.5 µg/g of lithium, depending on soil conditions.
• Vegetables: Potatoes, tomatoes, and leafy greens like spinach can have trace amounts (0.1–0.5 µg/g), with levels varying by region and soil lithium content.
• Legumes: Beans and lentils may contain small amounts (around 0.2–0.6 µg/g).
• Nuts and Seeds: Limited data suggest almonds and sunflower seeds may have trace lithium, though exact amounts depend on cultivation conditions.
• Animal-Based Foods: Lithium is present in low levels in animal products, as animals consume lithium-containing plants or water:
• Meat: Beef and poultry may contain 0.1–0.3 µg/g.
• Dairy: Milk and eggs have minimal amounts, typically less than 0.1 µg/g.
• Fish and Seafood: Some fish, like mackerel, may have slightly higher lithium due to water exposure, but data is sparse (around 0.2–0.5 µg/g).
• Mineral-Rich Soils and Rocks: Lithium is naturally present in certain geological formations, like pegmatite rocks or brine deposits, which can influence local water and food sources. Areas with lithium-rich soils (e.g., parts of Nevada or the Andes) may have higher lithium in crops and water.
• Hot Springs and Mineral Waters: Some natural springs, particularly in volcanic regions, have elevated lithium levels (up to 100–200 µg/L in certain cases), historically consumed for perceived health benefits.

Note: Dietary lithium intake is typically low (estimated at 0.6–3.1 mg/day in most populations) and varies widely based on geography, soil composition, and water sources. While these sources contribute to lithium intake, they are generally insufficient to reach therapeutic levels (e.g., for Alzheimer’s prevention as suggested by the Harvard study), which may require supplementation under medical supervision. Always consult a healthcare provider before considering lithium supplements, as excessive intake can be toxic.

Correlation Evidence

Research on the correlation between Alzheimer’s disease (AD) prevalence and lithium-rich drinking water primarily stems from epidemiological studies examining trace levels of lithium naturally occurring in water supplies. These levels vary geographically due to soil and rock composition, typically ranging from <1 µg/L to over 100 µg/L in some areas. While not all studies agree, a majority suggest an inverse correlation: higher lithium concentrations in drinking water are associated with lower rates of dementia (including AD) incidence or mortality. However, the relationship is often nonlinear, and some findings indicate no association at very low levels. Below is a summary of key evidence.

Studies Showing an Inverse Correlation

• A 2017 Danish nationwide study of over 73,000 individuals with dementia and 733,000 controls found that long-term exposure to higher lithium levels in drinking water (above ~10 µg/L) was linked to a lower incidence of dementia, including AD and vascular dementia subtypes. The association was nonlinear, with the lowest dementia rates at the highest lithium exposures (15–27 µg/L) and elevated rates at intermediate levels (5–10 µg/L). This study used individualized data on residence and water measurements, adjusting for factors like age and sex.
• A 2017 Texas study across 234 counties analyzed changes in age-adjusted AD mortality rates from 2000–2015 alongside lithium levels in public water supplies (mean ~0.002–0.056 mg/L). It reported that higher trace lithium was associated with slower increases in AD mortality, suggesting a protective effect.
• A 2022 Japanese epidemiological study covering 808 regions (91% of the population) found that lithium levels in drinking water (collected from 988 samples) correlated inversely with AD prevalence, based on health insurance claims data. The effect was more pronounced in women and at concentrations around 0.002–0.056 mg/L.
• Recent Harvard research (2025) on brain lithium deficiency in AD patients aligned with these population trends, noting that higher environmental lithium, including in drinking water, tracks with lower dementia rates in observational data.
• A 2024 systematic review of five studies (including the Danish, Texas, and Japanese ones) concluded that trace lithium in water (as low as 0.002 mg/L) is associated with reduced dementia incidence or mortality in most cases, though levels below 0.002 mg/L showed no benefit. The review emphasized potential neuroprotective effects at microdoses, warranting further trials.

Studies Showing No or Mixed Correlation

• A 2023 Scottish study examined lithium exposure up to 9.19 µg/L in drinking water and found no reduced rate of all-cause dementia, contrasting with findings from higher-exposure regions.
• Some analyses, like a 2018 University of Chicago review, reported no link between lithium in tap water and dementia risk, based on earlier data. Critics note that these null findings may stem from lower average lithium levels or unaccounted variables like healthcare access, socioeconomic factors, or dietary lithium intake.

Overall Insights and Limitations

The evidence points to a potential protective role for lithium at low environmental levels, possibly by modulating pathways like GSK3β kinase activity, reducing amyloid-beta and tau accumulation, or supporting brain cell function. However, correlations do not prove causation, and studies are observational, relying on ecological or registry data without controlling for all confounders (e.g., bottled water use, migration, or other environmental factors). Nonlinear patterns suggest a “sweet spot” for lithium concentrations, with benefits tapering at very low or intermediate levels. Related research on therapeutic lithium use (higher doses) also shows reduced dementia risk in bipolar patients, supporting the hypothesis. Clinical trials are needed to test if microdose lithium supplementation could prevent or delay AD, as natural water sources alone may not provide sufficient levels in low-lithium areas. Consult a healthcare provider before considering any lithium intake, as excess can be toxic.