Lithium used to be used in 7-Up as an upper. Then, it became the workhorse mood stabilizer for mood disorders. Now, it seems to be getting a third act as a plausible player in the Alzheimer’s disease (AD)story.
A few lines of evidence suggest this. First, we’ve got long term, small randomized clinical trials in older adults suggesting low doses of lithium carbonate can slow cognitive decline and nudge biomarkers in the right direction1–3. Second, we’ve got a new Nature paper reporting early reductions in brain lithium availability that proposes amyloid binding as a lithium sink, and showed serious improvements in aged AD-model mice using lithium orotate4. The third line comes from large registries of bipolar disorder that find lower subsequent dementia incidence among patients treated with lithium compared to those not on lithium or on other mood stabilizers5. A fourth line comes from large sample environmental epidemiology work showing those living in areas with higher ground levels of lithium had lower dementia rates than those in areas with lower ground levels6.
Because many readers may be asking “which kind of lithium is best?”, and because I’m not a doctor, I won’t be making any claims of which is better for people to take. That said, the current human outcomes evidence sits with low-dose lithium carbonate under clinical monitoring. Lithium orotate is mechanistically promising in mice, but we have no human AD efficiency data yet. And we’ve all seen mouse studies that looked at some drug, and were promising, but turned out to have no real impact when applied to humans.
What we know about lithium and AD in humans
Some clinical trials give hope here. A two-year, double-blinded placebo controlled RCT in older adults with mild cognitive impairment showed patients treated with lithium to achieve subtherapeutic levels (0.25-0.5 mEq/L) remained stable on cognitive measures across the two-year period, showing better performance on memory and attention tests. The placebo group had the expected decline seen in MCI patients2. The lithium group also showed changes in the AD biomarkers in their cerebrospinal fluid. Similarly, a randomized controlled study of daily lithium microdoses of 300 mg for 15 months showed no decrease in cognitive performance while the control group had the expected decline1. A recent meta-analysis on therapeutic lithium studies reported lower dementia risk and prevalence overall, while noting heterogeneity and differing study qualities to temper any strong claims3.
Treatment cohorts in bipolar populations
A nationwide registry study of people with bipolar disorder showed maintenance lithium treatment was associated with a lower subsequent incidence of diagnosed dementia when compared with non-lithium mood stabilizers or no lithium7. But it’s a bit more nuanced than “taking lithium reduced risk of dementia.” Continued lithium treatment seemed to reduce dementia back to levels of that in the general population compared to those who did not continue treatment. This is observational data that is consistent with the trials but is vulnerable to things like confounding by indication and adherence.
Lithium in water
A study out of Denmark linked where people lived over time to how much lithium was in municipal drinking water sources6. Big sample, careful linkage, and a result that isn’t a neat straight line. Compared to 2–5 µg/L, dementia rates were a bit higher at ~5–10 µg/L, flat around ~10–15 µg/L, and lower by about 17% for those in areas above ~15 µg/L. That same pattern showed up for Alzheimer’s and vascular dementia, and it stuck around after adjusting for urban vs. rural. One point people miss: the high-lithium East (Copenhagen region) also has more health care access; if anything that should mean more dementia being diagnosed there, not less. So the “it’s just better doctors” explanation cuts the wrong way for the high-lithium end. Still, this is a map and not a clinical trial. Everyone in a municipality gets the same number; individual income, education, and comorbidities can’t be fully untangled. It’s the best observational anchor we’ve got in the form of a strong signal, though still not causal6,8,9.
The proposed biology (in short)
A new paper in Nature by Aron and colleagues combined mouse models and post-mortem brain and serum data from humans to argue that the Alzheimer’s brain is short on available lithium, partly because amyloid binds it like a sponge. In aged AD-model mice, giving lithium (in the form of lithium orotate) improved pathology and behavior. In people, lithium was uniquely reduced in the prefrontal cortex in MCI and AD, while serum lithium didn’t differ from controls, consistent with a brain-availability problem. The mice also showed that dietary lithium depletion worsened amyloid, phospho-tau, inflammation, synaptic/axonal/myelin loss, and memory. Repletion with lithium orotate (which was chosen to evade that amyloid binding) reversed the pathology and rescued memory. This seems to lend some more credence to amyloid being a
Which lithium, practically speaking
The human outcomes data live with low-dose lithium carbonate under clinical monitoring1–3. Lithium orotate looks good in mice and shows brain uptake in healthy volunteers by 7Li-MRI, but there are no human AD efficacy data for orotate4,10.
What I’d like to see funded
A multi-site, adequately powered low-dose lithium carbonate vs lithium orotate vs placebo RCT in MCI plus fluid/MRI/PET endpoints over 24–36 months.
Where I land (for now)
Promising biology meets small, encouraging human trials. If lithium helps here, it’s probably via restoring availability, not a vitamin-style quick fix. Until those trials report, the only defensible, practical answer to “which lithium?” is carbonate, low-dose, monitored in a research or clinical setting. At the doses tested for MCI/AD, lithium’s short-term AE profile looked like placebo, and the long-term signal was mostly thyroid/weight/arrhythmia/diabetes (stuff geriatricians already monitor) with no renal hit in that cohort11. That’s “feasible under supervision,” not a zero-risk supplement.
The Edge of Epidemiology is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Citations
1. Nunes MA, Viel TA, Buck HS. Microdose Lithium Treatment Stabilized Cognitive Impairment in Patients with Alzheimer’s Disease. Curr Alzheimer Res. 10(1):104-107. doi:10.2174/1567205011310010014
2. Forlenza OV, Radanovic M, Talib LL, Gattaz WF. Clinical and biological effects of long-term lithium treatment in older adults with amnestic mild cognitive impairment: randomised clinical trial. Br J Psychiatry. 2019;215(5):668-674. doi:10.1192/bjp.2019.76
3. Lu Q, Lv H, Liu X, Zang L, Zhang Y, Meng Q. Lithium Therapy’s Potential to Lower Dementia Risk and the Prevalence of Alzheimer’s Disease: A Meta-Analysis. Eur Neurol. 2024;87(2):93-104. doi:10.1159/000538846
4. Aron L, Ngian ZK, Qiu C, et al. Lithium deficiency and the onset of Alzheimer’s disease. Nature. Published online August 6, 2025:1-10. doi:10.1038/s41586-025-09335-x
5. Burdick KE, Millett CE, Russo M, et al. The association between lithium use and neurocognitive performance in patients with bipolar disorder. Neuropsychopharmacology. 2020;45(10):1743-1749. doi:10.1038/s41386-020-0683-2
6. Kessing LV, Gerds TA, Knudsen NN, et al. Association of Lithium in Drinking Water With the Incidence of Dementia. JAMA Psychiatry. 2017;74(10):1005. doi:10.1001/jamapsychiatry.2017.2362
7. Kessing LV, Søndergård L, Forman JL, Andersen PK. Lithium Treatment and Risk of Dementia. Arch Gen Psychiatry. 2008;65(11):1331-1335. doi:10.1001/archpsyc.65.11.1331
8. Fraiha-Pegado J, de Paula VJR, Alotaibi T, Forlenza O, Hajek T. Trace lithium levels in drinking water and risk of dementia: a systematic review. Int J Bipolar Disord. 2024;12(1):32. doi:10.1186/s40345-024-00348-5
9. Iordache AM, Voica C, Roba C, Nechita C. Lithium Content and Its Nutritional Beneficence, Dietary Intake, and Impact on Human Health in Edibles from the Romanian Market. Foods. 2024;13(4):592. doi:10.3390/foods13040592
10. Neal MA, Strawbridge R, Wing VC, Cousins DA, Thelwall PE. Human brain 7Li-MRI following low-dose lithium dietary supplementation in healthy participants. J Affect Disord. 2024;360:139-145. doi:10.1016/j.jad.2024.05.128
11. Aprahamian I, Santos FS, dos Santos B, et al. Long-term, low-dose lithium treatment does not impair renal function in the elderly: a 2-year randomized, placebo-controlled trial followed by single-blind extension. J Clin Psychiatry. 2014;75(7):e672-678. doi:10.4088/JCP.13m08741
