Written 9/19/25; updated 09/23/25; reviewed 2/26/26

Can low doses of lithium prevent dementia? Quite possibly. After a brief history of this idea, I’ll show you some key recent research. Then an even bigger question: can a little lithium reverse a dementia that’s already happening? Maybe. Have a look.

A brief history

Lithium helps neurons grow, make connections and survive stresses. So, might it help in conditions where neurons shrink and die, like Alzheimer’s?

In early experiments, lithium showed no benefit compared to placebo controls. But maybe it was started too late. Perhaps if it was started earlier, it could prevent Alzheimer’s changes? (development of plaques and tangles)

This start-earlier approach has been tested in 2 small studies. In both, people on lithium had less cognitive decline than those on placebo — but in a total of less than 200 people. Compare a single recent drug company study to slow Alzheimer’s: it had 1700 participants.

So it was easy for skeptics to say “we just don’t have enough evidence.” But then came a key study, published in a top journal, Nature. Most of the work was done in mice, so the skeptics will still say “we need studies in humans”. True, we do. But once you see the results of the mouse studies, you’ll see why I think this Nature paper opens the door to consideration of lithium for dementia prevention, now.

The key new experiments: Nature, September 2025

This important report describes many separate experiments in mice that have been bred to be “Alzheimer’s-prone” (they make they same plaques and tangles that are seen in human Alzheimer’s brains.) Here are the 5 main discoveries.

1. Brains contain lithium. But people with Alzheimer’s dementia have much less.
2. Giving mice a low-lithium diet reduced their brain lithium and increased their brain’s plaques and tangles.
3. Giving mice extra lithium decreased their brain’s plaques and tangles.
4. Lithium carbonate (the standard psychiatric form) did not work as well as lithium orotate.
5. Even if given late in life, it lithium orotate restored the mice’ cognitive function.

These findings give strong support to the idea that lithium can prevent dementia. The Nature paper even presents results that show how lithium does this. I think this research is so important — and impressive — that I prepared a “translation” of their main findings. Here’s that video (then below I’ll link you to the my video playlist about lithium for dementia prevention, for the whole story including doses to consider).

The “whole story” video playlist is here. By the end, you might want to learn more about lithium orotate. Here’s my page on that.

Evidence for Reversal

Allow me to show you one key graph. Along with the results summarized above, this graph makes lithium orotate worth considering for people with cognitive impairment, I think.

The graph is adapted from the same Nature article’s mouse studies It shows the results of a mouse test of cognitive function. So notice: we’re not looking at plaques and tangles. We’re looking at the reason you’re probably reading this: can lithium orotate actually recover some lost cognitive function? Got it? Okay, let’s go…

Oh, wait: how do you test a mouse’s cognitive function? The research team used several tests for this. One is “time spent in target area”: the mice were put back in a maze where they’d previously received a reward. They had to remember where that was, to get another reward. The graph shows how much time they spent in the correct part of the maze. So, we’re looking at memory.

And one more thing: in the studies summarized above, the mice received small amounts of lithium beginning early in their lives. But in the experiment graphed below, the lithium did not begin until they were late-middle-aged (e.g. 17 months, until really old, 22 months). In these Alzheimer’s-prone mice, the development of Alzheimer’s changes (plaques and tangles and cognitive decline) is already extensive by late middle age.

You might understand this graph immediately: great. If not, allow me to walk you through it.

Step 1: what is “wild type”?

In this study, the researchers included “wild type” mice. Unlike the mice they used in all the rest of their studies, wild-type mice are not prone to develop Alzheimer’s-like cognitive impairment.

In grey you see the cognitive performance of the wild type mice. They’re old, but they’re still capable of learning (I’m on their team!). They spend more time in the correct part of the maze, because they remember where to look.

Step 2: what is “water”?

In blue are results for Alzheimer’s-prone mice who were allowed to grow up with no lithium in their water, developing plaques and tangles and loss of memory. They don’t spend as much time in the portion of the maze where they found a reward before. They don’t remember where to look.

Step 3: lithium orotate results

You can see where this is going now, right? In green are results for the mice who received lithium orotate starting in early old age: almost as good as the wild-type mice. Their memory is almost unimpaired.

In the experiments summarized above, lithium was given to Alzheimer’s-prone mice from a very young age with the goal of preventing development of plaques and tangles and cognitive impairment. This was very successful, when using lithium orotate.

Here, the idea is to undo damage that had already occurred by late middle-age. This also appears to have been very successful, using lithium orotate.

Lastly, Step 4: lithium carbonate results

You’ve already got this: mice who were given lithium carbonate from 17 to 22 months, well, it was better than water but not as good as lithium orotate.

Two Conclusions

1. Lithium orotate, in mice, appeared to prevent the development of Alzheimer’s-like brain changes (plaques and tangles). So did lithium carbonate, but less successfully. These results support those of small clinical trials of lithium (carbonate) in humans, which showed a slowing of Alzheimer’s progression.
   
2. In further limited experiments in mice, lithium orotate appeared to reverse the changes of Alzheimer’s dementia, including cognitive impairment. Could this happen in humans too? We have no such research, so the answer is not known.
   
3. What to do while awaiting further research? That requires a comparison of lithium orotate and the main alternative approach, lecanemab. I’ll link that page here soon. Until then, the video below might help.