Could adding lithium to tap water reduce dementia levels?


"Adding lithium to tap water could prevent thousands of dementia cases," reports The Daily Telegraph. The report is based on research from Denmark that found people who had lived in areas with higher levels of naturally occurring lithium (a type of metal) in the drinking water were slightly less likely to get dementia.

The study included 73,731 people with dementia and 733,653 without. However, it was not clear what level of lithium may be beneficial, and the authors say other factors might also have affected the results. Without further research, it's too early to be thinking about adding lithium to water supplies.

Lithium is the standard treatment for bipolar disorder, where it's used to stabilise mood. Animal studies suggest low doses of lithium may help improve learning and memory, while some previous studies in humans have found that low doses of lithium in older age may delay dementia.

The present study suggests the theory that long-term exposure to lithium is linked to dementia risk may be correct. But much more work will be needed to find out whether adding lithium to drinking water is safe and effective.

One practical consideration is that, despite international recommendations to add fluoride to tap water as a means of preventing tooth decay being in place since the 1960s, many English local authorities are reluctant to do so because of public opposition. Adding lithium to tap water could be an even harder "sell" in the face of public scepticism.

Where did the story come from?

The study was carried out by researchers from the University of Copenhagen, the University of Southern Denmark, the Geological Survey of Denmark and Greenland, Aarhus University and the National University of Singapore.

It was funded by Geocenter Denmark and published in the peer-reviewed journal JAMA Psychiatry on an open-access basis, so it's free to read online

BBC News gave a good overview of the study and, despite the overenthusiastic headline, the Daily Telegraph also covered the study accurately, making it clear that adding lithium to the water supply would be a premature step.

What kind of research was this?

This was a nationwide, population-based nested case control study. This type of study can identify patterns and links between factors, but cannot prove that one factor (in this case, naturally occurring lithium levels in drinking water) causes or protects against another (dementia).

According to the Drinking Water Inspectorate (DWI), lithium levels are not routinely measured in England and Wales. The British Geological Survey recently completed an analysis of several trace elements, including lithium, for the DWI, but the results have not yet been published.

What did the research involve?

Researchers used Denmark's population databases to identify anyone with a hospital diagnosis of dementia made between January 1 1995 and December 31 2013. For each person with dementia, 10 individuals of the same age and sex were matched as controls.

They used records of residence to establish where people had lived since 1986 then cross-referenced these to records of drinking water composition from the country's 275 municipalities.

They looked to see if exposure to lithium in drinking water was linked to the chance of being diagnosed with dementia.

People were excluded from the study if there was insufficient data about their places of residence. All control participants had to be alive and not diagnosed with dementia by the date the person they were matched with had been diagnosed with dementia.

Researchers used water samples from 151 waterworks around the country, taken between 2000 and 2010. They assumed the levels of lithium in different areas remained stable over time.

Because there's some evidence that people living in big cities have a different risk of some mental health conditions than those in smaller towns or the countryside, the researchers looked to see if the "urbanicity" of people's places of residence affected dementia risk.

They carried out calculations to see if there was an effect on dementia diagnosis at four levels of lithium dose:

  • 2.0 to 5.0 micrograms per litre
  • 5.1 to 10 micrograms per litre
  • 10.1 to 15 micrograms per litre
  • 15 micrograms or more per litre

What were the basic results?

Lithium levels in the different municipalities varied from 0.6 micrograms per litre in east Denmark to 30.7 micrograms per litre in west Denmark, with an average across the country of 11.6 micrograms per litre.

When the researchers compared the average level of exposure to lithium in drinking water between those with and without a diagnosis of dementia, they found:

People with dementia had an average level of 11.5 micrograms per litre. People without dementia had an average level of 12.2 micrograms per litre. Compared with the lowest levels of lithium (up to 5 micrograms per litre), they found:

  • People exposed to 5.1 to 10 micrograms per litre had a 22% higher risk of dementia (incidence rate ratio [IRR] 1.22, 95% confidence interval[CI] 1.19 to 1.25).
  • People exposed to 10.1 to 15 micrograms per litre had about the same risk of dementia (IRR 0.98, 95% CI 0.96 to 1.01).
  • People exposed to 15 micrograms or more per litre had a 17% lower risk of dementia (IRR 0.83, 95% CI 0.81 to 0.85).
  • Researchers found that people living in cities were the least likely to get dementia and those living in rural areas were the most likely to get it.

How did the researchers interpret the results?

The researchers said their results "confirmed the hypothesis that higher long-term lithium exposure from drinking water may be associated with a lower incidence of dementia", even though the results did not show a linear trend (a trend that points in one direction).

They also cautioned that they cannot rule out an effect of "confounding from other factors associated with municipality of residence".


The study is intriguing because we already know that lithium affects how the brain and nervous system work through many different pathways. However, the results are difficult to interpret.

The study seemed to suggest that lithium levels of more than 15 micrograms per litre could be protective against dementia in comparison with the lowest levels. However, that doesn't explain why levels of 5 to 10 micrograms per litre seemed to increase the risk of dementia in comparison with the lowest levels.

It's possible that some other factors – linked to where people live but not necessarily to the drinking water – are at work. More clinical studies are needed on the effects of long-term low-dose lithium so we can better understand whether one specific level of exposure might be protective.

There are plenty of things you can do to reduce your risk of dementia, although there are no guarantees. They include:

  • eating a healthy diet
  • maintaining a healthy weight
  • exercising regularly
  • not drinking too much alcohol
  • giving up smoking
  • trying to keep your blood pressure at a healthy level

Read more about dementia prevention.

Article Metadata Date Published: Tue, 29 Aug 2017
Author: Zana Technologies GmbH
NHS Choices