Disclosure of interest: I help Rosbacher and Römer Brunnen develop their markets in Taiwan. That background is directly relevant here, especially when a discussion moves from mineral-water composition to health.
Before working with the brand, I had never imagined that a bottle of water would lead me into neuroscience papers.
My earlier article placed Bad Vilbel’s water culture, a mineral-analysis report, and a 2025 study of lithium and Alzheimer’s disease on one timeline, describing it as “tradition arriving first, science following.” The connection interested me, but it also took the article across a gap that research has not yet bridged.
A community’s long history of drinking a particular water and an experiment suggesting that lithium may contribute to brain function require separate verification. Together, they have not established that drinking this water protects memory.
I want to retain my curiosity about water culture while keeping that gap visible to readers.
Start by separating the people from the mice
In August 2025, Aron, Yankner, and colleagues published Lithium deficiency and the onset of Alzheimer’s disease in Nature. The study analyzed human brain tissue and conducted mouse experiments. “A Harvard study proved it” cannot adequately describe both.
In the human data, the team observed lower brain lithium levels in people with mild cognitive impairment and a relationship between amyloid and lithium availability. This part provides clues from humans, not treatment results from randomly assigning people to receive lithium.
The mouse experiments restricted dietary lithium to observe changes in pathology and memory; the study also reported results from lithium-orotate intervention. This allowed the researchers to pursue causal mechanisms, but the subjects were still mice.
Harvard Medical School’s account likewise cautions that confirmation through human clinical trials is needed. These findings can guide the next experiment; they do not yet provide a drinking-water prescription for readers.
A “sponge” is an explanatory analogy, not a complete account of the disease
Comparing amyloid to a sponge that takes up lithium helps explain the mechanism proposed in this study. It cannot, by itself, explain every case of someone who has plaques but remains cognitively normal.
A more accurate account is that the team reported evidence of lithium binding to plaques and of reduced lithium availability, then proposed a disease mechanism worth testing. This does not overturn the entire history of amyloid research or establish that replenishing lithium will prevent cognitive decline in people.
Nor can a compound’s name settle whether lithium orotate has a particular advantage. A scholarly commentary published in 2026 questioned its chemical state in the body and its advantages over lithium carbonate. That disagreement requires experiments to resolve; this article has not adjudicated it as settled fact.
Keeping the disagreement in view makes the research look less tidy, but it better reflects what we currently know than a story running directly from discovery to product.
Human trials need more than “about to begin”
Clinical progress should be tied to named trials and registration records. “Human trials are about to begin,” without the trial’s name, substance, or status, leaves readers unable to judge what is happening.
There are other human studies of lithium and cognition. For example, a low-dose lithium-carbonate pilot trial published in JAMA Neurology in 2026 followed older adults with mild cognitive impairment for two years. Of 83 people randomized, 80 began treatment, and none of the six coprimary outcomes met the prespecified significance threshold.
It offers estimates and exploratory results to inform subsequent research, but signals in some measures cannot be rewritten as established efficacy. It tested lithium carbonate, not a direct human validation of the 2025 lithium-orotate mouse findings, and it did not test mineral water.
The trial neither establishes that all lithium research lacks promise nor endorses lithium-containing water. Stating the substance, population, and results clearly lets us identify the next question.
A laboratory report describes composition, not efficacy
The Römer Brunnen SGS Institut Fresenius report retained on this site is dated March 30, 2020. The Lithium entry on page one records 1.1 mg/L.
That number has a defined purpose: describing the lithium concentration of the sample in that report. It does not tell us how much to drink each day, much less test cognitive changes after drinking it. Nor can this old report guarantee the composition of every current batch.
Comparing mg/L in water, daily intake in mice, and daily milligrams of a medicine requires consistent treatment of concentration, intake volume, body weight, and whether a quantity refers to elemental lithium or a lithium salt. Simply dividing the numbers does not create a human-dose comparison. This article offers no self-dosing conversion.
A lithium concentration in natural water also does not, by itself, establish its chemical form or differences in absorption from a research formulation. An elemental-concentration test has not verified chemical speciation or human pharmacokinetics.
Local traditions deserve attention; health promises require their own evidence
The first article discussed German water culture, the second labels, and the third pairing water with meals. I remain interested in where water comes from, how communities understand their springs, and how people choose what they drink each day.
But the name Heilwasser, a spa town’s history, and long-standing drinking habits cannot jointly establish prevention of dementia. Discussing a product’s medical uses requires checking its actual authorized indications and applicable conditions, rather than inferring them from its name or lithium content.
My market-development relationship with the brands makes this distinction particularly necessary. “This sample report contains this amount of lithium” can be tied to a document; “therefore it helps memory” needs different evidence. Adding “may” or “not medical advice” does not make the missing research appear.
Questions a bottle of water can still hold
I am still willing to learn about geology, local life, and food culture through a bottle of water. Research into lithium and brain degeneration should be tested on its own terms.
This article has not established that lithium-containing water prevents or treats dementia. It should not be used as a reason to buy lithium supplements or change treatment independently. Relevant health needs should be discussed with the medical professionals caring for you.
The next time I discuss this water, I want to put the composition report on the table first: which entry was measured, and which statement still needs evidence?
Sources and scope of reading
- Aron et al., Nature, 2025: the original PubMed abstract and figure legends were checked against Harvard Medical School’s account. The Nature and PMC full-text entry points could not be read in full during this revision; no complete revalidation of the methods is claimed. Original abstract and figure legends · Institutional account
- Gildengers et al., Low-Dose Lithium for Mild Cognitive Impairment, JAMA Neurology, 2026: a human pilot trial whose substance, population, and main outcomes differ from drinking mineral water. Original study
- Hajek et al., Lithium orotate: distinct compound or simply Li+ after administration?: a scholarly commentary published online in June 2026. Its abstract was checked as evidence of questions about the advantages of a particular lithium salt, not as another efficacy experiment. Abstract
- SGS Institut Fresenius, March 30, 2020, report 4740147, page 1: the date, sample, and lithium concentration in the site’s retained scan were checked. These are not generalized to all batches or health effects. Complete file


💬 Comments
Loading...