Credit: doug m on Shutterstock
A study shows that the same chemical in both your salad and your tap water might not act the same in the body.
In A Nutshell
- In a 27-year Danish study, nitrate from vegetables was linked to lower dementia rates, while nitrate from processed meat and tap water showed the opposite pattern.
- The “same chemical” may behave differently depending on what comes with it: plants bring helpful compounds, while processed meats may make harmful byproducts more likely.
- Tap-water nitrate was tied to higher dementia rates even at levels far below current regulatory limits.
- Early-onset dementia showed stronger patterns, but those cases were uncommon, so the bigger swings come from a smaller group.
A glass of tap water and a plate of spinach contain the same chemical compound, yet research suggests the two sources of nitrate have opposite effects on the brain. Danish scientists tracking more than 54,000 adults for 27 years found that nitrate from vegetables was associated with an 11% lower dementia rate, while nitrate from drinking water correlated with a 14% increase, even at concentrations ten times below regulatory safety limits.
People who ate the most vegetable-sourced nitrate had an 11% lower rate of dementia compared to those who ate the least. On the other hand, those consuming high amounts from meat showed an 11% higher rate, while the highest tap water nitrate exposure was associated with a 14% increase.
The patterns intensified for early-onset dementia, which develops before age 65. Though these cases were rare (just 191 out of more than 54,000 participants) vegetable eaters in this group saw rates 39% lower, while those getting nitrate from animal sources had rates 73% higher.
Scientists recruited participants between 1993 and 1997, when they were ages 50 to 64, and followed them until 2020. During that period, 4,750 people developed dementia. The research team used detailed food questionnaires and Denmark’s health registries to track both dietary habits and dementia diagnoses.
Why Nitrate Acts Differently in Plants, Meat, and Water
Nitrate converts in the body to nitric oxide, a molecule that maintains blood vessel flexibility and supports brain oxygenation. Vegetables deliver nitrate alongside protective compounds (vitamin C and polyphenols) that may prevent it from forming N-nitrosamines, molecules animal research has linked to brain cell damage.
Meat, meanwhile, presents a different biochemical environment. It contains heme iron and proteins called amines that may promote N-nitrosamine formation. Even at relatively low intake levels, processed-meat nitrate and nitrite tracked with higher dementia rates in this dataset.
The group with the lowest dementia rates consumed about 64 milligrams of nitrate daily from vegetables, or roughly half a cup to one-and-a-half cups of raw leafy greens. The higher-risk group consumed around 11 milligrams daily from animal sources, equivalent to a few hundred grams of beef or several servings of yogurt.
These are translations of the nitrate amounts researchers measured, not specific portions tested in the study. Lettuce and potatoes were the primary vegetable contributors in this Danish population.

The Surprising Link Between Tap Water and Dementia Risk
Researchers tracked nitrate levels at every address where participants lived over 15 years and found elevated dementia rates at concentrations between 1.8 and 5.1 milligrams per liter. That’s roughly one-tenth of what government regulators consider safe.
The EPA and similar agencies worldwide set limits at 44 to 50 mg/L, standards designed primarily around short-term risks like infant methemoglobinemia rather than long-term neurological effects in adults.
Agricultural fertilizer runoff accounts for most tap water nitrate. Whether the elevated risk stems from nitrate itself, N-nitrosamines that may form in water, or other agricultural contaminants remains uncertain.
The association held when researchers accounted for people moving between homes with different water quality. Those exposed to higher (but still below regulatory limits) levels over 15 years showed 12% to 16% elevated dementia rates.
Why Early-Onset Dementia Shows Stronger Effects
Every association was two to four times stronger for early-onset dementia. Early-onset is less common than late-onset but particularly disruptive because it strikes during working years and while people are raising families.
Blood vessels in younger people may be more responsive to both beneficial and potentially harmful nitrate effects. By the time someone reaches their 70s or 80s, accumulated factors may overshadow dietary influences. The median age for early-onset diagnosis in this study was 62.
What This Long-Term Study Can and Can’t Prove
The research team collected dietary information through questionnaires completed at enrollment. They matched this data against Denmark’s health registries, which track hospital visits and prescriptions.
The study, published in Alzheimer’s & Dementia, accounted for smoking, exercise, education, weight, and alcohol intake. The associations persisted across these adjustments.
Several limitations warrant consideration. Dietary information came from a single baseline questionnaire, missing any changes over 27 years. When researchers examined just the first decade of data, the vegetable association strengthened from 11% to 43%, suggesting dietary changes may have weakened long-term findings.
Food frequency questionnaires carry inherent recall bias and may underestimate intake of less common high-nitrate foods. The study included primarily white individuals, limiting generalizability to other populations.
Ultimately, this is observational research, capable of showing associations but not proving causation. Still, the source-specific pattern was consistent and biologically plausible, adding to growing evidence that nitrate’s source may matter as much as quantity.
Effect Modification by Diet Quality
An unexpected finding emerged among those showing the strongest associations with vegetable nitrate. Lower dementia rates appeared primarily in people not already consuming substantial vitamin C and polyphenols from other sources.
Those already eating berries, citrus, and colorful vegetables throughout the day showed minimal additional benefit from leafy greens. But for people with lower baseline intake of these compounds, nitrate-rich vegetables correlated with clearer benefits.
This pattern suggests a threshold effect. Once vascular health reaches a certain level, additional nitrate may not provide extra advantages. For those with marginal diets, leafy greens may offer measurable benefits.

Implications for Public Health
These findings align with existing dietary guidance emphasizing plant consumption and limiting processed meats, and adds potential brain health benefits as yet another upside tied to eating vegetables.
The water findings complicate public health messaging. Nitrate levels associated with elevated dementia rates fall below consumer warning thresholds. People in agricultural areas might consider checking water quality reports or using filtration, though researchers stopped short of recommending policy changes.
Current water standards focus on preventing acute toxicity rather than chronic neurological effects. Whether regulations adequately protect brain health over decades of exposure remains an open question.
The research adds specificity to broad dietary recommendations. The protective range described by this work (half a cup to one-and-a-half cups of leafy greens daily) represents an achievable target for most people. Whether increasing vegetable intake or reducing animal-sourced nitrate actually prevents dementia would require controlled trials, but the observational evidence now spans multiple large cohorts with consistent findings.
Disclaimer: This article is for general information only. It does not offer medical advice. If you have questions about diet, dementia risk, or water quality, talk with a qualified clinician or local water expert.
Paper Summary
Limitations
The study’s dietary information came from a single questionnaire completed at enrollment, missing any changes in eating habits over the 27-year follow-up. This likely weakened the observed associations over time. The food frequency questionnaire method carries recall bias and may have underestimated intake of less common high-nitrate foods. Drinking water estimates covered residential addresses but not workplace exposure. The water nitrate range in the study was relatively narrow, mostly below 9 mg/L, and may not reflect higher exposures in other areas of Denmark. Dementia diagnosis relied on clinical records and prescriptions, potentially missing cases, and researchers couldn’t distinguish between dementia subtypes like Alzheimer’s versus vascular dementia. The predominantly Caucasian Danish population limits how well these findings apply to other ethnic groups. Despite adjusting for many confounding variables, unmeasured factors related to socioeconomic status, environmental exposures, or other dietary components could have influenced the results.
Funding and Disclosures
The Danish Diet, Cancer and Health Study was funded by the Danish Cancer Society. This analysis received additional support from the World Cancer Research Fund (IIG_FULL_2020_020), Independent Research Fund Denmark (1030-00307B), and the Novo Nordisk Foundation Challenge Programme through BERTHA – the Danish Big Data Centre for Environment and Health (Grant NNF17OC0027864). Individual researchers received support from the National Health and Medical Research Council of Australia, Western Australian Future Health Research and Innovation Fund, and National Heart Foundation of Australia. The authors declared no conflicts of interest.
Publication Details
The study was published in Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association in 2025 (Volume 21, Article e70995). The research team included Catherine P. Bondonno and colleagues from Edith Cowan University (Australia), The University of Western Australia, The Danish Cancer Institute, Aarhus University (Denmark), the Geological Survey of Denmark and Greenland, Herlev & Gentofte University Hospital, Murdoch University, and the University of Copenhagen. DOI: 10.1002/alz.70995.







