Eating more high-fat cheese and high-fat cream may be linked to a lower risk of developing dementia, according to a new study published on December 17, 2025, in Neurology®, the medical journal of the American Academy of Neurology. This study does not prove that eating high-fat cheese and high-fat cream lowers the risk of dementia, it only shows an association.
High-fat cheeses contain more than 20% fat and include varieties such as cheddar, Brie and Gouda. High-fat creams typically contain 30-40% fat and include whipping cream, double cream and clotted cream. These are commonly labeled as "full-fat" or "regular" versions in stores.
For decades, the debate over high-fat versus low-fat diets has shaped health advice, sometimes even categorizing cheese as an unhealthy food to limit. Our study found that some high-fat dairy products may actually lower the risk of dementia, challenging some long-held assumptions about fat and brain health."
Emily Sonestedt, PhD, Lund University, Sweden
Researchers analyzed data from 27,670 people in Sweden with an average age of 58 at the start of the study. They were followed for an average of 25 years. During the study, 3,208 people developed dementia.
Participants kept track of what they ate for a week and answered questions about how often they ate certain foods during the past few years. They also talked with researchers about how they prepared their food.
Researchers compared people who ate 50 grams or more of high-fat cheese daily to people who ate less than 15 grams daily. For example, 50 grams of cheese is about two slices of cheddar or half a cup of shredded cheese and is approximately 1.8 ounces. A typical serving of cheese is one ounce. Of those who ate more high-fat cheese, 10% developed dementia by the end of the study. Of those who ate less, 13% developed dementia.
After adjusting for age, sex, education and overall diet quality, researchers found that people who ate more high-fat cheese had a 13% lower risk of developing dementia compared to those who ate less. When looking at specific types of dementia, they found people who ate more high-fat cheese had a 29% lower risk of vascular dementia.
Researchers also found a lower risk of Alzheimer's disease among those who ate more high-fat cheese, but only among those not carrying the APOE e4 gene variant-a genetic risk factor for Alzheimer's disease.
Researchers also compared people who consumed 20 grams or more of high-fat cream daily to people who consumed none. For example, 20 grams of high-fat cream is about 1.4 tablespoons of heavy whipping cream. A recommended serving is about 1-2 tablespoons.
After similar adjustments, researchers found that those who consumed high-fat cream daily had a 16% lower risk of dementia compared to those who consumed none.
No associations were found between dementia risk and eating low-fat cheese, low-fat cream, high- or low-fat milk, butter or fermented milk, which includes yogurt, kefir and buttermilk.
"These findings suggest that when it comes to brain health not all dairy is equal," said Sonestedt. "While eating more high-fat cheese and cream was linked to a reduced risk of dementia, other dairy products and low-fat alternatives did not show the same effect. More research is needed to confirm our study results and further explore whether consuming certain high-fat dairy truly offers some level of protection for the brain."
A limitation was that study participants were all from Sweden, so results may not be the same for other populations. Sonestedt noted that in Sweden, cheese is often eaten uncooked, while in the United States, cheese is often heated or eaten with meat. Therefore, she says it is important that studies also be conducted in the United States.
The study was supported by Swedish Research Council, Swedish Heart and Lung Foundation, Crafoord Foundation, Magnus Bergvall Foundation and Albert Påhlsson Foundation.
Source:
Journal reference:
Du, Y., et al. (2025). High- and Low-Fat Dairy Consumption and Long-Term Risk of Dementia. Neurology. doi: 10.1212/wnl.0000000000214343. https://www.neurology.org/doi/10.1212/WNL.0000000000214343