Higher levels of a protein in urine may signal an increased risk of developing dementia, particularly forms related to blood vessel damage in the brain. A large Swedish study published in the Journal of Internal Medicine suggests that measuring this protein, known as albumin, could help doctors identify older adults who are more likely to experience cognitive decline.
The findings point to a relatively simple and non-invasive way to flag people at greater risk for certain types of dementia. Since treatments that lower albumin in the urine are already available, this research could help inform early prevention strategies, especially for conditions that are difficult to diagnose in early stages.
Dementia is a broad term that refers to a decline in cognitive function severe enough to interfere with daily life. It includes conditions such as Alzheimer’s disease, vascular dementia, and mixed forms that share features of both. As the population ages, researchers have been working to identify early markers that might predict who will develop these diseases.
While kidney disease has already been linked to a range of health problems, recent evidence suggests it might also be connected to brain function. Two main indicators are often used to assess kidney health: the estimated glomerular filtration rate, which measures how well the kidneys are filtering waste, and albuminuria, which indicates the presence of albumin protein in urine. This protein is usually retained in the bloodstream, so its appearance in urine can be a sign that small blood vessels are damaged.
Although earlier studies have explored the relationship between kidney function and dementia, many focused only on the filtration rate and overlooked albuminuria. The researchers behind this new study wanted to know whether albuminuria on its own could predict who might develop dementia, and whether this risk held true for specific types of the disease.
“Dementia currently has no cure, so it is crucial to identify people at risk for prevention or early intervention,” said study author Li Luo, a postdoctoral researcher at the University Medical Center Groningen. “Chronic kidney disease is reported to increase dementia risk. Albuminuria is a urine marker of early kidney damage in chronic kidney disease, it can be treated, but its role in dementia is not well understood. We therefore want to examine the risk of all-cause and type-specific dementia associated with albuminuria.”
The researchers used data from the Stockholm Creatinine Measurements (SCREAM) project, which includes health records for nearly all residents in Stockholm, Sweden. They focused on over 132,000 people aged 65 and older who had at least one test for urine albumin between 2006 and 2019 and no prior diagnosis of dementia.
The participants were followed for a median of nearly four years. During that time, about 7 percent developed dementia. To analyze the connection between albumin levels and dementia risk, the researchers grouped participants based on the amount of albumin in their urine, measured as a ratio with creatinine, another waste product. They used three categories: low (<30 mg/g), moderately high (30–299 mg/g), and high (≥300 mg/g).
After adjusting for age, sex, education, existing health conditions, and kidney filtration rate, they found that people with moderately high levels of albumin had a 25 percent higher risk of developing any form of dementia compared to those with low levels. Those with high albumin had a 37 percent higher risk.
“Our research found that people with albuminuria above 300 mg/g have nearly a 40% higher risk of developing dementia,” Luo told PsyPost. “This is particularly relevant for those aged over 65, who often have conditions like heart disease or diabetes that already involve albuminuria testing. Our results suggested these tests can also provide meaningful insights into brain health.”
“The kidneys and the brain may seem like very different organs, but they share an important characteristic: both depend on a delicate network of small blood vessels. When the blood vessels in the kidneys are damaged, the same process often occurs in the brain,” added senior author Hong Xu, an assistant professor at Karolinska Institutet.
The findings indicate that “albuminuria is a non-invasive marker that can help identify people at higher risk of dementia, potentially allowing for early interventions that may help prevent or delay its onset,” Luo said.
These elevated risks were particularly tied to vascular dementia and mixed forms that involve both vascular and Alzheimer’s-related changes. Specifically, those with moderately high albumin had a 33 percent greater risk of vascular dementia, and those with high levels had a 25 percent greater risk. The same patterns were found for mixed and unspecified forms of dementia.
The researchers did not find a statistically significant link between albumin levels and the risk of developing Alzheimer’s disease on its own. This suggests that albumin in urine may be more closely related to damage in blood vessels, which is a known driver of vascular and mixed dementias.
“We were surprised to find that albuminuria is more strongly associated to vascular dementia than Alzheimer’s disease,” Luo said. “This aligns with its role as a marker of endothelial dysfunction, highlighting its relevance to vascular pathways in dementia development.”
To confirm these findings, they repeated the analysis using another common method of measuring urine protein: a dipstick test. Even with this simpler method, the relationship between higher urine protein and increased dementia risk remained.
The team also tested for other possible explanations. For instance, they excluded early dementia cases that occurred within one year of the albumin test to rule out the possibility that the disease was already developing. They also looked for signs that unmeasured health issues might have influenced the results but found no strong evidence of this.
Importantly, they compared people with different levels of kidney filtration and found that the link between albumin and dementia held steady, regardless of how well the kidneys were functioning overall. This suggests that albumin in urine may be telling us something distinct from what kidney function alone can show.
This study provides evidence that protein in urine, specifically albumin, is linked to an increased risk of developing dementia. However, several limitations should be taken into account.
The participants were drawn from a Swedish healthcare database, so the findings may not apply to people in other countries with different health systems, genetics, or lifestyles. Also, the study relied on clinical diagnoses of dementia, which may miss early or mild cases, especially among frailer individuals. Some participants were diagnosed with unspecified dementia, which could mean that the exact type was unclear due to incomplete testing.
Another issue is that albuminuria may not be the cause of dementia but rather a signal of broader vascular damage happening throughout the body. While the findings are strong, observational studies like this one cannot prove cause and effect.
“Albuminuria is often seen as only a kidney marker, but it can actually indicate broader organ damage,” Luo noted. “This is particularly relevant for diseases in which vascular dysfunction plays a role, such as heart disease, diabetes, cancer, and dementia suggested in this study.”
The researchers also noted that they did not have data on some potentially important factors, such as diet, physical activity, or early signs of memory problems. More detailed testing and brain imaging could help clarify the pathways between kidney and brain health.
Even so, the results suggest that measuring albumin in urine could offer a useful clue in identifying people at risk for vascular-related forms of dementia. Since albumin levels can be lowered with medications and lifestyle changes, the authors are planning future studies to examine whether these interventions might also reduce the risk of cognitive decline.
The study, “Albuminuria is associated with increased risk of dementia, independent of eGFR: The SCREAM project,” was authored by Li Luo, Ron T. Gansevoort, Lyanne M. Kieneker, Yuanhang Yang, Alessandro Bosi, Rudolf A. de Boer, Casper F. M. Franssen, Maria Eriksdotter, Juan-Jesus Carrero, and Hong Xu.
