On Dec. 17, 2025, A survey of Alzheimer’s disease prevalence in Norway confirms earlier estimates and might show how education level relates to risk. Nearly one in ten people over the age of 70 have Alzheimer’s disease dementia, shows a first-of-its-kind study that paired blood-based markers and clinical assessments to study the disease in Norway. That prevalence is in line with previous estimates for some other white populations. But there were also unexpected differences, including higher disease rates than anticipated in individuals older than 85.

To assess the prevalence of Alzheimer’s disease (AD), an international team of researchers turned to the Trøndelag Health (HUNT) study, a prospective research study that started in 1984 and that has collected health data and biological samples from 250,000 Norwegians.

Using blood samples from 11,486 individuals in the study aged 58 and above, the team looked at levels of a protein, called tau, that has been phosphorylated at a specific site. Known as pTau217, this blood marker serves as a proxy for the build-up of amyloid plaque in the brain, a hallmark of AD. HUNT-study participants over the age of 70 have undergone cognitive testing, enabling the researchers to compare pTau217 levels with the presence of dementia.

Around 10% of participants over the age of 70 had dementia and AD pathology, showing both cognitive impairment and high pTau217, they report. Another 10% had mild cognitive impairments and high pTau217. And 10% had high pTau217 but no signs of cognitive impairment, which the authors refer to as preclinical AD.

These findings are broadly in line with expectations, but there were surprises, too. Some 25% of people aged 85–89 had dementia and AD pathology, up from previous estimates of around 7% for men and 13% for women in this age group in Western Europeans. And the incidence of preclinical AD in younger individuals was 8% in those aged 70–74, down from a previous estimate of around 22%.

The discrepancies probably reflect selection bias, says study co-author Anita Lenora Sunde, a physician and dementia researcher at Stavanger University Hospital in Norway. Previous estimates were made by recruiting participants for brain scans, and people with dementia might not have wanted to or been able to participate.

But other factors could also be at play, says Villain. The latest study uses a high pTau217 threshold to define its preclinical AD population, and so excludes people with intermediate levels of the protein and emerging pathology