A major international study involving more than 214,000 older adults has found that preventable dementia risk factors differ substantially between countries, suggesting that a single global prevention strategy may not work equally well everywhere.
The USC-led research examined ageing populations across 14 countries and regions, including the United States, China, India, Brazil, Mexico, Malaysia, Korea and parts of Europe. Researchers found major differences in the prevalence of factors associated with dementia, including limited education, high blood pressure, excess body weight and smoking.
The findings were published in The Lancet Healthy Longevity and presented at the 2026 Alzheimer’s Association International Conference in London. The analysis was conducted by researchers from the University of Southern California, Brown University and Johns Hopkins University.
Major Differences Between Countries
Researchers found that the proportion of older adults exposed to particular risk factors varied dramatically depending on location.
Low educational attainment affected approximately 85.6 per cent of older adults included in the Chinese dataset, compared with only 12 per cent of participants in the United States.
The opposite pattern appeared for high body mass index. About 44.9 per cent of American participants had a high BMI, compared with 13.3 per cent of those studied in India.
These differences indicate that dementia-prevention programmes should reflect the most pressing health and social challenges within individual populations.
A country where poor access to education is widespread may need to prioritise childhood learning and lifelong cognitive development. Another country with high rates of obesity, diabetes or hypertension may benefit more from programmes focused on cardiovascular health, exercise and nutrition.
The study does not suggest that a person’s country of residence directly determines whether they will develop dementia. Instead, it shows that the distribution of recognised risk factors is strongly influenced by local social, economic, behavioural and healthcare conditions.
Twelve Modifiable Risks Examined
The researchers analysed 12 potentially modifiable dementia risk factors identified by the Lancet Commission.
These included low education, hearing loss, hypertension, obesity, depression, physical inactivity, social isolation and lifestyle behaviours such as smoking and excessive alcohol consumption.
The team measured how common each factor was, how its prevalence differed according to age, gender and education, and how frequently several risks appeared together in the same person.
The information came from harmonised surveys collected between 2009 and 2023 through the Gateway to Global Aging Data project. Using harmonised information allowed researchers to make more meaningful comparisons across national ageing studies that were originally conducted separately.
Much of the previous evidence about dementia prevention has come from wealthy Western countries. The new analysis was designed to determine whether patterns identified in the United States and Western Europe also appeared in low- and middle-income populations.
Similar Risk Clusters Found Worldwide
Despite the major national differences, researchers also discovered unexpected similarities.
Certain risk factors repeatedly appeared together across many populations. Cardiovascular conditions such as hypertension and high cholesterol frequently formed one cluster, while behaviours including smoking and alcohol consumption tended to form another.
Lead researcher Emma Nichols said these shared patterns were among the study’s most surprising findings.
Their consistency suggests that some prevention programmes could address several connected risks simultaneously rather than treating each health problem separately.
For example, a healthcare programme established to help patients manage diabetes could be expanded to screen for and treat high blood pressure, excess weight and high cholesterol. Addressing the full group of cardiometabolic conditions could potentially produce greater benefits than targeting diabetes alone.
Similarly, community programmes aimed at reducing social isolation could incorporate physical activity, hearing support and mental-health services for older residents.
Prevention Must Reflect Local Needs
The researchers said governments and public-health organisations should use local evidence when deciding where to direct dementia-prevention resources.
A standardised programme based largely on evidence from wealthy countries may overlook the most common risks affecting people elsewhere.
National strategies could therefore share broad goals while using different methods. Improving education may be the most important long-term measure in one country, while controlling cardiovascular disease, expanding hearing care or reducing smoking may be more urgent in another.
The results also highlight the role of wider social conditions. Individuals can make changes that may lower their risk, but access to education, healthcare, safe spaces for exercise and supportive social networks is often shaped by government policy and economic circumstances.
Nichols said dementia risk is not entirely predetermined and that people may be able to influence it throughout their lives. However, she also emphasised that broader societal conditions affect the choices and opportunities available to individuals.
Findings Do Not Guarantee Individual Outcomes
The risk factors examined in the study are associated with dementia at the population level, but they do not guarantee that a particular individual will or will not develop the condition.
A person may have several recognised risks and never experience dementia, while another person without obvious modifiable risks may still develop it because of age, genetics or other biological factors.
The research primarily compared how common risk factors were across different populations. It did not prove that the differences observed directly caused different dementia rates.
Nevertheless, the large sample and the inclusion of diverse countries provide health authorities with a broader picture than studies based mainly on North American and Western European populations.
Research Expected to Expand
Future work is expected to examine additional potentially modifiable risks, including poor sleep, as researchers gain access to more comparable international data.
Similar ageing surveys are also being developed in additional countries, including Kenya and Egypt, which could allow future studies to represent a wider range of African and Middle Eastern populations.
The project was led by Nichols through the USC Schaeffer Institute’s Center for Economic and Social Research. Other researchers included Jinkook Lee, Zachary Kunicki, Alden Gross, Michael Markot, Drystan Phillips and Jenny Wilkens.
The study was supported by the United States National Institutes of Health.
The findings reinforce the idea that many dementia risks can potentially be reduced, but the most effective response will depend on understanding which health, behavioural and social factors are most common in each community.
