Risk for these late-life outcomes isn't predetermined
A sweeping study of more than 214,000 older adults across 14 nations has revealed that dementia does not threaten all people equally or in the same ways — where one lives shapes which risks loom largest, even as the deeper architecture of how those risks relate to one another holds steady across cultures. Led by researchers at USC and published in The Lancet Healthy Longevity, the work challenges the long-held assumption that prevention wisdom earned in wealthy nations can be transplanted wholesale to the rest of the world. What emerges is a quieter, more hopeful truth: dementia is not fate, and the path toward preventing it must be drawn from the ground up, in each place, for each people.
- The urgency is demographic — hundreds of millions of aging adults in low- and middle-income countries face dementia risk profiles that existing prevention frameworks were never designed to address.
- The disruption cuts through a foundational assumption in global health: that a universal checklist of risk factors can guide prevention everywhere, when in reality low education burdens 85% of older Chinese adults while obesity weighs on nearly 45% of Americans.
- Beneath the divergence, researchers found an unexpected stabilizing pattern — cardiovascular risks cluster together, risky behaviors cluster together, and this underlying architecture repeats itself across wildly different societies.
- Health systems are now being pointed toward a more efficient model: treat entire clusters of related risks at once, tailoring interventions to local realities rather than importing strategies built for different populations.
- The work is already expanding — Kenya, Egypt, and newer risk factors like sleep quality are next, as scientists race to map the full global terrain of a disease that waits for no border.
When researchers at USC set out to study dementia risk across 14 countries and more than 214,000 older adults, they expected to find differences. What surprised lead scientist Emma Nichols was how much the similarities mattered too.
The differences are real and consequential. Low education affects 85.6 percent of older adults in China but only 12 percent in the United States. High BMI burdens nearly 45 percent of Americans while touching just 13 percent of people in India. These gaps represent fundamentally different populations with fundamentally different vulnerabilities — and they expose the limits of applying prevention strategies developed in wealthy nations to the rest of the world.
Yet the study, presented at the Alzheimer's Association International Conference in London and published in The Lancet Healthy Longevity, found something steadying beneath all that variation. Across every country examined, risk factors cluster in recognizable ways: cardiovascular risks group together, risky behaviors travel in pairs. The prevalence shifts; the underlying logic does not.
Drawing on harmonized data collected between 2009 and 2023 from aging studies in England, Korea, Brazil, Malaysia, Mexico, and beyond, the team examined 12 modifiable risk factors — hearing loss, depression, physical inactivity, social isolation among them — tracking not just how common each was, but how they combined within individuals.
The practical payoff is a new model for prevention: rather than importing universal programs, health organizations can design interventions that match their own population's actual risk profile, treating whole clusters of related conditions at once. A diabetes program, for instance, might simultaneously address cholesterol and blood pressure — efficiency born from understanding how risks travel together.
Nichols is careful to carry the message beyond policy. Dementia risk accumulates over a lifetime, she notes, and individuals retain agency even as education systems, healthcare infrastructure, and economic conditions shape the terrain around them. The research team is already planning to extend the work to Kenya and Egypt, and to incorporate newer factors like sleep quality as data becomes available.
What this global portrait ultimately offers is not a single answer, but a more honest question: what does prevention look like here, for these people, given these realities? The underlying logic may be universal. The work of acting on it is always local.
Researchers studying dementia across the globe have discovered something counterintuitive: the risk factors that matter most vary dramatically depending on where you live, yet the way those risks cluster together remains strikingly consistent. A team led by scientists at USC analyzed data from more than 214,000 older adults spread across 14 countries and regions—from the United States and Western Europe to China, India, Brazil, and Mexico—and found that a universal approach to preventing dementia simply won't work everywhere.
The study, presented this week at the Alzheimer's Association International Conference in London and published in The Lancet Healthy Longevity, challenges the assumption that what we know about dementia prevention from wealthy nations applies equally to low- and middle-income countries. The differences are stark. Low education, for instance, affects 85.6 percent of older adults in China but only 12 percent in the United States. High body mass index tells a similar story of geographic divergence: it impacts 44.9 percent of Americans while affecting just 13.3 percent of people in India. These aren't minor variations. They represent fundamentally different populations with fundamentally different vulnerabilities.
Yet beneath these differences lies an unexpected pattern. Emma Nichols, the lead researcher and a scientist at USC's Center for Economic and Social Research, found herself more surprised by the similarities than the differences. Certain risk factors tend to cluster together in the same ways across all the countries studied. Cardiovascular risks—high cholesterol, hypertension—group together. Risky behaviors like smoking and drinking cluster as well. This consistency suggests that while the prevalence of individual risk factors shifts from place to place, the underlying architecture of how they relate to one another remains recognizable.
The research drew on harmonized data collected between 2009 and 2023 from long-running aging studies in places including England, Ireland, Korea, Mexico, Malaysia, and Brazil. The team examined 12 modifiable risk factors identified by the Lancet Commission on dementia: hearing loss, depression, physical inactivity, social isolation, and others. They looked not just at how common each factor was, but how they varied by age, gender, and education level, and crucially, how often multiple risk factors appeared together in the same person.
The practical implication is significant. Health organizations and policymakers can now design prevention strategies that reflect their own populations' actual risk profiles rather than importing wholesale approaches developed elsewhere. A diabetes care program, for example, could be expanded to address the entire cluster of related cardiometabolic risks simultaneously—treating not just blood sugar but cholesterol and blood pressure at the same time. This kind of targeted efficiency could make prevention efforts more effective and more efficient.
Nichols emphasizes that the findings carry a message for individuals as well. Dementia risk is not predetermined or inevitable. The factors that drive it accumulate over a lifetime, and people have agency in shaping their own risk, even as broader societal forces—access to education, healthcare infrastructure, economic opportunity—shape the landscape in which that individual choice occurs. The research team is already planning to expand the work, incorporating newer risk factors like sleep quality and extending the analysis to additional countries including Kenya and Egypt as more data becomes available.
What emerges from this global snapshot is a more nuanced picture of dementia prevention: local in its specifics, universal in its underlying logic. The challenge now is translating that insight into action, tailoring interventions to the realities of each population while learning from patterns that transcend geography.
Notable Quotes
I was less surprised by the differences and more surprised by some of the similarities, particularly in the ways these risks are patterned across settings.— Emma Nichols, lead researcher, USC Schaeffer Institute
The Hearth Conversation Another angle on the story
Why does it matter that risk factors cluster the same way everywhere if their prevalence is so different?
Because it tells us something about causation. If cardiovascular risks cluster together in China the same way they do in the U.S., it suggests the biological mechanisms linking them are universal—even if the percentage of people experiencing them varies wildly. That means some interventions might actually work everywhere.
So you're saying a prevention program designed in one country could work in another, just aimed at different people?
Exactly. The architecture is the same; the population you're targeting is different. In China, you'd focus heavily on education access because that's where the vulnerability is. In America, you'd focus on weight and metabolic health. But the underlying logic of how to address clusters of risk could be similar.
What about the countries that weren't included—why does that matter?
Most of what we know about dementia comes from wealthy Western countries. This study finally includes low- and middle-income regions where most of the world's older adults actually live. But Africa, Southeast Asia, parts of the Middle East—we still don't have good data. The patterns might be different there.
If someone reads this, what should they actually do differently?
Understand that your risk isn't fixed. The factors that matter in your country, in your community—those are the ones worth paying attention to. And recognize that preventing dementia isn't just about individual choices. It's about the systems around you: education access, healthcare quality, whether you can afford to be physically active.