For decades, studies have found that people with more education are less likely to develop Alzheimer’s disease and other dementias. The hard part has been knowing whether school itself does anything, or whether the people who stay in school longer simply differ in other ways, such as their genes, childhood circumstances or family money1. A new study published in the journal PNAS uses a change in British law from the 1970s to get closer to an answer1.
“We have known for a long time that people with more education are less likely to develop dementia, but it has been difficult to know whether education itself is responsible,” said Silvia Barcellos, a health economist at the University of Wisconsin-Madison who led the work1.
What happened
In 1972, the Raising of School-Leaving Age Act raised the minimum age at which pupils in England, Scotland and Wales could leave school from 15 to 161. The change applied only to students born after 1 September 19571.
That sharp cutoff created what researchers call a “natural experiment”: a situation where chance, not personal choice, decides who gets something. People born just after the date had to stay in school a year longer. People born just before could leave at 15. Otherwise, the two groups should be very similar1. The group made to stay longer is now in its late sixties1.
The researchers explained the logic in their paper: because people born right before and right after the date would have had similar dementia rates without the law, “any observed discontinuity could be attributed to the difference in schooling engendered by the policy”1. A discontinuity here means a sudden jump or drop in the numbers right at the cutoff date.
Using health data from the UK Biobank, a large long-running study of British volunteers, the team found what they were looking for. “There is a sharp decline in ADRD cumulative incidence among those born after September 1, 1957,” they wrote, using the shorthand ADRD for Alzheimer’s disease and related dementias1. “Our findings indicate that education has a negative causal effect on ADRD incidence,” meaning more schooling led to fewer cases1.
The evidence so far
The team’s calculation put the reduction at about 0.88 percentage points in the share of people who developed dementia1. The researchers themselves call this implausibly large, because fewer than 1 percent of people in the sample had dementia at all1. They suggest statistical factors may explain the odd number1.
“Although our sample is large, ADRD remains a rare outcome in these relatively young cohorts, and regression discontinuity designs require very large samples to achieve precise estimates. We can credibly estimate the direction of the causal effect of education, but not its precise magnitude.”
That quote, from the paper, is the key limit of the finding1. A regression discontinuity design is the formal name for comparing people just either side of a cutoff like a birth date.
Genes and schooling
The researchers also used participants’ DNA to sort them into higher and lower genetic risk for dementia, based on gene variants including APOE4, a well-known Alzheimer’s risk gene1. Among people born before the law change, those with higher genetic risk had markedly more dementia. Among people born after it, the extra school year was tied to a significant drop in dementia in the higher-risk group1.
“For people who know they have a higher genetic risk of dementia, this is a hopeful finding,” said co-author Leandro Carvalho, a health economist at the University of Southern California. “Genes help shape disease risk, but they do not necessarily determine the outcome.”1
What it does not show yet
- How big the effect is. The authors say plainly that their estimate of the size is not reliable, only its direction1.
- Why it happens. The study does not show what about the extra year mattered. The researchers suggest that benefits that come with more education, such as better jobs and higher social and economic standing, may build up over time2. That is an explanation they put forward, not something the study measured directly.
- What happens at older ages. The people studied are in their sixties, when dementia is still uncommon1. Whether the gap holds, grows or shrinks as they reach their seventies and eighties is not yet known.
- Whether it applies elsewhere. The data come from one country and one specific policy change in the 1970s1.
The coverage reviewed for this article does not say who funded the work.
Who is behind it
The study comes from health economists rather than brain scientists, including Barcellos at the University of Wisconsin-Madison and Carvalho at the University of Southern California1. Economists often use policy changes like this one to test cause and effect when a randomized trial is not possible. No one could ethically assign children at random to leave school early just to see what happens to their brains 50 years later.
The same kind of approach has been used on other questions with UK Biobank data. An earlier study this year looked at people born during wartime sugar rationing in Britain and compared their adult health with children who did not face those limits1.
How it fits the field
The link between schooling and dementia is well documented in observational studies, which watch what happens to people without changing anything. A US study using the Health and Retirement Study from 2000 to 2018 found dementia rates fell steadily with each added year of education, with an extra drop at 12 years of school (finishing high school) seen mainly in men and White adults3. An earlier USC study found college graduates kept good thinking skills for more of their later years than people who did not finish high school4.
Researchers have also used US laws before. A 2024 study led by Min Hee Kim, then at the University of California San Francisco, used differences in state schooling requirements and looked at more than 20,000 older Black and white adults. It found extra schooling caused by those laws was linked with better memory and thinking scores later in life5. The new UK study looks at dementia diagnoses themselves rather than test scores, which is why its authors describe it as the first evidence of its kind for a causal link to dementia1.
Not every finding points the same way. A 2007 USC study of Americans over 70 found that more educated people did better on memory tests at any given age but lost remembered words faster than less educated people, though the authors cautioned other factors could explain the difference6.
The idea that dementia risk can be shifted by life experience has also come from a randomized trial. In February, researchers reported that older adults who did about 10 hours of computer-based speed training in the late 1990s were about 25 percent less likely to be diagnosed with dementia over the next two decades, in a study of 2,802 people7.
What to watch next
The clearest test will come with time. As the people affected by the 1972 law move into their seventies and eighties, more dementia cases will appear in the data, which could let researchers estimate the size of the effect, something the authors say they cannot do yet1. Watch also for work that tries to explain why the extra year mattered, and for similar school-law studies in other countries that could confirm or challenge the finding.
