People in Britain who were required to stay in school an extra year under a 1972 reform were less likely to be diagnosed with dementia by age 65 than near-identical peers born just weeks earlier, according to a study published Oct. 23 in the Proceedings of the National Academy of Sciences. The protective effect held even among people with high genetic risk, the authors report, and they describe the work as the first to document a causal link between education and diagnosed dementia.
The peer-reviewed study, led by health economist Silvia Barcellos of the Center for Economic and Social Research at the USC Schaeffer Institute and the University of Wisconsin-Madison, used a natural experiment. In 1972, England, Scotland and Wales raised the minimum school-leaving age from 15 to 16. Anyone born on or after Sept. 1, 1957, had to stay in school longer; those born just before could still leave at 15. Comparing people born days apart on either side of the cutoff, a design known as regression discontinuity, separates the effect of schooling from family background and other factors that normally travel with education.
The reform raised the share of students staying in school to at least 16 by 14 percentage points, the researchers found. Using UK Biobank data, they followed participants through age 65 with hospital records, death registries and self-reported medical history. Those born just after the cutoff were less likely to receive a diagnosis of Alzheimer’s disease or a related dementia than those born just before. The additional schooling also raised household income and reduced the incidence of stroke and heart attack. The USC release does not state the number of participants analyzed, and the authors caution that because dementia is still uncommon in a cohort this young, their estimates of the size of the effect are imprecise.
Combining the reform with participants’ genetic data, the team found the reduction in dementia occurred even among those at the highest inherited risk, and they suggest education may be especially protective for that group. “Genes help shape disease risk, but they do not necessarily determine the outcome,” said co-author Leandro Carvalho, associate professor of economics at the Center for Economic and Social Research and USC Dornsife.
“Our findings provide evidence that education itself can reduce dementia risk decades later,” Barcellos said. The link between more schooling and lower dementia rates has been observed for years, she noted, but “it has been difficult to know whether education itself is responsible.” Co-authors include Patrick Turley and Sneha Nimmagadda of USC Dornsife and Kenneth Langa of the University of Michigan. The work was funded by the National Institute on Aging, part of the National Institutes of Health, under five grants, and by the Russell Sage Foundation.
Why it matters
Education is routinely listed among modifiable risk factors for dementia, but most of that evidence is observational and vulnerable to confounding: people who stay in school longer differ in wealth, health and childhood circumstances. A policy change that imposed extra schooling on one birth cohort and not another gets closer to a true experiment. The paper’s title, “Education Reduces Dementia Incidence and Mitigates Genetic Risk,” states the authors’ conclusion plainly.
The genetic finding may matter most for how risk is communicated. If an exposure as ordinary as a year of secondary school can blunt the effect of inherited risk, then a high genetic score is not destiny, the authors argue. The study cannot say which pathway did the work; the researchers were unable to determine the mechanisms behind the reduction.
The caveats
The cohort was followed only to 65, an age at which dementia diagnoses are rare, so the estimates are imprecise and the authors say continued follow-up is needed to see whether the reductions persist as cases accumulate. Diagnoses came from hospital records, death registries and self-report, which capture only part of true dementia incidence. The design identifies the effect of one extra year of schooling at age 15 to 16 in 1970s Britain, which may not generalize to other education systems or to higher education. The USC release gives no effect sizes or confidence intervals, and the authors themselves say it is unclear how large the reduction in risk is.
What to watch
The authors say longer follow-up of the same cohort will show whether the gap in diagnoses widens as participants move into their 70s. The paper appears in the Oct. 27 print issue of PNAS, volume 123, issue 43.
This article is for general information and does not replace advice from a qualified clinician. Corrections: email the editor using the address on the Contact page.
