One of the questions I am most often asked in clinical practice, usually by an adult arriving for a cognitive evaluation in their fifties or sixties, is some version of: “Doctor, does this number actually mean anything for my life?”
For decades, I had to give a cautious, qualified answer. The honest one is now this: yes. Probably more than you would expect.
In 1932, the Scottish Council for Research in Education did something audacious. On a single Wednesday in June, they administered the same intelligence test to nearly every eleven-year-old child in Scotland, 87,498 of them. They did it again in 1947 with another 70,805 children. For nearly seventy years, almost no one knew what to do with these forgotten files. Then, in 2001, a psychologist named Ian Deary at the University of Edinburgh tracked down what had happened to the children of 1932.
The result quietly launched an entire scientific discipline called cognitive epidemiology, and produced one of the most replicated findings in modern behavioral science: a child’s IQ at age eleven predicts how long they will live.
Not a little. A lot.
Across more than 1.1 million people pooled in the largest meta-analysis to date, a single standard deviation of IQ, fifteen points, the gap between an average score of 100 and a “high-average” score of 115, corresponds to a 24 percent lower risk of death over the following 17 to 69 years. The relationship is graded: very smart people outlive smart people, who outlive average people, who outlive below-average people. The curve only flattens around an IQ of about 163.
If you are someone who tracks your VO2 max, your resting heart rate, your LDL cholesterol, or your biological age, there is a strong empirical case that you should also know your cognitive baseline. This article is the long answer to the question my patients keep asking, written for the people who never get to ask it in a clinic.
The Scottish Schoolchildren Who Quietly Solved a Mystery
Before we discuss what cognitive epidemiology actually found, you need to understand why the Scottish Mental Surveys are arguably the most important public health datasets nobody has ever heard of.
In the 1930s, intelligence testing was in its infancy and viewed with suspicion. The Scottish Council for Research in Education wanted to settle a scientific debate about whether the Scottish population was becoming less intelligent due to differential birth rates. They administered the Moray House Test, a standardized cognitive assessment, to virtually every Scottish child born in 1921 on June 1, 1932. They repeated this for children born in 1936 on June 4, 1947.
These two surveys created something unprecedented: nearly complete population-level cognitive data for two entire birth cohorts of an industrialized nation. The original researchers analyzed the immediate questions and then filed the records away. They sat in archives for decades.
In the 1990s, Ian Deary and his colleagues realized what they had: a time machine. The children of 1932 were now in their seventies. Their National Health Service records, mortality data, hospital admissions, and cause-of-death information were all accessible. By matching the original test scores to subsequent health outcomes, they could ask a question that had never been answerable before: does childhood intelligence predict adult health and longevity?
The first paper, published by Whalley and Deary in the BMJ in 2001, traced 2,230 participants from the 1932 Aberdeen sample. The finding was striking. A 15-point IQ disadvantage at age 11 reduced the probability of being alive at 76 to 79 percent of peers’ survival rate. For women, the effect was larger: 71 percent. For men, 83 percent.
This single statistic launched cognitive epidemiology as a field. It also raised a question that researchers have spent the past two decades trying to answer: why?
The Numbers That Followed
The Scottish finding could have been a fluke. It was not. Over the next twenty years, replication studies poured in from around the world.
The definitive synthesis came in 2011, when Calvin and colleagues published a systematic review and meta-analysis in the International Journal of Epidemiology covering 16 prospective cohort studies from the United States, United Kingdom, Sweden, Denmark, and Australia. The pooled sample included 22,453 deaths among 1,107,022 participants, with follow-up periods ranging from 17 to 69 years.
The result was extraordinarily consistent. A 1-standard-deviation advantage in childhood or early-adulthood IQ was associated with a 24 percent lower risk of death from any cause, with a 95 percent confidence interval ranging from 23 to 25 percent. The relationship held across countries, cohorts, and historical periods.
What I find most clinically interesting about the Calvin meta-analysis is how the effect survived statistical controls. Adjusting for childhood socioeconomic status barely changed the relationship at all. Adjusting for adult socioeconomic status reduced it by 34 percent. Adjusting for education reduced it by 54 percent. Adjusting for both reduced the effect substantially but did not eliminate it.
The interpretation is important: education and adult socioeconomic status mediate a meaningful portion of the IQ-longevity relationship, but they do not fully explain it. Something about cognitive ability itself, beyond the social and economic advantages it tends to bring, predicts how long you live.
A 2017 follow-up analysis of the 1936 Scottish cohort, tracking 65,765 children to age 79, refined this picture further. Top-decile childhood IQ was associated with two-thirds lower risk of respiratory-disease death and roughly half the risk of death from coronary heart disease, stroke, smoking-related cancers, digestive disease, and external causes like accidents and injury.
The dose-response curve was monotonic. There was no threshold effect. Every additional IQ point was associated with incrementally better survival, all the way up the distribution.
The 1.2-Million-Person Swedish Study That Changed the Game
While the Scottish data established the longevity finding, a series of Swedish studies provided the largest and most rigorous tests of the relationship.
Sweden’s universal male military conscription system created a remarkable scientific resource: nearly every Swedish man born between 1950 and 1976 underwent standardized cognitive testing at age 18 or 19. By the early 2000s, when these men were reaching middle age, researchers could analyze IQ-mortality relationships in samples large enough to detect even small effects with high precision.
Aberg and colleagues’ 2009 study, published in Proceedings of the National Academy of Sciences, examined 1,221,727 Swedish male conscripts including 268,496 full-sibling pairs and 1,432 monozygotic twin pairs. The sibling and twin design was crucial because it allowed researchers to control for shared family environment. If two brothers raised by the same parents in the same house show different cognitive scores and different mortality outcomes, the difference cannot be attributed to family circumstances. It must reflect something about the individual.
The Swedish data confirmed the IQ-longevity relationship and added an important wrinkle: cardiovascular fitness at age 18 predicted IQ at age 18, and this relationship held even within identical twin pairs. The integration of fitness and cognition in young adulthood pointed toward a deeper biological connection that we will discuss shortly.
A separate Swedish analysis by Batty and colleagues followed approximately one million conscripts into middle age and confirmed the dose-response relationship between adolescent IQ and mortality. Hemmingsson and colleagues, using a different Swedish cohort, calculated a hazard ratio of 1.15 per one-point decrease on the 9-point Swedish stanine scale, meaning each step down the cognitive distribution was associated with a 15 percent increase in mortality risk.
The Wisconsin Longitudinal Study provided the American replication: 10,317 Wisconsin high school graduates of 1957, tracked from age 18 to 68. The pattern matched the Scottish and Swedish findings.
A 2023 meta-analysis in Intelligence covering newer cohorts produced a pooled hazard ratio of 0.79 (95 percent confidence interval: 0.76 to 0.81) per standard deviation of IQ. A 2025 multilevel meta-analysis in Communications Psychology found that a 15-point lower IQ was associated with a 22 percent higher risk of physical and mental illness, confirming the relationship in the most recent available data.
Why? Four Theories That Are Not Mutually Exclusive
The IQ-longevity relationship is one of the most robust findings in epidemiology. The question that has consumed researchers for two decades is: what is the mechanism?
Four major hypotheses dominate the literature, and current evidence suggests all four contribute to varying degrees.
The first is the health literacy hypothesis. People with higher cognitive ability are better at understanding medical information, navigating healthcare systems, adhering to medication regimens, recognizing warning signs of illness, and making complex risk-benefit decisions. They are more likely to engage in preventive behaviors like screenings, vaccinations, and routine check-ups. They are less likely to smoke, less likely to develop alcohol use disorders, and more likely to maintain healthy body weight. Linda Gottfredson, in her influential 2004 paper, argued that modern healthcare effectively rewards intelligence: those who can decipher prescription labels, follow complex treatment plans, and weigh evidence about competing therapies survive longer than those who cannot.
The second is the socioeconomic cushion hypothesis. Higher cognitive ability predicts higher educational attainment, which predicts higher income, which predicts better access to healthcare, safer working conditions, healthier neighborhoods, lower psychosocial stress, and reduced exposure to environmental toxins. From this perspective, IQ does not directly affect longevity. It affects the social and economic circumstances that determine longevity. The Calvin meta-analysis showed that adult socioeconomic status and education together account for roughly half of the IQ-mortality relationship, providing strong support for this mechanism.
The third is what Ian Deary has called the system integrity hypothesis. The basic idea is that IQ is not just a cognitive capacity but also a marker of how well the body’s biological systems are functioning overall. The brain is the most metabolically demanding organ in the body. It depends on cardiovascular, immune, and metabolic systems working efficiently. A person whose neural systems develop and operate at high efficiency may also have circulatory, immune, and cellular systems operating at high efficiency. Cognitive performance, in this view, is a kind of biological fitness signal: it reflects the health of the whole organism. The Aberg PNAS study, showing that cardiovascular fitness predicts IQ even within identical twin pairs, supports this interpretation.
The fourth, and perhaps most provocative, is the genetic hypothesis. In 2016, Arden and colleagues published an analysis in the International Journal of Epidemiology using three genetically informative samples: 377 male U.S. WWII twin pairs, 246 Swedish twin pairs, and 784 Danish twin pairs. By comparing concordance rates between identical and fraternal twins for both intelligence and lifespan, they could partition the IQ-longevity relationship into genetic and environmental components.
Their result was striking: approximately 95 percent of the IQ-longevity correlation was attributable to shared genetic influence. Only about 5 percent reflected environmental factors that affected both IQ and longevity.
This finding does not mean that IQ causes long life or that long life causes high IQ. It means that whatever genes contribute to higher cognitive ability also tend to contribute to longer lifespan. There is a deep biological link between the substrate that supports intelligence and the substrate that supports longevity, and that link is largely written into our DNA.
The contemporary scientific consensus is that all four mechanisms operate simultaneously. Health literacy explains some of the relationship. Socioeconomic factors explain a substantial portion. System integrity provides a biological framework that connects cognitive efficiency to physiological efficiency. And genetics underlies a meaningful share of why the link exists at all.
What About Specific Causes of Death?
The all-cause mortality finding is interesting, but the cause-specific patterns are more clinically informative.
Cardiovascular disease accounts for the largest portion of the IQ-mortality relationship. Multiple cohort studies have found that childhood IQ predicts adult risk of coronary heart disease and stroke even after controlling for traditional cardiovascular risk factors. The Scottish 1936 cohort showed roughly half the cardiovascular mortality risk in the highest-IQ decile compared to the lowest.
Smoking-related cancers show a similar pattern, though here the mechanism is more directly behavioral: lower-IQ individuals are more likely to take up smoking and less likely to quit, and the resulting differential exposure produces differential cancer mortality decades later.
Respiratory disease shows the largest cause-specific effect. The 1936 Scottish cohort showed two-thirds lower respiratory mortality in the top IQ decile, reflecting a combination of reduced smoking, better occupational protection from inhaled hazards, and possibly differences in immune function.
External causes of death, including accidents, injuries, and violence, show meaningful IQ-related differences as well. Higher cognitive ability is associated with better risk assessment, fewer impulsive behaviors, safer driving habits, and reduced involvement in high-risk environments.
Dementia represents an interesting case. Higher childhood IQ predicts lower dementia risk in old age, but the relationship operates partly through “cognitive reserve,” the brain’s ability to tolerate age-related pathology before clinical symptoms emerge. People with higher baseline cognitive function can sustain more neuropathological damage before demonstrating dementia symptoms, which delays clinical diagnosis and effectively extends cognitive lifespan.
Suicide is the one notable exception to the general pattern. Some studies have found a small positive association between very high IQ and suicide risk, though the effect is small relative to the protective effects across other causes of death, and the broader pattern remains overwhelmingly favorable for higher cognitive ability.
The Dose-Response Curve: Higher Is Better, All the Way Up
One of the most clinically important features of the IQ-longevity relationship is that it is graded across the entire distribution, with no apparent threshold.
This matters because the cognitive epidemiology literature is sometimes summarized as “very low IQ is bad for your health.” That summary is true but incomplete. The data show that the difference between an IQ of 100 and 115 produces roughly the same proportional mortality benefit as the difference between 85 and 100, and that benefits continue accruing well into the high-average and superior ranges.
The curve appears to flatten only at extremely high IQ levels, around 163 and above, though the sample sizes at these extremes are too small for confident estimates.
This dose-response pattern has practical implications. It means that knowing you are slightly above average rather than average corresponds to a meaningful health advantage. It means that knowing you are below average rather than average corresponds to a meaningful health risk that warrants additional attention to lifestyle factors. And it means that the question “is my IQ a health-relevant variable?” has the same answer regardless of where you fall on the distribution: yes.
The proportional effect sizes are substantial. A 15-point IQ difference produces a hazard ratio comparable to the difference between never-smoking and smoking a pack a day for several decades. It is comparable to the difference between regular moderate exercise and a sedentary lifestyle. It is comparable to the cardiovascular benefit of treating moderate hypertension.
In other words, your IQ is genuinely in the same league as the established health metrics that doctors monitor obsessively. The difference is that almost no one knows their cognitive baseline, while most people know their blood pressure.
What This Means for You
Everything I have described in this article points toward a single practical implication: cognitive ability is a health-relevant variable that most adults have never had measured.
You probably know your blood pressure. You probably have some sense of your resting heart rate. If you have engaged with the longevity culture of the past few years, you may have measured your VO2 max, your biological age, or your cardiovascular calcium score. These are all reasonable things to know about yourself.
Childhood IQ predicts mortality more strongly than total cholesterol does. Adolescent IQ predicts mortality at effect sizes comparable to those of major lifestyle factors. Adult cognitive function predicts cognitive trajectory in aging and dementia risk decades into the future.
And yet most adults have never had a validated cognitive baseline established. The school-administered tests of childhood, if they happened at all, are long gone or unreliable. The informal cognitive estimates we make about ourselves through professional achievement and educational attainment are confounded by motivation, opportunity, and luck. The internet quizzes most people have taken are not psychometrically valid.
Establishing a current cognitive baseline accomplishes several things that matter for your long-term health. It gives you a quantitative measure of cognitive function at this point in your life, against which you can track future change. It identifies your cognitive profile, your relative strengths and weaknesses across verbal reasoning, perceptual reasoning, working memory, and processing speed, which lets you target lifestyle interventions to the systems most relevant to your individual brain. And it provides early warning signal capacity. If your scores at sixty are meaningfully lower than they were at forty, that is information your physician needs.
The longevity research suggests that knowing your cognitive baseline is not optional self-knowledge. It is closer to knowing your blood pressure: useful to have, problematic not to have, and increasingly easy to obtain through validated online cognitive assessments that meet professional psychometric standards.
What You Can Actually Do With This Information
If childhood IQ predicts mortality decades later, the natural question is whether anything can be done about it. The answer is more hopeful than the headlines suggest.
The IQ-longevity relationship exists at the population level and reflects average tendencies, not individual destinies. Even within IQ ranges, lifestyle factors substantially modify the actual mortality risk. A person with average cognitive ability who exercises regularly, does not smoke, maintains healthy body weight, manages cardiovascular risk factors, and engages in lifelong intellectual challenge can outperform the population baseline for their cognitive group by a meaningful margin.
The protective behaviors that the longevity literature identifies are largely the same regardless of cognitive ability: regular aerobic exercise, adequate sleep, social connection, intellectual engagement, cardiovascular health management, and avoidance of major preventable risks like smoking and excessive alcohol. These are within reach for almost everyone.
What knowing your cognitive baseline adds is targeting and motivation. If your verbal reasoning is your strength but your processing speed is your weakness, deliberate engagement with novel cognitively demanding activities can specifically exercise the neural systems supporting processing speed. If your working memory is in a relatively high range, you have additional cognitive reserve to invest in protecting through deliberate use. The general principles of cognitive longevity apply universally; the specific applications benefit from knowing your individual profile.
This is also where the genetic finding from Arden and colleagues becomes useful rather than discouraging. The fact that 95 percent of the IQ-longevity link reflects shared genetic influence does not mean that 95 percent of your individual longevity is genetically determined. Heritability operates at the population level. At the individual level, environmental factors retain enormous power. People with identical genetic profiles can live dramatically different lifespans depending on their behavior, environment, and healthcare access.
The most defensible interpretation of the cognitive epidemiology literature is this: cognitive ability is one signal of underlying biological integrity, and the correlation between IQ and longevity reflects that shared substrate. But knowing this signal early gives you the leverage to optimize the environmental and behavioral factors that determine where you actually fall within your genetic range.
Final Thoughts
In three decades of administering cognitive assessments, I have watched the field shift from treating IQ as a fixed cognitive label to understanding it as a dynamic biological signal that predicts a wide range of life outcomes, including how long you will live.
The Scottish schoolchildren of 1932 inadvertently gave us one of the most important datasets in modern epidemiology. Their children and grandchildren, tested in 1947 and beyond, confirmed and refined the picture. Today, with more than three million participants across multiple countries and follow-up periods exceeding seventy years, we can say with high confidence that intelligence and longevity are linked at population scale, and that the link operates through a combination of behavioral, socioeconomic, biological, and genetic mechanisms that we are still working to fully understand.
What this means for you, individually, is that your cognitive ability is not a curiosity. It is a health metric. Like blood pressure, like cholesterol, like cardiovascular fitness, it is a quantitative measure of underlying physiological function that predicts long-term health outcomes.
Most people have never had it properly measured. That made sense in an era when cognitive testing was expensive, time-consuming, and inaccessible outside specialized clinical settings. It makes much less sense now.
The ancient Greeks at Delphi inscribed two words above the entrance to the temple of Apollo: gnothi seauton. Know thyself. Modern cognitive epidemiology has given that instruction a quantitative dimension the ancients could not have anticipated.
Knowing yourself, in the most concrete and actionable sense, increasingly means knowing your cognitive baseline.
