Smoking and IQ: Do Smokers Have Lower IQs, and Does Smoking Lower Them?

Smoking and IQ are linked: in a study of more than 20,000 Israeli 18-year-olds, smokers scored about 6 IQ points lower than non-smokers, or about 4 points after statistical adjustment. Most of that gap appears to come from lower IQ predicting who starts smoking, while smoking in pregnancy shows no effect on a child's IQ in sibling studies, and smoking into middle age is linked to somewhat faster cognitive decline.

This page covers what the large studies of young smokers found, why sibling comparisons changed the story on smoking in pregnancy, what smoking does to the ageing brain, and why nicotine's short-term boost to attention is easy to misread.

The clearest single study comes from Mark Weiser and colleagues, who used the Israeli military's draft-board testing. Each year a random sample of recruits filled in a smoking questionnaire, and every recruit took the military's standardized cognitive tests, which the authors describe as equivalent to IQ. Of 20,221 males aged 18, 28.5% smoked at least one cigarette a day.

Unadjusted, smokers scored 0.41 standard deviations lower than non-smokers. On the IQ scale, where one "standard deviation" (the typical spread of scores around the average) is 15 points, that is roughly 6 points. After statistical adjustment the gap was 0.27 standard deviations, about 4 points. The gap grew with the number of cigarettes:

• 1 to 5 a day: 0.14 standard deviations lower, about 2 points.

• 6 to 10 a day: 0.22 standard deviations lower, about 3 points.

• 11 to 20 a day: 0.33 standard deviations lower, about 5 points.

• 21 or more a day: 0.5 standard deviations lower, about 7.5 points.

Two further findings point to direction. Young men who had not smoked by 18 but took it up between 18 and 21 already had lower scores at 18 (0.14 standard deviations, about 2 points), before they started. And among brothers who differed in smoking, the smoking brother scored 0.27 standard deviations lower. A brother comparison removes much of what families share, though not everything, since brothers share only about half their genes and can differ in many other ways.

Cohort studies that measured ability in childhood, long before anyone smoked, tell the same story. Michael Daly and Mark Egan pooled 16,653 people from two British birth cohorts born in 1958 and 1970, tested at ages 10 or 11 and followed through adulthood. Each standard deviation of higher childhood ability predicted a 0.5 percentage point lower probability of smoking and a 2.9 percentage point higher probability of quitting. Differences in education explained about half of these links, which suggests that brighter children smoke less partly because they stay in school longer.

For years, studies found that children of mothers who smoked while pregnant scored lower on intelligence tests. The difficulty is that women who smoke in pregnancy differ from those who do not in education and in their own IQ, and those differences pass to children through genes and upbringing as well. Comparing siblings whose mother smoked in one pregnancy but not the other holds most of that constant.

The largest such study used Swedish conscription tests for 205,777 men born between 1983 and 1988. Frida Lundberg and colleagues found that, across unrelated men, sons of smokers scored about 0.65 points lower on a nine-point scale with a standard deviation of 2, roughly 5 IQ points. Adjusting for parental characteristics cut that to 0.24 points, under 2 IQ points. Within 14,722 pairs of full brothers the difference disappeared once birth order and maternal age were taken into account. Smoking only in the second pregnancy was not linked to poorer scores in either son, which is hard to square with a direct effect of the exposure.

An American study pointed the same way. Stephen Gilman, Hannah Gardener and Stephen Buka analysed 52,919 children from the US Collaborative Perinatal Project, followed from birth to age 7 between 1959 and 1974. In sibling comparisons, smoking during pregnancy was still linked to lower birth weight and a higher chance of being overweight at 7, but its associations with intelligence and academic achievement were entirely eliminated.

A Canadian study by Sonia Kafouri and colleagues tested 503 adolescents aged 12 to 18 on 33 cognitive tasks, matching those exposed to maternal smoking in pregnancy with unexposed teenagers by maternal education and school. It found no differences in any cognitive domain.

None of this makes smoking in pregnancy safe. The same sibling studies still find lower birth weight, and the research covers IQ and test scores only. On IQ specifically, though, the best designs find that the gap is explained by family background.

The evidence here concerns change over time in middle and later life, and it is stronger than the evidence for any effect on young people's IQ. In the Whitehall II study of British white-collar civil servants, Séverine Sabia and colleagues tested 5,099 men and 2,137 women, average age 56, three times over 10 years. Male current smokers declined faster than men who had never smoked, by 0.09 standard deviations in global cognition and 0.11 in executive function over the decade. Counting participants who dropped out or died made those differences 1.2 to 1.5 times larger. Men who had quit at least 10 years earlier declined at the same rate as never-smokers, and among women smoking status made no difference.

A 0.09 standard-deviation difference is about 1.4 IQ-point equivalents over 10 years, small for any one person but measurable across a population. A 2007 meta-analysis by Kaarin Anstey and colleagues pooled 19 prospective studies of older adults, average age 74. Current smokers had a 1.79 times higher risk of Alzheimer's disease, a 1.78 times higher risk of vascular dementia and a 1.27 times higher risk of any dementia than never-smokers, and their scores on a standard cognitive screening test fell faster each year.

These are observational studies, and smokers differ from non-smokers in ways that statistics can only partly adjust for. The finding that long-term quitters look like never-smokers fits a real effect of continued smoking, plausibly through the blood vessels and lungs, but it does not prove one.

People sometimes reason that nicotine sharpens thinking, so smokers should score higher. Nicotine does have short-term effects. A meta-analysis of 41 double-blind, placebo-controlled studies by Stephen Heishman and colleagues found small to moderate gains in fine motor skill, alerting and orienting attention, short-term memory and working-memory speed (effect sizes 0.16 to 0.44) in people who were not in withdrawal. These are brief changes in alertness and speed. They do not show up as higher IQ scores in smokers, who on average score lower.

The pattern across studies has two strands. Lower IQ in childhood and adolescence predicts starting to smoke and failing to quit, partly through education. Then, decades later, continued smoking is associated with faster decline. The same two-direction shape appears with IQ and alcoholism, and cannabis has its own, separately argued evidence base, covered in our page on cannabis and IQ.

The main confounds to keep in mind:

• Education and social class: both predict smoking and test scores, and education explained about half the childhood-ability link in the British cohorts.

• Parental IQ and family background: these explain the apparent effect of smoking in pregnancy in the sibling studies.

• Sex: the Israeli and Swedish conscript studies included only men, and in Whitehall II faster decline was seen in men only.

• Survival: smokers die earlier, so studies of older adults lose the smokers who may have declined most, which tends to understate effects.

Anyone who wants help to stop smoking should talk to a doctor or pharmacist, who can explain the options available.

On average, yes. In a study of 20,221 Israeli 18-year-old men, smokers scored about 6 points lower before adjustment and about 4 points lower after it, and heavier smokers scored lower still.

The evidence suggests lower IQ mostly comes first: teenagers who later start smoking already score lower. In middle age, continued smoking is linked to somewhat faster cognitive decline in men, and long-term quitters decline at the same rate as never-smokers.

Sibling studies from Sweden and the United States found no difference in IQ between brothers or siblings whose mother smoked in one pregnancy but not the other. Smoking in pregnancy is still linked to lower birth weight, so the absence of an IQ effect does not make it safe.

Nicotine gives small, short-lived improvements in attention, reaction speed and short-term memory in laboratory studies. It does not raise IQ, and smokers score lower on average.

In the Whitehall II study, men who had quit at least 10 years earlier showed the same rate of cognitive decline as men who had never smoked, while recent quitters still declined faster on executive function.

Smokers do score lower on IQ tests on average, by about 4 to 6 points among young men, but most of the gap comes from lower IQ and less education predicting who starts and who keeps smoking. The once-feared IQ effect of smoking in pregnancy vanishes when siblings are compared. Where smoking does seem to matter is in later life, through faster decline and higher dementia risk, and the evidence that long-term quitters look like never-smokers is encouraging. If you want to see where your own reasoning skills stand today, you can take the RIOT IQ test.

1. Weiser, M., Zarka, S., Werbeloff, N., Kravitz, E., & Lubin, G. (2010). Cognitive test scores in male adolescent cigarette smokers compared to non-smokers: A population-based study. Addiction, 105(2), 358-363. [doi.org/10.1111/j.1360-0443.2009.02740.x](. doi.org

2. Daly, M., & Egan, M. (2017). Childhood cognitive ability and smoking initiation, relapse and cessation throughout adulthood: Evidence from two British cohort studies. Addiction, 112(4), 651-659. [doi.org/10.1111/add.13554](. doi.org

3. Lundberg, F., Cnattingius, S., D'Onofrio, B., Altman, D., Lambe, M., Hultman, C., & Iliadou, A. (2010). Maternal smoking during pregnancy and intellectual performance in young adult Swedish male offspring. Paediatric and Perinatal Epidemiology, 24(1), 79-87. [doi.org/10.1111/j.1365-3016.2009.01073.x](. doi.org

4. Gilman, S. E., Gardener, H., & Buka, S. L. (2008). Maternal smoking during pregnancy and children's cognitive and physical development: A causal risk factor? American Journal of Epidemiology, 168(5), 522-531. [doi.org/10.1093/aje/kwn175](. doi.org

5. Kafouri, S., Leonard, G., Perron, M., Richer, L., Séguin, J. R., Veillette, S., Pausova, Z., & Paus, T. (2009). Maternal cigarette smoking during pregnancy and cognitive performance in adolescence. International Journal of Epidemiology, 38(1), 158-172. [doi.org/10.1093/ije/dyn250](. doi.org

6. Sabia, S., Elbaz, A., Dugravot, A., Head, J., Shipley, M., Hagger-Johnson, G., Kivimaki, M., & Singh-Manoux, A. (2012). Impact of smoking on cognitive decline in early old age: The Whitehall II cohort study. Archives of General Psychiatry, 69(6), 627-635. [doi.org/10.1001/archgenpsychiatry.2011.2016](. doi.org

7. Anstey, K. J., von Sanden, C., Salim, A., & O'Kearney, R. (2007). Smoking as a risk factor for dementia and cognitive decline: A meta-analysis of prospective studies. American Journal of Epidemiology, 166(4), 367-378. [doi.org/10.1093/aje/kwm116](. doi.org

8. Heishman, S. J., Kleykamp, B. A., & Singleton, E. G. (2010). Meta-analysis of the acute effects of nicotine and smoking on human performance. Psychopharmacology, 210(4), 453-469. [doi.org/10.1007/s00213-010-1848-1](. doi.org

Hero image: Cigarette butts in ash, by علاء, CC BY-SA 3.0. Via Wikimedia Commons.