Share this article on:

IQ in Early Adulthood and Mortality By Middle Age: Cohort Study of 1 Million Swedish Men

Batty, G Davida,b; Wennerstad, Karin Modigc; Smith, George Daveyd; Gunnell, Davidd; Deary, Ian J.b; Tynelius, Perc; Rasmussen, Finnc

doi: 10.1097/EDE.0b013e31818ba076
Intelligence and Health: Original Article

Background: High premorbid IQ test scores are related to a reduced rate of later total mortality, although little is known about the shape of this association (ie, dose-response versus threshold), or the role of mediating and confounding factors in explaining it. Additionally, the link between IQ and cause-specific mortality has been little explored.

Methods: A cohort of over 1 million Swedish men who underwent IQ testing at military service conscription at about 18 years of age was followed for mortality experience until middle age.

Results: An average of 20 years of follow-up gave rise to 14,498 deaths in an analytical sample of 994,262. In basic analyses adjusting for age, year of birth, and conscription testing center, lower IQ scores were associated with an elevated risk of all-cause mortality (HRper 1-SD decrease in IQ; 1.32; 95% confidence interval = 1.30–1.34). This relation was incremental across the full IQ range, and was robust to adjustment for indicators of childhood social circumstances. The association did not appear to be mediated by factors measured concurrent with IQ (blood pressure, body mass index, or cigarette smoking), nor was it attributable to reverse causality. However, controlling for education (a close correlate of IQ) led to marked attenuation. IQ was also associated with mortality from accidents, coronary heart disease, and suicides, but not cancer.

Conclusions: In this large cohort we found a robust stepwise relation between early adult IQ and risk of total and accident mortality in men.

From the aMRC Social & Public Health Sciences Unit, University of Glasgow, Glasgow, UK; bCentre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK; cDepartment of Public Health Sciences, Karolinska Institute, Stockholm, Sweden; and dDepartment of Social Medicine, University of Bristol, Bristol, UK.

Submitted 3 April 2007; accepted 22 April 2008.

The Medical Research Council (MRC) Social and Public Health Sciences Unit receives funding from the UK MRC and the Chief Scientist Office at the Scottish Government Health Directorates. The Centre for Cognitive Ageing and Cognitive Epidemiology is supported by the UK Biotechnology and Biological Sciences Research Council, the Engineering and Physical Sciences Research Council, the Economic and Social Research Council, the MRC, and the University of Edinburgh as part of the cross-council Lifelong Health and Wellbeing initiative. David Batty is a UK Wellcome Trust Fellow (WBS U.1300.00.006.00012.01).

Correspondence: F. Rasmussen, Child and Adolescent Public Health Epidemiology Group, Department of Public Health Sciences, Karolinska Institute, SE-17176 Stockholm, Sweden. E-mail: finn.rasmussen@ki.se.

In the emerging field of cognitive epidemiology,1 mental ability (IQ) measured between childhood and early adulthood (which can therefore be regarded as being premorbid) is inversely associated with all-cause mortality rates in cohorts followed for up to 6 decades.2 Most studies have appeared in the last 5 years and unresolved questions remain.

First, the shape of the IQ–mortality relation, important in assessing causation,3 is unclear. A recent systematic review found that of 9 studies in which the link between early life IQ and all-cause mortality was investigated,2 all reported an inverse relation. However, only 5 of these examined dose–response effects, and results were discrepant.4–8 Two4,5 found a stepwise increase in mortality risk in successively lower quartiles of IQ, whereas in 3 others,6–8 there was a suggestion of a threshold effect. Second, the inverse IQ–mortality relation does not appear to be confounded by early life socioeconomic position.5,7,8 This observation has prompted speculation about mechanisms that may explain this association. High IQ scores in early life are associated with a reduced prevalence of a range of established risk factors for premature mortality in adult life, including smoking,9–12 obesity,11,13–14 high blood pressure,12,15,16 and socioeconomic deprivation.17,18 There is therefore a strong prima facie case that such risk indices might mediate the IQ–mortality association.2,19 However, the few studies exploring this issue have, with few exceptions,20 focused on the role of the study participants’ adult socioeconomic position as a mediating variable in the IQ–death relation and have reached differing conclusions about its role.6–8,21,22 Third, few studies have examined the relation between IQ and cause-specific mortality.2

Very few studies include data on IQ, mortality, and risk factors other than socioeconomic deprivation with which to examine their mediating role. One exception is a cohort of male Swedish conscripts that uses data from over 1 million Swedish men who had their IQ assessed during medical examination at military service conscription in late adolescence. These data have been linked to a series of population registers, providing a dataset orders of magnitude larger than those for existing studies in this field.

Back to Top | Article Outline

METHODS

Study Participants and Record-Linkage of Registers

The record linkage methodology has been reported in detail elsewhere.8,23–25 In brief, the study cohort comprised all nonadopted men born in Sweden from 1950–1976 for whom both biologic parents could be identified in the Multi-Generation Register. Using unique personal identification numbers, we were able to link the Multi-Generation Register with the Military Service Conscription Register, the Cause of Death Register, Population and Housing Censuses records (1960, 1970, and 1990), and the Register of Education (1990–2001). This resulted in 1,346,545 successful matches, and a dataset containing a range of physiologic, social, psychologic, and behavioral characteristics of the men (and their parents) together with ascertainment of vital status. Study approval was obtained from the Ethics Committee, Stockholm.

Back to Top | Article Outline

Conscription Examination

The Swedish military service conscription examination involves a structured, standard medical assessment of physical health, mental status, and cognitive function (IQ). During the years covered by this study, this examination was required by law; only men of foreign citizenship or those with a severe medical condition or disability were excused from participation. The present dataset covers examinations from 1 January 1970 to 30 September 1994. Average age at examination was 18.3 years (range = 16.0–26.0 years).

During the conscription examination, IQ was measured by 4 subtests representing logical, spatial, verbal, and technical abilities.26 All test scores—including a global IQ score derived from the 4 subtests—were standardized to give a Gaussian-distributed score between 1 and 9. Higher values indicate greater intellectual capacity. The logical test measures the capacity to understand written instructions and apply them to solving a problem. Items from the spatial test depicted a plan drawing of an object in its preassembled, 2-dimensional state. Respondents were required to identify, from a series of drawings of fully assembled, 3-dimensional objects, which it represented. The verbal test measures the knowledge of synonyms. Here, the respondent is required to determine which of 4 alternatives is the synonym of a given word. The technical test measures knowledge of chemistry and physics and implies the assessment of a component of general knowledge. All tests are in the form of written questionnaires.

Systolic and diastolic blood pressure, height, and weight were also measured during the conscription examination by either a nurse or physician using standard protocols. Using height and weight, body mass index ([BMI], weight/height2) was computed. We used height as a proxy for early-life socioeconomic position, shorter stature being associated with disadvantaged social circumstances.27 Data on cigarette smoking habits were obtained in a subgroup of men who took part in the conscription examination between 1969 and 1970 (categorized into 5 groups: nonsmoker; 1–5; 6–10; 11–20; and >20 cigarettes per day), after which such enquiries were dropped. To ascertain comorbidity by the time of conscription we used 2 sources of data: the Cancer Register (for any mention of cancer) and the conscription examination itself. During the latter, a short physician interview was used to identify somatic and psychiatric illnesses, if any (codes as: psychoses [International Classification of Disease {ICD}8/9: 290–299], neuroses [ICD8/9: 300–316], mental disability [ICD8/9: 317–319], and cardiovascular disease [ICD8/9: 390–459]).28

The highest socioeconomic index of either parent was based on census assessments in the 1960s and 1970s (5 categories: nonmanual [high/intermediate], nonmanual [low], skilled, unskilled, and other). The men's highest achieved education level (1990–2001) was based on 4 categories (<9 years of primary school, 9–10 years of primary school, full secondary school, and higher education). In men born before 1960, we again used the Population and Housing Censuses records (1990) to ascertain their occupational social class position, coded into the same categories as parental social class shown previously.

Back to Top | Article Outline

Ascertainment of Mortality

Using the population-wide Cause of Death Register, deaths were categorized into coronary heart disease ([CHD]; ICD8/9: 410–414; ICD10: I20–I25), nonintentional accidents (E800–E929; V01–X59), cancer (140–239; C00–D48), and suicides (E950–E959; X60–X84). The follow-up period began at the date of conscription. Data in these men were censored at the time of death, emigration, or 31 December 2001; whichever came first.

Back to Top | Article Outline

Statistical Methods

Our analyses are based on 994,262 conscripts with complete information on IQ, potentially confounding and mediating variables, and cause of death. Pearson correlation coefficients among the 4 IQ subtests were moderate to high (range = 0.48–0.70). Given that patterns of association between each of the cognitive subtest scores and mortality were similar, we report on the association between mortality outcomes and global cognitive test scores only (denoted hereafter as IQ).

We examined the proportional hazards assumption graphically for IQ in relation to total mortality and found no evidence for violation. We therefore used a series of Cox proportional hazards models29 to assess the association between global IQ and mortality. In view of the well-documented secular increases in IQ30 and the wide range of birth years (26 years) in the present analyses, we controlled for birth year. The administration of the IQ testing protocol could have varied somewhat by conscript testing center, thus, we included this factor as a potential confounder.

Hazard ratios with accompanying 95% confidence intervals (CIs) were computed, which were initially adjusted for age, year of birth, and conscription testing center. (There were no differences between the hazards ratios in these analyses and in those in which age alone was added to the multivariable model.) In addition to the computations of effect estimates across the IQ categories, we also calculated hazards ratios per standard deviation (1.93 IQ ‘units’) increase in the 9-point version of the global IQ scale. All analyses were computed using STATA version 8.1 computer software.31

Back to Top | Article Outline

RESULTS

We examined the relationships between IQ and each of the study covariates (Table 1). In comparison to their lower-IQ-scoring counterparts, higher-IQ-scoring men were taller and less likely to have a comorbidity condition or a parent in an unskilled occupation. These gradients were incremental. The average age at conscription examination was marginally greater in the higher-IQ-scoring groups and a weak inverse association was seen between IQ and BMI, but there was no strong evidence of a relation between IQ and systolic or diastolic blood pressure.

TABLE 1

TABLE 1

Back to Top | Article Outline

IQ and All-Cause Mortality

In Table 2 and Figure 1 we report on the relation between IQ and subsequent mortality risk. A mean of 19.9 years of follow-up (range = 0.01–32.3 years) of 994,262 men gave rise to 14,498 deaths. In the basic model (adjustment for age, year of birth, and conscription testing center) there was an inverse relation between IQ and total mortality (hazards ratioper 1-SD decrease in IQ = 1.32; 95% CI = 1.30–1.34). This gradient was apparent across the full IQ range: each decrease in the 9-category IQ scale was associated with an elevation in death rate. With the highest IQ group as a reference, the mortality rate in the lowest-IQ-scoring group was over 3 times greater (3.33; 2.94–3.77). When we adjusted individually for indicators of early life socioeconomic position (height and parental occupational social class) these gradients were essentially unchanged. Similarly, controlling for physiologic factors (BMI and blood pressure) did not result in appreciable attenuation. When these potential covariates were simultaneously added to a multivariable model (1.26; 1.24–1.28), the effect estimate for a 1-SD decrease in IQ was again very similar to those in the basic model. However, after additional control for education in a subgroup of study members with this characteristic (985,007 men; 9760 deaths), the IQ–mortality association was clearly weakened. Effect estimates from the fully adjusted model were unchanged when we excluded deaths in the first 5 years of follow-up.

TABLE 2

TABLE 2

FIGURE 1.

FIGURE 1.

A subgroup of men took part in the earliest conscription examination in 1970 (n = 34,314; 1008 deaths) when enquiries were made about cigarette smoking habits. IQ was incrementally related to smoking habit across the range of IQ scores, whereby the proportion of smokers in the lowest cognition group (69%) was twice that seen in the highest (35%). Having these data allowed us to examine the mediating effect, if any, of smoking in the IQ–mortality gradient. The IQ–mortality relation for the basic model (age, year of birth, conscription test center) in this subgroup was 1.27 (1.20–1.35), with a modest attenuation following control for smoking (1.22; 1.15–1.30).

We conducted further subgroup analyses of men who were 29 years of age or older by 1990 who had information on later life occupational social class together with covariate data (n = 388,417; 5829 deaths). As with smoking, IQ was incrementally associated with social class across the full array of IQ results, such that proportion of unskilled workers in the lowest cognition group (44%) was 8 times larger than that seen in the highest (5%). There was some attenuation of the IQ–mortality gradient when this variable was added to the multivariable model (1.28; 1.25–1.32) compared with the relation seen after the basic adjustments in this subgroup (1.37; 1.34–1.40), but it was not considerable.

Back to Top | Article Outline

IQ and Cause-Specific Mortality

We also examined the relation of IQ with cause-specific mortality outcomes: CHD (n = 665 deaths; 5% of all deaths), accidents (3954 deaths; 27%), suicide (3335 deaths; 23%), cancer (2061 deaths; 14%), and “other” causes (4483; 31%). The associations of CHD25 and suicide23 with IQ in this cohort have been presented before. We have also reported on the relation between IQ and first notification of a cancer, most of which were nonfatal malignancies (10,273 cancer registrations with 51 deaths).24 The results for mortality due to cancer and accidents have therefore not been previously shown, whereas those for CHD and suicide are included here for the purposes of comparison. The 2 categories at opposing ends of the IQ spectrum were combined owing to the low number of CHD deaths.

In Table 3 we present the multiple-adjusted relation of IQ with each of the mortality outcomes with and without education in the model. Adjustment for individual covariates had a similar modest impact of IQ–mortality effect estimates to those seen for multiple adjustments (results available on request from the corresponding author). In the multiple-adjusted analyses that did not include education, the magnitude of the relation between IQ and each of the mortality outcomes varied. The strongest inverse association was seen for “other” deaths (1.41; 1.36–1.45), followed by CHD (1.31; 1.21–1.42), accidents (1.22; 1.18–1.27), and suicide (1.22; 1.18–1.26). These cause-specific outcomes generally revealed stepwise relations with IQ across the full range of scores, the only exception being CHD where there was a suggestion of a threshold effect at about an IQ score of 4. Controlling for the subjects’ educational level markedly weakened these gradients. There was no apparent IQ–cancer relation. The hazards ratios from Table 3 (with the exception of the null IQ–cancer result) are depicted in Figure 2.

TABLE 3

TABLE 3

FIGURE 2.

FIGURE 2.

Back to Top | Article Outline

DISCUSSION

This cohort of almost 1 million Swedish men was well characterized for IQ, childhood socioeconomic position risk factors, and mortality. We found an inverse relation between IQ and all-cause mortality. This association was strong, incremental across the broad IQ range, apparently independent of early life socioeconomic position, and risk factors for premature mortality, and did not appear to be explained by reverse causality. Controlling for BMI and blood pressure had little impact on the IQ–mortality gradient. This is unsurprising given the weak associations between these variables and IQ in our analyses. Cause-specific mortality (CHD, suicide, accidents) revealed similar associations to those for IQ and total mortality. In all of our analyses, adding education to the multivariable model led to pronounced attenuation of the IQ–mortality relation. However, in this dataset (r = 0.49) and others,32 education and IQ are moderately to highly positively correlated. The inclusion of the subject's own education in the multivariable models is therefore debatable: on the one hand, heavy attenuation by education may signal a pathway through which IQ might operate to influence mortality; on the other, the IQ–education correlation raises issues of colinearity.

Back to Top | Article Outline

Alternative Explanations: Reverse Causality, Confounding, and Selection Bias

In middle- and older-age persons, chronic somatic disease, such as diabetes, hypertension, and atherosclerosis may lead to reduced cognition.33–35 An advantage of utilizing IQ data ascertained in late adolescence, therefore, is the lower prevalence of such conditions that might impact on both IQ and later mortality risk, so leading to a spurious negative IQ–mortality effect. We further addressed the issue of reverse causality by controlling for comorbidity and excluding persons who died in the first 5 years of follow-up. Taking these approaches had very little impact on the strength of the IQ–mortality association.

Socioeconomic position in early life is most frequently posited as an important confounder in the IQ–mortality link,19 given its well-established relation with both IQ36,37 and mortality risk.38,39 We took account of differences across IQ scores according to socioeconomic position by adjusting for the occupational social class of the parents, the height of the study participant (both confounding variables), and their educational qualifications (a mediator). Only the adjustment for education had a pronounced effect. The specificity of the associations with several mortality outcomes but not cancer suggests that confounding is an unlikely explanation for our results (particularly confounding by socioeconomic position, which in itself is related to all of our mortality endpoints).

Selection bias would occur in the present study if the IQ–mortality gradients differed markedly between persons included in the analyses and those excluded. Approximately one quarter of the original study sample was dropped from the analytical sample and their characteristics, where available, were compared with those who were retained. Differences between the groups in terms of IQ and covariates were not marked (data not shown, but available from the corresponding author). Furthermore, the point estimate for the IQ–mortality relation (age-adjusted) in the excluded men (HRper SD decrease in IQ = 1.32; 95% CI = 1.27 – 1.39) was the same as in the analytical sample (1.32; 1.30–1.34), suggesting an absence of selection bias.

Back to Top | Article Outline

Plausible Mechanisms

We have previously outlined 5 plausible mechanisms to explain these associations,2 2 of which, at least partly, could be tested here: adult socioeconomic advantage (higher IQ leads to educational success, high social status, and well-remunerated employment), and improved disease and injury prevention (higher IQ leads to behaviors more conducive to health, such as not smoking, higher levels of physical activity, prudent diet, seatbelt use etc.). Given the recently reported relations between early life IQ and behaviors not assessed in the present study—including physical activity,9,40 food intake,40,41 and heavy drinking42,43—examination of their role is also warranted. Other possible mechanisms include better disease/injury management (high IQ results in the superior management of serious illness/injury such as heart disease or head trauma via increased uptake of appropriate healthcare); reduced psychiatric disease (high-IQ-scoring persons have a lower risk of later psychiatric illness and therefore subsequent mortality); and “body system integrity” (function of the neurologic system, which may correlate with function of other physiologic systems). Examination of the importance of these mechanisms as explanations for the IQ–death relation is warranted.

Back to Top | Article Outline

Comparison With Previous Studies

In a recent review,2 we identified 9 studies that had examined the relation between childhood or early adult IQ (assessment prior to age 24 years) and later mortality risk. An inverse relation was apparent irrespective of the IQ test used, the era of data collection (ie, pre- or postwar), or study population (although only apparently white groups were examined). Given the similar study design, it is possible to draw direct comparison between our results and those of 1 previous study. Hemmingsson et al8 also sampled Swedish conscripts (n = 49,323) with a 2% overlap with our study population. Their point estimate for a single category decrease in IQ score in relation to death (1.15; 1.13–1.18) was exactly the same as our own (1.15; 1.14–1.16; Table 2). The confidence intervals in the present analyses were tighter owing to the larger numbers. Also consistent with our results, Hemmingson et al8 found no evidence for confounding by early socioeconomic position and some modest attenuation after adjusting for differences across the IQ groups in later life social circumstances.

Two studies reported an incremental association between early IQ and later mortality4,5 as we have, although this is not a universal finding.6–8 In a further study, a dose–response IQ–death gradient was evident even in a nonrepresentative sample of intellectually gifted children followed into older age.44 In the 1946 British Birth Cohort7 (the only study to control for an adult risk factor other than socioeconomic disadvantage), adjusting for smoking in early adulthood did not affect the association between childhood IQ and later mortality, consistent with our results.

There are few studies of IQ and cause-specific deaths. Previous publications from the present cohort have reported an inverse relation of both suicide23 and CHD risk25 mortality with IQ. The observation that higher IQ scores seems to confer protection against suicide risk has been made elsewhere in men but not in women.28,45,46 These studies contradict the results of geographic correlation studies,47,48 although such ecologic studies provide very weak evidence regarding causal processes. Previous studies of IQ and mortality from accidents49,50 have found similar strong gradients to those seen in our study. CHD is perhaps the most frequently examined disease-specific outcome in relation to premorbid cognition, and results are highly consistent: in a recent review9 of relevant studies,8,25,51–53 an apparent cardioprotective effect of higher IQ scores is evident, as in the present study. There is a suggestion that, in smaller scale studies than our own, the effect of IQ may be mediated via later risk factors, in particular socioeconomic disadvantage.54 Previous studies of IQ and cancer have been null6,8 or have found inverse associations.55

Back to Top | Article Outline

Study Strengths and Limitations

The strengths of this study lie in it size, which facilitated an examination of dose–response effects, and the availability of a range of collateral data that allowed us to examine the mediating effects of components of blood pressure, adiposity, and socioeconomic position on the IQ–mortality link. Among the limitations of this study is the fact that some risk factors (blood pressure, BMI, and smoking) were measured simultaneously with IQ. The extent to which these factors can be regarded as temporally on the pathways between IQ and mortality is questionable. Although these risk indices “track” from early adulthood to middle age,56 we may have underestimated their true mediating effects. In a related point, it is possible that IQ influences the extent to which, later in life, people modify their risk behaviors. Second, we did not have data on other important risk factors for total mortality such as blood lipids, blood glucose, dietary characteristics (including alcohol intake), and physical activity with which to test their possible mediating role. Third, without data on women it was not possible to examine differential IQ–mortality effects by sex. Finally, we were able to investigate IQ–mortality associations only among men age 50 or younger. Associations observed in this analysis may differ from those seen in cohorts with longer follow-up, during which time the distributions of the main causes of death may change.

Back to Top | Article Outline

Public Health Implications

There is some evidence that IQ can be increased in childhood.57 Two recent overviews of early learning and school readiness programs,58,59 one of which included only randomized trials,58 suggested that interventions resulted in important improvements on tests of reading, arithmetic ability, and general intelligence that appeared to extend to secondary (high) school ages. With a short duration of follow-up in these trials, it is not clear to what extent these improvements are maintained across the life course. Given that the general learning and reasoning ability captured by IQ tests may be important in the successful management of a person's health, a consideration of individual cognition levels may increase the success of health promotion campaigns and the health professional–client interaction.2

Back to Top | Article Outline

REFERENCES

1. Deary IJ, Batty GD. Cognitive epidemiology. J Epidemiol Community Health. 2007;61:378–384.
2. Batty GD, Deary IJ, Gottfredson LS. Premorbid (early life) IQ and later mortality risk: systematic review. Ann Epidemiol. 2007;17:278–288.
3. Bradford Hill A. The environment and disease: association or causation? Proceed Royal Soc Med. 1965;58:295–300.
4. Whalley LJ, Deary IJ. Longitudinal cohort study of childhood IQ and survival up to age 76. BMJ. 2001;322:1–5.
5. Osler M, Andersen AM, Due P, et al. Socioeconomic position in early life, birth weight, childhood cognitive function, and adult mortality. A longitudinal study of Danish men born in 1953. J Epidemiol Community Health. 2003;57:681–686.
6. Hart CL, Taylor MD, Davey Smith G, et al. Childhood IQ, social class, deprivation, and their relationships with mortality and morbidity risk in later life: prospective observational study linking the Scottish Mental Survey 1932 and the Midspan studies. Psychosom Med. 2003;65:877–883.
7. Kuh D, Richards M, Hardy R, et al. Childhood cognitive ability and deaths up until middle age: a post-war birth cohort study. Int J Epidemiol. 2004;33:408–413.
8. Hemmingsson T, Melin B, Allebeck P, et al. The association between cognitive ability measured at ages 18–20 and mortality during 30 years of follow-up–a prospective observational study among Swedish males born 1949–51. Int J Epidemiol. 2006;35:665–670.
9. Batty GD, Gale CR, Chandola T, et al. Quociente de inteligência na infância, fatores de risco e doença cardiovascular na vida adulta: revisão sistemática e dados recentes do estudo de coorte Inglês de 1958. In Alves JG, Sampaio-Carneiro M, eds. Prevenção de Doenças do Adulto na Infância e na Adolescência, Rio deJaneiro: Medbook; 2007.
10. Taylor MD, Hart CL, Davey Smith G, et al. Childhood mental ability and smoking cessation in adulthood: prospective observational study linking the Scottish Mental Survey 1932 and the Midspan studies. J Epidemiol Community Health. 2003;57:464–465.
11. Batty GD, Deary IJ, MacIntyre S. Childhood IQ in relation to risk factors for premature mortality in middle-aged persons: the Aberdeen Children of the 1950s study. J Epidemiol Community Health. 2007;61:241–247.
12. Batty GD, Deary IJ, Schoon I, et al. Mental ability across childhood in relation to risk factors for premature mortality in adult life: the 1970 British Cohort Study. J Epidemiol Community Health. 2007;61:997–1003.
13. Sorensen TI, Sonne-Holm S. Intelligence test performance in obesity in relation to educational attainment and parental social class. J Biosoc Sci. 1985;17:379–387.
14. Chandola T, Deary IJ, Blane D, et al. Childhood IQ in relation to obesity and weight gain in adult life: the National Child Development (1958) Study. Int J Obes (Lond). 2006;30:1422–1432.
15. Lindgarde F, Furu M, Ljung BO. A longitudinal study on the significance of environmental and individual factors associated with the development of essential hypertension. J Epidemiol Community Health. 1987;41:220–226.
16. Starr JM, Taylor MD, Hart CL, et al. Childhood mental ability and blood pressure at midlife: linking the Scottish Mental Survey 1932 and the Midspan studies. J Hypertens. 2004;22:893–897.
17. Deary IJ, Taylor MD, Hart C, et al. Intergenerational social mobility and mid-life status attainment: Influences of childhood intelligence, childhood social factors, and education. Intelligence. 2005;33:455–472.
18. Neisser U, Boodoo G, Bouchard T Jr, et al. Intelligence: knowns and unknowns. Am Psychol. 1996;51:77–101.
19. Batty GD, Deary IJ. Early life intelligence and adult health. BMJ. 2004;329:585–586.
20. Batty GD, Shipley MJ, Mortensen L, et al. IQ in late adolescence/early adulthood, risk factors in middle-age, and later all-cause mortality in men: the Vietnam Experience Study. J Epidemiol Community Health. 2008;62:522–531.
21. Snowdon DA, Greiner L, Kemper S, et al. Linguistic ability in early life and longevity: findings from the Nun Study. In Robine J-M, Forette B, Franceschi C, Allard M, eds. The Paradoxes of Longevity. New York: Springer; 1999;103–113.
22. O'Toole BI, Adena MA, Jones MP. Risk factors for mortality in Australian Vietnam-era national servicemen: a case-control study. Community Health Stud. 1988;12:408–417.
23. Gunnell D, Magnusson PK, Rasmussen F. Low intelligence test scores in 18 year old men and risk of suicide: cohort study. BMJ. 2005;330:167.
24. Batty GD, Modig Wennerstad K, Davey Smith G, et al. IQ in early adulthood and later cancer risk: cohort study of one million Swedish men. Ann Oncol. 2007;18:21–28.
25. Silventoinen K, Modig-Wennerstad K, Tynelius P, et al. Association between intelligence and coronary heart disease mortality: a population-based cohort study of 682 361 Swedish men. Eur J Cardiovasc Prev Rehabil. 2007;14:555–560.
26. David AS, Malmberg A, Brandt L, et al. IQ and risk for schizophrenia: a population-based cohort study. Psychol Med. 1997;27:1311–1323.
27. Gunnell D. Can adult anthropometry be used as a ‘biomarker’ for prenatal and childhood exposures? Int J Epidemiol. 2002;31:390–394.
28. Allebeck P, Allgulander C, Fisher LD. Predictors of completed suicide in a cohort of 50,465 young men: role of personality and deviant behaviour. BMJ. 1988;297:176–178.
29. Cox DR. Regression models and life-tables. J R Stat Soc [Ser B]. 1972;34:187–220.
30. Flynn JR. Searching for justice: the discovery of IQ gains over time. Am Psychologist. 1999;54:5–20.
31. StataCorp. Stata statistical software (release 8). College Station, TX: Stata corporation; 2003.
32. Deary IJ, Strand S, Smith P, et al. Intelligence and educational achievement. Intelligence. 2007;35:13–21.
33. Manolio TA, Olson J, Longstreth WT. Hypertension and cognitive function: pathophysiologic effects of hypertension on the brain. Curr Hypertens Rep. 2003;5:255–261.
34. Awad N, Gagnon M, Messier C. The relationship between impaired glucose tolerance, type 2 diabetes, and cognitive function. J Clin Exp Neuropsychol. 2004;26:1044–1080.
35. Gale CR, Deary IJ, Batty GD. Intelligence and carotid atherosclerosis in older people: cross-sectional study. J Am Geriatr Soc. 2008;56:769–771.
36. Lawlor DA, Batty GD, Morton SM, et al. Early life predictors of childhood intelligence: evidence from the Aberdeen children of the 1950s study. J Epidemiol Community Health. 2005;59:656–663.
37. Lawlor DA, Najman JM, Batty GD, et al. Early life predictors of childhood intelligence: findings from the Mater-University study of pregnancy and its outcomes. Paediatr Perinat Epidemiol. 2006;20:148–162.
38. Galobardes B, Lynch JW, Davey Smith G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. Epidemiol Rev. 2004;26:7–21.
39. Lawlor DA, Sterne JA, Tynelius P, et al. Association of childhood socioeconomic position with cause-specific mortality in a prospective record linkage study of 1,839,384 individuals. Am J Epidemiol. 2006;164:907–915.
40. Batty GD, Deary IJ, Schoon I, et al. Childhood mental ability in relation to food intake and physical activity in adulthood: the 1970 British Cohort Study. Pediatrics. 2007;119:e38–e45.
41. Gale CR, Deary IJ, Schoon I, et al. IQ in childhood and vegetarianism in adulthood: 1970 British cohort study. BMJ. 2007;334:245.
42. Batty GD, Deary IJ, MacIntyre S. Childhood IQ and life course socioeconomic position in relation to alcohol induced hangovers in adulthood: the Aberdeen children of the 1950s study. J Epidemiol Community Health. 2006;60:872–874.
43. Batty GD, Deary IJ, Schoon I, et al. Childhood mental ability in relation to alcohol drinking problems and consumption in adulthood: the 1970 British Cohort Study. Am J Public Health. (In press).
44. Martin LT, Kubzansky LD. Childhood cognitive performance and risk of mortality: a prospective cohort study of gifted individuals. Am J Epidemiol. 2005;162:887–890.
45. Andersson L, Allebeck P, Gustafsson JE, et al. Association of IQ scores and school achievement with suicide in a 40-year follow-up of a Swedish cohort. Acta Psychiatr Scand. 2008;118:99–105.
46. Jiang GX, Rasmussen F, Wasserman D. Short stature and poor psychological performance: risk factors for attempted suicide among Swedish male conscripts. Acta Psychiatr. Scand. 1999;100:433–440.
47. Voracek M. National intelligence and suicide rate: an ecological study of 85 countries. Personality & Individual Differences. 2004;37:543–553.
48. Voracek M. National intelligence, suicide rate in the elderly, and a threshold intelligence for suicidality: an ecological study of 48 Eurasian countries. J Biosoc. Sci. 2005;37:721–740.
49. Lawlor DA, Clark H, Leon DA. Associations between childhood intelligence and hospital admissions for unintentional injuries in adulthood: the Aberdeen Children of the 1950s cohort study. Am J Public Health. 2007;97:291–297.
50. Osler M, Andersen AM, Laursen B, et al. Cognitive function in childhood and early adulthood and injuries later in life: the Metropolit 1953 male birth cohort. Int J Epidemiol. 2007;36:212–219.
51. Batty GD, Mortensen EL, Nybo Andersen AM, et al. Childhood intelligence in relation to adult coronary heart disease and stroke risk: evidence from a Danish birth cohort study. Paediatr Perinat Epidemiol. 2005;19:452–949.
52. Hart CL, Taylor MD, Davey Smith G, et al. Childhood IQ and cardiovascular disease in adulthood: prospective observational study linking the Scottish Mental Survey 1932 and the Midspan studies. Soc Sci Med. 2004;59:2131–2138.
53. Hemmingsson T, Essen J, Melin B, et al. The association between cognitive ability measured at ages 18–20 and coronary heart disease in middle age among men: a prospective study using the Swedish 1969 conscription cohort. Soc Sci Med. 2007;65:1410–1419.
54. Batty GD, Shipley MJ, Mortensen L, et al. IQ in late adolescence/early adulthood, risk factors in middle-age, and later coronary heart disease mortality in men: the Vietnam Experience Study. Eur J Cardiovasc Prev Rehabil. 2008;15:359–361.
55. Deary IJ, Whalley LJ, Starr J. IQ at age 11 and longevity: results from a follow-up of the Scottish Mental Survey 1932. In Christen Y, ed. Brain and Longevity: Perspectives in Longevity. Berlin: Springer; 2003.
56. Tate RB, Manfreda J, Krahn AD, et al. Tracking of blood pressure over a 40-year period in the University of Manitoba Follow-up Study, 1948–1988. Am J Epidemiol. 1995;142:946–954.
57. Sternberg R, Grigorenko E. Environmental Effects on Cognitive Abilities. Mahwah, NJ: Erlbaum; 2001.
58. Ramey CT, Ramey SL. Early learning and school readiness: can early intervention make a difference? Merrill-Palmer Quarterly J Develop Psychol. 2004;50:471–491.
59. Barnett WS, Camilli G. Compensatory preschool education, cognitive development, and “race”. In Fish JM, ed. Race and Intelligence. Separating Science From Myth. London: Lawrence Erlbaum Associates; 2002;69–406.
© 2009 Lippincott Williams & Wilkins, Inc.