American Journal of Epidemiology Advance Access published June 15, 2009

Transcription

1 American Journal of Epidemiology Advance Access published June 15, 2009 American Journal of Epidemiology ª The Author Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please DOI: /aje/kwp132 Original Contribution Growth Trajectories and Intellectual Abilities in Young Adulthood The Helsinki Birth Cohort Study Katri Räikkönen, Tom Forsén, Markus Henriksson, Eero Kajantie, Kati Heinonen, Anu- Katriina Pesonen, Jukka T. Leskinen, Ilmo Laaksonen, Clive Osmond, David J. P. Barker, and Johan G. Eriksson Initially submitted November 18, 2008; accepted for publication April 29, Slow childhood growth is associated with poorer intellectual ability. The critical periods of growth remain uncertain. Among 2,786 Finnish male military conscripts ( ) born in , the authors tested how specific growth periods from birth to age 20 years predicted verbal, visuospatial, and arithmetic abilities at age 20. Small head circumference at birth predicted poorer verbal, visuospatial, and arithmetic abilities. The latter 2 measures were also associated with lower weight and body mass index (weight (kg)/height (m) 2 ) at birth (for a 1-standard-deviation (SD) decrease in test score per SD decrease in body size 0.05, P s < 0.04). Slow linear growth and weight gain between birth and age 6 months, between ages 6 months and 2 years, or both predicted poorer performance on all 3 tests (for a 1-SD decrease in test score per SD decrease in growth 0.05, P s < 0.03). Reduced linear growth between ages 2 and 7 years predicted worse verbal ability, and between age 11 years and conscription it predicted worse performance on all 3 tests. Prenatal brain growth and linear growth up to 2 years after birth form a first critical period for intellectual development. There is a second critical period, specific for verbal development, between ages 2 and 7 years and a third critical period for all 3 tested outcomes during adolescence. adolescent; body mass index; child; cognition; growth; intelligence Abbreviations: BMI, body mass index; CI, confidence interval; SD, standard deviation. Prematurity (1, 2) and a small body size at birth (3) predict poorer cognitive performance in subsequent life. Slow childhood growth has also been linked with poorer cognitive performance (4 12). It remains uncertain which periods of growth are the most critical. We report here the effects of growth in body size on intellectual abilities, as estimated from measurements ranging from birth to an average age of 20 years, among Finnish men conscripted into the Finnish Defense Forces. The men were participants in the Helsinki Birth Cohort Study who were born during and had monthly measurements or estimates of height, weight, and body mass index (BMI) for the period between birth and age 2 years and annual measurements or estimates for the period between ages 2 and 11 years (13 15), in addition to measurements taken at conscription. MATERIALS AND METHODS Study population The study cohort comprised men born at Helsinki University Central Hospital (Helsinki, Finland) during the period We identified 4,630 men who had birth and child welfare clinic records and were still residents of Finland in 1971, when a unique personal identification number was allocated to each member of the Finnish population. The majority of the men (77%) had also gone to school in Helsinki and had school health-care records. The cohort has been described in detail elsewhere (13 15). We were able to identify 2,786 (60%) men who served in the Finnish Defense Forces between 1952 and At Correspondence to Prof. Katri Räikkönen, Department of Psychology, Faculty of Behavioural Sciences, University of Helsinki, P.O. Box 9, Siltavuorenpenger 20 D, Helsinki, Finland ( katri.raikkonen@helsinki.fi). 1

2 2 Räikkönen et al. conscription, their mean age was 20.1 years (standard deviation (SD), 1.4; range, years). The sample of conscripts in this study tended to have been born later during the period, as the Finnish Defense Forces only began systematic testing of conscripts intellectual abilities after the mid-1950s. The conscripts studied did not differ from the cohort as a whole in terms of body size at birth, length of gestation, postnatal growth from birth to age 11 years, father s occupational status, mother s age and height at delivery, parity, or history of breastfeeding (P s > 0.06). Because not every member of the cohort went to school in Helsinki, data on growth between ages 7 and 11 years were missing for nearly 20% of the conscripts (Table 1). Thus, we also tested whether conscripts with and without missing data differed with regard to pre- and postnatal growth from birth to age 2 years. The cohort members with no missing data had been born heavier (weight: P ¼ 0.05; BMI: P ¼ 0.03) and had gained weight faster between age 6 months and age 2 years (P ¼ 0.03). The 2 groups did not differ with regard to other phases of growth in weight or BMI or in length at birth and growth in height up to age 2 years (P s > 0.36). Power calculations indicate that with 80% power and a 2-sided a level of 0.05, the 20% attrition in sample size between ages 7 and 11 years is associated with a reduction in the observable standardized b coefficient (correlation coefficient) of SDs, an effect unlikely to be of major importance. The Helsinki Birth Cohort Study was approved by the Ethics Committee of the National Public Health Institute, and military service data were linked with the permission of the Finnish Defense Command. Anthropometric data and other neonatal and childhood characteristics Data on a newborn s date of birth, weight (g), length (cm), and head circumference (cm) and the mother s age at delivery (years), date of last menstrual period, and parity (primiparous vs. multiparous) were extracted from birth records. Data on monthly changes in weight (kg) and height (cm) from birth to age 2 years and annual changes up to age 6 years were estimated from child welfare records; annual changes thereafter, up to age 11 years, were estimated from school health records, as previously described (13 15). Social class based on father s occupation (manual worker, lower middle class, upper middle class) was extracted from school, child welfare clinic, and birth records, and history of breastfeeding (yes vs. no) was extracted from child welfare clinic records. Height (cm) and weight (kg) were measured again at conscription. BMI was calculated as weight (kg) divided by height (m) squared. Intellectual abilities at conscription The ability test scores were obtained from the Finnish Defense Forces Basic Ability Test, developed by the Finnish Defense Forces Education Development Center. The obligatory test is given to all new recruits during the first 2 weeks of their military service and is used when the conscripts are selected for leadership training during their service. The test battery and its psychometric properties are described in detail elsewhere (16). In brief, the ability test battery, which is designed to measure general ability and logical thinking, is composed of verbal, arithmetic, and visuospatial reasoning subtests. Each subtest is timed and consists of 40 multiple-choice questions that are ordered by difficulty. Correct answers are summed to obtain a test score. The verbal and arithmetic subtests comprise 4 types of questions. In the verbal reasoning test, the subject has to choose synonyms or antonyms of a given word, select a word belonging to the same category as a given word pair, identify which word of a word list does not belong in the group, and discern similar relations between 2 word pairs. In the arithmetic reasoning test, the subject has to complete a series of numbers that have been arranged to follow a certain rule, to solve verbally expressed short problems, to complete simple arithmetic operations, and to choose similar relations between 2 pairs of numbers. The visuospatial reasoning subtest comprises a set of matrices containing a pattern problem with 1 part removed; it is analogous to Raven s Progressive Matrices (17). The subject is asked to decide which of the given single figures completes the matrix, and the test requires the subject to conceptualize spatial relations ranging from the very obvious to the very abstract. Statistical analysis Measurements of body size and adult intellectual abilities were converted into z scores (13 15, 18). A z score represents the difference from the mean value for the whole cohort and is expressed in SDs. First, multiple linear regression analyses were used in testing the effects of prenatal growth on intellectual abilities. Prenatal growth variables were height, weight, BMI, and head circumference at birth adjusted for gestational age. Second, for the analyses of postnatal growth, the focus was on measurements taken at ages 6 months and 2, 7, 11, and 20 years, to represent the periods of infancy (up to age 2 years), childhood (up to age 11 years), and adolescence (up to age 20 years). Thus, we used multiple linear regression analyses again to examine the effects of growth on intellectual abilities. Postnatal growth variables were standardized residuals from linear regression models of height, weight, and BMI (head circumference was measured at birth only), where body size at each time point was regressed on corresponding measures at earlier time points, creating completely uncorrelated residuals reflecting growth conditional on previous history (13 15, 18, 19). The 2-sided P values were obtained from the multiple linear regression models and were based on comparison of the regression coefficient and its standard error. In this paper, we extend the analyses of growth by also presenting growth trajectories using unadjusted mean z scores of monthly estimates of growth from birth to age 2 years, annual estimates of growth from ages 2 to 11 years, and measured growth at age 20 years according to quartiles of intellectual abilities. Analyses of growth were adjusted for father s occupational status, mother s age and height, parity, history of breastfeeding, and age at and year of intellect testing. Previous research has implicated these variables as risk factors

3 Growth Trajectories and Intellectual Abilities 3 Table 1. Childhood and Adult Characteristics of Men Born in and Conscripted Into Military Service in Helsinki, Finland, Between 1952 and 1972 Characteristic No. of Subjects Mean (SD) No. (%) At birth Weight, kg 2, (0.5) Length, cm 2, (2.0) Body mass index a 2, (1.2) Head circumference, cm 2, (1.5) Gestational age, days 2, (12.9) Mother s age at delivery, years 2, (5.4) Mother s height at delivery, cm 2, (5.8) Parity (primiparous) 2,786 1,362 (48.9) Breastfeeding (yes) 2,786 2,352 (84.4) Father s occupational status 2,679 in childhood Manual worker 1,721 (64.2) Lower middle class 613 (22.9) Upper middle class 345 (12.9) At age 6 months Weight, kg 2, (0.9) Height, cm 2, (2.3) Body mass index 2, (1.4) At age 2 years Weight, kg 2, (1.2) Height, cm 2, (3.2) Body mass index 2, (1.2) At age 7 years Weight, kg 2, (2.7) Height, cm 2, (4.8) Body mass index 2, (1.1) At age 11 years Weight, kg 2, (4.6) Height, cm 2, (5.9) Body mass index 2, (1.5) At conscription Age, years 2, (1.4) Weight, kg 2, (9.1) Height, cm 2, (6.2) Body mass index 2, (2.5) Test of intellectual ability (range, 0 40 points) Verbal reasoning 2, (8.4) Visuospatial reasoning 2, (6.1) Arithmetic reasoning 2, (10.0) Abbreviation: SD, standard deviation. a Weight (kg)/height (m) 2. for slower growth and/or poorer development of intellectual abilities (1 15). Our data showed that, in comparison with conscripts from working-class families, conscripts from lower-middle-class and upper-middle-class families scored 0.45 and 0.78, 0.40 and 0.69, and 0.63 and 0.65 SDs higher on verbal, visuospatial, and arithmetic reasoning tests, respectively (all P s < 0.001). Conscripts with taller mothers scored higher on all 3 subtests (score increases of 0.07, 0.07,

4 4 Räikkönen et al. A) Visuospatial Reasoning Test Score P < Birth Weight, kg B) D) Arithmetic Reasoning Test Score P < and 0.09 SDs per SD increase in maternal height), as did conscripts whose mothers were primiparous (score increases of 0.17, 0.13, and 0.12 SDs). All test scores increased by age (score increases of 0.21, 0.16, and 0.19 SDs per SD increase in age on the verbal, visuospatial, and arithmetic reasoning tests, respectively) and year of testing (score increases of 0.05, 0.04, and 0.05 SDs per year on the verbal, visuospatial, and arithmetic reasoning tests, respectively) (all P s < 0.001). Finally, we tested whether differences in intellectual abilities at age 20 years translated into significant differences in occupational status and educational level in adulthood by using v 2 statistics and logistic regression analyses (contrasting the lowest and highest categories of occupation and education and adjusting for father s occupational status in childhood and age at and year of intellect testing). We derived the participants highest achieved occupation and level of education, which had been recorded at 5-year intervals between 1970 and 2000 by Statistics Finland. Occupation (manual worker, 22.6%; self-employed, 5.6%; low official, 27.0%; high official, 44.8%) and education (basic/primary Visuospatial Reasoning Test Score Arithmetic Reasoning Test Score Birth Weight, kg C) or less, 39.1%; upper secondary, 26.3%; lower tertiary, 22.3%; upper tertiary, 12.3%) were categorized according to the classification system of Statistics Finland. RESULTS P < BMI at Birth, in Quintiles P < BMI at Birth, in Quintiles Figure 1. Performance on visuospatial and arithmetic reasoning tests at age 20 years among men born in and conscripted into military service in Helsinki, Finland, between 1952 and 1972, according to birth weight category (parts A and B) and quintile (cutpoints were 12.5, 13.2, 13.7, and 14.4, respectively) of body mass index (BMI; weight (kg)/height (m) 2 ) at birth (parts C and D). Shown are mean test scores adjusted for length of gestation, father s occupational status, mother s age and height at delivery, parity, history of breastfeeding, and age at and year of intellect testing. Bars, 95% confidence interval. Table 1 shows characteristics of the study sample and the distribution of the intellectual ability test scores, which could range from 0 to 40, in the study population. Prenatal growth and intellectual abilities We tested how body size at birth, adjusted for gestational age and the covariates, was associated with intellectual abilities. With each SD reduction in head circumference at birth, scores for verbal, visuospatial, and arithmetic abilities fell by 0.05 (95% confidence interval (CI): 0.00, 0.09; P ¼ 0.03), 0.05 (95% CI: 0.00, 0.09; P ¼ 0.04), and 0.07 (95% CI: 0.03, 0.11; P ¼ 0.002) SDs, respectively. Figure 1 shows that lower birth weight and BMI at birth also predicted poorer visuospatial and arithmetic abilities. With each

5 Growth Trajectories and Intellectual Abilities 5 Table 2. Change in Intellectual Abilities (in SD units) at Age 20 Years Per 1-SD Change in Length, Weight, and Body Mass Index at Birth and Per 1-SD change in Later Growth in Height, Weight, and Body Mass Index from Birth to Age 20 Years Among Men Born in and Conscripted Into Military Service in Helsinki, Finland, Between 1952 and 1972 a Intellect Test and Growth Phase (Age, years) SD reduction in birth weight, scores for these abilities fell by 0.06 (95% CI: 0.02, 0.11; P ¼ 0.008) and 0.05 (95% CI: 0.00, 0.10; P ¼ 0.03) SDs, respectively, and with each SD reduction in BMI they fell by 0.06 (95% CI: 0.01, 0.10; P ¼ 0.01) and 0.06 (95% CI: 0.01, 0.10; P ¼ 0.02) SDs, respectively. Weight (P ¼ 0.27) and BMI (P ¼ 0.49) at birth were not associated with verbal ability. Length at birth was not associated with the intellectual abilities (P s > 0.39). Postnatal growth and intellectual abilities Length/Height Weight Body Mass Index b Effect Size c 95% CI Effect Size 95% CI Effect Size 95% CI Verbal reasoning Birth , , , * 0.01, , , *** 0.05, ** 0.02, , ** 0.02, , , , , , ** 0.02, , , 0.02 Visuospatial reasoning Birth , ** 0.02, ** 0.02, ** 0.01, * 0.01, , ** 0.02, * 0.01, , , , , , , , ** 0.03, , , 0.01 Arithmetic reasoning Birth , ** 0.03, ** 0.03, , , ** 0.01, * 0.01, ** 0.02, , , , , , , , * 0.01, , , 0.01 * P < 0.05; **P < 0.01; ***P < Abbreviations: CI, confidence interval; SD, standard deviation. a All associations were adjusted for father s occupational status, mother s age and height at delivery, parity, history of breastfeeding, and age at and year of intellect testing. b Weight (kg)/height (m) 2. c SD change in test score. The associations between postnatal growth in body size and intellectual abilities in young adulthood are presented in Table 2. Between birth and age 2 years, slow growth in height predicted worse performance in verbal and visuospatial reasoning, and between ages 6 months and 2 years, it predicted worse performance in arithmetic reasoning. Between birth and age 2 years, slow weight gain predicted worse visuospatial reasoning, and between ages 6 months and 2 years, it predicted worse verbal and arithmetic reasoning. Slow gain in BMI between birth and age 6 months was also associated with worse performance on the arithmetic test. Between ages 2 and 7 years, slow linear growth was associated with worse verbal reasoning but not with performance on the other 2 subtests. Between ages 7 and 11 years, neither linear growth nor weight nor BMI gain predicted test performance. However, at conscription, a reduced z score for height in comparison with that at age 11 years was associated with worse performance on all 3 subtests. The analyses of growth are displayed graphically in Figures 2 4. These figures extend the findings further by showing the trajectories of growth of length/height, weight, and BMI over the monthly estimates from birth to age 2 years, the annual estimates from age 2 years to age 11 years, and measurements taken at conscription at an average age of 20 years according to quartiles of verbal (Figure 2), visuospatial (Figure 3), and arithmetic reasoning (Figure 4) scores. These differences in intellectual abilities at age 20 years translated into significant differences in occupational status.

6 6 Räikkönen et al. Figure 2. Unadjusted mean z scores for monthly estimates of length/height (part A), weight (part B), and body mass index (BMI; weight (kg)/ height (m) 2 ) (part C) from birth to age 2 years, annual estimates from ages 2 to 11 years, and measurements taken at age 20 years among men born in and conscripted into military service in Helsinki, Finland, between 1952 and 1972, according to quartile (lowest referring to the poorest performers) of score on the verbal reasoning test. The mean values for all subjects are set at 0, with deviations from the mean expressed as standard deviation (SD) scores. Of the men scoring in the lowest quartile on verbal, visuospatial, and arithmetic reasoning tests at age 20 years, 31.4%, 30.8%, and 33.1%, respectively, had achieved a high official occupation in adulthood; of those scoring in the highest quartile on verbal, visuospatial, and arithmetic reasoning tests at age 20 years, 81.9%, 78.9%, and 79.2%, respectively, had achieved a high official occupation in adulthood (all P s < 0.001). Similar results were obtained when the highest achieved level of education in adulthood was used as an outcome (all P s < 0.001; data not shown). These associations were not explained by father s occupation in childhood, age at intellect testing, or year of intellect testing (all P s < 0.001). DISCUSSION We found that slow prenatal growth in head circumference, slow linear growth and weight gain between birth and age 6 months and between ages 6 months and 2 years, or both predicted poorer intellectual abilities across all 3 domains of intellectual functioning in young adulthood. Slow prenatal gain in weight and in BMI also predicted poorer arithmetic and visuospatial abilities. Furthermore, slow linear growth between ages 2 and 7 years predicted poorer verbal ability. These findings were not explained by potentially confounding variables, which, consistently with previous studies (1 14), included childhood socioeconomic conditions, breastfeeding, maternal height, parity, and age, and age at and year of intellect testing. Our study thus suggests that the fetal and infancy periods are critical for intellectual abilities in subsequent life. The effect of small head circumference at birth on intellectual functioning persisted after adjustment for gestational age and therefore reflected slow brain growth in utero. This finding agrees with previous observations (7, 11). However, Figure 3. Unadjusted mean z scores for monthly estimates of length/height (part A), weight (part B), and body mass index (BMI; weight (kg)/ height (m) 2 ) (part C) from birth to age 2 years, annual estimates from ages 2 to 11 years, and measurements taken at age 20 years among men born in and conscripted into military service in Helsinki, Finland, between 1952 and 1972, according to quartile (lowest referring to the poorest performers) of score on the visuospatial reasoning test. The mean values for all subjects are set at 0, with deviations from the mean expressed as standard deviation (SD) scores.

7 Growth Trajectories and Intellectual Abilities 7 Figure 4. Unadjusted mean z scores for monthly estimates of length/height (part A), weight (part B), and body mass index (BMI; weight (kg)/ height (m) 2 ) (part C) from birth to age 2 years, annual estimates from ages 2 to 11 years, and measurements taken at age 20 years among men born in and conscripted into military service in Helsinki, Finland, between 1952 and 1972, according to quartile (lowest referring to the poorest performers) of score on the arithmetic reasoning test. The mean values for all subjects are set at 0, with deviations from the mean expressed as standard deviation (SD) scores. our findings also showed that lower birth weight and lower BMI at birth, adjusted for gestational age, were associated with worse performance on tests of visuospatial and arithmetic reasoning but not verbal reasoning. Brain growth begins early in gestation, and small head circumference at birth may reflect fetal undernutrition in early or midgestation. Lower BMI at birth may be a result of fetal undernutrition in late gestation (20). Thus, our findings may suggest that verbal reasoning might be related to a different, perhaps earlier, stage of fetal brain growth than arithmetic and visuospatial reasoning. Slow postnatal growth in height before the age of 2 years, but also later on, is an accepted indicator of adverse childhood socioeconomic conditions (21). It may reflect malnutrition, recurrent minor infections which divert nutrition away from growth (15, 22, 23), or suboptimal parenting. Our finding that slow growth during infancy predicts poor intellectual function in early adult life is consistent with the observation that men in this cohort who had slow linear growth during infancy had lower incomes 50 years later (4). These associations were specific to linear growth during infancy and were independent of the socioeconomic status of the family into which the infant had been born. One interpretation of this is that biologic processes that retard infant growth also impair cognitive development during a critical period of brain development in early postnatal life. Development of intellectual abilities could be impaired either through direct effects on brain growth or through lack of intellectual stimulation. Malnourished infants conserve energy by moving less, which triggers less environmental stimulation as well as caregiver interaction and reduces stimulation to the brain (24, 25). Whatever the underlying processes, our observations suggest that an adverse environment during infancy has major effects on intellectual function. Our data show that this translates into worse economic performance in adult life. Small body size at birth and slow infant growth predict a number of adverse health outcomes in later life, including coronary heart disease, stroke, and type 2 diabetes (13 15, 26, 27). These associations are thought to reflect the phenomenon of programming (28), whereby adverse influences, including malnutrition, during critical periods of development permanently change the structure and function of the body s organs. For most organs and systems, the critical periods occur in utero, but the brain and liver have critical periods after birth. Poor cognitive functioning and low educational achievement are related to increased risk of chronic disease and mortality (16, 29 33). Our findings thus raise a question as to whether poor educational attainment and chronic disease may share a common origin in pre- and early postnatal life. We found that short stature at conscription in comparison with that predicted by height at age 11 years was associated with worse performance in all 3 domains of intellectual functioning that were tested. One possible explanation for this is that there is another critical period for intellectual development during adolescence. Final height reflects both the speed and duration of the adolescent growth spurt and hence the timing of puberty. We had no measurements of body size between ages 11 and 20 years and could not examine this further. There were limitations to our study. We had no data with which to determine critical periods of postnatal growth in head circumference. We used postnatal growth as a marker of living conditions. Growth during infancy may be a more sensitive marker than childhood growth. An infant uses up to 40% of its energy for growth (34). This falls to approximately 2% after 2 years (34). Infants more readily divert energy away from growth to meet other needs, such as combating infection, than do older children. Therefore, the lesser effect of growth after age 2 years on later intellectual function could reflect its lesser sensitivity as a marker of living conditions, rather than the lesser importance of this phase of life for intellectual development. Persons who died before 1971 were excluded from our cohort, as were persons who had not visited the voluntary child-welfare clinics,

8 8 Räikkönen et al. although attendance was free and most children visited them on an average of 8 occasions. Loss of follow-up for reasons disqualifying men from service in the Finnish Defense Forces could not be determined. A bias towards inclusion of healthier participants might have diminished rather than increased the statistical power of our study. Our cohort was born between 1934 and Because intelligence test performance has been shown to have improved since testing began (35), our results may not be generalizable to cohorts born more recently. However, the mean-level increase in intelligence does not necessarily affect the rank-order associations. Finally, our sample comprised Caucasian men, limiting the external validity of the findings. To conclude, our study demonstrated that the fetal and infancy periods are critical for the development of intellectual abilities. There is a second critical period, specific for verbal ability, during the childhood growth phase between ages 2 and 7 years. There may also be a third critical period for the development of intellectual abilities during adolescence. ACKNOWLEDGMENTS Author affiliations: Department of Psychology, Faculty of Behavioural Sciences, University of Helsinki, Helsinki, Finland (Katri Räikkönen, Kati Heinonen, Anu-Katriina Pesonen); Institute of Health and Welfare, Helsinki, Finland (Ton Forsén, Eero Kajantie, Johan G. Eriksson); Centre for Military Medicine, Finnish Defense Forces, Lahti, Finland (Markus Henriksson, Ilmo Laaksonen); National Defense University, Finnish Defense Forces, Helsinki, Finland (Jukka T. Leskinen); MRC Epidemiology Resource Centre, University of Southampton, Southampton, United Kingdom (Clive Osmond, David J. P. Barker); and Department of General Practice and Primary Health Care, Institute of Clinical Medicine, University of Helsinki, Helsinki, Finland (Johan G. Eriksson). This work was supported by the British Heart Foundation, the Academy of Finland, the University of Helsinki, the European Science Foundation, EuroSTRESS program, the Päivikki and Sakari Sohlberg Foundation, the Finnish Diabetes Research Foundation, the Finnish Foundation for Cardiovascular Research, the Finnish Foundation for Pediatric Research, the Finnish Medical Society Duodecim, the Yrjö Jahnsson Foundation, the Signe and Ane Gyllenberg Foundation, the Juho Vainio Foundation, the Sigrid Juselius Foundation, and Finska Läkaresällskapet. Conflict of interest: none declared. REFERENCES 1. Bhutta AT, Cleeves MA, Casey PH, et al. Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis. JAMA. 2002;288(6): Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet. 2008; 371(9608): Shenkin SD, Starr JM, Deary IJ. Birth weight and cognitive ability in childhood: a systematic review. Psychol Bull. 2004; 130(6): Barker DJ, Eriksson JG, Forsén T, et al. Infant growth and income 50 years later. Arch Dis Child. 2005;90(3): Pearce MS, Deary IJ, Young AH, et al. Growth in early life and childhood IQ at age 11 years: the Newcastle Thousand Families Study. Int J Epidemiol. 2005;34(3): Gale CR, O Callaghan FJ, Godfrey KM, et al. Critical periods of brain growth and cognitive function in children. Brain. 2004;127(2): Gale CR, O Callaghan FJ, Bredow M, et al. The influence of head growth in fetal life, infancy, and childhood on intelligence at the ages of 4 and 8 years. Pediatrics. 2006;118(4): Montgomery SM, Ehlin A, Sacker A. Pre-pubertal growth and cognitive function. Arch Dis Child. 2006;91(1): Richards M, Hardy R, Kuh D, et al. Birthweight, postnatal growth and cognitive function in a national UK birth cohort. Int J Epidemiol. 2002;31(2): Silva A, Metha Z, O Callaghan FJ. The relative effect of size at birth, postnatal growth and social factors on cognitive function in late childhood. Ann Epidemiol. 2006;16(6): Heinonen K, Räikkönen K, Pesonen AK, et al. Prenatal and postnatal growth and cognitive abilities at 56 months of age: a longitudinal study of infants born at term. Pediatrics. 2008; 121(5):e1325 e Victora CG, Adair L, Fall C, et al. Maternal and child undernutrition: consequences for adult health and human capital. Lancet. 2008;371(9609): Barker DJ, Osmond C, Forsén TJ, et al. Trajectories of growth among children who have coronary events as adults. N Engl J Med. 2005;353(17): Eriksson JG, Osmond C, Kajantie E, et al. Patterns of growth among children who later develop type 2 diabetes or its risk factors. Diabetologia. 2006;49(12): Osmond C, Kajantie E, Forsén TJ, et al. Infant growth and stroke in adult life. The Helsinki Birth Cohort Study. Stroke. 2007;38(2): Tiihonen J, Haukka J, Henriksson M, et al. Premorbid intellectual functioning in bipolar disorder and schizophrenia: results from a cohort study of male conscripts. Am J Psychiatry. 2005;162(10): Raven J, Raven JC, Court JH. Manual for Raven s Progressive Matrices and Vocabulary Scales (Section 3): Standard Progressive Matrices (Including the Parallel and Plus Versions). Oxford, United Kingdom: Oxford Psychologists Press; Royston P. Constructing time-specific reference ranges. Stat Med. 1991;10(5): De Stavola BL, Nitsch D, dos Santos Silva I, et al. Statistical issues in life course epidemiology. Am J Epidemiol. 2006; 163(1): Harding JE. The nutritional basis of the fetal origins of adult disease. Int J Epidemiol. 2001;30(1): World Health Organization. Physical Status: The Use and Interpretation of Anthropometry. Report of a WHO Expert Committee. (WHO Technical Report Series no. 854). Geneva, Switzerland: World Health Organization; Rona RJ. The impact of the environment on height in Europe: conceptual and theoretical considerations. Ann Hum Biol. 2000;27(2): Grantham-McGregor SM, Powell CA, Walker SP, et al. Nutritional supplementation, psychosocial stimulation and mental development of stunted children: the Jamaican study. Lancet. 1991;338(8758):1 5.

9 Growth Trajectories and Intellectual Abilities Bigsby R, Coster W, Lester BM, et al. Motor behavioral cues of term and preterm infants at 3 months. Infant Behav Dev. 1996;19(3): Hadders-Algra M. Variability in infant motor behavior: a hallmark of the healthy nervous system. Infant Behav Dev. 2002;25(4): Eriksson JG, Forsén TJ, Kajantie E, et al. Childhood growth and hypertension in later life. Hypertension. 2007;49(6): Kajantie E, Barker DJ, Osmond C, et al. Growth before 2 years of age and serum lipids 60 years later: the Helsinki Birth Cohort Study. Int J Epidemiol. 2008;37(2): Barker DJ. In utero programming of chronic disease. Clin Sci (Lond). 1998;95(2): Batty GD, Gale CR, Mortensen LH, et al. Pre-morbid intelligence, the metabolic syndrome and mortality: the Vietnam Experience Study. Diabetologia. 2008;51(3): 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(11): 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(5): Hemmingsson T, von Essen J, Melin B, et al. The association between cognitive ability measured at ages and coronary heart disease in middle age among men: a prospective study using the Swedish 1969 conscription cohort. Soc Sci Med. 2007;65(7): Batty GD, Deary IJ, Gottfredson LS. Premorbid (early life) IQ and later mortality risk: systematic review. Ann Epidemiol. 2007;17(4): Butte N. Fat intake of children in relation to energy requirements. Am J Clin Nutr. 2000;72(5):1246S 1252S. 35. Neisser U, Boodoo G, Bouchard T Jr, et al. Intelligence: knowns and unknowns. Am Psychol. 1996;51(2):