American Journal of Epidemiology Advance Access originally published online on June 6, 2006
American Journal of Epidemiology 2006 164(2):110-121; doi:10.1093/aje/kwj193
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original Contribution |
Developmental Origins of Midlife Physical Performance: Evidence from a British Birth Cohort
1 Medical Research Council National Survey of Health and Development, Department of Epidemiology and Public Health, Royal Free and University College London Medical School, London, United Kingdom
2 Medical Research Council Epidemiology Resource Centre, University of Southampton, Southampton, United Kingdom
Correspondence to Professor Diana Kuh, Medical Research Council National Survey of Health and Development, Department of Epidemiology and Public Health, Royal Free and University College London Medical School, Gower Street Campus, 1-19 Torrington Place, London WC1E 6BT, United Kingdom (e-mail: d.kuh{at}nshd.mrc.ac.uk).
Received for publication June 10, 2005. Accepted for publication January 19, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The authors hypothesized that 1) physical growth, as a marker of the early development of muscle fibers, and 2) advanced childhood motor and cognitive abilities, as markers of central nervous system development, would be positively related to midlife standing balance and chair rising, independently of later life experiences. They tested these hypotheses in a representative British sample of 1,374 men and 1,410 women aged 53 years in 1999 with prospective childhood measures of heights and weights, age at first standing and walking, cognitive ability, and motor coordination. Weight gain before age 7 years was positively related to adult performance in men but not women, independently of later body size, social class, physical activity, and health status. Attainment of motor milestones at the modal age and higher scores on tests of cognitive ability and motor coordination were associated with better performance, independently of other factors. This study is the first to show that childhood growth and development affect midlife performance; prevention of disability and frailty in later life may need to start early.
aging; cohort studies; growth and development; longitudinal studies; middle aged; psychomotor performance
Abbreviations: CI, confidence interval; ln, natural logarithm; SD, standard deviation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Tests of physical performance, such as rising from a sitting to a standing position ("chair rising") and standing balance, are summary markers of functional status in older people and of underlying biologic aging processes (1
), and these tests are strong predictors of frailty, disability, and subsequent death (2–7). To understand the dynamics of frailty and disability, we need to study physical performance before old age and consider its lifetime determinants; this may offer possibilities for intervention in earlier life to develop and maintain peak levels of performance or to slow its rate of decline with age. We have previously shown that standing balance and chair rising were strongly related to current weight, physical activity, health status, and socioeconomic conditions in a British cohort of men and women aged 53 years (8).
Far less is known about the role of early life factors with regard to these aspects of adult physical performance. There is growing evidence that early growth and development affect other functional measures, such as muscle strength (9–11), and adult health more generally (12–14). In this paper, we use prospective data from a British birth cohort study to test the hypothesis that optimal prenatal and childhood growth and development have long-term beneficial consequences for adult physical performance. We assessed performance in terms of length of time a one-legged balance with eyes closed could be maintained and speed of chair rising. Balancing on one leg requires muscle strength, mental concentration, and subtle motor control. The central nervous system must integrate input from the visual and vestibular systems, muscle spindles, and proprioceptive information from many sources. Chair rising requires muscle power and speed as well as strength and balance control.
We first investigated the role of birth weight and lifetime weight and height trajectories in order to identify any phases of growth or of adult weight or height change that were associated with midlife standing balance and chair rising. If the early development of muscle fibers has critical effects on adult performance, we hypothesized that childhood physical growth would be positively related to standing balance and chair rising, independently of later life experiences. We then investigated whether age at reaching motor milestones and puberty, early childhood cognitive ability, and adolescent motor coordination were associated with these performance measures. These developmental characteristics and later midlife performance are, to varying degrees, regulated by the central nervous system. If initial differences in the structure or function of the central nervous system have critical effects on the execution of physical tasks in midlife, we hypothesized that advanced early development would be positively associated with midlife standing balance and chair rising, independently of later life experiences. We controlled for childhood growth to test whether any associations were mediated through the development of muscle fibers. We controlled for later changes in adult body size, lifetime socioeconomic conditions, and current physical activity levels and health status to test whether good childhood growth and development are critical for good adult physical performance (and hence independent of these factors) or whether they represent the beginning of a pathway through life that promotes better performance through various associated protective life experiences.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The first maternity survey in Great Britain took place in 1946, based on 13,687 of the 16,695 registered births that took place between March 3 and 9 of that year. Medical officers of health in 453 of 458 local authorities in England, Wales, and Scotland at that time agreed to send health visitors to interview the mother 8 weeks after the birth. The Medical Research Council National Survey of Health and Development is a socially stratified sample of these births, consisting of 2,815 males and 2,547 females followed up 22 times, most recently when they were 53 years of age, when 3,035 were successfully contacted. This represents 57 percent of the original cohort sample and 83 percent of the target sample (15
). The rest were either living abroad (11 percent of the original cohort), had withdrawn from the study at earlier follow-ups (12 percent), or had died (9 percent). The cohort remains nationally representative in most respects (15).
At the home visit at age 53 years, trained nurses conducted tests of physical performance designed to elicit maximal performance, using standardized protocols described elsewhere (8). A test of standing balance was taken by 1,374 men and 1,410 women and measured as the longest time, up to a maximum of 30 seconds, that a one-legged stance with eyes closed could be maintained. The distribution was positively skewed and was normalized by use of a natural logarithm transformation. A test of chair rising was taken by 1,357 men and 1,400 women and measured as the time taken to rise from a sitting to a standing position and then to sit down again 10 complete times. In order for high scores to indicate good performance in line with the other tests, the reciprocal of the time taken (multiplied by 100) was used. Many of those who did not take or complete these tests were unable to do so because of chronic disease or disability (113 of 251 for standing balance and 176 of 278 for chair rising).
Birth weight (kg), height (cm), and weight (kg) in childhood (at ages 2, 4, 7, and 15), height (cm) at age 53 years, and weight (kg) at 26, 36, 43, and 53 years were measured by use of standardized protocols (16), except at age 26 years when they were self reported. Yearly rates of change were derived between all the consecutive weights ("weight velocities"), starting with birth weight, and between consecutive heights up to age 15 years ("height velocities"), starting with height at 2 years. The exact differences in age at measurement (in months) were used in the denominator. All heights, weights, and weight and height change variables were standardized to have a mean of 0 and a standard deviation of 1.
When the child was 2 years of age, mothers recalled the age (in months) of the child's first standing and walking unaided. The modal age for both was 12 months. At medical examinations at age 15 years, with the study member seated, the school physician recorded the maximum number of times in 15 seconds that they could tap the dorsum of their right hand with their left finger (and vice versa) and tap the ground with their right and then left foot. These were grouped in multiples of 10 for analysis. Early cognitive ability was measured as a standardized score, based on the average scores obtained on tests of reading comprehension (sentence completion), pronunciation, vocabulary, and nonverbal reasoning (picture intelligence) taken at age 8 years (17).
The age at menarche was classified as occurring at 11 years or earlier, 12 years, 13 years, and 14 years or later, on the basis of questions asked by school physicians at a medical examination at age 15 years. The physicians determined the stage of pubertal development of boys, classified as "mature" (advanced development of genitalia, profuse pubic and axillary hair, and voice broken); "advanced" (advanced development of genitalia but not fully mature according to at least one other marker); "early" (early development of genitalia and some pubic or axillary hair or voice starting to break); or "infantile" (infantile or early adolescent genitalia but no pubic or axillary hair and voice not broken).
We chose measures of socioeconomic conditions, physical activity, and health status as potential confounders or mediators, as these are important risk factors for physical performance (8). Socioeconomic conditions were defined by the British Registrar General's classification of social class, grouped to distinguish between the manual and nonmanual social classes and based on father's occupation in childhood and own occupation in adult life. Physical activity distinguished between those who reported taking part in sports, exercise, or other physical activities in their leisure time in the previous 4 weeks and those who did not. Three measures of current health status, detailed elsewhere (8), were used that were based on standardized questions and a checklist of common health problems: disabling or life-threatening conditions (diabetes, cancer, epilepsy, or cardiovascular disease); severe respiratory symptoms; and musculoskeletal symptoms (pain and stiffness in knees or hands for at least a month or severe backache).
For analysis, we used the measures of physical performance as continuous outcomes to study the full range of function and to provide extra statistical power, because there are a linear increase in disability across these scores and no agreed-upon clinical thresholds (8). First, chair rising and standing balance were examined by use of separate multiple regression models in relation to standardized heights and weights at various ages. Childhood weights were adjusted for height at the same age, and adult weights were adjusted for height at age 53 years. To identify how best to characterize the lifetime weight and height trajectories in relation to adult physical performance, we ran a multivariable model, including birth weight, height at age 2 years, and all the weight and height velocities up to age 53 years, using those with complete data. We then adjusted for lifetime social class and current physical activity and health status. These models were conducted separately for men and women, as previous work (8) had shown that the effect of weight on these performance measures was stronger for women than men. Sex interaction terms were added to models including both sexes to test whether the effects of growth or development on standing balance and chair rising were different for men and women. We also added interaction terms to test whether the association of each weight and height velocity with performance varied by childhood or adult social class.
Second, we examined chair rising and standing balance in relation to motor milestones and coordination, cognitive ability, and age at puberty and, in the subsamples with complete data, tested in separate models if any relations were explained by the growth trajectories, childhood social class, or adult risk factors. These models are not presented separately for men and women, as there was no evidence that the effect of these characteristics on physical performance differed by sex. All relations with continuous risk factors were tested for deviation from linearity by adding a quadratic term to the regression models (18).
As we carry out multiple statistical tests, thus increasing the possibility of false positive results, we report the null findings as well as significant ones and present estimates and confidence intervals rather than just the p values (19).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The characteristics of the total sample and the sample with complete growth data are provided in table 1. There were very few differences between those with and without complete data. Men without complete data were more likely to have lower childhood cognitive scores, to come from the adult manual social class, and to be physically inactive compared with men with complete data.
|
Standing balance and chair rising at age 53 years in relation to heights and weights at various ages and covariates
In men, standing balance at age 53 years was positively related to weight at ages up to 15 years, and the relation was significant at age 4 years, where a 1-standard deviation increase in weight was associated with a mean difference in balance time of 0.07 (natural logarithm (ln))seconds (95 percent confidence interval (CI): 0.02, 0.13) (table 2). In women, there was a positive effect of birth weight on standing balance, but otherwise no associations between childhood weights and standing balance were observed. At adult ages, the effect of weight on standing balance was consistently and significantly negative for men and women. Height from age 4 years through to adulthood was positively related to standing balance in women. The detrimental effect of weight at ages 43 and 53 years on standing balance and the beneficial effect of height at age 15 years were significantly stronger in women than in men (all tests for interaction: p < 0.05).
|
Similarly, chair rising in men was positively related to childhood weight, particularly at age 7 years (mean difference = 0.21 (1/seconds x 100), 95 percent CI: 0.05, 0.36 per 1 standard deviation (SD)), although strongly and negatively related to adult weight (e.g., at age 26 years, mean difference = –0.16 (1/seconds x 100), 95 percent CI: –0.28, –0.05 per 1 SD) (table 2). In women, there were negative effects of weight at all ages, which were particularly strong in adult life. The effect of weight at age 2 years was significantly different in women and men and, from age 36 years, the detrimental effect of weight was stronger for women than for men (all tests for interaction: p < 0.05). Height from 7 years through to adult life was negatively related to chair rising.
Those from the adult manual social class had worse performance scores (table 2). Men and women from a manual class background in childhood had poorer standing balance than did those from a nonmanual class background (p < 0.001); the effect on chair rising was less evident (table 2). Men and women who were inactive, had disabling health conditions, or had musculoskeletal or respiratory symptoms had worse performance scores than did their active and healthy peers (not shown).
Standing balance and chair rising in relation to weight and height velocities
For men, there were positive effects of childhood weight gain and negative effects of adult weight gain on standing balance at age 53 years, after adjustment for birth weight and the height trajectory (table 3, model 1; figure 1). For a given birth weight and height trajectory, weight gain between 2 and 4 years was particularly beneficial with a 1-standard deviation increase in velocity being associated with a mean difference in balance time of 0.14 (ln)seconds (95 percent CI: 0.04, 0.21). In contrast, weight gain between 15 and 26 years was particularly detrimental (mean difference = –0.18 (ln)seconds, 95 percent CI: –0.26, –0.11 per 1 SD). A similar picture for men was seen for chair rising (table 4, model 1; figure 2). For women, there was little effect of weight gain on standing balance or chair rising before age 7 years; only the difference in the effect of weight gain between birth and 2 years on chair rising was significantly different for men and women (test for interaction: p < 0.05) (tables 3 and 4, model 1). After age 7 years, weight gain was significantly and negatively related to standing balance and chair rising, and these effects were stronger in women than in men for weight gained between ages 7 and 15 years (with respect to chair rising) and after age 26 years (both tests for interaction: p < 0.05) (tables 3 and 4, model 1).
|
|
|
|
For women, height gain before age 4 years was beneficial for standing balance and significantly different from the negative relation observed for the men (test for interaction: p = 0.01) (table 3, model 1). In contrast, height at age 2 and height gain were negatively associated with chair rising, significantly so for men up to age 7 years, taking into account the rest of the growth trajectory (table 4, model 1). In women, this effect was weaker but not significantly so (table 4, model 1).
Up to age 7 years, those from the nonmanual social class had greater height and weight velocities than did those from the manual class; after age 7 years, those from the manual social class had greater weight velocities, and there were no differences in the height velocities (not shown). However, there was no evidence that the effects of the components of the growth trajectory on midlife physical performance varied systematically by childhood or adult social class. Adjustment for childhood and adult social class and for current physical activity and markers of health status slightly weakened the estimates for the growth trajectory but did not change the pattern of relations observed (tables 3 and 4, model 2).
Standing balance and chair rising in relation to other developmental factors
The age at first walking was associated with standing balance and chair rising at age 53 years, and the age at first standing was associated with chair rising. These relations were best described by a quadratic function (table 5). Chair rising was better in those who started standing and walking around the modal age (12 months) than in those who reached these developmental milestones earlier or later (figure 3). Standing balance was best in those who started walking at around 16 months (figure 4). Higher scores on the cognitive tests at age 8 years and on the motor coordination tests at age 15 years were associated with better standing balance and chair rising (table 5). The timing of puberty was not associated with either test (table 5).
|
|
|
The effects of age at first walking, cognitive score, and finger tapping on midlife performance were hardly changed after adjustment for the growth trajectory first up to age 7 years and for the whole trajectory (tables 6 and 7). Adjustment for childhood and adult social class generally had the most effects on the estimates. Current physical activity and health status did not confound these relations between developmental factors and midlife performance, except that the effect of physical activity reduced quite markedly the effect of cognitive ability on chair rising (table 7).
|
|
It was not possible to test whether motor competence was mediated by cognitive ability because, within the group with complete data on these factors, the size of the effects on midlife performance, adjusted only for gender, was much reduced.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In this cohort, we have demonstrated that childhood weight gain before age 7 years, in contrast to adult weight gain, was generally beneficial for midlife standing balance and chair rising in men. These findings were not explained by later weight gain, lifetime social class, and current physical activity levels and health status, nor were they modified by social class. Furthermore, we have shown that reaching motor milestones at the modal age, higher cognitive function, and improved motor coordination in adolescence have long-term benefits for these measures of functional performance in men and women. The underlying mechanisms need consideration.
There is little evidence of childhood obesity in this cohort (20), and childhood weight gain may be a good marker for gain in lean as well as fat mass, particularly for boys. Boys carry proportionately less body fat, even before puberty (21–23), and while the percentage of body fat increases with age in girls, it remains relatively steady in boys (21, 22). Our findings probably reflect muscle growth and tracking of muscle size, as well as possibly an increase in the numbers of muscle fibers (24). They are consistent with other findings in this cohort that show that childhood growth is associated with adult muscle strength (11). The effects of early height gain on physical performance were less consistent, being positive in women and negative in men, perhaps because childhood height and height gain are better markers of the accrual of lean mass in women than are weight and weight gain, whereas the opposite appears true for men.
We have shown previously that heavier men and women were worse at standing balance and chair rising compared with lighter men and women, and that the effect was stronger in women than men (8). Here we show that weight gain during puberty and throughout adult life was detrimental to women's performance. Weight gain in adult life represents gain in fat rather than lean mass, and women carry proportionately more fat than men do. For men, weight gain in early adult life (>15–26 years) was the most detrimental for standing balance and chair rising, perhaps because it reflected the adverse effects of long-term overweight or obesity.
The relations between age at first standing and walking and midlife performance were independent of childhood growth and were not linear. This suggests that the relations are unlikely to be due to the tracking of muscle size throughout life. Infant motor development reflects variation in structural and functional maturation of brain motor systems (25); if these differences were somehow sustained, we would expect infant motor development to be associated with structural or functional measures of adult brain motor systems. A recent study in a Finnish birth cohort to age 35 years has shown, for the first time, a long-term normative continuity between the timing of motor development and the anatomic integrity of adult motor systems (25). To our knowledge, motor milestones and motor coordination in childhood have not been previously linked to physical tasks that assess adult motor function. Our finding that midlife performance benefits from motor development that is "on time" with no benefit of advanced development needs investigation in other longitudinal studies.
The development of motor and cognitive function in early life and the aging of motor and cognitive function in later life are highly integrated (26, 27). Thus, the relations of motor milestones, motor coordination, and early cognitive ability with midlife physical performance may be because these developmental characteristics are markers of more complex cortical-subcortical neural circuits involved in higher cognitive function in adult life. This is supported by findings from the Finnish cohort study that demonstrated an anatomic overlap between brain systems associated with infant motor development and adult executive function (25), as well as an association between infant motor and adult executive performance (28). This pathway is currently being investigated in the National Survey of Health and Development cohort. Our measure of early cognitive ability may be similarly linked to adult physical performance. These possible pathways may explain why adjustment for lifetime social class, which is strongly associated with lifetime cognitive function (29), reduced the effects of early cognitive function and other developmental factors on midlife physical performance. Alternatively, other unmeasured factors may be implicated.
The study had a number of limitations. It excluded 9 percent of participants with disability or chronic disease who could not attempt or complete the tests. This was done in order to retain the performance tests as continuous measures and to investigate the full range of function on these tests. Further analysis (not shown) indicated that the growth trajectories and developmental characteristics of those excluded from these tests because of illness and disability were similar to those with the lowest performance scores. The growth trajectories were inevitably limited by the ages when height and weight were measured and, as rates of maturation vary considerably between individuals, parameters of growth such as peak growth velocity cannot be derived. However, we were able to characterize prepubertal, pubertal, and postpubertal growth, showing the differential effects of these parameters on later performance. Although dealing with multiple measures of body size is not straightforward because of the relatively high correlations between measures, the growth-velocities approach used here has a valid interpretation (30, 31). The analyses using the growth trajectory were also restricted to those with complete data and thus reduced the statistical power of the study. However, there were few differences between those with and without complete data and no reason to expect that relations between growth and midlife performance should vary between the two samples. The effects of childhood growth and development on adult performance were not strongly attenuated by adjustment for socioeconomic conditions, physical activity, and health status. However, residual confounding remains a possible explanation for the observed associations, because either these covariates were measured imprecisely or there were other unmeasured confounders. We found no evidence that the effects of childhood growth were modified by social class, although the power to detect interactions was limited by sample size.
It will be of interest to see whether the effects of early factors have a similar impact on physical performance or its decline when the cohort members reach their sixties. So far, our findings suggest that interventions in early life that help to build muscle mass and motor competence, such as increased physical exercise, may have long-term beneficial effects on adult physical performance and, hence, the risk of disability and frailty. There is already evidence of a critical window of opportunity before puberty when the skeleton is most sensitive to the anabolic effects of bone loading (32), and this may also apply to muscle development. Evidence from this cohort and others of environmental effects on growth (33) and development (29) indicates that these characteristics are modifiable. Our findings also confirm the importance of adult characteristics, and interventions to control weight gain in adult life should remain a key preventive strategy.
As the study is a nationally representative sample of men and women born in Britain in 1946, our findings should be generalizable to the national population. Whether our findings are generalizable to younger cohorts or in US cohorts with higher levels of obesity in childhood (34, 35), as well as in adult life, remains to be seen. Of relevance, the 50th percentiles of the heights and weights at ages 2, 4, and 7 years in this cohort are very similar to the recent revised growth charts produced by the Centers for Disease Control and Prevention (36). Where they differ is that the 50th percentile of the weights and heights at age 15 years in this cohort is closer to the 25th percentile on the revised charts. The similarity in prepubertal growth gives some support for the generalizability of our findings on prepubertal growth in relation to adult performance, although it is not known whether the proportion of fat to lean mass in children at certain ages differs across time and place.
In summary, this study is the first to show that childhood growth and development impact on midlife performance; prevention of disability and frailty in later life may need to start early.
|
ACKNOWLEDGMENTS |
|---|
The authors thank Dr. Joan Bassey and Dr. Graham Murray for their help and advice.
Conflict of interest: none declared.
|
NOTES |
|---|
Editor's note: An invited commentary on this article appears on page 122, and the authors' response appears on page 126.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
- Aihie Sayer A, Cooper C. A life course approach to biological ageing. In: Kuh D, Ben-Shlomo Y, eds. A life course approach to chronic disease epidemiology. 2nd ed. Oxford, United Kingdom: Oxford University Press, 2004.
- Stel VS, Pluijm SM, Deeg DJ, et al. Functional limitations and poor physical performance as independent risk factors for self-reported fractures in older persons. Osteoporos Int 2004;15:742–50.[Medline]
- Guralnik JM, Ferucci L, Pieper CF, et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci 2000;55:M221–31.
[Abstract/Free Full Text] - Guralnik JM, Ferucci L, Simonsick EM, et al. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med 1995;332:556–61.
[Abstract/Free Full Text] - Onder G, Penninx BW, Ferucci L, et al. Measures of physical performance and risk for progressive and catastrophic disability: results from the Women's Health and Aging Study. J Gerontol A Biol Sci Med Sci 2005;60:74–9.
[Abstract/Free Full Text] - Wolinsky FD, Miller DK, Andresen EM, et al. Further evidence for the importance of subclinical functional limitation and subclinical disability assessment in gerontology and geriatrics. J Gerontol B Psychol Sci Soc Sci 2005;60:S146–51.
[Abstract/Free Full Text] - Melzer D, Lan TY, Guralnik JM. The predictive validity for mortality of the index of mobility-related limitation—results from the EPESE study. Age Ageing 2003;32:619–25.
[Abstract/Free Full Text] - Kuh D, Bassey EJ, Butterworth S, et al. Grip strength, postural control, and functional leg power in a representative cohort of British men and women; associations with physical activity, health status, and socioeconomic conditions. J Gerontol A Biol Sci Med Sci 2005;60:224–31.
[Abstract/Free Full Text] - Aihie Sayer A, Cooper C, Evans JR, et al. Are rates of ageing determined in utero? Age Ageing 1998;27:579–83.
[Abstract/Free Full Text] - Aihie Sayer A, Syddall HE, Gilbody HJ, et al. Does sarcopenia originate in early life? Findings from the Hertfordshire cohort study. J Gerontol A Biol Sci Med Sci 2004;59:930–4.[Web of Science]
- Kuh D, Bassey EJ, Hardy R, et al. Birthweight, childhood size and muscle strength in adult life: evidence from a birth cohort study. Am J Epidemiol 2002;156:627–33.
[Abstract/Free Full Text] - Barker DJP. Mothers, babies and health in later life. Edinburgh, United Kingdom: Churchill Livingstone, 1998.
- Kuh D, Ben-Shlomo Y. A life course approach to chronic disease epidemiology: tracing the origins of ill-health from early to adult life. Oxford, United Kingdom: Oxford University Press, 2004.
- Gluckman P, Hanson M. The fetal matrix: evolution, development, and disease. Cambridge, United Kingdom: Cambridge University Press, 2004.
- Wadsworth MEJ, Butterworth S, Hardy RJ, et al. The life course prospective design: an example of benefits and costs associated with study longevity. Soc Sci Med 2003;57:2193–205.[CrossRef][Web of Science][Medline]
- Braddon FEM, Rodgers B, Wadsworth MEJ, et al. Onset of obesity in a 36 year birth cohort study. Br Med J 1986;293:299–303.
[Abstract/Free Full Text] - Richards M, Shipley B, Fuhrer R, et al. Cognitive ability in childhood and cognitive decline in mid-life: longitudinal birth cohort study. BMJ 2004;328:552–4.
[Abstract/Free Full Text] - Clayton D, Hills M. Statistical methods in epidemiology. Oxford, United Kingdom: Oxford University Press, 1993.
- Rothman KJ. Modern epidemiology. Boston, MA: Little, Brown and Co, 1986.
- Stark O, Atkins E, Wolff OH, et al. Longitudinal study of obesity in the National Survey of Health and Development. Br Med J 1981;283:13–17.
[Abstract/Free Full Text] - van der Sluis IM, de Ridder MA, Boot AM, et al. Reference data for bone density and body composition measured with dual energy x ray absorptiometry in white children and young adults. Arch Dis Child 2002;87:341–7.
[Abstract/Free Full Text] - Taylor RW, Jones IE, Williams SM, et al. Body fat percentages measured by dual-energy x-ray absorptiometry corresponding to recently recommended body mass index cutoffs for overweight and obesity in children and adolescents aged 3–18 y. Am J Clin Nutr 2002;76:1416–21.
[Abstract/Free Full Text] - Arfai K, Pitukcheewanont PD, Goran MI, et al. Bone, muscle, and fat: sex-related differences in prepubertal children. Radiology 2002;224:338–44.
[Abstract/Free Full Text] - Bailey P, Holowacz T, Lassar AB. The origin of skeletal muscle stem cells in the embryo and the adult. Curr Opin Cell Biol 2001;13:679–89.[CrossRef][Medline]
- Ridler K, Veijola JM, Tanskanen P, et al. Motor development in early childhood normally predicts adult frontal cortical structure but not in people with schizophrenia. Presented at the 11th Annual Meeting of the Organization for Human Brain Mapping, Toronto, Ontario, Canada, June 12–16, 2005.
- Thelen E, Smith L. A dynamic systems approach to the development of cognition and action. Cambridge, MA: Branford Books/MIT Press, 1994.
- Kluger A, Gianutsos JG, Golomb J, et al. Patterns of motor impairment in normal aging, mild cognitive decline, and early Alzheimer's disease. J Gerontol B Psychol Sci Soc Sci 1997;52:P28–39.[Abstract]
- Murray GK, Veijola J, Moilanen K, et al. Infant motor development is associated with adult cogntive categorisation in a longitudinal birth cohort study. J Child Psychol Psychiatry 2006;47:25–9.[CrossRef][Web of Science][Medline]
- Richards M, Sacker A. Lifetime antecedents of cognitive reserve. J Clin Exp Neuropsychol 2003;25:614–24.[Web of Science][Medline]
- De Stavola BL, dos Santos Silva I, McCormack V, et al. Childhood growth and breast cancer. Am J Epidemiol 2004;159:671–82.
[Abstract/Free Full Text] - De Stavola BL, Nitsch D, dos Santos Silva I, et al. Statistical issues in life course epidemiology. Am J Epidemiol 2006;163:84–96.
[Abstract/Free Full Text] - Bassey J, Aihie Sayer A, Cooper C. A life course approach to musculoskeletal ageing: muscle strength, osteoporosis, and osteoarthritis. In: Kuh D, Hardy R. A life course approach to women's health. Oxford, United Kingdom: Oxford University Press, 2002.
- Li L, Manor O, Power C. Early environment and child-to-adult growth trajectories in the 1958 British birth cohort. Am J Clin Nutr 2004;80:185–92.
[Abstract/Free Full Text] - Chinn S, Rona RJ. Prevalence and trends in overweight and obesity in three cross sectional studies of British children, 1974–94. BMJ 2001;322:24–6.
[Abstract/Free Full Text] - Gillman MW. A life course approach to obesity. In: Kuh D, Ben-Shlomo Y, eds. A life course approach to chronic disease epidemiology. 2nd ed. Oxford, United Kingdom: Oxford University Press, 2004.
- National Center for Health Statistics. 2000 CDC growth charts: United States. Hyattsville, MD: Centers for Disease Control and Prevention, Department of Health and Human Services, 2006. (http://www.cdc.gov/growthcharts).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. M. Blyth, G. T. Jones, and G. J. Macfarlane Musculoskeletal health--how early does it start? Rheumatology, October 1, 2009; 48(10): 1181 - 1182. [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. Pitcher, A. L. Robertson, R. A. Cockington, and V. M. Moore Prenatal Growth and Early Postnatal Influences on Adult Motor Cortical Excitability Pediatrics, July 1, 2009; 124(1): e128 - e136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kuh, R. Cooper, R. Hardy, J. Guralnik, M. Richards, and the Musculoskeletal Study Team Lifetime Cognitive Performance is Associated With Midlife Physical Performance in a Prospective National Birth Cohort Study Psychosom Med, January 1, 2009; 71(1): 38 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bergman, L. Ferrucci, J. Guralnik, D. B. Hogan, S. Hummel, S. Karunananthan, and C. Wolfson Frailty: An Emerging Research and Clinical Paradigm--Issues and Controversies J Gerontol A Biol Sci Med Sci, July 1, 2007; 62(7): 731 - 737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Wang Invited Commentary: Beyond Frequencies and Coefficients--Toward Meaningful Descriptions for Life Course Epidemiology Am. J. Epidemiol., July 15, 2006; 164(2): 122 - 125. [Full Text] [PDF]
|
|
|
|
|
|
R. Hardy, Y. Ben-Shlomo, and D. Kuh Hardy et al. Respond to "Beyond Frequencies and Coefficients" Am. J. Epidemiol., July 15, 2006; 164(2): 126 - 127. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||