SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

We examined whether behavioral problems in childhood and adolescence are associated with young adults' BMI and obesity, and tested whether childhood behavioral problems have a greater impact on young adults' obesity than adolescent behavioral problems. The data were from the Mater-University of Queensland Study of Pregnancy (MUSP) and Its Outcomes, a population-based birth cohort study commenced in Brisbane, Australia, in 1981. A subsample of 2,278 children for whom we had prospective information on their behavioral problems at ages 5 and 14 and measured BMI, and its categories (normal, overweight, and obese) at age 21 was chosen. Young adults who experienced behavioral problems at ages 5 or 14 had a greater average BMI and were more likely to be obese compared to young adults without behavioral problems at both ages. The childhood onset group was at greater risk of becoming obese by age 21 compared to the adolescent onset group (P = 0.04). These associations remained consistent after adjusting for a variety of potential covariates including maternal characteristics (i.e., demographics and life style), child dietary patterns, family meals, television (TV) watching, and participation in sports and exercise at 14 years. Childhood as well as persistent behavioral problems during childhood and adolescence predicts young adults' BMI and obesity. Although further studies are needed to confirm this association, there is a need for close monitoring of children presenting with behavioral problems.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

The population prevalence of childhood obesity has increased by two- to threefold during the last few decades (1). During the same period there is some evidence that the prevalence of behavioral problems in children has also increased (2), which may be partly due to the development of better recognition and diagnosis tools rather than a true epidemiological change. One area, which has been receiving increasing attention, is the association between behavioral problems and obesity. This relationship could potentially be bi-directional (3), or even synergistic. If mental disorders at younger ages were shown to have a causal relationship with adult obesity, this could provide an important opportunity for intervention.

Prospective studies examining childhood mental health and later onset weight gain have generally found an association with various psychopathologies, including an association with behavior disorders among 8–11 year olds (4), with major depression among 6–17 years (5), with depressive symptoms prior to 17 years of age (6), and in children aged 6–12 years, an association between total body fat and binge eating and dieting but not depression (7).

Investigations of adolescent depression and adult obesity in longitudinal studies have often found a significant association between depression in older adolescent females, including multiple ethnic groups, and obesity in adulthood (8,9,10). Another prospective study found that depressed adolescent males and females were at greater risk of developing obesity or remaining obese (11). Other prospective studies have broadened the range of behavioral problems in adolescents associated with increasing weight, including conduct disorder (12) and anxiety disorder in adolescent females (8). These prospective studies have experienced limitations, such as short follow-up times (4,7,11), at least some use of self-reported BMI (5,8,9,11,12), clinical sample (5), and retrospective recall for childhood symptoms (6).

Proposed mechanisms by which childhood behavioral problems may influence weight include a neurobiological link (13), physiological changes (14), and unhealthy environment (8). The neurobiological connection theorized that low serotonin levels in depressed individuals cause them to eat foods high in carbohydrates (13). Depression may cause physiological changes in their hormone and immune systems (14), which may influence appetite control. Depressed children have more difficulty taking good care of themselves because of symptoms and consequences of depression, such as difficulty adhering to fitness regiments, overeating, and having negative thoughts. Parents of depressed children may be more likely to use unhealthy foods or television (TV) as rewards, or to succumb to children's demands for these things (15). Alternatively, it appears that children who experience difficulties with peer socialization are more likely to be sedentary in adulthood (16).

There appears to be a correlation between depressive symptoms in childhood and also in adolescence (17,18,19), and examining childhood as well as adolescent data enables inspection for potential intervention periods for later obesity.

The aims of this study are to examine whether behavioral problems in childhood and adolescence are associated with young adults' BMI and obesity, and to test whether childhood behavioral problems have a greater impact on young adults' obesity than adolescent behavioral problems using a prospective follow-up birth cohort study in Australia.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Participants

The data used in this study are from the Mater-University of Queensland Study of Pregnancy (MUSP). The MUSP is a prospective study of 7,223 women, and their offspring, who received antenatal care at a major public hospital in Brisbane between 1981 and 1983 and delivered a live singleton child who was not adopted before leaving hospital (20). Both mothers and children were followed prospectively from the prenatal period until 21 years postdelivery, with intervening follow-up occurring at key points during development: 3–5 days, 6 months, 5 and 14 years postdelivery. At the 6-month follow-up, 6,720 women (93%) responded to the questionnaire. At the 5-year follow-up 5,259 mothers (73%) provided data for their child, 5,234 mothers (72%) provided data on their own health and 4,010 (56%) children attended the physical assessments. A similar pattern is evident at the 14-year follow-up with 5,185 mothers (72%) and 5,172 children (72%) responding, but child assessments were only completed for 3,799 children (53%). At 21 years 4,024 (56%) mothers and 3,778 (53%) children attended the follow-up assessment and provided usable data. Only 2,664 young adults' undertook the physical assessment. Due to limited funding fewer participants had physical assessments than were interviewed at all phases of the study. Full details of the study participants and measurements have been previously reported (20,21). Written informed consent from the mothers was obtained at all data collection phases and from the young adults at the 21-year follow-up of the study. Ethics committees at the Mater Hospital and the University of Queensland approved each phase of the study.

In this study the main analyses are restricted to 2,278 children for whom we had prospective information on their behavioral problems at ages 5 and 14 and measured BMI at age 21 years. Participants who were lost to follow-up (did not attend the 14-year follow-up and 21-year follow-up but attended at delivery) were more likely to be males and non-white (all P values <0.001). Their mothers were more likely to be teenagers at their birth, to be less educated, single or cohabitating, have three or more children, use tobacco and alcohol during pregnancy, and to be anxious and depressed at their first antenatal visit (all P values <0.001).

Outcome measurements

The main outcome in all analyses is the young adults' BMI and categories of BMI (normal, overweight, and obese) at the 21-year follow-up examination. Young adults' height was measured using a portable stadiometer to the nearest millimeter. Their weight was measured as the average of two measures of weight, lightly clothed with a scale accurate to 0.2 kg. BMI (weight in kilograms by the square of height in meters i.e., kg/m2) was categorized into normal (<25kg/m2), overweight (25–29 kg/m2), and obese (≥30 kg/m2) using the World Health Organization classification of BMI cutoffs (22).

Measurements of exposure

The main exposure of this study was child emotional and behavioral problems, which were assessed from maternal reports of child behavior using modified versions of Achenbach's Child Behavior Checklist (CBCL) at age 5 and the full CBCL at age 14 (refs. 23,24).

The CBCL is a widely used, standardized, empirically based parental report instrument designed to assess behavioral problems and competencies of children aged 4–16 years (23). This instrument has been shown to have construct validity and is standardized for age and sex. In a 3-year national study of 4- to 16-year-old children those with at least one “sign of disturbance” on the CBCL had increased risk for later social, academic, emotional, and behavioral problems (25). Though we recognize that our definition of emotional/behavioral problems using the CBCL does not constitute a clinical diagnosis, we describe those children with scores above the 90th centile as having behavioral problems at ages 5 and 14 years. We have chosen a 10% cutoff to ensure adequate numbers for analysis, rather than the t-score that is used for clinical purposes. Based on this cutoff, we classified the exposure for the main analysis into four mutually exclusive groups: (i) normal (<90th centile on CBCL) at age 5 and 14 years (without behavioral problems); (ii) had problems (≥90th centile on CBCL) at age 5 but normal at 14 (childhood behavioral problems); (iii) normal at age 5 but had problems at 14 years (adolescent behavioral problems); and (iv) persistent problems at 5 and 14 years (persistent behavioral problems).

Measurements of confounders and mediators

The following maternal and child characteristics were considered to be potential confounding or mediating factors on the basis of a priori knowledge (26) and their potential association with childhood behavioral problems and BMI or categories of BMI at 21 years. Maternal age at birth, maternal education (did not complete secondary school, completed secondary school, completed further/higher education) at birth, maternal depression at 5-year follow-up (assessed with the Delusions-Symptoms-States Inventory (27), mothers were classified as depressed if they reported three or more of seven symptoms related to depression), maternal smoking status at 5-year follow-up (nonsmoker, 1–9 cigarettes per day and ≥10 cigarettes per day), and pre-pregnancy maternal BMI were considered as confounders. Child demographic characteristics considered were sex and exact age at the 21-year follow-up. Child diet, family attitude to eating together, sports, and TV watching at 14-year follow-up were considered as mediators. Childhood diet, TV watching, and physical activity data were obtained from maternal report at the 14-year follow-up. Mothers were asked to report the frequency of their child's consumption of fast food, salad, soft drinks, and red meat (all with response options of rarely or never, at least two or three times a week, most days), family attitude to having meals together (at least once a day, few times/once/less than once a week), and to report the amount of time her child spent watching TV (<1 h per day, 1 to <3 h per day, 3 to <5 h per day and ≥5 h per day) and the amount of time they spent on sports or exercise (4–7 days per week; 0–3 days per week). Additionally, family communication (28) at 14-year follow-up was considered as a mediator.

Statistical analyses

The mean difference of BMI by mental health scores was compared by one-way analysis of variance and computation of an F-test (Table 1). Multinomial logistic regression was used to estimate the unadjusted odds ratio of being overweight and obese at age 21 years. Statistical evidence for a difference in effect between males and females was assessed by computing a likelihood ratio test of the interaction with sex. As there was no statistical evidence that the effect differed between the sexes, results are presented for males and females combined.

Table 1.  Mean BMI, prevalence (%) of overweight and obesity, and the unadjusted OR of being overweight and obese at age 21 by child psychopathology
inline image

A series of multiple linear regression models (see footnotes to Tables 2 and 3) were used to determine the mean difference in young adults' BMI by child behavioral problems, taking into account potential confounding and mediating factors (Table 2). Similarly, a series of multinomial logistic regression models were used to estimate adjusted odds ratios of being overweight and obese in young adulthood by child behavioral problems from ages 5 to 14 years adjusting for the potential factors.

Table 2.  Difference in mean BMI at age 21 years by behavioral problems at age 5 and 14 years adjusting for potential factors (N = 2004)
inline image
Table 3.  Adjusted odds ratios of young adults being obese at age 21 by change in behavioral problems between ages 5 and 14 years adjusting for potential factors (N = 2004)
inline image

In the sensitivity analysis, we excluded overweight or obese children at age 5 years from the present analysis (overweight or obese status was defined using Cole et al. (29) standard definition), to see whether any association may be influenced by reverse causality (individuals who are overweight or obese by age 5 years having behavioral problems).

The BMI distribution in this sample at age 21 years was slightly positively skewed as expected. A sensitivity analysis was conducted using a Quantile regression (30) model, which is used to estimate the various quantiles (e.g., median) of a population and is thus more robust in response to large outliers. In the sensitivity analysis, we found that the estimated results are not substantially different from the results presented. All analyses were undertaken using Stata version 9.2 (Stata, College Station, TX).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Of 2,278 young adults with prospective information on behavioral problems at ages 5 and 14 years and measured BMI at age 21 years, 20.85% were overweight and 12.77% obese at age 21 years. Young adults who experienced behavioral problems at ages 5 or 14 had on average a greater BMI and were more likely to be obese compared to young adults who did not have behavioral problems at both ages (Table 1).

Table 2 shows the mean difference in BMI at age 21 comparing children who experienced different patterns of behavioral problems to those who did not experience behavioral problems with adjustment for potential confounding factors in a series of multiple regression models. The results are presented for the 2,004 young adults with complete data available on all variables included in any of the multivariable models (Table 2). In the age and sex adjusted model, young adults who experienced persistent behavioral problems at age 5 and 14 years had on average 3.44 kg/m2 (95% confidence interval 1.81, 5.07) greater BMI compared to the young adults without behavioral problems at both ages. After adjusting for maternal and child characteristics including child diet, sports, and TV watching, the association remained consistent (models 2 and 3). However, adjustment for pre-pregnancy maternal BMI attenuated the association by nearly 16%, although it remained significant.

Young adults who had childhood but not adolescent behavioral problems were 2.08 (95% confidence interval: 1.13, 3.83) times more likely to be obese by age 21 compared to children who were without behavioral problems (model 1 in Table 3). In the same model, we found that young adults who had persistent behavioral problems were 3.86 (95% confidence interval: 1.77, 8.42) times more likely to be obese by age 21 years compared to children without behavioral problems. For the childhood case, the association remained consistent after adjusting for potential confounders. However, for the persistent case there was a little attenuation of the odds ratio, though it remained significant, after adjusting for maternal pre-pregnancy BMI.

When we repeated the analyses to test whether young adults who had childhood but not adolescent behavioral problems were at greater risk of becoming obese compared to adolescent behavioral problems (i.e., changing reference category from without behavioral problems at ages 5 and 14 to adolescent behavioral problems), we found that the childhood group were at greater risk of becoming obese by age 21 compared to the adolescent group (P = 0.04). When the analyses were repeated after removing the 600 children who were overweight or obese at age 5 the point estimates did not differ from those presented here (results are not shown). Similarly, when we repeated the analyses with further adjustment for parent-adolescent communication at age 14, the results were not altered.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

Using the 21-year prospective follow-up of mother and child study, we found that young adults' BMI as well as the prevalence of obesity was greater in those who experienced persistent behavioral problems at childhood and adolescence compared to those who did not have behavioral problems at both ages. Young adults who experienced childhood but not adolescent behavioral problems were also at greater risk to become obese by age 21 compared to young adults who had no history of behavior problems. We also found that this childhood group was at greater risk to become obese by age 21 compared to the adolescent behavioral problems group. These associations remained consistent after adjustment for a variety of potential covariates including childhood dietary patterns, family meals, family communication, TV watching, and participation in sports and exercise. Findings of this study suggest that childhood behavioral problems as well as persistent behavioral problems are independent predictors of young adults' BMI and obesity. These findings also provide some evidence for a long-term impact of childhood behavioral problems on the later development of obesity.

Our study is in broad agreement with the results from other studies, as overall there does appear to be evidence of behavioral problems in childhood influencing later BMI. Our study has added to this knowledge by examining both children and adolescents and finding that childhood behavioral problems are the stronger predictor. This is analogous to Moffitt's suggestion that adolescence-limited antisocial behavior emerges alongside puberty and, providing the adolescent does not encounter a “snare,” should be relatively temporary, whereas childhood onset of antisocial behavior persists throughout life (31).

Risk factors for childhood behavioral problems include gene-environment interactions (32), parental distress (33), poor self-regulation (31), low family socioeconomic status (33,34), reductions in perinatal brain levels of docosahexaenoic acid (35), and a poor social and academic performance by the child (33). Further support for our results are obtained from prospective studies showing that these risk factors in childhood result in obesity or greater weight in early adulthood, including parental neglect (36), childhood adversity, and paternal maladaptive behavior (37).

For behavioral problems to cause obesity/overweight it could be expected to affect caloric intake or exercise/inactivity. Though not measured precisely, inclusion of these mediators in the model did not substantially affect the estimates of association. It is possible that differences in sleep patterns (38,39) or biochemical changes, resulting from epigenetic modification of the hypothalamic pituitary adrenal axis (40) or the autonomic nervous system (41) could affect both behavior and weight gain. There was no evidence to suggest that the association was secondary to confounding by social adversity.

Our results should be seen in the context of some limitations. The loss to follow-up in our cohort was considerable. However, our results would be biased if the associations we have assessed were nonexistent or in the opposite direction in nonparticipants, that is to say if, among nonresponders, those young adults who became overweight or obese were less likely to have greater childhood/adolescence behavioral problems, which is unlikely. Our estimates of overweight or obesity in young adulthood were compared to a similar age category in the Australian National Nutrition Survey 1995 (ref. 42), and results are comparable. National Nutrition Survey is the most recent Australian survey, which collected physical measurements on participants in a comparable age group. For participants aged 21 years in MUSP, and for the age group 20–24 years in National Nutrition Survey, the prevalence of overweight was the same (34%).

Other limitations are the measuring tools for data collection, such as the use of a parent-reported questionnaire to measure behavioral problems, rather than a clinical interview with the child. Since in a large epidemiological study a clinical interview is not feasible, the use of a questionnaire is optimal. Similarly, variables which could potentially confound the results, such as diet and physical activity levels, also relied on parental report rather than more direct measurements. Nonetheless, strengths of this study include the large sample size, the long follow-up time period, inclusion of a large variety of potential confounders, and measured BMI.

In conclusion, this study shows that childhood as well as persistent behavioral problems during childhood and adolescence predicts young adults' BMI and obesity. It is important to alert mental health professionals and carers to the link between child behavioral problems and poor physical health outcomes in adulthood. The findings of this study further highlight the need for broad prevention programs for childhood behavioral problems. If further studies confirm this association it will be important to monitor weight in children presenting with behavior problems, especially those with other risk factors for cardiovascular disease. It will also be important to clarify the nature of the mental health differences predicting weight gain and the causal mechanisms.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES

We thank all participants in the study, the MUSP data collection team, and Greg Shuttlewood, University of Queensland who have helped to manage the data for the MUSP. A.A.M. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. A.A.M. is funded by NHMRC Career Development Awards in Population Health (ID 519756). S.C. was funded by the University of Queensland Early Career Grant. This study was funded by the National Heart Foundation of Australia (G07B3135). The core MUSP study was funded by the National Health and Medical Research Council of Australia but the views expressed in the paper are those of the authors and not necessarily those of any funding body.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Disclosure
  9. REFERENCES
  • 1
    Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity: public-health crisis, common sense cure. Lancet 2002; 360: 473482.
  • 2
    Sourander A, Santalahti P, Haavisto A et al. Have there been changes in children's psychiatric symptoms and mental health service use? A 10-year comparison from Finland. J Am Acad Child Adolesc Psychiatry 2004; 43: 11341145.
  • 3
    Onyike CU, Crum RM, Lee HB, Lyketsos CG, Eaton WW. Is obesity associated with major depression? Results from the Third National Health and Nutrition Examination Survey. Am J Epidemiol 2003; 158: 11391147.
  • 4
    Lumeng JC, Gannon K, Cabral HJ, Frank DA, Zuckerman B. Association between clinically meaningful behavior problems and overweight in children. Pediatrics 2003; 112: 11381145.
  • 5
    Pine DS, Goldstein RB, Wolk S, Weissman MM. The association between childhood depression and adulthood body mass index. Pediatrics 2001; 107: 10491056.
  • 6
    Hasler G, Pine DS, Kleinbaum DG et al. Depressive symptoms during childhood and adult obesity: the Zurich Cohort Study. Mol Psychiatry 2005; 10: 842850.
  • 7
    Tanofsky-Kraff M, Cohen ML, Yanovski SZ et al. A prospective study of psychological predictors of body fat gain among children at high risk for adult obesity. Pediatrics 2006; 117: 12031209.
  • 8
    Anderson SE, Cohen P, Naumova EN, Must A. Association of depression and anxiety disorders with weight change in a prospective community-based study of children followed up into adulthood. Arch Pediatr Adolesc Med 2006; 160: 285291.
  • 9
    Franko DL, Striegel-Moore RH, Thompson D, Schreiber GB, Daniels SR. Does adolescent depression predict obesity in black and white young adult women? Psychol Med 2005; 35: 15051513.
  • 10
    Richardson LP, Davis R, Poulton R et al. A longitudinal evaluation of adolescent depression and adult obesity. Arch Pediatr Adolesc Med 2003; 157: 739745.
  • 11
    Goodman E, Whitaker RC. A prospective study of the role of depression in the development and persistence of adolescent obesity. Pediatrics 2002; 110: 497504.
  • 12
    Pine DS, Cohen P, Brook J, Coplan JD. Psychiatric symptoms in adolescence as predictors of obesity in early adulthood: a longitudinal study. Am J Public Health 1997; 87: 13031310.
  • 13
    Wurtman JJ. Depression and weight gain: the serotonin connection. J Affect Disord 1993; 29: 183192.
  • 14
    Ludwig DS. The glycemic index: physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. JAMA 2002; 287: 24142423.
  • 15
    Anderson SE, Cohen P, Naumova EN, Must A. Relationship of childhood behavior disorders to weight gain from childhood into adulthood. Ambul Pediatr 2006; 6: 297301.
  • 16
    Pulkki-Raback L, Elovainio M, Kivimaki M, Raitakari OT, Keltikangas-Jarvinen L. Temperament in childhood predicts body mass in adulthood: the Cardiovascular Risk in Young Finns Study. Health Psychol 2005; 24: 307315.
  • 17
    Canals J, Domenech-Llaberia E, Fernandez-Ballart J, Marti-Henneberg C. Predictors of depression at eighteen. A 7-year follow-up study in a Spanish nonclinical population. Eur Child Adolesc Psychiatry 2002; 11: 226233.
  • 18
    Haavisto A, Sourander A, Multimäki P et al. Factors associated with depressive symptoms among 18-year-old boys: a prospective 10-year follow-up study. J Affect Disord 2004; 83: 143154.
  • 19
    Aronen ET, Soininen M. Childhood depressive symptoms predict psychiatric problems in young adults. Can J Psychiatry 2000; 45: 465470.
  • 20
    Keeping JD, Najman JM, Morrison J et al. A prospective longitudinal study of social, psychological and obstetric factors in pregnancy: response rates and demographic characteristics of the 8556 respondents. Br J Obstet Gynaecol 1989; 96: 289297.
  • 21
    Najman JM, Bor W, O'Callaghan M et al. Cohort Profile: The Mater-University of Queensland Study of Pregnancy (MUSP). Int J Epidemiol 2005; 34: 992997.
  • 22
    World Health Organization. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation on Obesity, 3–5 June 1997. World Health Organization: Geneva, Switzerland, 1998.
  • 23
    Achenbach TM. Integrative guide for the 1991 CBCL/4-18, YSR, and TRF profiles. University of Vermont Department of Psychiatry: Burlington, VT, 1991.
  • 24
    Najman JM, Hallam D, Bor WB et al. Predictors of depression in very young children—a prospective study. Soc Psychiatry Psychiatr Epidemiol 2005; 40: 367374.
  • 25
    Stanger C, Achenbach TM, McConaughy SH. Three-year course of behavioral/emotional problems in a national sample of 4- to 16-year-olds: 3. Predictors of signs of disturbance. J Consult Clin Psychol 1993; 61: 839848.
  • 26
    Hernan MA, Hernandez-Diaz S, Werler MM, Mitchell AA. Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am J Epidemiol 2002; 155: 176184.
  • 27
    Bedford A, Foulds G. Delusions-Symptoms-States Inventory of Anxiety and Depression. NFER: Winsdor, England, 1978.
  • 28
    Barnes GE, Olson DH. Parent-adolescent communication scale. In: Olson DH, McCubbin HI, Barnes A, Larsen M, Muxen M, Wilson M (eds). Family Inventories: Inventories Used in a National Survey of Families Across the Family Life Cycle. Family Social Science, University of Minnesota: St. Paul, MN, 1982, pp 3348.
  • 29
    Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 12401243.
  • 30
    Koenker R, Hallock K, Quantile regression. J Econ Perspect 2001; 15: 143156.
  • 31
    Moffitt TE, Caspi A. Childhood predictors differentiate life-course persistent and adolescence-limited antisocial pathways among males and females. Dev Psychopathol 2001; 13: 355375.
  • 32
    Kim-Cohen J, Caspi A, Taylor A et al. MAOA, maltreatment, and gene-environment interaction predicting children's mental health: new evidence and a meta-analysis. Mol Psychiatry 2006; 11: 903913.
  • 33
    Essex MJ, Kraemer HC, Armstrong JM et al. Exploring risk factors for the emergence of children's mental health problems. Arch Gen Psychiatry 2006; 63: 12461256.
  • 34
    Najman JM, Aird R, Bor W et al. The generational transmission of socioeconomic inequalities in child cognitive development and emotional health. Soc Sci Med 2004; 58: 11471158.
  • 35
    McNamara RK, Carlson SE. Role of omega-3 fatty acids in brain development and function: potential implications for the pathogenesis and prevention of psychopathology. Prostaglandins Leukot Essent Fatty Acids 2006; 75: 329349.
  • 36
    Lissau I, Sorensen TI. Parental neglect during childhood and increased risk of obesity in young adulthood. Lancet 1994; 343: 324327.
  • 37
    Johnson JG, Cohen P, Kasen S, Brook JS. Childhood adversities associated with risk for eating disorders or weight problems during adolescence or early adulthood. Am J Psychiatry 2002; 159: 394400.
  • 38
    Taheri S. The link between short sleep duration and obesity: we should recommend more sleep to prevent obesity. Arch Dis Child 2006; 91: 881884.
  • 39
    Gregory AM, Rijsdijk FV, Dahl RE, McGuffin P, Eley TC. Associations between sleep problems, anxiety, and depression in twins at 8 years of age. Pediatrics 2006; 118: 11241132.
  • 40
    Bornstein SR, Schuppenies A, Wong ML, Licinio J. Approaching the shared biology of obesity and depression: the stress axis as the locus of gene-environment interactions. Mol Psychiatry 2006; 11: 892902.
  • 41
    Nagai N, Matsumoto T, Kita H, Moritani T. Autonomic nervous system activity and the state and development of obesity in Japanese school children. Obes Res 2003; 11: 2532.
  • 42
    McLennan W, Podger A. National Nutrition Survey: nutrient intakes and physical measurements, Australia, 1995. Australian Bureau of Statistics: Canberra, Australia, 1998. (Catalogue no. 4805.0).