Eur Child Adolesc Psychiatry

DOI 10.1007/s00787-016-0870-1

ORIGINAL CONTRIBUTION

Psychosocial risk factors underlie the link between attention

deficit hyperactivity symptoms and overweight at school entry
Ursula Pauli‑Pott1 · Alexander Reinhardt1 · Elena Bagus1 · Birgit Wollenberg2 ·
Andrea Schroer2 · Monika Heinzel‑Gutenbrunner1 · Katja Becker1

Received: 8 December 2015 / Accepted: 17 May 2016

© Springer-Verlag Berlin Heidelberg 2016

Abstract The link between symptoms of attention def-
icit hyperactivity disorder (ADHD) and increased body
weight is well established, while the underlying mech-
anisms are not yet clear. Since increased body weight
and ADHD symptoms have been found to be associated
with psychosocial risk factors in childhood, we ana-
lyzed whether the psychosocial risks explain the associ-
ation between the two conditions. The sample consisted
of 360 children (age range 6–7 years, 173 boys) attend-
ing the obligatory medical health exam before school
entry. The childrens’ height and weight were measured
during the examination. ADHD symptoms were ascer-
tained by parent-report questionnaires. Psychosocial
risks were ascertained by a structured interview. The
link between ADHD symptoms and body weight could
be completely explained by cumulative psychosocial
risks while controlling for gender, symptoms of depres-
sion/anxiety and oppositional defiant disorder of the
child, maternal smoking during pregnancy, parental
body mass index, and potential diagnosis of ADHD in
the parents. In current models pertaining to the etiol-
ogy of overweight/obesity and ADHD, chronic stress
caused by psychosocial adversity is assumed to act as
a trigger for these conditions. Psychosocial risks expe-
rienced during childhood may activate processes that
* Ursula Pauli‑Pott
Ursula.pauli‑pott@med.uni‑marburg.de
1
Department of Child and Adolescent Psychiatry,
Psychosomatics and Psychotherapy, Philipps University
of Marburg, Hans Sachs Str. 6, 35039 Marburg, Germany
2 tors of ADHD symptom development. The prediction by
Department of Public Health, District administration
Marburg-Biedenkopf, Schwanallee 23, 35037 Marburg,
Germany
specifically lead to the combined ADHD-overweight
phenotype.
Keywords Family adversity factors · Body weight ·
Attention deficit hyperactivity symptoms · Mediator
analysis
Introduction
In recent research, ample evidence was found for an asso-
ciation between attention deficit hyperactivity disorder
(ADHD) and overweight/obesity [1]. Several large-scale
community-based studies revealed that individuals show-
ing more ADHD symptoms possess a higher body weight
and are more frequently overweight than those who dis-
play fewer symptoms of ADHD [2–6]. The mediating
mechanisms that link these two conditions are still largely
unknown. Beyond the common symptoms such as impul-
sivity, uncontrolled eating, inattention to internal states, and
sleep disturbance [7–9] the two conditions share environ-
mental risk factors.
Regarding ADHD, there is broad evidence that psy-
chosocial adversity is involved in the pathogenesis of the
disorder. Specifically, family adversity factors have been
found to increase the risk of developing ADHD. Among
the indicators of family adversity (i.e., large sibship and
overcrowded living conditions, low socioeconomic status
(SES), parental antisocial behavior and psychopathology,
and family conflict) that had been derived from the Isle of
Wight studies [10], maternal psychopathology and fam-
ily conflict were found to be the most significant predic-
these risk factors held even after adjusting for confounders
such as maternal ADHD and smoking during pregnancy.

13
 Eur Child Adolesc Psychiatry

Particularly in boys, the risks were associated with defi- Methods

cits in global functioning [11–13]. The risk factors have
been assumed to act as nonspecific triggers of an under- Sample and procedure
lying genetic predisposition and to modify the course of
the disease [11]. Accordingly, Nigg et al. [14] showed The sample consisted of 360 children (age range 5–7 years,
that, depending on the presence of ADHD-related suscep- 173 boys) living in Marburg, Germany, who attended the
tibility genes, psychosocial risks (in particular interparen- obligatory health exam before school entry that is carried
tal conflict) were associated with an increase in ADHD out by the local public health department. The inclusion cri-
symptoms. teria for the participants were as follows: participation in the
Regarding overweight/obesity, there is also broad evi- health exam, more than 29 weeks gestation, normal to high
dence for an association with the presence of psychoso- body weight (i.e., age- and sex-related percentile of body
cial risk factors. Low SES, in particular a low parental mass index (BMI) above 10), and a mother with decent Ger-
education level, was found to be closely associated with man language skills. Only one child was included from each
a higher body weight and the development of overweight family.
and obesity in childhood [15, 16]. This association held At the health exam, parents were asked to take part in
after adjustment for many confounders such as over- a telephone interview regarding child development and
weight parents, child´s birth weight, and maternal smok- to fill out questionnaires pertaining to the child’s mental
ing during pregnancy [16]. Moreover, a socioeconomic health status. Weight and height of the child were meas-
disadvantage experienced during childhood and psy- ured during the health examination. The study was con-
chosocial risks such as parental divorce, physical abuse, ducted in cooperation with the Marburg Health Depart-
and prolonged separation from the parents, were found ment. The study protocol was approved by the medical
to be significantly associated with obesity in adulthood
[17, 18]. Prospective longitudinal studies demonstrated
that living with single mothers, exposure to social risks Table 1  Descriptive data of the sample
such as maternal depression, substance use, violence
Variable Statistics
between partners, and poverty, as well as family tur-
moil and crowded living conditions, predicted weight Age of child
gain between preschool and adolescent ages [19–22]. Years: m(s); range 5.95 (0.38); 5–7
As described by Boynton-Jarrett et al. [20], these asso- Age of mother
ciations may be stronger in girls in comparison to boys. Years: m(s); range 40.5 (5.3); 26–55
However, there are also contradictory results. Hanc et al. Age of father
[23], for example, found no association between self- Years: m(s); range 40.5 (6.1); 24–59
reported adverse events and body weight in 10–15 years- Gender of child
old children. Male: n (%) 173 (48.1)
Taken together, in the research summarized above, most Female: n (%) 187 (51.9)
studies showed that psychosocial risk scores are associated Psychosocial risks
with the presence of ADHD and an elevated body weight in Number: m(s); range 0.82 (0.93); 0–3
childhood. In particular, separation of parents, interparental ADHD-symptoms of child
conflict, a low level of education among parents, and indi- Composite score: m(s); range −0.04 (0.85); −1.15–2.75
cations of poverty appear to be involved in the development Positive ADHD-symptom screening
of both conditions. Thus, even though opposing effects of   No: n (%) 346 (96.1)
gender may exist, it seems possible that the presence of   Yes: n (%) 14 (3.9)
psychosocial risks explains the link between ADHD symp- BMI-SDS of child: m(s); range 0.03 (0.79); −1.27–2.90
toms and an elevated body weight in childhood (i.e., that BMI of child: m(s); range 15.6 (1.51); 13.61–23.52
psychosocial risks act as a third variable which explains the BMI of mother: m(s); range 24.4 (4.39); 17.44–43.93
relationship between ADHD symptoms and body weight). BMI of father: m(s); range 26.3 (3.51); 18.61–47.32
Therefore, we aim to examine this assumption. It is hypoth- Maternal smoking during pregnancy
esized that the link between ADHD symptoms and body No: n (%) 327 (90.8)
weight can be explained by the presence of psychosocial ≥1 cigarette per day: n (%) 33 (9.2)
risks. Furthermore, we explore whether gender of child
Parental ADHD-diagnosis
moderates (i.e., acts as a moderator variable of) the asso-
No: n (%) 351 (97.5)
ciations between psychosocial risks, ADHD symptoms and
Yes: n (%) 9 (2.5)
body weight.

13
Eur Child Adolesc Psychiatry
faculty’s local ethics committee at the Philipps University
of Marburg. Written informed consent was obtained from
all of the participants. Table 1 contains descriptive data of
the sample.
Variables
ADHD‑symptoms of the child
For the assessment of ADHD symptoms, we used the
ADHD rating scale “FBB-ADHS-V” by Döpfner et al. [24]
and the Hyperactivity scale of the German versions of the
Strength and Difficulties Questionnaire (SDQ) by Woerner
et al. [25]. The parents filled out the questionnaires.
The FBB-ADHS-V measures ADHD symptoms accord-
ing to ICD-10 in preschool children (norms are available
for children between 3, 0 and 7, 11 years). Internal consist-
ency of the parent version was 0.94 (Cronbach’s alpha).
The questionnaire accurately differentiates between chil-
dren with and without ADHD [24, 26].
Internal consistency of the Hyperactivity Scale of the
German version of the SDQ was 0.77 (Cronbach’s alpha)
[25]. The factor structure was in good correspondence with
the original version [27]. In the present study, the correla-
tion between the FBB-ADHS-V and SDQ-Hyperactivity
score was r = 0.70. For further computations, the two
scores were z-transformed and added up to increase relia-
bility of measurement. Of the 360 children, n = 14 (3.9 %)
exceeded the clinical cutoff score of either questionnaire.
This rate corresponds well to the rate of ADHD diagno-
sis according to the child’s age [28, 29] indicating that the
sample includes the upper range of the dimensionally dis-
tributed (see [30]) ADHD symptoms.
Child body weight
The medical doctors at the Public Health Service measured
each child’s body weight and height at the school entry
examination. Children were weighed and measured while
wearing lightweight clothing without shoes, using a cali-
brated scale (Seca 709) and stadiometer (Seca 220). Each
child’s BMI was calculated. On the basis of the German ref-
erence data for children [31], the BMI calculated was trans-
formed into an age- and gender-corrected standard deviation
score (BMI-SDS) using the least mean square method by
Cole et al. [32], which normalizes the resulting distribution.
Psychosocial risks
To assess adverse psychosocial conditions we used the psy-
chosocial risk index by Laucht et al. [33] that represents an
expanded family adversity index (see [13]). During the struc-
tured telephone interviews with the mother, the following
psychosocial risks were assessed: (1) at least one parent with-
out occupational qualification, (2) at least one parent with a
low education level, (3) overcrowded living conditions (more
than one person per room or a residence space less than
50 m2), (4) at least one parent with a psychiatric disease, (5)
at least one parent with a broken home background, (6) early
parenthood (at least one parent was less than 18 years old at
the time of the child’s birth), (7) parental separation, and (8)
unwanted pregnancy. We calculated a sum score of these risks.
Control variables
For controlling purposes, we used the BMI and potential
ADHD diagnoses of parents to approximate the familiar-
ity of the conditions. We further assessed maternal smoking
during pregnancy with the child as well as anxiety/depres-
sive and aggressive/oppositional symptoms of the child,
because these variables might also explain the link between
ADHD symptoms and body weight [34].
During the interviews, each mother was asked to report
their own and the father’s current weight and height to cal-
culate the BMI of each parent. Each mother was also asked
whether the father or the mother of the child had ever been
diagnosed with ADHD and whether and how many ciga-
rettes she smoked per week during her pregnancy with
the child. For the assessment of anxiety/depressive and
aggressive/oppositional symptoms of the child, the SDQ-
subscales emotional symptoms and conduct problems were
used.
Statistical analysis
To assess associations between ADHD symptoms (as well
as positive versus negative ADHD symptom screening),
BMI-SDS and the psychosocial risks correlation and partial
correlation statistics were calculated. Student’s t-tests were
calculated to compare children with positive versus nega-
tive ADHD symptom screening. However, because of the
relatively low number of children who screened positive for
ADHD (and therefore a small variance of the dichotomous
variable), we refrained from considering this variable in
the multivariate analyses. To explore the interaction effects
between the psychosocial risks and gender of the child on
ADHD symptoms or BMI-SDS, we used multiple hierar-
chical regression analyses. To test whether psychosocial
risks act as a third variable which explains the common
variance (i.e., the overlap) between ADHD symptoms and
BMI-SDS, we used the bootstrapping procedure (in com-
parison to Sobel tests, assumptions made by the procedure
are more realistic) recommended by Preacher and Hayes
[35]. For these calculations, we used the SPSS Macro
‘Indirect’ [35]. Here, a “mediation” path model is esti-
mated using ordinary least square (OLS) regression. This
13

procedure is suitable for analyzing the role of third vari-
ables (e.g., mediators, confounders, suppressor variables)
in relationships between two variables [36]. It is analyzed
whether psychosocial risks explain the common variance
between ADHD symptoms and body weight. A significance
level of p < 0.05 is used. All calculations were conducted
using IBM SPSS Statistics software (IBM Corp.).
Results
Associations between ADHD symptoms, body weight,
and psychosocial risks
ADHD symptoms of the child correlated significantly with
his/her BMI-SDS (r = 0.14, p = 0.006). The children who
screened positive for ADHD had a significantly higher
BMI-SDS than children with a negative screening (positive
screening: mean (m) = 0.49; standard deviation (s) = 1.02;
negative screening: m = 0.01; s = 0.77; t = 2.24,
p = 0.025). After controlling for gender of the child,
maternal smoking during pregnancy, BMI of the mother
and father, diagnosis of ADHD in the parents, SDQ-emo-
tional symptoms, and SDQ-behavior problems, the asso-
ciation remained statistically significant (partial r = 0.12,
p = 0.020). The psychosocial risk score correlated signifi-
cantly with the child’s BMI-SDS (r = 0.22, p < 0.001) and
ADHD symptoms (r = 0.23, p < 0.001). The children who
screened positive for ADHD displayed a greater number
of the psychosocial risks in comparison to children who
screened negative (positive screening: m = 1.6; s = 1.1;
negative screening: m = 0.8; s = 1.0; t = 2.72, p = 0.007).
After adjusting for the control variables, the significance of
the correlations held (psychosocial risks with ADHD symp-
toms: r = 0.15, p = 0.003; psychosocial risks with BMI-
SDS: r = 0.20, p < 0.001). Because of the small number
of children with positive screening for ADHD, the variable
was not considered in the multivariate analyses.
The child’s gender did not affect the association
between the psychosocial risk score and ADHD symp-
toms (interaction effect between gender and psychosocial
risks on ADHD symptoms: F(1,352) = 1.60, p = 0.210)
or BMI-SDS (interaction effect between sex and the pres-
ence of psychosocial risks on BMI-SDS: F(1,352) = 1.11,
p = 0.293).
Role of psychosocial risks
Table 2 shows the results of the “mediation” analy-
ses (with adjustment for all control variables). The link
between ADHD symptoms of the child and his/her BMI-
SDS was completely explained by the psychosocial risk
score. In other words, after adjusting for the effect of the
13
Eur Child Adolesc Psychiatry
Table 2  Analysis of the “third variable” effect of psychosocial risks
Link between ADHD symptoms and BMI-SDS: total effect: t = 2.39,
p = 0.017
Indirect (third variable) Bootstrap CI95 Remaining direct effect
effect of
Psychosocial risk score 0.0051–0.0292 t = 1.88, p = 0.061
Parental education 0.0015–0.0226 t = 2.07, p = 0.039
Overcrowded living ns t = 2.34, p = 0.020
conditions
Parental broken home ns t = 2.33, p = 0.020
Parental separation 0.0008–0.0283 t = 2.00, p = 0.046
  a. Parental education 0.0018–0.0261
  b. Parental separation 0.0009–0.0285
  Total a + b 0.0054–0.0367 t = 1.69, p = 0.091
Control variables: gender of the child, BMI of the mother and father,
diagnosis of ADHD in the parents, maternal smoking during preg-
nancy, SDQ-emotional symptoms, and SDQ-conduct problems
ns not statistically significant
psychosocial risks, the direct link between ADHD symp-
toms and BMI-SDS was no longer statistically significant.
To further explore this result, we analyzed whether and
which single risk factor in particular contributed to the sig-
nificant effect. In these additional exploratory analyses,
the single risk factors that showed sufficient variance (i.e.,
were present in more than 10 % of the sample) could be
considered that applied to the following risks: overcrowded
living conditions, low parental education, parental sepa-
ration, and parents with a broken home background. We
analyzed the path models with each of these single risks
as “third” variables. The results of these analyses are also
shown in Table 2. Parental separation and low parental edu-
cation explained the link between ADHD symptoms and
BMI-SDS significantly, but in each case just partially. In
a subsequent analysis, we found that the significant third-
variable effects of these two risks were independent of each
other and that the two risks together completely explained
the association between ADHD symptoms and BMI-SDS
(see Table 2).
Discussion
In a community-based sample of children aged between
5 and 7 years, we found that small but significant asso-
ciation between ADHD symptoms and body weight (that
applied also to children with positive versus negative
ADHD screening and that held after controlling for sev-
eral important covariables) could be completely explained
by the presence of shared psychosocial risks. Among the
risks, parental separation and parents with a low level of
Eur Child Adolesc Psychiatry
education in particular proved to be significant independent
third variables that underlie ADHD-body weight link.
Overweight and ADHD have been assumed to be trig-
gered by chronic stress caused by psychosocial adversity
factors. Regarding the development of overweight during
childhood, it was assumed that stress and inadequate par-
enting engendered by psychosocial adversity impede on
adequate self-regulation development that leads to uncon-
trolled eating [21, 37]. In the domain of ADHD, psychoso-
cial risk factors are regarded as rather unspecific stress-elic-
iting factors that trigger the biological vulnerability for the
disease [38]. Psychosocial adversity might act as a general
predictor of social adjustment and emotional health devel-
opment in childhood [11].
In the present study, as far as we know, we showed for
the first time that cumulative psychosocial adversity fac-
tors completely explain the well-established association
between the two conditions. An interesting aspect is that
specifically two risk factors, a low level of parental edu-
cation and parental separation, independent of each other
explained parts of this association. That means that on the
one hand there are children who develop ADHD symp-
toms and overweight associated with separation/divorce of
their parents, and on the other hand, there are children who
developed ADHD symptoms and overweight in families
with low parental education. These “two pathways” might
correspond to the two routes of psychosocial influences
on weight gain distinguished by Davis et al. [17]: a physi-
ological route that is associated with chronic stress and a
psychosocial route that is associated with inadequate health
behaviors. The stress route might primarily involve parental
conflict/separation that most likely causes stress and emo-
tional insecurity in children [39, 40]. Low parental educa-
tion might be associated with inadequate health behaviors
and less adequate parenting. Both routes could affect the
development of ADHD symptoms as well as weight gain.
These interpretations do not rule out the possibility that
ADHD symptoms and elevated body weight have a com-
mon genetic basis and therefore show common symptoms
and biological markers [41–43]. In part, the psychosocial
risks assessed in the present study also may be caused by
these genetic factors. Our study design does not allow dis-
tinguishing between genetic and environmental influences.
Given current theorizing and empirical findings, it is prob-
able that genetic factors also in part underlie the associa-
tion between ADHD symptoms and body weight and that
interaction effects between the genetic vulnerability and
psychosocial risk conditions leading to the combination of
ADHD symptoms and elevated body weight [44].
Our study may contain limitations which should be dis-
cussed. Firstly, we assessed ADHD symptoms based on
parent reports that might be biased. However, several other
studies on the link between ADHD and overweight used
parent reports as well and Cortese et al. [1] recently found
that this measuring method did not influence the strength
of the link. Secondly, we used a dimensional definition of
ADHD instead of a categorical one. A dimensional meas-
urement approach most likely fits the empirical nature of
the distribution of ADHD [30, 45]. However, we demon-
strated that children who exceeded the symptom threshold
of a validated screening questionnaire (a criterion that had
been used in corresponding research, see [1]) showed sig-
nificantly higher body weights and the presence of more
psychosocial risks than children who scored below the
threshold. Moreover, for economical reasons we just asked
whether parents had ever been diagnosed with ADHD
and did not use structured interviews or questionnaires to
assess parental ADHD symptoms. Thirdly, the effect sizes
in our study were rather small. ADHD and overweight/obe-
sity are heterogeneous conditions and our findings apply
to the subgroups of children with ADHD symptoms and
heightened body weight. Therefore, larger effects cannot
be expected within a community based sample. Finally, we
excluded children with very low gestational age and current
underweight to control for these conditions. Therefore, the
results cannot be generalized to these children.
In all, our results point to the following: at early school
age, the presence of psychosocial risks that are associated
with chronic stress and less adequate parenting might have
contributed (e.g., by interaction effects with genetic factors)
to the development of symptoms of ADHD combined with
high body weight. It might be that the psychosocial risks
predict the worsening of the symptoms in the following
years and subsequently the full-blown co-morbid ADHD-
obesity phenotype. This possibility should be analyzed by
future longitudinal studies. Results of such studies could be
useful for tailoring future prevention programs [46] to meet
the needs of the children who are at risk for developing this
co-morbid phenotype.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest except for K. Becker who was on the Advisory Board of Eli
Lilly/Germany, was a member of the Scientific Committee of Shire,
and was paid for public speaking by Shire. She was not involved in
research/clinical trials sponsored by pharmaceutical industry within
the last five years.
References
1. Cortese S, Moreira-Maia CR, St Fleur D, Morcillo-Penalver C,
Rohde LA, Faraone SV (2016) Association between ADHD and
obesity: a systematic review and meta-analysis. Am J Psychiatry
173(1):34–43
2. de Zwaan M, Gruss B, Muller A, Philipsen A, Graap H, Martin
A, Glaesmer H, Hilbert A (2011) Association between obesity
13

and adult attention-deficit/hyperactivity disorder in a German
community-based sample. Obes Facts 4(3):204–211
3. Erhart M, Herpertz-Dahlmann B, Wille N, Sawitzky-Rose B,
Holling H, Ravens-Sieberer U (2012) Examining the relationship
between attention-deficit/hyperactivity disorder and overweight
in children and adolescents. Eur Child Adolesc Psychiatry
21(1):39–49
4. Fuemmeler BF, Ostbye T, Yang C, McClernon FJ, Kollins SH
(2010) Association between attention-deficit/hyperactivity disor-
der symptoms and obesity and hypertension in early adulthood: a
population-based study. Int J Obes 35(6):852–862
5. Lam LT, Yang L (2007) Overweight/obesity and attention deficit
and hyperactivity disorder tendency among adolescents in China.
Int J Obes 31:584–590
6. Waring E, Lapane KL (2008) Overweight in children and ado-
lescents in relation to attention-deficit/hyperactivity disorder:
results from a national sample. Pediatrics 122(1):e1–e6
7. Puder JJ, Munsch S (2010) Psychological correlates of child-
hood obesity. Int J Obes (Lond) 34(Suppl 2):S37–S43
8. Davis CR, Levitan RD, Smith M, Tweed S, Curtis C (2006)
Associations among overeating, overweight, and attention
deficit/hyperactivity disorder: a structural equation modelling
approach. Eat Behav 7:266–274
9. Owens J, Gruber R, Brown T, Corkum P, Cortese S, O’Brien L,
Stein M, Weiss M (2013) Future research directions in sleep and
ADHD: report of a consensus working group. J Atten Disord
17(7):550–564
10. Rutter ML (1999) Psychosocial adversity and child psychopa-
thology. Br J Psychiatry 174:480–493
11. Biederman J (2005) Attention-deficit/hyperactivity disorder: a
selective overview. Biol Psychiatry 57(11):1215–1220
12. Biederman J, Faraone SV, Monuteaux MC (2002) Differential
effect of environmental adversity by gender: Rutter’s index of
adversity in a group of boys and girls with and without ADHD.
Am J Psychiatry 158:1556–1562
13. Biederman J, Milberger S, Faraone SV, Kiely K, Guite J, Mick E,
Ablon S, Warburton R, Reed E (1995) Family-environment risk
factors for attention-deficit hyperactivity disorder. A test of Rut-
ter’s indicators of adversity. Arch Gen Psychiatry 52(6):464–470
14. Nigg JT, Nikolas M, Burt SA (2010) Measured gene-by-environ-
ment interaction in relation to attention-deficit/hyperactivity dis-
order. J Am Acad Child Adolesc Psychiatry 49(9):863–873
15. Lamerz A, Kuepper-Nybelen J, Wehle C, Bruning N, Trost-
Brinkhues G, Brenner H, Hebebrand J, Herpertz-Dahlmann
B (2005) Social class, parental education, and obesity preva-
lence in a study of six-year old children in Germany. Int J Obes
29:373–380
16. Wijlaars LP, Johnson L, van Jaarsveld CH, Wardle J (2011) Soci-
oeconomic status and weight gain in early infancy. Int J Obes
(Lond) 35(7):963–970
17. Davis CR, Dearing E, Usher N, Trifiletti S, Zaichenko L, Ollen
E, Brinkoetter MT, Crowell-Doom C, Joung K, Park KH, Man-
tzoros CS, Crowell JA (2014) Detailed assessments of child-
hood adversity enhance prediction of central obesity independ-
ent of gender, race, adult psychosocial risk and health behaviors.
Metabolism 63(2):199–206
18. Sobal J, Stunkard AJ (1989) Socioeconomic status and obesity: a
review of the literature. Psychol Bull 105(2):260–275
19. Chen AY, Escarce JJ (2010) Family structure and childhood obe-
sity, Early Childhood Longitudinal Study—Kindergarten Cohort.
Prev Chron Dis 7(3):A50
20. Boynton-Jarrett R, Fargnoli J, Suglia SF, Zuckerman B, Wright
RJ (2010) Association between maternal intimate partner vio-
lence and incident obesity in preschool-aged children: results
from the Fragile Families and Child Well-being Study. Arch
Pediatr Adolesc Med 164(6):540–546
13
Eur Child Adolesc Psychiatry
21. Suglia SF, Duarte CS, Chambers EC, Boynton-Jarrett R (2012)
Cumulative social risk and obesity in early childhood. Pediatrics
129(5):e1173–e1179
22. Wells NM, Evans GW, Beavis A, Ong AD (2010) Early child-
hood poverty, cumulative risk exposure, and body mass index
trajectories through young adulthood. Am J Public Health
100(12):2507–2512
23. Hanc T, Janicka K, Durda M, Cieslik J (2014) An association
between adverse events, anxiety and body size of adolescents. J
Biosoc Sci 46(1):122–138
24. Döpfner M, Görtz-Dorten A, Lehmkuhl G (2008) DISYPS-II
Diagnostik-System für psychische Störungen nach ICD-10 und
DSM-IV für Kinder und Jugendliche - II. Huber, Bern
25. Woerner W, Becker A, Friedrich C, Klasen H, Goodman
R, Rothenberger A (2002) Normierung und Evaluation der
deutschen Elternversion des Strengths and Difficulties Question-
naire (SDQ): Ergebnisse einer repräsentativen Felderhebung. Z
Kinder Jugendpsychiatrie Psychother 30(2):105–112
26. Breuer D, Dopfner M (2008) Development of a questionnaire
for the assessment of attention-deficit-/hyperactivity disorder
(ADHD) in preschoolers by parents and/or teacher ratings. Z
Entwickl Padagogis 40(1):40–48
27. Koglin U, Barquero B, Mayer H, Scheithauer H, Petermann F
(2007) Deutsche Version des Strengths and Difficulties Ques-
tionnaire (SDQ-Deu). Diagnostica 53(4):175–183
28. Huss M, Holling H, Kurth BM, Schlack R (2008) How often are
German children and adolescents diagnosed with ADHD? Preva-
lence based on the judgment of health care professionals: results
of the German health and examination survey (KiGGS). Eur
Child Adolesc Psychiatry 17(Suppl 1):52–58
29. Egger HL, Angold A (2006) Common emotional and behavioral
disorders in preschool children: presentation, nosology, and epi-
demiology. J Child Psychol Psychiatry 47(3–4):313–337
30. Coghill D, Sonuga-Barke EJ (2012) Annual research review:
categories versus dimensions in the classification and conceptu-
alisation of child and adolescent mental disorders–implications of
recent empirical study. J Child Psychol Psychiatry 53(5):469–489
31. Kromeyer-Hauschild K, Wabitsch M, Kunze D, Geller D, Geiss
HC, Hesse V, von Hippel A, Jaeger U, Johnsen D, Korte W,
Menner K, Muller G, Muller JM, Niemann-Pilatus A, Remer
T, Schaefer F, Wittchen HU, Zabransky S, Zellner K, Ziegler A,
Hebebrand J (2001) Percentiles of body mass index in children
and adolescents evaluated from different regional German stud-
ies. Monatsschr Kinderheilk 149(8):807–818
32. Cole TJ (1990) The LMS method for constructing normalized
growth standards. Eur J Clin Nutr 44:45–60
33. Laucht M, Skowronck MH, Becker K, Schmidt MH, Esser G,
Schulze TG, Rietschel M (2007) Interacting effects of dopamine
transporter gene and psychosocial adversity on attention-deficit/
hyperactivity disorder symptoms among 15-year-olds from a
high-risk community sample. Arch Gen Psychiatry 64:585–590
34. Pauli-Pott U, Neidhard J, Heinzel-Gutenbrunner M, Becker K
(2014) On the link between attention deficit/hyperactivity disor-
der and obesity: do comorbid oppositional defiant and conduct
disorder matter? Eur Child Adolesc Psychiatry 23(7):531–537
35. Preacher KJ, Hayes AF (2008) Asymptotic and resampling strat-
egies for assessing and comparing indirect effects in multiple
mediator models. Behav Res Methods 40(3):879–891
36. MacKinnon DP, Krull JL, Lockwood CM (2000) Equivalence
of the mediation, confounding and suppression effect. Prev Sci
1(4):173–181
37. Anderson SE, Gooze RA, Lemeshow S, Whitaker RC (2012)
Quality of early maternal-child relationship and risk of adoles-
cent obesity. Pediatrics 129(1):132–140
38. Faraone SV, Asherson P, Banaschewski T, Biederman J, Buite-
laar J, Ramos-Quiroga JA, Rohde LA, Sonuga-Barke EJ,
Eur Child Adolesc Psychiatry
Tannock R, Franke B (2015) Attention deficit/hyperactivity dis-
order. Nat Rev Dis Prim 1:15020. doi:10.1038/nrdp.2015.20
39. Cummings EM, Davies PT (2011) Marital conflict and children:
an emotional security perspective. Guilford Publications, New
York
40. Davies PT, Cummings EM (1994) Marital conflict and child
adjustment—an emotional security hypothesis. Psychol Bul
116(3):387–411
41. Cortese S, Konofal E, Dalla Bernardina B, Mouren MC, Lecen-
dreux M (2008) Does excessive daytime sleepiness contribute to
explaining the association between obesity and ADHD symp-
toms? Med Hypotheses 70(1):12–16
42. Cortese S, Angriman M (2014) Attention-deficit/hyperactivity
disorder, iron deficiency, and obesity: is there a link? Postgrad
Med 126(4):155–170
43. Hanc T, Slopien A, Wolanczyk T, Dmitrzak-Weglarz M, Szwed
A, Czapla Z, Durda M, Ratajczak J, Cieslik J (2015) ADHD and
overweight in boys: cross-sectional study with birth weight as a
controlled factor. Eur Child Adolesc Psychiatry 24(1):41–53
44. Cortese S, Morcillo Penalver C (2010) Comorbidity between
ADHD and obesity: exploring shared mechanisms and clinical
implications. Postgrad Med 122(5):88–96
45. Thapar A, Langley K, O’Donovan M, Owen M (2006) Refining
the attention deficit hyperactivity disorder phenotype for molec-
ular genetic studies. Mol Psychiatry 11(8):714–720
46. Cortese S, Castellanos FX (2014) The relationship between
ADHD and obesity: implications for therapy. Exp rev Neurother-
apeut 14(5):473–479
13