Correlation between Obesity and Autistic Symptoms in Children: Study

This research report presents a cross-sectional case-control study conducted at Ankara Children’s Health and Diseases Hematology-Oncology Training and Research Hospital, investigating the correlation between obesity and autistic symptoms in children. The study compared 204 obese children and adolescents with 143 normal-weight controls, utilizing socio-demographic data forms, the Social Communication Questionnaire (SCQ), and the Autism Behavior Checklist (ABC) to assess autism spectrum disorder and autistic symptoms. The results indicated a higher prevalence of autistic symptoms in the obese cohort, with significant differences in ABC scores and a higher proportion of children exceeding the SCQ cutoff. The study's findings suggest a potential association between obesity and autistic symptoms, highlighting the need for further research into this relationship. The study also found that 3 children had a positive ASD diagnosis, all from the obese cohort. Statistical analysis revealed significant differences in several metrics, including the total ABC score and relating (social skills) and sensory scores, which were higher in the study cohort. The research underscores the importance of considering both psychological and physical aspects of obesity and its potential links to neurodevelopmental conditions.

Running head: AUTISTIC SYMPTOMS AND ASD IN OBESE CHILDREN 1
Investigation in the Correlation between Obesity and Autistic Symptoms and Autism Spectrum
Disorder in Children
Name
Institution

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 2
INVESTIGATION IN THE CORRELATION BETWEEN OBESITY AND AUTISTIC
SYMPTOMS AND AUTISM SPECTRUM DISORDER IN CHILDREN
Abstract
Objective: Obesity remains a chronic disorder which triggers both psychological and physical
challenges and need not to be probed only from the biological viewpoint. Various obesity-linked
psychiatric diseases have been documented in past studies. Nevertheless, there is never adequate
information on the correlation between obesity and autism/autistic symptoms in literature.
However, high rates of obesity among autistic adolescents and children remained further stressed
in the latest studies. The purpose of this study is to compare autistic and autistic symptoms in
healthy and obese children.
Method: The study cohort entailed 204 adolescents and children with obesity whereas the
control cohort entailed 143 children and adolescents in normal weight drawn from Ankara
Children’s Health and Diseases Hematology-Oncology Training and Research Hospital. The
participants were handed socio-demographic data form, Social Communication Questionnaire
(SCQ), and Autism Behavior Checklist (ABC). The assessment of autism spectrum disorder and
autistic symptoms were done utilizing questionnaires and clinical interviews based on DSM-V.
Conclusion: 3 children had positive ASD diagnosis. All of them were from obese cohort. The
two groups remained identical based on SCQ’s total score and SCQ’s subscales (p>0.050).
Nonetheless, the quantity of children with SCQ (positive score) beyond cutoff stood higher
significantly in study cohort with p equals 0.0430. The total score of ABC (p=0.0140) alongside
relating (social skills) and sensory scores subgroups remained higher significantly in study
cohort (p=0.0040; p=0.0000).

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 3
Discussion: The study autistic symptoms were discovered to be more amongst adolescents and
children with obesity than children with normal weight in similar age and gender cohort.
Key Words: Obesity, social communication questionnaire, autistic symptom, autism
Introduction
Obesity is among the most severe global health issue in the 21st century. The World
Health Organization (WHO) defines obesity as the “abnormal/excessive accumulation of fat
which presents a real risk to health”. It is a metabolic disease that is triggered by the interaction
of genetic and environmental factors. Studies have found that approximately two-hundred genes
take part in the development of obesity. Obesity is amongst the most shared chronic illnesses of
childhood. It affects nearly between 25 and 30 percent of adolescents and children. Currently,
over forty million obese or overweight children exit globally (Baron-Cohen, Wheelwright,
Skinner, Martin & Clubley, 2001).
The prevalence of obesity continues to rise. It is probably that there exists an interaction
between environment and genetic predisposition whereby genetically vulnerable children react
with put on weight on the environment with a surged existence of energy-dense foods and
reduced energy spending. Obesity has social, mental and physical complications that severely
affect life. For this reasons, studies about mental health and obesity have increasingly gained
momentum in the recent years (Avcil, Baykara, Baydur, Münir & Emiroğlu, 2015). A strong
statistical relationship between anxiety-and depression- disorders has been acknowledged in a
community-wide study. Mental illness linked with low self-esteem, depression, eating disorder,
body dysmorphic and social interactions problems have been more prevalence in obese
adolescents than normal peers. Obese children have been less successful in school. They also
experience additional problems among friends and families compared to non-obese children.

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 4
Obese adolescents showed weaker cognitive functions when given neurocognitive
assignments. Recent studies have recognized that attention deficit hyperactivity disorder
(ADHD) is more prevalent among the obese adolescents and children than normal weight peers.
Children with ADHD show higher prevalence of obesity. According to the study by Lundström
et al. (2015) conducted to compare the prevalence of annual autism symptom phenotype and of
registered diagnose for ASD during a ten year period in children, it was found that the annual
prevalence of autism phenotype stood stable during the ten year period (p=0.870 for the linear
time trend). In comparison, the study established a monotonic significant increase in prevalence
of registered diagnose of ASD in national patient register (p<0.0010) for liner time trend)
((Lundström et al. 2015)). The study further established that autism symptom phenotype’s
prevalence has stayed stable in children in Sweden whereas the official prevalence for registered,
clinically diagnosed, ASD has increased significantly (Lundström et al. 2015). This suggested
that administrative changes that affect registered prevalence, instead of secular factors that affect
the pathogenesis, are significant for the increase in reported ASD prevalence (Lundström et al.
2015). It has also been reported that autistic symptoms can further be observed usually in
unaffected societal individuals. Nevertheless, only limited studies have focused on autism and
obesity in scientific literature. In such limited studies, high risks of obesity and unhealthy gain in
weight were recorded among autistic adolescents and children. It has been discovered that
obesity and unhealthy weight gain prevalence stood higher amongst autistic children aged
between two and five years as opposed to the general population (Hill, Zuckerman & Fombonne, 2015).
Very little study exists in literature on whether or not obese children and adolescents showcase
autism and autistic symptoms. It was hypothesize that there is a close correlation between

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 5
obesity and autism and autistic symptoms. In the present study, obese adolescents and children
were compared to normal weight peers in relation to autism and autistic symptoms.
Methods and Material
Study Setting and Design
The cross-sectional case-control prospective study was used. The study cohort
encompassed two-hundred and four (204) adolescents and children applied to Department of
Pediatrics at Ankara Children’s Health and Diseases Hematology-Oncology Training and
Research Hospital between 2016 June and 2017 June. The subjects were aged between seven-
and seventeen years. The parents of the subjects and patients showed fluency in Turkish
language and had been diagnosed with obesity. The patients with intellectual abilities, applied to
the psychiatric clinic, chronic illness, and genetic syndrome of obesity were excluded from the
study cohort. No subject was diagnosed with a known genetic syndrome with obesity alongside
neurodevelopment characteristics like Bardet-Billed Syndrome, Fragile X or Prader-Willi
Syndrome based on clinical interview (DSM-V). The control cohort entailed one-hundred and
forty-three (143) volunteer normal weight adolescents and children chosen for health check-ups
purposes in same hospital. The control group had no chronic illness and was not applied to the
psychiatric clinic. The control group also had similar age and gender as the treated group. A
consecutive approach was used to reduce bias and it included all referred patients. The process of
participants’ recruitment is shown in figure one below:

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 6
Both control and treated cohorts were given socio-demographic data form, Social
Communication Questionnaire (SCQ) and Autism Behavior Checklist (ABC). Psychometric tests
or psychiatric interviews were performed when required by specialists in adolescents and
children mental health for feasible intellectual insufficiency or psychopathologies detections in
the control and study group based on clinical interview (DSM-V). A total of one-hundred and
seven (107) amount them, 46 controls and 61 obese in the study group, diagnosed (clinical
interview-DSM-V) with intellectual disability or any illness (anxiety, depression, or ADHD
among others with an exception of autism) and were subsequently excluded from the study.
Diagnosis of autism and associated autistic symptoms were determined based on a

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 7
comprehensive critical ABC, SCQ scales and clinical interviews (DSM-V). The participants with
higher ABC and/or SCQ cut-off scores were subsequently re-evaluated for Autism Spectrum
Disorder (ASD for a 2nd round amongst adolescents and children by a specialist in the mental
health. The Local Ethics Committee of Ankara Children’s Health and Diseases Hematology-
Oncology Training Research Hospital approved the study. All families were informed regarding
the study and the corresponding informed consents were taken in adherence to the Declaration of
Helsinki requirements.
Tools Utilized
Demographic Information Form (DIF): The socio-demographic of participants were surveyed
by questionnaires that the researchers prepared besides the regularly utilized forms in clinic
(Parner et al., 2011). All the participants filled the DIF.
Autism Behavior Checklist (ABC): The researcher used this scale in assessing the symptoms of
autism particularly among school children with severe complaints. Comprising fifty-seven items
in five subscales (measuring sensory, relating (social skills,), body and object use, language, and
social and self-help skills) ABC scale was completed by parents or teachers who knew the
children best (Oro, Navarro-Calvillo & Esmer, 2014). The reliability and validity study of
Turkish version was carried out by Yilmaz-Irmak et al.
Social Communication Questionnaire (SCQ): The SCQ was initially known as Autism
Screening Questionnaire (ASQ). It is a reliable and valid measure utilized as a screening tool for
Pervasive Developmental Disorder (PDD) (Barnard-Brak, Richman, Chesnut & Little, 2016).
This tool consists of forty items hinged on Autism Diagnostic Interview-Revised (ADI-R). It is
completed by primary caregivers. The forty items query reciprocal interactions, language-
communication and restricted, repetitive and stereotyped behavior. These queries remain

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 8
consistent with autism criteria in DSM-V. The questions on scale were answered “yes” or “no”.
About ten minutes is taken to apply and about five minutes is taken to score.
Body Mass Index (BMI): Using the BarimedR Electronic Body Scale (SC-105 with 0.10 kg
accuracy), the researcher conducted body weight measurements after a ten-hour fasting, barefoot
and daily clothes retained. The researcher measured heights barefoot using Ayrton R Stadiometer
Model S100 with 0.1 centimeter accuracy. The researcher calculated the BMI using formula,
kg/m2. The researcher defined obesity as a BMI of over 95th percentile for sex and age. The
researcher then defined normal weight as a BMI between fifth and eighty-fifth percentile for sex
and age (Locke et al., 2015). The researcher used Centers for Disease Control and Prevention
(CDC) data to calculate the BMI standard deviation score (BMISDS).
Statistics: One hundred- and forty (140) patients was the minimum sample size needed to
compare age groups based on symptoms in every group with a value of 0.050 and an eighty-five
percent statistical power. The researcher expressed categorical variables as frequency (n) and
percentage (%). The researcher expressed the continuous variables as minimum-maximum,
mean, median, and standard deviation. The researcher analyzed the fitness to normal distribution
using the Kolmogorov-Smirnov test. The categorical variables were compared by the researcher
using Father’s exact test and Chi-Square test. The continuous variables were compared by the
researcher using the Mann-Whitney U test. The level of significance was accepted as p<0.050.
Results
The two groups remained identical in respect of sex distribution (p>0.050) and material
range of parental education and birth (p>0.050). The researcher discovered that the quantity of
siblings was mainly three or two. The two cohorts remained identical in respect of family history
of psychiatric disorder (p>0.050). The history of postnatal (p>0.027) and parental (p<0.001)

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 9
problem remained significantly high in study (treated) cohort. The 2 cohorts remained identical
on the basis of variable of type of birth and preterm-post term birth (p>0.05). Albeit peer
association problems were frequent in study cohort (18.60%-14.0%), the variance was
statistically insignificant (p>0.05). The poor school performance stood significantly pronounced
in study cohort (p<0.0010). Each of these demographic alongside clinical variables is shown in
table 1 below:
Total (n=347) Normal (n=143) Obese (n=204) p
n (%) n (%) n (%)
Gender
Girl 208 (59.9) 85 (59.4) 123 (60.3) 0.650
Boy 139 (40.1) 58 (40.6) 81 (39.7)
Maternal age of birth
18-24 years 153 (44.1) 65 (45.5) 88 (43.1) 0.719
25-34 years 147 (42.4) 58 (40.6) 89 (43.6)
35-40 years 45 (13.0) 20 (14.0) 25 (12.3)
41 years and over 2 (0.6) 0 (0.0) 2 (1.0)
Mother education
Primary 187 (53.9) 80 (55.9) 107 (52.5) 0.791
High school 110 (31.7) 44 (30.8) 66 (32.4)
University 50 (14.4) 19 (13.3) 31 (15.2)
Father education
Primary 181 (52.2) 79 (55.2) 102 (50.0) 0.604
High school 106 (30.5) 40 (28.0) 66 (32.4)
University 60 (17.3) 24 (16.8) 36 (17.6)
Number of siblings
Zero 34 (9.8) 30 (21.0) 4 (2.0) 0.000*
One 123 (35.4) 56 (39.2) 67 (32.8)
Two 118 (34.0) 41 (28.7) 77 (37.7)
Three 49 (14.1) 12 (8.4) 37 (18.1)
Four 17 (4.9) 4 (2.8) 13 (6.4)
Five 5 (1.4) 0 (0.0) 5 (2.5)

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 10
Six 1 (0.3) 0 (0.0) 1 (0.5)
History of
Family’s psychiatric
disorder
33 (9.5) 13 (9.1) 20 (9.8) 0.824
Prenatal problem 31 (8.9) 2 (1.4) 29 (14.2) 0.000*
Postnatal problem 34 (9.8) 8 (5.6) 26 (12.7) 0.027*
Type of delivery
Vaginal 201 (57.9) 77 (53.8) 124 (60.8) 0.419
C/S 138 (39.8) 62 (43.4) 76 (37.3)
Forceps or vacuum 8 (2.3) 4 (2.8) 4 (2.0)
Preterm birth
No 331 (95.4) 137 (95.8) 194 (95.1) 0.941
Yes 13 (3.7) 5 (3.5) 8 (3.9)
Postterm birth 3 (0.9) 1 (0.7) 2 (1.0)
Peer relationship
Good 289 (83.3) 123 (86.0) 166 (81.4) 0.492
Few friends 44 (12.7) 17 (11.9) 27 (13.2)
Having friend problems 10 (2.9) 2 (1.4) 8 (3.9)
Cannot establish
relationship
4 (1.2) 1 (0.7) 3 (1.5)
School success
Bad 27 (7.8) 3 (2.1) 24 (11.8) 0.000*
Good 81 (23.3) 23 (16.1) 58 (28.4)
Perfect 239 (68.9) 117 (81.8) 122 (59.8)
Two groups did not differ significantly in developmental milestones (P=0.484, 0.587). Two
groups were not significantly different in terms of birth type and preterm-post term birth
variables (p=0.492) (table 1). However, the two groups differed in BMI (P=0.0320) with the
obese group having significantly higher BMI (mean=24kg/m2; SD=0.5kg/m2) than the none-
obese group (mean=1.2kg/m2; SD=0.1) (table 2).
Table 2. Analysis of continuous variables in two groups
Total (n=347) Normal
(n=143)
Obese
(n=204)
Mean (SD)
Median (min-
Mean (SD)
Median (min-
Mean (SD)
Median (min-
p

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 11
max) max) max)
Child age (months) 10.5 (2.5)
10 (7-17)
10.2 (2.4)
10 (7-17)
10.6 (2.7)
10 (7-17) 0.246
Education (years) 5.3 (2.7)
5 (1-11)
4.8 (2.1)
4 (2-11)
5.1 (2.2)
5 (1-11) 0.115
Mother age (years) 33.4 (4.2)
33 (27-42)
32.3 (4.8)
32 (25-48)
33.0 (4.3)
33 (23-42) 0.445
Father age (years) 36.6 (5.2)
35 (27-57)
36.0 (5.4)
34 (29-57)
37.0 (4.9)
36 (30-51) 0.405
Developmental
milestones
Walking (months) 12.1 (0.7)
12 (10-16)
12.1 (0.7)
12 (10-16)
12.1 (0.6)
12 (10-16) 0.484
Talking (months) 24.1 (2.6)
24 (16-36)
24.1 (2.7)
24 (16-36)
24.2 (2.5)
24 (16-36) 0.587
BMI (kg/m2) (SDS) 1.9 (0.7)
1.9 (1-4.1)
1.2 (0.1)
1.2 (1-1.5)
2.4 (0.5)
2.3 (1-4.1) 0.0320*
SD: standart deviation, min: minimum, max: maximum, SDS: standart deviation score
*: Fisher’s exact test
The obesity group had a higher percentage of children and adolescents with ASD as 1.5% of the
children and adolescent had ASD. However, the difference in the percentage of children and
adolescents with ASD in the obese group and in the control group was not statistically significant
(P>0.05). Nevertheless, there was a significant difference in SCQ positive and negative scores
(P=0.043), ABC total scores (P=0.144), and SCQ subscale scores in the sensory group (p=0.004)
between the control group and the obese group (figure 3). This may suggest that the prevalence
of obesity is higher (1.5 fold) in children and adolescents with ASD than those with normal
weight. It also suggest that children and adolescents with ASD are more likely (1.5 fold) than the
non-obese children to be obese than normal children. The significantly higher SCQ positive,
ABC total scores and SCQ subscale scores in the sensory group in the obese group may be
attributed to the 3 cases of ASD confirmed in the obese group.

AUTISTIC SYMPTOMS AND ADSD IN OBESE CHILDREN 12
Table 3. Analysis of the scales categorically in two groups
Total
(n=347)
Normal
(n=143)
Obese
(n=204)
n (%) n (%) n (%) p
Social Communication
Questionnaire
Negative 330 (95.1) 140 (97.9) 190 (93.1) 0.043*
Positive 17 (4.9) 3 (2.1) 14 (6.9)
Autism Behavior Checklist
Negative 340 (98.0) 142 (99.3) 198 (97.1) 0.144
Positive 7 (2.0) 1 (0.7) 6 (2.9)
Table 4. Analysis of the two groups in terms of mean scores of the scales
Total (n=347) Normal
(n=143)
Obese
(n=204)
Mean (SD)
Median (min-
max)
Mean (SD)
Median (min-
max)
Mean (SD)
Median (min-
max)
p
Social Communication
Questionnaire
Total 7.6 (4.0)
7 (0-21)
7.3 (3.7)
8 (0-20)
7.7 (4.3)
7.0 (0-21) 0.397
Reciprocal social interactions 2.7 (1.7)
3 (0-9)
2.5 (1.5)
2 (0-9)
2.8 (1.9)
3 (0-9) 0.951
Communication 3.5 (2.0)
3 (0-11)
3.5 (2.0)
3 (0-11)
2.8 (1.9)
3 (0-9) 0.228
Stereotypical behavior 1.3 (1.5)
1 (0-7)
0.3 (1.0)
1 (0-6)
0.6 (2.0)
1 (0-7) 0.437
Autism Behavior Checklist
Total 12.4 (8.7) 10.9 (6.9) 13.5 (9.7) 0.014*