http:// [Link].

ir
Original Article (Pages: 543-553)

Investigation on the Effectiveness of Holistic Multi-dimensional

Treatment Model (HMTM) in Improvement of CARS Test
Indicators in Children suffering from Autism Spectrum Disorder
Maryam Hojjati11, *Ghasem Sadeghi Bejestani23, Farah Ashrafzadeh3
1

Noor Hedayat, Center of Autism Spectrum Disorders, Mashhad, Iran.

Department of Biomedical Engineering, Imam Reza International University, Mashhad, Iran.
3
Department of Pediatrics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran.
2

Abstract
Introduction
The goal of research is investigation on the effectiveness of Holistic Multi-dimensional Treatment Model
(HMTM) in improvement of CARS test indicators in children with Autism Spectrum Disorder (ASD) and
it is done without pharmaceutical interventions, without names and accompanied by parental satisfaction
questionnaire.
Materials and Methods
In this study with purposeful sampling method, 60 autistic children chosen randomly among 6-10 yearold patients were taken one year observation. At the beginning of this research they were taken the CARS
pre-test and then in each month the complete behavioral HMTM checklist and the CARS test was taken.
Observing HMTM treatment model, includes intuitive checklist with the aim of intuitive adjustment,
abnormal behaviors checklist with the aim of decreasing them, individual behaviors checklist with the
aim of increasing them and emotional interactive behaviors checklist with the aim of improvement that
are conducted in statistical tables.
Results
Results shown a positive impact of HMTM in treatment trend of autism spectrum disorder considering
sum of CARS points, also the effect on each CARS test parameters and CARS parameters improvements
are meaningful holistic approach in HMTM method has interpreted some vague CARS parameters
furthermore (P<0.05). The statistical properties of CARS test with HMTM treatment model, indicates that
there was a high correlation coefficient and a significant relationship in behavior tables of HMTM and
CARS test.
Conclusion
Holistic multi-dimensional treatment model has caused considerable and improvement childrens
individual interactions and intuitive balance.
Key Words: Autism, ASD, CARS test, Children, HMTM.
*Corresponding Author:
Ghasem Sadeghi Bejestani, Department of Biomedical Engineering, Imam Reza International University, Mashhad,
Iran.
Email: [Link]@[Link]
Received date: Jan 5, 2015 ; Accepted date: March 22, 2015

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

543

The Effects of HTMT Model in Improvement of CARS Test Andicators

Introduction
Autistic spectrum disorders (ASD)
describe a range of conditions classified as
neurodevelopmental disorders, which are
known by deficit in communication, social
interactions and presence of repetitive
patterns in behavior, interests and
activities(1). These features often appear in
the childhood. Recent epidemiologic studies
have estimated the prevalence of autism
spectrum disorders as one in every 88
children(2). Despite extensive research,
there is still much controversy about the
morphological, functional and neurological
characteristics of people with autistic brains
(3, 4). However, the neural basis of behavior
changes in ASD has remained largely
unknown.
Autism is characterized with disorders in
three main areas of social interaction,
communication and behavior. Autism is a
kind
of
developmental
disorder
characterized by abnormal communication
and verbal behaviors. The symptoms of this
disorder, which are more common in boys
than in girls, appear in the first years of life,
and the main cause is unknown.
In 1940, two scientists, an Austrian
American psychiatrist and physician, Leo
Connor,
American
psychiatrist
and
physicians, and an Austrian pediatrician,
medical theorist, and medical professor,
Hans Asperger, independently described a
new neurodevelopmental disorder and
incidentally both called the disorder with a
similar name as autism syndrome (derived
from the Greek word 'autos' meaning self).
Presence and contact with mobile phones
and other devices relying on wireless
technology due to creation of magnetic
fields density - air pollution, genetic factors
and maternal mental conditions during
pregnancy are as causes of autism. The
prevalence of autism spectrum disorders has

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

been much more reported in the developed

countries. This remarkable increase in the
prevalence of the disorder is attributed to the
changes made in methods of detection and
reporting rather than to be a real biological
increase (5).
There are two ways to study a system: Top down and bottom-up approaches (6). In the
bottom-up approach, called as componentoriented method, it is believed that a system
(here, the child) is made up of components
that the study of these components will
result in the recognition and understanding
the whole. Today, many treatment methods
are based on this idea. In the top-down
method, which is called holism, it is
supposed that the whole is more than just the
sum of components, and to recognize a
system,
the
interactions
between
components must also be considered in
addition to the components themselves. The
HMTM
(Holistic
multidimensional
treatment model) method is based on this
idea (7).
Materials and Methods
Research Tools
The diagnostic tools in this study
specialized diagnosis by a Child and
Adolescent Psychiatrist and a diagnosis by
the Evaluation Organization of Special
Education. Also, the CARS test was used as
pre-test and post-test at the beginning and
end of each month to determine the severity
of the disease. To this end, each child has
two test scores per month, and the rate of
change in the relevant items is the scores
difference that the child is given at the end
of each year. The research tools include the
following:
Multidimensional Holistic Treatment Model
(MHTM)

544

Hojjati et al.

This attitude has been formed through longterm direct experience with children with
pervasive developmental disorder, and
emphasizes on 6 axes:
1. Holistic view of the child;
2. Unconditional acceptance of the
child and making a deep emotional
connection between caregiver and
child;
3. Physiology;
4. Movement & Perception;
5. Speech & Expression;
6. Affective,
emotional,
communicational and play aspects.

Fig.1: HMTM model

In this method (Figure.1), for evaluating the

behavioral characteristics of the child, the
child is examined and assessed in five areas
of speech status, emotional and affective or
engagement behaviors, senses, abnormal
behaviors and individual behaviors.
According to the test procedure, the relevant
scores for the characteristics related to these
five areas are recorded in each month, and
finally, some graphs are plotted that can help
to assess the treatment process. These charts
that are given in the following were
provided for 60 children with autism in four
seasons of 2014; the patients' scores were
extracted and used in the study (Fig 2).
CARS Test

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

This diagnostic and screening test for autism

was developed and provided by Schopler in
1988 and has 15 items as follows:
1.
2.
3.
4.
5.
6.
7.
8.
9.

Communication with people;

Imitation;
Bodily movements;
Emotional response;
Application of objects;
Adapt to changes;
Visual response;
Auditory response;
Response and use of taste, smell, and
touch senses;
10. Fear or nervousness;
11. Verbal communication;
12. Non-verbal communication;
13. Function or activity level;
14. Level and stability of rational
responses;
15. Overall impression (8-10).
Each item is scored in a seven-point grading
scale. Low scores indicate the lower
intensity of autism symptoms, while high
scores indicate the severity of autism. Also,
the scores sum on this scale indicates an
overall score about the severity of autism
spectrum disorders (Review of psychometric
indices of autism diagnostic test). Each item
includes the options of "No problem",
"Little abnormal", "Moderate" and "Severe",
which were given in order 1, 2, 3 and 4
scores. The total test score will be between
15 and 60 that regarding scores above 30,
the diagnosis would be autism. The scores
between 30 and to 36 indicate low to
moderate autism, while scores between 36
and 60 reflect the severe autism (10).
Data analysis tools
Statistical methods of P-value and
correlation coefficient were used to
determine the psychometric indices of
HMTM test such as significance level and
reliability. Pearson product moment

545

The Effects of HTMT Model in Improvement of CARS Test Andicators

correlation coefficient (r) is a method

typically used to connect two quantitative
variables. To calculate r, there must be at
least two values (sizes) for each subject. It is
often assumed that these two sizes should be
at a distance level. The correlation
coefficient can range from +1 to 0 and from
0 to -1. The value of +1 indicates a complete
or positive correlation, and -1 indicates the
complete and negative correlation.

Fig 2: An example of HMTM behavioral checklist

A correlation coefficient of zero indicates

that there is no correlation between the two
variables. It must be however understood
that there is rarely a full correlation between
the two variables in the behavioral
sciences(11, 12).

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

The correlation coefficient also describes the

type of relationship, meaning a being
negative or positive. If a negative value
means when one variable increases, the
other decreases. If the value is positive, it
means that both variables are likely to
increase or decrease (12).
Relationship or difference between variables
can be examined through probability. The
relationship or difference between two
variables may be random. Therefore, the
relationships or differences between
variables should be examined according to
the significant level. The minimum
acceptable level is 95% probability. This
means if a relationship or difference is seen
between two variables and statistical
calculations show that the relationship or
difference is due to chance by 5%
probability, and by 95%, the relationship or
difference is non-randomized and arising
from the impact of a trial practice, the

546

Hojjati et al.

relationship or difference would be

considered significant; or if the p-value is
small, and represents a value of less than
0.05(12).

Results
At first, the CARS data was divided As
follows:
Number 1: Indicates the absence of
disease;
Numbers between 1 and 2 indicate
weak abnormality;
Numbers between 2 and 3 indicate
moderate abnormality;
Numbers between 3 and 4 indicate
severe abnormality.
In fact, by reduced CARS score in March
2014, one can say that improvement has
occurred. Then, only the fear or nervousness
features and the CARS overall scores were
studied (Table 1).

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

Variables
Fear or
nervousness

The standard deviation, as the most

important dispersion index, was also used to
demonstrate the scattering rate of data.
Reporting the average value, without
mentioning the variability or dispersion
indices, will lead to misinterpretation of the
data. The researchers tend to focus on
measures of central tendency and neglect of
their distribution pattern, while they are also
equally important. The average of two data
series may be the same, but their distribution
would be different. If the distribution scores
are similar, they would be called
homogeneous (with low variability), and if
the scores are not the same, they would be
called heterogeneous (with high variability).
Dispersion measures show the distribution
rate or changes between the scores of a
distribution(13, 14). Bar, pie and box graphs
were also used to investigate the treatment
trend.

Table 1: Comparison of fear or nervousness in

the pre-test and post-test

Absence
Weak
Moderate
Severe
Total

Frequency
May
0
13
31
16

March
0
20
30
10
60

Percent
May
0
21.6
51.6
26.6

March
0
33.3
50
16.6
100

In Table 1, characteristics of fear or

nervousness in May and March 2014
(pretest and posttest) were compared with
each other. In (Fig 3), for each month, the
frequency and frequency percentage was
calculated, and the following bar graph was
plotted.

Fig 3: Comparison of frequency percentage

distribution of characteristics of fear or nervousness
in A: May and B: March

It can be concluded that in May, most of the

children referred were moderately abnormal
in this feature. Given that the decline in the
children CARS scores in March compared to
May indicates an improvement, it can be
concluded regarding the characteristics of
fear or nervousness, only the scores of
children with moderate and severe abnormal
disease levels have reduced, and transferred
to the weak abnormal group. In contrast, the
number of children with weak abnormal
condition has increased. In the following,
the box diagram of fear or nervousness
changes were examined at the beginning and
end of the study.

547

The Effects of HTMT Model in Improvement of CARS Test Andicators

According to (Fig 5), one can conclude that

the scattering of data in May is more than in
March, and in March, the children CARS
scores have decreased, showing an
improvement in patients. Also in March, due
to occurrence of mean in the middle of the
rectangle, data distribution is more
symmetrical than in May. Due to the lowed
maximum in March compared to May, we
can conclude that the CARS overall score of
children has reduced, which represents the
improvement in their condition.
Standard Deviation (SD)

Fig 4: Box graphs for fear or nervousness feature: A.

May, 2014; B. March, 2014

In (Fig 4), the box diagram shows that the

distribution of data is symmetric. Given that
the length of the rectangles in both months is
the same, the scattering of data is the same
as well. The lower mean in March indicates
that 50% of data are less than 2.5,
suggesting progress in the treatment trend.

The aim is to determine the distribution of

data in each of the characteristics. Given that
lower and closer to zero standard deviation
value represents the higher congruence and
similarity in the studied feature among the
people, based on standard deviation results,
we can say that the studied children with
autism are similar in all CARS
characteristics. However, the greatest
similarity in March and May was seen in the
level of activity, respectively as 0.6731 and
0.5785.
Correlation coefficient (r)

Fig 5: Box plots for the sum of

CARS: A. May; B. March 2014

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

As mentioned above, one of the research

objectives was to investigate the correlation
between behavioral characteristics in
HMTM and CARS tests in order to assess
the HMTM method as a diagnostic test and a
therapeutic method. To this end, the features
of these two tests that would seem to be
clinically relevant were studied and
determined by Noor Autism Center experts.
The results obtained include the followings:
With larger sample size, one can expect to
obtain higher correlation coefficient, and the
samples size clearly related to the value of
correlation coefficient. Also, usually with
higher correlation coefficient, a smaller pvalue is obtained, which means the
increased chance of an actual relationship

548

Hojjati et al.

between the two variables, here the features

of the two tests. Also, there is a lower
probability that the correlation coefficient
obtained would be due to chance. For
example, one can say that if p-value is equal
to 0.05, the analysis suggests by 95%
probability that the correlation coefficient
obtained represents the relationship between
the two variables, and there is only a 5%
chance that the value is obtained by chance
and coincidence. Thus, the smaller the pvalue, the more reliable is the correlation
coefficient.
It is also clear that most of the features that
seem to be clinically relevant are really
connected. Since, as dealing with human and
study of human behaviors, even correlation
coefficients
greater
than
0.5
are
considerable. But in some cases, the features
were not associated much that can be due to
reasons such as errors in experimental and
clinical techniques, low volume samples and
the unavailability of complete scores for all
month for everyone.
It should be noted that since the purpose of
CARS test was reducing the score, while the
aim of HMTM test in these features was
increasing the behaviors, obtained negative
values for the correlation would be
reasonable.
Among the features surveyed, the following
characteristics in CARS HMTM tests
showed the highest correlation, i.e., greater
than 0.7. It should be in these cases, the size
of samples was more suitable.
1. Characteristic
of
verbal
communication in CARS test with
features of phonation, speech
imitation and comprehension of onestep sentences in HMTM test.

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

2. Visual response feature in CARS test

with appropriate eye contact feature
in HMTM test.
3. Characteristic of using objects in
CARS test and sensorimotor games
in HMTM test.
If we consider the correlation coefficients
greater than 0.5 adequate and reliable, the
following results can be found regarding the
relationship between the characteristics of
these two tests:
Every feature of CARS is associated with
several features of HMTM. This means
given the complexity and magnitude of
autism spectrum disorders, the patients
should be evaluated more precisely with
more details for diagnosis and assessment of
treatment process. For example, the
characteristic of communication with the
people examined in CARS test contains
subsets that are important and raise various
aspects of this type of communication. It can
be concluded that HMTM test, which
examines
each
of
the
patient's
communication and behaviors separately in
details, is more suitable for diagnosis and
assessment of disease treatment process.
Some of CARS features having many
relationships that with some features of
HMTM according to the correlation
coefficients obtained are specified in (Table
2).
In fact, these tables show the
connections that have been proven
quantitatively.
In Table 3, another study of correlation
coefficients between HMTM features, which
were typically calculated in a chart and the
features had high significant correlation
coefficients
are
given.

549

The Effects of HTMT Model in Improvement of CARS Test Andicators

Conclusion
The data were related to children who
were treated for one year by HMTM method
that their CARS and HMTM tests were
separately evaluated and then compared
with each other.
We can say the children have progressed in
many of the characteristics, and their
conditions have changed from sever
abnormal condition to moderate and even
mild abnormal conditions. These findings
indicate the effectiveness of the treatment
method as behavioral therapy and based on
special behavioral check lists. Another
considerable point is the high relationship
and correlation between CARS and HMTM
features, which due to the complexity of
autism spectrum disorders, may reflect the
insufficiency of CARS features to examine
and diagnose the children with autism. Each
feature has subsets, and in addition to a
holistic view of the children, their
behavioral communicational characteristics
should be examined in more details and
more accurately. This can be effective in
diagnosing the disease and assessment the
treatment process. In this regard, HMTM
can be more helpful than CARS for more
accurate interpretation. In other words, the
holistic approach in HMTM method could
well interpret some ambiguous indicators in
CARS
method.
In
addition,
the
implementation of this model resulted in
rapid and dramatic progress of children in
improvement of individual transactions and
sensory balances. Given the breadth and
complexity of autism spectrum disorders, by
increasing the sample size and receiving a
more complete data set, other various
aspects of the disease such as the impact of
age, gender, environment and family and
community connections on the current
findings as well as analyzing the relationship
or difference of brain signals of patients

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

together and with healthy children can be

studied.
Conflict of interest: None.
Acknowledgment
We want to thank all of dedicated
employees and authorities in Noore Hedayat
Therapy Center, Mashhad- Northeastern
Iran, for their full cooperation in doing this
research.
References
1. Association AP. Diagnostic and statistical
manual of mental disorders, (DSM-5):
American Psychiatric Pub; 2013.
2. Baio J. Prevalence of Autism Spectrum
Disorders: Autism and Developmental
Disabilities Monitoring Network, 14 Sites,
United States, 2008. Morbidity and Mortality
Weekly Report. Surveillance Summaries
2012; 61 (3): 1-19.
3. Calderoni S, Retico A, Biagi L, Tancredi R,
Muratori F, Tosetti M. Female children with
autism spectrum disorder: an insight from
mass-univariate and pattern classification
analyses. Neuroimage 2012;59(2):1013-22.
4. Schipul SE, Keller TA, Just MA.
Interregional brain communication and its
disturbance in autism. Front Syst Neurosci
2011;5:10.
5. Verhoeven JS, De Cock P, Lagae L, Sunaert
S. Neuroimaging of autism. Neuroradiology
2010;52(1):3-14.
6. Von Bertalanffy L. General system theory:
Foundations, development, applications:
George Braziller New York; 1968.
7. Mashhadi A, Soltanifar A, Hojati M, Reebye
P, Hosainzadeh- Maleki ZMA. Comparison
of
the
Effectiveness
of
Holistic
Multidimensional
Treatment
Model
(HMTM) and Applied Behavioral Analysis
Approach (ABA) in Treatment of Children

550

Hojjati et al.

with ASD. Journal of Clinical Psychology

2010;2(2): 27-35.
8. Mayes SD, Calhoun SL, Murray MJ, Pearl A,
Black A, Tierney CD. Final DSM-5 underidentifies mild Autism Spectrum Disorder:
Agreement between the DSM-5, CARS,
CASD, and clinical diagnoses. Research in
Autism Spectrum Disorders 2014;8(2):68-73.
9. Mayes SD, Black A, Tierney CD. DSM-5
under-identifies
PDDNOS:
Diagnostic
agreement between the DSM-5, DSM-IV,
and Checklist for Autism Spectrum Disorder.
Research in Autism Spectrum Disorders
2013;7(2):298-306.
10. Sturmey P, Dalfern S. The Effects of DSM5
Autism Diagnostic Criteria on Number of
Individuals
Diagnosed
with
Autism
Spectrum Disorders: A Systematic Review.
Review
Journal
of
Autism
and
Developmental Disorders 2014;1(4):249-52.

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

11. ZHOU Z-h, BIN X-n, LEI Y-x. The

Application of SPSS in Teaching of Medical
Statistics [J]. Practical Preventive Medicine
2004;5:101.
12. Wang J-T, Jiang E-H, Wang X-G, Zhao X-R,
Yao L. Application of SPSS technique in
demonstration study data for statistics and
analysis.
Jisuanji
Gongcheng
yu
[Link]
Engineering
and
Applications 2006;42(36):201-3.
13. Ling DZYPC. Application of SPSS Software
in Orthogonal Design and Result Analysis
[J]. Computer Study 2009;5:15-7.
14. Stakenas JA, Merrick NP. Application of
SPSS to management of computer-assisted
instruction usage statistics. Bulletin of the
Medical Library Association 1982;70(2):231.

551

The Effects of HTMT Model in Improvement of CARS Test Andicators

Table 2: Correlated characteristics of HMTM and CARS based on quantitative data

CARS

Communication with
the people

Emotional and
social interaction
Social skills,
communication

HMTM
Good eye contact
laugh
Emotional states face

r
-0.778
-0.561
-0.541

Frustration tolerance

-0.517

Appropriate social response

-0.513

Voice

-0.645

Imitation behavior

-0.534

Reticence

-0.521

Understanding sentences, one-stage

-0.689

Understanding sentences, two-stages

-0.570

Observing the shift

-0.594

Emotional states face

-0.561

Frustration tolerance

-0.578

Games sensorimotor

-0.767

Game Performance

-0.613

Frustration tolerance

-0.660

Voice

-0.754

0.000

Imitation speech

-0.728

0.000

Proper use of the pronoun

-0.564

0.000

The ability to define the pictures

-0.575

0.000

Recitation with no picture

Speech clarity
Tribute speech chain and joined
together
Reticence
Understanding sentences, one-stage

-0.500
-0.609

0.003
0.000

-0.534

0.000

-0.556
-0.724

0.000
0.000

Understanding sentences, two-stages

-0.632

0.000

Understanding sentences, one-stage

-0.694

0.000

Appropriate social response

-0.667

0.000

Voice

-0.581

0.000

Understanding sentences, one-stage

-0.634

0.000

Language

Imitation
Speech perception

Emotional response

The use of objects

Adapt to change

Emotional and
social interaction

Games
Emotional and
social interaction

Language

Verbal communication

Speech perception

The level and stability of

rational responses
Non-verbal communication

Speech perception
Social skills,
communication
Language
Speech perception

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

P-value
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.0000
0.000
0.000
0.000
0.000

552

Hojjati et al.

Table 3: Display of correlation coefficient between HMTM items with comparing the means
HMTM

Cases

53

0.645

29

0.540

26

0.571

29

0.767

40

0.722

38

0.602

33

0.614

30

0.577

29

0.557

28

0.670

39

0.810

Understanding sentences, one-stage

33

0.651

Understanding sentences, two-stages

32

0.712

Voice
Voice
Proper use of the pronoun
Speech clarity
Speech Clarity
Speech clarity
Speech clarity

Tribute speech chain and joined together

Tribute speech chain and joined together
Tribute speech chain and joined together
Understanding sentences, one-stage

Int J Pediatr, Vol.3, N.2-2, Serial No.16, April 2015

P-value
0.000

HMTM
Imitation behavior

0.002

Proper use of the pronoun

0.002

The ability to define the

pictures
Tribute speech chain
and joined together
Understanding
sentences, one-stage
Understanding
sentences, two-stages
Observing the shift

0.000
0.000
0.000
0.000
0.000
0.001
0.000

Understanding
sentences, one-stage
Understanding
sentences, two-stages
Observing the shift

0.000

Understanding
sentences, two-stages
Observing the shift

0.000

Observing the shift

0.000

553