attention deficit disorder with hyperactivity, child - adolescents
transtorno do déficit de atenção com hiperatividade, criança - adolescente
Attention-deficit/hyperactivity disorder (ADHD) is a neurobehavioral disorder characterized
by the combination of attention deficit, hyperactivity and impulsivity[1 ].
The diagnosis is fundamentally clinical and is based on criteria from classification
systems such as the Diagnostic and Statistical Manual of Mental Health Disorder, Fourth
Edition, 2000 (DSM-IV), effective at the beginning of this research[2 ]. Attention-deficit/hyperactivity disorder is classified into three types: (a) predominance
of hyperactivity-impulsivity; (b) predominance of attention-deficit; and (c) a combination
of both, which can be associated with comorbidities[1 ].
The hyperactive-impulsive individual presents with greater impairment in family/social
life, the inattentive child has greater impairment in school performance, and the
child affected by the combination has impairment both in family/social life and education[2 ]. The prevalence of ADHD is 5% to 10% in children and is more common in boys than
in girls, at a ratio of up to 3:1[3 ].
The mechanisms by which ADHD occurs are still not fully understood, but genetics have
a very important contribution[1 ],[4 ] in the dysfunction of brain dopaminergic and noradrenergic activities that lead
to inadequate control of the ventral prefrontal cortex areas over the lower structures,
related to automated responses, damaging cortical control of responses to certain
stimulae[5 ],[6 ]. Therefore, uncontrolled emotions may influence the decision-making process, interfering
in cognitive processes[7 ].
The decrease in dopamine and noradrenaline, and the late maturation of brain circuits
associated with the prefrontal cortex lead to a decrease in information processing
speed and to a lower self-control capability[8 ],[9 ]. The impairment in the functioning of the ventral anterior cingulate cortex is directly
associated with the control of primitive responses and impulsive emotional expressions[10 ]. In a normal state, the action of the prefrontal cortex over the subcortical areas
leads to an inhibitory influence on the amygdala and prevents undesirable automatic
reactions[11 ], and also preserves and restores attention, memory and emotional control (executive
functions)[12 ]. Therefore, a broad range of brain control processes connect, prioritize and integrate
functions that are needed for self-control.
When some emotions are overtly expressed, they can be of a negative nature, such as:
irritability, emotional lability, sadness, dysphoria and crying; or of a positive
nature: friendship, joy, happiness and spontaneity[13 ]. There needs to be monitoring of emotional expression during medical monitoring
or drug treatment that will act on the affected brain areas improving brain circuits[13 ].
Emotional expression refers to the ability to moderate negative emotions and potentiate
positive emotions in a balanced way without repressing or exaggerating the information
they convey[14 ].
It has been noted that “dysphoric” emotional effects arise during medication treatment
and, therefore, a parent-reported scale was developed to measure the “negative emotional
expression,” including affective blunting, mood lability, and the “zombie effect”.
The Expression and Emotion Scale for Children (EESC) aims to quantify the impact of
medication on emotional expression in children[15 ].
Kratochvil et al.[15 ] developed the EESC, taking into consideration the symptoms presented by children
with ADHD. This scale assesses the intensity and character of the emotions expressed
by the child or adolescent with ADHD. It considers the child’s mood and emotional
state (irritability, poor concentration, isolation, crying, liveliness, spontaneity,
sympathy, maturity, etc.) over the previous two weeks.
This scale has 29 questions divided into three domains: positive, with 13 questions;
negative, with 10 questions; and lability, with 5 questions; question 19 was not considered
for the statistical calculation. The items are quantified according the Likert scale:
1 – none; 2 – a little; 3 – reasonable; 4 – very much; 5 – totally. A high level of
emotional control is indicated by a lower score. The maximum and minimum values in
the positive, negative and lability domains are, respectively: 65-13, 50-10, 25-5;
noting that the positive domain is inverted for the sum of the overall score[15 ].
Considering the impairments in the emotional expression of individuals with ADHD caused
by neuroanatomical, chemical, and physiological conditions, the purpose of this study
was to validate, for the Portuguese language, the EESC for the assessment of emotional
expression control by patients with ADHD between six and 15 years old, who used stimulants
or were simply being monitored medically (see [Appendix ]).
METHODS
This cross-sectional study was approved by the Ethics Committee for Analysis of Research
Projects (CAPPesq) of the Hospital Clinical Board with the number 0613/11. For the
validation of the EESC patient version, the study children were all recruited from
the Learning Disabilities Clinic of the Children’s Institute of the University of
São Paulo Medical School.
Patients
A total of 126 patients with ADHD were assessed and diagnosed according to the Swanson,
Nolan and Pelham Teacher and Parent Rating Scale[16 ] following the DSM-IV criteria (DSM V had not yet been edited when the research began);
of these patients, 91% were being treated with a stimulant and 9% were medication-free.
The average age of the ADHD group was 11.24 years old and that of the control group
(n = 126) was 10.51 years old. Of the 126 patients with ADHD, 48 had mild comorbidities
(40%) and 78 had no comorbidities (60%). Data on comorbidities had been previously
collected and registered in the medical records. The assessment included the Child
Behavior Checklist[17 ]. The comorbidities identified in the ADHD group were: learning disorder 20 (15%);
depression 12 (10%); oppositional defiant disorder 7 (6%); conduct disorder 6 (5%);
and bipolar disorder 5 (4%). The characteristics of the patients’ education, socioeconomic
profile and hobbies are shown in [Table 1 ].
Table 1
Clinical characterization of the patients with and without ADHD.
Variable
ADHD
Control
Male gender
83%
52%
Female gender
17%
48%
Slight comorbidity
40%
-
Public school
74%
100%
Electronic games
47%
56%
Years of parent’s education
9.4
11.7
ADHD: attention-deficit/hyperactivity disorder.
Of the 126 patients, it was selected age groups for stablish the cutoff point, at
a total 85 subjects, because they were paired according to number, age and gender:
Group I – six to eight years old, n = 13 (17.46%); Group II – 9 to 11 years old, n
= 40 (46.82%); and Group III – 12 to 15 years old, n = 32 (35.71%).
The study excluded patients under six and over 15 years old; those with a cognitive
deficit; with an intelligence quotient lower than 80 (assessed by a neuropsychologist);
and those who did not have diligent medical care, or who had interrupted clinical
supervision. The control group also had exclusions, which included individuals under
six and over 15 years of age; those with a cognitive deficit; those with an intelligence
quotient lower than 80 (assessed by a neuropsychologist); and those diagnosed with
any psychopathology (assessed by a neuropsychologist).
The tests used to assess the control group were: the School Achievement Test[18 ]; the Swanson, Nolan and Pelham-IV Teacher and Parent Rating Scale; and the Child
Behavior Checklist.
Protocol
Between February and December of 2012, a validation of the parent-rated EESC was carried
out. The version was applied to 33% of the individuals with intra-observer and inter-observer
variations. The average duration of the application was 12 minutes. It was applied
directly to ADHD group parents and indirectly to control group parents. The members
of the latter group were parents of students from the Escola Municipal de Ensino Fundamental
Presidente Professor João Pinheiro (Vila Matilde, São Paulo, SP) with previous authorization
from the school principal.
Initially, a certified translation of the tool was carried out. Then, after the choosing
a native speaker for the translation, the document was culturally and semantically
adapted. The validation followed the steps proposed by Guillemin et al.[19 ]: 1) authorization from the authors; 2) initial translation into Portuguese; 3) cultural,
conceptual, experimental and idiomatic adaptation to the target population; 4) retroversion;
5) assessment by a revision committee; and 6) pre-test in two stages.
Description of the validation steps
The process of validation of the EESC questionnaire translated into Portuguese had
previous authorization (1st step) by Kratochvil et al[15 ]. It was translated by an English language native professional translator and by
a certified translator (2nd step). After the comparison of both translations, it was observed that the translation
made by the English language native translator was non-literal and, therefore, culturally
more appropriate, closer to colloquial language and more effective for the process
of linguistic adaptation than the certified translation.
Between the steps of translation and retroversion, the Probe technique[20 ] was included to assess only the comprehension of the items of the EESC by the patients’
parents. The understanding scale of 1 to 5 was: (1) no understanding, (2) little understanding,
(3) average comprehension, (4) understanding almost total, and (5) total understanding.
The comprehension level reached was above 85%.
In order to reach satisfactory comprehension, the Probe technique was applied to three
groups of 20 patients’ parents each in three phases. During these applications for
the improvement of the instrument, a cultural adaptation was made using the following
criteria: semantic, idiomatic, conceptual and experimental equivalences (3rd step).
When the ideal comprehension level was reached, the retroversion step began (4th step). The Portuguese version, resulting from the Probe technique, was retranslated
into English by an English language native translator and compared to the original.
The analysis of this retranslation enabled the verification of linguistic equivalence.
After this step, the version resulting from the Probe technique was assessed by the
revision committee (comprising two neuropediatricians, one pedagogue and one phonoaudiologist),
which suggested some changes (5th step).
The 6th step was divided into two periods: the reapplication of the Probe technique in order
to achieve the satisfactory comprehension level, applied to 41 individuals after the
changes suggested by the revision committee and, after this, the questionnaire was
applied to one parent of 126 patients for the assessment of the child’s emotional
expression, and the results were submitted to statistical calculation for psychometric
evaluation.
The evaluation of the psychometric quality of the EESC was carried out according to
the following distribution: internal consistency (reliability of questions), n = 126;
external consistency (inter-observer reliability), n = 35; external consistency (test-retest
reproducibility – twice in the period of three months), n = 24; sensitivity, specificity,
accuracy and over all cutoff point, n = 126.
Statistics
The calculation of internal consistency, item reliability, was made using the Cronbach’s
Alpha Coefficient test and was classified as: very low (alpha ≤ 0.30), low (alpha
between 0.30 and 0.60), moderate (alpha between 0.60 and 0.75), high (alpha from 0.75
to 0.90) and very high (alpha above 0.90), with alpha values above 0.70 being considered
satisfactory[21 ]. This same classification was applied to the other statistical tests used, as shown
below. Statistical significance was determined at p < 0.05.
For the calculation of external consistency, the reliability analysis was made through
Pearson’s Linear Correlation[22 ] and reproducibility was analyzed using the Intraclass Correlation Coefficient test
(ICC)[23 ],[24 ].
The Receiver Operating Characteristic Curve (ROC) is used for the assessment of sensitivity
and specificity of the instrument and to indicate optimal cutoff and accuracy. The
most elevated point of a curve, corresponding to the upper left angle of the graph,
represents 100% of sensitivity and 0% of false positives; in this case, the ideal
value of the diagnostic test is the gold standard (d = 0). Line ‘d’ in the curve indicates
proximity to the axis of abscissas, which represents the gold standard. When d = 0.20,
the distance to the gold standard is low, the distance of d = 0.50 is considered moderate,
the distance of d = 0.80 is considered high[25 ].
The ROC is a method for assessment, organization and selection of diagnostic and/or
prediction systems. It is commonly used in medicine to analyze the quality of a given
clinical test[26 ],[27 ]. The tools used were the SPSS Statistics version 2.2 for Cronbach’s Alpha and the
BioStat version 5.3 for the Pearson’s, ICC, and ROC tests. A series of tests was adopted
as the gold standard to guarantee the validity of the tool[28 ].
RESULTS
The level of comprehension reached through the Probe technique was 96%, which is above
the required minimum of 85%. The average of the overall score for patients and for
the controls was of 60.21 ± 14.02 for patients and 46.96 ± 10.61 for controls; group
I, ADHD patients 48.69 ± 9.3 and controls 47.69 ± 10.97; group II, ADHD patients 57.15
± 11.67 and controls 46.52 ± 13.42; group III, ADHD patients 61. 37 ± 14.13 and controls
51.19 ± 12.97. The averages and standard deviations of the domains can be seen in
[Table 2 ].
Table 2
Mean and standard deviation (SD) of patients with and without ADHD.
Variable
Positive domain
Negative domain
Lability domain
Overall score
ADHD vs control
ADHD mean - SD
24.69 ± 6.91
21.59 ± 7.41
14.38 ± 4.06
60.21 ± 14.02
Control mean - SD
21.51 ± 7.16
14.73 ± 4.84
14.381 ± 4.63
46.96 ± 10.61
Group I vs control
ADHD mean - SD
21.92 ± 8.1
16.38 ± 4.42
12.92 ± 3.82
48.69 ± 9.3
Control mean - SD
20 ± 4.45
14.54 ± 3.81
13 ± 4.06
47.69 ± 10.97
Group II vs control
ADHD mean - SD
22.57 ± 4.35
20.02 ± 6.76
15.1 ± 4.22
57.15 ± 11.67
Control mean - SD
20.57 ± 6.92
13.47 ± 3.96
11.7 ± 4.38
46.52 ± 13.42
Group III vs control
ADHD mean - SD
25.34 ± 6.40
22.31 ± 7.56
13.75 ± 3.8
61. 37 ± 14.13
Control mean - SD
22.84 ± 7.34
16.06 ± 5.78
12.62 ± 5.29
51.19 ± 12.97
ADHD: attention-deficit/hyperactivity disorder.
The internal consistency through Cronbach’s Alpha Test in the reliability assessment
of the items in the instrument showed an overall score of α = 0.76 and p < 0.001.
This score, and that of other domains, where the values guarantee psychometric reliability
of the items, are shown in [Table 3 ].
Table 3
Values of Cronbach’s alfa, Pearson’s linear correlation coefficient, intraclass correlation
(ICC) of the domains of the EESC questionnaire.
Variable
n
Overall score
Positive
Negative
Labile
p-significance
Cronbach’s alpha
13
0.76
0.76
0.75
0.74
p < 0.001
Pearson’s r
35
0.91
0.95
0.97
0.91
p < 0.001
ICC
24
0.66
0.78
0.70
0.76
p < 0.001
EESC: expression and emotion scale for children.
External consistency – reliability – through Pearson’s linear correlation test was
r = 0.91 and p < 0.001 in the overall score. The confidence interval of 95% between
the observer and the inter-observer in all domains and p < 0.0001 are shown in [Table 3 ]. For external consistency – reproducibility – the patients were assessed in twice,
with an interval of six months, through the ICC test, reaching the value of 0.66;
values of the domains are in [Table 3 ].
The results of the ROC Curve of the EESC for the overall score in all age groups were:
sensitivity = 0.75, specificity = 0.67, accuracy = 71% and cutoff = 51. The remaining
results of the different domains by age group are shown in [Table 4 ] and the [Figure ].
Table 4
Values of sensibility, specificity, accuracy and cutoff of the domains of the EESC
questionnaire.
Variable
Group A
Group B
Group C
All groups
Specificity
Positive domain
0.39
0.53
0.44
0.69
Negative domain
0.62
0.8
0.6
0.69
Labile domain
0.54
0.65
0.5
0.69
Overall score
0.69
0.7
0.53
0.67
Sensibility
Positive domain
0.85
0.78
0.72
0.60
Negative domain
0.69
0.88
0.81
0.79
Labile domain
0.46
0.68
0.69
0.53
Overall score
0.54
0.7
0.84
0.75
Accuracy (%)
Positive domain
62
65
58
65
Negative domain
65
84
70
74
Labile domain
50
66
59
57
Overall score
62
70
69
71
Cut off
Positive domain
17
20
22
23
Negative domain
15
15
17
16
Labile domain
14
13
12
14
Overall score
52
51
52
51
EESC: expression and emotion scale for children
Figure The ROC Curve of the Overall Score of Children with and without ADHD.
DISCUSSION
The EESC is capable of being self-administered and is easy to fill in. It requires
12 minutes to be completed. It is a recommended instrument for clinical supervision
in the quantification of the emotional expression of the patient with ADHD in drug
treatment to assess the oscillation of the emotions before the impact of medication[13 ],[15 ],[29 ].
A less literal translation was used as a strategy to improve comprehension[30 ]. The Probe technique was used to measure the level of comprehension; this enabled
a better cultural, idiomatic and semantic adaptation. This was necessary due to the
fact that words and expressions carry values and meanings inherent to a culture and
they must be incorporated into the semantics of the instrument, with the purpose of
expressing what the questionnaire intends to show while remaining true to the original
version[30 ]. After the application of the Probe technique, a satisfactory level of comprehension
was achieved to enable the collection of data and submission to statistical analysis
for the assessment of the psychometric quality of the instrument.
Internal consistency in each domain was moderate to high (α = 0.76). External consistency
indicated a strong correlation, guaranteeing reliability (r = 0.95). Test-retest reproducibility
was considered satisfactory (ICC = 0.76), qualified as reproducible according to the
criteria established by Shrout and Fleiss[23 ]. The results obtained in this validation are in accordance with those of the original
parent-rated EESC[29 ]. The present validation also measured sensitivity (0.75), specificity (0.67), accuracy
(71%) and cutoff (51) through the ROC Curve test. The sensitivity and the specificity
were significant, since they remained above the reference line[25 ]. The level of accuracy confirmed that the instrument adequately assessed that which
it aimed to. The validity and efficacy follow the measures of the gold standard, since
they are above the diagonal line ([Figure ])[25 ],[29 ].
By means of the application of the instrument to the age groups, it was possible to
establish a cutoff value related to age and domain. Depending on the age group the
individual is part of, there is variation in the way they perceive their surroundings,
which points to a higher or lower control over emotional expression and over quality
of life[28 ],[31 ]. Through the standardization of the cutoff regarding the age group, it was possible
to assess, with greater precision, the patient’s emotional expression, considering
that, in this study, values lower than 51 (overall score) indicated a better emotional
expression and, consequently, increased quality of life[28 ],[31 ].
The difficulty faced by children and adolescents with ADHD regarding emotional management
is notorious[12 ],[31 ]. The measurement of the level and quality (positive, negative or labile domains)
of emotional expression, which is part of their symptomatic state, must be part of
daily therapeutic practice to establish a parameter for the results of medication
on the patient’s life[12 ]. There are studies that confirm the need for the application of an instrument to
measure emotional expression[12 ],[13 ],[15 ],[30 ]. The efficacy of the application of the EESC has also been confirmed in the literature[15 ],[24 ].
There is little literature regarding this issue, which was restrictive for this study.
It also had other limitations: the EESC was validated in a sample with individuals
with comorbidities and was based only on the responses of the parents.
According to the psychometric data on internal and external consistency (reliability
and reproducibility), sensitivity, specificity and accuracy, the parent-rated EESC
was useful in assessing the intensity and quality of the emotions expressed by the
child or the adolescent between six and 15 years old, diagnosed with ADHD and under
clinical supervision, verifying the treatment results on the patient’s life.
