Patterns of temperament and character in a clinical sample of Korean children with attention-deficit hyperactivity disorder

Authors


*Jun-Won Hwang, MD, PhD, Division of Child and Adolescent Psychiatry, Department of Psychiatry, Seoul National University College of Medicine, Seoul National University Hospital, Yeongeon-dong, Chongno-gu, Seoul 110-744, Korea. Email: huangjw1@snu.ac.kr

Abstract

Aims:  The purpose of the present study was to assess patterns of temperament and character in a clinical sample of Korean children with attention-deficit hyperactivity disorder (ADHD) and to investigate the relationship between patterns of temperament and character and the severity of ADHD symptoms.

Methods:  Fifty-one children who met DSM-IV criteria for ADHD and 51 age- and sex-matched healthy comparison subjects completed the Junior Temperament and Character Inventory (JTCI) and the DuPaul's ADHD Rating Scale (ARS-IV).

Results:  Children with ADHD had significantly higher scores of Novelty Seeking and lower scores of Self-directedness in both parents' rating and children's self-rating. In addition, low Self-directedness scores predicted higher total scores of ARS-IV.

Conclusions:  In a clinical sample of Korean children with ADHD, a distinct pattern of temperament and character was found. The current findings suggest that low Self-directedness may be related to the severity of ADHD symptoms.

ATTENTION-DEFICIT HYPERACTIVITY disorder (ADHD) is characterized by developmentally inappropriate inattention, impulsivity, and hyperactivity.1 It has been considered as a highly heritable but clinically heterogenous disorder with earlier onset of symptoms before 7 years of age, suggesting the continuity between temperament and ADHD.2 Moreover, considering its persistence into adulthood and developmental outcome,3 exploring the relationship between character and ADHD may be needed.

The biosocial model of Cloninger's temperament and character is based on biogenetic hypotheses underlying human behaviors. Temperament dimensions of Novelty Seeking, Harm Avoidance, Reward Dependence, and Persistence have been reported to be genetically independent from one another, moderately heritable and stable across time.4,5 Each represents specific stimulus–response characteristics. Character dimensions include Self-directedness, Cooperativeness, and Self-transcendence. Each refers to individual differences in self-concepts, goals, and values. In contrast to temperament, character has been regarded as being more influenced by social learning rather than by genetic factors. It is expected to mature during the developmental stages throughout the lifespan.4

Several studies have investigated patterns of temperament and character in subjects with ADHD. McIntosh reported that children with ADHD had temperaments with higher activity levels, higher distractibility, and lower persistence compared to healthy peers using the Temperament Assessment Battery for Children.6 Recently, lower task orientation and higher activity were related to ADHD versus controls using the Revised Dimensions of Temperament Scale–Child.7 In addition, adults with ADHD had high scores in extraversion and neuroticism among the big five personality traits,8 which were thought to correlate with Novelty Seeking and Harm Avoidance, respectively.9,10

In line with this finding, Downey et al. reported that adults with ADHD scored higher on Novelty Seeking and Harm Avoidance scales using the Tridimensional Personality Questionnaire (TPQ).4,11,12 Using the Junior Temperament and Character Inventory (JTCI),13 Tillman et al. found that patterns of temperament and character in children with bipolar disorder were similar to those in children with ADHD.14 However, although major depressive disorder was excluded, other concurrent comorbidities including anxiety disorder, which may also influence patterns of temperament and character,15 were not sufficiently considered in that study. In addition, sociocultural differences across ethnicity or sample source should be considered in interpreting these findings. Our group has recently reported on the patterns of temperament and character in Korean children with higher ADHD symptoms in the community,16 which also needs to be confirmed in a clinical sample of children with ADHD.

Until now, the relationship between temperament and character or personality and the severity of ADHD symptoms has been studied in the adult population.17,18 In addition, the associations between temperament and the diagnosis of ADHD or other related conditions have been suggested in children.15 Recently, in a community sample of Korean children with higher ADHD symptoms, our group found some correlation between temperament and character and the severity of ADHD symptoms including inattention and hyperactivity–impulsivity.16 But the correlation has not been investigated in a clinical sample of children with ADHD.

The aim of the present study was to evaluate biogenetic temperament and character in a clinical sample of Korean children with ADHD, using a Korean version of the JTCI.19 In addition, a specific relationship between patterns of temperament and character and the severity of ADHD symptoms was investigated. Based on prior reports, we hypothesized that ADHD children will have higher Novelty Seeking, lower Reward Dependence, and lower Persistence in temperament (hypothesis 1) and lower Self-directedness and lower Cooperativeness in character (hypothesis 2) relative to healthy comparison subjects. As an auxiliary hypothesis, we expected that there would be correlations between characteristics of temperament and character and the severity of ADHD symptoms in the present sample (hypothesis 3).

METHODS

Subjects

Fifty-one children with ADHD (46 boys and five girls) were selected from the ADHD Clinic in the Seoul National University Hospital, Seoul, South Korea. The diagnoses of ADHD were made according to DSM-IV,1 confirmed by parent interviews using Korean version of Kiddie–Schedule for Affective Disorder and Schizophrenia–Present and Lifetime Version (K-SADS-PL-K).20 In addition, children with ADHD were also required to have total scores of Korean version of DuPaul's ADHD Rating Scale-IV (ARS-IV)21,22 at least 1.5 SD above age and sex norms.23 The K-SADS-PL-K and the ARS-IV were conducted by two board-certified psychiatrists (J.W.H and B.N.K).

Exclusion criteria were comorbid DSM-IV current Axis I disorders except oppositional–defiant disorder, conduct disorder, and enuresis, which would be the most common comorbid disorders among children with ADHD and have considerable overlaps with ADHD in genetic and environmental influences,24,25 current or past neurologic illness, mental retardation, and substance abuse as evaluated by history, physical examination, and laboratory testing (complete blood count, urinalysis, liver function test, and serology). After excluding eight children who had comorbidities in the exclusion criteria (one due to major depressive disorder, four due to specific phobia, two due to separation anxiety disorder and/or obsessive–compulsive disorder, and one due to Tourette disorder) and 16 children who did not complete more than 10% of total items of the JTCI, information on the remaining 51 children with ADHD was used for statistical analysis. There were no significant differences in demographic variables between excluded and included subjects.

Fifty-one age- and sex-matched, healthy comparison subjects were recruited through advertisement at the same institution. Children were excluded from analysis if they had T scores >60 (85th percentile) in any subscales of Korean version of the Child Behavior Checklist26 or any DSM-IV current Axis I disorders that were also confirmed by two board-certified psychiatrists (J. W. H and B. N. K). Other exclusion criteria included current or past neurologic illness, mental retardation, and substance abuse as evaluated by history and physical examination.

The present study was conducted as a part of the Integrative Neuroimaging Study in Children with ADHD. Current study protocol was approved by the Institutional Review Board at Seoul National University Hospital. After complete description of the study to the subjects, written informed consent was obtained.

Assessments

Junior Temperament and Character Inventory

The JTCI, a 108-item questionnaire, is used in order to assess the biogenetic temperament and character of children 7–14 years old.13 In the present study both children's self-rating and parent's rating version were completed. Cronbach alpha values of the Korean version of the JTCI scales ranged from 0.48 to 0.80 for temperament scale and from 0.64 to 0.68 for character scales. Test–retest correlations ranged from 0.62 to 0.85 for temperament scales and from 0.76 to 0.79 for character scales.19

ADHD Rating Scale-IV–Parent and Teacher version

The ARS-IV consists of 18 items: nine items for inattention and nine items for hyperactivity–impulsivity, based on the symptom list for ADHD delineated in the DSM-IV.21 Internal consistency of the Korean version of the ARS-IV by age is 0.77–0.89. Concurrent validity from correlations with Child Behavior Checklist and Korean version of Conner's scale is also high.22

Statistical analysis

Differences in demographic variables between two groups were tested using an independent t-test for continuous variables and Fisher's exact test for 2 × k table. Correlations between continuous data were analyzed using Pearson's correlation analysis.

It should be noted that whether temperament causes psychopathology or vice versa remains controversial. In the present study, based on the assumption that temperaments and characters may affect behaviors,4 we explored influence of temperament and character on the severity of ADHD symptoms in children with ADHD.

Initially, scales of the JTCI were compared using an independent t-test. As a next step, the influence of temperament and character on ARS-IV score was analyzed using hierarchical multiple regression. After an independent t-test, only scales in the JTCI that had significant differences were entered into the multiple regression model, controlling for age and sex. < 0.05 was taken as significant. All statistical analysis was two-tailed and conducted using the Windows version of SPSS 11.0 (SPSS, Chicago, IL, USA)

RESULTS

Demographic and clinical characteristics are presented in Table 1. Forty-four children (86.3%) with ADHD had never taken any psychotropic medication and seven children (13.7%) were not treated with psychotropic drugs in at least 2 weeks preceding evaluation.

Table 1.  Characteristics of children with ADHD and healthy comparison subjects
 Children with ADHD
(n = 51)
Mean ± SD
Healthy comparison subjects
(n = 51)
Mean ± SD
  1. ADHD, attention-deficit hyperactivity disorder; ARS-IV, ADHD Rating Scale-IV–Parent and Teacher version; –, not applicable.

Age (years)10.0 ± 2.010.3 ± 0.7
Gender
 Male, n (%)46 (90.2)46 (90.2)
 Female, n (%)5 (9.8)5 (9.8)
ADHD subtype
 Combined type22
 Predominantly inattentive type15
 Predominantly hyperactive-impulsive type4
 Not otherwise specified10
Comorbidity
 Oppositional defiant disorder10 (1, past)
 Enuresis6 (3, past)
ARS-IV
 Total score26.1 ± 11.9
 Inattention subscale score15.7 ± 7.1
 Hyperactivity-impulsivity subscale score10.4 ± 6.1

Scores of the JTCI are presented in Fig. 1. Overall correlations of JTCI scores between children's self-rating and parents' rating were significant in all dimensions (r = 0.252–0.338, P < 0.05). In parents' rating, children with ADHD had higher scores in Novelty Seeking and lower scores in Persistence, Self-directedness, and Cooperativeness than healthy controls (independent t-test: d.f. = 100, t = 5.810, P < 0.001; d.f. = 100, t = −4.347, P < 0.001; d.f. = 100, t = −5.559, P < 0.001; d.f. = 100, t = −3.917, P < 0.001, respectively; Fig. 1a). In addition, on children's self-rating, children with ADHD had higher scores in Novelty Seeking and Self-transcendence and lower scores in Reward Dependence and Self-directedness than healthy controls (d.f. = 100, t = 2.555, P = 0.012; d.f. = 100, t = 2.942, P = 0.004; d.f. = 100, t = −2.073, P = 0.041; d.f. = 100, t = −2.312, P = 0.022, respectively; Fig. 1b). In parents' rating of Reward Dependence and Self-transcendence and children's self-rating of Persistence, there were tendencies toward lower scores in children with ADHD (d.f. = 100, t = −1.906, P = 0.060; d.f. = 100, t = −1.689, P = 0.094; d.f. = 100, t = −1.763, P = 0.081, respectively).

Figure 1.

Comparison of Junior Temperament and Character Inventory (JTCI) Scores reported by (a) parents and (b) children. (▪) Children with ADHD; (□) healthy control subjects. ADHD, attention-deficit hyperactivity disorder; C, Cooperativeness; NS, Novelty Seeking; HA, Harm Avoidance; P, Persistence; RD, Reward Dependence; SD, Self-directedness; ST, Self-transcendence. Results are reported as mean with 95% confidence interval otherwise indicated. *< 0.05; **< 0.01; ***P < 0.001.

Hierarchical multiple regression was performed to examine the influence of temperament and character on symptoms in children with ADHD controlling for age and sex (Tables 2,3). In both parents' rating and children's self-rating, only low Self-directedness scores significantly predicted higher total scores in ARS-IV, indicating more severe symptoms, controlling for age and sex (in parents' rating, multiple regression analysis, β, standardized regression coefficients = −0.356, t = −2.164, P = 0.036; in children's self-rating, β = −0.342, t = −2.041, P = 0.048). In detailed analysis, Self-directedness scores had negative correlations with both hyperactivity–impulsivity scores and inattention scores (parents' rating, β = −0.312, t = −1.878, P = 0.067; β = −0.327, t = −1.976, P = 0.055; children's self-rating, β = −0.333, t = −2.001, P = 0.052; β = −0.286, t = −1.687, P = 0.099, respectively). But scores of other scales in the JTCI (Novelty Seeking, Persistence, and Cooperativeness in parents' rating and Novelty Seeking, Reward Dependence, and Self-transcendence in children's self-rating) could not predict any scores including total scores, hyperactivity–impulsivity scores, and inattention scores in ARS-IV.

Table 2.  Influences of temperament and character on symptom severity in children with ADHD (parent's report)
ARS-IVJTCI subscaleβtP
  1. ADHD, attention-deficit hyperactivity disorder; ARS-IV, ADHD Rating Scale-IV–Parent and Teacher version; JTCI, Junior Temperament and Character Inventory.

Total scoreNovelty seeking−0.082−0.4830.632
Persistence0.1600.9780.334
Self-directedness−0.356−2.1640.036
Cooperativeness0.2211.2370.223
Inattention subscale scoreNovelty seeking−0.057−0.3320.742
Persistence0.1701.0370.306
Self-directedness−0.327−1.9760.055
Cooperativeness0.1670.9290.358
Hyperactivity–impulsivity subscale scoreNovelty seeking−0.094−0.5440.589
Persistence0.1130.6840.498
Self-directedness−0.312−1.8780.067
Cooperativeness0.2351.3040.199
Table 3.  Influences of temperament and character on symptom severity in children with ADHD (children's self-report)
ARS-IVJTCI subscaleβtP
  1. ADHD, attention-deficit hyperactivity disorder; ARS-IV, ADHD Rating Scale-IV–Parent and Teacher version; JTCI, Junior Temperament and Character Inventory.

Total scoreNovelty seeking0.0070.0400.968
Reward dependence0.1040.6470.521
Self-directedness−0.342−2.0410.048
Self-transcendence0.0320.1950.846
Inattention subscale scoreNovelty seeking−0.052−0.2920.772
Reward dependence0.1330.8240.415
Self-directedness−0.286−1.6870.099
Self-transcendence−0.001−0.0060.995
Hyperactivity–impulsivity subscale scoreNovelty seeking0.0740.4220.675
Reward dependence0.0470.2940.770
Self-directedness−0.333−2.0010.052
Self-transcendence0.0630.3890.699

DISCUSSION

In the present study we found that children with ADHD had higher Novelty Seeking and lower Self-directedness relative to healthy controls in both parents' and children's self-rating. In addition, lower Persistence and lower Cooperativeness in parents' rating and lower Reward Dependence and higher Self-transcendence in children's self-rating were found in children with ADHD. The patterns in the current study are in partial accord with those in previous studies regarding the biogenetic temperament and character in both children and adults with ADHD.11,12,14,16

Downey et al. reported that adults with ADHD scored higher Novelty Seeking11,12 and Harm Avoidance12 than the established US norm using the TPQ. In their detailed analysis, ADHD adults with comorbidity scored lower than those without comorbidity on the TPQ, including pessimism and fatigability subscales in Harm Avoidance and dependence subscale in Reward Dependence. In addition, ADHD adults without comorbidity scored lower than 1 SD below the established US norm only on the TPQ pessimism subscale. Therefore, the higher Harm Avoidance seems to be due to the higher rate of comorbid anxiety or depressive disorders in the Downey et al. sample (47.4% of total ADHD subjects).

In the Tillman et al. study parent-reported data showed that children with ADHD were more Cooperative than children with bipolar disorder.14 In addition, while there were no differences in their child-reported data; children with ADHD had higher Novelty Seeking, less Reward-dependence less Persistence, less Self-directed, and less Cooperativeness than normal controls in the parent-report data.

Our group has recently reported that community children with higher ADHD symptoms, whose ARS-IV score were above the 90th percentile of the total study population, had higher Novelty Seeking and lower Self-directedness and Cooperativeness in both parents' and children's rating as well as lower persistence in parents' rating only.16 In summary, the current results support the idea that specific patterns of temperament and character may exist in children with ADHD across ethnicity and sample source.

Although the patterns in the parent rating were the same as the current results, a few different patterns were also found in children's rating in our prior reports in the community.16 The differences in the children's rating may be partly explained by the following reasons. The current study was based on clinically referred subjects, who represent a subgroup of greater symptoms and subjective distress. In addition, in our previous study the formal diagnosis was not made because subject selection was based on ARS-IV score only; risks for false-positive diagnosis and influences of other comorbid psychopathology could be inevitable. Although overall correlations were maintained satisfactorily (r = 0.252∼0.338, P < 0.05), there were statistical discordances between parent–child report of temperament and character other than Novelty-Seeking, Harm Avoidance, and Self-directedness. The statistical discordances between children's and parents' report in the current study are consistent with previous studies.14,27 And because the Cronbach alpha for the subscales of the Korean version of the JTCI is relatively low compared to the adult TCI,19 readers should be cautious in interpreting the present findings.

High Novelty Seeking has consistently been reported in children and adults with ADHD,11,12,14,16 as well as in parents with a history of ADHD who had an ADHD-affected sib pair.18 Novelty Seeking is linked to behavioral activation and describes a genetic disposition towards being excitable, impulsive, exploratory, and quick tempered.5 It has been considered to be related to monoaminergic neurotransmitters.28 Currently the associations among Novelty Seeking, DRD4, and ADHD have been a topic of interest and debate.11,29 Future studies are needed to investigate to what extent these monoamine-related genes including DRD4, play a role in the genetic liability to ADHD, Novelty Seeking, or both.

The present results also indicate that greater severity of ADHD symptoms may be explained by low Self-directedness of children with ADHD. Self-directedness refers to self-determination and willpower, or the ability of an individual to control, regulate, and adapt behavior to fit the situation in accord with individually chosen goals and values.5 Low Self-directedness has been associated with various mental health problems including juvenile delinquency and eating disorder as well as ADHD in children and adolescents, and is expected to exist across externalizing disorders.16,30–32 The symptom severity in the present clinical sample might be outcomes of low self-directedness in character, along with adverse environmental influences associated with children with ADHD even in childhood.

Limitations of the current study are as follows. First, it was not possible to test developmental changes of the association between temperament and character and the severity of ADHD symptoms due to the cross-sectional nature of the symptoms. Second, although the results were in partial accord with those in previous studies, they may not be easily generalizable to other sociocultural environments. Third, we recruited children from clinical settings for ADHD who might have different patterns of temperament and character and symptom severity, which limits the use of these findings in the community. Fourth, because of the small sample size, the current findings may not be replicated in the future, large-scale studies.

In summary, the present study reports specific patterns of temperament and character and their association with symptom severity in children with ADHD. In the future, in order to investigate the association among patterns of temperament and character, symptom severity in childhood and adolescence, and the later psychosocial outcome in adulthood, recruitment of cohorts with longitudinal follow-up on a large scale are needed.

Ancillary