Relation between temperament dimensions and attention-deficit/hyperactivity disorder symptoms.

INTRODUCTION: The relation between temperament and attention-deficit/hyperactivity disorder (ADHD) is complex and understood in different ways, with the most common being risk model and spectrum model. However, the evidence is mixed and emerging. AIM: To assess the relationship between ADHD symptoms and temperament dimensions in a clinical sample of school-aged children.
METHODS: A retrospective assessment of temperament of 50 children with ADHD was done on temperament measurement schedule. The mean and standard deviation was computed for continuous variables and frequency and percentage for discontinuous variables and correlation and regression analysis was computed.
RESULTS: Children with ADHD were high on activity level, intensity of reaction, approach, and distractibility and low on persistence and threshold of responsiveness. The strength of significant correlations between temperamental dimensions and ADHD symptoms (P < 0.05) ranged from 0.32 to 0.41. On regression analysis, temperament could explain 22% variance of inattention subscale and around 20% variance in hyperactivity/impulsivity subscale.
CONCLUSION: This moderate level of relation suggests that though certain temperamental traits are related to symptoms of ADHD, temperament and ADHD are phenotypically separate constructs, further favoring the risk model. Copyright:

Temperament or the unique and innate psychobiological characteristics in children present since birth[12] have been considered important in the development of psychopathology.[3] The relation between temperament and psychopathology has been explained in two major ways: first, temperament as a risk/vulnerability factor for later psychopathology and second, temperament and psychopathology lie on a continuum or spectrum; i.e., the disorder is a severe form of temperament.[4] Attention-deficit/hyperactivity disorder (ADHD) is a childhood-onset, neurodevelopmental disorder characterized by inattention, impulsivity, and hyperactivity with significant impairments in functioning in various domains.[5] Studies have demonstrated the presence of “difficult” or high maintenance temperament characterized by higher levels of emotionality, activity, and negative affect and lower effortful control in individuals with ADHD.[678] These temperamental dimensions when characterized as problem behaviors are similar to symptoms of ADHD.[7] Associations have been demonstrated between temperament and ADHD mainly in adults and adolescents[4691011] and recently in children with ADHD.[3781213141516] Hence, some authors have asserted that ADHD can be considered as an extreme temperament type, i.e., the continuum hypothesis.[68171819] On the contrary, there is growing research that considers temperament as a risk factor for ADHD.[342021] In midst of these somewhat contradicting theoretical positions, the direction and magnitude of relation between temperamental dimensions and ADHD may help.[2122] In this background, we explored the relation between temperament and ADHD in a clinical sample of school-aged children.

METHODS

The study was carried out in child and adolescent psychiatry clinic, department of psychiatry of a tertiary care hospital in North India. It was approved by the institutional ethics review committee. It had a cross-sectional study design. Fifty children aged between 4 and 14 years diagnosed with ADHD as confirmed by Mini International Neuropsychiatric Interview for children and adolescents[23] were recruited using purposive sampling. Children with moderate-to-severe intellectual disability (intelligence quotient of <55 on a standardized measure), autism spectrum disorder, seizure disorder, organic brain syndrome, and those not giving assent/consent for the study were excluded. Written informed consent from parents and assent from the children (the latter whenever possible) was obtained before recruitment.

Temperament was assessed using the temperament measurement schedule (TMS),[24] which is an Indian adaptation of Thomas and Chess Temperament Questionnaire.[25] TMS is a parent interview schedule that measures temperamental traits retrospectively. It provides information about nine temperamental dimensions, namely, approach withdrawal, adaptability, threshold of responsiveness, mood, persistence, activity level, distractibility, and rhythmicity. Each temperamental dimension is assessed on the basis of 4 or 5 questions, scored from 1 to 5, with 1 indicating absence and 5 indicating maximum level of manifestation. Five factors are derived from these nine dimensions, i.e., Factor I – sociability (consisting of dimensions of approach-withdrawal, adaptability, and threshold of responsiveness), Factor II – emotionality (mood and persistence), Factor III – energy (activity level and intensity of reaction), Factor IV – distractibility, and Factor V – rhythmicity. Conner's Parent Rating Scale-Revised: Short Form (CPRS-R: S)[26] was used for assessment of severity of ADHD. CPRS-R: S has been used in earlier studies rendering the findings of this study comparable to existing literature. Sociodemographic and clinical details were recorded in structured formats. Parents were interviewed using the TMS for retrospective assessment of temperament during infancy and they were requested to rate the CPRS-R: S.

TheIBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp. Released 2011. was used for statistical analysis. Means and standard deviation were calculated for continuous variables, and frequency and percentages for discontinuous variables. Correlation matrix and analysis was carried out to assess correlation of TMS with CPRS-R: S. Linear regression analysis by stepwise method was conducted to measure for the variance in ADHD subscale scores as assessed by CPRS- R:S explained by temperament dimensions as measured by TMS depending on the results of correlation analysis.

RESULTS

The sociodemographic and clinical profile of the children is presented in Table 1.

Table 1

Sociodemographic and clinical profile of children with attention-deficit/hyperactivity disorder

Sociodemographic variables	Mean±SD	n (%)
Age (years)	10.2±2.6	-
Gender
Male	-	45 (90)
Female		5 (10)
Education (years)	4.6±2.9	-
Family type
Nuclear	-	34 (68)
Extended		12 (24)
Joint		4 (8)
Locality
Urban	-	40 (80)
Rural		10 (20)
Revised Kuppuswamy’s socioeconomic status
Upper	-	7 (14)
Upper middle		24 (48)
Middle/lower middle		16 (32)
Lower/upper lower		3 (6)
Age at onset (years)	4.6±1.5	-
Duration of illness (years)	5.5±2.3	-
Intelligence quotient
Mean	86.5±17.1	-
Comorbid psychiatric disorders
Learning disability	-	18 (36)
Externalizing disorders	-	13 (26)
Others*	-	14 (28)
Conner’s parent rating scale- revised: short form
ADHD subscales	Scores
	Minimum score	Maximum score	Mean±SD
Oppositional	0	18	7.3±4.1
Inattention	2	17	12.3±3.7
Hyperactivity	2	17	9.3±3.6
ADHD index	6	33	22.94±6.27

*Borderline intelligence (n=8), Tic disorder and Obsessive compulsive disorder (n=2 each), Stuttering and enuresis (n=1 each). ADHD – Attention-deficit/hyperactivity disorder; SD – Standard deviation

Attention-deficit/hyperactivity disorder symptom severity as assessed by Conner's parent rating scale-revised: short form in children with attention-deficit/hyperactivity disorder

Children with ADHD had scores toward maximum in symptom domains of inattention, hyperactivity, and combined type, indicating that these children had moderate-to-severe level of symptoms. Table 1 depicts the severity of ADHD symptoms in children with ADHD as assessed by CPRS-R: S.

Temperament of children with attention-deficit/hyperactivity disorder as measured by temperament measurement schedule

The temperamental characteristics of children with ADHD as measured by TMS are depicted in Table 2. Children with ADHD scored high on the energy factor (4.01 ± 0.72) followed by distractibility factor (3.7 ± 1.0) and had a low score on the emotionality factor (2.73 ± 0.92). They were found to score highest on activity level (4.1 ± 0.80) and least on threshold of responsiveness (2.25 ± 1.1).

Table 2

Temperamental characteristics of children with attention-deficit/hyperactivity disorder

	Scores, mean±SD
	Raw score	Weighted score
Temperamental traits
Approach-withdrawal	14.7±4.1	3.6±1.0
Adaptability	17.7±5.2	3.5±1.0
Threshold of responsiveness	11.2±5.8	2.25±1.1
Mood	12.6±4.2	3.17±1.05
Persistence	11.4±5.4	2.29±1.1
Activity level	20.8±4.0	4.1±0.80
Intensity	19.3±5.0	3.86±1.01
Distractibility	18.5±5.2	3.7±1.04
Rhythmicity	15.1±4.08	3.0±0.81
Temperament factors
Factor I- sociability	43.7±9.04	3.1±0.64
Factor II- emotionality	24.1±8.4	2.73±0.92
Factor III- energy	40.1±7.27	4.01±0.72

SD – Standard deviation

Relation of attention-deficit/hyperactivity disorder symptoms as measured by Conner's parent rating scale-revised: short form and temperament as measured by temperament measurement schedule in children with attention-deficit/hyperactivity disorder

Inattention subscale of CPRS-R: S had significant positive correlations with temperamental dimensions of approach-withdrawal (i.e., higher on approach), adaptability, and activity level and the energy factor. Hyperactivity subscale had a significant positive correlation with activity level and energy factor. ADHD index had a significant positive correlation with temperamental dimensions of approach-withdrawal and activity level and energy factor. None of the other temperamental dimensions had significant correlation with any of the subscales of CPRS-R: S. The relation between temperament as measured by TMS and ADHD symptoms as rated on the CPRS-R: S in children with ADHD is shown in Table 3.

Table 3

Relation between temperament and subscales of Conner’s Parent Rating Scale- Revised: Short form in children with attention-deficit/hyperactivity disorder

ADHD subscales	Temperament traits
	Approach withdrawal	Activity level	Adaptability	Energy factor
Inattention	0.372** (0.00)	0.362** (<0.01)	0.285* (0.04)	0.306* (0.03)
Hyperactivity	-	0.445** (0.00)	-	0.297* (0.03)
ADHD index	0.298* (0.03)	0.398** (0.00)	-	0.344* (0.01)

*P<0.05, **P<0.01. ADHD – Attention-deficit/hyperactivity disorder

Variance of attention-deficit/hyperactivity disorder symptoms as explained by temperament

Stepwise linear regression was calculated to predict inattentive subscale scores based on approach, activity level, and adaptability. Energy factor was not taken into account while computing regression analysis as activity level (which has significant correlation with ADHD subscales) is a subcomponent of energy factor. A significant regression equation was found (F (1, 48) =7.703, P < 0.01), with an R2 = 0.138 with approach alone as an independent variable, and (F (2, 47) =6.203), with an R2 = 0.209. Stepwise linear regression was calculated to predict hyperactivity subscale scores based on activity level. A significant regression equation was found (F (1, 48) =11.872, P < 0.01), with an R2 = 0.198. On conducting stepwise linear regression to predict ADHD index scores based on activity level and approach-withdrawal, a significant regression equation was found (F (1, 48) =9.012, P < 0.01), with an R2 = 0.158 for activity level alone. Thus, 13%–21% of variance in ADHD symptoms as measured by CPRS-R: S was explained by temperament. Temperamental dimensions of approach withdrawal explained 21% of variance for inattention. Activity level explained 19% and 15% variance for hyperactivity and ADHD index, respectively Table 4.

Table 4

Variance in attention-deficit/hyperactivity disorder symptoms (as measured by Conner’s parent rating scale-revised: Short form) by temperament

ADHD subscale	Score, R2
Inattention subscale
Approach-withdrawal	0.138**
Approach withdrawal, activity	0.209**
Hyperactivity subscale
Activity level	0.198**
ADHD index subscale
Activity level	0.158**

**P<0.01. ADHD – Attention-deficit/hyperactivity disorder

DISCUSSION

Our study explored the relation between temperament and ADHD in a clinical sample of children and adolescents. Till recently, there were only few studies in this area, and our study adds to the growing literature.[1011121427]

Our children scored higher than average on activity, intensity of reaction, distractibility, and approach temperamental dimensions, while they had scored lower on persistence and threshold of responsiveness. Thus, these children were temperamentally “always on the move, jump (s), rather than walk (s),” had extremes of reaction (roaring with laughter or very angry/annoyed), had difficulty attending to task at hand, being highly distractible along with difficulty in persisting at task, easily bothered by noise, pain, temperature, and other sensory perceptions. They also “went and talked spontaneously, rushed into new places, spontaneously touched or held things.”[28] Such a temperament profile is similar to the “high maintenance temperament” described by McClowry[17] and characterized by high activity, negative reactivity, and low task persistence. Similar findings have been reported earlier by different authors using different temperament assessment schedules based on different models of temperament.[821222429]

In the present study, the inattention subscale, hyperactivity subscale, and ADHD index of CPRS-R: S had a significant positive correlation with temperamental dimensions of activity level and energy factor comprised of activity level and intensity of reaction. Also, inattentive children tended to be high on approach and were easily adaptable. Temperamental dimension of 'high approach' is characterized by eagerness, readily seeking new experiences with curiosity and openness but reacting impulsively. Hence, it consists of elements in common with high surgency and low effortful control of Putnam and Rothbart[30] Child Behavior Questionnaire. High surgency comprises high activity level, high-intensity pleasure seeking, low shyness, and impulsivity. Low effortful control is characterized by poor inhibitory control and low attentional focusing. High surgency has been found to be associated with inattention and hyperactive/impulsive symptoms of ADHD and low effortful control with hyperactive/impulsive symptoms.[19] Nigg et al., 2002, Nigg et al., 2004, and Martel et al.[42122] found that the latter was also associated with inattentive symptoms or subtype of ADHD, as seen in our study. Lemery et al.[20] had also reported association between activity level, attentional focusing, inhibitory control, and behavioral symptoms of inattention and impulsivity. Foley et al.[8] had also reported that all three subtypes of ADHD were strongly associated with high activity, high impulsivity, high negative reactivity, and low task persistence and higher approach primarily with hyperactivity/impulsivity symptoms.[21] Thus, the findings of the present study are consistent with those in literature.[48192021]

In the present study, strength of significant associations between temperamental dimensions and ADHD symptoms (at P < 0.05) ranged from 0.32 to 0.41. This moderate level of relation suggests that though certain temperamental traits are related to symptoms of ADHD, temperament and ADHD are phenotypically separate constructs. Moreover, on regression analysis, temperament could explain around 22% variance of inattention subscale and around 20% variance of hyperactivity/impulsivity subscale. Earlier studies have also reported modest associations between temperament and ADHD.[820] Our study in conjunction with the previous studies suggests that there is an association, but not identity between ADHD symptoms and key temperament domains, as proposed by Nigg et al.[21] In such a case, temperament may be considered as a risk or vulnerability factor for ADHD, with early deficits in regulation disrupting effortful control[21] and leading to ADHD symptoms in interaction with environmental variables[27] or executive functioning.[14]

CONCLUSION

To summarize, modest associations were found between temperament dimensions and ADHD symptoms, and temperament explained less than a quarter of the variance, indicating that temperament is a vulnerability factor for ADHD.

The present study has several limitations. First, the sample size is small, thus increasing the chances of Type 1 error. Hence, Bonferroni's correction was applied. Second, a major limitation is that the study had a cross-sectional design and the subjects had comorbid disorders; hence, causal inferences cannot be drawn. Also, temperament assessments were done retrospectively and are vulnerable to recall bias. Longitudinal study design with prospective assessments, though difficult would be ideal.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.