Is the temperamental trait of high persistence protective in siblings? - A comparative, exploratory study of healthy siblings, and attention deficit hyperactivity disorder probands.

Background: Study of temperament in first-degree relatives is an important line of inquiry to substantiate temperament as an etiological marker. Aim: This study aims to compare temperament in children with attention deficit hyperactivity disorder (ADHD) and their healthy siblings and to assess the association between ADHD symptoms and temperament dimensions in healthy siblings. Settings and Design: The study was carried out in the outpatient department of psychiatry in a tertiary care teaching hospital. A cross-sectional design with nonprobabilistic sampling technique was used for data collection. Materials and
Methods: A hundred children (50 children with ADHD and 50 siblings-one for each child with ADHD) were assessed retrospectively on temperament measurement schedule (TMS) and conners parent rating scale-revised: short form (CPRS-R: S). Statistical Analysis: IBM SPSS Statistics for Windows, Version 20.0 was used for statistical analysis. Mean and standard deviation and frequency and percentage were computed for continuous and categorical variables, respectively. Student's t-test was computed to compare means of the two groups and regression analysis was computed to see for the variance in ADHD subscale scores explained by temperament scores on TMS.
Results: Siblings scored highest on the intensity of reaction and lowest on threshold of responsiveness. Compared to probands, siblings scored significantly higher on persistence and lower on activity level, even after controlling for gender. Persistence trait had a significant negative correlation with and explained 7.4% to 21% of variance of all CPRS-R: S subscales. Persistence and distractibility together explained 23.2% of inattention scores.
Conclusion: Higher persistence in siblings appears to offer protection to these at-risk individuals who do not have ADHD; favoring the dual pathway model of ADHD. Copyright:

HIGHLIGHTS

The association of temperament and ADHD is well established with multiple and dual pathway models explaining this association

Compared to ADHD probands, healthy siblings scored highest on intensity of reaction and lowest on threshold of responsiveness

Siblings scored significantly higher on temperamental trait of persistence and it had a negative correlation with and explained 7.4%–21% of variance ADHD symptoms

Persistence and distractibility together explained 23.2% of inattention score on conners parent rating scale-revised: Short form.

INTRODUCTION

Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder affecting 5%–7% of school-aged children globally[12] and a pooled prevalence of 7.1% in India.[3] It is characterized by excessive and developmentally inappropriate levels of inattention, activity and impulsivity.[4] ADHD runs a chronic course with persisting impairments in academics, work, social and inter-personal relations.[56] Despite being a common disorder and associated with significant morbidity and co-morbidities,[78] its exact cause is not known. The risk of ADHD is thought to be multi-factorial, with both genetic and environmental factors contributing to illness.[9] The most prominent theories of causal pathways integrate the role of executive functioning deficits, delay aversion,[10] and certain temperamental traits.[1112]

Temperament refers to unique and innate, psychobiological characteristics in children, present since birth which overtly manifest within a few months of birth.[13] While, Thomas and Chess, defined temperament as a “behavioral style”[14] and identified nine dimensions-activity level, approach-withdrawal, mood, rhythmicity, persistence-attention span, adaptability, threshold, intensity, and distractibility,[15] Buss and Plomin, described three temperamental traits-emotionality, activity, and sociability.[16] Goldsmith and Campos, conceptualized temperament as individual differences in emotional domain and its regulatory aspects[17] while Rothbart and Bates, referred to temperament as individual differences in affective, motor, attentional, sensory sensitivity and reactivity;[18] and self-regulation processes such as effortful control that modulate this reactivity.[19]

Temperamental characteristics of high negative reactivity, activity, novelty-seeking, impulsivity, low agreeableness/hostility, low self-directedness, task persistence, low attentional focusing, inhibitory control, low conscientiousness/effortful control are associated with ADHD[2021222324252627282930] and studied in clinical samples of children,[21232627] community samples[2022242531] and in adolescents with ADHD[3233] in whom effortful control is considered to mediate some of the executive functioning deficits. A recent systematic review in adults with ADHD revealed associations with temperament traits of lability, irritability, and excessiveness of emotional responses.[34] Despite this overwhelming evidence of association of temperamental characteristics with ADHD, coupled with their moderate heritability and their role as putative risk factors, there is scarce and inconclusive research regarding temperamental characteristics in first-degree relatives of children with ADHD.[3536] While proposing their multiple pathway model, Nigg et al.[12] suggested the study of temperament in first-degree relatives as an important line of inquiry to substantiate temperament as a marker of etiological process. In this background, our study aimed to compare temperamental dimensions in children with ADHD and their healthy siblings, and to assess contribution of temperamental traits to ADHD symptomatology in siblings.

MATERIALS AND METHODS

The study had a cross-sectional design and was carried out in Child and Adolescent Psychiatry Services, Department of Psychiatry of a governmental postgraduate teaching hospital in north India between March 2015 and August 2016. Ethical clearance was obtained from the Institute Ethics Committee and strict data confidentiality was maintained. Since, it was an exploratory study and no previous studies had examined child-sibling pairs, thus sample size was mainly drawn from clinical studies in children with ADHD alone and study sample of 50 children with ADHD and 50 healthy siblings (at least one for each child with ADHD) was drawn by nonprobabilistic (convenience) sampling method. Child-healthy sibling pairs were recruited for the study. Children of either gender aged 4–14 years diagnosed clinically with ADHD or hyperkinetic disorder according to DSM IV[37] or International Classification of Diseases 10[38] and confirmed using the Mini International Neuropsychiatric Interview for children and adolescents (MINI KID)[39] and having a healthy sibling aged 4–14 years of either gender were included in the study. Healthy sibling was defined as a sibling of a child diagnosed with ADHD who had never been diagnosed with any psychiatric illness, was never referred by any teacher, medical or nonmedical agency, and had never sought psychiatric evaluation. The age range of 4–14 years was considered to maintain homogeneity in the two groups of the study sample and also because temperament assessment in the index study was done till the 3rd year of age for both child-sibling pair (essentially to ensure that the same symptoms may not contribute toward measuring temperamental trait and symptom severity of ADHD). In addition, most earlier clinic-based studies had included a broader age range including adolescents aged 14–18 and young adults. However, this was considered a limitation, as symptom profile in older adolescents and young adults may be different from those in the children and younger adolescents and therefore we restricted our study to children aged 4–14 years. The status was further confirmed by using the childhood psychopathology measurement schedule (CPMS).[40] Those siblings with a CPMS score of >10, were assessed on MINI-KID and none were found to have any diagnosable psychiatric disorder. Those children with ADHD who had moderate to severe intellectual disability, autism spectrum disorder, epilepsy, or any other neurological disorder; did not have a sibling or did not provide assent/consent for the study were excluded. In case of siblings, those with any known psychiatric disorder, intellectual disability, epilepsy, or chronic physical disorder were excluded from the study. Written informed consent from parents and assent from all participants was obtained prior to recruitment into the study. Results regarding the temperamental characteristics of the patient group have been presented elsewhere.[41]

Temperament measurement schedule (TMS)[42] which is an Indian adaptation of Thomas and Chess temperament questionnaire[14] was used for assessment of temperament till the 3rd year of age for both child-sibling pair. It is a bilingual parent interview schedule measuring nine temperamental traits, namely - Approach-withdrawal, adaptability, threshold of responsiveness, quality of mood, persistence, activity level, distractibility, and rhythmicity. Each temperamental trait is assessed based on 4 or 5 questions, scored from 1 to 5 (1-absence and 5-maximum level of manifestation). Five factors are derived from the nine traits, namely, Sociability (approach-withdrawal, adaptability, threshold of responsiveness), Emotionality (mood, persistence), Energy (activity level, intensity of reaction), distractibility and rhythmicity. TMS has been used in Indian children[26] and is found to be culturally valid.

Conners’ parent rating scale-revised: Short form (CPRS-R: S)[43] is a 27-item 4-point Likert-type scale that yields 4 mutually exclusive scale scores defined by factor analysis: The cognitive problems/inattention scale (6 items), the hyperactivity scale (6 items), oppositional scale (3 items), and the ADHD index (12 items). It is a diagnostic instrument as well as used to assess the severity of ADHD symptoms. CPRS-R: S is also used as a screening tool and ADHD index subscale identifies children at high risk of ADHD. It has good reliability and validity with alphas between 0.86 and 0.94; and 6-week test-retest correlations between 0.72 and 0.85.[43]

Children with ADHD fulfilled the inclusion and exclusion criteria and their parents were approached for recruitment in the study. Those providing written informed consent were included in the study. Socio-demographic and clinical details were recorded in structured formats. Intelligence quotient (IQ) of probands was extracted from routine clinical records while IQ testing using standard progressive matrices/CPM[44] was carried out for all sibling participants. Parents were interviewed using the TMS for retrospective assessment of temperament and rating on CPRS-R: S for children with ADHD and their siblings. The data was collected from March 2015 to March 2016.

Statistical analysis

The IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY, USA: IBM Corp. (Released 2011).[45] was used for statistical analysis. Means and standard deviation (SD) were calculated for continuous variables and frequency and percentages for discontinuous variables. Student’s t-test was computed to compare CPRS-R: S and TMS scores of probands and their healthy siblings. Analysis of covariance (ANCOVA) was used to control for gender while comparing the TMS and CPRS-R: S scores. Correlation matrix and analysis were carried out to assess the correlation of TMS with CPRS-R: S scores of siblings. Linear regression analysis by stepwise method was conducted to measure for the variance in ADHD subscale scores for siblings as assessed by CPRS-R: S explained by temperament dimensions as measured by TMS in them, depending on the results of correlation analysis.

RESULTS

A total of 80 children clinically diagnosed with ADHD were screened for recruitment in the study. Sixty-five patient-sibling pairs fulfilled the inclusion/exclusion criteria and consent was provided for 57 out of them. The final sample with complete assessments included 50 patient-sibling pairs as assessments could not be completed for 7 pairs.

Socio-demographic profile of children with attention deficit hyperactivity disorder and their siblings

Table 1 shows sociodemographic profile of the probands and their siblings.

Table 1

Comparison of Sociodemographic profile of patients with attention deficit hyperactivity disorder and their siblings

Variables	Mean (SD); Frequency (%)		t-test/Chi- square test (P)
	Patient (n=50)	Sibling (n=50)
Age (years)	10.2 (2.6)	9.2 (3.5)	1.7 (0.08)
Education (years)	4.6 (2.9)	3.9 (3.2)	0.8 (0.07)
Gender
Male	45 (90)	22 (44)	χ2=23.93 (<0.01)**
Female	5 (10)	28 (56)
Socioeconomic statusa
Lower middle and below	19 (38)		NA
Upper middle and above	31 (62)
Religion
Hindu	34 (68)		NA
Non-Hindu	16 (32)
Type of family
Nuclear	34 (68)		NA
Extended/joint	16 (32)
Locality
Urban	40 (80)		NA
Rural	10 (20)

**P≤0.01, aSocioeconomic status was determined by using Modified Kuppuswamy socioeconomic scale. SD – Standard deviation; χ2 – Chi-square value; t – t-test; NA – Not applicable

Clinical profile of children with attention deficit hyperactivity disorder

In children with ADHD, the mean age at onset of ADHD was 4.6 (SD 1.5) years with mean 5.5 (SD 2.3) years of duration of illness till the index presentation. Almost two-third (n = 36) of them received medication for ADHD and methylphenidate was most commonly used (66% of children receiving medications). Mean IQ of children with ADHD was 86.5 (SD 17.1) and there was no significant difference when compared to mean IQ of siblings (89.2 [SD 8.0]). 36% of children with ADHD had comorbid learning disability with another one-fourth having other comorbid externalizing disorders (oppositional defiant disorder [n = 4], conduct disorder [n = 9]).

Temperament measurement schedule and conners parent rating scale-revised: Short form scores in siblings and comparison with that in probands

Weighted scores on TMS showed that siblings scored highest on intensity of reaction (3.65 ± 0.79) and lowest on threshold of responsiveness (2.36 ± 0.99). Temperamental profile of siblings was characterized by high intensity of reaction and distractibility, average activity levels, adaptability, approach withdrawal and persistence, and generally calm mood. When compared to probands with ADHD, healthy siblings scored significantly higher on temperamental trait of persistence (P < 0.005) and low on activity level (P < 0.001). These results remained significant on controlling for gender. Table 2 shows the results.

Table 2

Comparison of Conners Parent Rating Scale scores and temperament of children with attention deficit hyperactivity disorder with their siblings

Variables	Mean (SD)		Mean difference (SE difference)	Comparison of patients and siblings, t-test (P) (95% CI)	ANCOVA test F (P) (gender)
	Patient (n=50)	Sibling (n=50)
CPRS-R: S scores
Oppositional	7.4 (4.1)	2.3 (1.8)	5.1 (0.6)	7.9 (0.000)*** (3.8–6.4)	42.9 (0.000)***
Inattention	12.3 (3.7)	3.1 (3.0)	9.1 (0.7)	13.3 (0.000)***(7.8–10.4)	131.6 (0.000)***
Hyperactivity	9.4 (3.7)	1.2 (1.8)	8.1 (0.6)	14.1 (0.000)*** (6.9–9.3)	154.0 (0.000)***
ADHD index	22.9 (6.3)	5.9 (4.8)	17.0 (1.1)	15.3 (0.000)*** (14.8–19.3)	168.2 (0.000)***
TMS dimension scores
Approach-Withdrawal	14.7 (4.1)	12.8 (4.1)	1.9 (0.82)	2.4 (0.021)* (0.30–3.6)	5.0 (0.027)*
Adaptability	17.7 (5.2)	16.0 (3.9)	1.7 (0.93)	1.8 (0.071) (−0.15–3.55)	4.4 (0.038)*
Threshold of responsiveness	11.3 (5.8)	11.8 (4.9)	−0.56 (1.1)	−0.52 (0.607) (−2.7–1.6)	0.22 (0.635)
Mood	12.7 (4.2)	13.7 (3.9)	−0.98 (0.81)	−1.2 (0.228) (−2.6–0.62)	0.27 (0.602)
Persistence	11.5 (5.8)	15.8 (4.7)	−4.4 (1.1)	−4.3 (0.000)*** (−6.5–2.3)	9.2 (0.003)**
Activity level	20.8 (4.0)	15.7 (3.9)	5.2 (0.79)	6.5 (0.000)*** (3.6–6.7)	33.5 (0.000)***
Intensity	19.3 (5.1)	18.3 (3.9)	1.1 (0.91)	1.2 (0.240) (−0.73–2.9)	0.29 (0.593)
Distractibility	18.5 (5.2)	17.4 (5.1)	1.1 (1.0)	1.1 (0.278) (−0.92–3.2)	0.31 (0.582)
Rhythmicity	15.1 (4.1)	15.2 (3.2)	−0.12 (0.74)	−0.16 (0.871) (−1.6-1.3)	0.05 (0.821)

CPRS-R: S – Conners Parent Rating Scale-Revised – Short form; ADHD – Attention deficit hyperactivity disorder; TMS – Temperament measurement schedule; SD – Standard deviation; SE – Standard error; CI – Confidence interval; ANCOVA – Analysis of covariance. *<0.05, **<0.01, ***<0.001

Siblings of children with ADHD scored significantly lower than the probands on all subscales of CPRS-R: S, i.e., oppositional (P < 0.001), inattention (P < 0.001), hyperactivity (P < 0.001) and ADHD index (P < 0.001). ANCOVA showed that these differences remained significant even after controlling for gender.

Correlation of temperamental dimensions with conners parent rating scale domain scores of siblings

Persistence as a temperamental trait emerged to have a significant negative correlation with all the four subscales (oppositional, inattention, hyperactivity, ADHD index) of CPRS-R: S in siblings of children with ADHD as shown in Table 3. Furthermore, distractibility had a significant positive correlation with the inattention subscale of CPRS-R: S. No other significant correlation was found between temperament dimensions and ADHD subscales.

Table 3

Correlation of temperamental dimensions with Conners Parent Rating Scale domain scores of siblings

Temperament dimensions	ADHD subscales
	Oppositional	Inattention	Hyperactivity	ADHD index
Approach-withdrawal	0.163 (0.257)	−0.177 (0.219)	0.050 (0.730)	−0.081 (0.574)
Adaptability	0.043 (0.768)	−0.195 (0.174)	0.033 (0.823)	−0.112 (0.439)
Threshold of responsiveness	−0.200 (0.163)	0.035 (0.807)	0.048 (0.742)	−0.102 (0.483)
Mood	−0.120 (0.405)	0.119 (0.411)	0.251 (0.078)	0.150 (0.300)
Persistence	−0.305 (0.031)*	−0.399 (0.004)**	−0.366 (0.009)**	−0.476 (0.000)***
Activity level	−0.054 (0.711)	−0.202 (0.159)	−0.158 (0.272)	−0.091 (0.528)
Intensity	0.003 (0.983)	0.167 (0.247)	0.272 (0.056)	0.203 (0.157)
Distractibility	−0.012 (0.931)	0.285 (0.045)*	0.164 (0.256)	0.239 (0.095)
Rhythmicity	−0.219 (0.126)	−0.278 (0.051)	−0.113 (0.434)	−0.289 (0.042)*

ADHD – Attention deficit hyperactivity disorder. *<0.05, **<0.01, ***<0.001

Contribution of temperament to variance in conners parent rating scale-revised: Short form subscale and attention deficit hyperactivity disorder index scores in the siblings

To explain the variance in ADHD symptoms due to temperament, stepwise linear regression analysis was computed with ADHD subscales of CPRS-R: S as dependent variables and temperamental dimension of persistence as independent variable. The adjusted R square (here indicated as r2) indicates the percentage of the variance of all explained by a variable or a set of variables. The results of regression analysis are shown in Table 4. Persistence explained 7.4% to 21.0% of the variance of all subscales of CPRS-R: S. Persistence and distractibility together explained 23.2% of Inattention scores.

Table 4

Regression analysis: Variance of Conners Parent Rating Scale-Revised: Short form subscales explained by temperament measurement schedule dimensions

Dependent variable (CPRS-R: S)	Predictor variable (TMS)	Standardized beta coefficient	Adjusted R2	F	Significance
Oppositional	Persistence	−0.305	0.074	4.916	0.031
Hyperactivity	Persistence	−0.366	0.116	7.403	0.009
Inattention	Persistence	−0.428	0.232	8.387	0.001
	Distractibility	0.323
	Persistence	−0.259	0.142	9.115	0.004
ADHD Index	Persistence	−0.476	0.210	14.044	0.000

CPRS-R –S – Conners Parent Rating Scale-Revised – Short form; TMS – Temperament measurement schedule; ADHD – Attention deficit hyperactivity disorder. There were two sets of predictor variables for inattention. Persistence and distractibility as one set and Persistence alone as another set

DISCUSSION AND CONCLUSIONS

Although the association between temperament and ADHD is well established,[2021222324252627282930] the research on temperament in first-degree relatives especially siblings of children with ADHD is lacking.[3536] Such studies are essential to understand if temperament may be a marker in the etiological process of ADHD.

Our study contributes to this much-needed line of inquiry by comparing the temperamental profiles of healthy siblings and the ADHD probands; and further, by exploring the contribution of such temperamental traits to ADHD symptom profile in the healthy siblings.

The healthy siblings were comparable to the ADHD probands in terms of age and intellectual capacity. However, our groups differed significantly on gender distribution, possibly because ADHD is commoner in males.[234] Despite this, there was no significant difference in TMS or CPRS scores between male and female children on sub-group analysis. We found that the healthy siblings scored significantly high on temperamental trait of persistence and low on activity level, adaptability, and approach as compared to their counterparts with ADHD. This profile differs from the “high maintenance” temperament profile of high activity, negative reactivity, and low task persistence[212327] classically associated with ADHD. Interestingly, the healthy siblings in our study had only slightly less distractibility; this difference being nonsignificant. Persistence refers to continuation of an activity in the face of obstacles to the maintenance of the activity direction,[14] and Distractibility is the effectiveness of extraneous environmental stimuli in interfering with or in altering the direction of ongoing behavior.[14] Hence, our healthy siblings had almost same degree of Distractibility, but with higher Persistence, they could again come back to the task and maintain the activity direction. Persistence is similar to Effortful control in Rothbart and Bates model,[18] that describes the ability to suppress a dominant or automatic response, persist with the nondominant response, pay attention, detect errors and plan for future. Both persistence and effortful control reflect the regulatory aspect of temperament, as opposed to reactive aspects.[141946] These temperamental traits are considered to correspond to the executive function of response inhibition, which is the ability or capacity to inhibit the dominant response and persist with the nondominant response ADHD.[32] The pathway models have also emphasized the role of regulatory characteristics of temperament, specifically, effortful control in the development of ADHD.[101112]

Further, we found that the persistence trait had significant negative correlations with all subscales of CPRS-R: S with correlation coefficients ranging from −0.305 to −0.476. It is noteworthy that the correlations are at best modest, hence suggesting that ADHD symptoms and temperamental traits are two distinct constructs.[12] The temperamental trait of Persistence explained 21% of variance of the ADHD index, 11.6% of variance of hyperactivity and 14.2% of variance of inattention in the healthy siblings. Goldsmith et al.,[47] reported biometric model fitting on longitudinal twin data. They found that both genetic and environmental sources of variance in effortful control accounted for variance in later ADHD symptoms. Furthermore, all the genetic variance in later ADHD symptoms was in common with the genetic variance from earlier effortful control. Our finding in healthy siblings of clinically diagnosed cases is in keeping with the observation of Goldsmith.[47] The authors proposed that temperament presents as a liability to childhood psychopathology.

Taking the main findings of the present study, i.e., healthy siblings have higher persistence, significant negative association with all the fours subscales of CPRS-R: S and that persistence explained up to 21% of the variance in ADHD symptoms in the healthy siblings. Persistence appears to be the trait that confers vulnerability or protection. In other words, higher persistence in siblings appears to offer protection to these at-risk (genetic vulnerability/first-degree relatives) individuals who do not have ADHD. Our findings further support the dual pathway model proposed by Nigg et al.[12]

The study has certain limitations and the findings must be interpreted in the light of these limitations. The sample size was small and therefore, there are higher chances of type 1 error. Second, the patient group was not homogenous and had comorbid disorders which can potentially confound the reporting by parents. In addition, temperament was assessed retrospectively which can introduce recall bias. A prospective study design in at-risk individuals can overcome this limitation. Parents were not assessed for any psychopathology including neurodevelopmental disorders which can have a bearing on the child’s temperament. Further, a comparison with a healthy control group was not done, which limit the interpretation of results.

Overall, the preliminary findings of the index study suggest that in healthy siblings of ADHD, persistence, as a temperamental trait emerged as a protective factor against the development of ADHD. However, future research with larger sample size, homogenous sample of ADHD, and a healthy control group may shed further light on the association between temperamental traits and symptoms of ADHD.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.