Study setting {9}
Patients will be recruited at the child psychiatry and neuropediatric units of Montpellier University Hospital. Information leaflets are distributed by mental health professionals to families during consultations in the outpatient services. Full study information is given to parents and children by the investigator before the signature of the informed consent.
Eligibility criteria {10}
Inclusion criteria:
Inclusion criteria are: 1) Child/adolescent aged between 7 and 13 years; 2) diagnosis of ADHD (based on the DSM-5 diagnostic criteria) assessed with a structured clinical interview (Kiddie-Schedule for Affective Disorders and Schizophrenia Present and Lifetime version; K-SADS-PL), with a supplement to assess the presence of Severe Mood Dysregulation; 3) a score ≥180 at the CBCL-A-A-A when combining the “Aggression”, “Anxiety/Depression” and “Attention” subscales; 4) child/adolescent followed at Montpellier University Hospital; and 5) families covered by the national healthcare insurance.
Exclusion criteria:
Exclusion criteria are: 1) child with delayed development or severe language disorder; 2) parents who do not speak French; 3) lack of informed consent and assent by the parents and child for their participation in the study; and 4) child not living with at least one parent.
Study exit criteria: 1) participants lost to follow-up; and 2) consent withdrawal by the parents. Each study exit will be explained and described in detail.
Who will take the informed consent? {26a}
Informed consent and assent are obtained from the parents and children, respectively, before starting any trial-specific procedure. All participants are advised that participation in research is entirely voluntary, and that they can withdraw their participation at any time. Families are not paid for their participation. The child’s psychiatrist or pediatrician does the first presentation of the study. The protocol is then explained in detail by the clinical investigator before the signature of the informed consent by parents and child.
Interventions
Explanation for the choice of comparators {6b}
In this trial, two presumably active groups are compared: the CBT group based on the program “How to improve anger and frustration management” (« Mieux gérer sa colère et ses frustrations », Massé & al., 2012), and the TBI group in which a theater-based group activity is proposed to develop verbal and reciprocal communication, theory of mind, self-awareness of emotions and interpersonal trust (Glass & al., 2000; Corbett & al., 2017; Felsman & al., 2019). TBI improves social skills, socio-cognitive functioning, and social interactions in children (Corbett et al., 2016). TBI also improves self-esteem and shows medium effects sizes for anxiety and depression reduction (Krueger & al., 2019).
TBI was chosen as active comparison group, because the objective of our study is to evaluate the primary and secondary outcomes in two potentially effective groups, while controlling for non-specific effects. In both groups, conditions are similar: duration and number of sessions, group location and setting, recruitment procedures. Differences concern only on the program content (CBT versus TBI).
Intervention description {11a}
Cognitive Behavioral Therapy group:
The CBT group is based on the manualized program “How to improve anger and frustration management” (“Mieux gérer sa colère et ses frustrations”, Massé & al., 2012). The program consists of 1-hour sessions (one per week; 15 in total) led by a psychologist and a caregiver (nurse or educator) both trained in the management of behavioral disorders and ADHD. They also have in-depth knowledge of CBT that they practice regularly.
During the program, children learn to identify situations that cause anger and frustration, but also techniques to manage negative emotions. Each session is organized as follows: reminder of the previous session by correction of the task done at home, presentation of the theme of the session with practice in the form of role-plays or exercises, presentation of inter-session tasks, and session evaluation. Children evaluate each session on the following criteria: interest and pleasure in the proposed activities, quality of the documents, ease and usefulness of the session in everyday life. Children must achieve a behavioral goal during the session. This challenge is individualized and focuses on each child’s specific difficulties. At the end of the session, the child evaluates whether he has achieved the goal. If the challenge is reached, he/she obtains a reward. Children learn that every effort deserves a reward. Several studies have shown that children with ADHD are particularly sensitive to the reward system and have emotional aversion to delay (Banaschewski & al., 2012).
During the program, children learn how to identify signals of anger at the body level, triggering situations, behaviors, and thoughts that appear when they are angry. Children also learn ways to diminish angry outbursts, to prevent their increase (e.g. through relaxation, activity practice, thinking about something else), and to create a repertory of solutions to solve problems. They learn to interpret social situations and the importance of thoughts in triggering and maintaining anger. During the final sessions, children learn techniques to peacefully solve conflicts with others and to deal with the anger of others, whether or not this outburst is justified.
Theater-based Intervention (TBI):
TBI is a role-play activity that is part of our day-care activity schedule. The TBI group participates in 1-hour sessions (one per week; 15 in total) led by two mental health workers (nurse or special educator) trained in the management of behavioral disorders, ADHD, and experienced in theater-play. The TBI sessions are not based on a specific program, and their content was developed from existing acting techniques already used in psychiatry (Héril and Mégrier, 2005; Alix and Renard, 2015).
The program objectives are to develop self-control and self-confidence skills, pro-social skills, and appropriate expression of emotions through role-play exercises. During the program, children are also trained in other competences, such as impulsivity control.
Each session includes structured and tailored activities:
- A warm-up phase around active group exercises: children are in motion, while learning how to regulate their impulsivity;
- An individual phase with exercises of improvisation, expression of emotions, and memory;
- A phase of game in pairs that includes several exercise types: role-playing around everyday life, games where children learn how to use pro-social skills and how to work in teams.
- A fun and collective play time that lasts 5 minutes. This time allows children to let off steam.
- A quiet time: at the session end, children enjoy a time of relaxation or meditation. This moment allows children to refocus on themselves and to learn techniques to manage their emotions. These exercises help to increase physical self-control that can promote better impulse control in children with ADHD (Glass & al., 2000).
During each session, the program includes sensory games, movement/space games, imitation activities, self-control games, and emotion expression activities.
Each session is adapted according to the individual needs of the children in the group and their age.
Parents Group:
The parent program is the same in both groups because this study objective is to compare the two programs for children and to evaluate their effectiveness on aggressivity in the context of ADHD with ED. Parents’ sessions take place at the same time as the children’s sessions, for convenience.
The parent group is based on a parent training program in which parents are taught behavioral and cognitive-behavioral techniques to effectively manage their child’s behavioral problems (Zwi et al., 2011). There are many manual programs, such as Webster-Stratton’s Incredible Years program (Webster-Stratton, 2006) and “How to improve living with ADHD at home” (“Mieux vivre le TDAH à la maison”, Massé & al., 2011). Previous studies have shown that parent management training improves childhood behavior problems, allows the development of positive parenting competences, and reduces parental stress (Furlong & al., 2012). Parents learn techniques to manage their child’s tantrums through videos and information sheets based on the experiences described by other parents. One session was added to this program on psychoeducation about ADHD and ED (session 1). In sessions 2 and 3, parents learn techniques on how to reinforce positive behaviors in their child (e.g. token system, special moment, encouragement…). Often parents of children with ADHD and ED tend to have conflictual and negative relationships with their own child and put in place coercive systems. In sessions 4 and 5, parents receive psychoeducation on anger (e.g. how to identify signs of anger, triggers) and learn several techniques for outburst management (e.g. to ignore minor disruptive behavior, to encourage emotion verbalization by their child). In session 6, parents learn problem-solving techniques, and techniques that allow parents and child to calmly discuss about conflicts and find solutions that are acceptable to both of them. In session 7, parents are taught several techniques to help their child manage their frustration. In session 8, a parent is invited to share a problem encountered in their daily life with their child. The other parents in the group can ask questions to analyze the problem. Then, solutions are proposed, and a solution that seems effective is tried at home. This session allows putting into practice the techniques learnt during the program.
Each session lasts approximately one hour and it’s carried out every second week. Between sessions, parents are asked to implement the acquired techniques in their everyday life. At the next session, sometime is dedicated to discuss their implementation and the difficulties encountered in order to adjust the techniques to the individual needs.
Criteria for discontinuing or modifying the allocated interventions {11b}
Criteria for discontinuing the allocated interventions are: 1) disruptive behavior disorder that does not allow the child to benefit from the intervention; 2) behavior not stabilized by drugs for children under treatment; and 3) aggressive behavior of one participant toward the others.
The protocol does not allow modifying the allocated intervention.
Strategies to improve adherence to interventions {11c}
To improve adherence to the intervention protocols, behavioral goals are defined to increase motivation. These challenges are individualized and focus on each child’s specific difficulties. At the session end, children must evaluate whether they have achieved their challenge. If the challenge is reached, they obtain a reward. Children learn that every effort deserves a reward.
During the study, participants will be contacted by telephone if absent during two consecutive sessions to know the reasons of their absence. Between sessions, caregivers are available by telephone or email to answer any questions from the families.
Relevant concomitant care permitted or prohibited during the trial {11d}
During the interventions, pharmacological treatments are permitted. Moreover, pharmacological treatments can be modified if they are no longer appropriate. It is possible to begin a drug treatment during the trial if behavioral symptoms become too severe and have a negative impact on the child’s participation and implication in the group. The maintenance of other usual care components is authorized.
It is recommended that parents do not participate in other parental management training programs, because of similarities between the interventions proposed in the trial and those usually offered by care services.
Provisions for post-trial care {30}
A participant may exit the trial for the following reasons: severe behavioral problems not sufficiently stabilized to allow the participant to benefit from the program, strong opposition to participate in the group, and inadequate behaviors (physical or verbal aggression against other participants or caregivers) that interfere with the proper functioning of the group. When a child behaves inappropriately (opposition, verbal or physical aggression, provocation…) during a session, a discussion time is offered with the child, parents and caregivers. The objective is to find a solution together to solve the problem. This exchange may be enough to motivate the participant and stop the inappropriate behavior. Concomitantly, the child’s psychiatrist will be contacted and informed of the situation. He/she might decide to meet the child to modify the pharmacological treatment and to re-motivate him/her. Nevertheless, if the problematic behaviors persist and affect the group, the child exit from the program is decided together with the child, the family and the psychiatrist.
Outcomes {12}
Primary and secondary outcomes will be evaluated in the two groups (TBI and CBT) at the inclusion visit (baseline), at the intervention end, and at month 6 after the end of the intervention. The primary judgment criterion is the variation of the “Aggressive behaviors” score of the CBCL between baseline and month 6 post-intervention. The initial evaluation will be on outcomes at month 6 post-intervention; because we hypothesize that the effectiveness of such programs is observed several months after their end. Usually, it takes time for children and parents to assimilate and routinely implement the new techniques and also to observe the intervention benefits in the daily life. This will allow also assessing whether the beneficial effects of the interventions are maintained in the long term.
Secondary judgement criteria are the score variation between baseline at the intervention end of the CBCL-A-A-A, CBCL (total score), Kidscreen-27, Parental Quality of Life and Developmental Disorder (Par-DD-Qol), Children’s Global Assessment Scale (C-GAS), Parenting Stress Index-Short Form, 4th edition (PSI-4-SF), Beck Depression Inventory-II (BDI-II), and Strengths and Difficulties Questionnaire (SDQ) and the SDQ-dysregulation profile (SDQ-DP). Secondary judgement criteria include also the variation between baseline and month 6 post-intervention of the CBCL-A-A-A score, total CBCL score, Kidscreen-27 scores, Par-DD-Qol scores, C-GAS score, BRIEF scores, PSI-4-SF scores, BDI-II score, and SDQ and the SDQ-DP scores.
Other data will be collected during the initial visit: presence or absence of psychiatric disorders in children using the K-SADS-PL diagnostic tool, socio-demographic data, anamnestic data, and medication history. Pharmacological treatments will be recorded at each visit because the introduction of a new drug or the modification of posology may influence the research outcomes. Indeed, ADHD usual treatment effectively reduces ED (Tourian & al., 2014; Fernandez de la Cruz & al., 2015; Kultu & al., 2017; Gamli & al., 2018; Winters & al., 2018). The Hierarchical Personality Inventory for Children (HiPIC) will be completed at the end of the intervention to limit the duration of the initial visit.
Participant timeline {13}
Outcome measures are collected at pretreatment (T1), just after treatment end (T2), and at month 6 post-treatment (T3) (Table 1). Data for the primary and secondary outcomes will be collected at child psychiatry unit of Montpellier University Hospital (MPEA Saint Eloi). The child presence is not required for all the measures collected during these assessments.
Pre-treatment visit (T1):
Participants and their parents will participate in a visit to determine their eligibility (Enrollment, Table 1). Then, the T1 visit takes approximately 2 hours, because the diagnostic interview (K-SADS-PL) is carried out systematically with the parents to confirm the diagnosis of ADHD and the presence of comorbidities. During this visit, parents are interviewed by clinicians for clinical ratings of their child’s overall functioning impairment (C-GAS). Children and parents will also complete self-report questionnaires (see Table 1). A questionnaire (SDQ-Teacher) is completed also by the child’s teacher. This questionnaire is either given to parents who will directly transmit it to the teacher, or sent by mail.
Visit at treatment end (T2):
After the last session, questionnaires are sent to the parents for completing them at home and bring them back at the T2 visit. Children and parents will complete the same self-report questionnaires as at T1, but for the HiPIC and the satisfaction questionnaires (only at T2), and BRIEF (only at T1) (Table 1). During the T2 visit, parents are interviewed by a clinician to assess the child’s overall functioning (C-GAS). This visit does not exceed 30 minutes. If needed, questionnaires can also be completed with the help of a researcher. The SDQ-Teacher questionnaire is again completed by the child’s teacher
Visit at month 6 post-treatment (T3):
This follow-up visit determines whether the intervention effects persist at month 6 after the intervention end. Children and parents will be interviewed by a clinician to complete the C-GAS. Before the visit, they will complete the same self-report questionnaires as at T1 (at home) (Table 1). This visit takes approximately 30 minutes.
Table 1. Visits and data acquisition during the trial
|
Study period
|
TIMEPOINT
|
Enrollment
|
Before the intervention (T1)
|
At the intervention end (T2)
|
Six months after the intervention (T3)
|
Enrollment:
|
Eligibility criteria
|
X
|
|
|
|
Child Behavior Checklist (CBCL)
|
X
|
|
|
|
Informed consent
|
X
|
|
|
|
Randomization
|
X
|
|
|
|
Assessments:
|
Structured clinical interview (K-SADS-PL)
|
|
X
|
|
|
Parent-rated measures
|
Child Behavior Checklist (CBCL)
|
|
X
|
X
|
X
|
Strengths and Difficulties Questionnaire (SDQ-Par)
|
|
X
|
X
|
X
|
Behavioral Rating Inventory of Executive Function (BRIEF)
|
|
X
|
|
X
|
Hierarchical Personality Inventory for Children (HiPIC)
|
|
|
X
|
|
Kidscreen-27
|
|
X
|
X
|
X
|
Parental Quality of Life and Developmental disorder in their children (Par-DD-Qol)
|
|
X
|
X
|
X
|
Parenting Stress Index 4th edition (PSI-4-SF)
|
|
X
|
X
|
X
|
Beck Depression Inventory-II (BDI-II)
|
|
X
|
X
|
X
|
Child-rated measures
|
Kidscreen-27
|
|
X
|
X
|
X
|
Teacher-rated measures
|
Strengths and Difficulties Questionnaire (SDQ-Teacher)
|
|
X
|
X
|
X
|
Clinician-rated measures
|
Children’s Global Assessment Scale (C-GAS)
|
|
X
|
X
|
X
|
Satisfaction questionnaires:
|
|
|
|
|
Parent-rated measures
|
Satisfaction questionnaire on the parent program
|
|
|
X
|
|
Satisfaction questionnaire concerning the child program
|
|
|
X
|
|
Child-rated measures
|
Satisfaction questionnaire on the child program
|
|
|
X
|
|
Sample size {14}
As ED in children with ADHD has been only recently studied, it is difficult to find literature data for calculating the number of subjects required (particularly for assessing the changes in the “Aggressive behaviors”, “Anxiety/Depression” and “Attentional Problems” scores of the CBCL scale).
The study by Masi and colleagues (2015) reported a change in the “Aggressive Behavior” score of the CBCL from 68.3 to 65.5 in the context of the non-comparative evaluation of an intervention similar to the one assessed in the present trial. The standard deviation of the score at the baseline was 5.4.
By making the reasonable assumption of a slightly more effective intervention than what reported by Masi G. et al., (2015), a CBCL score of 65 in the intervention group and 69 in the control group, with a common standard deviation of 5.5, can be expected at month 6 post-intervention. To highlight such a difference with a power of 80% and an alpha risk of 5%, 31 subjects per group need to be included. By increasing this number by 10% to take into account participants lost to follow-up, the total number of subjects to be included in the study is 68 (34 per group).
Recruitment {15}
Patients will be recruited at the Child Psychiatry and Neuropediatric units of Montpellier University Hospital. Information leaflets will be distributed by psychiatrists and may be posted electronically to families. Researchers will give presentations on this trial to the general public and mental health professionals.
Assignment of interventions: allocation
Sequence generation {16a}
After verification of the eligibility criteria, family’s information and consent signature, each participant will be randomized. Intervention allocation will be done using a computer-based random number generator. Randomization will be performed as soon as the number of participants is reached to start a session (one CBT group and one TBI group) with a 1:1 ratio.
Randomization will be stratified by school level (primary vs middle school) and severity of psychopathological disorders using the CBCL-A-A-A score. Children will be divided in groups according to their school level (<11 years, and 11-13 years of age).
The investigator in charge of the study assessments is blinded to the group allocation.
Concealment mechanism {16b}
Randomization is done by the Department of Medical Information of Montpellier University Hospital using the Capture System software (Clinsight).
Implementation {16c}
If participants are interested in the trial, psychiatrists or pediatricians give information and enroll them. Then, participants are contacted by telephone by the investigator. The psychologist informs about the study, verifies that families are aware of the implications of participation in a trial, and proposes an appointment for the initial evaluation. She verifies that children are motivated and willing to participate in the trial. Indeed, often, parents register their child without asking their opinion, while it is important that they are involved and motivated in order to benefit from the intervention. After the initial evaluation, participants are assigned to an intervention group by the research coordinator after randomization.
Assignment of interventions: Blinding
Who will be blinded {17a}
The research psychologist who performs the baseline and outcome assessments will be blind to the treatment conditions. Due to the nature of the interventions, it will not be possible for the researchers who administer the intervention to remain blind. Researchers who administer the intervention will not be involved in the collection of the outcome data to avoid bias in measuring results.
Given the study design, parents and children are not blind and are informed about the group allocation after randomization.
Procedure for unblinding if needed {17b}
There is no procedure for revealing a participant’s allocated intervention during the trial, because it is not a protocol targeting a pharmacological treatment. Therefore, it did not seem necessary to provide circumstances for and an unblinding protocol.
Data collection and management
Plans for assessment and collection of outcome data {18a}
Outcome data will be collected by a research psychologist. Data will be verified by supervised psychology students to detect errors in the collected data. Finally, data will be verified by a methodologist who will process all statistical data.
Primary outcome
Child Behavior Checklist (CBCL; Achenbach & Rescorla, 2001):
The CBCL is a 118-item parent-completed questionnaire to measure emotional and behavior problems in 4 to 18-year-old children in the past six months. Parents rate each item as 0 (not true), 1 (somewhat or sometimes true), or 2 (very true or often true). The CBCL includes eight syndrome scores: Anxious/Depressed, Withdraw/Depressed, Somatic Problems, Social Problems, Thought Problems, Attention Problems, Rule Breaking, and Aggressive Behavior. A computer program calculates the T-scores for each scale. Raw scores are converted to gender- and age-standardized scores. A T-score of 50 indicates average functioning and standard deviation is every 10 points.
The objective of the primary outcome is to assess the effectiveness of the CBT program on the “Aggressive behavior” score (i.e. primary outcome) of the CBCL-A-A-A. The CBT program should allow reducing aggression.
The “Aggressive behavior” sub-score contributes to the CBCL-DP and “Deficient Emotional Self-Regulation” (CBCL-DESR). The two profiles are the sum of the CBCL “Anxious/Depressed”, “Attention Problems” and “Aggressive Behavior” (A-A-A) syndrome T-scores (Achenbach and Rescorla, 2001). The CBCL-DESR is present when the sum of the CBCL A-A-A scores is ≥ 180, but below 210 (T-scores >60). The “Dysregulation Profile” is present when the sum of the CBCL A-A-A scores is ≥ 210 (2 SD; T-scores >70). These profiles represent a continuum of the emotional and behavioral dysregulation severity (Biederman & al., 2012; Masi & al., 2015). They allow defining an ED phenotype and are a risk marker of a persisting deficit of self-regulation of emotion and behavior (Holtmann & al., 2011). CBCL-DP and CBCL-DESR allow identifying a group of children with disruptive behavior (Aitken et al., 2019) and at risk of severe dysfunction (Spencer & al., 2011; Holtmann & al., 2011; Elmaghrabi & al., 2020). The CBCL-DP and CBCL-DESR profiles are associated with higher risk of comorbidities (Holtmann & al., 2011; De Caluwé & al., 2013; Masi & al., 2015), hospitalization in psychiatry services (De Caluwé & al., 2013), and development of inappropriate personality traits (Peyre & al., 2015) in adulthood. The CBCL-DESR profile is a predictor of increased risk of opposition defiant disorder, conduct disorder, and anxiety disorders (Spencer & al., 2011). The CBCL-DP predicts high risk of mood disorders during adolescence (Masi & al., 2015).
Secondary outcomes
The trial includes several secondary outcome measures, and also the clinician ratings of impairment (C-GAS). Moreover, parents complete questionnaires about their child to evaluate:
- Global functioning (SDQ, C-GAS)
- Executive functions (BRIEF)
- Personality traits (HiPIC)
- Quality of life (Kidscreen-27).
Parents also complete questionnaires to evaluate:
- Parental stress (PSI-4-SF)
- Parental depression (BDI-II)
- The impact of their child disorders on the quality of family life (Par-DD-Qol).
Secondary outcomes include also teacher-report measures of the child’s global functioning (SDQ-Teacher), and child-report measure of the quality of life (Kidscreen-27).
The study objectives are to evaluate the CBT program impact on the socio-communicative capacities, emotional and behavioral self-regulation capacities, quality of life of child and family, overall functioning of child, and parental stress/depression in the short term (intervention end) and also at month 6 post-intervention. Finally, the personality profiles of children with ADHD will be analyzed to determine whether there are common personality traits in children with ADHD and ED.
Outcome measures
French version of the Strengths and Difficulties Questionnaire (SDQ-Fr, Goodman, 1997):
SDQ-Fr is a standardized questionnaire completed by parents or teachers that allows a brief emotional and behavioral screening of 2 to 17-year-old children and adolescents (http://www.sdqinfo.com). The SDQ-Fr includes 25 items divided in 5 subscales with 5 items/each to assess emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems, and prosocial behavior. The questionnaire is completed on paper in 5-10 minutes.
The SDQ-Fr allows obtaining a “dysregulation profile” (SDQ-DP). It is a combination of two items from the Emotional symptom subscale (“Many worries, often seems worried”, “Often unhappy, down-hearted or tearful”), two from the Conduct problem subscale (“Picked on or bullied by other children”, “Steals from home, school or elsewhere”), and one from the Hyperactivity/Inattention subscale (“Restless, overactive, cannot stay still for long”). A study suggested that children with ADHD and SDQ-DP display more angry reactions and are less capable to control their anger (Caro-Canizares & al., 2017). The SDQ-DP allows identifying children with ADHD at risk of severe difficulties in overall functioning. A study showed that SDQ-DP is a predictor of developing psychiatric disorders in the next 4 years (Wang & al., 2017).
Kidscreen-27 (KIDSCREEN Group, 2006):
Kidscreen-27 is a questionnaire that evaluates the generic health-related quality of life (HRQol) in 6 to 18-year-old children. The Kidscreen-27 items are derived from the Kidscreen-52 questionnaire. There are two versions, one completed by parents (or primary caregivers) and the other by children. Children evaluate the subjective perception of their wellbeing during the last week. Kidscreen-27 comprises five domains: “Physical well-being”, “Psychological well-being”, “Autonomy and parents”, “Social support and peers”, and “School environment”. The responses are scored on 5-point Likert scales for frequency (never to always), or intensity (not at all to extremely). Kidscreen-27 has been translated into 36 languages. Kidscreen-27 scores are converted to T-values (standardized mean=50, standard deviation= 10). Higher scores indicate good HRQol and well-being (Ravens-Sieberer & al., 2007).
Parental Quality of Life and Developmental Disorder in their children (Par-DD-Qol; Berdeaux, 1998):
The Par-DD-Qol is a parent-completed 17–item questionnaire to assess the impact of their child’s chronic disabilities on the parental quality of life (Qol). It is particularly adapted to parents of children with neurodevelopmental disorders, because these are chronic conditions that impact the parents’ Qol. The questionnaire includes several dimensions: “Emotional”, “Daily Disturbance” and “Global Qol”. It is adapted from the Par-ENT-Qol (Berdeaux & al., 1998) used in the general population with chronic ear, nose and throat (ENT) infections. Responses are scored on a 5-point Likert scale (not at all to very much). Several studies (Baghdadli & al., 2014; Raysse, 2011) showed a reliability coefficient higher than 0.82 (Cronbach’s alpha) for each dimension. A score below 40 indicates “no impact” on the parents’ Qol, a score between 40 and 57 a “moderate impact”, and a score higher than 57 a “high impact”.
Behavioral Rating Inventory of Executive Function (BRIEF, Gioa, Guy & Kentworthy, 2000):
BRIEF is a parent- or a teacher-completed 86-item inventory to assess the executive functioning of 5 to 18-year-old children at home and in school environments using a three-point scale for frequency (never to always). The instrument includes eight scales that measure executive functioning: initiate, work memory, plan/organize, organization of materials, and monitor (forming the metacognition index [MI]), and inhibit, shift and emotional control (forming the behavioral regulation index [BRI]). The BRI assesses the ability to use appropriate inhibitory control to shift cognitive sets and modulate emotions and behaviors. The MI assesses the ability to use working memory to initiate, plan and sustain future-oriented problem solving (McCandless & al., 2007). BRIEF includes two validity scales (Negativity and Inconsistency of responses) to identify the parents’ response styles and to validate the questionnaire quality. A T-score of 65 indicates clinically significant executive function impairment. Gioia et al., (2000) reported that the BRIEF reliability coefficient ranges from 0.80 to 0.97 for the two forms (Parent and Teacher), except for two scales (“Initiate” and “Shift”).
Parenting Stress Index, 4th edition, short form (PSI-4-SF, Abidin, 1983):
The PSI-4-SF is a 36-item parent-completed questionnaire to measure the parental stress and to detect difficulties in the parent-child dyad. This is defined as a state of psychological malaise in parents related to the parent-child relationship. Responses are provided on a 5-point Likert scale (totally disagree to totally disagree). The questionnaire includes three sub-scales (“Parental distress”, “Dysfunction in parent-child interactions”, and “Difficulties in children”) to assess the factors that may influence the level of stress experienced by parents in their relationship with their child. The first subscale “Parental distress” allows measuring the distress experienced by parents in the exercise of their role. The second subscale “Dysfunction in parent-child interactions” measures the parents’ satisfaction of their relationship with their child and whether the child meets their expectations. The third subscale “Difficulties in children" assesses the parents’ degree of distress due to their child’s difficult behavior. A score between the 85th and 89th percentile indicates high stress level, and a score ≥90th percentile is clinically significant. Abidin (2012) reported a reliability coefficient of 0.84 (test-retest, 6 months) and internal consistency coefficient of 0.95 (Cronbach’s alpha) for the total score. The three subscales have internal consistency coefficients of 0.90, 0.89 and 0.88, respectively. PSI-4-SF is strongly correlated with the original version of the PSI-4, with a correlation coefficient of 0.98 for the Total Stress Scale.
Beck Depression Inventory-II (BDI-II; Beck, Steer & Brown, 1996)
The BDI-II is the most popular screening instrument for depression in adolescents and adults. It is a 21-item self-report questionnaire that examines the behavioral, emotional, somatic and cognitive symptoms of depression in the past two weeks. Each item is rated on a 4-point Likert scale (ranging from 0 to 3), reflecting the symptom severity. Scores from 0 to 13 indicate “Minimal depression”, scores from 14 to 19 “Mild depression”, scores from 20 to 28 “Moderate depression”, and scores higher than 29 indicate “Severe depression”.
Wang and Gorenstein (2013) reported a mean alpha coefficient of 0.9 (ranging from 0.83 to 0.96) and excellent coefficients of retest reliability (0.73 to 0.96). The BDI-II is valid in different cultures and presents strong psychometric properties (Beck, Steer & Brown, 1996).
Hierarchical Personality Inventory for Children (HiPIC, Mervielde & De Fruyt, 1999):
HiPIC is a 144-item parent-completed questionnaire to obtain a profile of the personality of 6 to 12-year-old children according to a five-factor model. Responses are scored on a 5-point Likert scale (from “Very untypical” to “Very typical”). Five personality dimensions are assessed: “Emotional Stability”, “Extraversion”, “Benevolence”, “Conscientiousness”, and “Imagination”. HiPIC is scored on 18 facets composed of 8 items/each that are summed and averaged to obtain the dimensions. The “Extraversion” dimension includes positive emotionality, energy in children, and ease in social situations. The “Emotional stability” dimension assesses negative emotions and reactions towards their environment. The “Conscientiousness” dimension measures the determination and drive to achieve a goal. The “Benevolence” dimension evaluates agreeableness, attitude in relationships, and ability to empathize. The “Imagination” dimension represents creativity, curiosity, and openness to new experiences. Mervielde and De Fruyt (1999) reported a reliability coefficient of 0.70 (Cronbach’s alphas) for each domain and facet.
Children’s Global Assessment Scale (C-GAS, Shaffer & al., 1983):
C-GAS is a numeric scale to assess the general functioning in 4-16-year-old children. This scale is used and completed by mental health clinicians. A clinician interviews the child, parents and school staff to assess the child’s global functioning. C-GAS is adapted from the Global Assessment Scale for adults. Scores are divided in ten categories that range from “Extremely impaired” (1-10) to “Doing very well” (91-100).
Schedule for Affective Disorders and Schizophrenia for School-aged Children, Present and Lifetime version (K-SADS-PL; Kaufman & al., 1997; French version Mouren-Siméoni & al., 2002):
K-SADS-PL is a semi-structured diagnostic interview to assess all current and lifetime DSM-IV Axis I mental disorders in 6-18-year-old children. K-SADS-PL is administrated by trained mental health clinicians (master level). The additional module to measure severe mood disorders, according to the diagnostic criteria described by Leibenluft & al., (2003), will also be used. The objective is to obtain more information on ED and its impact on daily life.
Plans to promote participant retention and complete follow-up {18b}
Evaluation visits will be scheduled at the intervention end (T2) and at month 6 post-intervention (T3) (see Table 1). Two weeks before the intervention end, the psychologist will propose to parents an appointment for the T2 visit. For the appointment of the follow-up visit (T3), families will be contacted by telephone by the psychologist. For families difficult to reach, several telephone calls may be made. The child’s referring psychiatrist will be also contacted to improve data collection (e.g. he/she might call the family to emphasize the importance of completing the questionnaires, refer family to the psychologist after consultation…).
For families unable to attend the visit (health problems…), it might be exceptionally decided to complete the questionnaires and the Case Report Form (CRF) during a telephone conversation. This will be indicated in the CRF.
If parents decide to interrupt their trial participation, the investigator will contact them to know whether they agree to complete the final questionnaires.
To promote participant retention during the program, families will be contacted by telephone in case of absence during two sessions.
Data management {19}
For each participant, data will be reported in a CRF, first in paper format and then electronically. The used software, Capture System, complies with the FDA recommendations on computerized systems for managing clinical trials.
The data manager will perform additional computerized consistency tests to detect the presence of non-standard, missing, aberrant or incoherent data. These tests will be executed regularly during the participants’ recruitment and monitoring. Each identified incoherence will be the subject of a request for clarification to the researcher.
Data will be saved by the Clinical Research and Epidemiology Unit (CREU) of Montpellier Hospital and will be stored in ASCII type format.
Confidentiality {27}
Data will be collected in the CRF only by the research psychologist. Each CRF will be anonymized to respect the participants’ confidentiality.
Statistical methods
Statistical methods for the primary and secondary outcomes {20a}
An intention-to-treat approach will be used to analyze the primary and secondary data from the trial. Participants will be analyzed in their randomization arm and cannot change group during the study.
The primary outcome measure is the changes of the “Aggressive behavior” sub-score between T1 (baseline) and T3 (month 6 post-intervention). A Z test will be used to compare the changes of the primary (between T1 and T3) and secondary outcomes (between T1 and T2/T3) in the CBT group and TBI group. In the case of group non-comparability, a multivariate model by linear regression could be carried out to take into account the potential confounding bias.
Interim analyses {21b}
In this trial, statistical analyses of data will be realized when the statistician will have all data (i.e. when the number of subjects required will be reached and when all visits will have been carried out).
The final decision to finish the trial will be taken by mutual agreement of the principal investigator, the data manager, the research coordinator, and the independent safety monitor. Data will be locked when all data have been checked and all corrections done.
Methods for additional analyses (e.g. subgroup analyses) {20b}
In the case of non-comparability for one or more parameters, an adjustment will be made on this/these parameter(s) for analysis of the judgment criteria. A multivariate linear regression model might be performed to account for potential confounders.
The protocol does not plan analyses for additional subgroups.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}
Analysis of the primary and secondary outcomes will be carried out per protocol for all families who participated in two thirds of the intervention.
For missing data, under the hypothesis of a Missing at Random (MAR) or Missing Completely at Random (MCAR) mechanism, an imputation method using the Markov chain Monte-Carlo method will be used with 15 imputation cycles. The analysis will be performed independently for each complete database and the results will be taken into account for estimating the final parameters and their standard deviations.
Plans to give access to the full protocol, participant level-data, and statistical code {31c}
On request, participants will be informed about the trial overall results by the principal investigator. Families will not have access to personal data.
Oversight and monitoring
Composition of the data monitoring committee, its role and reporting structure {21a}
Data monitoring will be carried out by an independent safety monitor who will ensure the compliance with the trial regulatory aspects and will verify data collection. The safety monitor acts as a representative of the study promotor.
Adverse event reporting and harms {22}
Due to the intervention types, serious adverse events are not expected. Serious adverse events will be reported by the promotor only if they are directly related to the protocol. Each adverse event will be recorded in the CRF by the principal investigator and will be monitored until resolution or stabilization. The principal investigator will evaluate each event and its gravity and will contact the trial promotor.
Frequency and plans for auditing trial conduct {23}
Investigators accept to comply with the regulatory requirements of the promoter and the competent authority for a research audit. Audit may be carried out at any stage of the trial, from the protocol development to the publication of results and archival of the data used for the study.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
All protocol modifications must be submitted to and validated by the Local Human Subject Protection Committee before their implementation. When the protocol modification is accepted by the committee, the research assistant ensures that the changes are implemented accordingly.
Dissemination plans {31a}
All communication of results must receive the prior agreement of the principal investigator and promoter. Montpellier Hospital is the data owner and must be mentioned as the trial promoter. On request, participants will be informed about the trial overall results by the principal investigator.