Study setting {9}
This multicentre research study takes place in four different clinical health care settings for youth in the Netherlands: Child and Youth Trauma Centre Kenter Jeugdhulp, GGZ Rivierduinen, UMC Utrecht and Arq.
Eligibility criteria {10}
Children are eligible to participate in the study if they meet the following inclusion criteria: a) the child has been exposed to family violence; b) the acute safety in the family has been established; c) the child is between 12 and 20 years old; d) the child lives at home with its caregiver(s); e) caregivers, living with the child can participate in the systemic components; f) the child has posttraumatic stress symptoms, at least intrusions and avoidance; g) caregivers and children master the Dutch language. Exclusion criteria for participating in the study are: a) children and caregivers have acute psychotic symptoms or severe alcohol and drug addictions; b) children with a suicide attempt in the past 3 months. In- and exclusion criteria are established with an intake-interview with a trained therapist and standardized questionnaires.
All therapists participating in the study are trained in the overall treatment program and have additional training in the part of the treatment program they will perform. Exposure therapy will be performed by therapists who have basic training in cognitive behavioural therapy. EMDR therapy will be performed by therapists who completed the advanced EMDR training (VEN, Dutch EMDR Association). The bodily-oriented therapy is offered by psychomotor therapists or therapists with experience in trauma and bodily-oriented therapy under supervision of an experienced psychomotor therapist. Parent therapy is performed by a family therapist or family worker. Together with a child therapist the family therapist/worker will also offer the family sessions.
Who will take informed consent? {26a}
Children between the age of 12 and 20 years old and referred to one of the participating trauma centres after being exposed to family violence will be given a brochure about the study, so are their parents. Subsequently, the therapist asks the family permission to share their contact details with the researcher. After approval the therapist will send the researcher the contact details by email. Within two working days the researcher will contact the family members separately to explain the study. An information letter with detailed information about the study will be sent to the families before the first intake interview at the respective centre. The families will have at least one week to consider their decision about participating in the study. If the family agrees to participate, informed consent will be sought from both parents, and children during the first intake. After inclusion criteria are met and informed consent is obtained, participating families will be randomly assigned to one of the treatment conditions by an independent researcher using a computerized randomization procedure with lottery drawings.
Additional consent provisions for collection and use of participant data and biological specimens {26b}
Not applicable.
Interventions
Explanation for the choice of comparators {6b}
Intensive trauma therapy (ITT) is the comparator of Family-based Intensive Trauma Treatment (FITT) to examine the addition of a component with parenting therapy and family sessions to an intensive trauma-focused therapy component. It is important to understand if this addition is more effective in reducing trauma symptoms, improving parenting functioning, and increasing emotional and physical security in children, through improvement of parent-child relationships – and thereby establishing long-term effects on the reduction of trauma symptoms and continuation of security in the family.
The second comparator for FITT is treatment as usual (TAU) to examine the addition of an intensive component to a low-frequent trauma-focused therapy with parenting therapy and family sessions. This comparator examines if FITT is as effective as TAU at the end of treatment and if FITT results in faster symptom reduction. In the view of adolescence as a critical period for development and peer influence (31), this faster symptom reduction is of great importance.
Intervention description {11a}
Components of ITT
ITT is a three-week intensive trauma treatment program and contains a preparation phase in week 1, and an intensive phase in week 2 and 3.
Preparation phase: psycho-education and clustering
The preparation phase consists of two treatment days. On the first treatment day the adolescent receives a 90-minute online and interactive session of psycho-education about the impact of trauma and the treatment program. Aim is to inform the adolescent about (the consequences of) family violence, trauma and PTS disorder, stress regulation, the window of tolerance and the treatment program. On the second treatment day the adolescent is offered a 90-minute session in which the different traumatic events are clustered. The adolescent and the therapist choose the six most traumatic memories, ordered by the degree to which they cause flashbacks, nightmares and avoidance behaviour and elicit the most tension. These traumatic events will be the focus of the Exposure and EMDR sessions.
Intensive phase: Exposure, EMDR and bodily-oriented therapy
In the intensive phase the adolescent has three treatment days per week. Each treatment day has a fixed program consisting of:
Exposure
The exposure and EMDR in the program will be given by different therapists. One traumatic memory each day will be treated with exposure and EMDR. The exposure therapy is based on the Dutch translation (52)(53) of Foa’s prolonged exposure protocol (54). Prolonged exposure therapy is a form of cognitive behavioural therapy in which the adolescent repeatedly describes the traumatic event in detail and present tense, focusing on the anxiety-provoking parts of the traumatic event. Every 5 to 10 minutes the therapist checks the level of disturbance experienced by the adolescent. The adolescent learns fearful expectations stay away when being exposed to the trauma and there is no need for avoidance behaviour.
Bodily-oriented therapy
Between the sessions Exposure and EMDR individual Bodily-Oriented therapy is offered by psychomotor therapists or therapists with experience in trauma and bodily-oriented therapy under supervision of an experienced psychomotor therapist. Through physical activities the therapist supports the adolescent to:
-
Enhance the understanding of the association between the body and brain
-
Enhance the understanding of the association between trauma and the body
-
Improve stress-regulation skills
-
Recognize the effects of trauma on the body
-
Calm brain and body to prepare for the EMDR session.
The therapist will work with bodily-oriented exercises to achieve these goals.
EMDR
For the EMDR session the most recent Dutch EMDR protocol for children and youth will be used and all therapists completed the basic and advanced EMDR training of the Dutch EMDR Association. During EMDR therapy the adolescent is asked to focus on the most disturbing image of the traumatic memory and the therapist asks questions about the thoughts and feelings associated with the traumatic image and the level of disturbance. The therapist will ask the adolescent to focus on the image and the associated thoughts and feelings and at the same time focus on a distracting stimulus. This will bring a stream of associations in the form of thoughts, feelings, images and bodily sensations and helps to reprocess the traumatic memory. The therapist will then ask what the adolescent notices and repeat the distracting stimulus. After a while the adolescent will notice the level of disturbance will lower until the traumatic memory is desensitized. If necessary, the therapist can add cognitive interweaves during the desensitization phase to elicit new associations. An overview of ITT is presented in table 1.
Additional components in FITT
Additions in preparation phase
Psycho-education for family and network
In FITT the session psycho-education is together with the parents and the support system. This is a 90-minute online and interactive session of psycho-education about the impact of trauma and the treatment program. Aim is to inform the adolescent and family system about (the consequences of) family violence, trauma and PTS disorder, stress regulation, the window of tolerance, trauma-sensitive parenting and resilience and the treatment program.
Additions in Intensive phase
Parent therapy
In the intensive phase in week 2 and 3 parents receive three 60-minute sessions of individual parent therapy per week from one family therapist or family worker. In dialogue different topics are discussed: parental stress, support system, trauma-sensitive parenting, trauma’s within the family. The parent therapy and EMDR take place simultaneously and after these sessions there is a short parent-child interaction moment where the adolescents share their thoughts and feelings on which the parents can practice supportive and trauma-sensitive parenting. The aim of the parent therapy is to:
-
Lower levels of parental stress
-
Increase parental support
-
Increase knowledge about trauma-sensitive parenting and increase trauma-sensitive parenting skills
-
Increase the insight that the family has a joint family story in which the different family members have their own memories and meaning about the traumatic events.
Additional integration phase
Family therapy
The integration phase consists of three 90-minute sessions of family therapy. Two therapists support the family to talk about the experienced family violence and current conflicts. The goal of family therapy is to:
-
Create a shared trauma narrative
-
Recognize destructive communicative patterns
-
Recognize emotional individual vulnerabilities and survival behaviour and their mutual effects in family relationships
-
Improve skills to prevent escalations and cope with destructive patterns.
In table 2 an overview of FITT is presented.
Components of TAU
TAU is comparable across the four trauma centres and consists of low-frequent trauma-focused therapy with parenting therapy and family sessions. In contrast to the high intensity of FITT, TAU is offered about once a week and without therapist rotation. In TAU the adolescents receive therapy from one therapist and parents receive therapy from another therapist. This individual or group-based trauma-focused cognitive behavioural therapy for adolescents and their parents is focused on the following components:
-
Psycho-education trauma and PTSD: increasing knowledge about (the consequences of) trauma and family violence and trauma-sensitive parenting, understanding the window of tolerance.
-
Affect regulation and coping: learning children to recognize and regulate feelings of stress, enhancing coping skills.
-
Trauma processing: EMDR, Exposure, writing therapy or Bodily-Oriented therapy.
-
Parenting skills and competence: enhancing trauma-sensitive parenting skills, lowering levels of parental stress, increasing social support.
-
On indication: family relationships: sharing the trauma narrative and accompanying feelings and cognitions (for example feelings of shame or guilt) and strengthen skills to prevent escalations.
Criteria for discontinuing or modifying allocated interventions {11b}
In accordance with the Medical Research Involving Human Subjects Act the sponsor will suspend the study if there is plausible ground that continuation of the study will jeopardise subject health or safety. The sponsor will then notify the accredited Medical Ethical Committee (from Amsterdam UMC) without undue delay of a temporary halt including the reason for such an action. The study will be suspended pending a further positive decision by the accredited Medical Ethical Committee. The investigator will take care that all subjects are kept informed.
Strategies to improve adherence to interventions {11c}
To increase treatment fidelity a training protocol for (F)ITT is developed and therapists receive training in the (F)ITT protocol before data collection. Furthermore, therapists receive supervision on Exposure, EMDR, Parent and Family therapy and Bodily-oriented therapy during data collection. Exposure supervision is given by a healthcare psychologist who is a licensed supervisor from the Dutch cognitive-behavioural therapy association. EMDR supervision is given by a healthcare and clinical psychologist who both are licensed supervisors from the Dutch EMDR association. The supervisor for the Parent and Family sessions is a clinical psychologist with advanced experience in the field of trauma and family therapy. The supervisor for the Bodily-oriented therapy is an experienced psychomotor therapist who works with traumatized children. To monitor adherence to the treatment protocol therapists are asked to fill in a treatment evaluation form after each therapy session.
The trial will be monitored by the independent physician of the four participating youth mental health centres. Two times per year the independent physician will randomly check the research file of a study participant on:
-
The involvement of a therapist registered in the Dutch register for healthcare professionals
-
The presence of a safety plan in cases of suicidal thoughts/ideations and/or self-damaging behaviour
-
The compliance of in and exclusion criteria (the first year the first three study participants will be checked and thereafter every half year a participant will be randomly checked on in and exclusion criteria)
-
Serious Adverse Events (SAE’s) and if these are reported to the independent physician, Inspection for Healthcare and Youth, the research institute and the Medical Ethical Committee
-
The presence of the casus conceptualisation.
If any errors are observed, the files of all study participants within the specific centre will be checked.
To monitor the collection, storage and processing of the data, each year peer auditors trained in data management will randomly check the research file on:
-
Presence of informed consents
-
Inclusion pace and drop out percentage
-
Presence and completeness of the research data.
If any errors are observed, the files of all study participants within the specific centre will be checked.
Furthermore, daily monitoring of the study will be done by the coordinating investigator, project leader, and principal investigator. Frequent contacts with the four trauma centres and research team meetings will be held to evaluate the progress of the study. All participating trauma centres have a contact person, who will be involved in all phases of the research.
Relevant concomitant care permitted or prohibited during the trial {11d}
Other psychological care for the adolescent will be paused from the first measurement until follow-up. Ongoing practical support can be continued. In cases of new warning signals of unsafety or suicidal thoughts crisis interventions are offered by the involved centre participating in this study
Provisions for post-trial care {30}
After the follow-up measurement an evaluation with the family will take place according to the procedure of the participating centres. If needed, post-trial care will be offered.
Outcomes {12}
Primary outcomes
Child PTSD symptoms
The Clinician-Administered-Ptsd-Scale-Children-And-Adolescents (CAPS-CA) (55) is a standardized clinical interview, to assess PTSD conform the DSM-5 standards. A Dutch translation of the CAPS-CA (56) is administered to children aged 8-18. The core traumatic event is chosen based on the life-events checklist. If the core traumatic event is a sequence of events, which is often the case in family violence, a brief term is chosen to capture the core of these events. The CAPS-CA assesses the frequency and intensity in which each PTS disorder symptoms occur. A five-point severity rating scale is used for all symptoms, ranging from 0 (= absent) to 4 (= extreme/incapacitating). An example item is ‘In the past month, have you had upsetting thoughts, pictures or sounds of what happened come into your mind when you didn't want them to?’. A total symptom severity score is calculated by summing severity scores for items 1-20. A total symptom score can also be calculated per cluster; re-experiencing (items 1-5), avoidance (items 6 and 7) negative alterations in cognitions and mood (items 8-17) and hyperarousal (items 15-20). A symptom cluster score may also be calculated for dissociation by summing items 29 and 30. To determine the PTSD diagnostic status individual symptoms should be dichotomized as “present” or “absent”. A symptom is considered present only if the corresponding item severity score is rated 2 (=Moderate/ threshold or higher). Items 9 and 11-20 have the additional requirement of a trauma-relatedness rating of Definite or Probable. The DSM-5 diagnostic rule requires the presence of at least one Criterion B symptom, one Criterion C symptom, two Criterion D symptoms, and two Criterion E symptoms. Also, Criteria F (disturbance at least one month) and G (disturbance that cause either clinically significant distress or functional impairment) must be met. Reliability of the CAPS-CA in a Dutch sample for the total scale ranged from coefficient a of 0.77 to .86 and the interrater reliability of the total scale is 0.99. Caps-CA also demonstrated divergent, convergent, and discriminant validity (57).
The Children’s Revised Impact of Event Scale (58) is translated in Dutch (59) and is a questionnaire for assessment of traumatic impact across different types of trauma and life-threatening events. Children and parents will rate 13 items according to the frequency of their occurrence in the past 24 hours in relation to the traumatic experiences (1 = none, 2 = rarely, 3 = sometimes, ad 4 = often). Example items are: “Do you think about the ……. even when you don’t mean to?”, and “Do you avoid talking about ……?” The sum score is used as an indicator of traumatic impact (ranging from 0 to 65). A score of 30 and over has been suggested as the most efficient cut-off for discriminating heightened risk for PTSD. Within a semi-clinical study population Cronbach’s a was .89 for the 13-items child version (60).
The Trauma Symptom Checklist for Children (61); Dutch translation: Trauma Symptoom Controle Lijst voor Kinderen (62) is a questionnaire to assess self-reported posttraumatic stress symptoms in children (8–12 years). It consists of 54 items clustering in eight scales: two validity scales (under response, hyper response) and six clinical scales (anxiety (9 items), depression (9 items), post-traumatic stress disorder (10 items), dissociation (10 items), anger (9 items) and sexual concerns (10 items). A four-point severity rating scale is used for all symptoms, ranging from 0 (= absent) to 3 (= very often). An example item is ‘Feeling afraid something bad might happen’. Reliability was high for all subscales, with a Cronbach alpha’s ranging from 0.72 to 0.83 in a large sample (63).
Emotional security and physical safety
To assess the severity and intensity of the violence that children have been exposed to, a combined measure will be used including items from the Conflict Tactics Scale (78 items, 8-point rating scale) (64), the Conflict Tactics Scale parent child (27 and 39 items, 8-point rating scale) (64), the Parents Report of Traumatic Impact (21 items ) (65); and the Adverse Childhood Experience Questionnaire (10 items, 5-point rating scale) (66). An example item is ‘I punched or hit my partner with something that could hurt’. This combined questionnaire has been used in a previous study on the effectiveness of a psycho-educational prevention program for children exposed to IPV (67). The questionnaire covers the topics and duration of the violence, the nature of the arguments in the relationship with the (ex-partner), followed by items from the Conflict Tactics Scale parent child and Parents Report of Traumatic Impact about problems between parent and child, and traumatic events the child has experienced. The questionnaire also includes items about traumatic experiences in parents’ own childhood. A short questionnaire of 8 items for parent and child is added to assess if any new IPV incidents or other stressful events occurred.
To assess emotional security the Dutch translation of the Security in the Interparental Subsystem (68) will be used. The Security in the Interparental Subsystem consists of 43-items with a 4-point ordinal scale (1 = not at all true of me, 4= very true of me). Children are asked to answer questions about previous violence and conflicts between parents and current violence and conflicts between parents and partners. It has three scales and seven subscales: emotional reactivity (emotional reactivity (9 items), behavioural dysregulation (3 items), regulation of exposure to affect (avoidance (7 items), involvement (6 items)) and internal representations (constructive family representations (4 items), destructive family representations (4 items) and conflict spillover representations (4 items)). An example item is ‘When my parents have an argument the family is still able to get along with each other’. The subscales show satisfactory internal consistency and test–retest reliability and previous research supported the validity of the SIS scale (68). Children will be asked to answer the same questions with respect to past fights and arguments between their parents and current fights and arguments between their mother and partner at T0, T12/14, and T13.
The Security in the Family System scale (69) will be used to assess how much children perceive their families as a reliable source of protection, stability, and support. The subscale ‘Secure’ will be used, which assesses a secure pattern of emotional security. Children indicate the extent to which they agree with 7 statements using a four-point scale ranging from ‘Complete disagree’(1) to ‘Complete agree’(4). An example item is ‘It’s worth caring about family members, even when things go wrong’. Psychometric properties of this security subscale are good, Cronbach’s a = 0.85 and test-retest reliability = 0.82 (69).
Caregivers report on children’s emotional security, including their emotional reactivity, behavioural involvement, and avoidance, after witnessing arguments between their parents using the Dutch translation of the Security in the Marital Subsystem Scale (SIMS) (68). Items are completed on a 5-point ordinal scale from 1 (not at all like him/her) to 5 (a whole lot like him/her) and include feeling sad, angry, afraid, and upset (e.g., “Still seems upset after we argue”). The SIMS has been found reliable and has demonstrated discriminant, convergent, and predictive validity (70).
The Security Scale (71)(72) is a 15-item self-report questionnaire for children between the ages of 8 and 18 and measures self-reported attachment security with their parents (73). Children and parents (independently) report on responsiveness and availability of the parent in the parent-child relationship (attachment). Each item is rated on a 5-point Likert scale (1 = strongly agree, 5 = strongly disagree’). An example item is ‘I feel my mother really understands me’. Several studies indicate adequate reliability and validity (74). High internal consistencies for 10 and 12 year olds (α=0.82, α=0.79) are reported (75). Test-retest reliability over a two-week period was high (r=0.75) (73). The Security Scale is related to other attachment measures (75). In a Dutch sample, internal consistency for the child version was α= 0.81 (76).
The Family Interaction Task (FIT) (77) is an observational instrument that measures parent-child interaction and consists of four tasks in which parent and child are instructed to complete a series of interactive tasks together. The FIT consists of four structured tasks: (1) guessing words by getting clues, (2) etch a sketch, (3) conflict task, and (4) completing a difficult but solvable puzzle by untangling a fixed ring from a standard within 4 minutes. These tasks are designed to elicit variations in parental and adolescent autonomy-promoting behaviours and emotional affect. The tasks are originally developed for children in middle childhood, but the tasks have been adapted for adolescents. Rating scales range from 1 to 5 and a higher score indicates a greater presence of that particular construct. Ratings are based on the complete session. Each videotaped session is independently coded by two coders (researchers). Interrater reliability was adequate (n =101, ICC range .87 – .94) in a Dutch sample (76).
Secondary outcomes
Child functioning
The Strengths and Difficulties Questionnaire (SDQ) (78) is a 25-item questionnaire using a 3-point scale (0 = not true, 2 = very true) . The SDQ contains five subscales and one total scale. Subscales are: a) conduct problems (5 items), b) emotional functioning (5 items), c) hyperactivity/inattention (5 items), d) Peer problems (5 items), e) prosocial behaviour (5). There is a self-report and parent version. An example item is ‘I worry a lot’. The internal consistency of the teacher version is good. The parent and self-report versions have an internal consistency that is generally acceptable in Dutch samples (78). In a Dutch sample Cronbach’s alphas were between 0.76 and 0.79 for internalizing problems and between 0.82 and 0.85 for externalizing problems (79).
The Children’s Depression Inventory 2 (CDI2) (80) is a 28-item self-rated questionnaire that measures symptoms of depression in children (7-18 years) with four subscales: negative mood/physical symptoms (9 items), negative self-esteem (6 items), interpersonal problems (5 items), ineffectiveness (8 items). Per item the child is asked to choose one of three sentences that best fits his/her feelings and thoughts in the past two weeks (e.g. ‘I am sad all the time’). The answers are calculated in a total score (ranging from 0 to 54). The internal consistency in a Dutch sample was high (a = 0.79), just as the test-retest reliability (r = 0.79) (81). The CDI has high criterion validity and scores on the CDI correlate high with scores on other measures for depression (81).
The Child Dissociation Checklist (CDC) (82) is a 20-item parent-rated questionnaire with a 3-point Likert scale (0 = not true, 2= very true). The child dissociation checklist is a screening instrument for dissociative problems in children (5-18 years). An example item is ‘Child does not remember or denies traumatic or painful experiences that are known to have occurred’. In a sample of maltreated and non-maltreated children Cronbach’s alpha was 0.89 (83). Discriminant validity is high (82).
Parent functioning
To assess parental stress, the Nijmegen Ouderlijke Stress Index—short version (NOSI-K) (84) will be used. The NOSI-K is a short version of the Parenting stress index (85) and a 25-item parent-rated questionnaire. The NOSI-K measures stress related to the parent-child relationship and uses a 6-point scale (1 = totally disagree, 6 = totally agree). An example item is ‘I often have the feeling I cannot handle things very well’. Cronbach’s alpha for mothers and fathers are .90 and .91 respectively (86) and the NOS-K has moderate concurrent, discriminate and criterion validity (84)(87).
To measure caregivers’ trauma knowledge, we will use a short self-report questionnaire developed for the training ‘Caring for Traumatized Children’ (88). The questionnaire comprises 6 items that reflect the six primary training goals, rated at a 10-point scale (1 = not accomplished, 10 = totally accomplished). Higher scores indicate higher trauma knowledge. An example item is ‘You know what the impact of trauma is on the development and behaviour of your child’.
Trauma symptoms in caregivers will be assessed with the Impact of Events Scale – Revised (IES-R) (89). The Dutch version ‘Schokverwerkingslijst (SVL-22)’ was developed by (90). This questionnaire consists of 22 items measuring symptoms of PTSD during the last week. The SVL-22 measures three dimensions: intrusion ( 9 items), avoidance (8 items) and hyper-arousal (6 items). Parents rate the items on a 5-point Likert scale ranging from ‘not at all’ (0) to ‘extremely’ (4). An example item is: ‘Any reminder brought back feelings about it’. Internal consistency was high (alpha = .88) (91). Support construct validity is found in a Dutch study (90).
The Young Adult Self-Report (92) will be used to assess psychopathology symptoms in parents. The short version of 29 items will be used in this study with items rated on a 3-point scale (0 = not true, 2 = often true). An example item is: ‘I am too dependent on others’. The Dutch version has good reliability and validity (93). In a Dutch study Cronbach’s alpha was 0.92 (7).
Daily Psychological Availability Scale (94) will be used to assess parental availability for the child eight adapted items of the Daily Psychological Availability Scale will be used. Items are measured using 7-point scales (1 = totally disagree, 7 = totally agree). An example item is ’When I spent time with my oldest child after work today, I was not able to focus on my child’. A higher score on this scale represents more psychological availability for the child. Cronbach’s alpha was 0.78 for both fathers and mothers (94).
Participant timeline {13}
Sample size {14}
Power analysis was conducted to determine whether an informative trial could be conducted given limitations imposed upon the maximum sample due to the duration of the study and size of the population of children indicated for trauma treatment, using G*power 3 (95). Sample size calculation was made based on the assumptions of the primary objective and the first secondary objective. A priori we decided that adding systemic intervention elements to individual trauma therapy would be justified if it would mean that on average at least one out of five children would have clinically significant benefit when given the full FITT compared to treatment without systemic elements (Number Needed to Treat = 5; equivalent to f = 0.18).
Secondary objective 1 (Does intensive trauma therapy with a family-based component (FITT) lead to faster reduction of trauma symptoms and increase of structural emotional safety than low frequency trauma treatment with parent therapy and family sessions?) is based on two measurements (pre and intermediate/follow-up). In accordance with the first sample size calculation we tested with setting f = 0.18 with a power of 80%, a significance level of 5% (two-sided) was used. Again default settings were used for correlation among repeated measures (= 0.5) and Nonsphericity correction (ε = 1). Based on this calculation a total sample size of 64 children is required.
The FITT sample we use to test our first hypothesis will also be used to test the second hypothesis. Taking into account loss to follow-up and measurement unreliability (20%), we set the target sample size of the multicentre randomized controlled trial on n = 120. Based on the number of families treated in previous years, and based on the current intakes the recruitment of 120 families that meet the inclusion criteria seems feasible.
Recruitment {15}
Recruitment will take place in four mental health centres for traumatized children in the Netherlands: Child and Youth Trauma Centre Kenter Jeugdhulp, Arq, UMC Utrecht, and GGZ Rivierduinen. Children and their parents are informed about the study by their therapists and will receive written information. After their permission, contact details of potential participants are provided to the researchers, who will contact the children and parents separately to provide more details about participation in the study. Children and their parents who are eligible for the trial are asked for informed consent. To reach the targeted sample size brochures about the study are handed out to referrers and are available for families and therapists at sites.
Assignment of interventions: allocation
Sequence generation {16a}
Random allocation to either condition is needed, in order to reduce selection bias and to be able to attribute effects to parenting training and family sessions added upon an intensive trauma therapy. After informed consent is provided families will be randomly assigned within each site to one of the treatment conditions. Within the multicentre trial families are randomized by an independent researcher, following a computer-generated list with each participant having a 33.3% chance of being assigned to FITT, ITT, or TAU. The results of the randomization are communicated to the site per participating family.
Concealment mechanism {16b}
Participating families are randomized by an independent researcher, following a computer-generated list with each participant having a 33.3% chance of being assigned to FITT, ITT, or TAU. The results of the randomization are provided to the centre per participating family. Subsequently the measures and intervention will start.
Implementation {16c}
An independent researcher will generate the allocation and directly informs the treatment team by e-mail about the treatment arm that families receive according to the randomization list. The therapist informs the families.
Assignment of interventions: Blinding
Who will be blinded {17a}
Until randomization both families and therapists are blind to treatment allocation to mitigate bias due to expectation or disappointment. Nobody will be blinded during the intervention.
Procedure for unblinding if needed {17b}
Not applicable.
Data collection and management
Plans for assessment and collection of outcomes {18a}
The instruments with the best psychometric qualities and sensitivity to change in a short time period are selected for this study. For the primary outcomes different types of measurements are included: questionnaires for multiple informants, an interview and an observation task. A description of the study instruments can be found in section ‘Outcomes (12)’. After collection, the data will be cleaned, coded and analysed in IBM SPSS.
Plans to promote participant retention and complete follow-up {18b}
It is common in trauma treatment that clients show avoidance behaviour because of the confrontation with anxiety- provoking details of the trauma. To prevent drop out psycho-education about avoidance behaviour and the disadvantages of avoiding is offered during treatment. To detect drop out of selective groups, participants who discontinue the treatment program, will be asked to continue measurements.
Data management {19}
All research data will be handled confidentially and will be stored in closed facilities, which can only be assessed by researchers. The research data is confidential and will not be provided to third parties. Privacy of the participants will be protected by allocating key codes to the personalia, which will only be traceable with a separate list. The key codes are 3-digit numbers ranging from 000 to 150, and will be given in consecutive order. This list with personalia (names, addresses, phone numbers) that connects the participants with the research data, is exclusively accessible by two of the researchers via an encrypted file and will eventually be destroyed. DarkStor from DANS (Dutch research data repository) will be used to store sensitive data and ArchStor (local data archive of Vrije Universiteit Amsterdam) for datasets that can be used for possible future research projects.
During data collection records of all specifics about the collections (e.g. measurements, dates, details about participants) will be kept in a logbook.
The researchers will write a report based on the interview part of the CAPS-CA and other answers are entered in a SPSS file. All data collected by questionnaires will be collected through an online survey system, Survalyzer. Data will be stored and coded in a SPSS file. The FIT observation tasks will be coded by researchers. All data will be cleaned before analyses (e.g. checking on double entries, ranges, checking dates of data collections with treatment dates, coding missing data).
Confidentiality {27}
The research data will be stored and managed by the research team. All employees who work with confidential data will sign a confidentiality agreement, on which they state not to share the information with third parties. The digitalized research materials and confidentiality agreements will be stored in a secured folder on the network of VU Amsterdam. Data will be stored for 15 years, according to Medical Ethical Committee guidelines or for five years after the last publication based on this data.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}
Not applicable.
Statistical methods
Statistical methods for primary and secondary outcomes {20a}
Measurement level of each variable is at least ordinal. We control for duration and severity of family violence (Kitzmann et al., 2003), child functioning, parental psychopathology (Levendosky et al., 2003), and new incidents of family violence as a priori potential confounders. Analyses will be performed using the latest version of the software package SPSS (Statistical Package for the Social Sciences). First, to determine whether changes on continuous measures are different per treatment, we use Linear mixed models. Continuous measures with only one time-point will be analysed with Independent samples T-test. Changes in categorical outcomes over the course of treatment across the conditions are tested with Generalized Estimating Equations (GEE). To handle missing data we will use maximum likelihood estimation within the Linear Mixed Models analyses. For the planned GEE, we will explore possibilities to deal with missing data, depending on the amount, randomness, and type of data missing. Analyses will primarily be performed according to the principles of intention-to-treat, however they will be repeated as per-protocol analyses (see table 4).
Descriptive and exploratory analyses will be carried out by standard methods. All continuous scores will be examined for normality. If variable distributions deviate too much from normality (West et al., 1996), corrective steps will be taken either by transforming variables or by using nonparametric tests. We will control for child functioning, parental psychopathology, duration, severity, and intensity of interparental violence, and new IPV incidents. The variables are treated as covariables in the analyses mentioned below.
To examine our primary hypothesis linear mixed models will be used in case of continuous data. GEE will be used for the categorical data. We will use the end of treatment (T12/T14) and intermediate/follow-up time-point (T13) to compare FITT with ITT (objective 1).
To examine our secondary hypotheses linear mixed models will be used. We will use the end of treatment time-point (T12/T14) to compare FITT with TAU. Control measures are child functioning, parental psychopathology, duration, severity, and intensity of interparental violence, and new IPV incidents. The variables are treated as covariables in the up mentioned analyses. We will use the Benjamini-Hochberg method to correct for multiple comparisons.
Interim analyses {21b}
n.a.
Methods for additional analyses (e.g. subgroup analyses) {20b}
n.a.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}
If children and families drop out from treatment, we will try to continue measurements according to the randomized treatment condition and analyses will primarily be performed according to the principles of intention-to-treat. This means families are analyzed as participants of the treatment condition they are randomized to. However analyses will be repeated as per-protocol analyses, which means that families are analysed as participants of the condition they actual received.
Plans to give access to the full protocol, participant level-data and statistical code {31c}
With this publication public access to the full protocol is granted. Because of privacy regulation and data protection the participant-level dataset is not accessible. Statistical code will be available upon request.
Oversight and monitoring
Composition of the coordinating centre and trial steering committee {5d}
Daily monitoring of the study will be done by the coordinating investigator, project leader, and principal investigator. Frequent contacts with the four trauma centres and research team meetings will be held to evaluate the progress of the study. All participating trauma centres have a contact person, who will be involved in all phases of the research. Furthermore and independent physician of each participating centre monitors the trial on-site and randomly check the research of participants. Participants can consult an independent expert with questions about the trial.
Composition of the data monitoring committee, its role and reporting structure {21a}
To monitor the collection, storage and processing of the data, peer auditors trained in data management will randomly check the research file on:
-
Presence of informed consents
-
Inclusion pace and drop out percentage
-
Presence and completeness of the research data.
If any errors are observed, the files of all study participants within the specific centre will be checked.
Presence and completeness of the research data will also be done on a daily basis by the research team.
Adverse event reporting and harms {22}
Adverse events are defined as any undesirable experience occurring to a subject during the study, whether or not considered related to intervention. All adverse events reported spontaneously by the subject or observed by the investigator or the research team will be recorded. Should a child or a parent seem adversely affected by the questionnaires, interviews or observational task as observed by the researcher or therapist or as reported by the parents or children themselves, it may be decided to (temporarily) discontinue participation in the project. The investigator will report all SAE’s to the sponsor without undue delay after obtaining knowledge of the events.
The sponsor will report the SAEs through the web portal ToetsingOnline to the accredited Medical Ethical Committee that approved the protocol, within 7 days of first knowledge for SAE’s that result in death or are life threatening followed by a period of maximum of 8 days to complete the initial preliminary report. All other SAE’s will be reported within a period of maximum 15 days after the sponsor has first knowledge of the serious adverse events.
All AE’s will be followed until they have abated, or until a stable situation has been reached. Depending on the event, follow-up may require additional tests or medical procedures as indicated, and/or referral to the general physician or a medical specialist.
Frequency and plans for auditing trial conduct {23}
The investigators as well as peer auditors trained in data management will randomly check the research file. Both are not independent of the sponsor. The investigators will check the research file on a daily basis and the peer auditors will monitor the file once a year. Twice a year the trial will also be monitored by the independent physician of each of the four participating youth mental health centres.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
Since a pilot with the treatment protocol has been done before data collection starts, we do not expect important protocol modifications during the period of data collection. If any substantial modifications are necessary they will be first proposed to the Medical Ethical Committee and only after approval implemented. If according to the therapists immediate care is necessary in cases of concerns about the safety of the child and/or families modifications can be implemented directly.
Dissemination plans {31a}
After data collection the first results will be presented to experienced experts with a history of family violence to interpret the results of the trial. In a second expert meeting participants from the trial, professionals with a history of family violence and professionals in the field of family violence will be invited to discuss: 1) if and how FITT should be adjusted to special target groups like refugees, veterans, mentally retarded, different cultural backgrounds; 2) how FITT can be disseminated in the Netherlands and what is necessary for implementation to achieve a nationwide network; 3) how FITT can be processed in clinical guidelines in consultation with professional associations like the Dutch Society on Traumatic Stress.
A handbook for therapists about FITT will be published so the knowledge can be transferred, implemented and executed in a sustainable way nationwide. The guide will provide information about the organization and structure of the program, and the questionnaires necessary to monitor FITT. We will develop a training for practitioners. Next, if proven effective - FITT is submitted to the Netherlands Youth Institute to assess the intervention for the database of effective interventions. Scientific articles about the research results will be submitted to national and international journals.