Effects of an individualized and goal-oriented intervention in the chronic phase of pediatric acquired brain injury: the Child in Context Intervention (CICI)—a randomized controlled trial Authors: Nina Rohrer-Baumgartner, Ingvil Laberg Holthe, Edel Jannecke Svendsen, Hilde M Dahl, Ida M H Borgen, Solveig L Hauger, Cathrine Brunborg, Jennie Ponsford, Shari L Wade, Jens Egeland,, Cecilie Røe, Marianne Løvstad Published: 2026 ABSTRACT AbstractBackgroundThe objective was to investigate the effect of an individualized, goal-oriented intervention for children in the chronic phase of pediatric acquired brain injury (pABI).MethodsA randomized controlled trial (RCT) of the Child in Context Intervention (CICI) with a 1:1 randomization to an intervention or treatment as usual control group. Children (6–16 years) with pABI ≥ 1 year before recruitment and pABI-related challenges in daily life were included with their parents and assessed at time point 1 (T1) and again after 6 months (T2) and 9 months (T3). The manualized rehabilitation intervention lasted 5 months and contained seven family sessions, four school sessions, and a parent seminar. Outcome assessors were blinded to group allocation. Preestablished primary outcomes were parent-reported brain injury symptoms in the children (Health and Behavior Inventory; HBI) and parenting self-efficacy (Tool to measure Parenting Self-efficacy; TOPSE/Teen TOPSE). Secondary outcomes assessed unmet health care needs, parental mental health, family functioning, quality of life, executive functioning, and severity of pABI-related main challenges in daily life.ResultsFifty-eight children (30 boys) and 112 parents (53 fathers) participated. The median age of the children at T1 was 11.7 (range 6.3 -16.9) years. No significant mean between group effect was found from T1 to T3 on the primary outcomes. Significant mean between group effects in favor of the intervention group were found regarding unmet health care needs (FNQ-P: 0.42; 95% CI 0.15 to 0.69; p = .002), parental symptoms of anxiety (GAD-7: -1.83; 95% CI -3.32 to -0.35; p = .016) and depression (PHQ-9: -2.0; 95% CI -3.69 to -0.30; p = .021), and family functioning (FAD-GF: -0.20; 95% CI -0.32 to -0.07; p = .002) at the main end-point T3.ConclusionsAcknowledging that the success of an intervention is defined by interpretation of all available evidence, not just the primary outcome(s), results indicate that the CICI is effective and can help families in the chronic phase after pABI. This is the first RCT to our knowledge that offers an individualized and goal-oriented pABI-intervention to children in the chronic phase, while involving families and schools.Trial RegistrationClinicalTrials.gov NCT04798859, 2021–05-10 https://clinicaltrials.gov/study/NCT04798859?term=Child%20in%20Context%20Intervention&rank=1Supplementary InformationThe online version contains supplementary material available at 10.1186/s12887-026-06997-y.Keywords: Pediatric acquired brain injury, Brain injury, Child, Intervention, School, Community, Family functioning, Health care needs, RCT, SMART goals Introduction Pediatric acquired brain injury (pABI) is any kind of injury or insult to the brain after birth or the initial neonatal period, caused by e.g. accidents, anoxia, or illness such as tumor, stroke or encephalitis [1]. pABI is a leading cause of disability as survivors often face a variety of long-term impairments, with symptoms such as fatigue, cognitive impairment, difficulties regulating emotions and behavior, sleep disturbances, impaired vision, and reduced motor functioning, which also impact their families [1–5]. Due to a complex interplay between affected brain development and environmental demands, the full range and severity particularly of cognitive and social impairment due to pABI, may not become evident until the chronic phase, at least one year after injury [6–9]. Despite the well-documented, long-lasting impairments after pABI, evidence-based rehabilitation for children and adolescents in the chronic phase of pABI is scarce and levels of unmet health care needs and educational needs are high world-wide [10–12]. Moreover, having a child with pABI may negatively influence family functioning and parental mental health [13]. The associations between family factors and child outcome are reciprocal, making family factors modifiable risk factors for long-term outcome after pABI [8, 14]. Thus, it is imperative to include the family in pediatric rehabilitation [15–18]. In addition, interventions need to be tailored to the unique contexts and needs of each child and family [19, 20], due to the heterogeneity of potential sequelae after pABI and because every family is unique. A goal-oriented approach where families, children and therapists collaborate in goal setting is considered a key component in pABI rehabilitation [21, 22], and may help increase family engagement during rehabilitation [23]. Parents’ experiences with achieving their families’ goals, may in turn boost parenting-self efficacy (parents’ confidence in their ability to parent), which is associated with parents’ well-being [24] and favorable parental involvement [25, 26]. In the field of adult traumatic brain injury (TBI), a community based intervention for veterans in the chronic phase of TBI was initially evaluated [27] and later expanded upon and tested in a Norwegian context with promising results regarding high goal-attainment, improved quality of life and reduced TBI symptoms and anxiety [28, 29]. Comparable interventions for the pABI population have been lacking, although problem-solving interventions and parent skills training have proved effective for children and young people with TBI [30–33]. Another intervention that has been evaluated for children with ABI is Goal Management Training (GMT) for dysexecutive symptoms. In an RCT with children with ABI, GMT did not outperform an active control group [34], while a single case study using an adapted GMT approach with five participants with severe TBI showed promise, but struggled to engage the children’s families and schools [35]. Of note, GMT is usually administered in groups, limiting work on specific individualized goals with relevance for daily life [36]. In contrast, a recent small-scale intervention for children with ABI and executive dysfunction (N = 12) where participants chose their own goals, yielded positive goal-attainment [37]. However, the study did not include children without executive dysfunction, families or schools. Considering that most children spend a large part of their time in educational settings, school is – alongside the home environment—an arena with great potential for rehabilitation. Yet, the number of studies involving school in rehabilitation is low and most seem to mainly focus on an educational [38, 39] or service-coordination approach [40]. To our knowledge, there are no previous RCTs including a comprehensive, family- and goal-oriented intervention involving a family focus and rehabilitation collaboration with schools as part of children’s ABI-rehabilitation. Consequently, we developed the Child in Context Intervention (CICI). It is a manualized, goal-oriented, individualized rehabilitation program tailored for children with pABI and their families, conducted in close collaboration with the children’s schools. The intervention design was informed by the previously mentioned adult intervention [28, 29], with adaptations to the pediatric population, such as introducing a three-way focus on the child, parents, and school. The development of the CICI was also influenced by pre-existing, evidence-based problem-solving interventions for children and adolescents with TBI in the USA [33, 41], as well as clinical experience. The study protocol [42] has been published and the intervention was found feasible and acceptable to children, parents and teachers [43, 44]. Moreover, 94% of SMART-goals [45] in the intervention group had a positive goal attainment score [22]. Objectives and hypotheses The two main objectives of the present study were to establish whether 1) the CICI was effective in improving parent-reported brain injury symptoms in the children and 2) in improving parenting self-efficacy (pre-established primary outcomes). Furthermore, we aimed to establish whether 3) the CICI is effective in reducing unmet health-care needs, improving family functioning, parents’ emotional functioning, children’s quality of life, children’s self-reported brain injury symptoms, and their executive functioning in daily life, along with the severity of the families’ self-reported main pABI-related challenges (pre-established secondary outcomes). We hypothesized that the intervention group receiving the CICI would show improvements in the above-mentioned functions compared to the control group at T3, which was the main end-point. No potential harms, defined as the opposite of beneficial effects [46, 47], were prespecified as they were not expected. Methods This study adhered to the Consolidated Standards of Reporting Trials (CONSORT) guideline for RCTs [48]. Parents and adolescents of 16 years of age provided written informed consent. Children under the age of 16 received age-appropriate information about the study and assented to participation, in line with Norwegian legislation and ethical requirements. The study was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (2019/1283) and the Data Protection Officer at Sunnaas Rehabilitation Hospital (SRH) and complies with the Helsinki Declaration. Trial design The CICI study is a parallel 2-group RCT with 1:1 randomization to an intervention group or control group (treatment as usual). The intention-to-treat principle was followed. Participating children and parents were assessed at T1. Schools were then contacted and T1-assessments from teachers completed before randomization. Outcome assessments were conducted, at T2 (6 months after T1), and at T3 (9 months after T1) in both groups. Outcome assessors were blinded to group allocation. For further details, see the study protocol [42]. Regarding user involvement, the Norwegian Organization for People with Acquired Injuries («Personskadeforbundet») was involved in designing the study. Experts by experience contributed with feedback during the development of the psychoeducation materials for the study and were present at the parent seminars. Recruitment and eligibility Eligibility criteria for participants Inclusion criteria were, as described in the study protocol (42, p. 5): “(1) children 6–16 years of age with a CT- or MRI-verified diagnosis of ABI or loss of consciousness post-insult and verified neurological symptoms in cases where MRI was not administered, (2) time since insult at least one year, (3) the child had self- or parent-reported cognitive, emotional, behavioral and/or social difficulties that affected everyday functioning and/or participation, (4) the child attended school regularly, (5) the family was able to participate actively in goal-oriented work, including speaking sufficient Norwegian or English.” The following exclusion criteria were applied: “severe pre- or comorbid neurological, neuropsychiatric or psychiatric disorders, cancer in active treatment or at great risk of relapse, children living in institutions most of the time, parental severe psychiatric illness, drug abuse, and indications of a history of or risk of domestic violence.” Settings The clinical records of potentially eligible children were screened for inclusion and exclusion criteria at SRH, Oslo University Hospital, and at the Norwegian Service of Special Education (Statped). Eligible families were then contacted by phone by designated researchers (authors NRB, ILH, EJS, respectively). Most T1 assessments were conducted at SRH, while a few families who lived far from the hospital were assessed locally. At all time points, participants completed questionnaires in Viedoc™ (Viedoc Technologies AB, Uppsala, Sweden), an electronic data capture system. Interviews were conducted face to face at T1 and by telephone at T2 and T3. Data were collected between May 11th 2021 (first T1 assessment) and August 10th 2023 (last T3 assessment). Interventions Intervention group A visual overview of the intervention content is presented in Fig. 1. The intervention consisted of seven family sessions delivered by video conference, four school meetings and one parent seminar. Content of the Intervention The first weeks after T1 were used to recruit schools and complete T1 assessments, followed by the 5 month intervention. The family sessions included parents and, in most cases, the child. Siblings were welcome to join when relevant to the families’ goals. The first session took approximately two hours, while the time frame for the remaining six family sessions was 1.5 h. Every session included at least one break, and children were free to take additional breaks as needed. In the family sessions, SMART goals were established. SMART goals should be Specific, Measurable, Achievable, Realistic/Relevant, and Timebound [45]. A goal-oriented approach to brain injury rehabilitation is useful as it increases patients’ engagement and motivation [23, 49], and SMART goals may cover the broad array of needs after brain injury for the whole family [50]. The SMART goals were defined in close collaboration between therapist and family and focused on problems that the families struggled with and were motivated to work with. With each goal, strategies to reach the goal were agreed upon. In each session, strategies for the goal(s) were reviewed and refined. In the first session a new goal was set, and the Goal Attainment Scaling (GAS) for the goal was defined. The fifth session was the last one where a new goal could be established. In the seventh and final family session, all GAS were calculated. The mean number of goals per family was 3.21 and most goals (84%) were about children’s functioning, with the remaining 16% targeting challenges related to parents and/or families. For further details about the SMART goals and goal attainment, see Holthe et al. [22]. Families were also provided with the psychoeducational CICI handbook; a twelve chapter booklet with information about common challenges after pABI and advice to minimize negative consequences. The intervention was manualized. For family sessions, here is an overview of the main manualized components: Session one: The therapist explained the material that the family had received: the CICI handbook and a booklet with contact information, information about SMART goals, and pages where the family could write SMART goals, GAS and strategies for each goal. The therapist reminded the family of their most challenging, pABI-related areas in daily life. The first SMART goal with a corresponding GAS was established, preferably based on one of the child’s challenging areas. After the child/family selected a challenging area within which to set the first goal, the therapist helped define the goal. Also, strategies to reach the goal and GAS were defined. In sessions 2–5, additional SMART goals were set when the family had the capacity to and wished to work on a new goal. In contrast to the first goal, these were not always based on the challenging areas in daily life defined at baseline, as new problems not previously mentioned could become clear with time. Strategies and GAS for the new goal were established. Session 5 had an extra focus on the child’s understanding of their challenges. It was discussed if the child could use a sentence to tell others when asked why he/she was not always at school or why he/she appeared different. When wanted, the therapist helped the child form a sentence that he/she felt comfortable with. Also, there was extra focus in this session on what other healthcare providers the family might need to be in touch with. Throughout sessions 1–6, a chapter in the CICI handbook was selected as homework for the next session. The chapter was chosen based on the family’s needs and challenges and related to the individualized goals. In the following session, the chapter was discussed in relation to how the family could use this knowledge in everyday life. At the end of the session, the session content was recapped, new appointment scheduled. In every session, the therapist took extra care to connect with the child by talking about something he/she enjoyed. The family was shown a visual overview of the sessions and which session they were in. A plan was made for how long the child would join the session and the therapist helped the parents to make a plan for what the child could do in his/her breaks. The strategies of each SMART goal were reviewed, and therapists asked if the families had used them since the last session and how they worked. Positive feedback regarding what they accomplished and what they managed well along with realistic hope for improvement was provided and possible obstacles were explored. Strategies that did not work were adjusted or discarded, and new strategies were established when necessary. After school sessions, therapists and families discussed what had worked well and what to focus on in the next school session. Children were not present when the focus was on parents’ goals related to guilt, grief, psychological distress or related topics. In the last session, therapists and families went through the strategies of each goal, recapped the intervention, and established the current level of goal attainment. Therapists conveyed realistic hope for the future and discussed possible future challenges (especially regarding transitions, e.g. when the child would start a new school) and how to handle them using the knowledge and tools gained in the intervention. Key therapeutic principles were to build trust, convey realistic hope, and acknowledging that the families are the experts regarding their lives and families, thus therapists were “guiding, not driving” when selecting goals and strategies. Another key therapeutic principle was to give the families very specific skills or “tools” to solve their problems in daily lives. One such tool was the SMART goal technique itself, another tool was the knowledge the families gained about pABI during the intervention, and further tools were the strategies that were selected to reach SMART goals. The strategies were determined by the problems they were meant to contribute to solve for each family. Thus, the therapeutic approach is best described as integrative or eclective. Importantly, strategies to reach child-related SMART goals were often compensatory, meaning that they involved influencing or changing the child’s environment to reduce symptom impact in daily life. The four school meetings were conducted by video conference with the child’s main teacher and parents, which each lasted 1–1.5 h. The child was invited to participate in the school meetings and the school’s administration and other relevant school personnel such as assistants were encouraged to participate. Before school meetings, a special needs educator who was part of the CICI research group (see Acknowledgements) observed the participating child in the classroom, to gain an improved understanding of the child in the school setting and of the school environment. Teachers were also given the CICI handbook at the start of the intervention and the handbook was referred to in school meetings when needed. Parents were invited to a one-day seminar in groups of approximately 10 parents. Topics included experience with goal-oriented work, communication within the family, taking care of siblings of the affected children, emotional challenges as a parent and psychological self-care. Therapists (authors NR-B, ILH, and EJS) provided brief introductions to each topic. Parents exchanged experiences and advice throughout the day. The families were assigned one of the two experienced therapists (clinical neuropsychologists; authors NR-B and ILH) and in some cases a co-therapist (a specialist children’s nurse; author EJS) who led the family sessions. The therapists had worked closely together in the feasibility study [43, 44], where every family had two therapists, to ensure that interventions did not vary systematically between practitioners. Regular meetings in the research group were held to discuss challenges and ensure consensus. Fidelity was ensured by having a senior therapist (authors NR-B, ILH, SLH, and ML) attend at least one session per family, completing a checklist (see Additional file 1). For therapists NR-B and ILH this was in addition to the families they already worked with. Fidelity was very high and will be described in detail in the process evaluation [51]. In most cases, the main therapist of the family attended the first and last school meeting, while the special needs educator from Statped attended all four school meetings. Control group Families in the control group received oral feedback on results from the T1 assessment. Otherwise, the control group received their usual care. In accordance with the study protocol, no rehabilitation or other treatment was systematically removed or added. However, in a few cases (n = 4), therapists recommended some new treatments for children in the control group at T1, as this was perceived as an ethical obligation. Advice was given to contact a vision rehabilitation therapist (n = 1), a mental health professional (n = 2), and one child was referred to Statped, where a CICI therapist participated in two meetings with the service and the child’s municipal health services. In Norway, health care services in the chronic phase of pABI are typically provided by municipalities and specialized health related services are only provided after referral. Data collection T1 assessment Information regarding child developmental and medical-, family-, and school characteristics was gathered. Children completed the Matrices and Similarities subtests from the Wechsler Intelligence Scale 5th edition (WISC-V) [52] to estimate verbal and visual abstract reasoning abilities. Outcome measures All the following measures were predefined and administered at T1, T2 and T3. A variety of outcome measures was chosen, as challenges after pABI are typically heterogeneous, and families worked with different goals whose topics were not chosen beforehand. The final decision regarding primary and secondary outcomes was based on existing literature, study hypotheses and experiences from the feasibility study [43]. Of note, there were no comparable studies to base the final decision regarding primary vs. secondary outcomes upon. Primary outcome measures Parent-reported severity of children's brain injury symptoms was assessed with the Health and Behavior Inventory (HBI; [53]) on a scale from 0 (never) to 3 (often), where higher total scores indicate more severe symptoms. Parenting self-efficacy was assessed with "A Tool to Measure Parenting Self-Efficacy” (TOPSE; [54]), which exists in a child and an adolescent version. The latter was used for parents of children of 10 years of age and older. The TOPSE is rated on a scale from 0 (completely disagree) to 10 (completely agree). Higher mean total scores indicate higher parenting self-efficacy. All questionnaires completed by parents were completed separately by mothers and fathers. Secondary outcome measures Unmet healthcare needs of the families were assessed with the Family Needs Questionnaire—Pediatric version (FNQ-P; [55]) on a scale from 0 (not at all met) to 4 (completely met), reported by parents. Higher mean scores indicate a higher level of met needs. Family functioning, reported by parents, was assessed with the General Functioning subscale of the McMaster Family Assessment Device (FAD GF; [56, 57]) on a 4-point scale: strongly agree, agree, disagree and strongly disagree. Lower mean scores indicate better family functioning. Depressive symptoms in the parents were assessed with the Patient Health Questionnaire 9 (PHQ-9; [58]) on a scale from 0 (not at all) to 3 (almost every day). Higher scores reflect higher levels of depressive symptoms. Anxiety symptoms in parents were assessed with the Generalized Anxiety Disorder 7 (GAD-7; [59]) on a scale from 0 (not at all) to 3 (almost every day). Higher scores reflect higher levels of anxiety symptoms. Children's quality of life was assessed with the core module of the Pediatric Quality of Life Inventory (Peds-QL; [60]) scored on a scale from 0 (never a problem) to 4 (almost always a problem). Parents and children each completed the questionnaire according to the child’s age. Suitable versions of the Peds-QL are administered for age ranges 6–7, 8–12 and 13 + years. Higher scaled scores indicate higher quality of life, because scores are reversed in the process. Child-reported (≥ 7 years of age) severity of brain injury symptoms was assessed with the Health and Behavior Inventory (HBI; [53, 61]) on a scale from 0 (never) to 3 (often), where higher total scores indicate more severe symptoms. Children's executive functioning in everyday life was assessed with the Global Executive Composite (GEC) on the Behavior Rating Inventory of Executive Function 2 (BRIEF-2; [62]) on a 3-point scale: never a problem, sometimes a problem or often a problem. Parents and children from age 11 completed the questionnaire. Higher scores indicate higher level of problems with executive functioning. We measured the most prominent p-ABI challenges using a 5-point Likert Scale: 0 (not all difficult), 1 (a little difficult), 2 (moderately difficult), 3 (quite difficult), 4 (very difficult). High scores indicate high severity. Parents reported three challenges each, rating their severity on the Likert Scale. For ethical reasons, children were free to name less than three). At T1, severity was assessed in in-person interviews, while at T2 and T3, by phone by blinded outcome assessors. Sample size Sample size calculations were conducted before the trial and were performed using G*Power software, version 3 [63]. Since two primary outcomes were used, the calculation was conducted with an alpha level of 0.025. Two-sided t-tests were used as the basis for the analysis. Treatment effects were estimated to be at 0.8 (Cohen’s d), based on the very limited literature available and experience from the CICI feasibility study [43]. With a statistical power of 0.8 and equal randomization to the two groups, a sample size of 70, which means N = 64 after expected attrition, was indicated. Randomization Rehabilitation therapists were responsible for the randomization of participants, which was done in the electronic data catching system Viedoc™ (Viedoc Technologies AB, Uppsala, Sweden). Random block sizes of four, six, and eight were generated by a statistician not involved in the treatment provision. Within each block, an equal number of participants were assigned to both groups. The sequence of the group allocation within each block was random. Blinding Outcome assessors who made phone calls to the families at T2 and T3, asking about the severity of the main pABI-related challenges and services received, were blinded to group allocation. For obvious reasons, it was not possible to blind participants to group allocation. Statistical analyses were conducted by an independent statistician (author CB). The first author wrote the Results section while blinded to group allocation. Statistical analysis The primary and secondary outcomes were analyzed using linear mixed effects models to account for repeated measures over the follow up period (T1, T2, T3). We nested the data structure into three levels for these models: i) time, ii) child, and iii) parents. The BRIEF-2 was also nested for teacher. All models included fixed effects for time, and treatment group x time interaction. The main endpoint for the outcomes was at T3. The analysis of primary interest in establishing treatment efficacy was a time × group interaction in the direction of the intervention group improving above the levels of the control group at T3. Based on these models, mean between-group differences in change from T1 to T3 were analyzed with 97.5% confidence intervals (CI) for primary endpoints and 95% CI for secondary outcomes. As there may be systematic differences in how children and parents respond to health-related questionnaires [64] and as children with acquired brain injury may have cognitive deficits and reduced self-awareness [65] that may influence their answers, the linear mixed effects analyses were performed separately for children and parents. Teachers´ responses were also analyzed separately, for an added perspective. The intention-to-treat approach was used in analyses. Results The study flowchart is presented in Fig. 2. A total of 377 potentially eligible children were identified. Of these, the largest group of ineligible children were 131 who were excluded because they did not have a verified brain injury. A total of 58 families were randomized. Two families withdrew from the study, one in each group. In addition, two families in the control group were lost to follow-up at T3. At T1, data were collected from 28 schools in the intervention group and 27 schools in the control group. One additional school participated in the intervention group but did not provide data at T1. One school was lost to follow up in the intervention group at T2 and three in the control group, leaving school n = 28 and 24, respectively. At T3, two schools in the intervention group and one in the control group were lost to follow-up. Thus, at T3, 26 schools participated in the intervention group and 23 in the control group. Nine participating children had at least one change of teachers. Three out of 29 children did not participate in the family sessions, due to severe cognitive deficits. Parents from five families were unable to attend the parent seminar. In four cases, the main therapist attended all school meetings, as the special needs educator was unavailable. Study Flowchart Participants were recruited from April 2021 to October 2022. At this time, we had exhausted potential eligible participants with injuries acquired during the past 10 years at all three recruitment sites, and a pragmatic decision to stop inclusion was made, despite the somewhat lower sample size than initially planned for. Demographic and clinical characteristics for both groups are presented in Table 1. Descriptive data regarding teachers were not collected. There were equal numbers of girls and boys in both groups and the age distributions of children and parents were comparable. In both groups, most parents were highly educated. Also, time since injury, abstract reasoning abilities and receipt of services were comparable between groups. Most children in both groups had non-traumatic injuries. Table 1Demographic and Clinical Characteristics of the Groups at T1CharacteristicParticipant No.Intervention GroupParticipant No.Control GroupSex (%) Girls (% of children within group)14 (48.3)14 (48.3) Boys (% of children within group)15 (51.7)15 (51.7) Mothers (% of parents within group)29 (52.7)30 (52.6) Fathers (% of parents within group)26 (47.3)27 (47.4)Age in years (IQR) Children median11.7 (9.1–15.0)11.7 (9.3–15.4) Parents median44 (39.0–49.0)44 (37.5–47.5)Parents’ education (% within group) ≤ 10 years3 (5.5)2 (3.5) 11–13 years14 (25.5)13 (22.8) 14–16 years24 (43.6)28 (49.1) ≥ 17 years14 (25.5)14 (24.6)Type of injury (% within group) TBI11 (37.9)5 (17.2) Tumors7 (24.1)8 (27.6) Stroke4 (13.8)6 (20.7) Infection2 (6.9)6 (20.7) Anoxia4 (13.8)2 (6.9) Other1 (3.4)2 (6.9)Time since first injury in years,median (IQR)5.3 (1.8–8.5)6.3 (2.8–8.4)Similarities (WISC-V), scaled score median (IQR)9.5 (7.3–11.0)a10.0 (8–11.5)Matrixes (WISC-V), scaled score median (IQR)9.0 (7.0–11.0)a9.0 (7.5–11.0)Receipt of services T1 (% within group) No follow-up7 (24.1)7 (24.1) Municipal services9 (31.0)6 (20.7) Specialized health care services5 (17.2)4 (13.8) Municipal and specialized health care services8 (27.6)12 (41.4)an = 28, one child with too severe cognitive deficit for WISC-V Demographic and Clinical Characteristics of the Groups at T1 Time since first injury in years, median (IQR) an = 28, one child with too severe cognitive deficit for WISC-V To provide a better understanding of the clinical work conducted in the intervention, an example of a child goal and a parent goal are provided in Figs. 3 and 4. Example of a child goal, goal attainment scaling and strategies to reach the goal, with energy scale Example of a parent goal, goal attainment scaling and strategies to reach the goal Primary and secondary outcomes The full results of the linear mixed model analyses are presented in Table 2 and Fig. 5. At T3, no significant mean between-group differences between the intervention group and the control group were found for the two primary outcome measures: parent-reported brain injury symptoms of children (HBI total score) and parenting self-efficacy (TOPSE/Teen TOPSE mean score). Table 2Linear mixed-effects model results for primary and secondary outcomes Outcome Intervention Group Control Group Between-group change from T1 No. of participants Mean (CI) a Change from T1, mean (CI) a No. of participants Mean (CI) a Change from T1, mean (CI) a Mean (CI) a P value Primary Outcomes HBI, parent-report T15532.85 (29.14 to 36.56)-5732.54 (28.85 to 36.24)--- T25230.04 (26.10 to 33.99)−2.81 (−5.12 to −0.50)4931.79 (27.82 to 35.75)−0.76 (−3.09 to 1.58)−2.05 (−5.34 to 1.23).162 T35230.58 (26.20 to 34.97)−2.27 (−5.28 to 0.74)4830.60 (26.15 to 35.05)−1.94 (−5.03 to 1.15)−0.33 (− 4.64 to 3.99).866 TOPSE T1557.81 (7.50 to 8.13)-578.13 (7.82 to 8.44)--- T2498.02 (7.68 to 8.37)0.21 (−0.04 to 0.41)468.03 (7.69 to 8.38)−0.09 (−0.35 to 0.16)0.30 (−0.05 to 0.66).057 T3527.85 (7.48 to 8.21)0.03 (−0.23 to 0.30)488.06 (7.69 to 8.43)−0.07 (−0.34 to 0.21)0.10 (− 0.28 to 0.48).560Secondary Outcomes FNQ-P T1552.57 (2.34 to 2.80)-562.46 (2.23 to 2.69)--- T2503.19 (2.94 to 3.44)0.62 (0.46 to.78)472.41 (2.16 to 2.66)−0.05 (−0.21 to 0.11)0.67 (0.44 to 0.89).000 T3513.07 (2.80 to 3.35)0.50 (0.32 to 0.69)472.55 (2.27 to 2.82)0.09 (−0.11 to 0.28)0.42 (0.15 to 0.69).002 FAD-GF T1551.90 (1.77 to 2.03)-571.66 (1.53 to 1.79)--- T2501.85 (1.72 to 1.99)−0.05 (−0.13 to 0.04)471.77 (1.63 to 1.90)0.11 (0.02 to 0.19)−0.15 −0.28 to −0.03).018 T3511.79 (1.65 to 1.92)−0.11 (−0.20 to −0.03)471.75 (1.61 to 1.88)0.09 (−0.00 to 0.17)−0.20 (−0.32 to −0.07).002 PHQ-9 T1557.76 (6.51 to 9.00)-564.98 (3.74 to 6.22)--- T2506.04 (4.71 to 7.38)−1.71 (−2.72 to −0.70)476.46 (5.11 to 7.81)1.48 (0.45 to 2.51)−3.19 (−4.64 to −1.75).000 T3516.98 (5.52 to 8.45)−0.77 (−1.95 to 0.41)476.20 (4.71 to 7.70)1.23 (0.01 to 2.44)−2.0 (−3.69 to −0.30).021 GAD-7 T1557.11 (5.87 to 8.35)-574.82 (3.60 to 6.05)--- T2505.49 (4.21 to 6.76)−1.62 (−2.66 to −0.58)475.52 (4.24 to 6.81)0.70 (−0.36 to 1.76)−2.32 (−3.81 to −0.83).002 T3516.21 (4.95 to 7.48)−0.90 (−1.93 to 0.14)475.76 (4.47 to 7.05)0.94 (−0.13 to 2.00)−1.83 (−3.32 to −0.35).016 PedsQL, child-report T12777.62 (72.26 to 82.97)-2967.84 (64.65 to 76.12)--- T22371.96(66.82 to 77.11)−5.65 (−10.04 to −1.27)1867.28 (61.84 to 72.71)−0.56 (−5.33 to 4.20)−5.09 (−11.56 to 1.39).124 T32170.39 (64.65 to 76.12)−7.23 (−12.96 to −1.50)2070.60 (64.70 to 76.50)2.76 (−3.13 to 8.64)9.99 (−1.78 to −18.20).017 PedsQL, parent-report T15560.54 (54.84 to 66.24)-5662.23 (56.54 to 67.92)--- T25062.36 (56.35 to 68.36)1.31 (−1.80 to 4.43)4763.54 (57.50 to 69.59)1.82 (−1.22 to 4.85)0.50 (−3.85 to 4.86).820 T35162.19 (55.72 to 68.67)1.44 (−2.51 to 5.40)4763.67 (57.11 to 70.23)1.65 (−2.16 to 5.46)0.21 (−5.28 to −5.70).939 HBI, child-report T12621.77 (17.35 to 26.19)-2726.93 (22.59 to 31.26)--- T22425.94 (22.14 to 29.75)4.18 (1.12 to 7.23)1727.95 (23.86 to 32.05)1.03 (−2.42 to 4.47)3.15 (−1.46 to 7.75).180 T32224.10 (20.35 to 27.86)2.33 (−1.57 to 6.23)1926.88 (22.90 to 30.86)−0.04 (−4.15 to 4.07)2.37 (−3.29 to 8.04).412 BRIEF-2, child-report T11458.71 (52.76 to 64.66)-1560.47 (54.72 to 66.21)--- T21267.53 (62.18 to 72.88)8.81 (4.60 to 13.03)959.95 (54.27 to 65.63)−0.52 (−5.22 to 4.19)9.33 (3.02 to 15.65).004 T31260.47 (55.49 to 65.45)1.75 (−2.96 to 6.46)1057.96 (52.72 to 63.20)−2.51 (−7.51 to 2.49)4.26 (−2.61 to 11.13).224 BRIEF-2, parent-report T15565.52 (61.95 to 69.09)-5763.90 (60.34 to 67.45)--- T25061.97 (58.28 to 65.65)−1.01 (−2.67 to 0.64)4762.88 (59.18 to 66.58)−3.56 (−5.19 to −1.92)−2.54 (−4.87 to −0.22).032 T35062.98 (59.08 to 66.88)−0.41 (−2.54 to 1.71)4763.48 (59.56 to 67.41)−2.54 (−4.62 to −0.47)−2.13 (−5.10 to 0.84).160 BRIEF-2, teacher-report T12460.04 (55.71 to 64.36)-2661.26 (56.96 to 65.57)--- T22561.21 (56.99 to 65.43)1.17 (−1.78 to 4.11)2360.47 (56.15 to 64.80)−0.79 (−3.70 to 2.13)1.96 (−2.18 to 6.10).354 T32459.40 (55.01 to 63.80)−0.63 (−4.06 to 2.79)2460.70 (56.28 to 65.11)−0.57 (−3.84 to 2.71)−0.07 (−4.81 to 4.67).978 Severity of main challenges, child-report T1242.38 (1.98 to 2.78)-292.67 (2.30 to 3.03)--- T2231.89 (1.55 to 2.23)−0.19 (−0.51 to 0.13)262.47 (2.16 to 2.79)−0.49 (−0.83 to −0.15)−0.30 (−0.77 to 0.17).210 T3232.16 (1.81 to 2.51)−0.35 (−0.75 to 0.05)252.31 (1.98 to 2.65)−0.22 (−0.65 to 0.21)0.13 (−0.45 to 0.72).656 Severity of main challenges, parent-report T1553.06 (2.87 to 3.25)-573.04 (2.86 to 3.23)--- T2522.40 (2.18 to 2.63)−0.65 (−0.87 to −0.44)502.62 (2.39 to 2.85)−0.43 (−0.64 to −0.21)−0.23 (−0.53 to 0.08).144 T3522.53 (2.24 to 2.81)−0.53 (−0.81 to −0.25)492.53 (2.24 to 2.82)−0.51 (−0.79 to −0.23)−0.02 (−0.41 to 0.38).379Abbreviations: HBI Health and Behavior Inventory, TOPSE Tool to Measure Parenting Self-Efficacy (child version and teen version), FNQ-P Family Needs Questionnaire pediatric version, FAD-GF McMaster Family Assessment Device, General Functioning, PHQ-9 Patient Health Questionnaire 9, GAD-7 Generalized Anxiety Disorder 7, PedsQL Pediatric Quality of Life Inventory, BRIEF-2 Behavior Rating Inventory of Executive Function 2 (Global Executive Composite)aConfidence intervals for primary outcomes 97.5% and for secondary outcomes 95% Linear mixed-effects model results for primary and secondary outcomes 71.96 (66.82 to 77.11) Abbreviations: HBI Health and Behavior Inventory, TOPSE Tool to Measure Parenting Self-Efficacy (child version and teen version), FNQ-P Family Needs Questionnaire pediatric version, FAD-GF McMaster Family Assessment Device, General Functioning, PHQ-9 Patient Health Questionnaire 9, GAD-7 Generalized Anxiety Disorder 7, PedsQL Pediatric Quality of Life Inventory, BRIEF-2 Behavior Rating Inventory of Executive Function 2 (Global Executive Composite) aConfidence intervals for primary outcomes 97.5% and for secondary outcomes 95% Linear Mixed Model Results for Primary and Secondary Outcomes. Legend: A. Primary Outcomes. TOPSE: high mean score = higher parenting self-efficacy, HBI: low score = fewer brain injury symptoms. B. Secondary Outcomes. FNQ-P: high mean score = fewer unmet health care needs, FAD-GF: lower mean score = better family functioning, PHQ-9: lower score = fewer depression symptoms, GAD-7: lower score = fewer anxiety sympotms, PedsQL: high scaled score = higher quality of life, BRIEF-2 GEC: lower score = fewer executive problems in daily life, Severity of challenges daily life: lower score = lower severity of self-reported main challenges in daily life. HBI Higher total scores indicate more severe symptoms. TOPSE Higher mean scores indicate higher parenting self-efficacy. FNQ-P: Higher mean scores indicate higher levels of met needs. FAD-GF: Higher mean scores indicate worse family functioning (score of 2 or higher indicating problematic family functioning). PHQ-9: Score range 0 to 27. Higher scores indicate higher levels of depressive symptoms. GAD-7: Higher scores indicate higher levels of anxiety symptoms. PedsQL: Higher mean scaled scores indicate higher quality of life. BRIEF-2 (GEC): Higher scaled scores indicate more problems with executive functioning. Severity of main challenges: Higher mean scores indicate higher severity of pABI-related challenges in daily life For the secondary outcomes, the results showed a significant mean between-group difference at T3 in favor of the intervention group for the following measures: unmet health care needs (FNQ-P mean total score), family functioning reported by parents (FAD-GF mean score), parents’ depressive symptoms (PHQ-9 total score), and parents’ symptoms of generalized anxiety (GAD-7 total score). Regarding child-reported quality of life (PedsQL scaled score), there was a significant mean between-group difference at T3 in favor of the control group compared to the intervention group. No significant mean between-group differences at T3 were found for the remaining secondary outcome measures (presented in Table 2). As shown in Table 2 and Fig. 5, all significant between-group differences at T3 were already present at T2, except for child-reported quality of life (PedsQL scaled score), on which there was no significant mean between-group difference at T2. However, at T2 only, child reports showed significant between-group differences regarding executive functioning in daily life (BRIEF-2 GEC) in favor of the control group, while parents’ ratings showed the opposite: significant between-group differences regarding executive functioning in daily life (BRIEF-2 GEC) in favor of the intervention group. As previously noted, harms were not prespecified in the protocol. See the next section for a discussion of potential harms. Discussion The current study examined the effect of the Child in Context Intervention, a pragmatic RCT which has previously been proven feasible and acceptable to participants [43, 44]. The intervention did not lead to change in the predefined primary outcomes regarding parenting self-efficacy and parent reported brain injury symptoms, but it did show a positive effect in favor of the intervention group at T2 that was maintained at T3 with regards to unmet health care needs, family functioning and parents’ mental health. The success of an intervention cannot simply be defined by the primary outcome(s), but demands an interpretation of all available evidence [66]. In the present intervention, other evidence included positive findings on secondary outcomes and a very high goal attainment in the intervention group [22]. Overall, the findings suggest that a goal-oriented, individualized intervention for children with ABI in the chronic phase, which also involves parents and schools, can help families. While none of the positive observed between group effects were directly related to the children, improving the family environment and parental mental health may be beneficial for the affected children in the long-term. More specifically, there is consistent evidence to suggest that family factors significantly influence long-term psychosocial outcome of children with brain injury, potentially more so than the injury itself [67–71]. This underscores the importance of complex interventions such as the CICI. Moreover, the CICI not only addresses a gap in the literature but also adds to the short list of interventions that may be made available for children with ABI in the chronic phase and their families. Similarly to the CICI, the online problem-solving intervention TOPS has also been shown to lead to improved mental health for parents, as the content of the intervention focuses on both children’s/adolescents’ and parents’ problem-solving skills [31]. To fully comprehend the results, it is necessary to address certain methodological issues. Choosing outcome measures and ranking them as primary versus secondary is challenging when there are no comparable studies to base the decisions on, resulting in somewhat arbitrary decisions. Also, there might be a lack of sensitivity in questionnaires regarding improvement. To illustrate: in the HBI questionnaire, participants are asked to what degree a symptom is present, not to what degree they are able to deal with the symptom in daily life. In comparison, the rehabilitation approach in the CICI mainly targeted symptom management. Also, the TOPSE is a generic tool for measuring parenting self-efficacy, and may not be sensitive to brain injury specific parenting challenges. Another, and maybe the most important challenge in the present study was that it is generally difficult to obtain significant effects when measuring the consequences of a complex, multifaceted and highly individualized intervention at a group level, using predefined standardized outcome measures, as compared to at the individual level (e.g. goal attainment). For example, the lack of significant result regarding parenting self-efficacy may in part be explained by the CICI being a complex, individualized intervention where we worked on different types of rehabilitation goals, according to the families’ wishes. Some families in the current sample experienced more pressing challenges than parenting self-efficacy that they wished to prioritize when working with SMART goals. However, in qualitative interviews that are published elsewhere [44, 72], some parents expressed what can be interpreted as clinically improved parenting self-efficacy due to the CICI. Thus, clinically, at least a subgroup of parents did probably experience increased parenting self-efficacy. Despite these potential challenges in complex interventions, due to the complex array of factors that influence functioning in daily life after pABI, complex interventions are essential to meet the different rehabilitation needs. A process evaluation that will discuss the results of the CICI in more detail, is in preparation [51]. Regarding the secondary outcomes with significant results, it makes sense from a clinical perspective that unmet family healthcare needs decreased due to the CICI, as this might be at the very core of the intervention. The CICI is a healthcare intervention and the therapists worked with the families to meet their needs. Family functioning was likely improved as the CICI contributed to a common understanding between parents regarding the child’s symptoms and how to adapt to these in everyday life. Also, the CICI contributed with advice regarding how to communicate to increase understanding and decrease conflict in the family (e.g. in the parent seminar and the CICI handbook). In addition, when classifying the goals of the CICI according to the International Classification of Functioning (ICF; [73]), the code regarding support and relationship within the immediate family was used ten times, the code about interpersonal interactions was used six times and the codes about relationships and communication three times [22]. Parents’ symptoms of depression and anxiety may have decreased due to a combination of several factors: improved family functioning, emotional support provided by therapists, and the experienced success and relief in reaching SMART goals that directly affect daily life. In combination, these factors may have conveyed hope for the future. From a clinical perspective, it is not clear why the same factors did not also lead to increased parenting self-efficacy as measured by the TOPSE and we assume that the reasons for this are mainly methodological (mentioned above), or also that a lack of parenting self-efficacy was not one of the main problems for the parents who participated in this study. Children reported lower absolute levels of problems or dysfunction than parents on all measures at baseline. This is in accordance with previous literature, as agreement between children and parents on standardized health-related measures is known to vary or even be poor [64, 65, 74, 75]. Child-reported executive functioning and quality of life decreased after the intervention, leaving post intervention scores closer to those of the parents. This is perhaps not surprising, as children and parents worked on rehabilitation goals together and received psychoeducation about pABI, which may have increased the children’s self-awareness of symptoms. In general, lack of awareness influences the validity of child report, irrespective of brain injury. Children are, for example, generally more likely to use a single, recent event as a justification for rating an item on a questionnaire, compared to adults [76]. Cognitive problems in this sample of children with ABI may additionally affect validity [77, 78], which also contributed to the choice of parent reported primary outcomes. This does not mean that children’s answers should not be considered. In fact, we have argued that children’s concerns about their challenges after pABI should be included in assessments [20]. However, judging the severity of symptoms might be more challenging for children than adults, as it is an abstract concept and requires a frame of reference. Unintended effects and harms Because parent reports did not support the unexpected effects shown in child reports, and due to the possible explanations regarding increased awareness in children and questionable validity of child report, the results regarding children’s increase in some self-reported problem levels are not interpreted as harms. Moreover, if these results, at least in part, indicate heightened self-awareness, this change might prepare the children to face everyday challenges in a more realistic way, which could be beneficial for functioning [79, 80]. Strengths and limitations A major strength of the CICI study was that it introduced and evaluated a new, promising intervention for children in the chronic phase of ABI and their families, an area where effective complex interventions are sorely needed [17, 81]. Further strengths of the study included being well-designed and focusing on individual challenges in everyday life, which increases the ecological validity of the intervention. A particular strength was that the study included children and schools as well as parents, and that many fathers participated in the study, which is not common in the literature. Moreover, attrition was low in both groups. A further strength was that the present study was conducted in accordance with the UK Medical Research Council (MRC) framework for complex interventions [82]: a protocol and two feasibility studies were published before the RCT. A process evaluation exploring possible active ingredients of the intervention is being prepared [51]. Limitations of the study include being slightly underpowered (N = 58 instead of 64 as suggested by power analyses), and therefore not having enough statistical power to detect small to moderate differences and to investigate interactions such as how etiology or age affected the outcome. The sample was from a mainly urban area with a large proportion of highly educated parents; thus, generalizability to samples with more socioeconomically or culturally diverse backgrounds is unknown. Of relevance, an online problem-solving intervention for teenagers with TBI found greater effects in families with lower socio-economic status [32]. Future studies might improve diversity by recruiting from more non-urban regions and taking special care to recruit families from cultural and ethnic minorities by e.g. providing information about the study in different languages and being open to working with interpreters. Some of the children’s answers on questionnaires may have been biased, as children who needed help completed questionnaires with assistance from therapists at T1 and with help of parents at later time points. Teacher reports may have been influenced by some teachers being new to a class and completing a questionnaire while not knowing the child well. A further limitation was that the telehealth format was not ideal for assuring active engagement in children [43, 44], although it had other advantages, such as enabling children/families who live far from SRH and their schools to participate. Conclusions and clinical implications The current study is, to our knowledge, the first well-controlled RCT that offers an individualized, goal-oriented intervention for children with ABI in the chronic phase, and where the children, families, and schools were actively involved. The results suggest that the CICI rehabilitation can help families in the chronic phase of pABI, even several years after injury. Future replication studies should include larger and more diverse samples, allowing investigation of factors that might interact with effect, such as etiology, age, and family characteristics. It is also important that implementation research is conducted, as factors related to therapist skill, context and sample characteristics may play a role regarding effect. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements We would like to thank all the participants for their time and effort. We would also like to thank Anine Strand-Saugnes for her valuable work as a special needs educator in the part of the intervention concerning the schools and for screening clinical records at Staped for the recruitment of participants. Abbreviations Child in Context Intervention Pediatric acquired brain injury Randomized controlled trial Time point one Time point two Time point three Health and Behavioral Inventory Tool to measure Parenting Self-Efficacy Traumatic brain injury Consolidated Standards of Reporting Trials Computed tomography Magnetic resonance imaging Sunnaas Rehabilitation Hospital Wechsler Intelligence Scale 5th Edition Family Needs Questionnaire –Pediatric version Family Assessment Device -General Functioning subscale Patient Health Questionnaire 9 Generalized Anxiety Disorder 7 Pediatric Quality of Life Inventory Behavior Rating Inventory of Executive Function 2, Global Executive Composite Confidence Interval UK Medical Research Council Authors’ contributions All authors contributed with critical comments during the writing process and read and approved the final manuscript. NRB contributed to the planning and the development of the study, participated in the recruitment of participants, served as a therapist, served as a senior therapist to ensure fidelity, contributed to data cleaning, completed descriptive analyses, wrote the first draft of the manuscript and led the revisions of the manuscript. ILH contributed to the planning and the development of the study, participated in the recruitment of participants, served as a therapist, served as a senior therapist to ensure fidelity, and contributed to data cleaning. EJS led the recruitment of participants and served as a co-therapist. HMD participated in the recruitment of the participants by screening clinical records at Oslo University Hospital. IMHB contributed to the planning and the development of the study, data collection, including outcome assessment, and data cleaning. SLH contributed to the planning and the development of the study and served as a senior therapist to ensure fidelity. CB served as an advisor regarding design and statistical power and completed the statistical outcome analyses. JP, SLW, JE, and CR contributed to the planning and the development of the study and served as advisors throughout the study. ML is the PI of the study, she contributed to the planning and the development of the study and oversaw the execution of the study and served as a senior therapist to ensure fidelity. Funding This work was supported by the Norwegian Research Council under grant 288172. Data availability The data generated and analyzed during the current study are not publicly available due to data protection legislation but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate Parents and adolescents of 16 years of age provided written informed consent. Children under the age of 16 received age-appropriate information about the study and assented to participation, in line with Norwegian legislation and ethical requirements. The study was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (2019/1283) and the Data Protection Officer at SRH and complies with the Helsinki Declaration. Consent for publication The Norwegian consent forms are available upon request. All licensed measures were used under license. The FNQ-P was used under license from Holland Bloorview Kids Rehabilitation Hospital, Toronto. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement The data generated and analyzed during the current study are not publicly available due to data protection legislation but are available from the corresponding author on reasonable request.