Study design

The proposed study is a parallel RCT, comparing pGMT to an active control intervention (pBHW), using a repeated-measures design across three time-points (preintervention (T1), postintervention (T2) and 6-month follow-up (T3)) (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Study period illustrated by SPIRIT figure. BRIEF, Behavior Rating Inventory of Executive Functions; EF, executive function; pGMT, paediatric Goal Management Training; pBHW, paediatric Brain Health Workshop; SPIRIT, Standard Protocol Items for Reporting in Trials.

The trial will be conducted and reported according to Consolidated Standards of Reporting Trials guidelines67 and Good Clinical Practice (GCP).68 The study flowchart is provided in figure 2.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Flowchart for the study protocol. Cognitive Rehabilitation in paediatric ABI—a randomised controlled study. ABI, acquired brain injury; EF, executive function.

Study setting

Eighty participants with paediatric ABI will be invited to participate in the study based on hospital discharge diagnosis and record information. Interventions will be conducted at two sites in Norway—St Olav’s Hospital (SOH) in Trondheim and Oslo University Hospital, Rikshospitalet (OUH-RH). In addition, patients from the University Hospital of North Norway (UNN) will also be invited to participate. SOH, OUH-RH and UNN are the trauma referral centres for the Central, South-Eastern and Northern regions of Norway, respectively, and have a population base of more than two-thirds of the Norwegian paediatric population. Recruitment started in November 2017 and will continue until the summer of 2019.

Population and eligibility criteria

The study population will consist of children and adolescents diagnosed with ABI resulting from traumatic (TBI) and non-traumatic injuries (brain tumour, stroke, hypoxia/anoxia and brain infections/inflammations) from the age of 10 up to 17 years at the time of the intervention. Participation requires a period of at least 12 months since injury/illness or more than 12 months since ended cancer therapy, in addition to experiencing executive dysfunction in daily life as determined by a semistructured interview. Exclusion criteria are as follows: (1) injury acquired before 2 years of age; (2) cognitive, sensory, physical or language impairment affecting the capacity to attend regular school (ie, primarily follow educational goals of peers and regular classroom teaching) and/or complete the training programme; (3) preinjury neurological disease, severe psychiatric disorder and/or stimulant medication; (4) recently detected brain tumour relapse; (5) unfit for evaluation of outcome (independent evaluation by two investigators) and (6) not fluent in Norwegian.

Sample size

The proposed study aims at recruiting a total of 80 children and adolescents with pABI. Based on previous studies, the estimated annual incidence of TBI in Norway (aged 0–19 years) is 1400.69 The prevalence of paediatric brain tumours in Norway is lower, with 40 new cases per year (survival rate is 80%),70 with a high proportion of patients suffering from cognitive deficits. Incidence rates for childhood stroke vary from 1.3 per 100 000 to 13.0 per 100 000,71 while rates of paediatric encephalitis-related hospitalisation range from 3 to 13 admissions per 100 000 children per year (USA and Europe).72 Thus, the sample size is considered to be attainable, with reference to the total eligible population. Previous cognitive rehabilitation studies after pABI including metacognitive training have included 29,61 32,73 3839 and 16135 participants. As only one previous single-case study on pGMT in pABI exists to date,60 this represents a challenge in estimating the required sample size in this proposed study. To document a clinically relevant effect as experienced by the child/adolescent and their family, we used the Global Executive Composite scale from the parental report on BRIEF66 as the main outcome measure in the power analysis. Prior research on the effect of GMT in adults with ABI and spina bifida has reported moderate to large effect sizes. To detect an effect size of d=0.7 with a power=0.80 and α=0.05, 32 patients are needed in each group.

Randomisation and blinding

Participants will be randomised to either pGMT or pBWT in a 1:1 ratio, applying block randomisation and stratification by (1) research site (Trondheim or Oslo) and (2) age at the time of intervention (10–13 years or 14–17 years). The Unit for Applied Clinical Research, Norwegian University of Science and Technology will be responsible for the computerised randomisation but will not be involved in determination of participant eligibility, assessment or execution of the interventions. The block size is set to four due to the study design of the treatment. The block randomisation was set up in advance by the allocator and hidden from the executer to minimise the possibility of guessing the next allocation. Blinding will be applied to reduce systematic bias as a result of knowing the treatment allocations, with the following procedures: (1) Families and participants will not be informed about which intervention they have been assigned to (the term ‘brain training’ will be used consistently in both groups); (2) Test technicians who conduct baseline and follow-up assessments will be blinded to treatment allocation; (3) Therapists who administer the two interventions will be blinded to all test performance and evaluation; (4) The use of the Web Clinical Research Form (WebCRF) system, with data stored anonymously and only biostatistician and research assistant having access, minimises the likelihood of influence by the investigators. Furthermore, the interventions (pGMT and pBHW) will be conducted separately in time on both sites, thus reducing the chance of participants with differing allocations sharing information and experience. Additionally, participants will be explicitly asked not to discuss course content with test technicians assessing them or other potential participants outside of their group.

Procedures

Following a positive invitation response, written informed consent will be required from potential participants (>16 years) or primary caregivers (participants <16 years) before conducting a semistructured interview to determine EF symptoms and information of other inclusion/exclusion criteria that are not included in the hospital records. The interview section concerning daily EF consists of eight specific questions where the caregiver is asked to assess and describe the child’s function, for example: ‘Is it difficult for the child/adolescent to plan school assignments or leisure activities?’. Participants that have surpassed 16 years of age are invited to attend the interview. Successively, individuals that meet inclusion criteria will be designated a study number and randomly assigned to either of the two interventions by a computer-based algorithm in a webCRF. Once the participants have been randomised, a baseline assessment is completed on the first day of admission, prior to the first session of the group intervention, as illustrated in the flow chart in figure 2. Test technicians will receive training in test administration and scoring by a clinical neuropsychologist.

Baseline variables

At baseline (T1), the following information will be recorded in addition to tests included for primary and secondary outcome measures: age, sex, height, weight, medical status (medicines, sleep, appetite, sensory sensitivity), neurological status (cranial nerve, motor function, balance, sensibility), pain and hormone levels (thyroid-stimulating hormone, free thyroxine, adrenocorticotropic hormone, cortisol, insulin-like growth factor one hormone, luteinising hormone, follicle-stimulating hormone and serum osmolality), school attendance and special educational needs, as well as injury characteristics, age of injury and medical treatment. The demographic information will be collected in a standardised interview, medical examination (including neurological screening) by a physician and information in relation to injury/illness and treatment will be extracted from medical records. In addition, an optional blood test will provide hormonal level status. To provide an IQ estimate, the subscales vocabulary, similarities, digit span, coding, block design and matrix reasoning of the Wechsler Intelligence Scale for Children—Fifth Edition74 will be employed.

Primary and secondary outcome measures

The present study intends to investigate effects on a number of central domains pertaining to sequelae after pABI. Hence, a comprehensive assessment battery of common outcome measures has been selected (table 1). The primary outcome measure is the BRIEF, parent report.66 BRIEF is an 86-item standardised questionnaire that captures parents perceptions of a child’s EF in his or her everyday environment. Each item’s frequency of occurrence is rated on a 3-point Likert scale from 1 (never) to 3 (often). It has demonstrated good reliability, with high test-retest reliability (rs=0.88 for teachers, .82 for parents), internal consistency (Cronbach's αs=0.80–0.98), and moderate correlations have been detected between teacher and parent ratings (rs=0.32–0.34). The questionnaire has been applied to several clinical groups in Norway.75 Secondary outcome measures include: (1) domains specific to EF and (2) those related to other cognitive functions, as well as other functional areas such as emotional health, fatigue, social function and QoL (see table 1). Additionally, we have included measures of EF assumed to be more ecologically valid than traditional neuropsychological tests. Indeed, to better uncover challenges related to EF in daily activities we have translated and included a method that is designed as an ecological, open-ended assessment of EF in children, the ‘Children’s Cooking Task’ (CCT).76 77 In addition to evaluating intervention satisfaction after each intervention week, intervention compliance will be assessed by modules completed by each child/adolescent.

Interventions

The intervention programme will be conducted in the same manner at both research sites according to preplanned standardised procedures. The two interventions are matched regarding non-specific factors (eg, structure, intensity, duration and therapist contact). Hence, the pGMT and pBHW protocols are group based and consist of seven modules of approximately 2-hour duration each (figure 3), both with standardised PowerPoint slides and participant workbooks. Between sessions, there will be homework assignments aiming to promote generalisation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Description of the paediatric Goal Management Training and paediatric Brain Health Workshop interventions.

Two experienced clinical neuropsychologists will lead the group treatment sessions, each consisting of two to five participants, ideally four. The present study will aim at facilitating transfer from training to real life by including training and education of parents. A 1-hour review of the modules will be given to the parents by the same group therapists subsequent to each of the seven participant sessions, with the opportunity for questions and experience exchange. Following to the fourth session, the pGMT participants will start receiving a text message stating ‘STOP’ (a key instruction in pGMT), while the pBHW will get a text message stating ‘Brain training’ every other day for five consecutive weeks until the end of the intervention period. This constitutes 18 text messages per participant and is included to cue goal management and encourage transfer to daily life.78 79 The cueing time will be between 08:00 and 20:00 and will be changed routinely to prevent habituation. Two hours (2×1) of telephone counselling by one of the therapists, including both parent(s) and child when possible, will be conducted following the sessions during a 1-month period. The counselling mainly provides support to promote use of the techniques in everyday life by discussion of how the strategies (pGMT) or lifestyle advice and techniques (pBHW) may be implemented in home and school settings, in addition to clarifying possible obstacles and misunderstandings. Teachers will receive a 1-hour telephone counselling in the same period. The main purpose is to provide a brief review of the specific intervention strategy allocated and thus encourage teachers to consider how the strategies can be implemented in the school setting. In addition, the counselling may contribute to enhanced knowledge of pABI and cognitive dysfunction in general. Predetermined templates for the telephone counselling of parents/participants and teachers respectively ensures systematic and consistent procedures across sites, with specific protocol for each intervention.

Paediatric Goal Management Training and Paediatric Brain Health Workshop

In pGMT, participants are trained to use strategies such as stopping and orienting to relevant information, partitioning goals into subgoals, encoding and retaining goals, monitoring performance and mindfulness training.59 The training also uses discussion of participants’ real-life attention deficits, in-session practice on tasks of attention supplemented with periodic alertness cueing and homework assignments. The manualised pGMT protocol is an adaptation of the adult version of GMT which has been translated into Norwegian.46 47 ,63 The modifications of the GMT protocol for children included changes to make materials more child friendly and age appropriate (eg, examples and discussions more related to school and leisure activities). For example, ‘The mental blackboard’ (ie, working memory) became ‘The brain blackboard’, and ‘The mental laboratory’ (a way of thinking about, practising and experimenting with daily life skills) became ‘The brain workshop’. Finally, the protocol has been compressed into seven modules (from 9). Following the pilot work by Stubberud and colleagues,63 the protocol has been further revised, primarily by simplifying language, reducing the amount of text, making examples more relevant to child and adolescent activities and adding more visual material. Nevertheless, all sessions will follow the same procedure; introduction to key concepts, practical exercises and discussion using examples from the participants’ daily life. Homework assignments will include practical exercises and logging of activity and exercises in mindfulness. Similarly to the development of the pGMT, the pBHW is an adaptation of the adult version of BHW which has been translated into Norwegian.55 It comprises educational materials addressing brain injury and (dys)function, plasticity, memory and learning, executive function, fatigue and lifestyle interventions (eg, the effect of stress, sleep, exercise and nutrition on brain function) (figure 3). Homework and within-session activities will include sharing of the participant’s experiences with the intervention topics in home and school settings, brain games, practical exercises and lifestyle advice, in addition to quizzes testing their acquired knowledge. Furthermore, the participants and parents are invited to discuss how the techniques and advice may be incorporated into their everyday routine.

Statistical analysis

Data analysis will follow the intention-to-treat principle and will be specified in a separate Statistical Analysis Plan. The analysis populations will be defined according to the following strategy: (1) The primary analysis population (modified intention-to-treat) will include all randomised participants who have completed baseline assessments. (2) The per protocol population will include all participants who have missed a maximum of two out of the seven group sessions. Results from all primary and secondary outcome measures will be presented separately by treatment allocation. A description of participant characteristics and baseline measurements will take place when all groups have carried through the baseline assessment (T1). The final analysis of the RCT will be conducted following the 6-month (T3) follow-up for participants, anticipated to be December 2019. For continuous outcomes, results will be presented as means and SD of age-corrected standardised scores defined from normative samples. The primary analyses will be performed by a linear mixed model with time point, intervention and the interaction between intervention and time as fixed factors. Supplementary analyses will be performed using linear regression. Sensitivity analyses will include adjustment for baseline factors where there is an obvious imbalance between the two groups. For binary outcomes, results will be presented as the number and proportion with impairments for each of the outcomes presented separately by treatment groups. Comparisons between groups will be done by χ² tests. Additional sensitivity analyses adjusting for any imbalanced baseline factors will be performed by multiple logistic regression models.28 29 Significance level is set to p≤0.05. Due to the issues of multiple testing in assessment of a number of secondary outcomes, results will be interpreted according to magnitude of the group difference (effect size) rather than relying solely on significance levels.

Ethics and dissemination

Patient and public involvement

Development of cognitive rehabilitation is highly requested by user organisations. We have identified several potential users; patients, their families, patient organisations and municipality services. This study seeks user involvement in all project phases. User involvement was included in the planning process through meetings and interviews with previous patients with pABI, patient organisations and municipality services, as well as execution of pilot studies. As such, the development of the study design, research questions and choice of outcome measures have been informed by patients’ priorities, experience and preferences. The patients will not, however, be involved in the recruitment to and conduct of the study. Furthermore, during the execution phase and dissemination phase, an advisory panel with user representatives (participants from pilot study, parents of children with pABI and patient organisations) has been appointed and is scheduled to have regular meetings throughout this phase and will be involved in discussions to improve study design, secure a good consent process, relevant assessment instruments and publication in relevant media. Finally, findings to be made available to the participants at the end of study include main findings, where the information is tailored to the characteristics of the sample, in addition to an offer of receiving feedback on individual results.