Brief Parent Training for Children With Behavioral Difficulties in Primary Care Settings

STUDY DESIGN

We will conduct a two-armed randomized controlled trial. Children and their parents will be randomly assigned (simple and parallel randomization) to (a) the intervention condition in which parents receive the brief parent training and may receive care as usual, or (b) the control condition in which parents may receive care as usual only.

The trial will take place in several general practitioners' practices throughout the Netherlands. The brief parent training will be provided by child mental health workers (in Dutch: praktijkondersteuners GGZ kind en jeugd).

TIME FRAME

December 2023: start recruitment, screening and inclusion of parents and children, start running interventions.

December 2023 until September 2025: ongoing contacts with general practitioners practices; recruitment, screening, and inclusion of children; run interventions and conduct measurements; supervision of therapists; fidelity checks; other project management activities.

September 2025 until September 2027: data-analysis, dissemination of results, implementation activities.

SAMPLE SIZE ASSESSMENT

In our prior study conducted in specialized mental health care with experienced behavioral therapists (all with MA degree), we found moderate short-term effects on momentary assessed (i.e., daily) ratings of behavioral difficulties compared to waitlist (range of Cohen's d .54 to .66; Hornstra et al., 2021). The current trial will take place in a less specialized setting, with a more heterogenous group of mental health care providers (also BA degree), who are generally less educated in behavioral therapy. Therefore, we expect the effect to be between small to moderate (d = 0.35).

A power analysis has been performed using G*Power software (Faul, 2007). Based on an effect size of d = 0.35, two groups and two repeated measures (T0 and T1 for short term effectiveness, with r = .60 between-measurement correlation), a power of .80 and alpha = 0.05, we would need 27 participants per group, resulting in a total of 54 participants. Given that our data is clustered, we multiplied our total sample with 110%, resulting in a total number 60 participants (Twisk et al., 2013).

TREATMENT OF SUBJECTS - BRIEF PARENT TRAINING PROGRAM (INTERVENTION CONDITION)

The brief parent training program consists of three sessions. The first two sessions will take 120 minutes and the third session will take 60 minutes. All three sessions will be spread over a maximum of six weeks. In between sessions, parents are asked to read some materials and implement the plan as created during the session at home.

The brief parent training program uses the behavioral techniques that were identified as effective for behavioral difficulties in a microtrial of our group (Hornstra et al., 2021) and the broader literature (Kaminski et al., 2008; Leijten et al., 2019; Leijten et al., 2021). The program combines the stimulus control and contingency management techniques that, separately, reduced behavioral difficulties in the microtrial.

In the screening phase, parents indicate which four, daily occurring, target behaviors they prefer to address in the training. They also indicate in which specific situations these behaviors mostly occur. The selected behaviors and situations will be sent to the child mental health worker. Out of the four target behaviors, two will be addressed during the training.

The first session consists of four steps. First, psycho-education will be provided on executive function problems and motivational issues that may occur in children with behavioral difficulties. Second, using functional analyses, the child mental health worker and parents will determine which of the available techniques may address the first target behavior in the specific situation. Third, the child mental health worker and parents make an action plan for the target behavior in the specific situation, in which these techniques are included and explained to the parents. Fourth, the action plan will be practiced using role play and the parents are instructed to implement the action plan as soon as possible at home. In the second session, the plan that was made in the previous session will first be evaluated. Thereafter, the same steps will be repeated for a second problem behavior. In the third session, the child mental health worker and parents will evaluate both plans and discuss how to maintain and generalize the behavioral techniques that proved to work.

Parents allocated to the intervention condition may receive care as usual as well.

TREATMENT OF SUBJECTS - CARE AS USUAL (CONTROL CONDITION)

Care as usual may include all mental health care that is usually provided within or outside (e.g., at a child mental healthcare institution or at school) the general practitioners' practice, except from pharmacological treatment for children's behavioral difficulties and/or behavioral parent training/support, up until the first posttreatment assessment (T1). Care as usual can also imply that there is no support or treatment.

METHODS

Demographic information about the child and parents will be collected with a self-composed questionnaire, with questions about the age and gender of the child, household composition, native country of the child and parents, school type of the child, educational background of the parents, (possible) mental disorders of the child, and mental healthcare that the family previously received for this child. In addition, child mental health workers will fill out a self-composed questionnaire, with questions about their age, educational background and working experience.

EXPLORATORY MODERATOR VARIABLES

Besides our primary and secondary outcomes (see outcome measures) we will investigate several candidate moderators of treatment effects:

• Parent-reported severity of child disruptive behaviors as measured with the intensity subscale of the Eyberg Child Behavior Inventory (ECBI-I; Robinson et al., 1980);
• Attachment style of parents as measured with the 12-item version of the Revised Experiences in Close Relationship (ECR-R) questionnaire (Wei et al., 2007);
• Parental reward responsivity as measured with the 8-item Reward Responsiveness (RR) questionnaire (Van den Berg et al., 2010);
• Parent-reported child reward responsivity and punishment sensitivity as measured with 22 items of the Sensitivity to Punishment and Sensitivity to Reward Questionnaire for Children (SPSRQ-C) (Luman et al., 2012);
• Parent-reported child differential susceptibility to environmental factors as measured with the 12-item Highly Sensitive Child - Parent Report questionnaire (HSC-PR; Sperati et al., 2022);
• Parent-reported child emotion regulation as measured with the 7-item emotion regulation subscale of the Dutch version of the Behavior Rating Inventory of Executive Function questionnaire (BRIEF-2; Huizinga et al., 2023).

STUDY PROCEDURES

Parents will be recruited within the general practitioners' practices, by child mental health workers or general practitioners. The child mental health worker will inform parents about the study and hand out an information letter. Parents who are interested are asked for verbal consent for the child mental health worker to share their contact information with the researchers. The researchers will use this contact information to further inform parents about the study through telephone calls. If parents want to participate, the researchers will ask them to provide written informed consent. Both legal caretakers (if present) of the child need to provide consent. Parents will have at least 14 days to decide upon participation and ask the researchers any questions. If needed, parents can ask for more time or consult an independent expert to decide upon participation.

After providing informed consent, screening will take place to check whether the family meets all inclusion criteria. If so, parents will be asked to fill in the pre-intervention measurement (T0), whereafter they will be randomized into one of the conditions. A randomization list is generated by and kept by two researchers that are not in any way involved in the study.

To measure the primary and secondary outcomes, parents in both treatment arms will be asked to fill out questionnaires and answer short daily phone calls (at least four days in a row) on a maximum of three measurement occasions. The questionnaires concerning candidate moderators of treatment response will be assessed in both treatment arms during the pre-intervention measurement (T0). The questionnaires will take around 30 minutes to complete at T0, and around 20 minutes at T1 and T2. The daily phone calls will take around two minutes each.

The child mental health workers need to participate in a one-day training provided by two experts in behavioral parent training. Before the one-day training, they are asked to read the materials carefully and to submit a video in which they demonstrate certain aspects of the training. The child mental health workers will receive feedback from the experts or researchers on this video. There are no further requirements that child mental health workers must meet in order to participate, as the group of child mental health workers should be representative for the current situation in general practitioners' practices. Generally, there is a lot of variation between child mental health workers in this setting in terms of study background, years of working experience and postmaster education.

STATISTICAL ANALYSIS

Data will be analyzed on an intention-to-treat basis. Differences in demographic variables between the study-arms will be analyzed with ANOVA's (continuous variables) or chi-squared tests (categorical variables). To examine the effects of the intervention on the primary and secondary outcomes we will conduct multilevel analyses (mixed modeling). Mixed model analysis takes missing data into account (Twisk et al., 2013). We will test a three-level multilevel model: outcomes (level 1), nested within children (level 2) and nested within therapists (level 3). The intercept at therapist level will only be included if the Likelihood Ratio Test shows a significant improvement of the model fit. To assess the influence of exploratory moderating variables on treatment effects, we will include these variables as interaction effects in the multilevel analysis.

ETHICAL CONSIDERATIONS

The risk of any physical or mental harm of the study is very low and not to be expected higher than what may be expected in daily life of the parents and their children. We will exclusively use conventional methods and interventions. The program uses behavioral techniques that were identified as effective for disruptive behaviors in our recent microtrial (Hornstra et al., 2021) and the broader literature (Kaminski et al., 2008; Leijten et al., 2019; Leijten et al., 2021). Parents in both intervention conditions can eventually receive the brief parent training program. Potential adverse reactions will be assessed at T1 (both conditions) and T2 (intervention condition) with a questionnaire. The study is embedded in general practitioners practices.

The children will not be directly involved in the study. They will not be present during the brief parent training sessions and all measures will be parental measures.

Parents can decide to leave the study at any time for any reason if they wish to do so without any consequences.

To debrief participants, parents who wish to learn about the results of the study can request this information from the researchers after the study is finished. This concerns only information on group level. We will inform parents about the possibility to ask for this information in the information letter.

HANDLING AND STORAGE OF DATA AND DOCUMENTS

Digital data, both raw and processed/analyzed data, will be stored at a study specific highly secured folder within the Accare network. This study specific folder can only be accessed by authorized researchers involved in this study.

All questionnaire data, phone call data and audio files (recordings of the sessions) will be stored in one pseudonymized file, that is accompanied by a 'readme' text file that contains a code book explaining the meaning of all variables. The file contains separate pages ("tabs") for all different outcome variables at all different timepoints. A separate logbook-file will be created documenting all decisions that will be made during the process from raw to published data. Old versions of all these documents will be stored, document names will contain their date to ensure version control.

Management of the data and access permissions are with the principal investigator of the research project.

Access permissions are controlled by one of the research project's principal investigators. Raw data containing identifiable information such as the separate log containing participant identification number and study code will be kept strictly separate from the processed data and can only be accessed by the supervisors, research assistants and PhD student. The processed, pseudonymized data will be available for fellow researchers and made available (restricted access) for re-use. Requests for re-use of data after the completion of the research project will be evaluated by principal investigators BvdH, SvdO and TD, who will check whether the research question falls within the scope of the informed consent.

After the research project has been completed (i.e., data collection, data analysis and publishing of research articles completed) all the anonymized (i.e., the link with identifiable information will be deleted) digital data will be transferred to a study specific folder for long-term storage. The folder can only be accessed by the Principal Investigator and by authorized personnel after approval of the principal investigators. This data storage environment allows for careful access management and is only accessible with username and password. Data storage is backed-up automatically daily.

All the research data will be stored for 15 years after the data collection has been completed (i.e. last research assessment for the last participant has been performed). Raw audio-files will be stored for 15 years in an encrypted folder.

TREATMENT INTEGRITY

Intervention integrity will be assessed using procedures based upon work by Abikoff and colleagues (Abikoff et al., 2013; Abikoff et al., 2014). This implies that all intervention sessions will be audiotaped. For every child mental health worker delivering the training, the first session will be checked and child mental health workers will receive feedback on intervention integrity and fidelity by cognitive behavior therapists with extensive experience in behavioral parent training programs. Further, audiotapes of a subset of randomly selected sessions will be scored on intervention integrity and fidelity during the study by independent evaluators. After each session, the child mental health workers have to complete a fidelity checklist in which they will be asked which topics were covered.