A Randomized Controlled trial of the Remote Triple P Level 4 Group Intervention in Chile

Triple P level 4 group format delivered by online sessions

2026-07-20

2026-09-18

1) Total Difficulties Score from the Strengths and Difficulties Questionnaire (SDQ).
2) Prosocial Behavior subscale score from the SDQ.

The primary outcomes will be constructed from the Strengths and Difficulties Questionnaire (SDQ), a 25-item behavioral screening tool answered by caregivers using a 3-point Likert scale (0 = Not True, 1 = Somewhat True, 2 = Certainly True). After reverse-scoring the negatively worded items as per standard instrument guidelines, all scores will be calculated as the mean (average) score of their respective items rather than a simple sum. Consequently, all resulting variables will range continuously from 0 to 2.

SDQ Total Difficulties Score: This continuous variable will be constructed by calculating the mean score across 20 specific items that comprise the four problem domains. The final score ranges from 0 to 2, where higher scores indicate a higher severity of behavioral difficulties.

Prosocial Behavior Score: This continuous variable will be constructed by calculating the mean score of the remaining 5 items that measure positive social assets. The final score ranges from 0 to 2, where higher scores indicate greater prosocial strengths.

The questionnaire is divided into 5 distinct subscales (5 items each), which are defined as follows:

Emotional Symptoms: Assesses internalizing distress, including somatic complaints, worries, unhappiness, clinginess, and fears.

Conduct Problems: Measures externalizing behaviors, such as temper tantrums, compliance issues, fighting, bullying, and dishonesty.

Hyperactivity/Inattention: Evaluates core symptoms of ADHD, including restlessness, fidgeting, distractibility, impulsivity, and poor concentration.

Peer Relationship Problems: Captures difficulties in social interactions, such as isolation, lack of friends, being picked on, and a preference for adults over peers.

Prosocial Behavior: Measures positive social orientation, including consideration of others, sharing, helpfulness, and empathy.

The first four subscales (20 items in total) collectively determine the Total Difficulties Score. Additionally, these four individual subscales will be analyzed as separate outcomes, conditional on demonstrating satisfactory internal consistency reliability (McDonald's omega) within our study sample.

1) Total score and subscales scores of Parenting Style Questionnaire
2) Subscales scores of an adapted Parenting Questionnaire for the following dimensions: involvment, positive parenting, inconsistent discipline, monitoring and supervision, and developing new skills

1) Parenting Style Questionnaire: This instrument assesses standard parenting practices and styles. Items are scored on a 1-to-7 Likert scale. For this measure, we will construct both a global total score and individual subscale scores, which will be calculated as the mean (average) score of their respective items. Consequently, all resulting variables will range continuously from 1 to 7.
2) Adapted Parenting Questionnaire: This adapted version of the Alabama Parenting Questionnaire (APQ) will be used to measure specific domains of parenting using a 1-to-5 Likert scale. For this instrument, we will not construct a global composite score; instead, the analysis will be conducted exclusively at the subscale level. We will construct scores for five distinct subscales, distinguishing between the original and the adapted components of the instrument. The first three correspond to original subscales of the APQ: (1) Involvement, (2) Positive Parenting, and (3) Inconsistent Discipline. The remaining two represent dimensions that have been adapted for this specific study: (4) Monitoring and Supervision, and (5) Developing New Skills. Each subscale score will be calculated as the mean (average) score of its corresponding items, ranging continuously from 1 to 5. The analysis and reporting of these individual subscales will be conditional on each dimension demonstrating satisfactory internal consistency reliability within our study sample.

This study employs a two-arm randomized controlled trial (RCT) design to evaluate the impact of the online Triple P Level 4 Group intervention. The experimental process is structured into three sequential phases: recruitment, eligibility screening, and baseline assessment followed by randomization.

1. Recruitment and Eligibility ScreeningParticipants are recruited through a public outreach campaign conducted via social media platforms. To be eligible for the study, applicants must meet all of the following criteria:
Caregiver Status: Be the mother, father, or legal guardian of the target child.
Age Requirements: The caregiver must be of legal age (18 years old), and the target child must be between 4 and 12 years old.
Geographic Location: Reside in the Metropolitan Region of Chile.
Child Development: The target child must not have any caregiver-reported language disorders.
Program Exclusivity: The caregiver must not be currently participating, nor have previously participated, in any other parenting program.
Behavioral Concern Threshold: Caregivers must report concern regarding at least two child behavioral problems. This will be assessed during screening using a 10-item checklist adapted from the Strengths and Difficulties Questionnaire (SDQ) with binary response options ("concerned" vs. "not concerned"). Applicants endorsing two or more behavioral concerns wlil meet the eligibility threshold.

2. Baseline Assessment and Randomization
Once eligibility is confirmed, participants will complete a comprehensive baseline assessment. This assessment will be conducted face-to-face during home visits; however, the measurement instruments will be self-administered by the caregivers to minimize social desirability bias.
To prevent anticipation bias, randomization will occur strictly after the baseline data collection is completed. Randomization will be stratified based on two key variables: (1) whether the child has a developmental disorder (Autism Spectrum Disorder [ASD] or Attention Deficit Hyperactivity Disorder [ADHD]), and (2) the caregiver's educational level (whether they have completed higher education or not). This stratified assignment will be executed using the randtreat package in Stata, utilizing the global misfit method to handle uneven strata sizes and guarantee the targeted overall sample balance. Participants will be randomly assigned to either the treatment or control group using a 1:1 allocation ratio, resulting in an expected sample of 200 families assigned to the online Triple P intervention and 200 families assigned to the control group.

Not available

Randomization will be conducted in an office setting by a computer using statistical software (Stata, via the randtreat package).

Individual (the caregiver of the child)

No

400

400

200 caregirvers control, 200 caregivers Triple P level 4 group format

The Minimum Detectable Effect Size (MDES) for the main outcomes is 0.28 standard deviation units (Cohen's d).This power calculation was conducted via a sensitivity analysis for a fixed-effects ANCOVA model using G*Power, assuming the full target sample size of 400 families allocated into 2 groups. With a significance level (alpha) of 0.05, and a target statistical power of 0.80, the model can reliably detect a Cohen's f effect size of 0.14. For a two-group comparison, this translates directly to an MDES of 0.28 standard deviations.