The "Irie Classrooms Toolbox": a cluster randomised trial of a universal violence prevention programme in Jamaican preschools.

Violence is a leading worldwide public health problem with a very high prevalence in Jamaica. Interventions in early childhood are an important component in the primary prevention of violence. Training young children's caregivers in behavioural strategies to reduce child aggression and promote child social skills can 1) prevent the early development of antisocial behaviour in children and 2) reduce violence against children by caregivers. Corporal punishment is widely used in schools across Jamaica despite efforts by the Ministry of Education to eliminate it. Although banned by law in early childhood institutions, corporal punishment is still common in Jamaican preschools. The prevalence of conduct problems in young Jamaican children has not been accurately determined but estimates suggest it is between 12%-21%. Preschool teachers report child behaviour problems as a major concern and feel ill equipped to deal with them.
We have shown that teachers trained to use evidence-based child behaviour management strategies provide a more emotionally supportive classroom environment and use less corporal punishment. This translates to improvements in class-wide child behaviour; reductions in high-risk children's conduct problems and increases in their social skills at school and at home. Through several years working with preschool teachers in Jamaica, we have developed an intervention that involves guiding teachers in the use of a "toolbox" of key behaviour management strategies that are relevant, easy to use, effective and flexible. The intervention is delivered through participatory and practical methods and has been specifically designed to be scalable, low cost and suitable for use in low resource settings with teachers with limited formal teacher training.
We now plan to evaluate this intervention in 76 preschools (228 classrooms) to determine if benefits are obtained when the training is disseminated on a large scale. The preschools will be randomly assigned to a group in which all teachers receive the intervention in year 1 (38 preschools, 114 classrooms) and a second group that will receive the intervention a year later (38 preschools, 114 classrooms). The intervention will involve 5 full-day workshops, monthly in-class support and weekly text messages over the course of a school year. We will evaluate the effect of the intervention on class-wide measures of child aggression and teachers' use of corporal punishment and verbal abuse. We will also evaluate whether the intervention benefits the quality of the classroom environment, class-wide measures of children's prosocial behaviour, teachers' mental health and children's self-regulation, mental health and attendance. We will examine issues surrounding implementation of the intervention including measuring the degree to which teachers adopt the strategies in the classroom, factors affecting teachers' uptake of the intervention and teachers' views on the components of the toolbox and the training. This will help us to refine the toolbox to ensure its' suitability for implementation at scale. An economic evaluation will be conducted to ascertain the costs associated with delivering the intervention to teachers and the ratio of costs to the benefits obtained.
This study has the potential to make an enormous public health impact in Jamaica with benefits to reduced aggression and improved child mental health at the population level. We have close links with the government agencies and training institutions responsible for early childhood education in Jamaica and are confident that, if this study shows positive results, the intervention will be integrated into on-going teacher training initiatives. Furthermore, the intervention would be suitable for use at the preschool and early primary school level in other low and middle-income countries with potential to be an important component of global initiatives to prevent violence against children and to promote child mental health.

Aim: To determine the effects of school-based violence prevention programme, implemented in Jamaican preschools, on the levels of aggression among children in school and violence against children by teachers.
Study Design: Cluster randomized controlled trial
Sample: 76 preschools in urban areas of Kingston, Jamaica will be randomly assigned to intervention (n = 38) or control (n = 38). Each preschool comprises three classrooms (one 3, 4 and 5 year old class) and all teachers and classrooms will be included in the study (228 teachers/classrooms: 114 in each group). In addition, a random sample of 6 children in the 4 and 5-year old classes will be selected for evaluation of child-level outcomes (912 children: 456 children in each group).
Intervention: The intervention involves training teachers in classroom behaviour management and in strategies to promote children's social-emotional competence. Training is delivered through 5 full-day workshops, monthly in-class coaching over two school terms and weekly text messages.
Measurements: The primary outcome measures are: i) observed violence against children by teachers and ii) observed levels of child aggression in the classroom. Secondary outcomes include observations of the quality of the classroom environment, teachers' reports of their mental health, teacher reported child mental health, direct tests of child self-regulation, child reports of teachers' use of corporal punishment and child attendance.
Economic, Social, Qualitative Measures: Implementation science is incorporated throughout the study to determine the factors affecting the quality of implementation and how implementation is associated with outcomes. An economic evaluation will also be conducted.
Implications: This study involves integrating an intervention into an existing service that is universally accessed by young Jamaican children. The intervention thus has the potential to make an enormous public health impact at the population level.

The most direct beneficiaries of this research will be the participant teachers and children. In addition, if teachers continue to use these strategies, they will continue benefiting new cohorts of children over time.
There is a high level of interest and concern around the issue of corporal punishment in schools in Jamaica and we anticipate wide interest in the results of this study. Our dissemination activities (as listed in the communications plan) should raise awareness in Jamaican society that alternative behaviour management strategies are available and effective. There is also great concern about the increasing levels of violence and aggression by Jamaican children in school and this study will inform strategies to reduce this by demonstrating the value of appropriate behaviour management and explicit teaching of social skills for appropriate child behaviour.
If the intervention is effective, our greatest potential impact within Jamaica would result from wide-scale adoption by the MOE, the ECC, the teacher training colleges and NCTVET. These institutions are actively seeking appropriate strategies to eliminate the use of corporal punishment in schools and this intervention is suitable for integration into pre-service and in-service teacher training. Integrating the intervention into existing teacher training initiatives has the potential for a large impact in terms of reducing violence against children and improving child mental health at the population level in Jamaica. Given our close links with these organisations, we anticipate that uptake of this intervention should occur over the short to medium term with benefits seen within five years from the end of the study.
The intervention has been designed to be suitable for implementation in low-resource settings with teachers with low levels of training in education. If the intervention implemented at scale is effective at reducing violence against children by teachers and improving child behaviour in the Jamaica early childhood setting, we will work with colleagues in the UNICEF regional office and the Inter-American Development Bank (IDB) to disseminate the intervention more widely across Latin America and the Caribbean. The intervention also has the potential to be adapted for use with early childhood practitioners in other low and middle income countries as it uses practical and participatory training methods to teach key behaviour principles that we have found to be relevant, easy to use, effective and feasible for poorly trained teachers, with limited resources and high child-staff ratios. Professor Walker is currently PI on a grant from Grand Challenges Canada to produce and disseminate an evidence-based parenting intervention package suitable for use in LMIC on the web. If the proposed intervention is effective, we plan to add the teacher-training package to the website, thus making it available to the international community.
TMRI has an international reputation for conducting high quality research addressing important public health problems for the region and beyond and is a prominent training institute for public health researchers in the Caribbean region. This study will support several graduate students in Jamaica who have been working with Dr. Baker-Henningham to develop and evaluate this intervention. Building the research capacity of junior Jamaican staff will facilitate the continuation of this programme of research in the future.
The training is based on sound theoretical principles with demonstrated effectiveness in the Jamaican setting. We are closely monitoring implementation dosage and quality throughout the trial through qualitative and quantitative methods to evaluate how this affects behavioural outcomes. This aspect of the study will contribute to the literature on implementation and dissemination of evidence-based interventions in service settings regardless of the outcome of the trial on teacher and child behaviour.