A mixed-method feasibility trial of an early childhood, violence prevention, parenting program integrated into preschool provision in Jamaica

Baker-Henningham, Helen; Francis, Taja; Bowers, Marsha Working Paper A mixed-method feasibility trial of an early childhood, violence prevention, parenting program integrated into preschool provision in Jamaica IDB Working Paper Series, No. IDB-WP-1703 Provided in Cooperation with: Inter-American Development Bank (IDB), Washington, DC Suggested Citation: Baker-Henningham, Helen; Francis, Taja; Bowers, Marsha (2025) : A mixedmethod feasibility trial of an early childhood, violence prevention, parenting program integrated into preschool provision in Jamaica, IDB Working Paper Series, No. IDB-WP-1703, Inter-American Development Bank (IDB), Washington, DC, https://doi.org/10.18235/0013518 This Version is available at: https://hdl.handle.net/10419/324788 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ A Mixed-method Feasibility Trial of an Early Childhood, Violence Prevention, Parenting Program Integrated into Preschool Provision in Jamaica Helen Baker-Henningham Taja Francis Marsha Bowers WORKING PAPER No IDB-WP-1703 InterA merican Development Bank Gender and Diversity Division May 2025 * Caribbean Institute for Health Research (CAIHR), University of the West Indies (UWI), and Bangor University ** CAIHR, UWI A Mixed-method Feasibility Trial of an Early Childhood, Violence Prevention, Parenting Program Integrated into Preschool Provision in Jamaica Helen Baker-Henningham* Taja Francis** Marsha Bowers** InterA merican Development Bank Gender and Diversity Division May 2025 Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Baker-Henningham, Helen. A mixed-method feasibility trial of an early childhood, violence prevention, parenting program integrated into preschool provision in Jamaica / Helen Baker-Henningham, Taja Francis, Marsha Bowers. p. cm. — (IDB Working Paper Series ; 1703) Includes bibliographical references. 1. Violence-Prevention-Jamaica. 2. Early childhood education-Jamaica. 3. Children and violence-Jamaica. 4. Child abuse-Prevention-Jamaica. I. Taja, Francis. II. Bowers, Marsha. III. Inter-American Development Bank. Gender and Diversity Division. IV. Title. V. Series. IDB-WP-1703 http://www.iadb.org Copyright © 2025 Inter-American Development Bank ("IDB"). This work is subject to a Creative Commons license CC BY 3.0 IGO (https://creativecommons.org/licenses/by/3.0/igo/legalcode). The terms and conditions indicated in the URL link must be met and the respective recognition must be granted to the IDB. Further to section 8 of the above license, any mediation relating to disputes arising under such license shall be conducted in accordance with the WIPO Mediation Rules. Any dispute related to the use of the works of the IDB that cannot be settled amicably shall be submitted to arbitration pursuant to the United Nations Commission on International Trade Law (UNCITRAL) rules. The use of the IDB's name for any purpose other than for attribution, and the use of IDB's logo shall be subject to a separate written license agreement between the IDB and the user and is not authorized as part of this license. Note that the URL link includes terms and conditions that are an integral part of this license. The opinions expressed in this work are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent. Abstract* We adapted a violence-prevention, parenting program (the Irie Homes Toolbox, or IHT) for integration into Jamaican preschool services. The adapted IHT was evaluated in a mixed-method feasibility trial in Kingston, Jamaica. Twenty-four preschools were randomly assigned to intervention (n=12) or wait-list control (n=12). Ten caregivers per school were recruited (n=240, n=120/group). The program consisted of eleven 1-hour parenting sessions delivered by a preschool teacher with groups of ten caregivers of children aged 2-6 years. In the impact evaluation, the primary outcome was caregivers’ use of violence against their child (VAC). Secondary outcomes were caregivers’ involvement with their child, attitude to VAC, preferences for harsh punishment, self-efficacy, and child conduct problems. All outcomes were measured by caregiver-report, and we test for and find no evidence of social desirability bias. We measured fidelity of implementation on an ongoing basis. We also conducted in-depth interviews with participating teachers and kept ongoing logs on intervention implementation. Participants attended a mean (SD)=4.0(3.1) sessions. The IHT intervention led to reductions in caregivers’ use of VAC (ES=-0.22, p=0.04) and caregivers’ favorable attitudes to VAC (ES=-0.36, p=0.01), and increases in caregivers’ involvement with their child (ES=0.30, p=0.005) and parenting self-efficacy (ES=0.29, p=0.02). Reductions in caregiver preferences for harsh punishment were significant at p=0.07 (ES=-0.21). We found no benefits to child conduct problems. Through observations of session quality, interviews with preschool teachers, and research team logs, we identified enablers and barriers to intervention implementation and suggestions for improvement. The program has potential for large-scale dissemination to reduce VAC in Jamaica. JEL classifications: I10, I20, J12, J13 Keywords: Violence prevention, Violence against children, Preschool, Parenting intervention * Acknowledgements: The study was funded through GDLab, the Gender and Diversity Knowledge Initiative of the Inter-American Development Bank. We thank the advisory board for their support and for their useful comments and feedback. Special thanks to the advisors Amber Peterman and Andrés Moya for their detailed reviews of the working paper. We also thank the preschool teachers and caregivers who participated in the study and the research assistants for data collection. 2 1. Introduction Violence against children (VAC) is a global public health problem with high prevalence across Latin America and the Caribbean (LAC), where two-thirds of children aged 2 to 14 years experience violent discipline at home (UNICEF, 2022). Jamaica ranks third highest across the LAC region in terms of VAC (Haiti and Suriname rank first and second) (UNICEF, 2022). Evidence from meta-analyses show that caregiver-training interventions in the early childhood years have potential to reduce child maltreatment (Chen and Chan, 2015; Vlahovicova et al., 2017), with some evidence of sustained effects over the short term, albeit with diminished effects (Backhaus et al., 2023). There is also growing evidence from lowand middle-income countries (LMIC) that caregiver training programs can be effective in reducing VAC (Wang and Zhang, 2024), including in the LAC region (Mejía et al., 2015; Santini and Williams, 2017; Skar et al., 2017; Altafim and Linhares, 2019). However, few evidence-based parenting programs have been implemented at scale (Britto et al., 2018; Sanders et al., 2022). To maximize scalability and sustainability, evidence-based interventions need to be integrated into existing government services and be delivered by existing staff. We have previously developed an early childhood, violence prevention, parenting program in Jamaica, the Irie Homes Toolbox (IHT). The IHT targets caregivers of children aged 2 to 6 years and includes content on promoting child positive behavior, preventing child negative behavior, helping children understand their emotions, promoting caregiver emotional self-regulation, and managing child misbehavior. In previous trials, we have shown that the IHT is effective at reducing VAC by caregivers when delivered through in-person and through virtual modalities (Francis and Baker-Henningham, 2021; Dinarte-Díaz et al., 2023). The challenge now is to implement the program on a wider scale and to achieve this aim, we need to investigate how to integrate the IHT into the Jamaican early childhood educational network and how it can be delivered by preschool teachers as part of their routine duties (Baker-Henningham, Bowers, and Francis, 2023). In this paper, we report on a mixed-method feasibility study of the IHT program when delivered by preschool teachers. The evaluation included an ongoing process evaluation, an impact evaluation, and a qualitative investigation of the IHT program. In the study, twenty-four community preschools were randomized to an intervention group (n=12 preschools) or a wait-list control group (n=12 preschools). We recruited 10 parents in each preschool to participate in the intervention (n=120 parents/group). Two teachers in each intervention preschool were trained to 3 deliver the IHT sessions. The program was delivered through eleven, weekly, 1-hour group sessions, with groups of ten caregivers. Participants were also sent materials, messages and short videos via WhatsApp between sessions (Francis, Bowers, and Baker-Henningham, 2025). In the process evaluation, we measured caregiver attendance, preschool teacher compliance, and the quality of implementation. In the impact evaluation, we examined the effectiveness of the IHT in reducing caregivers’ use of VAC. We also measured impacts from the IHT to caregiver involvement, attitudes to VAC, preferences for harsh punishment, parenting self-efficacy, and child conduct problems. In the qualitative investigation, we investigated the enablers and barriers to implementation and suggestions for improvement from the perspective of the preschool teachers and the research team. Caregivers attended a mean (SD) of 4.0 (3.8) sessions. Across the twelve schools allocated to intervention, teachers delivered a mean (SD) of 10.1 (1.2) sessions. Observations of session quality indicated that teachers adhered to the session scripts, formed positive relationships with participants, engaged participants well, used clear demonstrations and gave all participants the opportunity to practice. Weaknesses included inadequate preparation, not highlighting key points, and insufficient support and feedback to participants during practice activities. Intention to treat estimates indicate that the intervention reduced VAC by caregivers by 0.22 SD (p=0.04). Caregivers in the intervention group also increased their involvement with their child by 0.30 SD (p=0.005), reported less favorable attitudes to VAC by 0.36 SD (p=0.01) and increased parenting self-efficacy by 0.29 SD, (p=0.02). Caregiver reports of preferences for harsh punishment reduced by 0.21 SD (p=0.07). No reductions were identified for caregiver reports of child conduct problems (effect size=-0.06 SD, p=0.49). As the outcomes were measured through caregiver self-report, we addressed experimenter demand effects by examining the effect of treatment on a social desirability index (SDI) that measured participants’ propensity to respond in a socially acceptable way. There was no difference in SDI across treatment groups. In addition, controlling for SDI in the multilevel regression analyses did not impact the magnitude or significance of the impacts. There was also no evidence of heterogeneous effects of SDI across treatment groups for any of the outcome measures. Key enablers to implementation from the perspective of the teachers and the research team were that teachers were very positive about the IHT program, valued the materials used, were motivated to conduct the sessions and had good relationships with caregivers. Teachers reported 4 benefits to their professional and personal development, and to caregivers’ knowledge and behavior. Key barriers to implementation were teachers’ difficulties in using higher-order facilitator skills in session delivery and poor caregiver attendance. The skills that were difficult for teachers included reflective listening, dealing with resistance, engaging participants in collaborative problem-solving, and addressing caregivers’ favorable attitudes toward VAC (especially as some teachers also had positive attitudes toward VAC). Teachers were also often insufficiently prepared to conduct the sessions. One suggestion for improvement was to give all teachers in the school a role in the IHT program to embed the intervention more effectively into the school culture. For example, other teachers could help with caregiver recruitment and engagement and with promoting the strategies with caregivers of children in their class during their regular interactions with caregivers. Some teachers suggested shortening the program and/or altering session frequency to make it more feasible to fit into the school termly timetable. The research team identified a need for more training for teachers to ensure that they believed in and were competent in using the strategies themselves, and to better equip them in addressing caregiver resistance and responding to queries. The overall objective of the study was to prepare the IHT for implementation at scale. In this study, we found that the IHT was effective at reducing caregivers’ use of VAC when delivered by preschool teachers and the teachers valued the program and showed high compliance in conducting sessions. This gives us confidence that the program has potential for wider dissemination within the early childhood educational network. Conducting a feasibility study has provided the opportunity to identify implementation problems prior to implementing the program on a larger scale. We will use the findings from this study to inform further revisions to the IHT program and to develop training, supervision, and monitoring tools to promote the fidelity of intervention implementation when implemented on a larger scale. 2. Literature Review Globally, over one billion children are affected by violence each year (Hillis et al., 2016). Violence against children (VAC) is a violation of child rights, has high prevalence in lowand middleincome countries (LMICs), is associated with long-term negative effects on child functioning, and with high economic and social costs (McCoy et al., 2022; Cuartas et al., 2019; Hillis et al., 2017; Heilmann et al., 2021; Perenieto et al., 2014). In Jamaica, 84% of caregivers of children aged 2-4 5 years report using physical violence, and 71% report using psychological aggression demonstrating an urgent need for violence-prevention programming (Lansford and DeaterDeckard, 2012). This need has been recognized at the national level, as Jamaica is a pathfinder country in the Global Partnership to End Violence Against Children and the government has launched “The National Plan of Action for an Integrated Response to Children and Violence” (Government of Jamaica, 2018). To respond to the need for violence-prevention programming, we have developed and evaluated two complementary early childhood programs: i) to reduce teachers’ use of VAC at school: the Irie Classroom Toolbox (BakerHenningham, 2018; Baker-Henningham et al., 2019; Baker-Henningham et al., 2021; Bowers, Francis and Baker-Henningham, 2022), and ii) to reduce caregivers’ use of VAC at home: the Irie Homes Toolbox (IHT) (Francis and BakerHenningham, 2020; Francis and Baker-Henningham, 2021; Dinarte-Díaz et al., 2023; Francis et al., 2024). Irie is a Jamaican term that means “good” and “at peace with oneself and the world.” The programs target caregivers of children aged two to eight years, and the aim is to reduce VAC at the population level by integrating these programs into the early childhood educational network. We have been working with the Early Childhood Commission (ECC) in Jamaica to scale up the Irie Classroom Toolbox. The ECC is an agency of the Ministry of Education that has oversight for all early childhood institutions and coordinates all activities within the early childhood sector. The IHT was developed as a complementary program to the Irie Classroom Toolbox to promote consistent behavior management strategies by both teachers and parents. The next step is to also begin to scale up the IHT. The IHT was specifically designed to be feasible, relevant, effective, and scalable in the Jamaican context (Francis and Baker-Henningham, 2020). The content of the IHT includes core components of evidence-based, violence-prevention parenting programs operationalized for the Jamaican context. These core components address the proximal drivers of VAC including caregivers’ knowledge, attitudes, and skills relating to positive and non-violent parenting, caregiver emotional self-regulation, and caregiver self-efficacy. Violence is a complex, multidimensional problem with drivers at the level of individual (e.g., child age and behavior), caregiver (e.g., education, exposure to adversity), household (e.g., interpersonal violence, poverty, number of children), community (e.g., community violence), and societal (e.g., social norms) levels. However, these distal drivers require a multi-sectoral approach to violence prevention, as 12 Panel 1 (continued) Intervention Materials for Facilitators: Materials to conduct the parenting sessions included: i) a fully-scripted manual containing a script for each session, ii ) visual aids used to introduce and practice strategies and to help participants understand child misbehaviors, iii) hand-held charts with key points to reinforce the strategies introduced, iv) a picture of the Irie Tower: a pictorial representation of the strategies introduced in the program, v) picture books and low-cost, shopbought toys to practice Irie Time, and vi) phone cards with US$3.50 credit per week to facilitate teachers’ communication with caregivers. Materials for the school-wide element included: i ) a weekly poster to be placed on the school information board with details of the Irie strategy of the week, and ii) a weekly guide for all teachers in the school on how to engage caregivers in the Irie strategy of the week Intervention Materials for Caregivers Enrolled in Parenting Sessions: included: i) a homework record form (Irie Challenge) for each session to encourage participants to use the strategies at home, ii) three small picture books for parents to use with their child during Irie Time, and iii) an Irie Parent Oath that each participant signs on completion of the program to commit to continued use of the strategies. Caregivers were also provided with a small snack (juice and peanuts) after each session. Who provided: Two preschool teachers within each school were trained to deliver the IHT sessions. The teachers were given autonomy in how to organize the sessions, with some teachers choosing to have a “lead” teacher taking primary responsibility for the sessions, some teachers taking it in turns to deliver sessions and some delivering the sessions together. Preschool teachers were trained and supervised by two members of the Irie Toolbox team (TF and MB). Where: The parent training sessions were held on the preschool compound. Teacher-training workshops were conducted in a centrally located community preschool. When and How Much: Parenting sessions were held either at the beginning of the school day (when caregivers were dropping off their child at school) or after school (when caregivers came to pick up their child). The sessions were held once a week, for up to 11 weeks, and each session lasted approximately one hour. Teachers were to be trained in a 2-day initial workshop facilitated by two members of the Irie Toolbox Team (TF and MB) in groups of 12 teachers, and to receive fortnightly support from the two members of the Irie Toolbox team with six schools allocated to each team member. Tailoring and Modifications: Each session was fully scripted and implemented in a similar way across all preschools. However, there was flexibility in the number of practice activities completed based on parent interest and the time available. Seven schools did not complete all eleven sessions due to delays in starting the intervention (see deviations from protocol section for more details). 13 3.4 Measures In this mixed-method study, we collected three main categories of data: i) implementation outcomes, ii) the impact of the IHT program on caregiver and child outcomes, and iii) a qualitative investigation of the perceptions of preschool teachers and the research team. Details of these measures are given in Panel 2. Implementation Outcomes. Implementation outcomes included caregiver attendance, preschool teacher compliance and the quality of the parenting sessions implemented by the preschool teacher. Impact Evaluation. The primary outcome in the impact evaluation was caregivers’ use of VAC measured through caregiver report. Secondary outcomes were caregiver involvement with their child, caregivers’ attitudes to VAC, caregivers’ preference for harsh punishment, caregiver selfefficacy, and child conduct problems, all by caregiver report. All primary and secondary outcomes were measured at baseline (December 2023 to March 2024) and post-intervention (May to July 2024), with the exception of caregiver self-efficacy, which was measured at post-intervention only. All outcome measures, except for caregiver preferences for harsh punishment, have been used previously with this population in Jamaica and show good psychometric properties (see Panel 2). The measure of caregiver preferences for harsh punishment was designed for this study using an adapted version of a questionnaire used in Uganda (Satinsky et al., 2024). As these outcomes are all measured through caregiver-report, we included a social desirability index (SDI) in post-test measurements. The aim was to measure respondent’s propensity to report in a socially desirable way on their own behavior and attitudes. We used the Marlowe-Crowne Social Desirability Scale – short form (Crowne and Marlowe, 1960; Reynolds, 1982) which consists of 13 questions with a response scale of “True” or “False” (for example, “I sometimes feel resentful when I don’t get my way,” “I have never deliberately said something that hurt someone’s feelings,” “I’m always willing to admit it when I make a mistake.”). Previous studies have demonstrated significant correlations between SDI and self-reports of physical and psychological violence (Bell and Naugle, 2007; Fernández-González et al., 2013). We piloted the SDI prior to collecting baseline measurements and test-retest over 2 weeks was ICC=0.60 and internal reliability was α=0.65 (n=31). 14 Qualitative Data. We also collected qualitative data including the perceptions of the implementing teachers and the perspectives of the research team on the enablers and barriers to implementation and suggestions for improvement. During our studies on the prevention of VAC, we follow the reporting requirements of the Jamaican Child Protection Act that mandates reporting of severe corporal punishment. The limits of confidentiality based on this reporting mandate are included in the informed consent process. The measure of caregivers’ use of VAC used in this study is a measure that was designed for use in prevention studies with community samples, and as such, we do not ask caregivers to report on severe corporal punishment. We follow a structured research protocol to collect data on caregivers’ use of VAC that includes guidelines for training interviewers on how to conduct interviews in a sensitive way (including how to respond to participant distress and how to respond to disclosure of severe VAC), interviews conducted out of listening range of other persons, and referral mechanisms to additional services if required. Panel 2. Description of Implementation, Primary, and Secondary Outcomes and Qualitative Evaluation Outcome Variables Measures Used Implementation Outcomes Caregiver attendance Measured weekly through attendance logs Preschool teacher compliance Teacher compliance with conducting sessions was measured weekly using project logs. Compliance with sending out WhatsApp messages to participating parents was documented through mobile phone records. Session quality Measured through observation by a member of the research team using a structured questionnaire that includes rating teacher preparation, skills in session delivery, skills in building positive relationships with caregivers, and skills in helping participants learn. Ratings were conducted on a five-point scale: 1=poor, 2=pass , 3=average, 4=good, 5=excellent. 15 Primary Outcome Caregivers use of violence against their child (VAC) Measured using items from the corporal punishment and psychological aggression subscales of the Conflict Tactics Scale Parent Child (Strauss et al., 1998). The scale had ten questions, five questions measure physical violence and five measure psychological aggression: 1) shake, 2) hit on the bottom with bare hand, 3) pinch, 4) hit on the bottom with something hard (e.g., belt, stick), 5) slap on the arms, hand or leg, 6) shout, yell or scream, 7) threaten to hit, 8) call names like idiot, dummy, 9) threaten to send the child away, 10) swear at child. Caregivers report on the last two weeks and responses are given on a seven-point frequency scale from 1=never to 7=more than once a day. Higher scores indicate higher caregiver use of VAC. Test retest over two weeks (n=20) was intra-correlation coefficient (ICC)=0.88. Internal reliability using Cronbach’s alpha (α)=0.71. Secondary Outcomes Caregiver involvement with their child The scale consists of twelve questions answered on a seven-point frequency scale (from 1=never to 7=more than once a day): 1) reading storybooks, 2) helping with homework, 3) playing games inside the home, 4) playing outside, 5) play with toys, 6) sit with child as they write, draw or color, 7) chat with child about school and/or friends, 8) involve child in chores, 9) chat with child during daily routines (e.g., dressing, bathing), 10) teach child household rules, 11) praise child, 12) spend 10-15 minutes with child doing something fun. Higher scores indicate higher caregiver involvement. Test retest over two weeks (n=20): ICC=0.96. Internal reliability: α=0.76. Caregiver attitude to VAC Measured using an adapted version of the UNICEF Multiple Indicator Cluster Survey (MICS) questionnaire used previously in Jamaica (Dinarte-Diaz et al., 2023). The scale consisted of five questions answered on a four-point scale from 1=disagree completely to 4=agree completely: three questions measured attitudes to physical violence and two measured attitudes to psychological aggression. Higher scores indicate more favorable attitudes to VAC. Test retest over two weeks (n=21): ICC=0.71. Internal reliability: α=0.63. Caregiver preferences for harsh punishment The scale consists of nine pictorial scenarios of children misbehaving: four scenarios in a public place (bank, market, supermarket, school) and five scenarios at home. The child misbehaviors include non-compliance, giving attitude, temper tantrum, hitting a younger child, doing something dangerous, being overly active, and making a mess. We selected these behaviors as in previous qualitative research, caregivers reported that they often responded to these behaviors with physical violence (Francis et al., 2023). For five of the questions, if the caregivers’ response(s) do not include physical punishment, we ask what they would do if the behavior continued. A pictorial response scale is provided, and multiple responses are accepted. The score includes the number of responses that include VAC (e.g., hit, shout or yell, threaten to hit). Higher scores indicate higher preferences for harsh punishment. Test retest over two weeks (n=30): ICC=0.80. Internal reliability: α=0.57. 16 Secondary Outcomes (continued) Child conduct problems Measured using the Eyberg Child Behavior Inventory (ECBI) frequency scale (Eyberg and Ross, 1978). The ECBI consists of 36 questions and caregivers report on their child’s behavior using a 7-point scale from 1=never to 7=Always. Higher scores indicate higher levels of conduct problems. Test retest over two weeks (n=20): ICC=0.99. Internal reliability: α=0.81 Parenting selfefficacy Measured using the Brief Parenting Self-Efficacy Scale (Woolgar et al., 2023). The scale consists of five questions and caregivers answered on a four-point scale from 1=disagree completely to 4=agree completely. Higher scores indicate higher parenting self-efficacy. Test retest over two weeks (n=30): ICC=0.58. Internal reliability: α=0.69. Qualitative data Teacher perceptions After teachers had completed all IHT sessions, we conducted in-depth interviews with teachers who conducted the sessions within each school. Interviewers used a topic guide to explore teachers’ perceptions of the enablers and barriers to intervention implementation (relating to the content, process, structure, and materials used in the program) and teachers’ suggestions for improvement and to promote sustainment within the preschool system. Research team perceptions The research team kept an ongoing log of the enablers and barriers to intervention implementation and suggestions for improvement based on their interactions with the preschool teachers and parents and their observations of the sessions. 3.5 Procedure for Measures Implementation Outcomes. Implementation outcomes were collected throughout intervention implementation by two members of the Irie Toolbox Team (TF and MB). Each team member was responsible for collecting implementation data in six preschools. Impact Evaluation. The questionnaires related to caregivers’ use of VAC, caregiver involvement with their child, attitudes to VAC, preferences for harsh punishment, self-efficacy, and child conduct problems were administered in face-to-face interviews by two research assistants, masked to study design and group allocation. Interviews were usually conducted at the study preschools with a minority of interviews conducted at the participants’ homes. All participating caregivers (in both intervention and control schools) received a small gift for their child (e.g., coloring book and crayons) and US$2.50 phone credit after baseline and again after post-test measurements. Research 17 assistants were rotated across schools and interviewed equal numbers of caregivers in each group at both time points. Research assistants were trained over a two-week period including 0.5 weeks in office and 1.5 weeks of field training. Ongoing quality control was maintained by the trainer accompanying each research assistant once a week throughout data collection. Completed questionnaires were checked on a weekly basis for completeness by the project manager. Qualitative Data. Teacher interviews were conducted within two months after the end of the intervention. Teachers were contacted by phone and a convenient time arranged for interview. Interviews were conducted at the preschool by the Irie Team member who was responsible for supporting the teachers. Interviewers used a structured topic guide, and the interview lasted approximately one hour. 3.6 Deviations from Protocol In the protocol, the intervention was described as a twelve-week intervention. We found that when integrating into the existing preschool system, we had to follow the pace of the schools. Teachers had responsibility for identifying the participant parents and setting up their parenting groups (including setting the date and time and creating a WhatsApp group of participants). These activities took more time than we anticipated. Hence the intervention was shortened to a maximum of eleven weeks. We added a measure of caregiver self-efficacy to the measurements at post-test. This measure was added after feedback from the advisory board overseeing the research. 3.7 Analysis This is a feasibility pilot study, and its power is limited by the small number of clusters. With a cluster size of 10 (10 parent/child dyads per school), and assuming an intracluster correlation coefficient of 0.02 on our primary outcome of caregivers’ use of VAC, (based on our previous data), 120 participants per group are sufficient to detect an effect of 0.35 SD with 80% power at a significance of p<0.1. All analyses were prespecified with details given in a previously published protocol (Francis et al., 2025). Outcome variables were assessed for normality and transformed where necessary. Caregivers’ use of VAC, caregivers’ attitude to VAC, and caregiver preferences for harsh punishment were negatively skewed, and were normalized with a square root transformation. 18 The effect of intervention was analyzed using MLWin v 3.04 (Rasbash et al., 2009), on an intention to treat basis, using multilevel linear regression models to take into account the hierarchical nature of the data (caregivers nested in schools). Fixed effects in the model included a constant, baseline score (where available), interviewer, child age and sex, intervention status and variables that were different between the groups at baseline, or that were different between those lost to analysis and those who continued, at a significance level of p<0.1. Preschool was entered as a random effect. As there was only one primary outcome (caregiver-reported frequency of violence against their child), we did not control for multiple hypothesis testing. Secondary outcomes are considered exploratory. Effect sizes were calculated by dividing the regression coefficient by the standard deviation of the control group at post-test for each outcome. As the outcome measures are assessed through caregiver self-report, they are susceptible to experimenter demand effects especially for outcomes related to caregivers’ use of and attitudes to VAC. We assessed experimenter demand effects using three sets of multilevel regression analyses. In all analyses, child age, sex, interviewer, caregiver age, and baseline caregiver involvement were entered as fixed effects and preschool as a random effect. First, we investigated the effect of the intervention on a social desirability index using multilevel regression analysis. Second, we repeated all the multilevel regression analyses on primary and secondary outcomes described above, with the social desirability index added as a covariate in the model. Finally, we investigated heterogeneity of experimenter demand effects by re-running the multilevel regression analyses for each primary and secondary outcome with the social desirability index (SDI) and an interaction term (SDI x study group) as fixed effects in the model. For all multilevel regression analyses, we used the restricted maximum likelihood estimator available in MLWin to take into account the small number of clusters (Elff et al., 2021). The in-depth interviews with teachers were audio-recorded. Due to resource constraints, we made notes from the audio-recording of each teacher interview rather than transcribe the interviews in full. The qualitative data, including the notes from the teachers interviews and the ongoing logs from the research team during the IHT implementation, were analyzed manually using thematic analysis (Braun and Clarke, 2006). We developed a list of key themes using a deductive approach separately for each set of qualitative data (teacher interviews and researcher logs). One of the Irie Team members (TF) created a thematic matrix linking each theme and 19 subtheme with corresponding extracts from the qualitative data. The analyses were conducted with ongoing input and consultation with HBH. 4. Results 4.1 Sample Characteristics Sample characteristics are shown in Table 1. Children had a mean (SD) age of 4.1 (1.1) years, 118 (49.2%) were boys, and 49 (20.4%) were in the clinical range for conduct problems by caregiverreport, with no significant differences by study group. In terms of caregivers and household characteristics, 172 (71.7%) had completed high school, 176 (73.3%) were employed, 227 (94.6%) were female, 205 (85.4%) were mothers, 91 (37.9%) fathers lived with their child and there was a mean (SD) of 2.2 (0.98) children per household. The only significant differences between the groups, (at p<0.1), were for caregiver age (p=0.07) and caregiver involvement with their child (p=0.07). Seven caregiver/child dyads were lost to follow up: four from the intervention group and three from the control group: reasons for loss are shown in Figure 1. There were no differences between those lost and followed up, (at a significance of p<0.1), for any primary or secondary outcome or for any caregiver characteristics (See Appendix). However, children lost to follow-up were significantly older than those who continued (Mean (SD): 5.0 (1.2) versus 4.1 (1.1) respectively, p=0.03). Caregiver age, caregiver involvement and child age were entered as fixed effects in all multi-level regression analyses. 20 Table 1 Child, Caregiver and Household Characteristics by Study Group Intervention (n= 120) Control (n= 120) p value Child age (in years) 4.07 (1.12) 4.08 (1.08) 0.91 Child sex (n (%) boys)) 56 (46.7) 62 (51.7) 0.44 Clinical range for conduct problems by parent report1 n (%) 23 (19.3) 26 (22.0) 0.61 Caregiver age (in years) 33.92 (9.99) 31.79 (8.13) 0.07 Caregiver sex (n (%) female) 115 (95.8) 112 (93.3) 0.39 High school completed n (%) 86 (71.7) 86 (71.7) 1.00 Caregiver currently employed n (%) 89 (74.2) 87 (72.5) 0.77 Mother lives with child n (%) 108 (90.0) 113 (94.2) 0.69 Father lives with child n (%) 41 (34.2) 50 (41.7) 0.31 Number of children < 18 years in household 2.13 (1.18) 2.33 (1.18) 0.19 Household possessions 2 9.44 (2.22) 9.30 (2.42) 0.64 Household sanitation (median (IQR)) 3 12.00 (8-12) 12.00 (8-12) 0.98 Crowding median (range)4 1.33 (0-5) 1.50 (0-8) 0.64 Notes: Scores are mean (SD) unless otherwise stated. 1Above cut off (>130) on Eyberg Child Behavior Inventory by parent report. 2Number of possessions from a list of 16 items. 3Water supply and toilet. 4Number of people per room. 4.2 Characteristics of Intervention Teachers Twenty-four preschool teachers (n=2 from each preschool) participated in program implementation. Teacher characteristics are shown in Table 2. All teachers were female, 6 (25%) were also the school principal, in addition to being a classroom teacher, and 11 (45.8%) had a formal teacher-training qualification (Diploma or B.Ed.). 21 Table 2. Characteristics of Teachers Implementing the Irie Homes Toolbox Program n=24 Teacher sex: n (%) female 24 (100%) Teacher age: • 25-34 8 (33.3%) • 35-44 8 (33.3%) • 45-54 4 (16.7%) • 55-64 4 (16.7%) Teacher role: n (%) • Teacher 18 (75%) • Teacher & Principal 6 (25%) Number of years teaching: median (IQR) 8 (5-16) Number of years teaching at this school: median (IQR) 5 (1.5-10) Received training in early childhood education: n (%) 22 (95.7%) Have a Diploma in Education: n (%) 1 (4.2%) Have a Bachelor’s in Education: n (%) 10 (41.7%) 4.3 Intervention Implementation Teacher Participation in Training. Teachers were trained in two rounds of implementation: one group (n=12 teachers) was trained in early January 2024, and one group (n=12 teachers) was trained in the February half term. In the January cohort, 8/12 teachers (66.7%) attended two days of training, 4/12 (33.3%) attended one day. In the February cohort, teachers were only available for one day of training: 10/12 teachers (83.3%) attended the training. So, overall, (8/24 (33.3%) attended two days, 14/24 attended 1 day (58.3%) and 2/24 (8.3%) attended no training. Schools were visited by an Irie Toolbox team member the day before each session for approximately 30 minutes to review the script and over 90% of sessions were attended by a member of the team. Number of Sessions Delivered and Teacher Compliance. Five schools completed all 11 sessions, five schools completed 10 sessions, one school completed 9 sessions, and one school only completed 7 sessions. The reason why some schools completed fewer than 11 sessions was due to a delay in setting up the program. This occurred because of delays in participant recruitment and/or in organizing the time and date of the first session. Teacher compliance with implementing the sessions was high and sessions were only missed due to poor weather, the session falling on a 28 4.5 Qualitative Investigation For the qualitative investigation, we selected teachers who had been actively involved in conducting the IHT session to participate in the in-depth interview. In six schools, both teachers implemented the sessions, while in six schools one teacher took primary responsibility for conducting the sessions. Hence we invited two teachers from six schools and one teacher from the other six schools to participate. Interviews were conducted during the time that Hurricane Beryl impacted Jamaica in July 2024, and one teacher was unavailable for interview. We therefore interviewed seventeen teachers from eleven preschools. Teachers’ Perspectives of the Enablers to Implementation. Teachers reported benefits to themselves from their involvement in the IHT delivery including building better relationships with parents, understanding parents more, increased skills in managing children’s behavior in the classroom, and increased job satisfaction (Table 7). Enablers at the level of the caregivers included caregivers’ participating in the sessions, seeing benefits to participants parenting skills and their involvement in school, and caregivers supporting each other with their parenting. In terms of enablers related to materials, most teachers liked using the scripts for each session and reported that it made the session easier to deliver. They reported that the bulletins and WhatsApp messages generated interest among caregivers who were not enrolled in the group sessions and that the videos were easy to understand and interesting. Teachers highly valued the training and the ongoing support. Teachers’ Perspectives of the Barriers to Implementation. Teachers reported some difficulties in facilitating the sessions including difficulty in doing the roleplays due to shyness and feeling uncomfortable or because it was difficult to play the role required (Table 8). A minority of teachers would have preferred for the sessions to be less scripted. Other difficulties included poor attendance of caregivers and some caregivers being unwilling to participate in the sessions and/or finding it difficult to understand and practice the strategies. Several teachers felt that there were too many practice activities within each session which led to disengagement of caregivers. Teachers’ Suggestions for Improvement. Teachers gave suggestions for caregiver recruitment that included the following: i) informing the caregivers about the program in face-to-face meetings, 29 either in groups during parent orientation, or individually to allow teachers to address individual caregivers’ needs; ii) using WhatsApp messages and videos; iii) advertising the program using posters around the school, and iv) asking caregivers who have participated in the program to encourage other caregivers to join the groups by sharing their experiences and how they have benefitted (Table 9). To promote attendance, they suggested ongoing encouragement during faceto-face meetings, phone calls, and WhatsApp messages. Some teachers recommended using incentives to motivate caregivers. The majority of teachers wanted more support from other teachers in the school in running the program. This support could involve all teachers being trained to conduct the sessions and/or support with other aspects of the program such as with preparation and with recruiting, mobilizing, and engaging caregivers. Some teachers reported that fortnightly or monthly sessions would be more feasible for teachers and/or caregivers and that the program should be shortened to fit better into the school term. They also gave suggestions for engaging caregivers with the videos, messages and bulletins by, for example, making them more interactive and more salient. 30 Table 7. Preschool Teachers’ Perceptions of the Enablers to Implementation of the Irie Homes Toolbox Program Theme Quote Enablers at the level of the teachers Teachers build better relationships with parents • “It brings us closer together, we interact more, they will ask certain questions.” T16 • “The parents are communicating more.” T3 • “We are closer. We have chit chats, talk about their child’s development when they drop off or pick up their child.” T4 Teachers understand parents more • “I got to understand them more. They were willing to open up and confide in me, and that was a benefit to me.” T5 • “There was one mommy who I thought was avoiding me, but I learnt that she is shy. So, I know how to interact with her now.” T3 • “I get to see them in a different light and how they interact with their child.” T15 Participating in the IHT program helped teachers to manage child behavior in the classroom • “Going through this program, I’ve learned many different ways of how to deal with children rather than hitting and shouting at them.” T3 • “It helped me to know how to deal with the children better and how to manage my anger.” T2 • “It helps me a lot, I am calmer with the children ... during the sessions you learn strategies that you can use.”T15 Teachers report increased job satisfaction • “I think she (the parent) looks to me more. When she comes, she asks for me, it makes me feel good that I was able to impact someone’s life, other than a child.” T1 • “It makes me feel accomplished because they can trust me to come and talk. …They now know that they can talk to me.” T3 Enablers at the level of the participants Caregivers participated in the sessions • “It was easy because the parents were receptive. I loved that the parents encouraged each other. We also cheered on the parents.” T11 • “Supporting them as they practiced was comfortable because they listened and responded accordingly.” T10 • “It was easy, because parents were willing to participate.” T12 Caregivers benefit from the IHT: (e.g., understand child behavior, skills in positive discipline strategies) • “It helped … parents to be able to manage the behavior of their children.” T2 • “They (the parents) expressed that the sessions helped them in their everyday life, dealing with their children, managing behavior.” T4 • “They get to understand more about children’s behavior and what causes them to behave a certain way, and how to deal with it.” T1 Caregivers were more involved in school • “They are a bit more attached to the institution. You are seeing those parents more often.” T17 Caregivers supported each other • “…they spoke about personal triumphs and issues in their personal lives. In the WhatsApp group, parents were praising each other; they sent videos in the groups, and the parents would comment on those because it was about them, their lives, and their accomplishments.” T4 31 Table 7 (continued) Theme Quote Enablers relating to materials Having a script to follow helped • “The script tells you what to do, gives the possible answers. It makes it less stressful and easier.” T1 • “So, when you have a structured thing, you are able to do it the right way and don’t spoil it.” T11 • “The experience of using a script is exciting. I prefer to use a script because it is a guideline of the session, so nothing is left out.” T6 Bulletins sparked caregivers’ interest in program • " When the bulletins are up, ..we had parents who were not in the workshop come and say, ‘Miss, me try some of these things you know.’" T9 • "Parents asked what is was about. The parents asked me to let them know when it was going to happen again." T2 • “I saw them take pictures of it, they stopped, and they read it, and they asked us about it.” T12 WhatsApp messages sparked interest from caregivers who are not participating in sessions • “Other parents wanted to be a part of the group, they asked ‘When is the other one going to start?’” T4 • “A parent asked about the message that encourages parents to praise children. So even though she wasn’t aware of the group, the message made her wonder, so she asked me about it.” T8 Videos were interesting/relatable, useful and easy to understand • “I liked that the videos showed real parents interacting with their real children. I also liked that fathers were in the videos.” T5 • “The videos clarify everything a little more. They were interesting and helpful.” T9 • “I got feedback from parents saying how it has helped them, and they are better able to connect with their children.” T1 Enablers related to training Training and practice prepared teachers to deliver the sessions • “The practice the day before helped because you knew what to do, where to stand, and how to approach the parent.” T4 • “I was comfortable enough because I did the training already. I had the understanding of what to do. I was aware of what to say and the reaction I would probably get.” T3 • “The fact that I went through it before, it wasn’t new. So, when it was time to do it in the workshop, it was clearer and good.” T5 Support from the research team was valued by teachers • “The fact that a person from the research team was there as a guide, in case the key points were not being highlighted, or it was not being done as it should be done.” T5 • “The support was very helpful. If we made a mistake or if we weren’t clear about something, a member of the research team would help us. I felt more comfortable having her there.” T12 32 Table 8. Preschool Teachers’ Perceptions of the Barriers to Implementation of the Irie Homes Toolbox Program Theme Quotes Barriers at the level of the facilitator Teachers report discomfort in doing roleplays • “For the first two sessions it was difficult but as time went on it became easy, and more comfortable. You feel silly sometimes because most of the actions are of a child, and I wondered how they would see me, but they saw me as what it was intended to do: a child.” T4 • “Some things I had to check myself on. So, if the child is doing something that you don’t want them to do, you have to really restrain yourself from saying ‘Stop do that’ or ‘Don’t do that’. Because it has to be positive, so that part was difficult.” T9 • “I remember the first training where I had to play the role of a ‘bad mommy’ I felt a way because it is something that I don’t do. It was hard for me to play the bad mommy.” T11 Teachers prefer more autonomy with delivery • “…if teachers were allowed to adlib, maybe it would be a bit more interesting. Not changing the script but say it in my own way.” T9 • " I would rather use my own words. I feel more comfortable using my own words.” T12 Barriers at the level of the participants Poor attendance of caregivers • “Sometimes it was hard because they didn’t show up.” T3 Caregivers’ lack of participation in sessions • “It made it difficult when the parents didn’t want to participate.” T3 • “Sometimes some persons didn’t want to talk because they are very shy, so that might be a little difficulty.” T6 • “Some parents are timid, so they don’t want to speak out, so you have to … encourage them along the way.” T14 Caregivers find aspects of the IHT program difficult • “Sometimes you are not getting what you want from the parents, so you have to work more to get the information.” T5 • “Some of them cannot read, that’s why some of them don’t fill out the home assignment.” T2 Barriers relating to materials Few teachers watched all the videos • “I watched a few. I think I watched the one about withdrawing attention.” T14 • “I didn’t pay any in-depth attention because I just figured the videos were for parents who were not at the sessions or for those who don’t understand.” T15 Barriers related to intervention Teachers reported that there were too many practice activities • “I think that we can minimize the amount of practice in each session ... We don’t need so many activities on the paper.” T12 • “Only thing is most of the demonstrations if we did two or three it is enough rather than doing all the examples.” T13 33 Table 9. Preschool Teachers’ Suggestions Relating to Implementation of the Irie Homes Toolbox Program Theme Quotes Suggestions relating to promoting participant recruitment and attendance Teachers can recruit participants face-to-face • “Talk about it during orientation in September. We can also tell them about it in parents’ month (November).” T5 • “Face to face because you would have a more meaningful conversation rather than on WhatsApp. Face to face they know that you really care.” T4 • “For the parents who really need it, I would recruit them face-to-face because then I can get their feedback and know which approach to take.” T3 • “I would approach them about the sessions individually, we want them to feel important.” T15 Teachers can recruit participants virtually • “I could also tell them about it via the WhatsApp group (the school group). Whoever indicates interest is told to come in and sign up.” T9 • “Flyer, video advertisement, something lively. The video ads could be sent to the school, and it could be posted in the WhatsApp group or the school’s social media page, or website.” T5 • “Highlight it more. Have a billboard in the schoolyard. Advertisement video sent in the group.” T7 Teachers can engage past participants in recruiting new caregivers for the IHT • “Let the other parents who have been through the program tell them how they benefitted.” T10 • “Have past parents speak about their experience and how it helps them and their children.” T16 Teachers’ suggestions for promoting caregiver attendance • “Encourage them when we see them in the mornings, or WhatsApp message, or phone call. We could say, ‘remember we are having the Irie meeting this afternoon. This will help you to be an Irie parent.” T8 • “You have to give incentives.” T13 • “Continue sending out the bulletins, messages on the phone.” T14 Suggestions relating to the school Train all teachers in the school to conduct the parenting sessions • “We want them to come on board to help us to do the sessions. We have 5 teachers so we could all take a turn in doing the sessions.” T14 • “Firstly, I want to introduce the program to the teachers at the school during the holiday so they can do sessions. Then all hands would be on deck for engaging new parents and to get everybody on board with Irie parenting.” T12 • “Do a roster where each teacher does a session, so it is not burdensome on one teacher.” T13 Ask other teachers in the school to provide support for the teachers conducting the parenting sessions • “Helping to set up the room, getting refreshments prepared, helping to do a week, not leaving it on one person.” T5 • “Be engaged and communicate the information sent out to parents so that there is a wider audience range, because we can’t have more than 10 parents in a session.” T4 34 4.6 Ongoing Researcher Logs The two members of the Irie Toolbox team who were training and supporting the teachers to implement the intervention kept an ongoing log of the enablers and barriers to implementation and suggestions for improvement. These logs included notes kept after each session they visited, combined with notes related to ongoing interactions with teachers. Research Team’s Perspectives of the Enablers to Implementation. All teachers were reported to be highly compliant with intervention protocols, enjoyed delivering the sessions, and utilized the strategies from the IHT in their personal and professional lives (Table 10). The majority of teachers had good relationships with the caregivers in their parenting groups, engaged caregivers well, and shared personal experiences of how they used the strategies themselves and the difficulties they faced. Some teachers showed good understanding of the content, demonstrated the skills well, were competent at using the script, and delivered the sessions enthusiastically and with a lively pace. Research Team’s Perspectives of the Barriers to Implementation. Key barriers to the quality of session delivery, included lack of preparation and insufficient support for caregivers during practice activities (Table 11). Some teachers found it difficult to maintain a lively demeanor, and their sessions were not engaging for caregivers. An important barrier was that many teachers found it difficult to address caregivers’ questions and concerns relating to their child’s misbehavior and lacked skills in dealing with resistance from caregivers. Furthermore, some teachers had positive attitudes to using harsh punishment. We faced ongoing challenges with caregiver attendance. In some instances, these challenges were due to caregivers’ work schedules and other responsibilities. However, it was also evident that in many schools, teachers had not checked caregivers’ willingness and availability prior to recruitment. Research Teams’ Suggestions for Improvement. The research team identified a need for more initial training to prepare teachers to facilitate the sessions. Optimally, the training would include the teachers participating in the training as though they are caregivers, prior to being trained to conduct the sessions (Table 12). This would ensure they are fully conversant with the strategies and have a thorough knowledge and understanding of the core content of the program. 35 Experiencing the IHT program as a participant would also help teachers to understand the rationale for the behavior change techniques used. For the school-wide element to be effective, we need to orient all teachers in the school to the IHT and provide guidelines for how schools can allocate tasks so that all teachers are invested and involved in the program. Another recommendation was to develop a school readiness index to identify schools that are ready to implement the program versus schools that may need to be strengthened prior to implementation. 36 Table 10. Research Team Perspectives of Preschool Teachers’ Strengths in Implementing the Irie Homes Toolbox Program Theme Extract from Research Logs Notes from Observations of Sessions Clear and Engaging Script Delivery: Teachers read the script enthusiastically, naturally, with clarity and at a good pace. • “Clear and engaging delivery, with appropriate emphasis on key points.” • “Reads the script as if having a conversation, very natural and kept a good pace.” Good Relationship with Caregivers: Teachers show warmth and foster open communication and caregivers share openly • “She has a good relationship with the parents – they share personal experiences, and she makes them laugh.” • “It is clear that she has a good relationship with the parents and that they like talking to her.” Understand Content: Teachers have a solid grasp of session content which helps them effectively guide discussions and clarify concepts. • “Understood the content regarding reason for child misbehavior.” • “Understands the importance of praise and can effectively demonstrate it." • “Understood the steps to teach the skill and could recognize it when the parents did it.” Relatability and Personal Experience: Teachers share personal stories or acknowledge their own learning journey that tend to build trust and rapport with parents. • “Shared personal experiences with her 1and 3-year-old children.” • “Shared how the program has helped her in her classroom, particularly managing emotions.” • “Shared personal instances where she almost lost her cool and exploded.” Engagement and Liveliness: Teachers interact with parents warmly and maintain an engaging session atmosphere through eye contact, smiling, humor, or liveliness. • “Was enthusiastic and encouraged parent participation." • “Demonstrated the strategies well and kept it fun.” • “Encouraged quieter parents to share their perspectives.” Clear Demonstration: Teachers demonstrate the use of the strategies clearly • “Good job demonstrating how to praise.” • “Good job helping them to use many concept words in Irie Time.” Notes from Ongoing Logs Teacher compliance was high • “The teachers were willing to deliver the sessions each week. They made time to meet with us to go through the script and then actually held the sessions. They sent out the messages to the parents and put up the bulletins. They even called parents to ask where they were if the session is about to start, and they were not yet present.” Teachers enjoyed conducting sessions • “Most teachers seemed to enjoy doing the sessions and getting to know the parents. They laughed and shared their personal stories; After sessions were finished, they sometimes hung around chatting with the parents.” Teachers used the program in their professional and personal life • “Some teachers were learning alongside the parents. They learnt strategies that they would then use in their classrooms as well as with their own children at home. They shared the videos on their personal social media pages, sent videos to family members and encouraged them to try the strategies. One teacher inquired about running sessions at her church.” 37 Table 11. Research Team Perspectives of Facilitators’ Needs in Implementing the Irie Homes Toolbox Program Theme Extract from Researcher Logs Notes from Observations of Sessions Lack of Preparedness & Organization: Teachers were often unprepared with materials and didn’t set up the room in advance. • “Did not take out materials beforehand.” • “Chairs set up but still needed a bit of rearranging and materials not set out.” Managing and Organizing Session Materials: Teachers struggle with organizing and handling materials, including charts and scripts simultaneously, and ensuring materials are visible for parents. • “Needs to work out how to hold the script and charts at the same time." • “Turned body away from the group." • “Needs to handle charts so parents can see the picture and/or text.” Lack of Script Familiarity and Engagement: Teachers had issues with sticking to the script, often attempting to deliver from memory or in their own words. This led to them changing the meaning at times or led to them missing certain sections/points. • “Needs to read with more enthusiasm and keep her head up from the script.” • “Tried to do it from memory instead of following the script." • “Tried to put the script in her own words, which changed the meaning." Poor Pacing, Low Energy and Engagement: Teachers spoke in a slow or monotonous tone, were low-energy and lacked enthusiasm which affected the pacing of the session and led to caregivers disengaging. • “Read in a flat tone, which made the session feel longer." • “Is too laid back. Reads very flat, monotone and all in one. Not vibrant.” • “…struggles with pacing and energy.” Does not Prompt, Scaffold or Support Caregivers: Teachers don’t always recognize when parents need help and don’t support caregivers when they practice the strategies. • “Often missed opportunities to prompt parents and to give specific feedback when parents needed encouragement.” • “Overall, she has difficulty or is uncomfortable with supporting parents to make sure they get it right.” Dealing with Sensitive Topics and Misconceptions: Teachers sometimes struggle to guide caregivers in understanding and addressing common misconceptions about child behavior and discipline. They also need help with dealing with resistance from participants. • “Parents and teachers believe that hitting works and sometimes the children need a good slap.” • “The teacher did not believe in changing negative thoughts to positive, and she said she would still be angry.” Notes from Ongoing Logs Poor and Inconsistent Parent Attendance: Attendance at the sessions was low in some schools. In others the attendance was inconsistent with different parents attending each week. • “Only three parents attend consistently, the majority of the parents in the group work.” • “Attendance was low this week at the school due a water issue in the community” • “Multiple parents were busy this week taking their children to register at the primary school.” Unavailable Parents Recruited and Enrolled: In multiple schools the parents enrolled in the study were unavailable to attend the sessions • ” In some schools, the teachers do not know the parents well, so they recruited parents who did not have time to attend.” • “Preschool teachers recruited caregivers who they thought needed the program rather than caregivers who were willing and able to attend." 44 from the outset for integration into the existing early childhood education system, ongoing adaptation and flexibility is required as government systems are not static (Herlitz et al., 2020). 5.1 Strengths and Limitations The strengths of the study are the use of a cluster-randomized design, outcome measures with good psychometric properties, assessors masked to the study design, hypothesis, and group allocation, low levels of attrition with only 2.9% lost to follow-up, and the mixed-method approach including qualitative data from the preschool teachers who are implementing the program. The limitations of the study are that all outcome measures are by caregiver self-report and parents are aware of their group allocation, so this may result in biased responses. Using a social desirability index (SDI), we found no evidence that caregivers assigned to the intervention were more likely to respond in a socially desirable way than caregivers assigned to the wait-list control and controlling for SDI in the analyses did not alter the magnitude or significance of the effects on caregiver outcomes. The use of parent questionnaires to measure use of harsh punishment, involvement and child behavior has the advantage of assessing parent and child behavior across all contexts. Another limitation is that this is a feasibility study, and the sample size was small. We had insufficient power for heterogeneity analyses and hence we did not examine heterogeneity by child age and sex, or by caregiver age, education, and socio-economic status. We were unable to recruit all caregivers prior to randomization, due to the need to conduct teacher training workshops in the school holidays. As preschool staff led the recruitment of caregivers into the study, this could lead to teachers in intervention schools inviting higher-risk parents to attend the parenting sessions (Francis and Baker-Henningham, 2023). However, the groups were balanced on the primary outcome of caregivers’ use of violence against their child and on other secondary outcomes of caregiver attitudes to violence, preferences for harsh punishment, and child behavior difficulties. We found a significant difference between the groups in caregiver involvement with their child and caregiver age and we controlled for these differences in all analyses. Caregivers were recruited into this study based on their interest and willingness to participate, while preschools were selected based on their willingness and feasibility to implement the program. Hence, the participants are not representative of the wider population of caregivers and preschools. The decision to recruit by interest and willingness was based on the diffusion of innovation theory which describes how innovations are spread through a population, starting with the early adopters, then spreading to the 45 early majority, then the late majority, and finally the laggards (Rogers, 2003). Implementing the program with interested participants maximizes the likelihood that the program will be successful, leading to the program gaining momentum, thus attracting other preschools and other caregivers (the early majority). 5.2 Summary This study demonstrated that the IHT can maintain effectiveness when delivered by preschool teachers as part of their routine duties. The next steps are to further adapt the IHT program based on lessons learnt. There is a need to test a shorter version of the IHT that may be more acceptable to teachers and caregivers. We also need to test the effect of school-wide messaging on caregivers’ attitude to and use of VAC. To maintain the quality of the IHT implementation as the program scales, it is important to devise strategies for adequate training and supervision of preschool teachers. Finally, it is important to increase the availability of evidence-based programs to reduce caregivers’ use of VAC. This involves designing flexible programs that are delivered through different modalities, and that cater to caregivers with different needs. 46 References Altafim, E. R. P., and M. B. M. Linhares, 2019. “Preventive intervention for strengthening effective parenting practices: A randomized controlled trial.” Journal of Applied Developmental Psychology. 62: 160-172. https://doi.org/10.1016/j.appdev.2019.03.003 Backhaus, S., P. Leijten, J. Jochim, et al. 2023. “Effects over time of parenting interventions to reduce physical and emotional violence against children: a systematic review and metaanalysis.” eClinicalMedicine. 60: 102003. https://doi.org/10.1016/j.eclinm.2023.102003 Baker-Henningham, H. 2018. “The Irie Classroom Toolbox: developing a violence prevention, preschool teacher training program using evidence, theory and practice.” Annals of the New York Academy of Sciences. 1419(1): 179–200. https://doi.org/10.1111/nyas.13713 Baker-Henningham, H., M. Bowers, and T. Francis. 2023. “Scaling early childhood, violence prevention programs across home and school settings in Jamaica.” Pediatrics. 151(S2): 115. https://doi.org/10.1542/peds.2023-060221M Baker-Henningham, H., M. Bowers, T. Francis, et al. 2021. “The Irie Classroom Toolbox, a universal violence-prevention teacher-training programmme, in Jamaican preschools: a singleblind, cluster randomised trial.” Lancet Global Health. 9: e456-68. https://doi.org/10.1016/S2214-109X(21)00002-4 Baker-Henningham, H., Y. Scott, M. Bowers, et al. 2019. “Evaluation of a violence-prevention program with Jamaican primary school teachers: A cluster randomized trial.” International Journal of Environmental Research and Public Health. 16: 2797. https://doi.org/10.3390/ijerph16152797 Bell, K. M., and A. E. Naugle. 2007. “Effects of social desirability on students’ self-reporting of partner abuse perpetration and victimization.” Violence and Victims. 22(2): 243-256. https://doi.org/10.1891/088667007780477348 Bernal, R., M.L. Gomez, S. Perez-Cardona, et al. 2023. “Implementation quality of an early childhood parenting program in Colombia and child development.” Pediatrics. 151(S2): e2023060221J. https://www.doi.org/10.1542/peds.2023-060221J Braun, V., and Clarke, V. 2006. Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2): 77-101. https://doi.org/10.1191/1478088706qp063oa Bowers, M., F. Francis, and H. Baker-Henningham. 2022. “The Irie Classroom Toolbox: Mixed method assessment to inform future implementation and scale-up of an early childhood 47 teachertraining, violence-prevention programme.” Frontiers in Public Health. 10: 1040952. https://doi.org/10.2289/fpubh.2022.1040952 Britto, P., M. Singh, T. Dua, et al. 2018. “What implementation evidence matters: scaling-up nurturing interventions that promote early childhood development.” Annals of the New York Academy of Sciences. 1419(1): 5-16. https://doi.org/10.1111/nyas.13720 Carey, R. N., L. E. Connell, M. Johnston, et al. 2019. “Behavior change techniques and their mechanisms of action: a synthesis of links described in published intervention literature.” Annals of Behavioral Medicine. 53: 693-707. https://doi.org/10.1093/abm/kay078 Caridad Araujo, M., M. Rubio-Codina, N. Schady. 2021. “70 to 700 to 70,000. Lessons from the Jamaica experiment.” In J. List, D. Suskind, and L. Supplee, editors. The scale-up effect in early childhood and public policy. Why interventions lose impact at scale and what we can do about it. New York: Routledge, pp. 160-183. Caron, E., B., K. Bernard, A. Metz. 2021. “Fidelity and properties of the situation. Challenges and recommendations.” In J. List, D. Suskind, and L. Supplee, editors. The scale-up effect in early childhood and public policy. Why interventions lose impact at scale and what we can do about it. New York: Routledge, pp. 160-183. Chan, K., Q. Chen, and M. Chen. 2019. “Prevalence and correlates of the co-occurrence of family violence: A meta-analysis on family polyvictimization.” Trauma, Violence, & Abuse. 22(2): 289-305. https://doi.org/10.1177/1524838019841601 Chan, A., J. M. Tetzlaff, D. G. Altman, et al. 2013. “SPIRIT 2013 Statement: Defining protocol items for clinic trials.” Annals of Internal Medicine. 158(3): 200-207. https://doi.org/10.7326/0003-4819-158-3-201302050-00583 Chen, M., and K. Chan. 2015. “Effects of parenting programs on child maltreatment prevention.” Trauma Violence Abuse. 17(1): 88-104. https://doi.org/10.1177/1524838014566718 Crowne, D. P., and D. Marlowe. (1960). “A new scale of social desirability independent of psychopathology.” Journal of Consulting Psychology. 24(4): 349-354. https://doi.org/10.1037/h0047358 Cuartas, J., H. Baker-Henningham, A. Cepeda, et al. (2022) “The Apapacho violence prevention parenting program: Conceptual foundations and pathways to scale.” International Journal of Environmental and Public Health. 19: 8582. https://doi.org/0.3390/ijerph19148582 48 Cuartas, J., D. C. McCoy, C. Rey-Guerra, et al. 2019. “Early childhood exposure to non-violent discipline and physical and psychological aggression in lowand middle-income countries: National, regional, and global prevalence estimates.” Child Abuse and Neglect. 92: 93-105. https://doi.org/10.1016/j.chiabu.2019.03.021 Dinarte-Díaz, D., Ravindran, S., Shah, M., Powers, S., and Baker-Henningham, H. 2023. Violent discipline and child Behavior. Shortand medium-term effects of parenting support to caregivers in Jamaica. NBER Working Paper No. 31338. https://www.nber.org/system/files/working_papers/w31338/w31338.pdf Elff, M., Heisig, J. P., Schaeffer, M., and Shikano, S. 2021. “Multilevel analysis with few clusters: Improving likelihood-based methods to provide unbiased estimates and accurate inference.” British Journal of Political Science. 51(1): 412-426. https://doi.org/10.1017/S0007123419000097 Elff, M., J. P. Heisig, M. Schaeffer, et al. 2021. “Multilevel analysis with few clusters: Improving likelihood-based methods to provide unbiased estimates and accurate inference.” British Journal of Political Science. 51(1): 412-426. https://doi.org/10.1017/S0007123419000097 Eyberg, S., and A. Ross. 1978. “Assessment of child behavior problems: The validation of a new inventory.” Journal of Clinical Child Psychology. 7(2): 113–116. https://doi.org/10.1080/15374417809532835 Fernández-Gonzáles, L., K. D. O’Leary, and M. J. Muñoz-Rivas. 2013. “We are not joking: Need for controls in reports of dating violence.” Journal of Interpersonal Violence. 28(3): 602620. https://doi.org/10.1177/08862605124555 Francis, T., D. Dinarte-Díaz, S. Powers, et al. (2024). “The Virtual Irie Homes Toolbox: Adaptation and remote delivery of an early childhood, violence prevention, parenting program in Jamaica.” Journal of Research in Childhood Education. 38(sup 1): S92-S111. https://doi.org/10.1080/02568543.2023.2281560 Francis, T., M. Bowers, and H. Baker-Henningham. (2025). “A feasibility trial of an early childhood, violence prevention, parenting program integrated into early childhood educational provision in Jamaica: A study protocol.” Developmental Psychology. https://doi.org/10.1037/dev0001899 Francis, T., and H. Baker-Henningham. 2020. “Design and implementation of the Irie Homes Toolbox.” Frontiers in Public Health. 8: 282961. 49 https://doi.org/10.3389/fpubh.2020.282961 Francis, T., and H. Baker-Henningham. 2021. “The Irie Homes Toolbox: A cluster-randomized controlled trial of an early childhood parenting program to prevent violence against children in Jamaica.” Children and Youth Services Review. 126: 106060. https://doi.org/10.1016/j.childyouth.2021.106060 Francis, T., D. Packer, and H. Baker-Henningham. 2023. “A qualitative evaluation of the mechanism of action in an early childhood parenting program to prevent violence against children in Jamaica.” Child: Care, Health, and Development. 49(3): 579-590. https://doi.org/10.1111/cch.13074 Gericke, C. A., C. Kurowski, M. K. Ranson, A. Mills. 2005. “Intervention complexity – a conceptual framework to inform priority setting in health.” Bulletin of the World Health Organization. 83(4): 285-293. PMID: 15868020. Gershoff, ET. 2010. “More harm than good: A summary of scientific research on the intended and unintended effects of corporal punishment on children.” Law and Contemporary Problems. 73: 31-56. PMID: 34446972 Government of Jamaica, (GOJ). 2018. National Plan of Action for an Integrated Response to Children and Violence (NPACV) 2018-2023. Kingston, Jamaica: Government of Jamaica. Heilmann, A., A. Mehay, R. G. Watt, et al. 2021. “Physical punishment and child outcomes: a narrative review of prospective studies.” Lancet. 398(10297): 355-364. https://doi.org/10.1016/s0140-6736(21)00582-1 Herlitz, L., H. MacIntyre, T. Osborn, Bonell C., et al. 2020. “The sustainability of public health interventions in schools: a systematic review.” Implement Science. 15:4. https://www.10.1186/s13012-019-0961-8. Hillis, S., J. Mercy, A. Amobi, et al. (2016). “Global prevalence of past-year violence against children: A systematic review and minimum estimates.” Pediatrics. 137(3): e20154079. https://doi.org/10.1542/peds.2015-4079 Hillis, S. D., J. A. Mercy, and J. R. Saul. 2017. “The enduring impact of violence against children.” Psychological Health and Medicine. 22(4): 393-405. https://doi.org/10.1080/13548506.2016.1153679 50 Hoffman, T.C., Glasziou, P.P, Boutron, I. et al., 2014. “Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide.” BMJ. 348: g1687. https://doi.org/10.1136/bmj.g1687 Heilmann, A., A. Mehay, R. G. Watt, et al. 2021. “Physical punishment and child outcomes: a narrative review of prospective studies.” Lancet. 398(10297): 355-364. https://doi.org/10.1016/s0140-6736(21)00582-1 Hillis, S., J. Mercy, A. Amobi, et al. (2016). “Global prevalence of past-year violence against children: A systematic review and minimum estimates.” Pediatrics. 137(3): e20154079. https://doi.org/10.1542/peds.2015-4079 Hillis, S. D., J. A. Mercy, and J. R. Saul. 2017. “The enduring impact of violence against children.” Psychological Health and Medicine. 22(4): 393-405. https://doi.org/10.1080/13548506.2016.1153679 Kasteng, F., J. Murray, S. Cousens, et al. 2018. “Cost-effectiveness and economies of scale of a mass radio campaign to promote household life-saving practices in Burkina Faso.” BMJ Global Health. 3(4). https://doi.org/10.1136/bmjgh-2018-000809 Leijten, P., M. Raaijmakers, B. Orobio de Castro, et al. 2015. “Effectiveness of the Incredible Years Parenting Program for Families with Socioeconomically Disadvantaged and Ethnic Minority Backgrounds.” Journal Of Clinical Child & Adolescent Psychology. 46(1): 5973. https://doi.org/10.1080/15374416.2015.1038823 Menting, A., B. Orobio de Castro, and W. Matthys. 2013. “Effectiveness of the Incredible Years parent training to modify disruptive and prosocial child behavior: A meta-analytic review.” Clinical Psychology Review. 33(8): 901-913. https://doi.org/10.1016/j.cpr.2013.07.006 Michie, S., van Stralen, M.M., and West, R. 2011. “The behaviour change wheel: a new method for characterising and designing behaviour change interventions.” Implementation Science. 6(42). https://doi.org/10.1186/1748-5908-6-42 Pacchiano, D., M. Connors, R. Klein, et al. 2021. “Embedding workforce development into scaled innovations to prevent declines in administration quality.” In J. List, D. Suskind, and L. Supplee, editors. The scale-up effect in early childhood and public policy. Why interventions lose impact at scale and what we can do about it. New York: Routledge, pp350-369. 51 Pereznieto, P., A. Montes, S. Routier, et al. 2014. The Costs and Economic Impact of Violence Against Children. London, UK: Overseas Development Institute. Rasbash, J., C. Charlton, W. J. Browne, et al. 2009. MLwiN Version 2.10. Bristol, UK: Centre for Multilevel Modelling. Reynolds, W. M. (1982). Development of reliable and valid short forms of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology. 38(1): 11925. https://doi.org/10.1002/1097-4679(198201)38:1 Rogers, M. E. (2003). Diffusion of innovations. (5th ed.). New York, USA: Free Pass. Sanders, M., Divan, G., M. Singhal, et al. 2022. “Scaling Up Parenting Interventions is Critical for Attaining the Sustainable Development Goals.” Child Psychiatry and Human Development. 53: 941-952 https://doi.org/10.1007/s10578-021-01171-0 Santini, P., and L. Williams. 2017. “A randomized controlled trial of an intervention program to Brazilian mothers who use corporal punishment.” Child Abuse and Neglect. 71: 80-91. https://doi.org/10.1016/j.chiabu.2017.04.019 Sarrassat, S., N. Meda, H. Badolo, et al. 2018. “Effect of a mass radio campaign on family behaviours and child survival in Burkina Faso: A repeated cross-sectional, clusterrandomised trial.” The Lancet Global Health. 6(3). https://doi.org/10.1016/s2214109x(18)30004-4 Satinsky, E. N., B. Kakuhikire, C. Baguma, et al. 2024. “Caregiver preferences for physically harsh discipline of children in rural Uganda.” Journal of Family Violence. 39: 861-874. https://doi.org/10.1007/s10896-023-00536-4 Schulz, K., and D. A. Grimes. 2002. “Sample size slippages in randomized trials: exclusions and the lost and wayward.” The Lancet. 359(781-785). https://doi.org/10.1016/S01406736(02)07882-0 Skar, A., L. Sherr, A. Macedo, et al. 2017. “Evaluation of Parenting Interventions to Prevent Violence Against Children in Colombia: A Randomized Controlled Trial.” Journal of Interpersonal Violence. 36(1-2). https://doi.org/10.1177/0886260517736881 Straus, M., S. Hamby, D. Finkelhor, et al. 1998. “Identification of child maltreatment with the parent-child Conflict Tactics Scales: Development and psychometric data for a national sample of American parents.” Child Abuse and Neglect. 22(4): 249-270. https://doi.org/10.1016/s01452134(97)00174-9 52 UNICEF. 2022. A statistical profile of violence against children in Latin America and the Caribbean. Technical Report. United Nations Children’s Fund (UNICEF). Vlahovicova, K., G. Melendez-Torres, P. Leijten, et al. 2017. “Parenting Programs for the Prevention of Child Physical Abuse Recurrence: A Systematic Review and MetaAnalysis.” Clinical Child and Family Psychological Review. 20(3): 351-365. https://doi.org/10.1007/s10567-017-0232-7 Wang, W., and H. Zhang. 2024. “A scoping review of parenting programs in preventing violence against children in lowand middle-income countries. Trauma, Violence & Abuse. 25(3): 2173-2188. https://doi.org/10.1177/15248380231207887 Woolgar, M., Humayun, S., Scott, S., and Dadds, M. R. 2023. “I know what to do; I can do it; it will work: the Brief Parental Self Efficacy Scale (BPSES) for parenting interventions.” Child Psychiatry and Human Development. https://doi.org/10.1007/s10578-023-01583-0 World Health Organization. 2016. INSPIRE: Seven strategies for ending violence against children. Geneva, Switzerland: World Health Organisation. 53 Table A. Child, Caregiver, Household Characteristics and Baseline Outcomes by Lost and Found Lost (n= 7) Found (n= 233) P-value Child age (in years) 5.0 (1.2) 4.1 (1.1) 0.03 Child sex (n (%) boys)) 5 (71.4%) 113 (48.5%) 0.23 Caregiver age (in years) 30.4 (6.7) 32.9 (9.2) 0.48 Caregiver sex (n (%) female) 7 (100%) 220 (94.4%) 0.52 High school completed n (%) 3 (42.9%) 65 (27.9%) 0.39 Caregiver currently employed n (%) 3 (42.9%) 61 (26.2%) 0.33 Mother lives with child n (%) 7 (100%) 214 (91.8%) 0.96 Father lives with child n (%) 4 (57.1%) 87 (37.3%) 0.29 Number of children < 18 years in household 1.9 (0.9) 2.2 (1.0) 0.42 Household possessions 8.0 (2.4) 9.4 (2.3) 0.11 Household sanitation (median (IQR)) 12.0 (8.0-12.0) 12.0 (8.0-12.0) 0.68 Crowding median (range) 1.3 (0.8-2.5) 1.4 (1.0-2.0) 0.91 Caregivers’ use of violence against their child (median (IQR)) 11.0 (4.0-17.0) 12.0 (7.0-17.0) 0.63 Caregiver involvement 31.3 (14.7) 35.3 (11.0) 0.12 Caregiver attitude to violence against children (median (IQR)) 6.0 (2.0-10.0) 5.0 (3.0-7.0) 0.67 Caregiver preferences for harsh punishment (median (IQR)) 4.0 (4.0-7.0) 5.0 (3.0-6.0) 0.89 Child conduct problems 106.3 (11.3) 114.2 (22.9) 0.37 Notes: Scores are mean (SD) unless otherwise stated.