Background and rationale

Child maltreatment–defined as physical, sexual, and emotional abuse or neglect–is a global public health crisis with self-reported prevalence rates worldwide ranging from 22.6% for physical abuse to 36.3% for emotional abuse [1]. Most child maltreatment is perpetrated in the home by parents or parental guardians [2, 3]. The potentially lifelong and intergenerational impacts of child maltreatment are well-documented and include mental health disorders (e.g., depression, anxiety, childhood behavioural problems), alcohol and drug abuse, suicidal behaviour, risky sexual behaviour, obesity, and increased risk of future violence perpetration and victimization [4, 5]. While child maltreatment is a global problem prevalent in both high-income and low- and middle-income countries (LMICs), evidence suggests that children in LMICs–particularly those living at the intersections of poverty, armed conflict and forced migration–may be at increased risk of victimization [3, 6–8]. War and war-induced displacement have been shown to increase known risk factors for child maltreatment including household poverty, parental psychological distress, and family dysfunction [3, 9, 10]. Growing evidence from low-resource, forced migration contexts suggests that exposure to war and displacement-related stressors heighten parental distress, weaken coping and emotion regulation strategies, and interact with cultural norms to increase violence against children in the home [11, 12]. Disruptions to social services and social support networks due to war and displacement further degrade caregivers’ capacity to protect children and provide positive, nurturing care in the face of widespread violence and instability [9].

Interventions that support parents and caregivers to practice positive parenting skills and behaviours have been shown to reduce the risk of child maltreatment as well as improve a range of caregiver and child mental health and behavioural outcomes [13, 14]. While parenting interventions can be delivered through different modalities (e.g., individual/family-based, group-based, virtual, in-person), they typically consist of a structured series of sessions built around evidence-based components for increasing positive parent-child interaction and reducing harsh parenting. In view of robust evidence of their effectiveness, parenting interventions are included as one of the seven INSPIRE strategies advocated by the World Health Organization for ending violence against children [15]. However, large-scale implementation of parenting interventions in LMICs remains limited, particularly in humanitarian and displacement settings where the need is arguably the greatest [16]. Delivery of parenting support to migrant and displaced families in LMICs is constrained by structural and contextual factors including limited infrastructure and workforce, inadequate financial resources, and low levels of education and basic and digital literacy among the target population [17]. Reach may be further compromised by ambivalence on the part of host governments to fund services for refugees and migrants, resulting in fragmented and under-resourced delivery of parenting support to these populations.

Film-based interventions, a form of entertainment-education, has the potential to overcome many of these barriers and reach large audiences at relatively low cost with minimal infrastructure and human resource support [18]. While there is accumulating evidence that film and other narrative entertainment-education approaches can be effective at addressing certain health outcomes in LMICs including HIV prevention, sexual and reproductive health, and child survival, to our knowledge such approaches have not been applied to reducing child maltreatment among migrant and displaced populations in low-resource settings [19–21].

Film-based interventions

Entertainment-education refers to “the intentional placement of educational content in entertainment messages” [22 p117]. Narrative approaches to entertainment-education, including film, are hypothesized to impact knowledge, attitudinal, and behavioural outcomes through a range of mechanisms [23]. For example, social learning theory posits that film or show characters may serve as potential role models and improve audiences’ self-efficacy for adopting new behaviours [24], while constructs such as transportation (i.e., engagement in the narrative) [25] and identification (i.e., involvement with characters) [26] highlight the potential for narrative approaches to reduce counter-arguing and resistance to intervention messages. By activating both emotional and cognitive processes, narrative film-based interventions may be able to influence culturally entrenched values and practices that contribute to negative outcomes for children and families.

There is currently limited evidence of the potential for film-based interventions to reduce child maltreatment or other forms of violence in the home, particularly in LMICs where most children and adolescents reside. A systematic review of universal interventions with a media component aimed at preventing child physical abuse found only 17 studies in five countries, all high-income [27]. While quality of studies was generally low, the review found significant reductions in child abuse incidence as well as a range of related outcomes including dysfunctional parenting. Only a few studies have examined entertainment-education interventions to address family violence in LMICs [28, 29] and these have focused on violence against women. For instance, evaluation of MTV Shuga in Nigeria found that exposure to this educational TV series improved men’s attitudes towards gender-based violence eight months later [30]. This emerging evidence underscores the need for greater investment in developing and rigorously evaluating the use of narrative entertainment-education approaches to reduce child maltreatment, particularly in low-resource displacement settings where access to parenting support is extremely limited.

Objectives

The current study will evaluate the impact of a film-based intervention on reducing two forms of child maltreatment, physical and emotional abuse, and increasing positive parenting behaviour among migrant and displaced families from Myanmar living in Thailand. To our knowledge, this will be the first randomized controlled trial evaluation of a film-based intervention to reduce child maltreatment among migrant and displaced families in a LMIC. A mixed methods process evaluation will be embedded within the trial to assess intervention delivery, acceptability, perceived impacts, and potential mechanisms of change. The study is being implemented as part of the Global Parenting Initiative, a global consortium of academic and research institutions and implementing organizations aiming to provide access to free, evidence-based, playful parenting support to promote child learning and prevent violence at scale in the Global South [31].

Aims

The primary study aim is to evaluate the effectiveness of a film-based intervention for reducing caregiver-reported physical and emotional abuse and increasing positive parenting. Secondary aims are to assess the impacts of the intervention on caregiver knowledge of positive parenting, caregiver attitudes towards harsh punishment, caregiver psychological distress, and family functioning. Additional aims are to explore intervention impacts on caregiver coping, behaviours to support early learning, social support, and child internalizing and externalizing symptoms. The study will also examine potential treatment moderators (e.g., lifetime trauma exposure, length of displacement) and mediators (e.g., caregiver psychological distress), as well as acceptability and perceived intervention impacts and mechanisms of change.

Trial design

The study is a pragmatic, superiority cluster randomized controlled trial with approximately 40 communities (clusters) randomized to the intervention or treatment as usual (TAU) arms in a 1:1 ratio. Participating families in communities allocated to the intervention arm will be invited to attend a community screening of the film intervention and a post-screening discussion. Each family will also receive a poster depicting key messages from the film.

Primary outcomes for the trial are changes in (1) physical and emotional abuse; and (2) positive parenting behaviour. Secondary outcomes include caregiver knowledge of positive parenting, caregiver attitudes towards harsh punishment, caregiver psychological distress, and family functioning. Exploratory outcomes include caregiver coping, behaviours to support early learning and education, social support, and child internalizing and externalizing symptoms. Outcomes will be assessed at baseline prior to randomization (T0), approximately 4 weeks post-intervention (T1), and 4-month follow up (T2) (S1 File). Qualitative group interviews with caregivers and adolescents will be conducted at T1 to examine acceptability of the film-based intervention and perceived impacts on caregiver, child, and family outcomes.

Study setting and implementation partners

The study will be conducted in Tak province in Thailand on the border with Myanmar. There is a long-standing history of migration and displacement from Myanmar into Tak province due to decades of armed conflict and political and economic instability in Myanmar [32]. As a result of the coup d’état in Myanmar in February 2021, the area has seen a significant influx of individuals from Myanmar across the border into Tak province [33]. A 2022 report indicates that Thailand hosts 658,023 individuals from populations of concern including an estimated 91,401 refugees from Myanmar who live in the 9 refugee camps along the border [33]. However, this figure does not include the large number of migrants and displaced people from Myanmar who live outside the camps. The 2019 Thailand Migration Report estimates that there are 3.9 million migrants living in Thailand, with approximately 70% originating from Myanmar [34]. Thailand is not a signatory to the 1951 Refugee Convention and there is currently no national legislation or legal framework governing the definition, rights, and protection of refugees, migrants, and other displaced people in Thailand [34].

Many migrants and displaced people from Myanmar do not have work or residence permits and are at risk of deportation, exploitative working conditions, and inequitable access to education, health, and other essential services [33, 35]. Reduced opportunities for mobility and employment as a result of the COVID-19 pandemic have further increased levels of poverty and food insecurity, which in turn has elevated risk of child maltreatment and other negative child and family outcomes [36]. Recent studies with migrant and displaced families from Myanmar in Tak Province highlight concerns about physical and emotional abuse in the home, child labour and neglect, as well as growing mental health difficulties such as depression and anxiety among both parents and children [32, 37]. Results from the 2015 Myanmar Demographic Health Survey indicate high prevalence of harsh discipline in Myanmar, with 74% of children aged 2 to 14 experiencing emotional abuse and 43% experiencing physical abuse [38]. These findings suggest that the use of harsh punishment is widespread among the Myanmar population and may be further intensified by the chronic adversity experienced by migrant and displaced families living across the border in Thailand.

The study will be implemented in partnership with local non-governmental and community-based organizations Mae Tao Clinic, Help Without Frontiers Thailand Foundation, Inclusive Education Foundation, and Sermpanya Foundation. All partner organizations have extensive experience providing health, education, and psychosocial support to migrant and displaced families in Tak province. We refer to the study population as “migrant and displaced” to recognize that individuals and families from Myanmar have varying legal status in Thailand and have left their country of origin for a combination of economic and conflict-related factors [39].

Intervention and control conditions

Participants

Procedures

Outcomes

Existing Burmese translations of measures were obtained from study authors where available. All other measures were translated into Burmese following recommended procedures including forward translation to Burmese, back translation to English by an independent translator, and consultations with bilingual team members to review and resolve any discrepancies [44]. Survey instruments were pilot tested with the target population to assess acceptability, relevance, comprehensibility, and length prior to data collection. The survey included approximately 100 items and took 1–2 hours to complete, which was deemed acceptable during piloting.

Sample size calculation

The sample size calculation was performed by the trial statistician GJMT and based on a two-group comparison of one of the primary outcomes (physical abuse, emotional abuse, or positive parenting) assessed at T1. Assuming a significance level of 5% with an intra-cluster correlation coefficient (ICC) of 0.02 and a two-tailed test with 0.9 power, 40 clusters with 50 caregivers per cluster would be required to detect an effect size of 0.2. The total sample size was set at 2,200 caregivers (40 clusters with 55 families per cluster) to account for up to 10% attrition. Our estimate of expected effect size was based on a review of parenting interventions for child maltreatment prevention which found an average effect of 0.2 [61]. It is possible that the number of recruited caregivers will vary in each community. We will continuously monitor realized power as we enrol caregivers into the study and stop recruitment once the harmonic mean of cluster size and number of clusters reach 90% power, or until we reach maximum capacity, whichever comes first.

Statistical analysis

Trial results will be reported following the updated recommendations of the Consolidated Standards of Reporting Trials (CONSORT) 2010 statement: extension to cluster randomised trials [62]. Initial analyses will compare baseline characteristics of participants across the two study arms as well as participants who completed T1 assessments and those who did not.

Process evaluation

A process evaluation will be embedded in the trial, drawing on components of the RE-AIM framework to assess intervention delivery [63]. A combination of quantitative and qualitative data will be collected as part of the process evaluation. Quantitative data will include total attendance at community film screenings (including study participants and other community members) to provide an indication of reach, as well as attendance by study participants specifically to assess intervention uptake. Study participants will also be asked if they attended the film screening at the endline assessment (T1). We will not collect identifying information from film screening attendees who are not enrolled in the study. To assess fidelity of the post-film audience discussion, trained research staff will observe a random sample of at least 25% of community film screenings and complete a structured observation form to score which and to what extent components of the discussion were completed.

Qualitative data will be collected from a subsample of participating caregivers and adolescents who attended the film screening to examine acceptability, relevance, and perceived impacts of the intervention. Up to 6 focus group discussions with approximately 8 caregivers and 4 focus group discussions with approximately 8 adolescents aged 14 to 17 years will be conducted after participants complete the endline assessment. We will first purposively select up to 4 communities that have received the intervention to ensure representation from urban and rural settings. Using endline survey data, we will select caregiver and adolescent participants who report having attended the film screening and recruit them for the focus group discussions. We will stratify selection by caregiver and adolescent gender and age of index child to ensure representation across these variables. Focus group discussions will be used to examine participants’ experiences and reactions to the film-based intervention, perceived impacts on primary and secondary outcomes, any unanticipated outcomes, and potential causal mechanisms. In addition, we will conduct focus group discussions with community stakeholders and implementation staff to examine feasibility, acceptability, appropriateness, relevance, and sustainability of the intervention. Focus group discussions will be led by two local research staff with qualitative research experience, and data will be transcribed and translated for thematic content analysis [64].

Data management

Assessments will be administered using Open Data Kit (ODK) on password-protected tablets and uploaded daily onto a secure encrypted server. All submissions will be cross-checked before data are permanently removed from the tablets each week. Only authorized members of the research team will have access to data on the server. Data on the ODK server is automatically backed up daily and an additional weekly back up will be made onto a password-protected OneDrive folder accessible only to authorized research team members. Any hard copies of forms will be scanned and uploaded to the server upon completion of data collection, after which they will be permanently destroyed. Audio recordings and transcription of focus group discussions will also be stored in OneDrive; once transcription is completed and verified, audio recordings will be permanently deleted from recording devices.

All data will be de-identified prior to analysis. Participant responses will be identified only by a unique study code. Information linking the unique study code to participants’ identifying information such as names and contact information will be stored separately on a password-protected OneDrive folder accessible only to authorized research team members and destroyed at the conclusion of the study. Transcripts of audio recordings of focus group discussions will be anonymized by removing all personal identifying data. Audio recordings will be destroyed upon verification of the transcripts. De-identified data will be stored securely in perpetuity for research purposes only. Participants will be able to notify the study team if they want to withdraw their data until the personal linked data is destroyed. Statistical code and fully anonymized data will be made publicly available on Oxford Figshare and OSF following UK Research and Innovation (UKRI) guidelines [65].

Trial and adverse events monitoring

A trial management committee consisting of principal investigators, co-investigators, and research coordinators will monitor the implementation of study procedures. Any deviations from the study protocol will be reported to the trial management committee and to the research ethics committees as necessary. Significant modifications to the protocol will be submitted to the research ethics committees for approval and trial protocols will be updated in the online registry.

The research team will be trained on a safety protocol developed for this study that will include information on reporting and referral procedures if a safety concern is identified. Local implementing partners have a Child Safeguarding Policy that includes a protocol for reporting suspected child maltreatment, exploitation, or any other violation. All staff will be required to adhere to this policy. Data collection staff will be trained to record and report all adverse events (e.g., extreme participant distress, child safeguarding concerns) to the local research coordinators and co-investigators (KZL, SEP) within 24 hours. A committee comprised of the co-PI (AS), local co-investigator and protection lead (NNO), and research coordinators and co-investigators (KZL, SEP) will be responsible for reviewing all reported adverse events within 48 hours and determining any necessary action, which may include referral to the local Child Protection Unit. Information is included in the informed consent form to notify participants that disclosures of harm to self or others may be reported.

To reduce risk of participant distress, data collection staff will be trained to remind participants that they can choose to skip any questions or withdraw from the study at any time with no negative consequences. All participants will be provided information about available mental health and psychosocial support services during each assessment point. No interim analyses are planned. The local research coordinators and co-investigators are responsible for ensuring timely follow up of any adverse events.

Ethics

Ethics approval was obtained from the Institutional Review Board of the Institute for Population and Social Research (IPSR) at Mahidol University in Thailand (COA. No. 2022/11-217), as well as the ethics committees of the research team’s respective institutions. The study protocol was also reviewed by the Community Ethics Advisory Board on the Thailand-Myanmar border.

Inclusivity in global research

Additional information regarding the ethical, cultural, and scientific considerations specific to inclusivity in global research is included in S4 File.

Trial status

Recruitment of the first study participant began on 23 February 2023. At the time of submission, baseline data collection has been completed in 33 communities and the film-based intervention has been delivered in 2 communities.