Developing a socio-ecological prevention framework for child sexual abuse in Gweru District, Zimbabwe: engaging stakeholders for contextual solutions

Context and relevance. Child sexual abuse remains a major public health and human rights issue, especially in low and middle-income countries where prevention systems are often fragmented and lack coherence. Although socio-ecological frameworks are widely used, they often remain descriptive rather than being translated into an integrated prevention architecture. Objective. To develop a context-specific CSA prevention framework for Gweru district, Zimbabwe, integrating global evidence with stakeholders' local expertise. Methods and materials. The study was conducted in Gweru District, Zimbabwe, involving purposively selected teachers, psychologists, and social workers with experience in child sexual abuse. A multi-phase participatory design was employed, incorporating a structured literature review, semi-structured interviews, a Delphi process, and a co-creation workshop. Data were analysed using socio-ecologically guided thematic analysis. Results. Findings informed the development of a socio-ecological prevention framework comprising coordinated interventions across the individual, microsystem, mesosystem, exosystem, and macrosystem levels. The framework emphasises coordination, collaboration, reporting, and monitoring as cross-cutting mechanisms for strengthening child protection systems. Stakeholders consistently preferred system alignment over isolated programming elements. Conclusions. The framework extends the practical application of the socio-ecological model by translating prevention activities across ecological levels into a coordinated and contextually grounded child protection framework. The findings highlight the importance of collaboration, reporting, monitoring, and accountability mechanisms in strengthening child sexual abuse prevention systems.

Introduction

Child sexual abuse (CSA) represents a profound and pervasive global public health and human rights concern, recognised for its devastating long-term consequences that profoundly impact individuals, families, and entire communities (Bryce, 2020). These consequences can manifest as severe mental health issues such as anxiety, depression, post-traumatic stress disorder, and even suicidal ideation; often persisting into adulthood. Survivors may also experience cognitive disturbances, difficulties with memory and attention, leading to poor academic performance and low self-esteem. Furthermore, maladaptive coping skills, including substance abuse and risky sexual behaviours are common, perpetuating a cycle of distress and hindering the formation of healthy relationships. In addition to individual suffering, the stigma and shame associated with CSA can lead to social isolation and complicate the pursuit of justice and healing (Laajasalo et al., 2022). This issue is particularly critical in low- and middle-income countries (LMIC) such as Zimbabwe, where prevention systems are often fragmented, under-resourced, and lack coordinated strategies to address the multi-faceted nature of the problem, thus exacerbating these long-term impacts (Adam, Ahmad, Kassim, 2024). Despite significant international and national efforts to combat this pervasive issue, CSA remains distressingly prevalent (Mensah, Abdullah, Cudjoe, 2024).

To adequately address this intricate problem, the socio-ecological model of Bronfenbrenner (1996) offers a robust theoretical framework, providing a comprehensive lens through which to understand the aetiology and prevention of CSA. This model conceptualises CSA not as an isolated event but as an outcome shaped by the dynamic interplay of influences across multiple interacting levels, namely the individual, microsystem, mesosystem, exosystem, and macrosystem. Adopting a public health perspective, this framework moves beyond simplistic, individual-blaming explanations, acknowledging that both risk and protective factors for CSA victimisation and perpetration are distributed and interact across these interconnected spheres. As Martinello emphasises, such a holistic approach is vital for developing integrated and effective prevention and intervention strategies that accurately target the systemic and social roots of sexual violence, rather than merely addressing its symptoms (Martinello, 2020). The socio-ecological model thus provides a critical foundation for developing contextually relevant and effective prevention programs, particularly in resource-constrained environments where vulnerabilities are often compounded by systemic weaknesses. Bronfenbrenner's socio-ecological theory, which underpins this model, posits that human development is influenced by interactions within a series of nested systems, from the immediate environment to broader cultural and societal contexts (Quadara, El-Murr, Latham, 2017). Understanding these interconnections is crucial for designing interventions that are truly impactful.

The existing literature demonstrates that vulnerability to CSA is influenced by a complex interplay of factors spanning these ecological levels. At the individual level, personal characteristics and developmental factors may influence vulnerability to CSA (Quadara, El-Murr, Latham, 2017). For perpetrators, individual-level risk factors are diverse, encompassing a range of psychological, behavioural, and experiential characteristics. In their systematic review and meta-analysis of risk factors associated with CSA perpetration, D.J. Whitaker and colleagues grouped the identified risk factors into broad categories, which are family factors, externalising behaviours, social deficits, sexual problems, and problematic attitudes or beliefs (Whitaker et al., 2008). They further highlighted that child sex offenders often differ significantly from non-sex offenders, particularly concerning sexual problems and attitudes, and cognitions that minimise perpetrators' culpability. In addition to that, A. Quadara, A. El-Murr and J. Latham indicate that these non-violent child sex offenders may employ primitive and immature defence mechanisms such as dissociation (Quadara, El-Murr, Latham, 2017). They went on to say that perpetrators often utilise sophisticated grooming tactics, building trust and making the child feel special, gradually introducing sexual content, and enforcing secrecy, effectively isolating the child.

Apart from those factors related to the offender, there are factors at the individual level that put victims of sexual abuse at risk. These factors include, but are not limited to, social isolation, family dysfunction, physical and intellectual disabilities, lack of secure attachments, a history of maternal CSA, and exposure to family violence (Quadara, El-Murr, Latham, 2017). Other factors associated with victimisation, as highlighted by Jones et al. (2020), include a child's use of alcohol or drugs, having a physical or learning disability, being cared for by someone other than parents, a history of sexual activity, or a history of psychiatric support. Evidence from meta-analytic research indicates that individual-level risk factors for victimisation include prior exposure to abuse, family dysfunction, and parental vulnerability, which may increase children’s susceptibility to further victimisation (Assink et al., 2019). Studies have also suggested that socio-economic factors like low education and low income can be associated with higher levels of sexual violence victimisation (Bows, 2017; Dallelucci et al., 2019). These elements at the individual level, therefore, underscore how individual predispositions and life circumstances create varied levels of risk.

Apart from the individual level, victimisation also occurred at the microsystem level. At this level, close relationships such as family and immediate caregivers play a critical role in increasing the risk of experiencing or perpetrating CSA. Within familial relationships, CSA is highly prevalent, particularly intrafamilial abuse, which involves a wide range of perpetrators from biological relatives to stepfamily members. Dysfunctional family dynamics, such as poor cohesion and domestic violence, create environments where sexual abuse can thrive. Children in single-parent or blended families may be at a heightened risk of victimisation, and maternal dissatisfaction with marriage can also contribute to an increased likelihood of abuse (Quadara, El-Murr, Latham, 2017). Moreover, the quality of the parent-child relationship is also a critical predictor of a child's vulnerability to CSA. Parents play a crucial role as protectors, both directly through supervision, monitoring, involvement, and communication, and indirectly by promoting their children's competence and well-being. As Rudolph et al. (2017) found, parental interventions show promise in enhancing communication and supervision practices among caregivers. Therefore, the absence of this parental involvement creates a good breeding ground for children's victimisation. Furthermore, Quadara, El-Murr and Latham (2017) postulate that peer and social interactions also contribute to the risk profile, as social isolation can leave victims with limited avenues for disclosure and support. Therefore, prevention strategies at this level include strong and healthy social connections, parenting programmes and peer-based interventions designed to strengthen communication, reduce conflict within families, and promote healthy relational dynamics (Finkelhor, 2009; Rudolph et al., 2017).

The third level, which is also key in promoting vulnerability and acts as a prevention ground, is the mesosystem. Interactions between immediate environments, such as relationships between families and schools, influence children’s exposure to risk on the one hand and protection on the other hand (Quadara, El-Murr, Latham, 2017). According to Morris et al. (2017) neighbourhood characteristics are important factors in shaping the prevalence of CSA, where communities exhibit high rates of child abuse and neglect. Risk factors at this level include inadequate social support systems within the community and weak community-level accountability mechanisms. Limited social sanctions against perpetrators have also been identified as key contributors to the persistence of CSA in these contexts. Prevention efforts at this level seek to identify and modify the characteristics of settings that increase or buffer against the risk for violence, particularly the social, economic, and environmental characteristics of neighbourhood, school, and workplace environments (Quadara, El-Murr, Latham, 2017). As Quadara, El-Murr and Latham describe, community-level approaches include providing children with safer environments by addressing "hot spots" of CSA, ensuring collective efficacy, and using bystander programmes. Additionally, Centers for Disease Control and Prevention (Fortson et al., 2016) also highlighted those modifications to the physical and social environment, such as creating safer community spaces, strengthening institutional safeguards, and implementing screening procedures in youth-serving organisations, are important strategies for reducing risk. Moreover, Quadara, El-Murr and Latham (2017) noted that community mobilisation, gender-sensitive programming, and faith-based involvement aiming to shift harmful social norms and reduce the stigma associated with sexual abuse are also key things that need to be considered.

Beyond the mesosystem, the fourth level, which also needed to be looked at, is the exosystem. This level refers to broader institutional and environmental factors that indirectly influence children’s experiences. These include social services, law enforcement systems, and workplace conditions affecting caregivers. As noted by Quadara, El-Murr and Latham (2017), on the one hand, weak institutional coordination, limited reporting mechanisms, and ineffective service delivery at this level increase vulnerability by constraining protective responses. On the other hand, strong and well-organised coordination within institutions and an effective reporting mechanism accompanied by proper handling of reported cases also increases protection of children from victimisation.

Finally, at the macrosystem level, broader cultural values, societal norms, economic conditions, and legal frameworks shape attitudes toward and responses to CSA (Quadara, El-Murr, Latham, 2017; Finkelhor et al., 2013). At this level, these broader socio-cultural norms, which include gender inequality, stigma surrounding disclosure, and silence around sexuality, perpetuate conditions in which CSA occurs (Adam, Ahmad, Kassim, 2024; Selengia, Thuy, Mushi, 2020). Together, these factors highlight that CSA is not solely an individual-level issue, but one that is embedded within broader structural and systemic contexts. Moreover, CSA at this level is further shaped by broader socio-economic and policy environments. Cultural values and belief systems significantly influence how CSA is perceived, reported, and addressed. In some collectivist contexts, lower reported rates of CSA may reflect underreporting due to stigma, cultural taboos, and norms discouraging disclosure rather than actual lower prevalence (Danaeifar et al., 2022; Selengia, Thuy, Mushi, 2020). Certain cultural ideologies, including beliefs that reinforce male dominance and sexual entitlement, may further contribute to the persistence of CSA and warrant attention in prevention efforts (Adam, Ahmad, Kassim, 2024). Additionally, socioeconomic conditions also play a critical role. Poverty and economic instability can increase vulnerability by disrupting family structures and caregiving arrangements. For example, economic hardship may lead to parental migration, leaving children in less supervised or alternative care arrangements, thereby increasing exposure to risk (Finkelhor et al., 2013; Quadara, El-Murr, Latham, 2017). However, legal and policy frameworks are also equally central to prevention and response. Laws that criminalise CSA and protect children’s rights are essential for improving reporting and accountability. Evidence suggests that many policy responses remain reactive rather than preventive, with a strong emphasis on identification and reporting rather than addressing upstream risk factors (Finkelhor, 2009).

Despite the widespread application of these socio-ecological frameworks in violence prevention research, a significant research gap persists. These frameworks are frequently used descriptively rather than effectively translated into a coordinated system architecture that comprehensively addresses the multi-level interplay of factors contributing to CSA. This fragmentation is particularly evident in LMICs like Zimbabwe, where prevention efforts often operate in parallel rather than in a coordinated manner, limiting their systemic leverage. Quadara, El-Murr and Latham (2017) observed a relatively fragmented understanding of prevention across different service systems and agencies. This fragmentation leads to inconsistent criteria for funding programs, poor program design, sluggish development of best practices, disconnected prevention activities, and weak evaluation methodologies. Individualised interventions are frequently limited, focusing only on the individual and potentially placing undue responsibility on children to prevent abuse. As Quadara, El-Murr and Latham emphasise, risk and protection can accumulate and interact across all levels of the socio-ecological model, contributing to the inherent complexity in designing and implementing truly effective prevention strategies. Research consistently demonstrates that individual-level interventions alone are insufficient to address the systemic drivers of child sexual abuse, underscoring the need for coordinated, multi-level prevention approaches (Fortson et al., 2016; Martinello, 2020; McKibbin et al., 2020). This fragmentation, compounded by issues like poor intersectoral cooperation, constitutes a primary obstacle to effective child protection, particularly in resource-constrained contexts where expertise and data are scarce (Adam, Ahmad, Kassim, 2024; Danaeifar et al., 2022).

Therefore, the overarching aim of this study is to develop a context-specific socio-ecological prevention framework for CSA. The study specifically sought to bridge the identified gap by moving beyond a descriptive application of the socio-ecological model to develop a coordinated system architecture. By integrating global evidence with locally grounded stakeholder expertise, the objective of this study was to identify key areas for intervention and provide a framework for developing targeted, effective, and coordinated prevention strategies. These strategies are designed to be responsive to the unique socio-ecological dynamics of Gweru District and offer transferable insights for comparable resource-constrained settings. This study extends the practical application of the socio-ecological model by integrating governance, institutional coordination, and stakeholder engagement within a contextually grounded prevention framework. By translating socio-ecological principles into practice, the study offers an applied socio-ecological prevention architecture for strengthening child protection in Zimbabwe and comparable resource-constrained settings.

Research questions

This study was guided by the following research questions:

1. What socio-ecological risk and protective factors influence CSA prevention in Gweru District, Zimbabwe?
2. What prevention strategies and intervention approaches are identified in the international CSA prevention literature?
3. Which contextually relevant prevention components can be prioritised through stakeholder consensus?
4. How can these components be integrated into a coordinated socio-ecological framework for CSA prevention?

Objectives of the Study

The overall aim of this study was to develop a contextually relevant socio-ecological framework for child sexual abuse prevention in Gweru District, Zimbabwe.

To achieve this aim, the study sought to:

1. Examine international evidence on CSA prevention strategies and intervention approaches.
2. Identify socio-ecological risk and protective factors influencing CSA prevention in Gweru District.
3. Establish stakeholder consensus on priority prevention components through a Delphi process.
4. Develop a coordinated socio-ecological child sexual abuse prevention framework informed by empirical evidence and stakeholder perspectives.

Materials and methods

Study design. This study employed a staged, multi-phase participatory research design to develop a context-specific socio-ecological prevention framework. The design integrated global evidence synthesis with locally grounded stakeholder engagement to ensure both empirical rigour and contextual relevance. Guided by Bronfenbrenner’s socio-ecological framework, the study examined CSA across interconnected levels namely, the individual, microsystem, mesosystem, exosystem, and macrosystem, allowing for a comprehensive understanding of risk and protective factors within the local context. The research was conducted in Gweru District, Zimbabwe, a setting characterised by socio-economic diversity and established child protection structures. The study combined four sequential and interrelated phases: (i) a structured literature review of international CSA prevention approaches; (ii) qualitative semi-structured interviews with key stakeholders; (iii) a two-round Delphi consensus process; and (iv) a stakeholder co-creation workshop. The literature review was undertaken to identify commonly reported risk factors, protective mechanisms, and prevention strategies documented in international studies. Findings from the review informed the development of the interview guide and provided an evidence base for subsequent stakeholder consultations and consensus-building activities.

Participants and sampling.

A total of 12 participants (Table 1) were purposively selected based on their professional experience in child protection and direct involvement in CSA prevention and response. The sample comprised 8 primary school teachers, 2 educational psychologists, and 2 social workers. Participants ranged in age from 30 to 50 years and included 8 females and 4 males. All participants were government employees and had a minimum of five years of professional experience working with children and child protection-related issues. Purposive sampling was employed to ensure that participants possessed relevant professional knowledge and practical experience regarding CSA prevention and response. Inclusion criteria required participants to be currently employed within the education, social welfare, or psychological services sectors and to have direct experience working with children or handling CSA-related cases. Individuals without a minimum of five years of professional experience in these sectors were excluded. All 12 participants took part in the qualitative interview phase. Ten participants subsequently participated in the first Delphi round, comprising all educational psychologists, all social workers, and six teachers. Nine participants completed the second Delphi round following the withdrawal of one teacher due to competing professional commitments. The same nine participants participated in the stakeholder co-creation workshop. The retention of participants across phases ensured continuity and strengthened the validity of the consensus-building process.

Given the exploratory and participatory nature of the study, sample adequacy was determined by the depth of professional expertise represented rather than statistical representativeness. The selected participants provided rich, practice-based insights from multiple sectors involved in child protection, making them well suited to contribute to the development of a contextually relevant prevention framework.

Table 1 / Таблица 1

Participant characteristics (N = 12)

Описание выборки (N = 12)

Participants were purposively selected based on their professional involvement in child protection and CSA prevention and response. All participants were government employees with at least 5 years of professional experience.

Data collection procedures. Data collection was conducted across four interconnected phases. The first phase involved a structured literature review of published literature on CSA prevention. Relevant studies were identified through searches of Google Scholar, Scopus, and PsycINFO. Searches were conducted between October and December 2025 and focused on literature published between 2010 and 2025. Combinations of keywords including "child sexual abuse", "prevention", "intervention", "child protection", "safeguarding", and "socio-ecological model" were used. The review prioritised peer-reviewed journal articles, systematic reviews, policy documents, and international frameworks relevant to CSA prevention. Studies were included if they addressed CSA risk factors, protective factors, prevention interventions, safeguarding approaches, or socio-ecological prevention strategies. Publications focusing exclusively on offender treatment, unrelated forms of child maltreatment, or studies lacking direct relevance to prevention were excluded. The initial search yielded 96 records. Following removal of duplicates and screening for relevance, 42 publications were retained for review. The review focused on identifying commonly reported risk factors, protective factors, intervention approaches, and implementation challenges across different socio-ecological levels. Findings from this review informed the development of interview schedules and the subsequent Delphi consensus process. The second phase comprised qualitative semi-structured interviews with selected participants. Interviews were conducted either in person or via online platforms, depending on participant availability, and lasted approximately 45–60 minutes. The interview guide was informed by findings from the structured literature review and structured around socio-ecological domains to explore perceived risk factors, protective mechanisms, systemic gaps, and opportunities for coordinated intervention.

A two-round Delphi process was conducted to refine and prioritise prevention components identified through the structured literature review and qualitative interviews. Participants were asked to rate the relevance, contextual applicability, and implementation feasibility of proposed intervention components using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). A consensus threshold of 75% agreement was established a priori for the retention of intervention components. Components not meeting this threshold in Round 1 were revised based on participant feedback and re-evaluated during Round 2. Ten participants completed Round 1, while nine completed Round 2. The Delphi process enabled systematic refinement of prevention components while reducing the influence of dominant voices and promoting collective expert judgement.

The final phase consisted of a stakeholder co-creation workshop, where participants collaboratively synthesised findings from previous phases into a structured prevention framework. The workshop included facilitated discussions, system-mapping exercises, and consensus-building activities.

Data analysis. Data analysis was conducted using an iterative and integrative approach across all phases of the study. Qualitative data from the semi-structured interviews were analysed using deductive thematic analysis guided by the socio-ecological framework. Transcripts were read repeatedly to ensure familiarisation, followed by coding of data segments according to ecological levels (individual, microsystem, mesosystem, exosystem, and macrosystem). Codes were refined and grouped into broader themes reflecting patterns of risk and protective factors. The analysis was conducted manually, with continuous comparison across participants to ensure consistency and depth.

Data from the structured literature review were analysed using a structured narrative synthesis approach. Extracted studies were examined to identify key intervention types, target populations, and implementation strategies. Findings were organised into thematic categories aligned with the socio-ecological framework. The aim was to identify recurring patterns, conceptual trends, and gaps in the literature to inform subsequent phases of the study. Data generated through the Delphi process were analysed using descriptive and consensus-based techniques. Initial qualitative inputs from Round 1 were thematically grouped to develop structured items for Round 2. Participants then rated these items using a five-point Likert scale. Measures of central tendency and percentage agreement were used to assess consensus, with a predefined threshold of 75% agreement applied to determine which components were retained, revised, or excluded. Finally, data from the stakeholder co-creation workshop were analysed through thematic synthesis, integrating workshop discussions, system-mapping outputs, and participant reflections. The analysis focused on identifying areas of convergence across phases and translating these into actionable prevention components. Particular attention was given to how participants collectively interpreted relationships between risk and protective factors across ecological levels.

Across all these phases, analysis was iterative and integrative, allowing insights from the structured literature review, interviews, Delphi process, and workshop to inform one another and strengthen the overall coherence of the study.

Ethical considerations. Permission to conduct the study was obtained from the Ministry of Primary and Secondary Education and the Ministry of Labour and Social Welfare in Zimbabwe prior to data collection. Participation was voluntary, and all participants provided informed consent before taking part in the study. Participants were informed of the purpose of the research, their right to withdraw at any stage without penalty, and the measures taken to ensure confidentiality and anonymity. To protect participant identities, no personally identifiable information was collected or reported. Data were securely stored and used solely for research purposes. Given the professional nature of the participant group and the non-clinical focus of the study, particular care was taken to ensure that discussions relating to CSA remained focused on professional experiences and prevention systems rather than personal disclosures.

Results

The findings of this study unfold in four interconnected phases, namely: (i) a structured literature review of global approaches to CSA prevention; (ii) qualitative semi-structured interviews with stakeholders; (iii) a two-round Delphi consensus process; and (iv) co-creation workshop that informed the development of a socio-ecological prevention framework (Table 4). Together, these phases show how CSA vulnerabilities are amplified in interconnected socio-ecological systems across the individual, microsystem, mesosystem, exosystem, and macrosystem levels and how the activation of protection requires deliberate coordination in the same areas. The results are grouped in three overarching themes, namely Theme 1, which presents the patterns and gaps identified in the global evidence base, Theme 2, which illustrates the socio-ecological risk factors and Theme 3, which also illustrates socio-ecological protective factors. All these themes are illustrated below.

Theme 1: Findings from the structured review of international CSA prevention approaches

The structured literature review synthesised the findings of the peer-reviewed studies examining interventions designed to prevent and respond to CSA in a variety of global contexts. Interventions were grouped into five socio-ecological domains, namely school-based prevention, family-oriented programming, community mobilisation and norm transformation, legislative and systemic reform, and trauma-informed therapeutic responses. While approaches vary in scope and intensity, the overall picture is one of innovation and fragmentation at the level of the environment. School-based prevention has been identified as the most consistently documented strategy, with programmes typically addressing body autonomy, distinguishing between safe and unsafe touch, promoting disclosure pathways, and identifying trusted adults. Across settings, these interventions have shown short-term improvements in children's knowledge and disclosure capacity. However, evidence has repeatedly highlighted persistent knowledge gaps, poor long-term monitoring and variability in the level of compliance. Hybrid and digitally adapted models have shown growing promise, especially in contexts affected by school disruption; however, technological inequalities limit their reach in LMICs.

Moreover, family-based interventions also emphasised caregiver education, communication, and supervision practices to strengthen relationship protection. Parenting programmes have been associated with increased caregiver awareness and a reduction in harsh disciplinary practices associated with CSA vulnerabilities. However, the involvement of male carers remained limited, and family strategies were rarely systematically integrated with school efforts, which reduced cross-system support and sustainability. Furthermore, faith-based dialogues, community mobilisation strategies, and gender transformation programming were seen to be associated with positive changes in attitudes and increased willingness to take action, while structural reforms, which include mandatory reporting laws, child-friendly court procedures, and one-stop shops, were seen to be strengthening the response and support mechanisms for survivors. However, enforcement gaps, limited resources, and institutional fragmentation have been recurrent problems, especially in the context of LMICs. Moreover, many evaluations relied on proxy indicators rather than direct measures of CSA incidence.

Theme 2: Socio-ecological risk factors

Qualitative stakeholder interviews revealed that CSA in Gweru District is driven by interrelated risk factors within individual, microsystem, mesosystem, exosystem, and macrosystem systems. For analytical clarity, these pathways are understood as comprising (a) factors that increase children’s vulnerability to victimisation, and (b) factors that enable or facilitate perpetration within specific relational and systemic contexts. The participants consistently framed CSA not as an isolated incident but as the result of systemic weaknesses interacting at different levels. As a result, several subthemes emerged from this theme, including constrained child agency and emerging digital exposure; family silence, economic migration, and relational instability; community norms, patriarchy, and cultural silence; and institutional fragmentation and weak enforcement.

Subtheme 2.1: Child vulnerability and emerging digital exposure

Stakeholders described children's developmental dependence, limited information on consent and economic vulnerability as key risk factors. These findings primarily reflect several child reliability factors at the individual and microsystem levels, rather than causes of perpetration. As one teacher explained:

“The unstable economy in our country has caused a lot. Because of economic hardship in our country, several families are left headed by either an older child (child-headed family) or young children are left in the custody of an uncle or in some worst scenarios, children are left with neighbours whilst parents are moving to other areas of the country looking for food. In some cases, parents even leave the country for greener pastures, hence exposing children to high levels of risk of abuse.” (Teacher)

Another issue which participants noted is that of children with disabilities. Several issues put these children at risk, for example, some of them could not speak properly, such that they cannot even be able to relate the abuse. Some depend much on caregivers for almost everything, which also increases their vulnerability. This was reiterated by one of the participants, from the Department of Social Welfare, who stated that:

“We have received several cases of abuse involving children with several disabilities. It is so pathetic that such children are also victims of abuse. The issue of overdependence on other people increases their vulnerability. We have also realised that even if these children are taken for an interview, some of them have challenges in giving a proper report, and this also gives advantage to the perpetrators who, at times, just go free without being convicted.” (Social Worker)

Additionally, participants noted that digital exposure has silently and unknowingly introduced new dimensions of risk regarding sexual abuse. One of the participants from the Ministry of Primary and Secondary Education mentioned that:

“Children now have phones that they use to store nude pictures and share with others at school. After viewing these nude pictures, some older students will then try to experiment with what they have seen with other young children at school or in their neighbourhoods, hence promoting abuse.” (Teacher)

Therefore, these accounts show how vulnerability to sexual abuse at the individual level is continuously shaped not only by the developmental stage but also by the relational and structural conditions that increase vulnerability within insufficiently protective environments.

Subtheme 2.2: Family silence and economic migration

Findings indicate that family systems can function as both protective and high-risk environments. Within this context, a distinction emerges between child vulnerability factors, such as disrupted caregiving arrangements due to economic migration, and perpetrator-enabling conditions, including the protection of offenders within family structures. The silence around sexuality and the protection of offenders were the most recurrent themes. Evidence gathered from participants proved that the microsystem-level dynamics that may enable perpetration include the protection of offenders by close relatives, especially in situations where the perpetrator is a close relative, and particularly where the perpetrator holds economic or social power within the household. One of the participants, a school psychologist, clearly mentioned that:

“Families tend to defend the abuse, especially when the perpetrator is a close relative, to protect the family's reputation, and this results in recurrence of abuse since the perpetrator will know that no one is going to report him or her. Imagine your uncle, who is looking after you, paying fees and buying food for you, is to be incarcerated for 10 years. No one will be comfortable letting that person go beyond bars.” (Psychologist)

The participant added, “In such cases, it will be very difficult for us, as educational psychologists, to fully assist victims and survivors.”

In addition to silence in reporting, the hard economic environment in Zimbabwe was also mentioned as another factor that exacerbates risks to sexual abuse in most communities. Participants noted that because of these economic hardships, most children are left behind by their parents as child-headed families or under the custodianship of relatives while their parents migrate for greener pastures. A sentiment from one participant reinforced this when she stated:

“The relocation of parents in search of greener pastures leaves children in the care of relatives who may end up being the abuser. I was surprised this other day on my way back from work, and it was late, when I saw some children playing football along the road. It was when I asked my colleague, who told me that these children were left alone when their parents left for South Africa. As a result, this kind of situation, where children are left alone, exposes them to sexual abuse." (Teacher)

These findings suggest that economic hardship, caregiving disruptions, and family instability increase children's vulnerability to CSA in various communities, hence requiring attention.

Subtheme 2.3: Community norms, patriarchy, and cultural silence

Beyond individual and microsystem-level risks, findings also revealed that the mesosystem-level contexts in which children interact are also replete with risks that render them vulnerable to abuse. It was revealed that societal norms and stigma are powerful risk amplifiers. These factors are best understood as mesosystem-level conditions that enable perpetration and inhibit disclosure, rather than characteristics of the child. This can be reinforced by sentiments from one participant who stated that:

“Apart from the efforts being made by the new curriculum, which encourages teaching children as early as preschool about body parts which they should not allow anyone to touch, it is still a taboo in our communities for one to be head openly speaking about sex, even with our own children. Therefore, by so doing, it will make it difficult for our own kids to speak out after being abused, hence promoting recurrence of victimisation.” (Teacher)

Stigma and victim-blaming attitudes were also found to reinforce silence and discourage disclosure. Participants described how children are sometimes blamed for abuse, especially when it involves girls; this further marginalises victims and insulates perpetrators from accountability. One psychologist said:

“Child blaming is still a big issue in our communities, even in schools. Both teachers and parents should give kids enough time to explain in the event of abuse. Instead, they tend to blame the child, saying it’s your fault that you being abused. They usually say to their children that they love accepting goodies from strangers, which leads them to be abused.” (Psychologist)

Subtheme 2.4: Institutional fragmentation and weak law enforcement

Additional factors contributing to vulnerability highlighted by participants are those related to institutional fragmentation and weak law enforcement. These findings reflect exosystem and macrosystem-level enabling conditions, which shape opportunities for perpetration and weaken protective responses, rather than increasing vulnerability at the individual level. It was noted that high levels of inconsistent enforcement of the laws by law enforcers, starting from police officers, extending to the judiciary system due to corruption, have undermined the reduction of child abuse, thereby promoting it. One participant highlighted that:

“When laws and regulations are not enforced effectively, the problem of CSA becomes very difficult to deal with.” (Psychologist)

This was also reinforced by another participant who noted:

“Corruption means perpetrators are left without being convicted since law enforcers are lured to accept bribes.” (Teacher)

These findings indicate that institutional fragmentation contributes significantly to CSA vulnerability. Consequently, participants indicated that CSA may be increasing in some communities.

Theme 3: Socio-ecological protective factors

While stakeholders have described several ways in which risks can be amplified in socio-ecological systems, they have also identified mechanisms that can counteract these risks and disrupt these trends. Protective mechanisms emerged not as isolated programmatic proposals, but as systemic conditions that, if deliberately aligned, can address vulnerabilities at individual, microsystem, mesosystem, exosystem, and macrosystem levels.

Subtheme 3.1: Knowledge, agency, and digital awareness

At the individual level, participants underlined the importance of providing children with knowledge and practical skills to strengthen their self-protection and disclosure capacities. Participants consistently highlighted education on body safety, understanding consent and digital literacy as basic safeguards. One participant explained:

“As teachers, we are trying our best to educate these young ones on how best they can safeguard themselves by not allowing anyone to touch their bodies. We have realised that children should be taught about body parts which they should always cover and make sure that no one touches them, even if someone they know or are staying with. Even if the person is of the same sex with them.” (Teacher)

Beyond body safety education, digital exposure also emerged as a modern area requiring protection activation. Participants described increased access to mobile phones and social media as a potential weakness and an opportunity for empowerment, provided it is accompanied by structured digital literacy education. A teacher explained that:

“The issue of technology gives headaches to us teachers. As teachers, we always urge children not to spend most of their time on phones and to use them only when researching their homework from the internet. We always teach them about the advantages and disadvantages that come with technology. Therefore, if they use technology responsibly and understand the dangers online, they can enhance their awareness and ability to seek help within supportive environments.” (Teacher)

These narratives, therefore, suggest that supporting children requires not only knowledge provision but also strengthening the protective environments in which they are embedded.

However, stakeholders have made it clear that individual empowerment alone is not enough without strengthening the microsystems in their entirety. Therefore, the following subtheme explains how factors within the microsystem can also help in protecting against CSA.

Subtheme 3.2: Communication and relational trust

Families were repeatedly described as key sites of increased protection. Where silence, denial, and economic pressure accentuated vulnerability, open communication and oversight were framed as protective forces. This was emphasised by one participant who mentioned that:

“One of the key issues which we always try to impart to parents is that of being approachable. Children should feel comfortable discussing or sharing issues related to abuse openly and without fear with their parents and guardians. Once children are comfortable doing that, it will be easy for us as professionals to quickly gather information surrounding the abuse. For that to be achieved, the home environment should always be conducive and friendly.” (Social worker)

Other participants also stressed that mutual trust allows timely disclosure and breaks cycles of silence surrounding abuse, which is the key mechanism perpetuating abuse. Structured parent education and guidance for carers was seen as an essential associated with reducing vulnerability. However, participants further recognised economic pressures and migration patterns as limiting protection at the microsystem level. Protection activation at this level has therefore been conceptualised not only as parental responsibility but also as a need for a supportive community and a structural scaffolding to maintain relational security.

Subtheme 3.3: Norm transformation and collective accountability

At this point, participants described cultural silences and taboos as strong drivers of risk. Yet these same structures have been identified as potential places of transformation. Traditional leaders, faith leaders, and school administrators were seen as powerful gatekeepers who could shift community norms about sexuality, disclosure, and responsibility. One participant observed:

“I have also realised that in most cases when the pastor speaks, people tend to listen. So, if this issue is given slots within church services, many people, both old and young, can benefit a lot. Pastors should try to explain to people the link between spirituality and child sexual abuse and how grooming children in a spiritual way can prevent them from being victims or perpetrators. Apart from teachers, we also have our own traditional headmen and chiefs who always have community gatherings with people in our communities. These key stakeholders should also utilise this opportunity to educate people about the danger of being found on the wrong side of the law. By so doing, this can also reduce cases of abuse in our communities.” (Teacher)

Participants also identified public dialogue, awareness-raising campaigns, and collective condemnation of abuse as mechanisms to reduce stigma and encourage reporting. They emphasised the need to present CSA as a community concern rather than a private family matter. This shift from secrecy to collective accountability was a critical turning point in the socio-ecological system.

Subtheme 3.4: Institutional credibility and coordination

Protection at the exosystem and macrosystem levels has emerged as a matter of institutional trust and coordination. Participants repeatedly expressed the view that weak law enforcement and corruption undermine the deterrent effect. On the other hand, consistent enforcement and transparent judicial procedures were described as strong protection signals. One participant stated:

“The greatest challenge we are facing as practitioners is that of perpetrators who, after being apprehended, are found walking scot-free without being sentenced. We have realised that the reason why this is happening is because of the levels of corruption within the judiciary systems. Corruption has destroyed our systems. It starts from the officer who did the dockets, the public prosecutor, up to the magistrate and also not leaving out the witnesses. This goes as far as even if someone engages a lawyer. After engaging a lawyer as the complainant, that same lawyer will go in as far as conniving with the accused person, such that if the accused person offers the lawyer more than the complainant has offered, that same lawyer will make sure the sentence of the accused person is reduced or make sure that he/she is released or given free bail. So, what it means here is that the lawyer benefits from both the accused person and the complainant. Therefore, if laws are to be properly enforced and everyone convicted is sentenced accordingly without fear, favour or bias, truly as societies we will witness great changes in relation to cases of child sexual abuse” (Psychologist)

Additionally, participants also stressed the need for coordination between schools, social services, health institutions, and law enforcement. Participants described fragmentation as allowing cases to slip through the cracks, while structured referral channels and information-sharing mechanisms act as safety nets. The activation of protection at the exosystem and macrosystem levels has not been conceived as a stand-alone policy reform, but as a cross-sectoral convergence. Participants further envisaged mechanisms whereby institutions would operate in synergy rather than in isolation, thereby strengthening accountability and support for survivors.

Therefore, it has been seen that across all ecological levels, protection activation has been characterised by deliberate alignment of relational openness, norm-shifting, institutional empowerment, and knowledge-empowerment (Table 2). This insight, therefore, informed the subsequent Delphi consensus phase, which sought to prioritise and operationalise these empirically grounded protective elements in a coherent framework of prevention.

Table 2 / Таблица 2

Summary of socio-ecological risk factors, protective factors,
and prevention framework components

Систематизация социоэкологических факторов риска,
защитных факторов и компонентов модели профилактики

From protective mechanism to system alignment: Delphi consensus prioritisation

The preceding qualitative findings demonstrated that both risk factors and protective factors coexist within the socio-ecological systems of Gweru District. While participants clearly articulated processes for activating protection, ranging from child empowerment and family communication to the transformation of community norms and institutional responsibility, these mechanisms were frequently described as uneven, fragmented, and inadequately coordinated. The qualitative stage thus served to identify the existing protective features and their potential activation points within the system.

The subsequent analytical step involved identifying which of these empirically grounded components were most contextually feasible, relevant, and strategically important for designing a structured prevention framework. To move from descriptive insights to operational prioritisation, a two-step Delphi process was employed to refine and prioritise preventive components derived from both the overarching structured literature review and the local qualitative findings. The Delphi method was selected for its capacity to generate structured expert consensus while mitigating dominance bias through anonymous feedback. Qualitative participants, including teachers, psychologists, and social workers, were invited to assess proposed intervention elements across all socio-ecological domains. In the initial round, participants rated each component based on its general relevance, contextual applicability, and implementation feasibility using a structured rating scale. They were also prompted to provide qualitative comments to clarify reservations, propose clarifications, or suggest omissions. A descriptive analysis determined the level of agreement, with a pre-specified consensus threshold of more than 75 per cent required for component retention. Components failing to reach this consensus were revised in light of narrative feedback and redrafted for the second round, aiming to clarify areas of disagreement and enhance contextual relevance.

Across the 68 statements assessed during Round Two, 65 achieved the predefined consensus threshold (≥ 75% agreement and median ≥ 4) and were retained within the final framework. Agreement levels among retained statements ranged from 77,8% to 100%, indicating strong stakeholder support for the proposed prevention components. Three statements relating to stakeholder referral systems, peer-support mechanisms for survivors, and the incorporation of traditional values into prevention programmes did not initially achieve consensus and were subsequently refined based on participant feedback.

To enhance transparency in the consensus-building process, Table 3 presents a summary of selected Delphi findings illustrating the progression of prevention components from Round One thematic identification to Round Two consensus ratings. The table includes representative components that achieved strong consensus across the major socio-ecological domains, as well as components that required revision following participant feedback before inclusion in the final framework.

The consensus process revealed strong agreement across domains regarding the necessity for multi-level alignment rather than fragmented, singular programming. At the individual level, comprehensive body safety education combined with digital literacy education achieved high consensus as a fundamental prerequisite for prevention. At the microsystem level, structured communication and supervision training for caregivers were prioritised as a central mechanism of relational protection. Mesosystem-level components, including awareness-raising campaigns led by traditional and religious leaders, anti-stigma dialogues, and bystander activation strategies, were consistently endorsed as essential for norm transformation. At the exosystem and macrosystem levels, reinforced inter-sectoral coordination protocols, clearer referral channels, and institutionalised monitoring and evaluation systems were identified as crucial building blocks for sustainable impact. Significantly, the Delphi findings did not introduce new concepts but rather validated and refined protective processes identified in earlier stages. What emerged was not a disparate list of interventions, but a prioritised set of mutually reinforcing components spanning ecological systems. Participants repeatedly emphasised that prevention efforts would be ineffective if implemented in isolation, underscoring the imperative for deliberate alignment and integration across sectors. The Delphi process thus served as a bridge between empirical identification and architectural conceptualisation, transforming locally defined safeguard mechanisms into a structured, consensus-based, and operational framework. These prioritised components subsequently informed a joint development workshop where stakeholders transitioned from validating components to developing the system’s architecture, culminating in a dynamic model of the socio-ecological prevention framework.

Table 3 / Таблица 3

Summary of Delphi consensus results

Сводные результаты согласованности мнений экспертов по методу Дельфи

Note: Round One generated 13 thematic domains that informed the development of 68 structured statements for Round Two. To maintain clarity and readability, Table 3 presents a selection of representative prevention components that illustrate the consensus-building process across the major socio-ecological domains, including components that achieved immediate consensus and those that required revision. Of the 68 statements assessed in Round Two, 65 achieved the predefined consensus threshold (≥75% agreement and median ≥4) and were retained within the final framework. Three statements (M5, SS5, and CU2) did not initially achieve consensus and were revised based on participant feedback before inclusion in the final socio-ecological prevention framework. Detailed results for all 68 Delphi statements, including item codes, statement wording, agreement percentages, measures of central tendency, and final decisions, are provided in Appendix A.

Примечание: в ходе первого раунда было выделено 13 тематических областей, которые легли в основу разработки 68 структурированных утверждений для второго раунда. Для обеспечения наглядности и удобства восприятия в таблице 3 представлена выборка репрезентативных компонентов профилактики, иллюстрирующая процесс достижения консенсуса по основным социально-экологическим сферам; в нее включены как компоненты, по которым консенсус был достигнут сразу, так и те, которые потребовали доработки. Из 68 утверждений, оцененных во втором раунде, 65 достигли заранее установленного порога консенсуса (≥75% согласия и медиана ≥4) и были включены в итоговую структуру. Три утверждения (M5, SS5 и CU2) изначально не достигли консенсуса и были доработаны с учетом отзывов участников перед включением в окончательную социально-экологическую структуру профилактики. Подробные результаты по всем 68 утверждениям метода Дельфи, включая коды пунктов, формулировки утверждений, процент согласия, показатели центральной тенденции и итоговые решения, представлены в Приложении А.

Contextually grounded socio-ecological framework for CSA prevention: Prevention components and mechanisms across ecological levels in Zimbabwe

The sequential phases of the study, comprising the structured literature review, qualitative stakeholder interviews, Delphi consensus process, and co-creation workshop, culminated in the development of a contextually grounded socio-ecological framework for CSA prevention (Table 4). Rather than proposing a new theoretical model, the framework operationalises the socio-ecological model by translating evidence-based prevention components into coordinated actions across ecological levels. The framework is organised around five interconnected socio-ecological domains: the individual, microsystem, mesosystem, exosystem, and macrosystem. At the individual level, prevention focuses on strengthening children's body safety knowledge, digital literacy, self-protection skills, and help-seeking behaviours. The microsystem emphasises parenting programmes, family communication, caregiver supervision, and supportive teacher-child relationships. At the mesosystem level, prevention efforts focus on community awareness campaigns, engagement of religious and traditional leaders, community surveillance, and the transformation of harmful social norms. The exosystem encompasses institutional responses, including school safeguarding policies, referral pathways, social welfare services, and collaboration between child protection stakeholders. At the macrosystem level, prevention is supported through legislation, policy implementation, resource allocation, and accountability mechanisms that create an enabling environment for child protection.

Table 4 / Таблица 4

Socio-ecological model for CSA prevention in Zimbabwe

Социоэкологическая модель профилактики CSA в Зимбабве

The framework recognises that no single intervention can effectively prevent CSA in isolation. Rather, prevention requires coordinated action across ecological levels, with each level reinforcing the protective functions of the others. The arrows linking the framework levels illustrate the interdependence of prevention efforts and the need for vertical integration between child-focused interventions, family support systems, community mobilisation, institutional responses, and policy structures. Four cross-cutting processes underpin the framework: coordination, collaboration, reporting, and monitoring and evaluation. These processes facilitate communication and accountability among stakeholders while ensuring that prevention efforts remain responsive to emerging challenges and contextual realities. Through these mechanisms, the framework promotes continuity between prevention, detection, reporting, and response systems. The framework ultimately aims to achieve improved child protection outcomes through reduced CSA risk, increased reporting, stronger safeguarding systems, and enhanced community awareness. By integrating empirically identified prevention components into a coherent socio-ecological structure, the framework provides a contextually grounded prevention framework that adapts the socio-ecological model to the Zimbabwean context and offers practical guidance for coordinated child sexual abuse prevention.

Discussion

The findings therefore contribute primarily to the practical application of the socio-ecological model by demonstrating how prevention activities can be coordinated across ecological levels within a specific LMIC context. Consistent with Bronfenbrenner’s ecological systems theory, the findings show that vulnerability and protection are not produced by isolated factors but emerge through the interaction of multiple social, institutional, and structural conditions. This reinforces growing evidence that effective CSA prevention requires coordinated interventions that address risks and protective factors simultaneously across ecological levels rather than relying on isolated programmes (Martinello, 2020).

At the individual level, participants identified limited knowledge of body safety, disability-related vulnerabilities, and unsafe digital exposure as important risk factors. These findings are consistent with previous studies that have highlighted how developmental dependence, limited awareness of abuse, and restricted disclosure capacities increase children’s vulnerability to victimisation (Assink et al., 2019; Quadara, El-Murr, Latham, 2017). The emergence of digital technologies as a risk factor is particularly significant. Participants described how increased access to mobile phones, social media platforms, and online content has created new pathways through which children may be exposed to exploitation and grooming. Similar concerns have been raised in international literature, which suggests that online environments are increasingly intersecting with traditional forms of CSA and therefore require prevention approaches that incorporate digital safety and digital literacy education (Jones et al., 2020). The findings therefore highlight the importance of equipping children not only with knowledge about body safety and consent but also with the skills needed to navigate digital environments safely.

The study further demonstrated that family relationships and caregiving arrangements play a central role in shaping both vulnerability and protection. Participants repeatedly described economic migration, disrupted caregiving structures, and family silence surrounding sexuality as factors that increase risk. These findings support previous research showing that family instability, limited parental supervision, and weak communication structures contribute to children’s vulnerability to sexual abuse (Rudolph et al., 2017; Quadara, El-Murr, Latham, 2017). At the same time, stakeholders identified open communication, relational trust, and supportive caregiving as critical protective factors. This suggests that families remain one of the most important sites for prevention and that strengthening parenting capacities should form a central component of child protection strategies. Consistent with socio-ecological theory, family-level interventions are likely to be most effective when reinforced by broader community and institutional support systems.

At the community level, participants highlighted the role of cultural silence, victim-blaming attitudes, and patriarchal norms in perpetuating abuse and discouraging disclosure. These findings align with research demonstrating that community norms and social attitudes shape how abuse is understood, reported, and responded to (Adam, Ahmad, Kassim, 2024; Selengia, Thuy, Mushi, 2020). The persistence of stigma surrounding sexuality and abuse creates environments in which perpetrators remain protected while victims are discouraged from seeking assistance. However, participants also identified opportunities for transformation through community dialogue, faith-based engagement, and the involvement of traditional leaders. These findings suggest that prevention efforts should not focus solely on individual behaviour change but should also seek to transform the broader social norms that sustain abuse. Community mobilisation and collective accountability therefore emerge as important mechanisms for strengthening prevention efforts in resource-constrained settings.

One of the most significant findings of this study concerns the role of institutional systems in either strengthening or weakening child protection. Participants consistently identified weak referral pathways, fragmented service delivery, poor coordination among institutions, and inconsistent law enforcement as major barriers to effective prevention. These findings support previous research which has argued that fragmented child protection systems undermine prevention efforts by creating gaps in reporting, investigation, referral, and service provision (Fortson et al., 2016). Stakeholders emphasised that schools, social welfare services, health institutions, law enforcement agencies, and judicial systems often operate independently, reducing the effectiveness of prevention and response mechanisms. The findings therefore highlight the importance of inter-sectoral collaboration and coordinated service delivery as foundational elements of effective child protection systems.

The Delphi process provided an important mechanism for translating these findings into practical prevention priorities. Rather than introducing entirely new concepts, the consensus-building process validated and refined prevention components identified through the literature review and stakeholder interviews. High levels of agreement emerged around child empowerment, parenting support, community awareness initiatives, institutional coordination, reporting mechanisms, and monitoring systems. Importantly, participants consistently emphasised that these interventions should not be implemented independently but as part of an integrated prevention strategy. This finding reflects a growing recognition within prevention literature that sustainable child protection depends on the coordination of interventions across multiple ecological levels rather than the implementation of isolated programmes (McKibbin et al., 2020).

A major contribution of this study is the development of a socio-ecological prevention framework that translates ecological theory into a practical architecture for prevention. While socio-ecological models have frequently been used to categorise risk and protective factors, they have often remained descriptive rather than operational. The framework developed in this study extends the practical application of socio-ecological approaches by identifying how coordinated interventions can be organised across ecological levels and connected through cross-cutting processes of collaboration, coordination, reporting, monitoring, and accountability. In this respect, the study contributes to the growing body of literature advocating for systems-based approaches to violence prevention and child protection.

The findings have important implications for policy and practice in Zimbabwe and other low- and middle-income countries. First, prevention strategies should prioritise coordinated multi-sectoral approaches involving education, health, social welfare, justice, and community structures. Second, child protection programmes should strengthen family communication, caregiver supervision, and parenting support as key preventive mechanisms. Third, prevention initiatives should engage religious leaders, traditional leaders, and community-based organisations in efforts to challenge harmful norms and reduce stigma associated with disclosure. Finally, the growing role of digital technologies requires child protection systems to integrate digital literacy, online safety education, and technology-responsive safeguarding measures into existing prevention frameworks.

Overall, the findings demonstrate that effective CSA prevention requires more than isolated interventions targeting individual behaviour. Sustainable prevention depends on strengthening protective mechanisms across families, communities, institutions, and policy environments while ensuring coordination among these interconnected systems. The socio-ecological prevention framework developed in this study provides a practical and contextually relevant foundation for achieving this goal within Zimbabwe and similar resource-constrained settings.

Conclusions

This study provides a contextually grounded understanding of CSA prevention in Gweru District, Zimbabwe, by integrating global evidence with locally derived stakeholder perspectives. The findings demonstrate that CSA is shaped by interacting factors across ecological levels and that effective prevention requires coordinated, multi-level responses rather than isolated interventions. The study extends the practical application of the socio-ecological model by translating socio-ecological principles into a coordinated and contextually grounded prevention framework. Rather than proposing a new theoretical model, the framework operationalises the socio-ecological model through integrated prevention strategies across the individual, microsystem, mesosystem, exosystem, and macrosystem levels. The findings highlight that strengthening alignment across ecological systems is critical for sustainable prevention and effective child protection. The framework adapts the socio-ecological model to the Zimbabwean context by incorporating locally identified priorities, stakeholder perspectives, and cross-sectoral coordination mechanisms. As such, it offers an applied socio-ecological prevention architecture that can guide prevention planning, implementation, and policy development in Zimbabwe and other resource-constrained settings. The study contributes to ongoing efforts to design culturally relevant and context-sensitive prevention strategies in African settings, particularly within education and child protection systems. It underscores the importance of integrated approaches that combine child empowerment, family strengthening, community norm transformation, institutional coordination, and accountability mechanisms to strengthen child protection and reduce the risk of child sexual abuse.

Limitations. Several limitations should be considered when interpreting the findings of this study. First, the study was conducted within a single district (Gweru), which may limit the transferability of findings to other contexts with different socio-cultural or institutional dynamics. Second, the study relied on a relatively small, purposively selected sample of professionals, including teachers, psychologists, and social workers. While this approach enabled in-depth, context-specific insights, it may not fully capture the perspectives of other relevant stakeholders, such as caregivers, community members, or children themselves. Third, the study focused on qualitative data and expert consensus, which, while valuable for understanding complex social processes, does not allow for statistical generalisation. Finally, the study did not include a longitudinal assessment of intervention outcomes. Future research could build on these findings by testing and evaluating the implementation and effectiveness of the proposed prevention framework over time and across diverse settings. Despite these limitations, the study provides valuable insights into the systemic nature of CSA and offers a practical foundation for developing coordinated prevention strategies.

Ограничения. При интерпретации результатов настоящего исследования следует учитывать ряд методологических ограничений. Во-первых, исследование было проведено в пределах одного района (Гверу), что может ограничивать применимость полученных результатов к другим контекстам, характеризующимся иными социокультурными или институциональными условиями. Во-вторых, в исследовании использована относительно небольшая выборка специалистов, отобранная методом целевой выборки, в которую вошли педагоги, психологи и социальные работники. Несмотря на то, что данный подход позволил получить глубокие, учитывающие локальный контекст данные, он может не в полной мере отражать позиции иных заинтересованных сторон, таких как лица, осуществляющие уход за детьми, представители местных сообществ или сами дети. В-третьих, исследование опиралось на качественные данные и экспертный консенсус, что, будучи ценным для понимания сложных социальных процессов, не позволяет осуществлять статистическую генерализацию результатов. Наконец, в исследовании не проводилась лонгитюдная оценка результатов внедрения предлагаемых мер. Перспективные направления дальнейших исследований могут включать апробацию и оценку реализации и эффективности разработанных основ профилактики в динамике и в различных социально-культурных контекстах. Несмотря на перечисленные ограничения, настоящее исследование вносит ценный вклад в понимание системной природы сексуального насилия в отношении детей и предлагает практическую основу для разработки скоординированных стратегий профилактики.

1. Adam, A., Ahmad, Y., Kassim, A. (2024). An uphill battle: A systematic literature review on the challenges of addressing child sexual abuse (CSA) in the developing countries. Environment-Behaviour Proceedings Journal, 9(SI22), 237—243. https://doi.org/10.21834/e-bpj.v9iSI22.5852
2. Assink, M., van der Put, C.E., Meeuwsen, M.W.C.M., de Jong, N.M., Oort, F.J., Stams, G.J.J.M., Hoeve, M. (2019). Risk factors for child sexual abuse victimisation: A meta-analytic review. Psychological Bulletin, 145(5), 459—489. https://doi.org/10.1037/bul0000188
3. Bows, H. (2017). Researching sexual violence against older people: Reflecting on the use of freedom of information requests in a feminist study. Feminist Review, 115(1), 30—45. https://doi.org/10.1057/s41305-017-0029-z
4. Bryce, I. (2020). Child sexual abuse: Complexities and contexts. In: I. Bryce, W. Petherick (Eds.), Child sexual abuse: Forensic issues in evidence, impact, and management (pp. 3—26). Academic Press. https://doi.org/10.1016/B978-0-12-819434-8.00001-5
5. Dallelucci, C.C., Bragiato, E.C., Areco, K.C.N., Fidalgo, T.M., da Silveira, D.X. (2019). Sexually risky behaviour, cocaine and alcohol use among substance users in an outpatient facility: a cross-sectional study. Substance Abuse Treatment, Prevention, and Policy, 14(1), Article 46. https://doi.org/10.1186/s13011-019-0238-x
6. Danaeifar, M., Arshi, M., Ostadhashemi, L., Moghanibashi-Mansourieh, A. (2022). Facilitators and barriers to child sexual abuse interventions: A qualitative study of interventions in Iran. Iranian Journal of Psychiatry and Behavioural Sciences, 16(4), Article e129326. https://doi.org/10.5812/ijpbs-129326
7. Finkelhor, D. (2009). The prevention of childhood sexual abuse. The Future of Children, 19(2), 169—194. https://doi.org/10.1353/foc.0.0035
8. Finkelhor, D., Turner, H.A., Shattuck, A., Hamby, S.L. (2013). Violence, crime, and abuse exposure in a national sample of children and youth. JAMA Pediatrics, 167(7), 614—621. https://doi.org/10.1001/jamapediatrics.2013.42
9. Fortson, B.L., Klevens, J., Merrick, M.T., Gilbert, L.K., Alexander, S.P. (2016). Preventing child abuse and neglect: A technical package for policy, norm, and programmatic activities. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. https://doi.org/10.15620/cdc.38864
10. Jones, C., Scholes, L., Rolfe, B., Stieler-Hunt, C. (2020). A serious game for child sexual abuse prevention: An evaluation of Orbit. Child Abuse & Neglect, 107, Article 104569. https://doi.org/10.1016/j.chiabu.2020.104569
11. Laajasalo, T., Ellonen, N., Horselenberg, R., Izura, C., Wager, N. (2022). Editorial: Child sexual abuse: Empirical research on understanding and helping victims and offenders. Frontiers in Psychology, 13, Article https://doi.org/10.3389/fpsyg.2022.844639
12. Martinello, E. (2020). Applying the ecological systems theory to better understand and prevent child sexual abuse. Sexuality & Culture, 24(1), 326—344. https://doi.org/10.1007/s12119-019-09629-z
13. McKibbin, G., Humphreys, C., Hamilton, B. (2020). Prevention of child sexual abuse: A review of contemporary approaches and evidence. Child Abuse & Neglect, 105, Article 104422. https://doi.org/10.1016/j.chiabu.2020.104422
14. Mensah, F., Abdullah, A., Cudjoe, (2024). Unpacking norms shaping disclosure of intrafamilial child sexual abuse. Discover Psychology, 4(1), Article 170. https://doi.org/10.1007/s44202-024-00286-4
15. Morris, M.C., Kouros, C.D., Janecek, K., Freeman, R., Mielock, A., Garber, J. (2017). Community-level moderators of a school-based childhood sexual assault prevention program. Child Abuse & Neglect, 63, 295—306. https://doi.org/10.1016/j.chiabu.2016.10.005
16. Quadara, A., El-Murr, A., Latham, J. (2017). The effects of pornography on children and young people: An evidence scan. Melbourne: Australian Institute of Family Studies. https://doi.org/10.13140/RG.2.2.16152.11522
17. Rudolph, J., Zimmer-Gembeck, M.J., Shanley, D.C., Hawkins, R. (2017). Child sexual abuse prevention opportunities: Parenting, programs, and the reduction of risk. Child Maltreatment, 23(1), 96—106. https://doi.org/10.1177/1077559517729479
18. Selengia, V., Thuy, H.N.T., Mushi, D. (2020). Prevalence and patterns of child sexual abuse in selected countries of Asia and Africa: A review of literature. Open Journal of Social Sciences, 8(9), 146—160. https://doi.org/10.4236/jss.2020.89010
19. Whitaker, D.J., Le, B., Hanson, R.K., Baker, C.K., McMahon, P.M., Ryan, G., Klein, A., Donovan Rice, D. (2008). Risk factors for the perpetration of child sexual abuse: A review and meta-analysis. Child Abuse & Neglect, 32(5), 529—548. https://doi.org/10.1016/j.chiabu.2007.08.005
20. World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects (2013). JAMA, 310(20), 2191—2194. https://doi.org/10.1001/jama.2013.281053