Background Violence against women and children (VAWC) are global issues that particularly affect South Africa. Conducting research on VAWC requires skilled and compassionate interviewers who can collect accurate and valuable data from multiple family members while being sensitive to participants' stories and potential distress. However, there is a lack of knowledge on how to train fieldworkers for such research. Objectives To address gaps in knowledge on appropriate training for fieldworkers, this paper presents key reflections and recommendations on challenges and opportunities for high-quality training and data collection from a research study on intergenerational violence transmission. Methods The study included three focus group discussions with six fieldworkers and debriefing sessions with seven study facilitators. The fieldworkers were aged 28-48 and resided in Mpumalanga, South Africa. The collected data (transcriptions from the audio recorded FDGs) was analyzed using a rapid analytical approach. Results The findings showed that continuous, experiential training is crucial in enhancing fieldworkers' skills and shifting attitudes, especially regarding gender norms and violence. Conclusion Training on sensitive topics should be tailored to address the specific demands of the study, considering the complexity of the subject matter and the interviewers' previous experience. Promoting a culture of self-care among the research teams is essential. This can be achieved by implementing techniques like meditation, frequent debriefings, and encouraging open communication to enhance the resilience and general well-being of fieldworkers.

BACKGROUND:Adverse Childhood Experiences (ACEs) are associated with poor adult mental health, particularly in low-and middle-income countries where ACE exposure is high. AIMS:This study used latent class analysis (LCA) to characterise patterns of ACEs in a longitudinal sample of South African adolescents and explored associations between ACE classes and adult mental health, including anxiety, depression, post-traumatic stress disorder, psychosis, and suicidality. METHODS:Fourteen ACEs were assessed in 1648 adolescents (54.92% girls) from Mpumalanga using data collected across three waves (2010/2011, 2011/2012, and 2022/2023). Each ACE was dichotomised (exposed vs. not exposed), with exposure defined as meeting threshold criteria at either childhood wave. RESULTS:A four-class solution was identified: Low ACEs, High Parental AIDS-Affectedness and High Parental Sickness, High Parental AIDS-Affectedness and High Parental Death, and High multi-type ACEs. Compared to the low ACEs class, other classes had significantly higher poverty levels with further differences in gender, age and location. Adults in the highest multi-ACEs class reported significantly worse symptoms across all five mental health outcomes. Contextually sensitive ACEs (such as parental AIDS-affectedness) played an important role in differentiating ACE classes. Adults who experience multiple ACEs may be particularly at risk for poor mental health. In the two classes with, on average, the worst mental health outcomes, classes were primarily characterised by expanded ACEs, suggesting these broader adversities may confer unique risks in low-resource settings. CONCLUSIONS:Findings highlight the value of person-centred approaches like LCA to understand the complexity and multiplicative effects of ACEs and their long-term mental health impacts.

Background Research on violence against children in Africa often relies on adult's qualitative accounts or on quantitative surveys that focus mainly on sexual abuse or on adverse childhood exposures. We seldom hear how young children describe their experiences and their perceptions of violence. Objective This study aimed to describe children's personal experiences and perceptions of violence. Participants and setting The study was conducted from 2022 to 2023 in peri-urban and rural Mpumalanga, South Africa with 23 children aged between 6 and 12 years who, together with their primary caregivers, were participants in a larger longitudinal cohort study. Methods We used in-depth interviews alongside arts-based methods for the children. Interviews underwent reflexive thematic analysis. Results Children were highly exposed to physical and emotional violence both in and outside of the home. Physical violence was used as a discipline method by teachers, parents, or children's caregivers. Severe occurrences involved adults using objects to discipline children, including steel pipes, shoes, or classroom materials. Most children associated violence with fear and pain. However, they described physical abuse as a morally “right” act if they understood it to be a form of discipline for their wrong behaviours. Attitudes condoning violence also enabled peer-to-peer violence. Children were not confident in reporting cases of violence inflicted by adults due to the fear of sanctions and the belief that violence from adults was “accepted”. Conclusion In this sample, children had limited spaces that they marked as free from violence. Violence was normalised by the adults inflicting it, creating a state of cognitive dissonance among children, leading to victim blaming and, in turn, inflicting violence against same-age peers.

We describe the validation, cultural adaptation, and updating of The International Society for the Prevention of Child Abuse and Neglect, Child Abuse Screening Tool, Retrospective version (ICAST-R) for use in a Danish context. Thorough Danish questionnaires on child abuse and neglect exist, but some are proprietary, and their development precedes widespread use of social media. As forms of abuse evolve, including online sharing of abuse material, updated tools are needed, and we are not aware of any recent, comprehensive and validated questionnaires that are culturally adapted for Danish. The free ICAST-R measure has been translated into at least 20 languages and is widely used in the field. To develop a Danish version, the questionnaire has been back-and-forth-translated and has undergone cognitive testing and focus group interviews by eight individuals within the target age range and seven experts by experience with abuse and neglect. Further adaptation was done through focus group interviews with professionals from research and practice. Categorisation of severity of abuse was done through a Delphi process, and acceptability of the adapted measure was assessed in an online survey distributed to 450 Danish adults. The ICAST-R has also been reviewed in focus group interviews with 11 professionals working with child abuse and neglect, representing practitioners, researchers and non-governmental organisations. The final Danish measure incorporates additional questions focused on online sexual abuse, free-text options for other experiences of abuse, and items describing acts of abuse with subcultural attributions of meaning. The neglect section is substantially altered, and the section on sexual child abuse has been restructured into non-contact sexual abuse, non-contact technology-facilitated sexual abuse, and moderate and severe contact sexual abuse. For sexual abuse, the structure was changed so that sexual experiences were recorded and subsequent questions explored consent and age at the event. For all events, a number of questions exploring disclosure was added.

Objectives: To address evidence gaps on intergenerational parenting in developing countries, this study aims to identify prospective relationships between parenting experiences in childhood (2010-2011) and parenting practices used in adulthood among young female adults in Mpumalanga province, South Africa (2022-2024). Design: Secondary analysis of a randomly sampled prospective longitudinal cohort study in a community setting. Setting: The research was carried out in one randomly selected rural and peri-urban site within one health district in Mpumalanga province, South Africa (Wave 1 2010, Wave 2 2011, Wave 3 2022-2024). Participants: A subsample of 549 female adolescents aged 10-17 enrolled in an observational cohort study on violence in one health district in Mpumalanga province, South Africa, and caring for a child aged 0-17 in young adulthood. Fieldworker-administrated interviews were conducted at three waves (2010, 2011, 2022-2024). Demographics were analyzed using Stata SE 18. Confirmatory factor analysis of parenting domains and structural equation modelling of prospective longitudinal pathways was completed using MPlus 7.11. Outcome measures: The primary outcome was use of parenting practices in adulthood. Pathway outcomes included adolescent anxiety, depression, post-traumatic stress, and suicidality symptoms, lifetime and past-year intimate partner violence experience, past-month food insecurity, and recent parental stress. Results: Eight domains of parenting practices were confirmed. Childhood experience of praise was associated with more inconsistent discipline in adulthood (β 0.13, p 0.028). Parental monitoring (β 0.13, p 0.024) and childhood physical abuse exposure (β 0.12, p 0.036) were associated with rigid parenting in adulthood. Childhood psychological abuse exposure was indirectly associated with use of violent parenting via recent IPV experience, recent food insecurity and parenting stress. Conclusion: Investment in parenting interventions with play-based activities, mental health services, social support, and economic empowerment may interrupt intergenerational transmission of negative parenting and promote transmission of positive parenting.

BACKGROUND:Violence against children is a violation of children's rights and a threat to their health and development, but young children are not often asked about violence they experience. OBJECTIVE:This study investigates the prevalence of violence reported by children in young childhood through adolescence, including how prevalence differs by age and gender, how children conceptualize violence, and associations with mental health. PARTICIPANTS AND SETTING:This study sampled children (n = 280; age 6-17; mean age: 8.8) from low-resource communities in South Africa who comprised a cross-sectional child sample within a longitudinal, multi-generational study. METHODS:Children completed an interviewer-led child-friendly questionnaire that asked about their experiences of and views on violence, their mental health symptoms, and their life broadly. Analyses assessed prevalence across age and gender, plus associations between mental health symptoms and violence via multivariate regression. RESULTS:Most children experienced emotional (56%) and physical abuse (55%) from a caregiver and repeated peer (53%) and sibling violence (21%), which was more prevalent for girls. Community violence was common (27%), and most children showed a moderate endorsement of violence norms. Younger children were more likely to experience physical abuse and higher peer violence scores, while adolescents were more likely to report suicidality. Regression models found peer, community, and domestic violence to be most consistently associated with poor mental health. CONCLUSIONS:Preventing and responding to violence against children requires efforts to undo normalized acceptance of the issue and holistic efforts to address it across multiple contexts, such as children's homes, schools, and communities.

Abstract Aims Prevention of child maltreatment – incorporating physical abuse, sexual abuse, emotional abuse, neglect and exposure to domestic violence – is a clearly defined global policy priority. Global Burden of Disease studies have focused on estimating burden attributable to childhood sexual abuse omitting other forms of child maltreatment. This study aims to estimate burden attributable to child maltreatment using data from the first comprehensive national study, the Australian Child Maltreatment Study (ACMS), accounting for the co-occurrence of multiple forms, the complex impact of multi-type maltreatment and the contribution of interrelated factors. Methods We estimated burden attributable to child maltreatment by age and gender for Australia in 2021. Risk–outcome pairs that met criteria for sufficient evidence for a causal relationship were included. Relative risks were estimated as a function of exposure based on data from the ACMS incorporating increased risk with multi-type maltreatment and adjustment for confounding. Levels of exposure in each of the 32 mutually exclusive combinations or patterns of child maltreatment were estimated based on ACMS data by age and gender. The theoretical minimum risk exposure level was determined as no exposure to child maltreatment in the population and population attributable fractions (PAFs) were calculated. Attributable mortality, years of life lost, years lived with disability and disability-adjusted life years (DALYs) were estimated by multiplying PAFs by the relevant burden of disease estimates by age and gender for Australia in 2021. Sensitivity analyses were conducted to assess the robustness of the results. Uncertainty was propagated into attributable burden estimates using Monte Carlo simulation methods. Results Overall, child maltreatment accounted for 6.6% (95% uncertainty interval (UI), 6.2–6.9%) of all DALYs for women and 6.4% (95% UI, 6.0–6.7%) of all DALYs for men in Australia in 2021. An estimated 71.2% of self-harm, 57.1% of anxiety disorders and 49.3% of major depressive disorder (MDD) DALYs in women, and 63.8% of self-harm, 55.9% of anxiety disorders and 42.9% of MDD DALYs in men were attributable to child maltreatment. Conclusions Child maltreatment contributes to a substantial proportion of burden of disease in Australia, equivalent to leading lifestyle-related risk factors such as high body mass index, high blood pressure and smoking. This research significantly advances knowledge of the disease burden attributable to child maltreatment and provides novel methodology for measuring the impact of all five forms of child maltreatment combined on mental health and health risk behaviours nationally and globally.

Objective: To map the extent, nature, and definitions of the literature on AI-facilitated child sexual exploitation and abuse (CSEA) perpetration, including (i) threats associated with AI and emerging technologies used to facilitate CSEA, (ii) how these threats are defined and operationalised, and (iii) interventions, policies, safeguarding practices, and regulatory responses relevant to the UK context. Introduction: AI systems are increasingly embedded in children and young people’s digital lives, and capabilities such as high-fidelity content generation and scalable automated interaction may enable new or intensified forms of CSEA. However, the evidence base is conceptually and terminologically inconsistent, with overlapping terms and variable operational definitions that limit comparability and surveillance. Inclusion criteria: This review will consider literature relating to AI-facilitated CSEA perpetration. Sources addressing definitions, typologies, and UK-relevant responses will be included. Context will include online, hybrid, and technology-mediated environments, with specific extraction of UK-relevant interventions and regulatory and safeguarding responses. Methods: The key information sources to be searched include Scopus, PsycINFO, CINAHL, IEEE Xplore, and Web of Science, supplemented by systematic grey literature searches of UK government and international NGO databases for English-language records from 2010 to the present. Following independent dual screening of titles and abstracts, data will be extracted using a standardised charting form.

BACKGROUND:Childhood experiences of domestic violence (EDV) are associated with short- and long-term adverse mental and physical health outcomes for victim-survivors; however, little is known about the health service utilisation among individuals with childhood EDV. OBJECTIVE:Examine associations between different forms of childhood EDV and health service utilisation. PARTICIPANTS AND SETTING:Data were drawn from the the Australian Child Maltreatment Study (n = 8503). METHODS:Associations between four forms of childhood EDV (physical violence, threats of harm, damage to property or pets, and intimidation or control) and past 12-month health service utilisation (hospitalisation, and consultation with seven health professionals) were analysed using survey-weighted logistic regression models. Estimates were calculated independently for each form of childhood EDV. Models were adjusted for each other form of EDV, four child maltreatment types, and socio-economic factors. Models were stratified by gender. RESULTS:Utilisation of mental health services (psychologists, psychiatrists and mental health nurses) were significantly more common among both women and men with childhood EDV compared to those without. The strongest associations were between any domestic violence and a primary care physician consultation (aOR = 1.49, 95% CI 1.25-1.78), and intimidation or control and a mental health professional consultation (aOR = 1.44, 95% CI 1.22-1.69), after full adjustment. Associations were typically stronger for men in the gender-stratified analyses. CONCLUSIONS:Childhood EDV significantly increases the likelihood of health service utilisation across the lifespan. Greater investment in domestic violence education and training for health professionals is needed.

Violence against children (VAC) is a global health issue. Research that includes child participants is essential for their perspectives to be fully represented in policy and interventions that affect them. There is a gap in our understanding of how VAC researchers have engaged with young children as participants. This scoping review maps the study design, research methods, and ethical practices undertaken in VAC research with children aged 2 to 12. Eleven databases were searched, in addition to gray literature and recommendations from the EndViolence Lab. Documents were assessed against inclusion criteria, including a focus on data collection with children aged 2 to 12 on experiences of violence. Both empirical studies and guidance articles and reports were included. Data extraction from 21 empirical studies and themes from 14 guidance articles were cross-referenced to further map current practices. Engaging young children in VAC research was shown to be possible by creating a safe and comfortable research environment, using child-specific measures and engagement techniques, managing adult gatekeepers and participants, training interviewers, gaining consent from children, and establishing referral protocols. By using the study designs, methods and ethical protocols compiled in this scoping review, future researchers can continue to engage young children in VAC research and expand our knowledge of children's experiences of violence in the movement to end VAC.

Young children should be included in research that investigates their life experiences, as this honors their rights to inclusion and allows their voices to be heard. This paper summarizes key themes, takeaways, and recommendations gathered from a multi-year research study regarding how to include young children in violence against children research, where feedback on the scope, methods, and consequences was sought from child participants and study staff. Children (age 4–17 years) in South Africa were included as the youngest participant group in a multi-generational study that also included their parents and grandparents, though data here focus on children aged 4–10. Rigorous policies and safeguards were employed to prioritize the ethical inclusion of child participants. Following completion of a child-friendly questionnaire, qualitative data were collected from a sample of participating children, as well as study staff (fieldworkers; social worker), wherein they reported their views on including young children in violence research. Following thematic coding of this qualitative data, the study management team engaged in multiple reflection sessions to discuss the fieldwork challenges, solutions, best practice takeaways, training implications, and beyond with regards to conducting VAC research with young children. Key themes and takeaways included: the feasibility of young children’s participation and how they understood and responded to participation, the interviewer qualities that allowed for meaningful participation from young children, the role of play, arts, and fun in the interview space, and post-participation considerations and actions. The takeaways and recommendations pointed to how young children can be safely included in violence-focused interviews and research, while noting that careful attention must be paid to how they are invited to and engaged in the research, as well as the attentiveness and sensitivity of the interviewers and research team.

Artificial intelligence (AI) is increasingly implicated in child sexual exploitation and abuse (CSEA), but inconsistent terminology has limited synthesis. This scoping review mapped reported AI-CSEA threats and enabling technologies and examined how AI systems participated in perpetration. English-language sources published from January 2010 to 1 April 2026 were eligible if they concerned children under 18 and where an AI system played a material role in the abuse process. Searches of the ACM Digital Library, CINAHL, Web of Science, EBSCOhost databases, and grey literature identified 48 sources. Three reviewers contributed to screening and data charting, and extracted data were summarised numerically and through qualitative content analysis. Overall, our review found that AI-CSEA nomenclature conflated abusive conduct, production methods and system roles. Five non-exclusive threat domains comprised i) AI-generated CSAM, ii) AI-manipulated CSAM, iii) grooming, iv) sextortion, and v) conversational AI sexual abuse. Six technology groups encompassed: i) generative systems; ii) model adaptation and training data; iii) media manipulation systems; iv) conversational systems; v) immersive and embodied technologies; and vi) access, hosting, and distribution infrastructure. Forms of AI involvement include: i) AI-enabled assistance with person-directed abuse; ii) AI-simulated co-production of user-initiated scenarios; and iii) AI-driven initiation or escalation of sexualised conduct. AI-CSEA largely reflected established forms of abuse reshaped by new modes of production and interaction. Existing act-based classifications of violence against children provide a suitable foundation but require supplementary recording of production route, child linkage, system role and point of steering.

Research assessing children's experiences of violence with child participants is essential to understand the burden of violence against children, to identify children most at risk and, consequently, enhance prevention. Yet many ethical review boards hesitate to allow violence research with children, and most will require active parental consent for child participation. However, this gatekeeping often results in refusal of parental consent and in children's voices remaining unheard, which is particularly risky in violence research where parents are most commonly inflicting specific types of violence or protecting perpetrators. When to demand parental consent should be carefully informed by scientific evidence, legal requirements and ethical principles. We provide (1) an overview of issues surrounding parental consent including child capacity and present a new analysis of country data with differing consent types and their implications on participation, disclosure and non-response, (2) summarise ethical issues in relation to consent, and (3) discuss opportunities for consent regulations in violence against children research. Our analysis demonstrates that type of consent affects rates of participation and violence disclosure and that fully-informed consent is more than a one-time agreement to participate. Considering the ethical and legal obligation of ensuring children's safety in violence research, and children's ability to understand the consequences of participation, children aged 13 years and older should be able to make informed decisions about participation independent of their legal caregivers providing necessary ethical and safeguarding requirements are met; alternatively, passive parental consent should be considered.

BACKGROUND:Recent efforts to quantify any distinct impact of out-of-home care (OOHC) experience have encountered challenges in separating the effects of OOHC with the earlier maltreatment experiences given their strong association. This study investigates whether an OOHC experience uniquely affects the odds of meeting clinical threshold for four common mental disorders when matched with a non-care-experienced group on the intensity of their experience of child maltreatment. METHODS:Drawing on data from the Australian Child Maltreatment Study (ACMS), the present study aimed to minimize selection bias by using propensity score matching. The care-experienced (intervention; n = 395) and non-care-experienced (comparison; n = 395) groups were matched on key maltreatment characteristics including age of onset, duration, frequency, perpetrator type, type and range of maltreatment. Participants were aged between 16 and 65 + . RESULTS:Logistic regression showed that when adjusted for child maltreatment intensity, there was no significant effect of OOHC on any mental disorders. The odds ratio (OR) for experiencing each of the four mental health disorders explored were non-significant (OR = 0.8; 95 % CI = 0.4-1.7) indicating that care-experienced individuals were no more likely to experience any mental health disorder than non-care-experienced people once they were matched on the intensity of their childhood maltreatment experience. CONCLUSIONS:These results suggest that the key risk factor for mental disorders is the prior experience of child maltreatment and the intensity and characteristics of child maltreatment experiences in care-experienced populations. The findings highlight the importance of assessing child maltreatment intensity as a risk factor for poor mental health in care populations, rather than attributing these risks solely to the experience of being in OOHC.

BACKGROUND:Preventing violence against children begins with current and accurate data on its scope and nature. Such information on violence against children in the United Kingdom (UK) was most recently collected nationally in 2011 and requires updating. A national research organisation in the UK has carried out a feasibility study on whether and how a national prevalence survey on violence in childhood can be conducted. OBJECTIVE:Phase two of this study includes the development of a questionnaire to collect information on childhood experiences of violence from participants aged 11-25. The aim of this paper is to report on the process of creating the questionnaire. PARTICIPANTS AND SETTING:Data collection was conducted with both UK-based and international child abuse and violence researchers and professionals, adults and children with lived experience of violence in the UK, and children and young people in the target age group in the UK. METHODS:The multistage development and validation process was extensive and thorough. It included a Delphi survey, focus groups, cognitive interviews, and individual consultations. Thematic analysis was conducted to identify feedback by violence type. RESULTS:The final questionnaire includes the following topics: physical abuse, corporal punishment, emotional abuse, neglect, witnessing domestic violence, sexual abuse, peer and sibling violence including bullying, criminal exploitation, community violence, and dating violence as well as factors associated with violence (risk factors, protective factors, outcomes). CONCLUSIONS:Relevance, comprehensibility, and comprehensiveness of items was established, and the questionnaire shows good content validity for use with the target age range in the UK.

To protect children against violence, we need robust data from every child population. In spite of this, not all child populations are well represented in research. This hampers the usability of results to those left out, and necessitate a discussion of potential solutions. Data collection efforts are hampered by a lack of validated instruments for a number of populations, including but not limited to young children, children with disabilities, and children on the move, including refugees and immigrants. This results in the exclusion from research of many children who are disproportionately affected by violence, but unable to exercise their right to be heard on this matter. In this discussion article, we draw on the co-authors’ research experience and relevant literature to present issues specific to measurement of violence, and recommendations on how to get precise estimates in these underrepresented groups. These include puppet-based interviews for young children, alternative response formats such as validated sign language instruments for children with disabilities, and a range of violence measures which reflect the diverse violence experienced by children on the move. However, the literature also reveals an overall lack of validation and participatory approaches, affecting both research measuring violence against children, and development of instruments for this purpose. This article outlines recommendations and examples of best practice for including these children in research, and for the development and validation of instruments suitable for capturing their self-report of violence. It is our hope that this discussion article, and the potential solutions presented, can inspire further studies on violence on how to include all child populations in research.

By 2050, one in four people in the world will live in Africa. The region's adolescents are our present and future but need effective services to support their holistic health and wellbeing, advance their education, allow them to flourish, and protect them from adversity. This paper reviews evidence for a new concept of 'development accelerators': interventions that can help achieve multiple Sustainable Development Goals simultaneously. Across 26 studies, the most frequently identified accelerators were government-provided social protection, positive parenting including good supervision of their adolescents, adolescent-friendly healthcare and childcare, positive gender norms, and psychosocial support. These have been shown to be highly cost-effective, giving substantial returns on investment, especially when applied in combination.

OBJECTIVES:To estimate the prevalence in Australia of intimate partner violence, each intimate partner violence type, and multitype intimate partner violence, overall and by gender, age group, and sexual orientation. STUDY DESIGN:National survey; Composite Abuse Scale (Revised)-Short Form administered in mobile telephone interviews, as a component of the Australian Child Maltreatment Study. SETTING:Australia, 9 April - 11 October 2021. PARTICIPANTS:8503 people aged 16 years or older: 3500 aged 16-24 years and about 1000 each aged 25-34, 35-44, 45-54, 55-64, or 65 years or older. MAIN OUTCOME MEASURES:Proportions of participants who had ever been in an intimate partner relationship since the age of 16 years (overall, and by gender, age group, and sexual orientation) who reported ever experiencing intimate partner physical, sexual, or psychological violence. RESULTS:Survey data were available for 8503 eligible participants (14% of eligible persons contacted), of whom 7022 had been in intimate relationships. The prevalence of experiencing any intimate partner violence was 44.8% (95% confidence interval [CI], 43.3-46.2%); physical violence was reported by 29.1% (95% CI, 27.7-30.4%) of participants, sexual violence by 11.7% (95% CI, 10.8-12.7%), and psychological violence by 41.2% (95% CI, 39.8-42.6%). The prevalence of experiencing intimate partner violence was significantly higher among women (48.4%; 95% CI, 46.3-50.4%) than men (40.4%; 95% CI, 38.3-42.5%); the prevalence of physical, sexual, and psychological violence were also higher for women. The proportion of participants of diverse genders who reported experiencing intimate partner violence was high (62 of 88 participants; 69%; 95% CI, 55-83%). The proportion of non-heterosexual participants who reported experiencing intimate partner violence (70.2%; 95% CI, 65.7-74.7%) was larger than for those of heterosexual orientation (43.1%; 95% CI, 41.6-44.6%). More women (33.7%; 95% CI, 31.7-35.6%) than men (22.7%; 95% CI, 20.9-24.5%) reported multitype intimate partner violence. Larger proportions of participants aged 25-44 years (51.4%; 95% CI, 48.9-53.9%) or 16-24 years (48.4%, 95% CI, 46.1-50.6%) reported experiencing intimate partner violence than of participants aged 45 years or older (39.9%; 95% CI, 37.9-41.9%). CONCLUSIONS:Intimate partner violence is widespread in Australia. Women are significantly more likely than men to experience any intimate partner violence, each type of violence, and multitype intimate partner violence. A comprehensive national prevention policy is needed, and clinicians should be helped with recognising and responding to intimate partner violence.

Interpersonal violence is a pervasive experience for women and girls in South Africa. Interpersonal violence occurs early in development and is sustained over the life course with risks of replication across generations. Most children in South Africa are raised by women as fathers are frequently absent, rendering this caregiving relationship particularly significant. We aimed to understand how the mother/caregiver-daughter relationship may impact intergenerational violence transmission in one province in South Africa. Drawing from the third phase of a longitudinal cohort study, young adult daughters (previously interviewed in adolescence) and their former mothers/caregivers, who reported experiences of violence, were purposively sampled for in-depth interviews (n = 34 interviews representing 17 dyads). Using a dyadic approach to analysis, moving beyond individual experience to how these influenced and related to each other dynamically within the dyad, and considering attachment theory as a lens to review and understand findings, four themes were identified. Violence was transmitted through generations via 1) insecure attachment bonds between mothers and daughters, 2) mothers’ inability to protect and support their daughters, and 3) through the perpetuation of patriarchy via gendered attitudes towards parenting and subscription to unequal gender norms. There was recognition of 4) intergenerational changes in attitudes and responses to interpersonal violence and marriage that represented more positive effects on IPV and the parent–child relationship. In situations of continuous violence exposure, strengthening this primary and weighty attachment relationship (working on violence prevention and intervention at the level of the mother/caregiver-daughter dyad), and offering opportunities to meaningfully reflect on violence experiences through one’s life, may represent a durable positive influence on curbing violence transmission in vulnerable families.