Violence behind closed doors: What new evidence from Cox’s Bazar tells us about protecting women and children

Our study used data from a large household survey within a cluster-randomised trial of an early childhood development (ECD) parenting programme. We looked at the prevalence and predictors of violence against children and intimate partner violence. The findings show an urgent need to integrate violence prevention into ECD and family-focused programmes in fragile settings.

Violence within households is alarmingly common

About half of all mothers in both the Rohingya and host communities reported experiencing some form of intimate partner violence in the past month. Similarly, violent discipline of children was widespread, with many mothers – and an even greater number of fathers – reporting physical or psychological punishment of their children.

Children were not just direct victims of violence; many witnessed their fathers being violent at home, especially in Rohingya families. Many women reported feeling fear of their husbands, particularly those living in the camps. These findings challenge the idea that domestic violence is mainly a ‘refugee problem’. Host communities living near refugee settlements often face similar economic and social challenges and are at comparable levels of risk.

Intimate partner violence and violent discipline are closely linked

There is a strong connection between intimate partner violence and violent discipline. Mothers who experienced intimate partner violence were more than twice as likely to report using violent discipline on their children, even after considering socioeconomic and demographic factors. This connection is important for both policy and practice. Children growing up in violent homes are at greater risk of emotional distress, behavioural issues and poor developmental outcomes. If nothing changes, violence at home can become a cycle, normalised and repeated over time.

Who is most at risk?

The study identified several factors linked to higher levels of violence in both communities:

• a child’s age – older children were more likely to experience violent discipline
• household poverty – mothers in poorer households were more likely to experience intimate partner violence
• parental education and norms – higher levels of education were associated with less acceptance of violent discipline, but patterns differed by community. Among Rohingya mothers, higher education often correlated with more reporting of intimate partner violence. This may suggest greater awareness and willingness to disclose abuse rather than a higher actual incidence.

These findings highlight that domestic violence in humanitarian contexts is shaped by complex, specific conditions rather than a single risk factor.

What do these findings mean for early childhood programmes?

The data was collected as part of an ECD parenting intervention delivered through home visits and interactive voice response messages. Although the programme did not specifically aim to prevent domestic violence, mothers who took part reported a lower acceptance of violent discipline and a greater awareness and reporting of intimate partner violence, which likely shows improved recognition of abuse and more confidence to speak out.

This suggests that parenting programmes that promote responsive caregiving and positive parent-child interactions may indirectly shape attitudes toward violence. However, it also emphasises that indirect effects alone are not sufficient.

Why integration is essential?

If ECD programmes want to enhance children’s long-term health and development, they cannot overlook the violence many children face at home. The evidence clearly indicates that preventing violence against intimate partners and children must be integrated into ECD and parenting programmes, especially in crisis-affected and displacement settings.

Integrated approaches present chances to:

• address harmful gender and parenting norms together
• support maternal wellbeing and mental health
• involve both fathers and mothers
• break cycles of violence early in a child’s life.

Moving from evidence to action

Domestic violence in Cox’s Bazar reflects larger structural inequalities, chronic stress and deep-rooted social norms affecting both refugees and host communities. Ignoring it could undermine efforts in ECD, education and health. To create safer environments for children to grow, policymakers and practitioners must go beyond isolated interventions. Integrating evidence-based violence prevention in scalable ECD programmes is essential.

By addressing what goes on behind closed doors, we can take significant steps toward breaking intergenerational cycles of violence and fostering healthier futures for women and children in vulnerable communities.

Further reading: Global perspectives on violence prevention

1. Mutavati, A., Zaman, M. and Olajide, D. (2020, 1 May). Fighting the shadow pandemic of violence against women and children during COVID 19. UN Sustainable Development Group (UNSDG). https://unsdg.un.org/latest/blog/fighting-shadow-pandemic-violence-against-women-and-children-during-covid19  
2. UNICEF. (n.d.). Child protection. Keeping children safe from violence, exploitation and abuse.. https://www.unicef.org/protection/strengthening-child-protection-systems
3. UNICEF Innocenti (2024, 03 July). Webinar on interventions to address intimate partner violence and violence against children. https://www.unicef.org/innocenti/webinar-interventions-intimate-partner-violence-and-violence-against-children  
4. United Nations High Commissioner for Refugees (UNHCR). (n.d.). Gender-based violence. https://www.unhcr.org/what-we-do/protect-human-rights/protection/gender-based-violence
5. United Nations Office on Drugs and Crime (UNODC). (n.d.). End Violence Against Children (END VAC). https://www.unodc.org/endvac
6. Seruwagi, G. . (2024, 16 December). Meeting the moment: Data to action in humanitarian settings. Together for Girls. https://www.togetherforgirls.org/en/blog/meeting-the-moment-data-to-action-in-humanitarian-settings