If you’re here because you’re worried about your children, you’re asking a fair question. The research gives some clear answers and some honest limits, and I’ll give you both.
Children who live with abuse between the adults who care for them are more likely, as a group, to have difficulties with emotions, behavior, school, friendships, sleep, and health. The research also shows that children differ from one another, that some adjust well, and that symptoms can improve with help.
This page is about children age 13 and younger. Most of the studies look at children from infancy through about age eleven. A few reach older ages or include adults looking back on childhood, and I’ll tell you when they do. Everything here comes from 16 peer-reviewed papers, listed at the end.
I stayed in my abusive marriage for 17+ years, thinking that I could protect my sons from harm. It wasn’t until near the end of my marriage that I realized how much hell they were growing up in. It didn’t matter how valiantly I thought I protected them. They were living in it too.
Hopefully, I can help heal some of the pain they suffered by educating them and taking responsibility for my part in their upbringing. It’s been years since I left their father, but I feel there is a long way to go before both my sons are mentally and emotionally healthy.
The Short Version
- Children don’t have to see an assault to be exposed. Hearing abuse, seeing what it leaves behind, and living with threats or control all count.
- When abuse starts in a child’s first years, studies link it to emotional and behavior difficulties, trouble with attention and self-control, and social difficulties later on.
- Studies also link it to lower school scores, language problems, sleep trouble, and asthma-related health visits. One study found higher blood pressure in children who were exposed at the time they were measured.
- A 2026 review of 36 studies links childhood exposure to anxiety, depression, and trauma-related symptoms.
- These are patterns across groups of children. They can’t tell you what will happen to your child, and the research does not support calling every exposed child permanently damaged.
- Some things are linked to better outcomes: a mother’s mental health, responsiveness, and engagement. In one trial, therapy improved children’s PTSD and anxiety symptoms.
What Does “Witnessing” Domestic Abuse Mean?
When researchers talk about children witnessing abuse, they mean more than seeing an assault. A child’s exposure can include hearing abuse, seeing what it leaves behind, and living with threats or control.
This page covers psychological abuse, coercive control, and physical violence between the adults who care for a child. If your child was in another room, that does not show your child was unaffected. That is a finding about exposure. It says nothing about how hard you worked to protect your children.
This applies to babies too. An infant can’t tell you what happened, and several of these studies measured abuse in a child’s first year or first few years anyway. They found links with difficulties at ages five, ten, and fifteen. They did not test whether a child remembers any particular incident.
Emotional and Behavior Difficulties
Two studies that followed children from birth looked at emotions and behavior. In France, researchers followed 9,639 children. When psychological abuse between partners continued during a child’s first two years, children had higher odds of scoring in the abnormal range on a screening questionnaire about emotions and behavior at age five. The odds were about 2.3 times higher. Children whose exposure decreased over that time also had higher odds of difficulties.
A screening score is a flag, not a diagnosis. The questionnaire points to children who may be struggling, and it can’t tell you a child has a disorder. Odds are also not the same as the share of children affected. And because psychological and physical abuse can overlap, the study can’t cleanly separate them.
An Australian study began with 1,507 families and tracked abuse across the first ten years of the children’s lives. Children whose exposure was early, increasing, or persistent each had higher odds of emotional and behavior difficulties at age ten, roughly two to three times higher.
For conduct problems in children with persistent exposure, the odds were more than six times higher. The range around that estimate is very wide, from about 2.3 to about 16.4, so it tells you the direction of the link and not its size.
None of this means your child will have these difficulties. It means that, across many children, they showed up more often in those who lived with abuse.
Attention and Self-Control
In a U.S. study of 4,338 children, abuse between caregivers during infancy was linked to poorer attention and behavior regulation at age five, which means more difficulty focusing and managing behavior.
The analysis also found statistical pathways running through parenting stress and maternal warmth. A pathway like this shows what an abused parent may need support with. It does not make you responsible for the abuse.
School and Learning
A U.K. study followed a cohort that began with about 19,000 children. By age eleven, children who lived with abused mothers had lower English and math scores. The children were assessed at ages seven and eleven. The authors adjusted for selection and say that factors they did not measure could still play a part.
These scores measure school and language performance. They do not show that a child is less intelligent.
Another study from the same Australian cohort followed 615 mother-child pairs. Mothers reported on abuse when their children were one, four, and ten, and the researchers measured outcomes at age ten. Early exposure alone was linked to roughly three times the odds of language problems at age ten.
Friends and Social Skills
Social skills show up in the research too. A third study from the same Australian cohort, which began with 1,507 mother-child pairs, linked abuse during a child’s first year to peer and social difficulties at age ten. A statistical pathway ran through the mother’s depressive symptoms, measured when the child was four. The study also found that a mother’s engagement in learning at home was linked to stronger social skills.
The same caution applies here. A pathway through a parent’s depressive symptoms points to a need for support, and it does not make that parent responsible for the abuse. The study also relies on mothers’ reports, and not every family stayed in it, which limits how far the results reach.
Children’s social skills did not all follow the same pattern. In a U.S. study of 1,678 children in the child welfare system, social skills from age three to age ten followed different patterns, and some children improved after an earlier decline. A mother’s responsiveness was linked to which pattern a child showed. These children also had other experiences of maltreatment, and the data were collected between 1999 and 2007, so the study can’t isolate the effect of abuse between caregivers.
Sleep and Physical Health
Sleep difficulties come up again and again in this research. A 2026 review combined 18 studies and found higher odds of sleep disturbance, about 1.5 times higher, in people who witnessed partner violence as children. The range around that figure was 1.16 to 1.92. The studies differed widely from one to the next, so the number is not a precise forecast for any child. The review covers childhood through adult life, not only children under fourteen.
The Australian age-ten study of 615 mother-child pairs also linked exposure to sleep difficulties. It found associations with asthma-related healthcare consultations, and early exposure alone was linked to roughly three times the odds of asthma at age ten. In children with concurrent exposure, meaning exposure at the time they were measured, it also found higher blood pressure.
Some of these outcomes come from parent reports and healthcare visits, and the study can’t show that abuse alone caused these conditions.
Anxiety, Depression, and Trauma Symptoms
Researchers have also looked at how children feel. A 2026 meta-analysis pooled 36 studies with 59,561 participants. It found associations between childhood exposure to partner violence and anxiety, depression, and trauma-related symptoms. In the paper’s terms, these were internalizing, depressive, and PTSD-related symptoms.
The review included children exposed before age eighteen, so the results are not limited to children under fourteen. Nineteen of the studies measured children at one point in time, and seventeen followed them over time. The larger and higher-quality studies generally found smaller associations.
Stress Biology and the Brain
You may have seen claims about abuse and children’s brains. Here is what the research supports. Some researchers study how the body’s stress systems and the brain respond. A scoping review first published in September 2026 mapped 62 quantitative studies of biological and brain-structure findings after childhood exposure to partner violence. It combines studies of children with studies of adults looking back on childhood exposure.
A scoping review maps what has been studied, and it does not produce one overall estimate. Exposure definitions and age ranges vary across the studies. It does not show one pattern of brain injury that applies to every exposed child, and it can’t be used to test an individual child. So if you hear that abuse has permanently changed a child’s brain, this research does not support saying that.
When One Caregiver Controls the Other
Not all of the abuse in this research is physical. A systematic review of 51 studies connected control between caregivers, often called coercive control, with difficulties for children and families. The review found effects on family relationships, children’s opportunities for social contact, and access to health care, along with emotional and behavior problems.
The review covers children and families broadly, not only children under fourteen. It also includes outcomes during pregnancy and cases where the child was directly maltreated, so not every finding comes from a child witnessing nonphysical abuse.
Later Outcomes
You may be wondering what all of this means for your child years from now. The research on later outcomes is limited, so I’ll go through it carefully.
A U.S. study followed 1,690 children. Abuse between caregivers when a child was one to three years old predicted more depression symptoms at age fifteen, directly and through pathways involving physical punishment and peer victimization, which means being targeted by other children. The overall association was small, and the study does not establish depression in adulthood.
A Manitoba study compared 10,731 children exposed to partner violence before age eighteen with 53,655 matched children who were not exposed. It followed them into young adulthood. The exposed group had poorer school outcomes and was more likely to be the victim of a crime later on. The study relied on recorded exposure, so it can miss abuse that was never reported, and its estimates cover children exposed at any age before eighteen.
A 2025 review of 26 studies found that people exposed to violence in childhood had about twice the odds of emotionally abusing a partner as adults (2.00, with a range of 1.78 to 2.23). The exposure measures include broader family violence and maltreatment, not only abuse between caregivers. Eighty-nine percent of the papers were rated at moderate or high risk of bias.
This does not mean your child is destined to repeat abuse. The research does not support describing every exposed child as permanently damaged or certain to become abusive.
What These Studies Can and Can’t Tell You
- They describe groups, not your child. These findings describe differences between groups of children. They do not determine what will happen to any one child.
- A link does not prove cause. Most of these studies observe families. They do not assign children to be exposed. Financial hardship, a parent’s health, and direct abuse of the child can overlap with abuse between caregivers.
- A screening score is not a diagnosis. A high score on a questionnaire, a behavior difficulty, and a diagnosis from a clinician are three different things.
- Odds are not probabilities. If you see “twice the odds” somewhere, it does not mean twice the chance, and it does not mean half of exposed children have the outcome. A correlation of 0.263, the overall figure in the 2026 mental health review, is not a percentage of children affected.
- Studies on the same families are not separate confirmations. Three of the Australian studies use the same group of families, the Mothers’ and Young People’s Study. Two of the U.S. studies use the same research program, Future of Families/Fragile Families. They ask different questions.
- Publication date is not data date. A study published in 2025 can use data collected between 1999 and 2007.
What Seems to Help
The research does not only describe risk. Researchers also look at what is linked to better outcomes. A review of 15 studies of children age zero to twelve who were exposed to abuse looked at factors linked to positive adjustment. A mother’s mental health was repeatedly linked to better adjustment. Support for emotion coaching, parenting, and a child’s temperament was preliminary, and broader social and community influences have been understudied.
Two findings from the studies above relate to children’s social skills:
- A mother’s responsiveness was linked to the pattern a child’s social skills followed from age three to ten, and some children improved after an earlier decline.
- A mother’s engagement in learning at home was linked to stronger social skills at age ten.
These are possible protective factors, not guarantees. The research does not show that any one change protects every child.
There is also evidence that treatment can help. In one U.S. randomized trial, 124 children age seven to fourteen had symptoms after being exposed to partner violence. Eight sessions of trauma-focused cognitive behavioral therapy improved PTSD and anxiety symptoms related to the abuse more than child-centered therapy did. The children were selected because they had symptoms, and the trial included children older than thirteen. It does not promise benefit for every child or every outcome. It does show that measured symptoms can improve.
If you want to look into this for your child, ask a counselor or doctor about trauma-focused cognitive behavioral therapy.
What This Means for You
Here is how I read all of this. The research supports paying attention to two things together: your child’s needs, and the safety and wellbeing of the adult who is being abused. If you are that adult, supporting you can also support your child. The studies with pathways through parenting stress, depressive symptoms, and warmth point to support needs. They do not put the blame for the abuse on you. This is a practical reading of the studies, not an assessment of your child.
The research does not tell you whether to stay or to leave. If you are weighing that, see Safety Planning for Domestic Abuse and Violence.
Honestly, I don’t know how much good knowing how witnessing domestic abuse would have helped me when my marriage and children were young. I didn’t think we were abused early on. But if I had family and friends around me, maybe I could have left my abusive marriage years before.
If You Need to Talk to Someone
In the United States, the National Domestic Violence Hotline is open 24 hours a day. Call 1-800-799-SAFE (7233), text START to 88788, or chat at thehotline.org. Outside the United States, contact your local domestic violence service or emergency number. If you or your children are in immediate danger, call 911 or your local emergency number.
If the person abusing you can see your phone or computer, think about using a device they have never had access to. Call logs and browser history can show what you did.
Sources
This page draws on 16 peer-reviewed papers. The research was checked on October 5, 2026. For several papers, the findings come from the authors’ or publisher’s abstract and not the full text. The hotline details were checked at thehotline.org on the same day.
Cohen, Mannarino, and Iyengar, 2011. Community Treatment of Posttraumatic Stress Disorder for Children Exposed to Intimate Partner Violence: A Randomized Controlled Trial. Archives of Pediatrics & Adolescent Medicine, 165(1).
Gregorini, Sacchi, and De Carli, 2026. A Scoping Review of Neurobiological and Brain Structural Alterations Following Child Exposure to Intimate Partner Violence. Trauma, Violence, & Abuse.
Abdul Hamid and colleagues, 2026. Witnessing intimate partner violence in childhood and sleep disturbances across the lifespan: A systematic review and meta-analysis. Social Science & Medicine, 403, 119373.
Xu, Xie, Li, Zhao, and Li, 2026. Childhood Exposure to Intimate Partner Violence: A Three-Level Meta-Analysis to Examine the Impact on Mental Health. Trauma, Violence, & Abuse.
Schulz and colleagues, 2026 (online 2025). The role of maternal depressive symptoms and maternal engagement as mechanisms between intimate partner violence exposure during infancy and social functioning in middle childhood. International Journal of Behavioral Development, 50(5).
Russell and colleagues, 2026. Educational, judicial, and social outcomes of children exposed to intimate partner violence. Child Abuse & Neglect, 176, 108075.
Bailhache and colleagues, 2025. Psychological intimate partner violence, child witnessing of parental arguments, and emotional-behavioral outcomes in five-years old: The French ELFE cohort. Child Abuse & Neglect, 160, 107185.
Jofre-Bonet, Rossello Roig, and Serra-Sastre, 2025. Intimate partner violence and children’s cognitive development. Empirical Economics, 69(2).
Zhang, Liu, and Wang, 2025 (online 2024). Understanding the link between intimate partner violence exposure and children’s self-regulation: The mediating role of parenting stress and warmth. Family Process, 64(1), e13054.
Holmes and colleagues, 2025 (online 2024). Examination of protective factors that promote prosocial skill development among children exposed to intimate partner violence. Development and Psychopathology, 37(1).
Osborne, Munasinghe, and Page, 2025. The Intergenerational Transmission of Emotional Intimate Partner Violence: a Systematic Review and Meta-Analysis. Journal of Family Violence.
Fogarty and colleagues, 2024. Patterns of intimate partner violence exposure across the first 10 years of life and children’s emotional-behavioural outcomes at 10 years. Child Abuse & Neglect, 153, 106814.
Xyrakis and colleagues, 2024 (online 2022). Interparental Coercive Control and Child and Family Outcomes: A Systematic Review. Trauma, Violence, & Abuse, 25(1).
Gartland and colleagues, 2021. Intimate partner violence and child outcomes at age 10: a pregnancy cohort. Archives of Disease in Childhood, 106(11).
Huang, Chen, and Cheung, 2021. Early childhood exposure to intimate partner violence and teen depression symptoms in the U.S.. Health & Social Care in the Community, 29(5).
Fogarty, Wood, Giallo, Kaufman, and Hansen, 2019. Factors promoting emotional-behavioural resilience and adjustment in children exposed to intimate partner violence: A systematic review. Australian Journal of Psychology, 71.
Featured photo by Luke Pennystan