Maternal Intimate Partner Violence Predicts Contact with Child Protective Services

Maternal Intimate Partner Violence (IPV) is a national epidemic, affecting 1 in 3 American women and exposing 15 million children to toxic stress annually. Current risk assessments often overlook infancy, failing to recognize early IPV as an independent driver of child welfare entry. This study examined whether early, later, and persistent exposure to maternal IPV predicts Child Protective Services (CPS) contact by age 9 and whether these associations persist after accounting for child, maternal, and socioeconomic factors.

Data were drawn from the Future Families and Child Wellbeing Study (FFCWS), a longitudinal birth cohort of 4,898 urban mother-child dyads. Analyses were restricted to participants with non-missing data on maternal IPV (year 1), CPS contact (years 5 and 9), and all covariates, yielding an analytic sample of 1,493 mother-child dyads. Maternal IPV was measured using self reported items that captured physical, emotional, sexual, economic, and coercive control behaviors perpetrated by the child’s father. The outcome was any CPS contact by year 9. Covariates included child characteristics (sex, low birth weight), maternal risk factors (education, age, race, substance use, depression), and household context (poverty, parental co-residence, number of children in the household). Multivariable logistic regression was used to examine associations of IPV at year 1 with CPS contact, adjusting for covariates. A second model used IPV trajectories across years 1 to 9 (no IPV, early, later, persistent) as the primary predictor, with the same covariates, among participants with complete data across all waves (N = 1,013).

Overall, 45.7% of children were exposed to IPV at year 1 and 18.4% had CPS contact by year 9. Multivariable results showed that maternal IPV at year 1 was associated with higher odds of CPS contact (OR = 1.67, 95% CI [1.26-2.21]) after adjustment for child, maternal, and household risk factors (Nagelkerke R² = .126; N = 1,493). In the trajectory model, IPV trajectories across childhood significantly predicted CPS contact. After adjustment for covariates, only persistent IPV remained significantly associated with higher odds of CPS contact (OR = 2.69, 95% CI [1.47-4.94]; Nagelkerke R² = .173; N = 1,013), whereas early and later IPV were not significant. Across both models, poverty, maternal substance use, and lower maternal education were significantly associated with higher odds of CPS contact, whereas child characteristics, maternal race, and family structure were not.

These findings indicate that IPV should be treated as an early warning signal of child welfare involvement. Screening and support should be embedded within early childhood and public health systems. Relatedly, preventive services should be initiated at first exposure, not after repeated harm. Practice should shift from incident-based responses to trajectory-informed assessment. Practitioners should not require survivors to repeatedly prove danger. Instead, maternal safety should be recognized as a primary pathway to child protection. Child welfare systems should integrate domestic violence advocacy through co-located services and develop early warning approaches that use longitudinal data to identify IPV exposure. Policy should prioritize prevention, as early support for families is more effective and less costly than later system involvement.