When Caregivers Carry Childhood Trauma: Nine-Year Study Traces How Abuse and Neglect Echo Into the Next Generation

In the mountain villages of rural western China, where parental labor migration often leaves grandparents holding the daily work of raising children, a team of researchers has spent nearly a decade quietly measuring something profound: whether the scars a caregiver carries from their own childhood leave visible marks on the development of the children they raise. The answer, drawn from five waves of longitudinal data collected between 2015 and 2024, is a cautious but striking yes—and the pattern of those marks changes in unexpected ways as children grow.

The study, published in SSM – Population Health, followed 1,243 children from infancy through middle childhood, generating 5,575 child-wave observations across twenty low-income counties in the Qin-ba Mountain region. At the final wave, researchers asked the primary caregiver about their own experiences of childhood abuse and neglect using the 28-item Childhood Trauma Questionnaire-Short Form, a validated retrospective instrument covering emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect. Half of all caregivers reported exposure to at least one subtype, and the burden fell unevenly across generations: 67 percent of grandparents reported childhood abuse or neglect, compared with 42 percent of parents.

What makes this study unusual is not just its setting but its duration. Children were assessed with age-appropriate standardized instruments at every wave—the Bayley Scales of Infant and Toddler Development in early years, then the Wechsler Preschool and Primary Scale of Intelligence, and finally the Wechsler Intelligence Scale for Children as they approached age ten. Social-emotional skills were tracked with the Ages and Stages Questionnaires: Social-Emotional and, later, the Chinese version of the Strengths and Difficulties Questionnaire. This design allowed the researchers to watch developmental trajectories unfold rather than capture a single snapshot, a luxury rarely afforded in low- and middle-income country research on intergenerational adversity.

The statistical machinery behind the findings was equally rigorous. The team fitted multilevel mixed-effects regression models in which repeated developmental assessments were nested within children, allowing them to account for the correlation among repeated observations from the same child while accommodating unbalanced data. Interaction terms between caregiver childhood abuse and neglect and survey wave tested whether associations changed over time. Propensity score matching with a caliper of 0.05 improved covariate balance between exposed and unexposed caregivers, and all absolute standardized mean differences fell below the conventional 0.10 threshold after matching. Analyses restricted to families who never received the original parenting intervention produced substantively similar results, suggesting the associations were not artifacts of the trial itself.

The headline result is that caregivers’ histories of childhood abuse and neglect were consistently associated with poorer outcomes in the next generation. In the full sample, caregiver exposure predicted lower standardized cognitive scores (β = –0.12, p = 0.002) and substantially lower social-emotional scores (β = –0.21, p < 0.001). In the consistent-caregiver subsample—children whose primary caregiver remained the same from baseline through Wave 4—the social-emotional association grew even stronger (β = –0.36, p < 0.001). But the temporal patterns diverged in a way that surprised the research team: a significant negative interaction between caregiver exposure and survey wave indicated that cognitive disparities widened as children aged, while the social-emotional gap remained stable across waves.

Subtype analyses sharpened the picture further. Physical neglect, the most prevalent form of caregiver adversity at 46 percent, was associated with poorer outcomes in both domains, and the cognitive disparities linked to physical neglect widened significantly over time. Emotional neglect, by contrast, showed a clearer association with social-emotional development (β = –0.27, p < 0.001 in the full sample) and less consistent evidence for cognition at the average wave, though a significant interaction in the consistent-caregiver subsample suggested cognitive disparities associated with emotional neglect emerged gradually as children grew. Abuse subtypes were too rare in this sample to support subtype-specific analysis, a sobering reminder that neglect, not violence, dominates the adversity landscape in this population.

Perhaps the most consequential finding concerns the pathway itself. Attachment theory has long suggested that caregivers’ early relational experiences shape their internal working models of caregiving—their capacity to read children’s cues and respond sensitively. If that theory holds, parenting practices should statistically account for part of the intergenerational association. The researchers measured both positive parenting, through caregiver-reported activities like reading, storytelling, singing, and playing, and neglectful parenting, using the Rural Child Neglect Assessment Scale covering physical, emotional, educational, medical, safety, and social domains. Mediation analyses used the product-of-coefficients approach with bias-corrected 95 percent confidence intervals from 1,000 non-parametric bootstrap replications.

The mediation results were revealing precisely because they were limited. For social-emotional outcomes, indirect associations ran chiefly through neglectful parenting measured in the same wave as the outcome (Wave 3: β = –0.07, 95% CI –0.11 to –0.03; Wave 4: β = –0.15, 95% CI –0.21 to –0.10), while positive parenting produced only a small indirect association at Wave 4. For cognitive outcomes, evidence of mediation was nearly absent; only the cross-wave estimate from neglectful parenting at Wave 3 to cognition at Wave 4 excluded zero, and it was small. The researchers are careful to frame these as statistical rather than causal mediation, and to note that caregiver mental health—another well-established intergenerational pathway—was not examined as a mediator and may account for additional variance.

The study also broke new ground by including grandparental caregivers, a population almost entirely absent from the intergenerational adversity literature. Grandparents in this sample reported more childhood adversity than parents, plausibly reflecting generational exposure to famine-related deprivation and material hardship. Point estimates were larger for parental than grandparental exposure on both outcomes, and the cognitive association was statistically significant only among parents (β = –0.14, p = 0.006 versus β = –0.06, p = 0.369 for grandparents). Formal tests of coefficient equality, however, found no statistically significant differences between caregiver types, so the researchers treat these patterns as hypothesis-generating rather than established heterogeneity.

The implications for intervention are cautiously optimistic. Because neglectful parenting is modifiable and is the direct target of many early childhood programs, the findings suggest that approaches recognizing caregivers’ own histories of adversity while strengthening responsive and stimulating caregiving may help address social-emotional difficulties in the next generation. In rural China and other settings where grandparents frequently assume primary caregiving, such efforts may need to include grandparental caregivers rather than focusing exclusively on mothers. The researchers acknowledge limitations—retrospective self-report may under-identify adversity, the observational design limits causal inference, and the parenting measures did not capture harsh discipline or coercive control. Yet the nine-year window itself is the study’s quiet triumph: it reveals that the developmental consequences of a caregiver’s childhood do not simply appear and fade, but in the cognitive domain, they compound—widening steadily as children age, a pattern invisible to any single-wave analysis.

Subject of Research: Caregiver childhood abuse and neglect and children’s cognitive and social-emotional development: The role of parenting practices

Article Title: Caregiver childhood abuse and neglect and children’s cognitive and social-emotional development: The role of parenting practices

Article References: Song, Q., Gong, H., Liu, M., Qiao, N., Yue, A., & Shi, Y. (2026). Caregiver childhood abuse and neglect and children’s cognitive and social-emotional development: The role of parenting practices. SSM – Population Health, Article 101988. https://doi.org/10.1016/j.ssmph.2026.101988

Image Credits: AI Generated

DOI: 10.1016/j.ssmph.2026.101988

Keywords: Caregiver, childhood, abuse, neglect, children, cognitive, social-emotional, development, role, parenting, practices, scientific research