Developmental Timing of Childhood Abuse May Impact Conduct Problem Trajectories

child sitting in chair, dad yelling
child sitting in chair, dad yelling
While the relationship between early life abuse and behavioral problems in children is well-documented, the effects of abuse on adolescent development remain poorly understood.

Study data published in BMC Psychiatry demonstrate a strong association between the developmental timing of childhood abuse and conduct problems in early life. In a cohort study of individuals in the United Kingdom, childhood-onset and adolescence-onset conduct problems were each associated with abuse exposure, suggesting that common risk factors subtend both trajectories. Abuse exposure persisting through both childhood and adolescence was the strongest predictor of conduct problems.

While the relationship between early life abuse and behavioral problems in children is well-documented, the effects of abuse on adolescent development remain poorly understood. To better inform this relationship, investigators abstracted data from the Avon Longitudinal Study of Parents and Children, an ongoing prospective birth cohort study conducted in the UK. Briefly, the Avon study enrolled pregnant women with an estimated delivery date between 1991 and 1992.

Follow-up questionnaires were administered at regular intervals to mothers and their offspring. The Strengths and Difficulty Questionnaire (SDQ) was used to rate behavioral problems when children were aged 4, 7, 8, 10, 12, 13, and 17 years. Exposure to physical, psychological, and sexual abuse was captured by retrospective self-report when offspring were aged 22 years. Childhood was defined as before age 11 years. Adolescence was defined as between ages 11 and 17 years.

Latent class growth analyses were used to identify developmental trajectories of conduct problems. Multinomial logistic regression was then used to estimate the association between abuse exposure and each trajectory.

Complete data on abuse exposure and covariates were available for 3172 individuals. Overall, 4 developmental trajectories were identified: (1) early-onset persistent conduct problems, which emerge in childhood and continue throughout the life course (4.8% of sample); (2) adolescence-onset conduct problems, which emerge in adolescence (4.5%); (3) childhood-limited conduct problems, which emerge childhood but resolve prior to adolescence (15.4%); and (4) low conduct problems (75.3%).

Across all exposure categories, 19.6% of the sample reported experiencing at least some form of abuse in childhood or adolescence. The prevalence of abuse exposure was substantially elevated in the early-onset persistent (40.9%) and adolescence-onset (37.5%) conduct problem classes compared to the childhood-limited (23.8%) and low (16.8%) classes.

Childhood-only abuse was associated with substantially increased odds of belonging to the early-onset persistent (odds ratio [OR], 3.94; 95% confidence interval [CI], 1.71-9.02) and adolescence-onset (OR, 3.24; 95% CI, 1.27-8.26) trajectories compared to the low conduct problems trajectory. Individuals exposed to abuse in both childhood and adolescence were also more likely to display early-onset persistent (OR, 6.75; 95% CI, 3.49-13.06) and adolescence-onset (OR, 6.48; 95% CI, 3.31-12.68) conduct problems.

These associations persisted across both physical and psychological abuse categories. Notably, abuse experienced in adolescence was not a significant predictor of membership to any behavioral trajectory.

Results from this study emphasize the relationship between childhood abuse exposure and behavioral trajectories in later life. Based on regression data, investigators hypothesized that the relationship between abuse and conduct problem trajectories are “more quantitative than qualitative in nature.” In other words, common risk factors inform all behavioral problem trajectories, and the degree of exposure to each risk factor contributes to overall behavioral problems. Study limitations include the small number of respondents to the abuse survey compared to the total birth cohort. The high attrition rate may have introduced bias and led to an underestimation of the true prevalence of abuse.

“Our findings…demonstrate the importance of adopting measures covering both childhood and adolescence when investigating the timing and persistence of child abuse, as harsh and abusive parenting may persist up to emerging adulthood,” the investigators wrote.


Bauer A, Hammerton G, Fraser A, Fairchild G, Halligan SL. Associations between developmental timing of child abuse and conduct problem trajectories in a UK birth cohort. BMC Psychiatry. Published online March 16, 2021. doi:10.1186/s12888-021-03083-8