Methods to Estimate the Community Incidence of Child Maltreatment

Existing estimates of child maltreatment incidence, such as the National Incidence Study (NIS) are often expensive to conduct, untimely, and may lack the precision and statistical power necessary to assess critical group differences in maltreatment (Drake and Jonson-Reid 2011). The purpose of this project was to develop a methodology that could be generalized to other states, could be used in conjunction with or in place of the NIS, and could produce estimates of the incidence of child maltreatment inclusive of both children who are identified as victims of abuse or neglect in a given year and those who are victims, but do not come to the attention of CPS. In the context of a population-level understanding of maltreatment exposure, linked administrative data have the potential to be prospectively leveraged as an alternative or complement to the NIS, generating more cost-effective, timely, local, and (potentially) more accurate estimates of the victims of child abuse and neglect.

Cross-sectional findings

• The project found variation in base and cumulative substantiation rates (+/ - 1, 2, 3 years) by county and by allegation type.
• The methodology developed in the cross-sectional work holds the promise of helping other jurisdictions develop alternative estimates of child maltreatment incidence through simple extensions of the window in which substantiation is counted.

Longitudinal findings

• The cumulative proportions of Black and Native American children who had child protection encounters were significantly higher than those of other children. Both groups also experienced termination of parental rights at twice the rate of White children in the cohort.
• The likelihood of child protection involvement exhibited an inverse relationship to both maternal age at birth and maternal education levels.
• Receiving public health insurance and missing paternity were associated with higher likelihoods of any level of CPS involvement.

Cross-sectional strategy

The project team linked California birth and death records with CPS records from 2012 through 2018 to identify a cross-section of children, across counties and demographics, identified as substantiated victims of maltreatment in 2015. The team calculated a series of revised cumulative victimization rates that considered both children who were substantiated victims in a base year, as well as those who were referred to CPS but were substantiated as a victim within three years on either side of the base year. In addition, the team explored maltreatment victimization rate estimates by allegation type (emotional, physical, and sexual abuse, neglect) and by geography, as well as interactions between allegation type and geography.

Longitudinal strategy

The project team organized CPS records linked with birth records longitudinally for a cohort of children born in 1999 to estimate the cumulative childhood risk of abuse and neglect in California. With these data, the team examined differences in annual incidence and cumulative prevalence rates by county and by demographic characteristics at birth. The team modeled probability of exposure of abuse or neglect symptoms to mandated reporters to assess a range of incidence rate estimates. In addition, the longitudinal approach allowed the team to examine “persistence,” which would identify children whose risk for long-term negative outcomes is heightened because of their repeated involvement with the child welfare system.

Putnam-Hornstein, E., Prindle, J., Cuccaro-Alamin, S., Wiegmann, W., Magruder, J., Webster, D., Ahn, E., and Foust, R. “Methods to estimate the community incidence of child maltreatment” OPRE Report #2021-253, Washington, DC: Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services, 2021.

Further details on the longitudinal analyses and findings can be found in:

1. Prindle, J., R. Foust, and E. Putnam-Hornstein. “Maltreatment Type Classifications and Transitions During Childhood for a California Birth Cohort.” Child Maltreatment, 2021. Available at https://journals.sagepub.com/doi/10.1177/10775595211006784 .
2. Putnam-Hornstein, E., E. Ahn, J.J. Prindle, J. Magruder, D. Webster, and C. Wildeman. “Cumulative Rates of Child Protection Involvement and Terminations of Parental Rights in a California Birth Cohort, 1999 — 2017.” American Journal of Public Health, 2021, pp. e1-e7. Available at https://doi.org/10.2105/AJPH.2021.306214 .