Beyond Household Adversities: Synthetic Imputation of Emotional, Physical and Sexual Abuse for Children Aged <18 Years in the National Survey on Children’s Health (NSCH) Through Parameter Bridging from the Behavioral Risk Factor Surveillance System (

The National Survey of Children’s Health (NSCH) is widely used to estimate adverse childhood experiences (ACEs) in pediatric populations, but it only captures household adversity and economic hardship, omitting direct measures of emotional, physical, and sexual abuse present in the original ACE Study. We quantified this measurement gap by projecting unmeasured abuse domains onto the pooled 2023–2024 NSCH (n = 104,838) using parameter-bridging models estimated from the Behavioral Risk Factor Surveillance System (BRFSS; adults aged 18–49 years, n = 27,499) and the Youth Risk Behavior Surveillance System (YRBSS; adolescents, n = 11,016), both of which directly measure these abuse domains. Raw NSCH data classify 93.9% of infants under 12 months and 61.4% of adolescents aged 14–17 years as free of any ACEs; incorporating projected abuse domains reduces these figures to approximately 53–57% and 36–46%, respectively, depending on donor source. Conversely, the proportion with four or more ACEs—0.2% and 3.8% in the raw NSCH data for these two age groups—rises to approximately 2% and 7–14% once projected abuse is incorporated. These projected values represent model-based structural risk conditional on a child’s observed household environment, not direct measures of accumulated exposure or observed prevalence. Out-of-sample calibration testing indicates that projections from each donor source carry substantial asymmetric uncertainty: BRFSS-derived estimates over-predict and YRBSS-derived estimates under-predict emotional and sexual abuse relative to each source’s own directly observed rates. Despite this uncertainty, the direction and scale of the gap are consistent across donor sources and sensitivity analyses. NSCH-based ACE indices that omit direct abuse likely substantially understate true childhood adversity, with implications for screening, resource allocation, and cross-population comparisons relying on this widely used pediatric surveillance tool.