For the period covered by the data (1980-2005), assault admissions more than doubled from 2.8 per 10,000 children in 1981 to 6.1 per 10,000 in 2005 (p less than 0.0001). Boys older than 12 years old were at greater risk of an assault, and children younger than 6 years old were more likely to be maltreated, as were those children from greater disadvantaged backgrounds. Aboriginal children were more likely to experience assault and maltreatment compared with non-Aboriginal children. Common characteristics of assault cases were injuries of the skull and facial bones and intracranial, wrist, hand, and abdominal injuries. Children admitted for maltreatment-related injuries were more likely to have superficial head or abdominal injuries, and a high proportion of maltreated children had infectious and parasitic diseases, particularly intestinal infections. The authors advise that broadening child maltreatment surveillance to include children's hospital admissions for assault and maltreatment injuries is an important public health initiative that can be improved with the increased use of external-cause codes. Health data are collected using international coding standards that enable comparability across States and countries. Such data have clinical implications by highlighting injuries related to child abuse and neglect. This was a retrospective cohort study of all children 0-17 years old in Western Australia from 1980 to 2005. Annual trends in prevalence of assault and maltreatment-related hospital admissions were calculated, and child characteristics were examined using logistic regression models. 3 tables, 2 figures, and 28 references