BACKGROUND: While it is known that trauma systems improve the outcome of injury in children, there is a paucity of information regarding trauma system function amid changes in policies and health care financing that affect emergency medical systems for children. OBJECTIVES: To describe the trends in the proportion of pediatric trauma patients acutely hospitalized in trauma-designated versus non-trauma-designated hospitals. METHODS: This was a retrospective observational study of a population-based cohort obtained by secondary analysis of a publicly available data set: the California Office of Statewide Health Planning and Development Patient Discharge Database from 1998 to 2004. Patients were included in the analysis if they were 0-19 years old, had International Classification of Disease, Ninth Revision (ICD-9) diagnostic codes and E-codes indicative of trauma, had an unscheduled admission, and were discharged from a general acute care hospital (N = 111,566). Proportions of patients hospitalized in trauma-designated hospitals versus non-trauma-designated hospitals were calculated for Injury Severity Score and death. Injury Severity Scores were calculated from ICD-9 codes. Primary outcomes were hospitalization in a trauma center and death two or more days after hospitalization. RESULTS: Over the study period, the proportion of children aged 0-14 years with acute trauma requiring hospitalization and who were cared for in trauma-designated hospitals increased from 55% (95% confidence interval [CI] = 54% to 56%) in 1998 to 66% (95% CI = 65% to 67%) in 2004 (p < 0.01). For children aged 15-19 years, the proportion increased from 55% (95% CI = 54% to 57%) in 1998 to 74% (95% CI = 72% to 75%) in 2004 (p < 0.0001). When trauma discharges were stratified by injury severity, the proportion of children with severe injury who were hospitalized in trauma-designated hospitals increased from 69% (95% CI = 66% to 72%) in 1998 to 84% (95% CI = 82% to 87%) in 2004, a rate higher than in children with moderate injury (59% [95% CI = 58% to 61%] in 1998 and 75% [95% CI = 74% to 76%] in 2004) and mild injury (51% [95% CI = 50% to 52%] in 1998 and 63% [95% CI = 62% to 64%] in 2004) (p < 0.0001 for each injury severity category and both age groups). Of the hospitalized children who died two or more days after injury (n = 502), 18.1% died in non-trauma-designated hospitals (p < 0.002 for children aged 0-14 years; p = 0.346 for children aged 15-19 years). CONCLUSIONS: An increasing majority of children with trauma were cared for in trauma-designated hospitals over the study period. However, 23% of children with severe injuries, and 18.1% of pediatric deaths more than two days after injury, were cared for in non-trauma-designated hospitals. These findings demonstrate an important opportunity for improvement. If we can characterize those children who do not access the trauma system despite severe injury or death, we will be able to design clinical protocols and implement policies that ensure access to appropriate regional trauma care for all children in need.