Objective This article describes the epidemiology of child pedestrian fatalities in British Columbia using data generated by the province's Child Death Review Unit, to demonstrate the unique capacity of child death review to provide an ecological understanding of child mortality and catalyse evidence based, multi-level prevention strategies.

Methods All child pedestrian fatalities in British Columbia from 1 January 1 2003 to 31 December 2008 were reviewed. Data on demographics, circumstance of injury, and risk factors related to the child, driver, vehicle, and physical environment were extracted. Frequency of sociodemographic variables and modifiable risk factors were calculated, followed by statistical comparisons against the general population for Aboriginal ancestry, gender, ethnicity, income assistance and driver violations using z and t tests.

Results Analysis of child pedestrian fatalities (n=33) found a significant overrepresentation of Aboriginal children (p=0.06), males (p<0.01), and children within low income families (p<0.01). The majority of incidents occurred in residential areas (51.5%), with a speed limit of 50 kph or higher (85.7%). Risky pedestrian behaviour was a factor in 56.7% of cases, and 33% of children under 10 years of age were not under active supervision. Drivers had significantly more driving violations than the comparison population (p<0.01).

Conclusion Child pedestrian fatalities are highly preventable through the modification of behavioural, social, and environmental risk factors. This paper illustrates the ability of child death review to generate an ecological understanding of injury epidemiology not otherwise available and advance policy and programme interventions designed to reduce preventable child mortality.