Aim Factors previously associated with the risk of unintentional injury in primary school aged children include a range of child and family variables. Risk factors in the environment have been seldom studied in this age group and their relative contribution to overall injury risk is unclear.

Methods Avon longitudinal study of pregnancy and childhood is a large, population-based longitudinal birth cohort study in Avon, England. It recruited mothers during their pregnancy and 14 602 children in the cohort were born in 1991–1992. We investigated the association of parentally reported injury data from 5752 children collected at 5, 6, 8 and 11 years of age with a range of potential risk factors using multivariable analysis with multiple imputation of missing data.

Results Using a hierarchical conceptual framework for modelling childhood injury risk, a greater number of factors related to the child's immediate environment (eg, home) were associated with an increased risk of injury than factors in the wider environment (eg, neighbourhood). Environmental factors showed weaker evidence against the null hypothesis of no association with injury than family or child factors. The relative contribution of environmental factors varied with age of the child. Comparisons with previous research will be made.

Conclusions This research provides robust evidence of the relative contribution of risk factors for unintentional injuries in primary school aged children in the UK. This is of value to inform the development of targeted interventions to prevent unintentional injuries in these children.