Injuries caused by accidents are a particular problem in young children and children's injury rates vary considerably from place to place. Low-income neighbourhoods have higher child injury rates than high-income neighbourhoods, and recent research suggests that only part of the difference is due to variations in the social, economic and demographic composition of local populations. This research aims to find out the relative contributions of family characteristics and neighbourhood features to the risk of injury to children aged 0-4 years in Bristol and its surrounding region, to discover some of the mechanisms that create patterns of inequality. The objectives are:

1) to determine the risk factors acting at the level of the child, the household and the neighbourhood, and whether these risks act independently.

2) to determine whether residents' perceptions of neighbourhood and local social networks are related to geographical variations in the risk of injury.

3) to determine the size of effects and interactions between risk factors, and identify modifiable risk factors that might be subject to intervention.

4) to identify the scale (size of neighbourhoods) at which any effects of place are most evident.