Background Injury events in homes constitute a major social problem. Falling, cutting and poisoning make up 85 per cent of all injury events in residential areas.

This study is based on a unique data set that includes several million cases of falling, cutting and poisoning in Sweden during the period 1990-2013 which lead to either to death, hospitalisation or to a visit to a health care provider, and a control group randomly selected from the population.

Three riskgroups are given special attention in the analysis: (1) old people, (2) children, (3) persons with disabilities and or long term illnesses.

Methods Multilevel regression analysis and geographical information systems, GIS.

Results The results show the probability for each riskgroup to be exposed to each type of injury event, and how this probability varies with place (GIS), previous exposure, type of household, socioeconomic status and type of housing.

Conclusions The project is ongoing. Our cross-sectorial group has demonstrated the importance of injury epidemiology as a guiding principle in architectural design, particularly for high-risk groups.