Understanding geographical variations in road traffic fatalities in South Africa

Abstract

South Africa (SA) faces an unprecedented burden of injuries from road traffic crashes, yet the distribution of these events has not yet been studied using a geographical approach in order to help understand the importance of putative social and environment drivers. Such an approach was used in this study to investigate the correlates of spatial variations in road traffic fatalities (RTFs) in SA. Variations in RTFs between 2002 and 2006 were studied for 993 police areas. A wide range of explanatory variables comprising physical, environmental and socio-demographic characteristics were generated, and multilevel negative binomial regression models were fitted to identify those associated with RTFs. An area measure of violence and crime was shown to be a significant predictor of RTFs in SA in addition to a range of factors associated with driver behaviour, traffic exposure and socio-economic deprivation. Our research provides new insights into the correlates of road traffic mortality in this less developed country, and our findings have implications for the development of integrated resource-efficient strategies that allow for enforcement and other broader structural interventions to target injuries and crime in general.

Keywords::

• South Africa
• road traffic fatalities
• geographical
• multilevel
• negative binomial

Acknowledgements

The National Department of Transport is gratefully acknowledged for providing the traffic fatality data for this study.

Notes

^1. Email: Email: [email protected]