Risks of Bicyclist Accident Involvement

Monash University Accident Research Centre - Report #2 - 1988

Full report in .pdf format [890KB]

Authors: A. Drummond & F. Jee

Abstract:

This report details the method and results of a study to investigate the relative safety of cycling on the road and footpath and of a variety of cycling behaviours.

A method was developed for the collection of cyclist exposure (duration of travel) and behavioural (including helmet wearing) information in 105 randomly selected observation zones in metropolitan Melbourne. Additional work was also undertaken in the area of accident data, comprising accident typology coding and an investigation of underreporting issues.

Results were provided on exposure patterns, accident involvement risk estimates and helmet wearing rates. A general interpretation of the comprehensive set of results presented in the report indicates that the safety benefits of allowing footpath cycling along arterial roads would be much greater than in the non-arterial environment. Significant improvements in the safety of footpath cycling could be accomplished if effective strategies to reduce the incidence of transitional, rideout cycling could be implemented.

Executive Summary

This report details the method and results of a study conducted by the Monash University Accident Research Centre to investigate the relative safety of cycling on the road and footpath and of a variety of different cycling behaviours. The study was commissioned by the Road Traffic Authority and the State Bicycle Committee.

The study was initiated to provide an empirical basis for policy decisions on the possible legalisation of footpath cycling for specific cyclist groups and /or specific locations. At the same time, additional information was collected which would enable the risk of various behaviours to be estimated, thus providing the capacity for further improvements in cycling safety.

A method was developed for the collection of cyclist exposure (duration of travel) and behavioural (including helmet wearing) information in 105 randomly selected observation zones in metropolitan Melbourne. These sample data were then scaled up to the entire study area for all non-holiday periods of the year, thus providing the necessary information for the denominator of the risk ratio (an accident set divided by its comparable exposure set).

Additional work was also undertaken in the area of accident data. Firstly, to enable accident involvement risk estimates for the various behavioural classes to be calculated, typology coding was conducted (classifying accidents into accident type categories on the basis of accident report narratives and diagrams). Secondly, to counter underreporting problems, accident information was also collected directly from the hospital system. Analysis of these data demonstrated that the road/footpath risk ratio was similar for both Police reported and hospital sourced data (around 2.5 : 1) This indicates that the relative risks presented in this report are valid measures of cycling safety.

Results were provided on exposure patterns, accident involvement risk estimates of both road and footpath cycling and their interactions, accident involvement risk estimates of selected behavioural components and helmet wearing rates. Illustrations of the basic results are given in Figures A and B. Figure A presents the risks of accident involvement (per 100,000 hours of cycling) by road and footpath cycling overall and by road class. It shows that road cycling is a much riskier activity by a factor of 2.6 overall. On road cycling on arterial roads is three times more dangerous than cycling on the footpath in these locations; the same ratio applies on the non-arterial network.

Sponsor: Road Traffic Authority and State Bicycle Committee