LAUSR

Abstract Objective: The present case-control study sought to explore at-risk riding behaviors associated with e-bike related traffic crashes among e-bike riders in China. Methods: Cases were recruited from residents aged 16 years and over in communities which stated “selected e-bikes as travel tools and experienced traffic crashes in the last year”. Two controls for each case were randomly selected from a population of e-bike riders who had not experienced a traffic crash in the past year. The cases and controls were matched by gender, age (within 5 years) and school education level. Data were collected using questionnaires and face-to-face interviews from July 2015 to September 2015 in China. After conducting univariate logistic analysis on study variables, a conditional logistic regression model based on the 1:2 matched case-control study design was developed. Results: Multiple-factor conditional logistic regression analysis of e-bike related traffic crashes showed that running red lights (always vs. never, AOR = 3.094, 95% CI, 1.077-8.891, P < .05), riding after drinking (yes vs. no, AOR = 1.578, 95% CI, 1.102-2.259, P < .05), carrying adults while riding (always vs. never, AOR = 2.140, 95% CI, 1.273-3.595, P < .05), turning without signaling (sometimes vs. never, AOR = 1.446, 95% CI, 1.805-1.928, P < .05), riding in the motor vehicle lane (always vs. never, AOR = 2.413, 95% CI, 1.576-3.695, P < .01), prior crash history (yes vs. no, AOR = 1.670, 95% CI, 1.257-2.220, P < .05), and type of e-bikes (scooter-style e-bikes vs. bicycle-style e-bikes, AOR = 1.471, 95% CI, 1.068-2.026, P < .05) were identified as possible risk factors for e-bike traffic crashes. Conclusion: The findings of this research provide evidence about specific risky behaviors related to road traffic crashes involving e-bikes and indicated that behavioral intervention and education need to be strengthened to reduce dangerous riding behaviors. These results will be helpful for design of e-bike road risk prevention programs.