The impact of implementing a public bicycle share program on the likelihood of collisions and near misses in Montreal, Canada Daniel Fuller a, ⁎, Lise Gauvin b , Patrick Morency c , Yan Kestens b , Louis Drouin c a Department of Community Health and Epidemiology, University of Saskatchewan, Health Sciences Building, 107 Wiggins Road, Saskatoon, SK S7N 5E5, Canada b Centre de recherche du Centre Hospitalier de l'Université de Montréal & Département de médecine sociale et préventive, Université de Montréal, C.P. 6128, Succursale Centre-ville, Montréal, QC H3C 3J7, Canada c Direction de santé publique de Montréal & Département de médecine sociale et préventive, Université de Montréal, 1301, rue Sherbrooke Est, Montréal, QC H2L 1M3, Canada abstract article info Available online 6 June 2013 Keywords: Cycling Injuries Collisions Population health Intervention Objective. This study aimed to estimate the impact of implementing a public bicycle share program (PBSP) on the likelihood of self-reported collisions and near misses between cyclists and motor vehicles among cyclists liv- ing in Montreal. Methods. A repeated cross sectional design was used. Surveys were conducted at the launch of the PBSP, at the end of the first and second seasons of implementation. Logistic regression estimated changes in the likeli- hood of reporting collisions or near misses. Results. There was no evidence of a change in likelihood of reporting a collision or near miss after implementing the PBSP. PBSP users were not at a greater risk of reporting a collision (OR = 1.53, 95% CI: 0.77–3.02) or near miss (OR = 1.37, 95% CI: 0.94–1.98), although confidence intervals were wide. The num- ber of days of cycling per week was associated with collisions (OR = 1.27, 95% CI: 1.17–1.39) and near misses (OR = 1.34, 95% CI: 1.26–1.42). Conclusions. There was no evidence of a change in the likelihood of reporting collisions or near misses in Montreal between the implementation of the PBSP and the end of the second season. Time spent cycling was associated with reporting a collision or near miss. © 2013 Elsevier Inc. All rights reserved. Public health researchers are increasingly interested in the relation- ship between transportation and health (Saelens et al., 2003; Sallis et al., 2006). One of the objectives of public health intervention in transpor- tation is to promote safe and sustainable increases in cycling for trans- portation in the population. Although many might suggest that an increase in the number of cyclists will likely result in a commensurate increase in the number of collisions, the safety in numbers (SIN) hy- pothesis has made an important contribution to thinking about this rela- tionship. The SIN hypothesis suggests an inverse though non-linear relationship between the prevalence of cycling and the cyclist's risk of injury due to collisions with motor vehicles (Dill and Carr, 2003; Elvik, 2009; Jacobsen, 2003; Robinson, 2005). Jacobsen (2003) shows that the safety in numbers relationship as a power curve of 0.4 and provides the example that a doubling of the prevalence of walking or cycling in a community would result in a 32% increase in the number of inju- ries (2 0.4 = 1.32), rather than a doubling of injuries and would thus re- sult in a 66% reduction in the individual risk of injury, for each cyclist (2 0.4 /2 = 2 -0.6 = 0.66). Despite the appeal of the safety in numbers hypothesis in cycling advocacy and policy making circles, there is no scientific consensus regarding the validity of the hypothesis (Bhatia and Wier, 2011). Of relevance in the present study are the temporal sequence with which changes in prevalence of cycling and likelihood of injuries occur and the potential mechanisms that underlie the SIN hypothesis. Jacobsen (2003) suggests that “it is improbable that the roadway de- sign, traffic laws or social mores, all of which change relatively slowly, can explain the relationship between exposure [to motor vehicles] and injury rates” (p. 208). Natural experiments, which can occur quickly, can provide insight into whether environmental changes could explain the safety in numbers hypothesis. Public bicycle share programs (PBSP) offer such a natural experi- ment. Widely implemented in Western Europe and China and increas- ingly popular in North America, PBSPs increase population access to cycling by making bicycles available at docking stations throughout an area within a city for a fee (Pucher et al., 2009; Shaheen et al., 2010). For example, Montreal's bicycle-taxi (BIXI) program, North America's largest in 2011, launched in May 2009 makes available 5050 bicycles at 405 docking stations. Bicycles are available for a check out fee of $7 for 24 h or $78 for the season. After paying the checkout fee, the first 45 min of usage is free. Users extending their usage beyond 45 min pay a usage fee of approximately $1.50 per 45 min. The implementation of a PBSP allows for the evaluation of whether implementation and use of the program are associated with increases Preventive Medicine 57 (2013) 920–924 ⁎ Corresponding author. E-mail address: dlf545@mail.usask.ca (D. Fuller). 0091-7435/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ypmed.2013.05.028 Contents lists available at ScienceDirect Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed