Are Barcelona’s Advanced Zones for Motorcycles Effective?

Abstract

Background In 2009-2010 advanced zones for motorcycles (ZAM) were implemented. The objective is to assess the impact of ZAMs on road traffic collisions Methods This is a quasi-experimental design of an intervention (ZAM) evaluation study with comparison group. The study population is people travelling through Barcelona. The ZAM is a grid of 4 meters length painted before the crosswalk. 35 ZAMs were implemented in 2009 (Phase I) and 16 in 2010 (Phase II). For each ZAM, a nearby zone of comparison is chosen. Two areas are defined: ZAM area and 30 meters before (ZAM34); crosswalk and intersection (CROSS). Data on crashes are provided by the police and includes geocodes. It allows to identify those occurring in the two defined areas. Dependent variables are total number of collisions, collisions involving motorcycle, people injured and motorcycle drivers involved. The study periods are: Phase I, 2002-2009 pre-intervention and 2010-2014 post; Phase II, 2002-2010 pre and 2011-2014 post. Relative risk (RR) and its 95% Confidence Intervals, in the post with respect to the pre, in both areas, are estimated using Poisson-lognormal regression models, with a random zone factor, for the ZAMs and for the comparison zones, adjusting for mobility and road characteristics Results In Phase I, in ZAMs, annual means of people injured in the ZAM34 area were: 50 pre and 63 post, and in the CROSS area 234 and 197. Adjusting for mobility and road characteristics, in the ZAM34 area of ZAMs, the risk of a collision increased significantly (RR = 1.34 [1.12-1.61]), motorcycle involvement (RR = 1.44 [1.17-1.78]), people injured (RR = 1.26 [1.01-1.58]) and being motorcycle driver involved (RR = 1.39 [1.10-1.75]). In the comparison areas, only the risk of injury increased (RR = 1.42 [1.09-1.85]). In the CROSS area of ZAMs, there were no significant changes, but in the comparison areas the risk of injury (RR = 0.86 [0.75-0.98]) and being motorcycle driver involved (RR = 0.85 [0.74-0.98]) decreased. In phase II, in ZAMs, annual means of people injured in the ZAM34 area were: 30 and 35 in pre and post, and in the CROSS 91 and 87. Adjusting for mobility and road characteristics, there were no significant changes in risk, neither in ZAMs nor in the comparison areas Conclusions Advanced zones for motorcycles are not effective in improving road safety. In some circumstances, they increase the risk of traffic injury in the ZAM area and the previous 30 meters.