Latent analysis of Complete Streets and traffic safety along an urban corridor. (March 2018)

• Record Type: Journal Article
• Authors: MacLeod, Kara E.; Sanders, Rebecca L.; Griffin, Ashleigh; Cooper, Jill F.; Ragland, David R.
• Is Part Of: Journal of transport & health. Volume 8(2018)
• Page Start: 15
• Page End: 29
• Publication Date: 2018-03
• Subjects: Complete Streets -- pedestrian safety -- traffic safety -- typologies -- vehicle conflict -- Vision Zero; Transportation -- Health aspects -- Periodicals; Transportation -- Periodicals; Public Health -- Periodicals; Noise, Transportation -- Periodicals; Air Pollutants -- Periodicals; 388
• DOI: 10.1016/j.jth.2017.05.001
• Languages: English
• ISSNs: 2214-1405

Abstract

Abstract: Background: To evaluate Complete Street implementations that covary, the present paper aims to: 1) explore the development of typologies of intersections; and 2) examine how these typologies relate to traffic safety. Methods: The study site is a five-mile segment in Los Angeles County, California. Multiple indicators of environmental features were collected in 2012 and were included in a latent analysis. Latent classes were then analyzed as a predictor of the number of pedestrian injuries/fatalities and injuries/fatalities for all modes in separate models using negative binomial regression and controlling for exposures. Injuries/fatalities for a 6-year period were used (2009-2014), representing the most recent crash data available surrounding the environmental data collection time point. We also examined the role of alcohol. Results: For a relatively short segment of an urban corridor, we identified two distinct classes of intersections. One class was more complete with respect to pedestrian features but was also associated with indicators of increased potential conflict and was predictive of higher overall injuries/fatalities for all modes. This class also had higher pedestrian volumes but was not predictive of higher pedestrian injuries/fatalities in the final models. The alcohol involvement in crash injuries at these locations did not differ by intersection class but was positively associated with injuries/fatalities for all modes and with severe/fatal injuries for pedestrians in the final models. Conclusions: Typologies can be used to understand the combination of features and to prioritize locations for treatment. While Complete Streets may help counter pedestrian injury trends, the efforts captured in this data are insufficient for municipalities aiming for Vision Zero. Ideally, future research can examine these intersections after the implementation of additional improvements in order to isolate treatment effects. These findings suggest additional intersection countermeasures are needed, in addition to efforts to address social problems such as alcohol use and traffic safety. Highlights: We conducted a cross-sectional environmental audit of an urban corridor in LA. Based on this audit, we identified two distinct classes of intersections. Complete intersections had greater pedestrian volumes and sources of conflict. Complete intersections also had more injuries for all modes but not for pedestrians. Complete Streets may help counter some injury trends but more work is needed.