Background Ridesharing companies (such as Uber, Lyft, and DiDi) have facilitated over 20 billion trips globally between paying passengers and owner/operator drivers. Given that ridesharing has changed patterns of human mobility, it could also affect injury incidence.

Method We conducted a series of longitudinal spatial analyses using trip-level rideshare data in Chicago, IL, and New York, NY, in the United States. Outcomes of interest were injury events that ridesharing could theoretically affect: bar-area assaults, alcohol-involved motor vehicle crashes, pedestrian vs. motor vehicle crashes, and bicyclist vs. motor vehicle crashes. The space-time units of analysis were small areas around injury event locations, and the exposure of interest was counts of rideshare trips that occurred in those units. Multivariable logistic regression models accounted for the space-time data structure and time-varying confounders, including taxi trip volume, temperature and precipitation.

Results An increase of 100 rideshare trips per hour was associated with increased incidence of bar-area assaults (OR=1.050, 95%CI:1.002,1.100), increased incidence of pedestrian vs. motor vehicle crashes (OR=1.061, 95%CI:1.035,1.087), and decreased incidence of alcohol-involved motor vehicle crashes (OR=0.819, 95%CI:0.670,0.905). There was no association between ridesharing and bicyclist vs. motor vehicle crashes.

Conclusions Ridesharing has mixed effects on injury outcomes. Public health benefits due to reduced incidence of alcohol-involved crashes may be offset by increases in other injury events.

Learning Outcomes

1. To understand that ridesharing has mixed effects on injury outcomes.

2. To critique space-time epidemiologic methods to examine injury outcomes.

3. To learn opportunities for rideshare-based interventions to reduce injury incidence.