Does Geofencing Reduce E-Scooter Reported Injuries? A Casual Mediation Analysis of Austin, Texas

Dockless e-scooter rentals have surged in popularity post-COVID-19 restrictions, resulting in a spike in hospital visits due to falls and collisions. Despite this trend, the effectiveness of measures like geofencing in reducing crash rates remains uncertain. This paper aims to answer the questions of: what’re the characteristics of e-scooter crashes and is geofencing effective at lowering crash rates? To address this problem statement, researchers present a multi-faceted and comprehensive analysis of e-scooter trip and injury data at The University of Texas at Austin (UT Austin) campus from 2018 to 2022. Findings reveal that the majority of crashes (55%) occurred in 2019, primarily attributed to riders falling off (70%) due to excessive speeds (50% of users falling off), resulting in abrasions or head injuries. Furthermore, to assess the impact of geofencing technologies and the COVID-19 pandemic on e-scooter crash rates, three causal mediation analyses were conducted. The results indicate that parking restrictive geofences and the pandemic indirectly contribute to a decrease in reported e-scooter crashes. Parking geofences accounted for 50% of the mediation, while the pandemic contributed 30%. ‘Indirectly’ can be interpreted as the parking restrictive geofences preventing e-scooter riders’ exposure to injuries. These findings suggest that implementing parking restriction geofences could mitigate e-scooter injuries in high-risk collision scenarios, such as areas with heavy pedestrian or vehicular traffic. Further research is highly recommended to evaluate the effectiveness of speed and parking geofences in reducing crash rates.