Uncategorized

Injury rates per mile of travel for electric scooters versus motor vehicles

a b s t r a c t

Objective: This study determined the vehicle-miles-traveled (VMT)-based injury rate for stand-up, dockless elec- tric rental scooters (e-scooters), and compare it with the VMT-based injury rate for motor vehicle travel.

Methods: In this secondary analysis of existing data, the e-scooter injury rate was calculated based on E-scooter injuries presenting to an emergency department or the emergency medical services system in Austin, TX between September and November 2018. Injuries were identified by Austin Public Health through a targeted e-scooter epidemiological injury investigation; e-scooter VMT data were reported by e-scooter vendors as a condition of their city licensing. Comparative injury rates for motor vehicle travel in Texas, and specifically in Travis County were calculated using annual motor vehicle crash (MVC) injury and VMT data reported by the Texas Department of Transportation.

Results: There were 160 confirmed e-scooter injuries identified by the e-scooter injury investigation, with 891,121 reported miles of e-scooter travel during the study period. This produces an injury rate estimate of 180 injuries/million VMT (MVMT). The injury rates for motor vehicle travel for Texas and for Travis County were 0.9 injuries/MVMT and 1.0 injuries/MVMT, respectively.

Conclusion: The observed VMT-based e-scooter injury rate was approximately 175 to 200 times higher than statewide or county specific injury rates for motor vehicle travel. These findings raise concerns about the poten- tial higher injury rate associated with e-scooters, and highlight the need for further injury surveillance, research and prevention activities addressing this emerging transportation technology.

(C) 2020

  1. Introduction

Since late 2017, fleets of dockless stand-up electric rental scooters (“e-scooters”) have been deployed in cities across the United States (U.S.) and around the world. Distinct from mobility scooters or gas- powered Vespa-type scooters, these lightweight battery-operated two-wheeled vehicles are intended for very short commutes and are rented through a Smartphone application. Their rapid proliferation has been accompanied by several reports inventorying e-scooter- related injuries requiring emergency department (ED) care [1-9]. Al- though most patients injured by e-scooters are ultimately discharged from the ED [2-5], nearly all of them undergo imaging and as many as 25% require a surgical procedure [2-7]. Head injuries are a particular concern as helmet use is uncommon among e-scooter riders: [2-8] 28% to 60% of e-scooter injuries involve head injuries [1-5,8] with 2% to 8%

? Presented (virtually) at the Society for Academic Emergency Medicine, Denver, CO, May 2020.

* Corresponding author at: Division of Emergency Medicine, 1400 N. IH 35, Suite 2.230, Austin, TX 78701, USA.

E-mail address: [email protected] (L.H. Brown).

involving serious head injuries such as concussions, Skull fractures and intracranial hemorrhages [2-4].

Despite these reports of injuries, some have promoted e-scooters as an alternative to motor vehicle travel, arguing e-scooter riders reduce traffic congestion and pollution while avoiding the crash risks associ- ated with automobiles [10,11]. To date, however, no studies have com- pared the injury risk associated with e-scooter travel to the injury risk associated with motor vehicle travel using a common, standard mea- sure of exposure. The purpose of this analysis was to compare the vehicle-miles-traveled (VMT)-based e-scooter injury rate to the VMT- based injury rate for motor vehicle travel in one large U.S. city.

  1. Methods
    1. Study design

This was a secondary analysis utilizing existing, publicly reported e-scooter and motor vehicle crash (MVC) injury data. The Office of Research Support and Compliance at the University of Texas-Austin affirmed this analysis did not constitute human subjects research.

https://doi.org/10.1016/j.ajem.2020.10.048

0735-6757/(C) 2020

Table 1

Texas Department of Transportation classification of motor vehicle crash injuries [15]. Injury classification Definition

Fatal injury Any injury sustained in a motor vehicle traffic crash that

results in death within thirty days of the motor vehicle traffic crash.

  1. Results

APH identified 160 confirmed e-scooter injuries during their targeted investigation [12]. For the same time period, e-scooter vendors reported 891,121 miles of e-scooter travel [14]. Thus, the injury rate for e-scooter travel was 180 injuries/MVMT (Table 2).

Suspected serious injury

Non-incapacitating injury

Any injury, other than a fatal injury, which prevents the injured person from walking, driving or normally continuing the activities he was capable of performing before the injury occurred.

Any injury, other than a fatal or an incapacitating injury, which is evident to observers at the scene of the crash in which the injury occurred.

In 2018, there were 252,880 MVC-related injuries in Texas–includ- ing 160,448 “possible” injuries–and 282,038 MVMT on Texas highways and roadways [15,16]. Thus, the statewide injury rate for motor vehicle travel was 0.9 injuries/MVMT. There were 11,838 MVC-related injuries in Travis County in 2018, including 6,020 “possible” injuries [15]. There were 11,445 MVMT in Travis County that year [16]. Thus, the

Possible injury Any injury reported or claimed which is not a fatal,

incapacitating or non- incapacitating injury.

2.2. Setting and data sources

The e-scooter injury rate was calculated using injury data reported by Austin Public Health (APH) following a targeted e-scooter epidemio- logical injury investigation [12]. Located within Travis County, TX, Aus- tin has a population of approximately 950,000 [13] and (as of November 2019) seven vendors operating more than 17,000 e-scooters distributed over an 85 mile2 service area [14]. APH added e-scooter-related injuries to its list of syndromic surveillance conditions on September 5, 2018, and identified potential e-scooter injuries occurring through November 30, 2018 using chief complaint data from the city’s nine EDs and Local emergency medical services (EMS) system data. APH classified e-scooter injuries as “confirmed” when their investigators could verify the incident involved a “rentable, dockless electric scooter.” Other types of scooters (e.g., mobility scooters; gas-powered scooters; kick-scooters) were excluded [12].

As part of Austin’s licensing process, e-scooter vendors submit daily reports to the city’s transportation department which include the num- ber of trips and the distances traveled during each trip [14]. The number of confirmed e-scooter injuries and the e-scooter distances traveled data for September 5 through November 30, 2018 were used to calcu- late the VMT-based e-scooter injury rate as injuries per million VMT (MVMT) (i.e., (injuries/VMT) x1,000,000).

To calculate the injury rate for motor vehicle travel, both statewide and Travis County MVC fatality and injury counts for 2018 were ob- tained from the Texas Department of Transportation’s (TXDOT) “Texas Motor Vehicle Crash Statistics” publication [15]. TXDOT classifies MVC-related injuries as “fatal,” “suspected serious,” “non-incapacitat- ing” and “possible” (see Table 1) [15]. All reported MVC-related inju- ries–regardless of vehicle type and including “possible” injuries–were included in the analysis. Statewide and county-specific VMT data were obtained from the state’s “Roadway Inventory Annual Report.” [16] The injury rate for motor vehicle travel was also calculated as injuries per MVMT.

2.3. Data analysis

incident rate ratios with 95% confidence intervals comparing the in- jury rate for e-scooter travel to the statewide and county-specific injury rates for motor vehicle travel were calculated using Stata, Version 16.0 (StataCorp, College Station, TX).

county-specific injury rate for motor vehicle travel was 1.0 injuries/ MVMT.

The incident rate ratio was 200 (CI: 170-234) when comparing the e-scooter and statewide motor vehicle travel injury rates, and 174 (CI: 148-203) when comparing the e-scooter and county-specific motor ve- hicle travel injury rates.

  1. Discussion

In this large U.S. city, the VMT-based e-scooter injury rate in late 2018 was approximately 175 to 200 times the VMT-based injury rate for motor vehicle travel that year. These findings raise concerns about the potential higher injury rate associated with e-scooter travel. Despite the differences in these travel modalities, these data provide a frame of reference for the risks associated with e-scooter travel. Further injury surveillance and research is needed to guide prevention activities, regu- lations and engineering solutions to reduce injuries associated with this emerging transportation technology.

Emergency physicians play an important role as advocates for crash injury prevention, at both the policy level and through bedside educa- tion [17]. The potential to reduce crash-related injuries is well- evidenced by the experience with automobiles [18]. Injury rates for motor vehicle travel have decreased from 1.7 injuries/MVMT to 0.9 inju- ries/MVMT just since 1988 [19]. Fatality rates have decreased from 22 deaths per 100 MVMT in 1923 to less than 1 death per 100 MVMT in 2017 [20,21].

This analysis likely underestimates the e-scooter injury rate. The APH investigation did not capture persons with e-scooter injuries whodid not seek medical care or who sought care at urgent care or primary care facil- ities. Further, this analysis includes only “confirmed” e-scooter-related injuries. Had the 32 “probable” e-scooter injuries identified by the APH investigation [12] been included in the calculations, the e-scooter injury rate and incident rate ratios would be 20% higher. On the other hand, the analysis likely overestimates the injury rate for motor vehicle travel by including reported “possible” MVC-related injuries in the calcula- tions. That is, the rates and comparisons we report are biased in favor of e-scooters; the true differential in injury rates is likely even greater. One other city has publicly reported concurrent e-scooter injury counts and VMT data: Portland, OR reported 176 e-scooter-related emergency department (ED) and urgent care visits and 801,888 e-scooter VMT between July and November 2018, for a slightly higher

injury rate of 219 injuries/MVMT [11].

No e-scooter-related deaths were identified during the APH investi- gation. For context, however, the 2018 Travis County MVC-related fatal- ity rate was 10.2 deaths per billion VMT–or approximately 1 death per

Table 2

Estimated injury rates for e-scooter and motor vehicle travel, per million vehicle miles traveled.

Injuries, N

Vehicle miles traveled

Injury rate

Incident rate ratio (95% CI)

E-scooters (Austin, TX, Sep. – Nov. 2018) [12,14]

160

891,121

180/MVMT

N/A

Motor vehicles (Texas, 2018) [15,16]

252,880

282,037,481,031

0.9/MVMT

200 (170-234)

Motor vehicles (Travis County, TX, 2018) [15,16]

11,838

11,444,638,556

1.0/MVMT

174 (148-203)

MVMT = Million Vehicle Miles Traveled; CI = Confidence Interval.

100 MVMT [15,16]. Less than 9 million miles of e-scooter travel were re- ported in the first 24 months after vendors began operating in Austin [14], preventing any meaningful calculation of an e-scooter travel fatal- ity rate requiring hundreds of millions of VMT exposure.

We compared injury rates for e-scooter and motor vehicle travel be- cause some have suggested that e-scooter riders avoid the risks associ- ated with automobile travel [10,11]. Comparing e-scooter injury rates to those of other two-wheeled vehicles would also be informative. While standardized exposure data for motorcycle and bicycle travel are not collected or reported with the same consistency as for MVCs [22], the data that are available suggest VMT-based e-scooter injury rates also ex- ceed both motorcycle and bicycle VMT-based injury rates. Nationwide, the motorcycle injury rate has been estimated at 4.5 injuries/MVMT [21]. In Texas, the motorcycle injury rate has been estimated at 4.1 inju- ries/MVMT to 5.5 injuries/MVMT, although those rates are based on self-reported motorcycle VMT estimates gathered from household travel surveys [23]. Injury rates during organized cycling events vary be- tween 11.4 injuries/MVMT to 35.5 injuries/MVMT for injuries requiring ED care, and between 100 injuries/MVMT to 127 injuries/MVMT when including all injuries (e.g., muscle strains; minor abrasions) [24-26].

    1. Limitations

This analysis is limited to e-scooter injuries in one U.S. city. E-scooter injury rates could differ in other locations, although the injury rate we observed was similar to that reported in Portland, OR [11]. The injury rates for motor vehicle travel in our analysis are based on annual data, while the e-scooter injury rates are based on approximately three months of data. It is not possible to isolate the state-reported motor ve- hicle VMT and injury data for that three-month period, and there might be seasonal variability in the e-scooter injury rate or a longitudinal “learning curve” among e-scooter riders that reduces the e-scooter injury rate over time. However, any such seasonality or learning curve is unlikely to negate the nearly 200-fold observed difference be- tween e-scooter and motor vehicle travel injury rates. The e-scooter injuries in our analysis were identified from public health surveillance and medical data sources whereas the MVC injury counts come from law enforcement reports, and the use of these summary data pre- cluded any comparison of injury severity. Still, our approach of counting only confirmed e-scooter injuries that required ED evalua- tion while including all MVC-related injuries, including “possible inju- ries,” favors e-scooters and likely underestimates the true difference in injury rates–including serious injury rates. Without patient- or crash- specific data, we could not adjust for or identify risk factors (e.g., alcohol involvement, helmet use, roadway type) that might dif- fer for e-scooter versus motor vehicle travel.

  1. Conclusion

In this large U.S. city, the VMT-based injury rate for dockless stand- up rental e-scooters between September and November 2018 was ap- proximately 174 to 200 times the VMT-based injury rate for motor ve- hicle travel during that year. Emergency physicians can play an important role in advocating for further injury surveillance and re- search, prevention activities, regulations and engineering solutions ad- dressing this emerging transportation technology.

Funding

None.

Author contributions

KR, NJD, DFZ and LHB conceived and designed the study and col- lected the data. LHB conducted the analysis. LHB and NJD created the

original draft manuscript. KR, NJD, DFZ and LHB reviewed and edited the final manuscript.

Declaration of Competing Interest

None.

References

  1. Choron RL, Sakran JV. The integration of Electric scooters: useful technology or pub- lic health problem? Am J Public Health. 2019;109:555-6.
  2. Trivedi TK, Liu C, Antonio ALM, et al. Injuries associated with standing electric scooter use. JAMA Netw Open. 2019;2(1):e187381. https://doi.org/10.1001/ jamanetworkopen.2018.7381.
  3. Badeau A, Carman C, Newman M, Steenblik J, Carlson M, Madsen T. Emergency de- partment visits for electric scooter-related injuries after introduction of an urban rental program. Am J Emerg Med. 2019. https://doi.org/10.1016/j.ajem.2019.05.003.
  4. Bekhit MNZ, Le Fevre J, Bergin CJ. Regional Healthcare costs and burden of injury as- sociated with electric scooters. Injury. 2019. https://doi.org/10.1016/j.injury.2019. 10.026.
  5. English KC, Allen JR, Rix K, et al. The characteristics of dockless electric rental scooter-related injuries in a large U.S. city. Traff Inj Prev. 2020;21:476-81.
  6. Liew YK, Wee CPJ, Pek JH. New peril on our roads: a retrospective study of electric scooter-related injuries. Singap Med J. 2019. https://doi.org/10.11622/smedj. 2019083.
  7. Mayhew LJ, Bergin C. Impact of e-scooter injuries on emergency department imag- ing. J Med Imag Radiat On. 2019. https://doi.org/10.1111/1754-9485.12889.
  8. Schlaff CD, Sack KD, Elliott RJ, Rosner MK. Early experience with electric scooter in- juries requiring neurosurgical evaluation in District of Columbia: a case series. World Neurosurg. 2019. https://doi.org/10.1016/j.wneu.2019.08.237.
  9. Brownson AB, Fagan PV, Dickson S, Civil ID. Electric scooter injuries at Auckland City Hospital. NZ Med J. 2019;132:62-72.
  10. Tillemann L, Feasley L. Let’s count the ways e-scooters could save the city. WIRED. December 7, 2018 Available from: https://www.wired.com/story/e-scooter- micromobility-infographics-cost-emissions/.
  11. Portland Bureau of Transportation. E-scooter findings report. Portland, OR: Portland Bureau of Transportation; 2018 Available from: https://www.portlandoregon.gov/ transportation/article/709719.
  12. Dockless electric scooter-related injuries study. Austin, TX: Austin Public Health; 2019 Available from: http://www.austintexas.gov/sites/default/files/files/Health/ Web_Dockless_Electric_Scooter-Related_Injury_Study_final_version_EDSU_5.14.19. pdf.
  13. U.S. Census. Annual estimates of the resident population for the United States, Regions, States, and Puerto Rico: April 1, 2010 to July 1, 2018 (NST-EST2018-01). Avail- able from: https://www.census.gov/newsroom/press-kits/2018/pop-estimates- national-state.html; 2019.
  14. Shared micromobility transportation data and performance hub. Austin, TX: City of Austin Transportation Department; 2019 Available from: https://data.mobility. austin.gov/micromobility-data/.
  15. Texas motor vehicle crash statistics – 2018. Austin, TX: Texas Department of Trans- portation; 2019 Available from: https://www.txdot.gov/government/enforcement/ annual-summary.html.
  16. Roadway inventory annual reports – 2018. Austin, TX: Texas Department of Trans- portation; 2018 Available from: https://www.txdot.gov/inside-txdot/division/ transportation-planning/roadway-inventory.html.
  17. Peterson TD, Jolly BT, Runge JW, Hunt RC. Motor vehicle safety: current concepts and challenges for emergency physicians. Ann Emerg Med. 1999;34:384-93.
  18. Anonymous. Achievements in public health, 1900-1999 motor-vehicle safety: a 20th century public health achievement. MMWR. 1999;48(18):369-74.
  19. National Highway Traffic Safety Administration. Table 2: Persons killed or injured and fatality and injury rates per population, licensed drivers, registered vehicles, and vehicle miles traveled, 1966-2017. National safety facts annual report tables: trends (general); 2019 Available from: https://cdan.nhtsa.gov/tsftables/tsfar.htm#. Accessed December 3, 2019.
  20. National Safety Council. Historical fatality trends. Available at: https://injuryfacts. nsc.org/motor-vehicle/historical-fatality-trends/deaths-and-rates/; 2019. Accessed

November 16, 2019.

  1. National Highway Traffic Safety Administration. Traffic safety facts 2017. Washington, DC: U.S. Department of Transportation; 2019.
  2. Beck LF, Dellinger AM, O’Neil ME. Motor vehicle crash injury rates by mode of travel, United States: using exposure-based methods to quantify differences. Am J Epidemiol. 2007;166:212-8.
  3. Shipp EM, Trueblood AB, Perez M, et al. Analysis of motorcycle crashes in Texas, 2010-2017. Texas A&M Transport Institute: College Station, TX; 2018.
  4. Emond SD, Tayoun P, Bedolla JP, Camargo CA. Injuries in a 1-day recreational cycling tour: bike New York. Ann Emerg Med. 1999;33:56-61.
  5. Yanturali S, Canacik O, Karsli E, Suner S. Injury and illness among athletes during a multi-day elite cycling road race. Phys Sportsmed. 2015;43:348-54.
  6. Pommering TL, Manos DC, Singichetti B, Brown CR, Yang J. injuries and illnesses oc- curring on a recreational bicycle tour: the great Ohio bicycle adventure. Wild Envi- ron Med. 2017;28:299-306.