Micromobility Products-Related
Deaths, Injuries, and Hazard
Patterns: 2017–2021
September 2022
James Tark
Directorate for Epidemiology
Division of Hazard Analysis
U.S. Consumer Product Safety Commission
4330 East West Highway
Bethesda, MD 20814
This report was prepared by the CPSC staff.
It has not been reviewed or approved by,
and may not necessarily reflect the views of,
the Commission.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
2
Table of Contents
Executive Summary ................................................................................................................. 4
Emergency Department (ED) - Treated Injury Estimates ........................................................ 4
Special Study on E-Scooters .................................................................................................. 5
Reported Fatalities ................................................................................................................. 5
Associated Hazard Patterns ................................................................................................... 6
Introduction .............................................................................................................................. 7
I. National Injury Estimates ..................................................................................................... 8
Figure 1.1: Estimated ED Visits Associated with Micromobility Products by Year ................... 9
Figure 1.2: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Sex (2017–2021 Total) ...................................................................................................10
Figure 1.3: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Age Group (2017–2021 Total) Compared to U.S. Population Age Distribution ...............11
Figure 1.4: Distribution of Estimated ED Visits Associated with Micromobility Product Type
and Location of Injury (2017–2021 Total) ..............................................................................12
Figure 1.5: Distribution of Estimated ED Visits Associated with Micromobility Product Type
and Month of Injury (2017–2021 Total) ..................................................................................13
Figure 1.6: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Race/Ethnicity, where known
+
, (2019–2021 Total) Compared to U.S. Population
Race/Ethnicity Distribution .....................................................................................................14
II. Special Study on E-Scooters..............................................................................................15
Table 2.1: Scooter Products: NEISS Product Code vs. Verified Scooter Product (2021) .......16
Table 2.2: Completed E-Scooter Investigations by Powered vs. Unspecified Scooter (2021) 17
Table 2.3: Distribution of Estimated Injuries by Rental Status (2021) .....................................17
Table 2.4: Distribution of Estimated Injuries by How Victim Was Injured (2021) ....................18
Table 2.5: Distribution of Estimated Injuries by Type of Riding Surface (2021) ......................18
Table 2.6: Distribution of Estimated Injuries by Visibility (2021) .............................................18
Table 2.7: Distribution of Estimated Injuries by Distraction (2021) .........................................19
Table 2.8: Distribution of Estimated Injuries by Rider Baggage (2021) ..................................19
Table 2.9: Distribution of Estimated Injuries by Safety Equipment (2021) ..............................20
III. Reported Fatalities Associated with Micromobility Products .........................................20
Table 3.1: Number of Reported Fatalities Associated with Micromobility Product Type and
Year ......................................................................................................................................20
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
3
Table 3.2: Number of Reported Fatalities Associated with Micromobility Product Type and
Gender (2017–2021 Total) ....................................................................................................21
Table 3.3: Reported Fatalities Associated with Micromobility Product Type and Age Group
(2017–2021 Total) .................................................................................................................21
Table 3.4: Reported Fatalities Associated with Micromobility Product Type and Associated
Hazards (2017–2021 Total) ...................................................................................................23
IV. Hazard Patterns Based on In-Depth Investigation Review .............................................23
Table 4.1: Reported Fatalities Associated with Micromobility Product Type and Associated
Hazards (2017–2021 Total) ...................................................................................................25
References ..............................................................................................................................26
Appendix A ..............................................................................................................................27
Methodology ..........................................................................................................................27
Appendix B ..............................................................................................................................29
Summary of Annual Injury Estimates and Trend Analysis ......................................................29
Appendix C ..............................................................................................................................31
Revised 2020 Special Study Results .....................................................................................31
Special Study Survey Questionnaire .....................................................................................34
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
4
Executive Summary
In this report, U.S. Consumer Product Safety Commission (CPSC) staff presents the latest
available statistics on injury estimates, fatalities, and hazard patterns associated with three
micromobility products: e-scooters (including dockless/rental e-scooters), hoverboards, and e-
bikes. The timeframe covered is 2017 through 2021. This report includes all data presented in
the previous annual report, Micromobility Products-Related Deaths Injuries and Hazard Patterns
2017-2020
1
and adds the 2021 data and the findings from a 2021 special study. For
micromobility-related fatalities, staff notes that, due to delays in death certificate reporting, the
number of reported fatalities may change in the future. CPSC staff conducted a special follow-
up study of e-scooter-related, emergency department (ED)-treated injuries between January 1,
2021, and December 31, 2021. The special study allowed the collection and analysis of more
detailed information on these injuries, such as the type of injury, the characteristics of the victim,
and the incident scenario. Staff found that the proportion of estimated injuries sustained by
Black consumers (among data with known race/ethnicity, 76% of ED visits from 2019-2021) was
31 percent for overall micromobility and 37 percent for e-scooters, which are substantially higher
proportions than the 13 percent Black population nationwide; however, staff notes some data
limitations to that comparison.
Emergency Department (ED) - Treated Injury Estimates
• All Micromobility Products
o Estimated total of 267,700 ED visits from 2017 through 2021;
o Estimated ED visits associated with micromobility increased from 34,000 in 2017
to 77,200 in 2021, which is statistically significant (p-value < 0.01).
• E-Scooters
o Estimated total of 117,600 ED visits from 2017 through 2021;
o Estimated ED visits associated with e-scooters increased from 7,700 in 2017 to
42,200 in 2021, which is statistically significant (p-value < 0.01);
o 2021 ED-treated injury estimate of 42,200 for e-scooters reflects an increase of
66 percent from the 2020 estimate, which is statistically significant (p-value <
0.01);
o Estimates for dockless/rental e-scooters did not meet the reporting criteria for
NEISS. These estimates accounted for 28 percent of ED visits for e-scooters.
2
However, this may be an underestimate because not all dockless/rental e-
scooters may have been identified as dockless/rental in the NEISS data by the
hospital staff.
3
1
The report, Micromobility Products-Related Deaths Injuries and Hazard Patterns 2017-2020 can be found at:
https://www.cpsc.gov/s3fs-
public/Micromobility-Products-Related-Deaths-Injuries-and-Hazard-Patterns-2017-
2020.pdf?VersionId=s8MfDNAVvHasSbqotb7UC.OCWYDcqena
2
The estimate for dockless/rental e-scooters based on NEISS data (2018-2021) and special studies (2020-2021).
3
The reporting criteria for NEISS require that the estimated number of injuries be 1,200 or higher, the sample size be
20 or larger, and the coefficient of variation be less than 33 percent.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
5
• Hoverboards
o Estimated total of 121,300 ED visits from 2017 through 2021;
o 2021 ED-treated injury estimate of 23,100 for hoverboards reflects a decrease of
15 percent from the 2020 estimate, which is statistically significant (p-value:
0.04).
• E-Bikes
o Accounted for 11 percent of the overall micromobility injury estimate from 2017
through 2021;
o Estimates did not meet the reporting criteria for NEISS.
3
Special Study on E-Scooters
• CPSC staff successfully followed up 307 e-scooter NEISS injury cases through investigations
for 2021.
• Rental e-scooters accounted for 44 percent of the e-scooter-related ED visits in the special
study.
• Seventy-four percent of the injuries occurred on paved roads;
• Twenty-one percent of the victims reported that it was dark or difficult to see.
• Twenty-six percent of the injured were carrying or holding something while riding the e-
scooter.
• Fifty-two percent of the riders were wearing a helmet while riding the e-scooter.
Reported Fatalities
4
• All Micromobility Products
o CPSC staff is aware of 129 fatalities from 2017 through 2021.
o The number of fatalities has been increasing steadily from 5 in 2017, to 48 in
2021.
• E-Scooters
o CPSC staff is aware of 68 (14 dockless/rental) fatalities from 2017 through
2021.
5
o Motor vehicle accident and user-control issues were top hazards associated with
e-scooter fatalities.
• Hoverboards
o CPSC staff is aware of 8 fatalities from 2017 through 2021.
o Four fatalities were associated with hoverboard-related fires.
• E-Bikes
o CPSC staff is aware of 53 fatalities from 2017 through 2021.
4
Reporting for 2020-2021 is ongoing. Counts may change in future reports.
5
Fatality reports associated with dockless e-scooters began to appear in the CPSC surveillance data in 2018.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
6
o Motor vehicle accident and user-control issues were top hazards associated with
e-bike fatalities.
Associated Hazard Patterns
CPSC Field staff completed 207 follow-up in-depth investigations related to all micromobility
products, based on reports of incidents in CPSC’s Consumer Product Safety Risk Management
System (CPSRMS) that occurred from 2017 through 2021. Of the 207 completed investigations,
48 involved an e-scooter, 148 involved a hoverboard, and 11 involved an e-bike.
• E-Scooters
o Forty-eight of the investigated incidents involved an e-scooter (32 of the 48 were
dockless/rental e-scooters).
o Brake problems were associated with more of the investigated incidents (18 out
of 48) than any other problem.
• Hoverboards
o Field staff investigated 148 incidents associated with hoverboards.
o Fire hazards were the most common problem reported, accounting for 118 of the
148 investigated incidents.
• E-Bikes
o CPSC staff reviewed 11 completed investigation reports on e-bikes.
o Fire hazards and brake problems were identified in 8 of the 11 investigated
incidents.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
7
Introduction
The use of micromobility products, including e-scooters, hoverboards, and e-bikes, has
increased in recent years with advancements in battery technology and the growing popularity
of commercial ride-sharing services. Consumers may rent commercial, dockless e-scooters and
e-bikes or purchase their own micromobility products. These products are popular with
consumers because they are perceived as eco-friendly, given that they have no tailpipe
emissions, and they are a convenient, cost-effective mode of transportation for short-distance
travel.
6
In addition, no motor vehicle license is required to operate these products, which makes
them very accessible to everyone.
This report summarizes the injuries, deaths, and hazards associated with the use of
micromobility products, based on data from the CPSC epidemiological databases from 2017
through 2021. Data from 2021 were added to the data from 2017 through 2020 previously
presented in the 2021 annual report on micromobility products.
7
In addition, this report
summarizes findings from a special follow-up study of emergency department-treated injuries
involving e-scooters in 2021. The micromobility products covered in this report are:
• electric scooters (e-scooters: electric-powered, motorized standing scooters), including ride-
sharing dockless/rental e-scooters;
• hoverboards (also referred to as self-balancing e-scooters that are electric-powered, two-
wheeled standing scooters with no handlebars); and
• electric bicycles (e-bikes: motorized bicycles powered by battery to assist riders’ pedal-power,
with a maximum speed of 28 mph).
Electric unicycles, three-wheeled e-scooters, non-electric kick scooters, gas-powered scooters,
mobility scooters, mopeds, motorized carts, and other seated motorized scooters are not in
scope for this report.
The first section of this report presents the national injury estimates for micromobility products,
in the following order: overall micromobility, e-scooters, and hoverboards. The individual injury
estimates for dockless/rental e-scooters and e-bikes were not presented in this report, due to
small sample sizes, high coefficient of variation (CV), and limitations in NEISS
8
data possibly
leading to undercounts because of the unavailability of sufficient information to identify the
product as a dockless/rental. This is followed by a section on the special study of e-scooter-
related injuries. The special study section is followed by a section on the fatalities reported to
CPSC. Following the fatalities is a section on hazard patterns, as identified from completed in-
depth investigation reports. For each of the sections on fatality and hazard patterns, the analysis
is presented for overall micromobility, then e-scooters (dockless/rental e-scooter statistics in
parentheses), hoverboards, and finally, for e-bikes. Lastly, Appendix A describes staff’s
methodology, including process for data extraction, scope determination, and discussion of the
6
See https://www.cpsc.gov/s3fs-public/safety-concerns-associated-with-micromobility-products.
7
The report, Micromobility Products-Related Deaths Injuries and Hazard Patterns 2017-2020 can be found at:
https://www.cpsc.gov/s3fs-
public/Micromobility-Products-Related-Deaths-Injuries-and-Hazard-Patterns-2017-
2020.pdf?VersionId=s8MfDNAVvHasSbqotb7UC.OCWYDcqena,
8
See https://www.cpsc.gov/Research--Statistics/NEISS-Injury-Data.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
8
raking methodology. Appendix B presents additional details about the injury estimates.
Appendix C includes the revised 2020 special study results using corrected raking analysis to
account for variability of NEISS weights within subgroups. Appendix D lists the questions used
in the follow-up study of NEISS on e-scooter-related injuries treated.
Staff’s national estimates of injuries are based on injury data collected by CPSC’s NEISS, which
is a nationally representative stratified probability sample of hospitals in the United States and
its territories. Each injury report contains a product code that identifies the type of product
involved; in addition, information on the injured victim’s sex, age, diagnosis, disposition, body
part injured, and a brief narrative description of the injury is available. Each injury in the sample
represents an estimated number of injuries that staff projects nationally.
CPSC staff conducted a special follow-up study of NEISS on e-scooter-related injuries treated in
hospital emergency departments between January 1, 2021 and December 31, 2021. These
investigations were completed through telephone interviews and self-administered online
surveys of injured victims to learn more about how the injury occurred, the type of injury, the
scooter type involved, the characteristics of the rider/victim, and the incident scenario.
The fatality statistics, as well as the hazard pattern review staff presents in this report, are
based on incidents reported to CPSC through the CPSRMS. See Appendix A for the codes and
keywords used in the database searches. Reports in CPSRMS come from various sources,
including consumer complaints, news clips, state/local authorities, medical examiners, death
certificates from the states, manufacturers and retailers, among others. Staff considers the data
in CPSRMS to be anecdotal and not nationally representative. Moreover, data collection is
ongoing, and staff considers the later years’ data to be incomplete. Specifically, for death
statistics, which rely on death certificates reported by the states, staff observes a lag of up to 2
years from the time of the death to the time of reporting to the CPSC. As such, the data included
in this report (especially from 2020 through 2021) are likely incomplete. In annual reports for
coming years, CPSC staff will update the statistics on an as-needed basis. Each incident report
contains a product code that identifies the type of product involved, as well as information about
the location of the incident (state and city) and the individual(s) involved or injured (age and
sex), and a narrative description.
I. National Injury Estimates
Staff estimates 267,700 injuries (sample size=7,365, coefficient of variation=0.14) related to all
micromobility products were treated in U.S. emergency departments over the 5-year period
2017 through 2021. The annual estimates for 2022 are not available until NEISS data for 2022
are finalized in spring 2023. The estimated emergency department (ED) visits associated with
micromobility increased from 34,000 in 2017 to 77,200 in 2021, which is statistically significant
(p-value < 0.01).
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
9
Products
Figure 1.1 shows the national annual estimates of ED-treated micromobility injuries and product
types from 2017 through 2021. For overall micromobility, it reflects an increase of 127 percent
from the 2017 estimate, which is statistically significant (p-value < 0.01). Comparing 2021 to
2020, it reflects an increase of 34 percent from the 2020 estimate, which is statistically
significant (p-value < 0.01). In prior years, the ED-treated injury estimates reflected a statistically
significant increase from 2017 to 2018 (p-value: 0.01), as well as from 2018 to 2019 (p-value:
0.05). See Appendix B for additional details.
Figure 1.1: Estimated ED Visits Associated with Micromobility Products by Year
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
*E-bike estimates did not meet the reporting criteria for NEISS and are not presented but are included in “All
Micromobility” estimates.
Staff estimates that 117,600 injuries related to e-scooters were treated from 2017 through 2021.
The 2021 ED-treated injury estimate of 42,200 for e-scooters reflects an increase of 66 percent
from the 2020 estimate, which is statistically significant (p-value < 0.01). There were statistically
significant increases from 2017 to 2018 (p-value: 0.05) and 2018 to 2019 (p-value: < 0.01), as
well. The narratives in the 2017 NEISS data did not provide enough information for staff to
identify any dockless/rental scooters; for 2018 through 2021, the annual estimated ED visits,
where the NEISS narrative provided enough information to determine that the product was a
dockless/rental e-scooter, did not meet minimum requirement for reporting.
3
Moreover, it is likely
that NEISS estimates on dockless/rental e-scooters is an underestimate, due to insufficient
information present to identify the product as a dockless/rental.
Staff estimated 121,300 ED-treated visits for hoverboards from 2017 through 2021. The 2021
ED-treated injury estimate of 23,100 for hoverboards reflects a decrease of 15 percent from the
2020 estimate, which is statistically significant (p-value: 0.04). Over the 5 years, the
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
2017 2018 2019 2020 2021
Estimated ED Visits
Year
E-Scooter Hoverboard All Micromobility*
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
10
hoverboard-related, ED-treated injury estimates have been fluctuating. The 2020 ED-treated
injury estimate of 27,100 for hoverboards reflected a statistically significant increase of 23
percent from the 2019 estimate (p-value: 0.02), but there was a statistically significant decrease
from 2018 to 2019 (26,300 to 22,100, p-value: 0.03).
Staff determined that the annual estimated ED visits for e-bikes did not meet the reporting
criteria for 2017 through 2021. Refer to Appendix B for details. Staff estimates the e-bikes’
share of estimated ED-treated injuries for all micromobility products to be about 11 percent over
the 5-year period.
Sex
Figure 1.2 shows the distributions of estimated micromobility-related injuries by product type
and sex. Males experienced a higher percentage of micromobility-related, ED-treated injuries in
e-scooters (64 percent) during the 5-year period. In contrast, females had a higher percentage
(56 percent) of hoverboard-related, ED-treated injuries.
Figure 1.2: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Sex (2017–2021 Total)
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Age Groups
Figure 1.3 shows the distribution of estimated micromobility-related injuries by age from 2017
through 2021, versus the general U.S. population distribution. Staff obtained the population by
age data from the U.S. Census Bureau,
9
corresponding to the average of 5 years, 2017–2021.
The distributions of estimated injuries sustained by the 15-to-24 and 25-to-44 age groups were
9
See https://www.census.gov/data/datasets/time-series/demo/popest/2010s-national-detail.html.
57%
64%
44%
43%
36%
56%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
All Micromobility E-Scooter Hoverboard
Percentage of Estimated ED
Visits
Male Female
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
11
25 percent and 39 percent, respectively, for e-scooters. These distributions were
disproportionately high compared to their proportions in the general U.S. population (13 percent
and 27 percent, respectively). Similarly, the percentage of estimated hoverboard-related injuries
for the 5-to-14 age group (66 percent) was disproportionately high, compared to its proportion in
the general U.S. population (13 percent).
Figure 1.3: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Age Group (2017–2021 Total) Compared to U.S. Population Age Distribution
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Location of
Injury
Figure 1.4 presents the proportions of injuries associated with micromobility by product type and
location of injury. A large proportion (37 percent for e-scooters, 48 percent for hoverboards, and
41 percent overall) of estimated injuries occurred at unknown locations. For the known
locations, the injuries associated with e-scooters occurred most frequently on streets or
highways (41 percent); whereas, the hoverboard-related injuries occurred most frequently at
home (38 percent).
16%
3%
3%
1%
25%
12%
14%
7%
27%
28%
39%
15%
13%
17%
25%
8%
13%
38%
18%
66%
6%
2%
1%
3%
Census All Micromobility E-Scooter Hoverboard
Percentage of Estimated ED Visits
Micromobility Product Type vs. Census
Age
65 and over
45–64 25–44 15–24 5–14
Under 5
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
12
Figure 1.4: Distribution of Estimated ED Visits Associated with Micromobility Product Type and
Location of Injury (2017–2021 Total)
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Time of Year
Figure 1.5 illustrates the monthly percentage distribution of the estimated emergency visits by
micromobility product type. If the distribution of ED visits were to follow the discrete uniform
distribution,
10
the expected monthly ED visits would be around 8 percent. The months of May
through October had the largest percentages for e-scooters ED visits; whereas, December and
January had the largest percentages for hoverboard-related ED visits.
10
The discrete uniform distribution is a symmetric probability distribution, where all 365 days are equally likely to be
observed. Under that distribution, every month has a probability between 7.7 percent and 8.5 percent.
41%
37%
48%
21%
7%
38%
27%
41%
7%
8%
12%
4%
<1%
<1%
<1%
2%
2%
3%
All Micromobility E-Scooter Hoverboard
Percentage of Estimated ED Visits
Not recorded Home Street or Highway
Other public property School Place of recreation or sports
*
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
13
Figure 1.5: Distribution of Estimated ED Visits Associated with Micromobility Product Type and
Month of Injury (2017–2021 Total)
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Race/Ethnicity Groups
NEISS data include three variables (“Race,” “RaceOth,” and “Hispanic”) that record information
about the race or ethnicity of ED-treated patients. These three variables were used when
designating race/ethnicity categories for each patient. This resulted in five categories used for
analysis on race/ethnicity in this report: (1) non-Hispanic Black or African American (Black), (2)
non-Hispanic White (White), (3) Hispanic (Any race) (4) all other patients with known
races/ethnicities or who identify as Multiracial (Other), and (5) patients with an unknown
race/ethnicity. This specific grouping aligns with the population data grouping presented by the
U.S. Census Bureau.
As of the writing of this report, the Hispanic ethnicity variable has only been recorded for NEISS
years 2019 through 2021. Therefore, analysis concerning race/ethnicity for the estimated
number of ED-treated micromobility injuries is based on the years 2019 through 2021.
Estimates of micromobility injuries involving races/ethnicities other than Black, Hispanic, or
White are not statistically reliable for the data available, and hence, not presented.
8%
3%
13%
5%
4%
6%
7%
6%
9%
8%
7%
9%
9%
9%
8%
9%
11%
7%
10%
13%
8%
10%
13%
6%
10%
13%
6%
7%
9%
5%
6%
6%
5%
10%
4%
18%
All Micromobility E-Scooter Hoverboard
Percentage of Estimated ED Visits
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
14
Figure 1.6 shows the distribution of estimated micromobility-related injuries versus the general
U.S. population distribution by race/ethnicity from 2019 through 2021. The population data
corresponding to the average of the 3 years, 2019–2021, are from U.S. Census Bureau.
9
As the
figure shows, the proportion of estimated injuries sustained by Black consumers was 31 percent
for overall micromobility and 37 percent for e-scooters. These injuries seem disproportionately
high compared to the proportion in the general U.S. population (13 percent Black Americans).
However, staff notes some limitation in interpretation of the data presented in Figure 1.6 for two
reasons. First, only 76 percent (144,200 out of 189,700) of the estimated micromobility injuries
between 2019 and 2021 provided race/ethnicity information. Second, micromobility injuries
(especially for e-scooters and e-bikes) are skewed towards younger (Figure 1.3) and possibly
more urban sub-populations, and staff is uncertain what reference basis of race/ethnicity
baseline is applicable.
Figure 1.6: Distribution of Estimated ED Visits Associated with Micromobility by Product Type
and Race/Ethnicity, where known
+
, (2019–2021 Total) Compared to U.S. Population
Race/Ethnicity Distribution
Source: NEISS, U.S. Consumer Product Safety Commission, 2019-2021.
Asterisks (*) indicate Race/Ethnicity designations are for reported single race only. The Other category includes the
proportion of the U.S. population that are not counted among the non-Hispanic Black, Hispanic, or non-Hispanic
White people in U.S. Census figures.
+
Race/Ethnicity percentages are based on the known data which were available in only 76% of cases.
13%
31%
37%
18%
19%
10%
10%
13%
60%
55%
48%
66%
9%
4%
4%
3%
Census All Micromobility E-Scooter Hoverboard
Percentage of Estimated ED Visits
Micromobility Product Type vs. Census
Black/African American, Non-Hispanic* Hispanic (Any Race)
White, Non-Hispanic* Other
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
15
Other Characteristics
The remaining characteristics, which do not vary much by product type, are as follows:
• Fractures, followed by contusions/abrasions, are the two most common diagnoses.
• The most frequently injured body parts
11
are the upper and lower limbs, as well as the head
and neck.
• Most of the injuries are attributed to unspecified falls. Loss of user-control, collisions with
other motor vehicles, and pavement issues are other notable hazards leading to the injuries.
• Overall, 89 percent of the injured are treated and released from the EDs. About 8 percent are
treated and admitted or transferred to another hospital. Disposition of the remaining 3 percent
of injuries included: “left without being seen,” “held for observation,” as well as fatalities.
12
II. Special Study on E-Scooters
In response to the changing environment for micromobility devices, CPSC revised the set of
product codes used to describe scooters and powered skateboards (e.g., hoverboards)
beginning in 2020. Previously two product codes 1329 (Scooters, unpowered) and 5042
(Scooters/skateboards, powered) were used to capture scooters and powered skateboards.
However, this pair of codes created a conundrum for incidents and emergency department visits
that only described a “scooter” without indicating whether the scooter was powered. In 2020,
the set was replaced with 5022 (Scooters, powered), 5023 (Scooters, unpowered), 5024
(Scooters, unspecified), and 5025 (Hoverboards, and powered skateboards). To obtain more
detailed information about the scooters associated with emergency department visits, CPSC
staff initiated a special study of scooter-related injuries in January 1, 2020,
13
which continued in
2021. This section focuses on the special study results from January 1, 2021 through December
31, 2021. More than 3,600 injuries, coded as 5022 (Scooters, Powered) and 5024 (Scooters,
Unspecified) were followed up through a survey questionnaire to obtain additional information
on the scooter type involved, how the injury occurred, the type of injury, the characteristics of
the rider/victim, and the incident scenario. The results showed some of product code 5024
(Scooters, unspecified) injuries (sample size=89) were powered scooters. Recalibrating the
weights for these cases (by taking the survey response rate into account), the 89 injury cases
were added to all completed survey responses for injuries under product code 5022 (Scooters,
powered). The e-scooter-related injuries (sample size=307) from the special study were
included as part of the 2021 estimate of 42,200 e-scooter injuries seen in emergency
departments. In the investigations, information was requested directly from the victim (or the
11
Body parts were grouped. For example, all body parts that would generally be considered a part of the lower limb
(e.g., toe, foot, ankle, knee, and leg) were grouped as “lower limb.”
12
Less than 0.1 percent of the estimated injuries were fatal. All fatal injuries from NEISS have been included in the
fatality discussion of this report.
13
Since the publication of the previous annual report in 2021, staff became aware of errors in the estimate derivation
process used for the 2020 special study. In this report, staff revised the 2020 special study results and presented
them in Appendix C.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
16
victim’s parent, if the victim was a minor) about the type of scooter involved, where the scooter
was obtained, and how the injury occurred.
Scooter Types
Table 2.1 shows the number of respondents by verified scooter type and the corresponding
NEISS product code in the special study. The respondents were asked a series of questions
related to scooter type and different components. They were asked to provide information on
the power source, whether it was an assisted mobility scooter, moped, motorcycle, or scooter
that requires a registration or license. In addition, the respondents were asked if their scooter
had steering handles and side-by-side wheels, as opposed to wheels that are one in front of the
other. Of the 647 completed follow-up investigations on scooter-related incidents, 47 percent
were e-scooters (307 out of 647), 33 percent were related to kick scooters, and the remaining
20 percent were other types of scooters. Of the 307 follow-up, e-scooter-related incidents, 71
percent were originally coded as powered scooters (217 out of 307), and 29 percent were coded
as unspecified scooters (89 out of 307).
14
Table 2.1: Scooter Products: NEISS Product Code vs. Verified Scooter Product (2021)
Verified Product Overall
Powered
Scooter (5022)
Unspecified
Scooter (5024)
Various Other
Product Codes
Kick (unpowered) Scooter
211
17
190
4
Mobility Scooter
27
18
9
0
Moped
45
25
19
1
Hoverboard
9
4
4
1
E-Scooter
307
218
89
0
Gas powered scooter
13
12
1
0
Other/missing
35
18
16
1
Total
647
312
328
7
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on responses from the special study survey questions V3 – V8. See Appendix C for details.
Table 2.2 shows the response rate for e-scooter investigations by Overall, Powered, and
Unspecified Scooter. Staff selected 3,642 cases for investigations, of which 307 were completed
and determined to be in scope, and 3,336 were either incomplete or out of scope. For the
findings to be generalizable to all those injured, raking ratio estimation was implemented to
create adjusted weights to reduce non-response bias due to differences in sex, race, stratum,
age category, and response rate. These benchmarked weights were used to produce the
national estimates from the special study survey. See Appendix A for additional details.
14
The 89 unspecified scooters have been included in the calculation of national injury estimates.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
17
Table 2.2: Completed E-Scooter Investigations by Powered vs. Unspecified Scooter
(2021)
Investigation Overall Percent
Powered
Scooter (5022)
Percent
Unspecified
Scooter (5024)
Percent
Completed Investigation
307
8%
218
15%
89
4%
Incomplete/Out of Scope
3,335
92%
1,276
85%
2,059
96%
Total Assigned
3,642
100%
1,494
100%
2,148
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Estimated Injuries - S
cenario-Specific Characteristics
In the remainder of this section, staff summarizes the 307 e-scooter cases from the follow-up
investigations. They are descriptive of the portion that participated in the special study.
Additionally, respondents did not respond to every question posed to them; as such, in many of
the tables below, the “Total” row does not add up to the total of 42,200 estimated injuries.
Table 2.3 shows the distribution of the estimated ED injuries based on the rental status of e-
scooter. Of the 40,000 injuries, rental e-scooters accounted for 44 percent, while nonrental e-
scooters accounted for 56 percent.
Table 2.3: Distribution of Estimated Injuries by Rental Status (2021)
Rental vs Non-rental N
Estimated ED-treated
Injuries+
Percent
Rental
149
17,400
44%
Non-rental
148
22,600
56%
Total
297
40,000
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on the responses from special study survey question: S1 Which of the following best describes the scooter?
Rental, Owned by victim, Borrowed, Other, Don’t know.
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table 2.4 shows the distribution of the estimated ED injuries while the victim was either riding
the e-scooter or was struck by the e-scooter. Of the 42,200 injuries, 96 percent of the injuries
occurred to riders of the e-scooter.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
18
Table 2.4: Distribution of Estimated Injuries by How Victim Was Injured (2021)
Riding/Struck by
Scooter
N Estimated ED-treated Injuries+ Percent
Riding the scooter
293
40,500
96%
Struck by scooter
5
*
*
Other
5
1,300
2%
Total
303
42,200
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on the responses from special study survey question: V10 You/the victim were/was: Riding the scooter; Struck
by scooter; Other.
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
*The reporting criteria for NEISS require that the estimated number of injuries be 1,200 or higher.
Table 2.5 shows the distribution of the ED visits based on the type of surface. Of the 41,700
estimated injuries, paved roads accounted for 74 percent, paved sidewalks for 23 percent, and
driveways for 3 percent.
Table 2.5: Distribution of Estimated Injuries by Type of Riding Surface (2021)
Type of Surface
N
Estimated ED-treated Injuries+
Percent
Paved road
177
30,900
74%
Paved sidewalk
113
9,700
23%
Driveway
6
1,400
3%
Other
11
*
*
Total
299
41,700
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on the special study survey question: A3 What type of surface were you/the victim on?
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates; percentages may not sum to 100 percent due to rounding, and also
because some riders may have ridden e-scooter on more than one surface type.
*The reporting criteria for NEISS require that the estimated number of injuries be 1,200 or higher.
Table 2.6 shows the distribution of the injuries by responses on visibility (whether it was dark or
difficult to see while riding the scooter). Of the 42,200 estimated injuries with a response to this
question, 21 percent of the victims responded that it was dark or difficult to see.
Table 2.6: Distribution of Estimated Injuries by Visibility (2021)
Visibility Issues
N
Estimated ED-treated Injuries+
Percent
No
235
33,300
79%
Yes
62
8,200
21%
Total
297
42,200
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on the special study survey question: A5 Was it dark or difficult to see?
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
19
Table 2.7 shows the percentage breakout of the ED-treated injuries based on source of
distraction, such as music, cell phone, or loud music while riding the scooter. Of the 41,700
estimated injuries with a response to this question, a source of distraction accounted for only 5
percent.
Table 2.7: Distribution of Estimated Injuries by Distraction (2021)
Distraction
N
Estimated ED-treated Injuries+
Percent
No
282
39,800
95%
Yes
15
1,900
5%
Total
297
41,700
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on responses to the special study survey question: A6 Was there anything else occurring at the time of the
accident such as music, cell phone interference, or loud music?
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table 2.8 shows the percentage of ED-treated injuries based on whether rider was
carrying/holding something (e.g., bag, purse, or backpack) while riding the scooter. Of the
39,500 estimated injuries with a response to this question, 26 percent of the injured were
carrying or holding something while riding the scooter.
Table 2.8: Distribution of Estimated Injuries by Rider Baggage (2021)
Carrying/holding something
N
Estimated ED-treated Injuries+
Percent
No
203
29,300
74%
Yes
81
10,300
26%
Total
284
39,500
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on responses to the special study survey question: A8 Was there anything else occurring at the time of the
accident such as music, cell phone interference, or loud music?
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table 2.9 shows the percent of ED-treated injuries by usage of safety equipment, such as a
helmet, blinking lights, head lamp, knee pads, elbow pads, and/or reflective vest while riding the
scooter. Because riders may use more than one type of safety equipment, these rows do not
add up to the total. Of the 19,100 estimated injuries where this information was available, the
rider was wearing a helmet while riding the e-scooter 52 percent of the time, had blinking lights
or head lamp 64 percent of the time, was wearing a reflective vest 6 percent of the time, was
wearing knee/elbow pads 5 percent of the time, and had other unspecified safety equipment 6
percent of the time.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
20
Table 2.9: Distribution of Estimated Injuries by Safety Equipment (2021)
Safety Equipment
N
Estimated ED-treated Injuries+
Percent
Helmet
71
9,900
52%
Blinking Lights/Head lamp
97
12,300
64%
Reflective vest
8
*
*
Knee/elbow pads
9
*
*
Other
18
*
*
Total
149
19,100
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2021.
Based on responses from the special study survey question: S12 I'm going to read a list of safety equipment that
riders might wear. Please tell me if the rider was wearing any of these at the time of the incident.
Riders may use more than one type of safety equipment as such rows do not add up to total.
+Injury estimates are rounded to the nearest 100. Percentages are calculated from the unrounded estimates.
*The reporting criteria for NEISS require that the estimated number of injuries be 1,200 or higher.
III. Reported Fatalities Associated with
Micromobility Products
CPSC staff is aware of 129 fatalities related to micromobility products that occurred in the
United States during the 5-year timeframe, 2017 through 2021. Some of the characteristics of
these reported fatalities are summarized in the tables below. Due to delays in death certificate
reporting, staff expects the number of reported fatalities to change in future reports. Compared
to the last annual report
, there are 10
additional fatalities. Out of the 10 fatalities, 7 e-bike, and 3
e-scooter-related fatalities occurred in 2020.
T
able 3.1 shows the fatality data for micromobility products by incident year from 2017 to 2021.
While data reporting is ongoing, as of the writing of this report, staff identified 129 fatalities
involving the micromobility products. E-scooter-related fatalities represent 68 (14 were dockless
e-scooter-related) out of the 129 (53 percent) total fatalities, increasing substantially from 2017
to 2021. E-bikes account for 53 (41 percent) of the total fatalities, mostly reported in the latter
part of the 5-year time frame 2017-2021. CPSC staff is aware of 8 fatalities involving
hoverboards during the same 5-year timeframe.
Table 3.1: Number of Reported Fatalities Associated with Micromobility Product Type
and Year
Year All Micromobility
E-Scooter
(Dockless/rental)
Hoverboard E-Bike
2017
5
1 (0)
4
0
2018
11
5 (2)
0
6
2019
31
25 (7)
0
6
2020
34
14 (2)
2
18
2021
48
23 (3)
2
23
Total
129
68 (14)
8
53
Note: Reporting for 2020-2021 is ongoing. Counts may change in future reports.
Source: CPSRMS, NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
21
Among the 129 micromobility-related fatalities, 103 were male decedents, 20 were females, and
gender was unknown for the remaining 6 decedents. Decedents in most of the fatalities related
to e-scooters (55, including 12 dockless e-scooter fatalities, out of 68), and e-bikes (46 out of
53) were males; of the 8 hoverboard-related fatality victims, 7 were females. The distribution of
decedents by product type by gender distribution is presented in Table 3.2.
Table 3.2: Number of Reported Fatalities Associated with Micromobility Product Type
and Gender (2017–2021 Total)
Gender All Micromobility
E-Scooter
(Dockless/rental)
Hoverboard E-Bike
Male
103
55 (12)
2
46
Female
20
7 (2)
6
7
Unknown
6
6 (0)
0
0
Total
129
68 (14)
8
53
Note: Reporting for 2020-2021 is ongoing. Counts may change in future reports.
Source: CPSRMS, NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Table 3.3 presents the fatality data for micromobility products by product type and age group of
the deceased from 2017 to 2021. One hundred and eleven out of 129 total reported
micromobility fatalities provided age information. The fatalities were divided into three age
groups: under 18 years of age; 18 to 59 years; and 60 years of age or older. Of the 129
fatalities, 79 (61 percent) were 18 to 59 years old, 26 (20 percent) were 60 years or older, 6 (5
percent) were under 18 years of age, and the remaining 19 (15 percent) were of unspecified
ages. Of the 79 fatalities in the 18-59 age group, 44 involved e-scooters, 33 involved e-bikes,
and 2 involved hoverboards. Of the 6 fatalities in the under 18 age group, 4 involved
hoverboards, and 2 (1 dockless/rental) involved e-scooters. The majority of the 26 fatalities in
the 60 and over age group involved e-bikes (17); 7 involved e-scooters; and 2 involved a
hoverboard.
Table 3.3: Reported Fatalities Associated with Micromobility Product Type and Age
Group (2017–2021 Total)
Age Group All Micromobility
E-Scooter
(Dockless/rental)
Hoverboard E-Bike
Under 18
6
2 (1)
4
0
18–59
79
44 (9)
2
33
60 and over
26
7 (1)
2
17
Unknown
18
15 (3)
0
3
Total
129
68 (14)
8
53
Note: Reporting for 2020-2021 is ongoing. Counts may change in future reports.
Source: CPSRMS, NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Table 3.4 shows the data for micromobility-related fatalities by product and hazard types. Out of
the 129 fatalities reported to CPSC staff, 68 (14 dockless/rental e-scooters) deaths involved e-
scooters, 8 involved hoverboards, and e-bikes accounted for 53 of these deaths.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
22
E-Scooters (Including Dockless/Rental E-Scooters)
Staff’s review of the 68 fatalities shows the following hazards:
• Motor vehicle accidents were the leading cause of death associated with e-scooters. Out of
the 68 fatalities reported to CPSC staff, 49 deaths involved motor vehicle accidents (e.g.,
collision with cars, SUVs, and trucks).
• Nine e-scooter fatalities (including 2 on dockless/rental e-scooters) were due to user-control
issues. User-control issues led to crashing into fixed object/tree, colliding with other riders,
striking road curbs, and/or getting thrown into oncoming traffic.
• Two fatalities were associated with e-scooter-related fires. Electrical fires started while electric
scooters were being charged.
• Pedestrian accidents were associated in 2 fatalities; they were struck by e-scooter riders.
• Other hazards accounted for 2 fatalities. One fatality was associated with intoxication, and
another fatality involved a crash with a commuter train.
• Unknown falls accounted for 4 e-scooter-related fatalities with falls resulting in deaths; staff
does not have sufficient scenario-specific information to determine what caused the falls.
Hoverboards
Staff’s review of the 8 hoverboard-related fatalities shows the following hazards:
• Out of the 8 fatalities reported to CPSC staff, 2 deaths involved motor vehicle accidents (e.g.,
collision with cars, SUVs, and/or trucks).
• Four fatalities were associated with hoverboard-related fires, where flames erupted while
charging the hoverboards.
• Unknown falls accounted for 2 hoverboard-related fatalities; staff does not have sufficient
scenario-specific information to determine what caused the falls.
E-B
ikes
Staff’s review of the 53 fatalities shows the following hazards:
• Collisions with motor vehicles were the leading cause of death associated with e-bikes,
accounting for 27 of the 53 reported deaths.
• Twelve e-bike fatalities were due to user-control issues, such as crashing into other fixed
objects (i.e., gate, sign, post, barricade, railing, dumpster, median), striking road curbs, and
getting thrown into oncoming traffic.
• Pedestrian accidents were associated with 6 fatalities; 2 e-bike riders were killed due to blunt
impact caused when they struck pedestrians; and 4 pedestrians were struck by e-bikes and
died from blunt impact/fall.
• One fatality was due to rider hitting a speed bump and becoming airborne, then crash-
landing.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
23
• Unknown falls accounted for 7 e-bike-related fatalities, but staff does not have sufficient
scenario-specific information to determine what caused the falls.
Table 3.4: Reported Fatalities Associated with Micromobility Product Type and
Associated Hazards (2017–2021 Total)
Hazard Pattern All Micromobility
E-Scooter
(Dockless/rental)
Hoverboard E-bike
Motor vehicle accident
78
49 (12)
2
27
User-control
21
9 (2)
0
12
Fire hazards
6
2
4
0
Pedestrian accident
8
2
0
6
Pavement
1
0
0
1
Other
2
2
0
0
Unspecified falls
13
4
2
7
Total
129
68 (14)
8
53
Note: Reporting for 2020-2021 is ongoing. Counts may change in future reports.
Source: CPSRMS, NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
IV. Hazard Patterns Based on In-Depth
Investigation Review
Given that the narratives available in NEISS focus on the injury sustained rather than on the
circumstances leading to the injury, and the death reports only cover the fatalities reported,
CPSC staff evaluated the available in-depth investigations for a more comprehensive look at
how incidents happened. Based on reports of incidents in CPSRMS that occurred between
2017 and 2021, CPSC Field staff completed 204 follow-up in-depth investigations related to all
micromobility products. The 204 follow-up in-depth investigations include 7 fatalities that have
been discussed in the fatality section above. Of the 204 completed investigations, 48 involved
an e-scooter (32 of the 48 were dockless/rental e-scooters); 145 involved a hoverboard; and 11
involved an e-bike. This does not necessarily reflect the current prevalence of incidents related
to micromobility products in the CPSRMS database. Staff initiated many more in-depth
investigations that could not be completed due to product unavailability or unwillingness of
consumer(s) to cooperate and provide product and injury information. Data collection is
ongoing, and staff expects the numbers to change in future reports. Staff discusses the types of
products and the reported hazards associated with each from the 204 investigations below.
E-S
cooters (Including Dockless/Rental E-Scooters)
Of the 48 e-scooter-related incidents that were investigated, 32 were dockless e-scooters.
Staff’s review of the 48 in-depth investigations shows the following hazards
15
:
15
Based on newly available information, one e-scooter incident that was included in the previously published annual
report was deemed out of scope and is excluded in this report.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
24
• Brake problem was associated with more of the reported incidents (18 out of 48) than any
other category. The investigations show that brakes not engaging at all, sporadically
engaging, or engaging excessively following a delay resulted in 15 of the 18 reported
incidents. In one other case, the complainant reported that the brake cable was not properly
attached to the adjustment bracket on the handlebar grip. The remaining 2 incidents
happened to the same consumer; other than an email message indicating mechanical brake
failures, the consumer did not share any additional scenario-specific detail.
• Fire hazard was reported in 12 of the 48 incidents; mostly occurred while charging the e-
scooter (8 out of 12); one incident happened after charging; and no other information is
available for the remaining 3 incidents.
• Unexpected power loss caused the rider(s) to tip over or get thrown off in 5 of the 48
reported incidents. In one of the 5 incidents, the rider was going downhill, and the e-scooter
lost power when it went over a curb.
• Multiple product-related issue such as brakes malfunctioning, throttles getting stuck, control
panels catching fire, and wobbliness were reported in 5 of the 48 incidents. In one of these 5
incidents, the rider went over a bump, which seemed to mark the onset of throttle and brake
problems.
• Control factor, environmental factor, and unknown factor played a major role in 5 (out of
48) incidents. The users lost control of the e-scooters in 3 cases after hitting a pothole,
coming off of a curb, or while being chased by an aggressive dog, respectively. Staff has
insufficient information to determine why the users in the remaining 2 incidents went through
an intersection against a red light, causing a collision with other motor vehicles that had the
right-of-way.
• Miscellaneous product-related issue, such as footboard or handlebar breaking/detaching
or e-scooter not powering up, were reported in the remaining 3 incidents.
Hoverboards
A review of the 145 in-depth investigations showed the following hazards:
• Fire hazard was the most common problem, accounting for 118 of the 145 reports. The
reports describe fire (sometimes after an explosion), smoke, or sparks emanating from the
product; some reports describe the product overheating or melting. Seventy-eight of the 118
incidents occurred when the board was being charged or had just completed charging; 23 of
the 118 reported that the incidents occurred during use or immediately after use; 11 of the
118 boards caught fire spontaneously. Two additional incidents reported that the board would
not shut-off and eventually started to smoke. In a different incident, the board caught fire after
a consumer had reset the battery, which was not charging. Another hoverboard emitted
sparks when the consumer attempted to remove the battery per the manufacturer’s
recommendation because the board was not charging. Staff had no scenario-specific
information for the 2 remaining incidents.
• Other electrical hazard was identified in 20 of the 145 investigated incidents. These included
board spinning out of control at high speed, spinning in circles/one side, board failing to shut
off, jumping off device, throwing off rider, and crashing into fixed objects;
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
25
• Control factor, environmental factor, and unknown factor played a role in 4 out of 145
incidents. Two of these incidents were related to losing balance while trying to perform a
trick/avoid a biker. Staff had no scenario-specific information for the other 2 incidents.
• Miscellaneous product-related issue resulted in 3 of the 145 investigated incidents. In all of
the cases, the board vibrated excessively, throwing off the rider. While uneven weight
distribution may have contributed, at least for 2 of these incidents, some other unknown factor
was at play.
E-B
ikes
CPSC staff reviewed 11 completed investigation reports.
• Fire hazard was identified in 4 of the 11 investigated incidents. The reports describe the fires
started while the e-bikes were being charged. In one case, the user removed the battery from
the e-bike, which then suddenly caught fire. In another case, staff has insufficient information
to determine whether the lithium battery was being charged or being removed from the e-bike.
• Brake issue was reported in 4 incidents. Reports describe brakes malfunctioned/stopped
working in 3 of the cases; the fourth report describes the owner of the e-bike needed to
“tighten” the brakes repeatedly until, on the day of the incident, the brakes failed completely.
• Miscellaneous product-related issue, such as structural integrity/design defect issue, was
reported in 2 incidents. In one incident, the consumer was riding an e-bike when the pedal
and crank assembly were separated from the bike frame, causing the person to fall. The other
incident reported the pin in the handle bars prevented making a left turn, and the front wheel
got stuck.
• Multiple product-related issue was identified in one incident as the consumer stated that he
was having issues with the bicycle chain and the throttle; the motor stalled while he was riding
the bicycle, causing him to fall.
Table 4.1 summarizes the hazards associated with the use of the various micromobility
products.
Table 4.1: Reported Fatalities Associated with Micromobility Product Type and
Associated Hazards (2017–2021 Total)
Hazard
All
Micromobility
E-Scooter
(Dockless/rental)
Hoverboard E-Bike
Fire Hazard
134
12 (1)
118
4
Brake Problem
22
18 (16)
0
4
Unexpected Power Loss
6
5 (5)
0
0
Other Electrical Hazard
19
0
20
0
Misc. Product-Related Issue
8
3 (3)
3
2
Control/Environmental/Unknown
Factor
12 5 (4) 7 0
Multiple Product-Related Issue
6
5 (3)
0
1
Total
207
48 (32)
148
11
Source: CPSC In-Depth Investigation File from CPSRMS, 2017-2021.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
26
References
lzreal, D., Hoaglin, D. C., & Battaglia, M. P. (n.d.). A SAS Macro for Balancing a Weighted
Sample. Retrieved October 2010, from Abt Associates Inc.: https://www.abtassociates.com.
NEISS sample design: https://www.cpsc.gov/Research--Statistics/NEISS-Injury-Data.
Lee, Douglas, “Safety Concerns Associated with Micromobility Products” U.S. Consumer
Product Safety Commission, March 2020. Online at https://www.cpsc.gov/s3fs-publ
ic/Report-on-
Micromobility-Products_FINAL-to-Commission.pdf?THHIorYXAZ.KiZnobh1o7.7.lN9nNCLo.
Schel
long, Daniel, Sadek, Philipp, Schaetzberger, Carsten, and Barrack, Tyler, “The Promise
and Pitfalls of E-Scooter Sharing”, May 2019. Online at:
https://www.bcg.com/publications/2019/promise-pitfalls-e-scooter-sharing.aspx.
Schroeder, Tom, “Trend Analysis of NEISS Data”, U.S. Consumer Product Safety Commission,
February 2000.
Schroeder, Tom, and Ault, Kimberly (2001), “The NEISS Sample (Design and Implementation),
1997 to Present” U.S. Consumer Product Safety Commission, June 2001. Online at:
https://www.cpsc.gov/s3fs-public/pdfs/blk_media_2001d011-6b6.pdf
.
Tark, James, “Micromobility Products-Related Deaths, Injuries, and Hazard Patterns 2017-
2020” U.S. Consumer Product Safety Commission, October 2021. Online at
https://www.cpsc.gov/s3fs-p
ublic/Micromobility-Products-Related-Deaths-Injuries-and-Hazard-
Patterns-2017-2020.pdf?VersionId=s8MfDNAVvHasSbqotb7UC.OCWYDcqena.
U.S. Census Bureau https://www.census.gov/
The Monthly Postcensal Resident Population counts can be found at:
https://www.census.gov/programs-s
urveys/popest/technical-documentation/research/evaluation-
estimates/2020-evaluation-estimates/2010s-national-detail.html
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
27
Appendix A
Methodology
CPSC staff queried epidemiology data from the (NEISS) and CPSRMS. Staff reviewed query
results to include only the incidents related to micromobility products.
Date of Data E
xtraction: 1/27/2022
Date of NEISS Weight Update: 4/8/2022
Incident Dates or Treatment Dates:
1/1/17-12/31/21
Product codes and narrative descriptions
E-scooter:
• Product codes: 5042 (Scooters/skateboards, powered) in 2017-2019, 5022 (Scooters,
Powered) from 2020 onwards, and 5024 (Scooters, Unspecified) as part of the 2020 and
2021 special studies.
• For 2020: 42 injury cases, originally coded as 5024, were identified to be e-scooters following
the investigations through the special study. These 42 cases were added to the all injury
cases coded under 5042 and 5022. To account for the response rate of the 2020 special
study, the weights for these 42 cases were recalibrated before derivation of the injury
estimates.
• For 2021: 89 injury cases, originally coded as 5024, were identified to be e-scooters following
the investigations through the special study. These 89 cases were added to the all injury
cases coded under 5042 and 5022. To account for the response rate of the 2021 special
study, the weights for these 89 cases were recalibrated before derivation of the injury
estimates.
• Narrative/Text contains any of the following: “electric scooter,” “e-scooter,” “stand up scooter,”
“standup scooter,” “motorized scooter,” “power scooter,” “dockless scooter,” “rental scooter,”
“scooter sharing,” any brand known to be dockless/rental e-scooters and other variant
spellings.
• For CPSRMS data, searched brand, manufacturer fields when their information were
available to identify in-scope products.
Hoverboard:
• Product code: 5042 (Scooters/skateboards, powered) in 2017–2019 and 5025 (Hoverboards
and powered skateboards) from 2020 onwards.
• Narrative/Text contains any of the following: “self-balancing scooter,” “hoverboard,” and other
varied spellings.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
28
• For CPSRMS data, searched brand, manufacturer fields when their information were
available to identify in-scope products.
E-bike:
• Product code: 3215 (Mopeds or power-assisted cycles) in 2017-2021
• Narrative/Text contains any of the following: “electric bike,” “e-bike,” “electric bicycle,” “e-
bicycle,” “power (assisted) bike,” “power (assisted) bicycle,” “motorized bike,” “motorized
bicycle,” and other varied spellings.
• For CPSRMS data, searched brand, manufacturer fields when their information was available
to identify in-scope products.
For this report, an incident was deemed out of scope if any of the following criteria was satisfied:
• not electric or battery-powered
• not two-wheeled
• seated scooters
• mopeds, motorcycle
• mobility scooter/wheelchair
• any brand names that are known to be not of interest
• incidents occurred outside of the U.S.
For CPSRMS data, CPSC staff consolidated multiple reports that pertain to a single incident as
one incident prior to analysis.
Raking Methodology and Raked Injury Estimate
This section summarizes the survey nonresponse and the estimation techniques implemented
in this study in order to handle the nonresponse and to generate estimates of injuries. A unit
nonresponse occurs when an assigned (sampled) subject cannot be reached or refuses to
participate. In this case, no information is collected from the subject. In calculating the number
of emergency department-treated, e-scooter-related injuries based on the results of this study,
the biggest area of concern is unit nonresponse.
A popular method of dealing with unit nonresponse is raking, also known as "raking ratio
estimation" or "sample balancing." This method uses an iterative proportional fitting algorithm to
adjust weights to known population marginal totals to handle unit nonresponse within the
survey. Further information can be found in (Tu and Garland 2012).
From January 1, 2020 to December 31, 2021, CPSC staff conducted a special study for injuries
related to the NEISS product codes 5022 (Scooters, powered), and 5024 (Scooters,
unspecified). The product code 5022 is used when an emergency department-treated injury is
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
29
reported to have been associated with an e-scooter, whereas the product code 5024 is used
when specific type of scooter is unknown. Any injury in the specified timeframe that had a
corresponding product code of either 5022 or 5024 were assigned for a follow-up survey. Many
surveys were not completed due to missing contact information for the victim, inability to contact
a victim that had contact information available, or a victim's refusal to participate. Raking was
implemented to handle this type of nonresponse. The weights recorded for nonresponding
subjects are distributed among the subjects with completed surveys. That is, raking adjusts the
weights of the completed surveys to compensate for the nonresponse.
To handle the nonresponse in this study via raking, the demographic variables, such as gender,
age, race, hospital size (stratum), and e-scooter response rate were used to rake the sample
against the assigned marginal totals. The following variables had multiple-level responses and
had to be collapsed into categories to limit the number of possible combinations to a
manageable number: age, race, and stratum. Age variable was split into two age groups: “13
years old or younger” and “Over 13 years old”; Race into “White,” “Other race,” and “Not stated”;
Stratum was grouped into “Very large and Children’s hospitals” and “Other hospitals.” Estimates
for e-scooter-related, emergency department-treated injuries can be determined based on the
information collected for each completed survey. Raking uses the population marginal totals for
all the aforementioned variables. For this study, the population marginal totals were generated
through the NEISS database using the data known for all assigned cases for this special study
survey. Using the corresponding weights, the population marginal totals can be obtained. The
survey weights within the study were raked (adjusted) to match the population marginal totals
for each variable. The raking macro (lzreal, Hoaglin, & Battaglia) was used to generate the
raked weights.
Appendix B
Summary of Annual Injury Estimates and Trend Analysis
Statistically significant year-to-year estimated ED visits:
• All Micromobility Products
o For 2021, estimated ED visits were significantly higher than for any of the years
2017 (p-value: < 0.01), 2018 (p-value: < 0.01), 2019 (p-value: 0.03), and 2020 (p-
value: < 0.01).
o For 2020, estimated ED visits were significantly higher than for 2017 (p-value: <
0.01).
o For 2019, estimated ED visits were significantly higher than for either of the years
2017 (p-value: < 0.01) or 2018 (p-value: 0.05).
o For 2018, estimated ED visits were significantly higher than for 2017 (p-value:
0.01).
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
30
• E-Scooters
o For 2021, estimated ED visits were significantly higher than for any of the years
2017 (p-value: < 0.01), 2018 (p-value: < 0.01), and 2020 (p-value: < 0.01).
o For 2020, estimated ED visits were significantly higher than for 2017 (p-value: <
0.01).
o For 2019, estimated ED visits were significantly higher than for either of years
2017 (p-value: < 0.01) or 2018 (p-value: < 0.01).
o For 2018, estimated ED visits were significantly higher than for 2017 (p-value:
0.05).
• Hoverboards
o For 2021, estimated ED visits were significantly lower than for 2020 (p-value:
0.04).
o For 2020, estimated ED visits were significantly higher than for 2019 (p-value:
0.02).
o For 2019, estimated ED visits were significantly lower than for 2018 (p-value:
0.03).
Annual Injury Estimates, Corresponding Sample Sizes, and Coefficients of Variation by
Product Type, 2017-2021
Year
All Micromobility E-Scooter
Dockless/Rental E-
Scooter
Hoverboard E-Bike
N
Est. ED
Visits
C.V. N
Est. ED
Visits
C.V. N
Est. ED
Visits
C.V. N
Est. ED
Visits
C.V.
N
Est. ED
Visits
C.V.
2017
935
34,000
0.11
185
7,700
0.18
--
--
--
683
22,800
0.11
67
3,500
0.32
2018
1,149
44,000
0.12
369
14,500
0.25
47
1,800
0.80
714
26,300
0.13
66
3,200
0.48
2019
1,561
54,800
0.15
761
27,700
0.27
163
4,900
0.59
664
22,100
0.16
136
5,000
0.36
2020
1,690
57,700
0.17
652
25,400
0.28
78
3,400
0.27
922
27,100
0.14
116
5,200
0.41
2021
2,030
77,200
0.18
945
42,200
0.23
216
11,200
0.34
794
23,100
0.18
290
11,800
0.35
Total
7,365
267,700
0.14
2,913
117,600
0.21
504
21,300
0.36
3,777
121,300
0.13
675
28,800
0.36
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Grey shaded cells in the table do not meet the NEISS reportability criteria. Reporting criteria for NEISS require that
the estimated number of injuries be 1,200 or higher, the sample size be 20 or larger, and the coefficient of variation
be less than 33 percent.
-- Staff did not identify any dockless/rental scooters in the 2017 NEISS data.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
31
Trend Analysis
Staff observed significant increases for all micromobility product-related injuries from
2017 to 2018, 2018 to 2019, as well as from 2020 to 2021. However, even after running
trend analyses using different statistical models (results for one of the models shown
below), no significant trend was observed at α = 0.05 level. This is counterintuitive;
however, staff opines a relatively small number of data points is the reason for absence
of significance. It is likely that additional years of data would produce significant results.
Trend Analysis Results Based on Unstructured Variance/Covariance Matrix
Effect
Parameter
Estimate
Standard Error
Degrees of
Freedom
t-value p-value
Intercept
-14,690,000
4,870,851
3
-3.02
0.06
Year
7,299
2,415
3
3.02
0.06
Source: NEISS, U.S. Consumer Product Safety Commission, 2017-2021.
Appendix C
Revised 2020 Special Study Results
Since the publication of the previous annual report in 2021, staff became aware of
errors in the estimate derivation process used for the 2020 special study. In the earlier
process, the sampling variations between subgroups were not retained. In this report,
staff included 17 additional cases, revised the process, and estimates. The revised
tabulations are presented below.
Table C2.1: Scooter Products: NEISS Product Code vs. Verified Scooter Product (2020)
Verified Product Overall
Powered
Scooter (5022)
Unspecified
Scooter (5024)
Various Other
Product Codes
Kick (unpowered) Scooter
381
19
357
5
Mobility Scooter
29
16
12
1
Moped
39
23
11
5
Hoverboard
16
4
8
4
E-Scooter
143
101
42
0
Gas powered scooter
17
14
3
0
Other/missing
26
11
15
0
Total
651
188
448
15
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on responses from the special study survey questions V3–V8.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
32
Table C2.2: Completed E-Scooter Investigations by Powered vs. Unspecified Scooter
(2020)
Investigation Overall Percent
Powered
Scooter
(5022)
Percent
Unspecified
Scooter (5024)
Percent
Completed Investigation
143
5%
101
12%
42
2%
Incomplete/Out of Scope
2,462
95%
732
88%
1,730
98%
Total Assigned
2,605
100%
833
100%
1,772
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Table C2.3: Distribution of Estimated Injuries by Rental Status (2020)
Rental vs Non-rental
N
Estimated ED-treated Injuries+
Percent
Rental
48
8,300
34%
Non-rental
95
16,200
66%
Total
137
24,500
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on the responses from special study survey question: S1 Which of the following best describes the scooter?
Rental, Owned by victim, Borrowed, Other, Don’t know.
+ Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table C2.4: Distribution of Estimated Injuries by How Victim Was Injured (2020)
Riding/Struck by
Scooter
N Estimated ED-treated Injuries+ Percent
Riding the scooter
134
24,400
96%
Struck by scooter
5
*
*
Other
4
*
*
Total
143
25,400
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on the responses from special study survey question: V10 You/the victim were/was: Riding the scooter; Struck
by scooter; Other.
+ Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
* Estimated ED-treated injuries under 1,200.
Table C2.5: Distribution of Estimated Injuries by Type of Riding Surface (2020)
Type of Surface
N
Estimated ED-treated Injuries+
Percent
Paved road
78
15,900
67%
Paved sidewalk
53
8,900
35%
Driveway
6
*
*
Other
7
*
*
Total
137
24,500
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on the special study survey question: A3 What type of surface were you/the victim on?
+ Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates; percentages may not sum to 100 percent due to rounding, and also
because some riders may have ridden e-scooter on more than one surface type.
* Estimated ED-treated injuries under 1,200.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
33
Table C2.6: Distribution of Estimated Injuries by Visibility (2020)
Visibility Issues
N
Estimated ED-treated Injuries+
Percent
No
108
16,800
71%
Yes
31
6,700
29%
Total
139
23,500
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on the special study survey question: A5 Was it dark or difficult to see?
+ Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table C2.7: Distribution of Estimated Injuries by Distraction (2020)
Distraction
N
Estimated ED-treated Injuries+
Percent
No
130
23,600
96%
Yes
5
*
*
Total
135
24,100
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on responses to the special study survey question: A6 Was there anything else occurring at the time of the
accident such as music, cell phone interference, or loud music?
+ Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
* Estimated ED-treated injuries under 1,200.
Table C2.8: Distribution of Estimated Injuries by Rider Baggage (2020)
Carrying/holding something
N
Estimated ED-treated Injuries+
Percent
No
106
19,300
84%
Yes
21
3,600
16%
Total
127
22,900
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on responses to the special study survey question: A8 Was there anything else occurring at the time of the
accident such as music, cell phone interference, or loud music?
+Injury estimates are rounded to the nearest 100 and may not sum to totals due to rounding. Percentages are
calculated from the unrounded estimates.
Table C2.9: Distribution of Estimated Injuries by Safety Equipment (2020)
Safety equipment
N
Estimated ED-treated Injuries+
Percent
Helmet
40
8,500
48%
Blinking Lights/Head lamp
24
4,800
65%
Reflective vest
4
1,700
6%
Knee/elbow pads
5
*
*
Other
6
*
*
Total
55
12,200
100%
Source: NEISS, U.S. Consumer Product Safety Commission, 2020.
Based on responses from the special study survey question: S12 I'm going to read a list of safety equipment that
riders might wear. Please tell me if the rider was wearing any of these at the time of the incident.
Riders may use more than one type of safety equipment as such rows do not add up to total.
+ Injury estimates are rounded to the nearest 100. Percentages are calculated from the unrounded estimates.
* Estimated ED-treated injuries under 1,200
.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
34
Special Study Survey Questionnaire
Q1 The U.S. Consumer Product Safety Commission (CPSC) collects data through the National Electronic
Injury Surveillance System (NEISS) on injuries treated in hospital emergency departments. CPSC
conducts follow-up investigations with a small number of people to learn more about how the injury
occurred. The results of these investigations will be used to determine if similar injuries can be prevented
in the future.
Your participation in this survey is completely voluntary and your identity and answers will be strictly
confidential. This survey will take between 10-15 minutes and data are used for statistical purposes only.
You should have received a letter with the following information needed to continue:
1. Investigation Task Number
2. Randomly generated password
To continue, you will have to enter the task number correctly below:
I1 CPSC would prefer that the person who answers this questionnaire is the actual person injured and
treated in the hospital emergency department. If the injured person is under the age of 16, CPSC would
prefer that a parent or guardian completes the questionnaire.
Was the injured person 16 years old or older?
Note: If you are the injured person and are under 16, please ask your parent or guardian to complete the
survey. If no one is available, it is okay to respond yourself.
o Yes (1)
o No (2)
I2 According to our records from the National Electronic Injury Surveillance System the injured person
was seen on {injury date} in the emergency department at {hospital name} for an injury that involved a
scooter. Is that correct?
o Yes (1)
o No (2)
o Don't know (3)
I3 What information is incorrect from the statement above?
○ Different date (1)
○ Different hospital (2)
○ (I/the victim) did not receive treatment in a hospital emergency department for a scooter injury (3)
I4 What is the correct date?
I5 Where did you/the victim receive treatment for your/their injury?
V1 Are you the:
o Injured person (1)
o Parent or guardian of injured person (2)
o Other (specify in next window) (3)
V2 Specify relationship:
V3 Was the scooter unpowered (e.g., a kick scooter or push scooter)?
Note: Powered scooters have a power source like electric or gas.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
35
o Yes (1)
o No (2)
o Don't know (4)
V4 Was the scooter an assisted mobility scooter to help people with physical limitations?
o Yes (1)
o No (2)
V5 Was the scooter a moped, motorcycle, or scooter that requires a registration or license?
o Yes (1)
o No (2)
V6 Did the scooter have handles for steering?
Note: Handles for steering are distinct from handles used for balancing purposes.
o Yes (8)
o No (9)
V7 If your scooter had only two wheels, were those wheels side-by-side?
Note: side-by-side wheels are distinct from wheels that are one in front of the other. Below is an example
of side-by-side wheels.
o Yes (1)
o No (2)
V8 What kind of power did the scooter run on?
o Gas (1)
o Electric (2)
o Other (specify in next window) (3)
V9 Specify.
V10 You/the victim were/was:
Interviewer instruction: If two scooters collided select "Riding the scooter."
o Riding the scooter (1) o Struck by scooter (2) o Other (specify) (3)
V11 Specify.
A1 Please describe how the accident happened. That is, what were you/the victim doing just before,
during, and just after the injury occurred? Please specify the location of the accident and any
environmental factors; such as weather, temperature, and anything else that may have contributed to the
accident.
A2 The following are specific questions about the incident that you may have already described. Please
bear with us as you fill out the next set of questions.
A3 What type of surface were you/the victim on?
○ Paved Road (1)
○ Paved Sidewalk (2)
○ Gravel (3)
○ Grass (4)
○ Driveway (5)
○ Other (specify in next window) (6)
○ Don't know (7)
A4 Specify.
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
36
A5 Was it dark or difficult to see?
o Yes (1)
o No (2)
o Don't know (3)
A6 Was there anything else occurring at the time of the accident such as music, cell phone interference,
or loud music?
o Yes (1)
o No (2)
o Don't know (3)
A7 Please specify the additional factors.
A8 Were/was you/the victim carrying or holding something such as a bag, purse, or backpack?
o Yes (1)
o No (2)
o Don't know (3)
A9 What were/was you/the victim carrying?
A10 Which of the following best describes how you were injured?
o Hit from the front (1)
o Hit from the side (2)
o Hit from behind (3)
o Other (specify in next window) (4)
o Don't know (5)
A11 Specify.
A12 Was there any warning before you/the victim were/was hit? (ex. bell, shouting, or other noise)
o Yes (1)
o No (2)
o Don't know (3)
S1 Which of the following best describes the scooter?
o Rental (1)
o Owned by victim (2)
o Borrowed (3)
o Other (specify in next window) (4)
o Don't know (5)
S2 Specify.
S3 Who was the scooter rented from?
S4 Do you know the brand and model names of the scooter or have a photo of the scooter involved in the
injury?
o Yes (1)
o No (2)
S5 Specify brand
S6 Specify model
Note: if brand is known but model is not, enter unknown below
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
37
S7 If you are able, please upload a photo of the scooter.
S8 It is very important for us to know what brands are involved in these injuries. Would you be willing to
go look at the scooter and record the brand name, model name, and take a photo of the scooter?
Note: You may also upload a pre-existing photo of the scooter if you have one. Select 'Yes' if you have a
pre-existing photo.
o Yes (1)
o No (2)
S9 Specify brand
S10 Specify model
Note: if brand is known but model is not, enter unknown below
S11 If you are able, please upload a photo of the scooter.
S12 Were/was you/the victim wearing any of these at the time of the incident. (Select all that apply)
○ Helmet (1)
○ Knee pads (2)
○ Elbow pads (3)
○ Wrist pads (4)
○ Reflective vest (5)
○ Blinking lights/Head lamp (6)
○ Other (specify in next window) (7)
○ None of the above (8)
S13 Specify.
C1 Is there anything else about this accident or the scooter involved that you would like to share?
o Yes (1)
o No (2)
C2 Explain.
C3 The following race and ethnicity questions will help the U.S. Consumer Product Safety Commission
better focus outreach and education efforts related to e-scooter safety.
C4 Are you/the victim Hispanic or Latino?
o Yes (1)
o No (2)
o Don't know (3)
o Prefer not to answer (4)
C5 What race(s) do you consider yourself to be? Please check all that apply.
○ White (1)
○ Black or African American (2)
○ American Indian or Alaska Native (3)
○ Asian (4)
○ Native Hawaiian or Pacific Islander (5)
○ Other (6)
○ Don't Know (7)
○ Prefer not to answer (8)
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
Micromobility-Related Deaths, Injuries, and Hazard Patterns | September 2022 | cpsc.gov
38
C6 Please specify "Other" race. Please be as specific as possible.
C7 We may be interested in sending a CPSC investigator to your home to gather more information about
how the accident occurred and take more detailed pictures of the scooter. This investigation would be set
up at your convenience. May we have an investigator contact you by phone to setup a visit?
o Yes (1)
o No (2)
C8 Please supply your phone number.
C9 When is a good time to call? (Check all that apply.)
○ Morning (1)
○ Afternoon (2)
○ Evening (3)
THIS DOCUMENT HAS NOT BEEN REVIEWED
OR ACCEPTED BY THE COMMISSION
CLEARED FOR PUBLIC RELEASE
UNDER CPSA 6(b)(1)
