Epidemiology of Fractures Following Electric Scooter Injury

Abstract

Background

The incidence of injuries associated with the use of electric scooter (e-scooter) has consistently increased since the launch of the first shared e-scooter systems in late 2017. The aim with this review was to investigate the epidemiological characteristics of fractures related to e-scooter accidents.

Methods

A systematic review of the all published papers was conducted regarding epidemiologic characteristics of traumatic fractures following e-scooter-related injuries. Multiple databases of PubMed, Web of Science, EMBASE, and Cochrane library were searched and relevant clinical studies reporting incidence for traumatic fractures from e-scooter-related injuries were included. All other types of studies and those including patients with non-orthopedic injuries were excluded. The Methodological Index for Non-randomized Studies (MINORS) was used to assess the quality of the included studies. The characteristics of patients were analyzed by the mean and the standard deviation for continuous variables.

Results

Eleven published studies (6579 patients) were identified which have been reported from Nov, 2019 to Sep, 2021. Periods of research in each study were different from May, 2017 to July, 2020. In the majority of cases, the mechanism of injury was a fall from the scooter (84.8%), which was followed by e-scooter motor vehicle accident (7.4%) and collision (5.5%). Regarding the fractures, upper limb injuries were the most prevalent (65%) and radial fracture was the most common fracture type of the upper limb (28.2%). Lower limb fractures were recorded in 475 (25.3%) patients, the most common being tibial fractures (8.4%).

Conclusions

As e-scooter market continues to considerably rise, the incidence of major orthopedic traumas is also likely to increase. Despite potential selection and reporting biases, our systematic review suggested up-to-date guidelines for the epidemiology of fractures following e-scooter injury. According to the results found in the present study, we believe that it is reasonable to mandate helmet use and the obligation to wear other types of safety equipment such as elbow and knee pads should be discussed.

Introduction

Electric scooters (e-scooters) are convenient and offer a sustainable urban mobility solution (Fig. 1). The widespread use of e-scooters as economical, accessible, and eco-friendly vehicles has changed the patterns of public transport since their launch worldwide [1]. According to the National Electronic Injury Surveillance System (NEISS), the estimated number of emergency department (ED) visits for e-scooter injuries increased from 4881 (95% CI 4086–5676) in 2014 to 29,628 (95% CI 14 919–44 338) in 2019, and the incidence of e-scooter accidents nearly doubled between 2018 and 2019 [2]. Over the past few years, e-scooters have emerged as a new, popular means of transportation in both the United States (US) and Europe [3, 4]. They have expanded to many metropolitan areas throughout the world and have changed the commuting patterns in many cities [5]

Fig. 1.

E-scooter is a motorcycle with a handlebar, a platform for the rider's feet, a motor and a brake

Even though e-scooters provide a convenient and eco-friendly mode of transportation, they pose safety issues and meaningful health problems. Previously, two studies have reported the musculoskeletal injuries related to e-scooters as the 2nd most common cause for trauma surgery in 2019 [5, 6]. While e-scooters are growing in popularity, it is essential to emphasize the risks and support suggestions for their proper use. Nevertheless, real-world data relating to injury patterns associated with e-scooter accidents are limited, and few studies have assessed the incidence of orthopedic fractures. In addition, distribution of the fractures with regard to body region has been inconsistently described in previous researches as only patients who admitted to one large hospital in one country were examined [7].Recent data from a prospective observational multicentre study indicated that the lower limbs were affected more frequently than the upper limbs (34 vs 30%) [8]. On the contrary, Störmann et al. [9] reported that the most commonly affected body region after injury was upper extremity, followed by injuries to the lower extremity and head. Current evidence regarding trauma frequency following e-scooter injuries depends on a small number of retrospective studies [10]. Our knowledge with regard to the characteristics and injury patterns in e-scooter-related accidents is limited to a few cases from single centers, although the incidence of injury may differ between individuals of different countries and ethnic backgrounds. Analysis of more institutions’ data of different nationalities may provide more universal and reliable results.

In this study, we performed a systematic review of the recent publications regarding traumatic fracture after e-scooter-related accidents to analyze and present epidemiological data of fractures following e-scooter injuries. In addition, we asked what proportion of these specific injuries requires surgical treatment.

Materials and Methods

Literature Search

A systematic review was conducted as recommended by the preferred reporting items for systematic reviews and meta-analyses guidelines [11]. An analysis of the available studies using different search strategies was performed. Web of Science, PubMed, Cochrane library, and EMBASE were searched using keywords “electric scooter”, “e-scooter”, “injury”, “trauma”, and “fracture” in different combinations. Spelling variations or synonyms for each search term were also used to capture as many studies as possible. No regional or language restrictions were applied. According to the Cochrane Collaboration [12], two authors evaluated the titles and abstracts of all relevant publications independently. We conducted the literature search until December 31, 2021.

Criterion for Inclusion

All full-text, peer-reviewed articles about fracture patients following e-scooter injury were reviewed. The identified studies were evaluated for inclusion by two reviewers independently. Inclusion criteria of this review were a specific focus on epidemiologic characteristics of orthopedic fractures associated with e-scooter collisions. All publications reporting fractures after accidents were included. The exclusion criteria were (1) articles that had no detailed data on the patterns of fractures (e.g. radial fracture or tibial fracture) [13, 14], (2) review articles, (3) articles related to non-orthopedic injuries [15], (4) non-English articles, and (5) case series included fewer than 10 cases [16].

A total of 817 articles were identified after conducting the database search. A systematic review flowchart is shown in Fig. 2. A total of 558 unique records were reviewed as 274 publications appeared in more than one of the four database searches. In addition, references from selected studies were examined to look for relevant publications that might be missed in the aforementioned database search. Finally, a total of 11 publications were included after rigorous exclusion criteria were applied. All included articles were freely available and there were no paid articles in this review.

Fig. 2.

Systematic review flowchart following the PRISMA guideline. PRISMA: preferred reporting items for systematic reviews and meta-analyses

Data Collection and Analysis

All data related to the study outcomes were extracted from each study by one author and were reviewed by two authors independently. The following data were obtained from each study: author, country, year of publication, data collection period, demographics of the patients (age, gender), study design, injury rate, traumatic mechanism, helmet use, type and frequency of fractures, treatment, and outcomes after surgery. Because fracture incidence was consistently reported in each study, we pooled these data to derive mean values. In the present study, all fractures were classified into twelve groups, according to anatomical area: clavicle, humerus, elbow, forearm, hand (carpal, metacarpal, phalanges), spine, pelvis, femur, knee (including patella), tibia, ankle, and foot. Despite limited response from data sources, we also tried to collect data on compounding factors regarding mechanism of injury (MOI) such as a fall from e-scooter and motor vehicle accidents.

Study Quality Assessment

The Methodological Index for Non-randomized Studies (MINORS) [17] was used to assess the quality of the included studies and was scored from 0 to 24 [17]. There are 8 items that are specific to non-comparative studies. Each item was scored as 0 (if not reported), 1 (when incomplete), and 2 (when completely reported). The maximum score was 16 for non-comparative studies and 24 for comparative studies. Two authors independently assessed the quality of the included studies. Any disagreement was resolved by a third author. All studies in our systematic review were non-randomized and major concerns included the lack of a prospective cohort study. The mean MINORS score was 9 ± 3 points. A meta-analysis could not be performed and effect sizes could not be obtained because there was substantial variability in the methodology of the included articles and heterogeneity in data reporting (I² value of 72% was identified according to the test statistic for evaluating heterogeneity).

Statistical Analysis

Excel spreadsheet was used to record extracted data. Owing to considerable heterogeneity in the data reported as mentioned, only simple descriptive statistics were used to summarize primary findings. The characteristics of patients were analyzed by the mean and the standard deviation for continuous variables. Descriptive statistics of the number of individuals treated for fractures, was performed using frequency and percentage.

Results

Search Results

Literature search of multiple databases yielded a total of 6579 patients from 11 publications. Two studies [18, 19] limited to only upper extremity fractures (without data on lower extremity injuries) were excluded. Publication dates ranged from Nov, 2019 to Sep, 2021. Disagreements were resolved by discussion between the two authors in the quality assessment of included articles. Even though complete data could not have been extracted in all included studies, data such as anatomical area and the incidence of fractures were clearly reported in detail in all the selected articles.

Demography

Most articles retrieved patients’ data retrospectively through ED or trauma services databases. Periods of research in each study ranged from May, 2017 to July, 2020. No prospective study was identified and the individuals treated for traumatic fractures were included in this review. Age, sex, and characteristics of patients included are shown in Table 1. The results show that the mean age of each study ranged from 20 to 30 years and male patients represented more than 50% of the injured patients. Four studies [5, 20–22] were conducted in USA. One multi-center [8] study was included.

Table 1.

Demographic information in the studies group (NA, not available)

Author	Journal	Year	Country	Time interval	Total patients	% Male	Age
Tischler et al.	J Clin Orthop Trauma	Sep, 2021	USA	Jan 2019 to December 2019	1541	56.7	23
Coelho et al.	J Clin Med	July, 2021	Spain	May 2019 to May 2020	397	59.9	30.8 ± 10.1
Shichman et al.	J Bone Joint Surg Am	June, 2021	Israel	May 2017 to February 2020	3331	NA	33.9 ± 11.49
Uluk et al.	Emerg Med J	June, 2021	Germany	June 2019 to December 2019	248	52	29
Mukhtar et al.	Clin imaging	June, 2021	USA	September 2018 to December 2019	140	56.4	28
Harbrecht et al.	Eur J Orthop Surg Traumatol	June, 2021	Germany	July 2019 to July 2020	59	40.7	30.03 ± 9.32
Nielsen et al.	Dan Med J	Jan, 2021	Denmark	June 2019 to September 2019	49	75.5	26
Siow et al.	J Orthop Trauma	Nov, 2020	USA	November 2017 to January 2020	485	61.9	37.1 ± 14
Störmann et al.	J Clin Med	May, 2020	Germany	June 2019 to March 2020	76	69.7	34 ± 12.4
Ishmael et al.	J Bone Joint Surg Am	Mar, 2020	USA	September 2017 to August 2019	73	50.7	35.4
Brownson et al.	N Z Med J	Nov, 2019	New Zealand	October 2018 to February 2019	180	60	28

With regards to the injury mechanisms, eight studies (n = 3280 participants) reported MOI in all riders with e-scooter accidents (Table 2). 84.8% were attributed to ground level fall or crash, 7.4% were e-scooter-motor vehicle accident, and 5.5% were due to collision. Other reasons for MOI (2.4%) included scooter malfunction, medical event, and incomplete information.

Table 2.

Mechanism of injury

Author	Rider fall (%)	Hit by a vehicle (%)	Collision (%)	Others (%)
Tischler et al.	1342 (87.1)	107 (6.9)	63 (4.1)	29 (1.9)
Coelho et al.	341 (85.9)	18 (4.5)	38 (9.6)	0 (0)
Shichman et al.	613 (85.6)	78 (10.9)	25 (3.5)	0 (0)
Uluk et al.	141 (57)	21 (8.5)	45 (18)	41 (16.5)
Nielsen et al.	43 (87.8)	0 (0)	5 (10.2)	1 (2)
Störmann et al.	70 (92.1)	5 (6.6)	1 (1.3)	0 (0)
Ishmael et al.	65 (89)	8 (11)	0 (0)	0 (0)
Brownson et al.	165 (91.7)	5 (2.8)	3 (1.7)	7 (3.9)
Total	2780 (84.8)	242 (7.4)	180 (5.5)	78 (2.4)

The Frequency and the Type of Fractures

Ten articles reported all cases with e-scooter-associated injury and the number of individuals who sustained fractures (Table 3). Overall, there were 6778 total injuries (number of total injuries was not reported in one study [5]). Among them, fractures were found in 27.8% (1881/6778) of patients and 36.2% (range: 7.7–62%) required surgical treatment due to a fracture. Craniofacial injuries were reported in 8 studies, and only 5.3% of the patients had worn a helmet (76/1445).

Table 3.

Patient and accident characteristics associated with e-scooter-related injuries (NA, not available)

Author	Total injuries	Total fractures	% of fractures	Cranial or facial fracture	Helmet use	Surgery
Tischler et al.	1541	543	35.2	24	NA	NA
Coelho et al.	422	193	45.7	2	55	98
Shichman et al.	3331	716	21.5	6	NA	225
Uluk et al.	323	59	18.3	NA	3	NA
Mukhtar et al.	198	75	37.9	43	NA	NA
Harbrecht et al.	99	16	16.1	NA	0	5
Nielsen et al.	58	13	22.4	1	NA	1
Siow et al.	485	157	32.4	8	14	49
Störmann et al.	100	29	29	10	1	18
Ishmael et al.	NA	75	NA	NA	NA	NA
Brownson et al.	221	80	36.2	22	3	40

The upper limbs were the primary body region affected, with a total of 1272 fractures (65%). A radial fracture was the most common type of upper extremity fractures and overall (551 fractures [28.2%]). Lower limb fractures were recorded in 475 (25.3%) patients and the most common type of lower extremity fracture was tibial fractures (164 fractures [8.4%]) (Table 4). The exact types of surgical procedure or outcomes after surgery were not consistently described in all reports, making collection and analyzing of these data challenging.

Table 4.

The distribution of fracture sites in each study

Author	Clavicle	Humerus	elbow	Radius (ulna)	hand	Spine	Pelvis	Femur	Knee	Tibia (fibula)	Ankle	Foot
Tischler et al.	23	39	62	250	29	19	15	3	11	24	32	36
Coelho et al.	9	13	5	77	20	2	1	10	9	28	15	4
Shichman et al.	42	40	112	112 (53)	133	9	5	35	15	50 (20)	31	7
Uluk et al.	3	7	9	12	11	0	1	2	0	5	6	3
Mukhtar et al.	6	1	19	11(5)	14	0	0	1	0	9 (2)	3	3
Harbrecht et al.	0	0	8	2	1	0	0	0	1	1	1	2
Nielsen et al.	0	0	3	1	8	0	0	0	0	0	0	0
Siow et al.	13	3	0	47 (9)	29	8	2	11	0	14	11	10
Störmann et al.	0	7	7	8	0	0	1	0	2	0	3	1
Ishmael et al.	3	0	10	15	4	0	0	11	6	17	6	3
Brownson et al.	4	3	10	16	11	1	0	2	0	16	6	9
Total (%)	103 (5.3)	113 (5.8)	245 (12.5)	551 (28.2)	260 (13.3)	39 (2)	25 (1.3)	75 (3.8)	44 (2.2)	164 (8.4)	114 (5.8)	78 (4)

Discussion

The use of e-scooters has increased considerably after the introduction of the shared e-scooter services [23–25]. Accordingly, the number of fractures from accidents has increased substantially and several previous publications have reported the increase in injuries associated with e-scooter use, especially on emergency department [1, 26]. However, few literatures have specifically focused on the characteristics of scooter-related fractures. The purpose of this review was to investigate previous reports on epidemiological characteristics of fractures following e-scooter accidents to suggest improvements for the use of e-scooters.

The main finding of this study is that orthopedic fractures are very common owing to the high incidence of severe accidents following risky behaviors associated with scooter riding, and they were found in one in four cases (27.8%) of overall injured patients. Our study has shown a significant number of electric scooter-related fractures (range 16.1–45.7%). Furthermore, 36.2% of the patients with fractures required surgical treatment in our review. It has been reported that nearly 50% of the patients admitted from injuries associated with e-scooters riding had extremity fractures as their initial diagnosis [27]. Compared to craniofacial trauma, these individuals were much more likely to need surgical treatment during same hospital admission (odds ratio of 9.5 versus 1.6) [27]. Even though the majority of injuries following scooters accidents might be minor injuries such as contusion and abrasions, e-scooter riding always has risks of sustaining major limb injuries, especially under the influence of alcohol [28] or when breaking local laws [14]. Alcohol is assumed to be a major contributing factor to severe injuries. According to a recent research from US, alcohol abuse was identified in 45.8% of accidents (93 of 206 cases) [29]. Some patients may experience high-energy trauma following e-scooter accidents. Reportedly, 12% of patients undergoing surgical procedures following e-scooter-related injuries had open fractures and 9% of patients had hip fractures [5].

The incidence of scooter-related extremity injuries has also increased dramatically in the US over the last decade, and the upper limb fractures including wrist, elbow, and hand are especially common, making up over 50% of all e-scooter-related injuries [19]. With regard to upper limb and hand injury, contradictory results have been reported in the literature. Recently, after a retrospective review of 3331 cases, Shichman et al. [30] reported that there were 433 (72.3%) upper limb fractures and 166 (27.7%) lower limb fractures out of a total of 599 fractures. LaGreca et al. [19] demonstrated that between 2010 and 2019, the incidence of scooter-related injuries in the US has risen by more than 200% with the wrist and forearm as the most commonly injured body parts. Similarly, in our pooled analysis, the most common fracture sites following e-scooter accidents were the upper limbs as expected. Any type of injury may occur after falling off a two-wheeler. In our cohort, the high prevalence of upper extremity fracture may be attributed to the common MOI (rider fall) [31]. A sudden and rapid loss of balance obviously lead to the rider’s instinct to fall onto an outstretched hand to avoid head and body injuries. As a consequence, it seems that hand and radial fractures emerge as the most common fractures [18]. Recently, it has been reported that the injury patterns of e-scooter and other 2 wheeler (bicycle)-related trauma were different and the two groups of patients should not be regarded as a single trauma entity [32]. Compared to bicycle accidents, e-scooter users suffered more commonly head and face injuries due to less frequent helmet use [32]. E-scooter accidents more frequently occur in association with alcohol and on weekends [28]. In addition, an increased risk of injury to the tibialis posterior tendon from sharp-edged components of the e-scooter was demonstrated. Bicycle-related injuries decreased in last decade despite increased public utilization [33]. In this study, we tried to investigate and review the injury patterns related to e-scooter accidents which are increasing dramatically.

According to our pooled analysis, lower limb fractures were recorded in 475 (25.3%) patients and the most common fracture of the lower limb was tibial fractures. Even though the incidence is relatively low (reportedly 20–40% [8, 34]), the lower limb fractures should not be ignored because they often need higher rates of admission and longer hospital stay than upper limb fractures. The observed fracture patterns of lower extremity are mostly secondary to high-energy trauma, which is stressed by the higher Injury Severity Score (ISS) and the high rates of open fractures [7]. Consequently, in the majority of cases, fractures of lower extremity were mostly need surgical treatment compared to fractures of upper extremity. Furthermore, treatments of these fractures often requires a staged surgical procedure in which initial temporary stabilization device is applied at the early stage and later, as the soft tissue swelling goes down, is converted to definitive internal fixation [10, 35].

Not surprisingly, one thing that should be noted in the present study is the fact that wearing a helmet on a scooter appears to be extremely rare, with several publications reporting safety compliance as low as 1 ~ 2% [9, 13]. Head injuries are known to be the most common injuries in individuals with e-scooter-related accidents. In all patients visiting the emergency department with scooter-related trauma, almost 30% of them were diagnosed with head and neck trauma [36]. Low helmet compliance level contributing to high rates of associated head injuries has been reported in many literatures [22, 37]. It can be argued whether it should be mandatory or not. We believe that it is reasonable to mandate helmet use in scooter riders and it is the only realistic solution. At least, an opportunity to rent a helmet should be provided when riding a scooter. In addition, the obligation to wear other types of safety equipment could also be discussed. For instance, wearing elbow or knee pads should be encouraged to avoid severe limb injuries as per extreme sports safety recommendations [38].

Limitations of the Study

There are some limitations to this review that ought to be considered. First, most publications included in our analysis were retrospective cohort studies that only reported on clinically relevant variables present in medical databases and thus both selection bias and reporting bias cannot be avoided. Second, owing to substantial heterogeneity in data reporting of the included articles, our study did not include any statistical analysis or hypothesis testing. Because our goal was to describe patterns of fractures, not to verify hypotheses, only simple descriptive statistics were used. Nevertheless, we believe that our systematic review suggested up-to-date guidelines for the epidemiological characteristics of fractures following e-scooter injury.

In conclusion, injuries occurred while riding scooters are a rapidly growing health issue, in spite of existing regulations. Based on our systematic review, orthopedic fractures requiring surgery are very common after e-scooter accidents, and the upper limbs are especially vulnerable with falls being the most common MOI. Physicians should be alert to and seek associated fractures during initial assessments of e-scooter riders. Additional clinical trials involving larger number of patients and including a wide range of ethnic backgrounds will demonstrate better evidence on the fracture patterns and severity of e-scooter-related injury, and help improve our ability to address this emerging phenomenon.

Funding

None.

Declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

This article does not contain any studies with human or animal subjects performed by the any of the authors.

Informed Consent

For this type of study informed consent is not required.