Article Text

Epidemiological profile of hospitalised injuries among electric bicycle riders admitted to a rural hospital in Suzhou: a cross-sectional study
  1. Wei Du1,2,
  2. Jie Yang3,
  3. Brent Powis4,
  4. Xiaoying Zheng1,
  5. Joan Ozanne-Smith5,
  6. Lynne Bilston2,
  7. JingLin He6,
  8. Ting Ma7,
  9. Xiaofei Wang7,
  10. Ming Wu3
  1. 1Institute of Population Research, Peking University, Beijing, China
  2. 2Neuroscience Research Australia, University of New South Wales, Randwick, NSW, New South Wales, Australia
  3. 3Jiangsu Provincial Centre for Disease Control, Nanjing, China
  4. 4School of Postgraduate Studies and Research, International Medical University, Kuala Lumpur, Malaysia
  5. 5Department of Forensic Medicine, Monash University, Melbourne, Victoria, Australia
  6. 6The World Health Organization Beijing Office, Beijing, China
  7. 7The Third Municipal Hospital in Zhangjiagang, Suzhou, China
  1. Correspondence to Dr Wei Du, Institute of Population Research/WHO Collaborating Center on Reproductive Health and Population Science, Peking University, Beijing 100871, China; kelvindu{at} or weidu{at} Dr Ming Wu, Jiangsu Provincial Centre for Disease Control, Nanjing, 210009, China;


Police reports indicate an increasing burden of electric bike (E-bike) casualties in China; however, hospitalised injury data have not been reported. The aim of the present work was to describe hospitalised injury patterns for E-bikers involved in road crashes and explore injury risk disparities among them. For the period October 2010 to April 2011, this cross-sectional study retrospectively collected information for hospitalised E-bikers involved in road crashes from hospital records, in Suzhou China, using the International Classification of Diseases, 10th revision (ICD-10) injury diagnosis codes. Injury nature and body region were further categorised using ICD-10 codes. Multivariate logistic regression was used to assess the risk of specific injury types. We found that hospitalised E-biker injuries (n=323) accounted for 57.2% of road traffic hospitalisations over the 6-month study period. The average age, length of stay and hospitalisation cost were 43.8 years, 10.0 days and ¥8229 (US$1286), respectively. Fractures and head injuries were common. The odds of traumatic brain injuries were significantly elevated for night-time E-bike crashes and incidents other than colliding with motor vehicles. These findings confirm E-bike injuries as an important population health problem and identify elevated injury odds in different E-biker groups. Future injury prevention initiatives should include encouraging helmet use among E-bikers.

Statistics from


Road crashes are a population health problem worldwide, and are projected to become the fifth leading cause of global mortality by 2030.1 In 2011, the United Nations called for global action to improve road safety in the next decade (the so-called ‘Decade of Action for Road Safety’), with the overall goal to stabilise and then reduce the forecast level of road traffic fatalities around the world by 2020; specifically, to save 5 million lives and prevent 50 million serious injuries.2 Accounting for about 15% of road deaths worldwide,3 China, along with many other low-income and middle-income countries, joined the global action to reduce road crash casualties. China is also facing unsurprising growth in mobility demand and vehicle ownership in the course of rapid economic growth; for example, enormous mobility demand has resulted in 120 million electric bikes (E-bikes) operating in China by 2011, and this trend is continuing.4

An E-bike is a two-wheeled bicycle propelled in part or solely by electric power, popular across China for its convenience of use, affordability (approximately US$300 on average) and low energy consumption.5 An E-bike's greater travel speed and acceleration differentiate it from a traditional pedal bicycle, making it more akin to a motorcycle; the latter have been banned for use in many cities in China.6 According to China's rules of the road,7 E-bikes are categorised as non-motor vehicles and E-bikers are regulated as if riding bicycles, eg, ‘no mandatory helmet use required’. However, many E-bikers travel in the motor vehicle lanes, mixing with fast moving motor vehicles, which might partly explain why E-bike casualties have been mounting over time.8

During the period 2004–2008, national police records reported that E-biker fatalities and non-fatal injuries increased 5-fold from 589 to 3107, and 3-fold again from 5295 to 17 303, respectively.4 During the same period in Hangzhou, police data indicated the E-bike-related casualty rate increased by 2.7 per 100 000 persons per annum while the overall road crash casualty rate decreased between 2004 and 2008 (1.1 per 100 000 persons decrease per annum).9 Undoubtedly, E-biker injuries are becoming an increasingly heavy burden to individuals and communities in China. Considering the rapid growth of E-bike sales in China (eg, a 25% increase in 2010 alone)10 and worldwide (eg, an estimated 466 million by 2016),11 policies and strategies to improve E-bikers’ safety are urgently needed.

Injury prevention strategies rely on quality injury surveillance data to set priorities and inform future interventions.12 Despite previously reported injury profiles using police data,4 ,8 ,9 no study has evaluated the injury burden using hospital-based admission data from seriously injured E-bikers. Considering routinely collected hospital separation data provides details of injury diagnoses and cost of treatment relating to external causes,13 analysis of hospitalisation data may provide additional information to complement previously established evidence using police reports, due to differences in reporting practices.14 Therefore this study aimed to describe the epidemiological profile of E-bikers hospitalised for injuries following crashes and to examine correlating factors for different types of injuries. The findings may provide insights to establish interventions such as new legislation and enforcement activities in China in keeping with the Decade of Action for Road Safety.


Data sources

The study protocol was approved by the Jiangsu Provincial Centre for Disease Control Ethics Committee. We used inpatient hospitalisation data from the Third Municipal Hospital in Zhangjiagang, a rural county of Suzhou, China (figure 1). This hospital is one of the best facilitated comprehensive hospitals in Zhangjiagang, located to the east of the city, serving a catchment area of approximately 100 km2 and a population of about 255 000 rural residents, with an annual net income per capita of US$3000 in 2009. Hospital admissions were 11 441 (2010) and 12 351 (2011) respectively. This hospital was selected because of its injury reporting system, enabling us to identify E-bike crashes as external causes of injury. This capacity is generally lacking in hospitalisation coding systems among many hospitals that we assessed. For each road traffic injury-related separation during the period October 2010 to April 2011, we extracted recorded information including age, sex, residential status, insurance status; admission time, emergency status, medical diagnoses, discharge status, length of stay, external cause and direct hospitalisation cost (including personalised medication, examination, inpatient care and surgery, if any). At the time of separation, the reason for hospital admission was coded in prespecified categories according to mode of travel; and the primary and secondary injury diagnoses during hospital stay were coded according to the International Classification of Diseases, 10th revision (ICD-10).

Figure 1

Geographic location of Zhangjiagang county of Suzhou in Jiangsu province, China.

Case definitions

We used ICD-10 codes to identify E-biker injuries by four body regions (head and neck, trunk, upper limbs and lower limbs), and further grouped four specific nature of injury categories: superficial injuries (eg, bruises, abrasions, or contusions), traumatic brain injuries (TBIs; eg, selected skull fractures, concussion or other intracranial injuries), fractures and internal organ injuries (see online supplementary appendix). We categorised all injured body regions and nature of injury in a binary manner (ie, yes or no).

We further categorised subjects by age as children (aged <18 years), adults (aged 18–64 years) or seniors (aged ≥65); sex as male or female; residential status as local or other; insurance status as insured or not; admission time of day as morning (00:00 to 11:59), afternoon (12:00 to 17:59) or night (18:00 to 23:59); admission day of week as weekday (Monday to Friday) or weekend (Saturday and Sunday); emergency status as urgent or other; discharge status as routine discharge to home or self-care or expected to have further medical attention; and crash counterparts as motor vehicles or other (eg, colliding with other E-bikes or non-collision).

Statistical analysis

Statistical analyses were performed using SAS V.9.2 (SAS Institute, Cary, North Carolina, USA). We calculated frequencies and proportions for injured body regions and nature of injury. Individual injury type counts were based on the subset of total cases with the relevant injury (ie, injured body region or nature of injury), and multiple injuries within one study subject were counted separately. We further used multivariate logistic regression to estimate the adjusted OR with 95% CIs for injured body regions and nature of injury, controlling for variables including age group, sex, residential status, insurance status, admission day of week, admission time of day and crash type. We set p < 0.05 as statistically significant.


During the study period, a total of 565 hospitalisations resulting from road crashes (approximating 43.3% of all-injury and 8.0% of all-cause hospitalisations) were reported, among which E-biker hospitalisations accounted for 57.2% (n=323) compared with 14.3% (n=81) bicyclist hospitalisations. E-bike hospitalisations accounted for approximately 24.8% of all hospitalised injuries. The majority of hospitalised E-bikers (95.7%, n=309) were not insured and paid their hospital costs out-of-pocket (table 1). No E-bikers died in hospital. The direct hospitalisation cost for E-bikers was approximately ¥2.7 million (US$400 000), which accounted for 50.0% of that for all road crash casualties (¥5.4 million US$800 000).

Table 1

Distribution of hospitalisation characteristics among E-bikers (n=323)

Approximately 48.6% (n=157) of injured E-bikers were involved in collisions with motor vehicles, 70.6% (n=228) were weekday crashes and 45.5% (n=147) night-time crashes (table 2). The head was the most common body region injured (46.4%, n=150) and more than one-third of the study population (35.9%, n=116) experienced traumatic brain injuries (eg, selected skull fractures, concussion or other intracranial injuries) (table 2). Fractures (51.1%, n=165) were the dominant nature of injury followed by superficial injuries (42.1%, n=136) (table 2).

Table 2

Distribution of injured body regions and nature of injury by characteristics of E-biker hospitalisations (n=323)

Compared with other types of crashes, colliding with motor vehicles was associated with elevated odds of superficial injuries (unadjusted OR: 2.27, 95% CI 1.44 to 3.56; adjusted OR 2.44, 95% CI 1.51 to 3.94) (tables 3 and 4). E-bike crashes other than colliding with motor vehicles were associated with increased odds of TBI (tables 3 and 4). Elevated odds of hospitalised head injuries including skull fractures, concussion or other intracranial injuries were associated with E-bike collisions at night rather than in the afternoon; whereas elevated odds of fractures was associated with local E-bikers compared with non-local residents (tables 3 and 4).

Table 3

Unadjusted ORs and 95% CIs for different injured body regions and nature of injury (n=323)

Table 4

Adjusted ORs and 95% CIs for different injured body regions and nature of injury (n=323)


The current findings profile the hospitalised injuries sustained by E-bikers admitted to the Third Municipal Hospital in Zhangjiagang, particularly with regard to the injured body regions and nature of injury, and demonstrate injuries to E-bikers are a significant population health burden. Consistent with the previous finding across China that E-bikers accounted for 53.4% of non-motor-vehicle-related crash casualties using published police data in 2008,8 E-bikes were the dominant mode of transport (57.2%) among all hospitalised injuries resulting from road crashes. This finding also echoes the previous finding that there may be a shift from bicycle injuries to E-biker injuries in China.9 Allowing for the epidemic problem of injuries to bicyclists in previous years,15 injuries to E-bikers may be expected to rise given the rapid growth of E-bikes on Chinese roads, especially when known effective countermeasures such as compulsory helmet use are lacking and general road safety rules such as speeding are not widely enforced for on-road E-bikers.16 Should all E-biker injuries be prevented in the current study, a direct hospitalisation cost of about ¥2.7 million could be saved before considering the indirect burden to individuals, their families, the communities and the wider healthcare system. Therefore, to inform policymakers, understanding the effectiveness of any E-bike injury countermeasures (eg, population-based comprehensive safety strategies including education, training, legislation and enforcement) would be the next step to quantifying the population-level benefit in terms of injury risk reduction and financial savings. Although proven countermeasures such as mandatory helmet use have not yet been established in China for E-bikers, evidence could be gleaned through population-based intervention trials with a focus on helmet use among E-bikers and speed limiters on E-bikes.

The finding of a high proportion of head injuries may relate to non-wearing of helmets while travelling on E-bikes, which is supported by a recent Suzhou study where only a 9% helmet-wearing rate among E-bikers was observed. Since hazards and impacts are likely to be similar, evidence of effectiveness of helmets in reducing bicyclist17 and motorcyclist18 head injuries clearly supports helmet use for E-bikers. The most likely explanations of the higher odds of traumatic brain injury for non-motor vehicle collision victims admitted to hospital are (i) the likely high case fatality rate among these vulnerable road users, with deaths from head injuries sustained in high energy collisions, particularly with motor vehicles occurring at the scene, before reaching hospital, or in the emergency department; (ii) some TBIs may have been relatively minor concussions, incidental to more severe fractures or other injuries that led to hospital admission. A Brazilian study of road traffic injuries reported an approximate 5% case fatality rate among bicyclists and pedestrians, followed by motorcyclists (3%).19 We found no E-biker fatalities among hospital admissions, which supports this interpretation of our findings. Despite some limitations, the current study complements existing evidence from police data,8 ,9 provides evidence of the need for further investigation, and points out the importance of head injury prevention.


The current study setting may not be generalisable to other contexts. Hospital admission criteria and length of stay may differ between hospitals. For example, the study hospital is a rural hospital catering for a rural population whose case mix and practice may differ from metropolitan counterparts. Exposure of E-bikers may vary by distance travelled, number of trips per rider and number of passengers carried in different locations. In addition, the current study covers only the cooler seasons (October to April), whereas injury risk exposure may increase in warmer months (May to September). Hence further studies are needed to address injury severity and seasonality variations. Nevertheless, the study finding of a high proportion of E-biker injuries among admissions due to road crashes (57.2%) indicates safety issues related to E-bike riding in China and is supported by emerging studies from police data and elsewhere in China.8 ,9 Another limitation is that while hospitalisation data can be used to describe the problem it cannot determine underlying causes of the observed injuries. Nor is such a study able to identify accurate exposure variables. For example, the hospital data include extensive detail of injury diagnoses and direct medical cost but lack details of in-crash factors such as the use of protective equipment and exposure or behavioural details such as travel speed. Nor does hospital data generally include blood alcohol levels for injured E-bikers. In addition, the current data did not specify travel mode (eg, drivers or passengers) or categorise crash type except for collisions with motor vehicles (eg, non-collision, multiple E-bike collision, or collision with fixed objects). Although the study found little direct evidence of exposure–injury causation, the common head injuries including skull fractures and intracranial injuries among hospitalised E-bikers warrant particular attention and further investigation, which may include a large-scale study using improved data collection methods and/or focusing on a helmet intervention trial.


This study found hospitalised injuries to E-bikers, especially traumatic brain injuries, were predominant for urgent attention among serious non-fatal road traffic injuries in Suzhou, China. It highlights the need to establish injury prevention policies and strategies for E-bike safety. Development and implementation of initiatives such as compulsory helmet use, speed limit enforcement and behavioural education should be considered to reverse the trend of the current increasing burden of E-biker injuries in China, echoing the initiatives of the Decade of Action for Road Safety.

What is already known on this subject

  • The popularity of electric bikes (E-bikes) is increasing in China and elsewhere in the world.

  • In China, police data show E-bike-related casualties to be increasing over time.

What this study adds

  • E-biker injuries represent a high proportion (57%) of serious non-fatal road traffic injuries in a rural hospital in Suzhou, China.

  • The head was the most commonly injured body region (46.4%), and traumatic brain injuries (TBIs) were common (35.9%) among hospitalised E-bikers.

  • Consideration of future injury prevention policies should include compulsory helmet use.

Maintaining declining teen driving deaths

In the USA, the number of teens who die in motor vehicle crashes has fallen by 64% since 1975. However, as Baker and Bishai observe in the Christian Science Monitor (22 June 2012), between 2007 and 2009, the death rate from crashes for teens dropped almost 30%. They attribute much of this to ‘graduated driver licensing’ but also to the economic recession. The increase in unemployment in 2008 coincided with a decline in car travel, a decrease in alcohol consumption, especially in bars and restaurants, and with a decrease in highway deaths. Accordingly, they propose increasing taxes on alcohol (higher prices reduce consumption) and gasoline, limiting where alcohol sales are permitted, and restricting alcohol advertising, especially ads aimed at adolescents. (noted by IBP).


We appreciate special support from the WHO China Office, the Ministry of Health in China, Jiangsu Centre for Disease Control, Suzhou Centre for Disease Control, Suzhou Bureau of Health and Suzhou Bureau of Public Security. We also thank Yilan Liao, Ning Li, Yihe Hu, Ziyi Jin, Jianfeng Liu, Qi Zhang, Xianglin Liu and Yan Lu for their assistance. WD and LB are supported by NHMRC fellowships.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors WD and JY contributed equally to the conceptual development of this work, conducted the literature review and data analysis, completed full first draft and contributed to subsequent drafts; all the other authors contributed to the conceptual development, data interpretation, critical revision of the first draft and subsequent drafts.

  • Funding This project received funding support from Bloomberg Philanthropies as part of the Road Safety in 10 Countries Project in China.

  • Competing interests None.

  • Ethics approval The Jiangsu Provincial Centre for Disease Control Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.