https://orcid.org/0000-0002-3291-8006
Abstract
Objective
Bus travel is relatively safe: however there remains a lack of understanding of passenger injury incidents onboard buses. The objective of this study was to understand more about onboard passenger incidents to help inform injury mitigation.
Methods
The UK national STATS19 data and Transport for London bus incident data (IRIS) were used to determine the size of the problem in Greater London. Other data including onboard incident reports from two bus operators and CCTV footage of 70 incidents were used to understand passenger injury in more depth and identify common themes and challenges.
Results
The STATS19 and IRIS analysis showed that there was a difference between nationally reported bus incidents compared to locally reported bus incidents. Non-collision incidents are prevalent in the data suggesting there is a large problem to tackle. The CCTV and bus incident data identified braking to be the single largest problem in onboard bus passenger injury incidents. Inconsistent reporting of passenger incidents and injury descriptions make it difficult to identify injury patterns and trends. Areas on the bus appear to contribute to higher injury incidents namely those seats facing and closest to the wheelchair area. Other challenges relating to expected passenger and driver behaviors were noted where blame for the incident and outcome can be attributed to both parties.
Conclusions
This combined analysis of incident reports and CCTV footage has enabled a better understanding of the events leading to on-board passenger injury incidents. Preventing harsh braking would appear to be the most effective way of reducing passenger injuries. Additionally improved data collection would assist both transport authorities and bus operators to identify and monitor the effect of bus safety improvements.
Introduction
There were over 4.3 billion bus passenger journeys in England in 2018-2019 (pre-COVID 19), over half of which occurred in London (DfT Citation2019). Buses are also considered one of the safest ways to travel in England, having the lowest fatality rate per billion miles traveled compared to all other transport modes (DfT Citation2019). Additionally, bus use is predicted to rise in the coming years, with Government campaigns encouraging people to use buses more instead of cars in concerted efforts to reduce congestion and carbon emissions. Last year the Department for Transport launched the ‘Bus Back Better’ campaign (DfT Citation2021), with aims to improve bus services across the country and increase passenger numbers.
However, injuries to bus passengers are still of serious concern; a previous review estimated over 6000 people a year were injured on UK buses (Kendrick et al. Citation2015). In particular, there is a need to understand injuries resulting from non-collision events, which are believed to represent the majority of bus passenger injuries in England, but which are often under-reported in national statistics (Kirk et al. Citation2003, Barnes et al. Citation2016, Edwards et al. Citation2019). Previous research found that older bus passengers were more at risk of injury and that sudden harsh braking or acceleration is commonly the cause of injury (Kendrick et al Citation2015, Barnes et al. Citation2016). Therefore, research into how these injuries occur, and how to prevent them in the future, is critical to the success of any campaigns to increase bus passenger numbers, as well as to ensure the safety of a growing group of transport users.
Recent research analyzed CCTV footage of passenger injury incidents on London buses (Edwards et al. Citation2019). This research specifically focused on the bus interior and passenger contact as injury mechanisms and suggested countermeasures to reduce injuries, e.g., handrail position. However, to date little in-depth research has been carried out to understand the injury mechanisms in direct correlation with the behavior of the passengers and driver (Barnes et al. Citation2019). Therefore, the objective of this research was to derive a better understanding of the nature and circumstances of bus passenger injuries in London, using in-depth data including CCTV footage, investigator notes, and incident report notes from bus operators.
Methods
For this analysis buses were defined as public transport buses with 4 seats per row with a central aisle, with some variation at the front of the bus depending on the luggage provision. To define the extent of the problem the UK national data (STATS19) was analyzed for Greater London as was the Transport for London (TfL) Incident Reporting Information System (IRIS) database for a 5-year study period 2012-2016. To understand the incident and injury causation in detail, a sample of 500 anonymised onboard bus incident reports were obtained from two bus operators in Greater London. The bus operators were asked to provide a sample of incident reports (excluding assaults or vandalism) for the same 5-year period. Both operators provided the incident data reports as Excel spreadsheets containing investigation notes and driver statement notes for passenger injury incidents. One operator supplied detailed incident investigation forms containing general notes, summary notes and where available, driver and passenger interviews (n = 300) compared to the other operator. Additional CCTV data was provided for 70 incidents by the same bus operators to help understand the passenger and driver behaviors onboard.
All bus operator incident data were analyzed using a qualitative approach and all data were uploaded to the software package NVivo. For consistency the sections comprising overall summary and investigation notes were included in the analysis. A theory driven thematic analysis was undertaken to understand what contributes to injury causation (Braun and Clarke V Citation2006). Overarching themes of bus, driver and passenger behavior were used to initially explore the data and more detailed sub themes identified to provide an overall understanding of the problem (Boyatzis Citation1998).
CCTV data were analyzed by at least two researchers identifying the following patterns of behavior on the bus; Passenger position and activity pre and post incident; driver behavior pre and post incident; deceleration speed and time to passenger incident; contact points for passengers. No assumption of injury severity was made for the passengers as this cannot be determined from video data.
Results
Extent of the problem
Analysis of STATS19 data for the 5-year period 2012-2016 (inclusive), showed that there were 14,462 accidents reported to the police in Great Britain where at least one bus or coach passenger was injured. Of these, 5,699 (39%) were in the Greater London area and the majority 93.8% were defined as ‘Slight’ accidents (). In contrast the IRIS data for the same period reported 20,490 incidents involving at least one passenger on the London bus network (this excludes non passenger incidents including vandalism, driver assault etc). typically shows the STATS19 options for passenger location at the time of the incidents and standing/sitting passengers predominated for Greater London passengers
Table 1. Injury severities of accidents involving bus passenger casualties in London (n = 5699) (STATS19 data 2012–2016).
Table 2. Passenger location at time of injury for bus passenger casualties in London (n = 7043) (STATS19 data 2012–2016).
The IRIS data passenger incident events are presented in where the main category of incident was identified as ‘slip trip fall’. The authors categorized events analyzed into a category, for example, ‘pushchair falling over’ was a personal injury event, and all falls were categorized under ‘slip trip fall’.
Table 3. Original and condensed event types (n = 20490) (IRIS data 2012–2016).
Injury causation and understanding injury patterns
There were 494 incident reports available for analysis; six were removed due to being vandalism incidents. Overarching themes were used to drive the initial analysis of bus, driver and passenger and further sub-themes were identified that helped explain the interrelationships between them. This resulted in three themes that contribute to injury causation and challenge our understanding of injuries and expected behaviors on-board buses. Incident report extracts and CCTV insights were used to illustrate the findings. Injury severity in the CCTV analysis was not inferred to prevent assumption and speculation due to the complex nature of injury biomechanics and the need for medical evidence to support injury definitions.
Braking: Braking incidents predominated incident causation; however they involved several factors including the driver, road and passenger behavior. The language describing braking was rarely for an ‘emergency’ situation and more likely to be ‘harsh’, ‘sharp’, ‘late’, ‘slammed’ and it was noted that many ‘pulling away’ incidents also included a ‘harsh’ braking moment.
the bus was coming up to a round about and the driver slammed his brakes really hard
Driver did not anticipate that the vehicle in front to stop, and to avoid a collision our driver braked harshly causing TP child to fall forward hitting the hand rail.
Late braking by our driver, passenger falls from rear seat.
Driver decision making was notable in the braking incidents indicating there were lapses in concentration when reading the road or driving inappropriately for the traffic conditions.
Learner did stop but was so far ahead with brake lights visible that our driver was way too late in braking and may have thought learner would pull off by the time the bus had reached it but no excuse - totally avoidable
driver is not paying attention to road ahead. A TP vehicle stops and indicates to turn right. Our driver does not brake until late causing the majority of his customers to be thrown forward on the bus.
Insights from the CCTV analysis found that London bus drivers were consistently having to negotiate congested roads, and if they were not able to read the road ahead appeared to be the biggest factor for the drivers resulting in numerous harsh braking passenger incidents. Drivers were able to ‘brace’ for the braking, but passengers were unaware of when it would happen. Drivers also tend to be wearing seatbelts and contained in their seats. Rapid speed changes were noted during the bus slowing or braking incidents (), some 8.58 mph in 2 seconds, which probably accentuated the inertial forces. High deceleration rates were noted for the bus moving off category, though the act of the bus slowing down or braking caused a noticeable negative deceleration.
Table 4. Speed change of bus maneuvers in CCTV incidents (n = 51).
Passenger consequences from braking incidents caused many falls or stumbling episodes.
Suddenly has to brake harshly to avoid a collision causing the passenger to fall (badly).
Not all incidents caused an actual fall to the passenger but generated a trajectory path instead where passengers encounter the internal structures on the bus.
Driver is pulling away from stop and suddenly brakes sharply and the passenger is thrown forward and hurts her wrist
The contact sources for this kind of incident tended to be the poles, seats, handrails, and panels (described as glass, modesty partition) and depended on whether and where the passenger was seated or standing at the time. However, it was difficult to establish where all passengers were situated on the bus as this did not appear in the text for all incidents. Where it was stated there were some more common areas for example behind the wheelchair area, doors, aisle, rear of bus.
One female passenger who is standing by the wheelchair area is swung around and lands on the stairs.
When our driver looks back, he has to brake to avoid colliding into the rear but as a result an elderly passenger is thrown from her seat into the wheelchair area. He suffered a bloody nose and complained of neck pain.
Language of injury and incidents: The variation in language, between and within the bus operator reports was notable, such that identifying passenger injury and severity was difficult. Often it was reported that a passenger had ‘hit’ their head or ‘banged’ their knee or ‘caught’ their shoulder which indicates the body area that was injured. These descriptions do not provide enough evidence of actual injury type or severity and are only lay person assumptions.
The customer has been seriously injured in the process. This incident involved a bus driver on the XX route who closed the rear doors on his head as he was leaving the bus
Additional injury information to corroborate a serious head injury is not provided.
Overall, from the available data the commonly injured body regions were found to be the head/face, arms and legs with fewer reports of injury to back or chest.
Our driver is looking out of offside window and when he turns back, sees car has stopped and brakes harshly causing customer to fall from seat and hit head on pole.
Although falls were common, there were differences in fall descriptions and distances i.e., ‘a few steps’, ‘downstairs’ and ‘to the floor’ which the latter unfortunately does not make it clear if they were ‘standing’ or ‘seated’ at the time.
Furthermore, the language used to describe the passenger movements in non-fall incidents varied with ‘thrown, flung, jerked, jumped, fall forwards, shunted and surged’ all used.
6 people can be seen being thrown forward. Two of them were thrown to the floor
Expected and observed behaviours: This was an interesting observation whereby passenger behaviour was considered contributory to their injury causation in particular ‘not holding on’ and being out of position and therefore liable to be partially at fault for the injury sustained.
As the bus bears to the left at the lights at (xx) she loses balance and hits the window of the bus whilst still seated. ……. the speed was correct it was only that the customer had her legs crossed so had no stability in the seat
footage would suggest that the sole cause of the accident was the customer’s omission to take precautions to ensure her own safety on a moving bus, given that bus passengers are expected to hold on to available supports. Drivers are not required to monitor passengers who appear to be safely seated.
Standing choices varied as the CCTV showed passengers were either forced to stand because of lack of seats or they chose to stand during the journey. The standing area tended to be the priority area on the buses, either pre-selected due to having a pushchair or just preferred. Leaning on infrastructures without holding onto any handrail or pole for support was common and considered in incident reports to be contributory to the passenger falling.
Similarly, passengers with pushchairs were often blamed for any consequences of braking if they were not holding onto the pushchair when it toppled over or involved hanging shopping or nappy changing bags on the handles.
The woman stands between the buggy and the offside of the bus and is not holding onto the buggy… bus turns right …. and as a result the buggy tips over.
Liability for the buggy tipping rests with the mother for failing to park the buggy safely and carrying shopping on the handlebars which renders the buggy unstable.
Our CCTV analysis found that the stability of the parked pushchairs varied particularly when bags were loaded onto the handles. Stability improved when there was more than one buggy in the area compared to when only one was present, but this was when they were parked across the bus with handles facing the middle doors.
One incident showed a mobility scooter parked at a similar angle topple over during a braking incident. Following this, they parked the scooter facing forward; notably no markings were on the floor to assist passengers to make the best decisions of how to park.
Incident reporting by the passengers did not always occur at the time nor in some instances was the driver aware of anything happening.
I went up to the driver to get his details, (which was xxxx), and he was surprised that I had fallen down, as he wasn’t aware, but he never got out of his cab to check if everyone was OK.
In some cases, the drivers show no apparent concern following sudden braking or acceleration maneuvers.
No-one approaches driver to state they need medical attention. But additionally, driver doesn’t ask them if they require any.
Insights from the CCTV analysis found that following an incident it tended to be other passengers who showed most concern to the injured person, often waiting with them for the ambulance rather than the driver who in most cases was elsewhere (outside the bus). The long delay in waiting for ambulances and or police was notable and is presumed to disrupt the service and journey times.
Discussion
Braking incidents appear to be the largest factor responsible for passenger injuries and data suggest that preventing harsh braking (except in an emergency) would impact positively on injury prevention. Harsh braking resulted from drivers reacting to the road situations or responding late due to distractions. Advanced Drivers Assistance Systems (ADAS) notably headway monitoring systems have been effective in reducing onboard bus incidents due to harsh braking for one London Bus operator by 60% in a recently completed pilot study (Mobileye Citation2019). However, these systems are not routinely fitted to buses or even retrofitted possibly due to the overall costs to the bus operators. There is potential to undertake a cost-effective analysis of these systems against buses being out of service waiting for emergency vehicles and potential compensation claims for incurred injuries to make justifications for widespread implementation. The interior infrastructure of buses is hard and unyielding and provides the contact sources for injuries to occur post braking. The work by Edwards et al. (Citation2019) in collaboration with TfL ‘bus safety standard’ have highlighted the hostility of internal structures and recommended adaptations to handrails and seat heights to mitigate injury severity. However, if harsh braking is prevented in most of the incidents, the internal structure changes would not be crucial and could even act as a second line of defence in severe injury mitigation. This would also prevent many of the incidents that occur in the rear seats or seats facing the wheelchair area where passengers are thrown forwards or fall into the aisle or priority areas. Design solutions have been suggested for these seats including removing of handrails or enabling the repositioning of them to prevent passengers impacting them on falling Edwards et al. (Citation2019), but these are not practical solutions as handrails serve a safety purpose and regulation requires them to be present on buses.
Our analysis also identified that many passengers do not hold onto the handrails which appeared to be a driver expectation in the incident reports. This suggests that more awareness of expected passenger behavior is needed for users of the bus, particularly for the seats facing the wheelchair area and for those choosing to stand. The perception of these seats in the analysis is that they are potentially unsafe as passengers are thrown out of them onto the floor and some appeared to travel some distance before landing due to the lack of full handrail or panel. Similarly, rear seats were also problematic as passengers sat in the middle of them were thrown forwards down the aisle. However, the root cause of passengers being thrown or falling stems predominantly from harsh braking.
Unfortunately, one of the biggest problems in studying bus passenger injury causation is the underlying data collected and shared. STATS19 has limited variables which allow analysis of bus incidents, and none capture the behavior of driver or passenger onboard except at the higher level of ‘boarding’, ‘alighting’, ‘standing’ or ‘sitting’. More comprehensive data would enable evaluation of safety targets or new safety strategies to be measured. TfL have a large dataset (IRIS) that could be used to its full potential to inform passenger safety if a change and consistency of language was implemented. It would also benefit from adding new variables to determine basic injury data including body region and supplementing the incident reports with follow up injury description data post-emergency care or hospital assessment. This data would provide a benchmark measure for any bus safety standard improvements expected in the future.
The high number of bus operator incidents reported to TfL compared to the national STATS19 data suggests there is a bigger problem to address. It is evident that prevention of harsh braking incidents would have a positive impact on the number of incidents and passenger injuries onboard buses. The use of ADAS might help to mitigate this but a shared responsibility with passengers is needed to ensure that handrails are used to try and prevent passengers being thrown forward or falling from seats. Future benchmarking and evaluation of bus safety features requires comprehensive data which is not currently available which means that analysis of injury severity, causation, and mitigation is not currently possible but would be beneficial in the future.
Limitations: The analysis was conducted on a sample of incidents provided by bus operators, and these were not in-depth documents. The CCTV analysis was also a sample of mainly non-collision incidents and might not represent a comprehensive review of all bus incidents. It was the intent to explore injury causation and examine the injury type and severity of bus passengers. However, this could not be ascertained from the IRIS dataset or bus incident reports in detail and furthermore, STATS19 is limited to the generic ‘slight’, ‘serious’ and ‘fatal’ descriptions. Other studies have been able to ascertain injury severity and description detail through linkage of police and hospital data (Barnes et al. Citation2016). The shift to the Collision Reporting and Sharing system (CRASH) by the police should enhance the injury descriptions in the future and future studies should consider using opportunities to link bus operator data, STATS19/CRASH, CCTV incidents, Hospital Episode Statistics (HES) and coroners’ data to fully explore bus passenger injury causation at an in-depth injury type and severity level.
Acknowledgements
This study was commissioned and supported by London TravelWatch. The contents of this paper reflect the views of the authors who are grateful to the participating Bus Operators and Transport for London for supplying data.
Additional information
Funding
References
- Department for Transport. 2019. Annual bus statistics. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/852652/annual-bus-statistics-2019.pdf.
- Department for Transport. 2021. Bus back better strategy. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/980227/DfT-Bus-Back-Better-national-bus-strategy-for-England.pdf.
- Kendrick D, Drummond A, Logan P, Barnes J, Worthington E. 2015. Systematic review of the epidemiology of non-collision injuries occurring to older people during use of public buses in high income countries. J Transp Health. 2(3):394–405. doi:10.1016/j.jth.2015.06.002.
- Kirk A, Grant R, Bird R. (2003) Passenger casualties in non-collision incidents on buses and coaches in Great Britain. In: Proceedings of 18th international technical conference on the enhanced safety of vehicles. Nagoya, Japan 19–22 May 2003
- Barnes J, Morris A, Welsh R, Summerskill S, Marshall R, Kendrick D, Logan P, Drummond A, Conroy S, Fildes B, et al. 2016. Injuries to older users of buses in the UK. Public Transp. 8(1):25–38.
- Edwards M, Edwards A, Appleby J, Beaumont D. 2019. Banging heads onboard buses: assessment scheme to improve injury mitigation for bus passengers. Traffic Inj Prev. 20(sup1):S71–S77. doi:10.1080/15389588.2018.1563293.
- Barnes J, Brown L, Morris A, Stuttard N. 2019. Research into the safety of London bus passengers. London Travel Watch. https://hdl.handle.net/2134/9772502.v1.
- Braun, V, Clarke V. 2006. Using thematic analysis in psychology. Qual Res Psychol. 3(2):77–101. doi:10.1191/1478088706qp063oa.
- Boyatzis R. 1998. Transforming qualitative information: thematic analysis and code development. Thousand Oaks: Sage Publications.
- Mobileye. 2019. Success story: Abellio London reduces collisions and on-board injuries with Mobileye. [accessed 2022 06 06]. https://static.mobileye.com/website/uk/post/files/abellio_london_case_study.pdf.