Analysis of Firetruck Crashes and Associated Firefighter Injuries in the United States

Abstract

Motor vehicle crashes are the second leading cause of death for on-duty firefighters. Firetruck crashes, occurring at a rate of approximately 30,000 crashes per year, have potentially dire consequences for the vehicle occupants and for the community if the firetruck was traveling to provide emergency services. Data from the United States Fire Administration and the National Highway Traffic Safety Administration shows that firefighters neglect to buckle their seatbelts while traveling in a fire apparatus, thus putting themselves at a high risk for injuries if the truck crashes, especially in rollover crashes. Despite national regulations and departmental guidelines aiming to improve safety on fire apparatuses, belt use among firefighters remains dangerously low. The results from this study indicate that further steps need to be taken to improve belt use. One promising solution would be to redesign firetruck seatbelts to improve the ease of buckling and to accommodate wider variations in firefighter sizes.

INTRODUCTION

Each year, an average of 100 firefighters die and 100,000 firefighters are injured in the line of duty from a variety of causes including, but not limited to, extreme physical exertion, underlying medical conditions, and motor vehicle crashes (United States Fire Administration, 2011). The United States Fire Administration (USFA), an agency of the Department of Homeland Security, cites motor vehicle crashes as the cause of death for between 20–25% of the annual line-of-duty fatalities. Motor vehicle crashes are the second highest cause of death for firefighters. The leading cause of death is stress and overexertion which accounts for approximately 50% of the fatalities. Other significant causes of death in the dataset include: caught/trapped (10%), fall (5%), collapse (3%) and other (7%) (United States Fire Administration, n.d.). Firetruck crashes, although rare in comparison to non-emergency vehicle crashes, tend to have grave consequences for firetruck occupants and for occupants in other vehicles involved in the crash. Despite revising national standards to improve firetruck safety and reduce firefighters’ risk of injury and fatality, the annual injury and fatality rate has remained essentially unchanged over the past decade.

The USFA has openly prioritized reducing firefighter risk as its number one goal (United States Fire Administration, 2010), intending to accomplish it through injury prevention and mitigation strategies to reduce the total number of line-of-duty injuries and fatalities.

This paper investigates the characteristics of fatal firetruck crashes and identifies some underlying issues that may lead to increased firefighter injury and fatality risk while riding in a fire emergency vehicle. The data presented comes from two different national databases with varying degrees of crash-level and occupant-level information.

METHODOLOGY

Defining Relevant Terms

Certain terms will reoccur throughout this paper. To ensure a uniform understanding, the key terms are defined in this section.

Firetruck

As defined in the FARS Data Dictionary, a firetruck is a vehicle that is specially designed and equipped to respond to fire, hazmat, medical and extrication emergencies. These vehicles typically contain special lights and markings that allow them to be easily identified. Firetrucks, in this paper, are synonymous with the term fire apparatus and include “medium and heavy vehicles such as engines, pumpers, ladder, platform aerial apparatus, heavy rescue vehicles, water tenders or tankers, brush or wilderness firefighting vehicles, etc.” (National Highway Traffic Safety Administration, 2009).

Line-of-duty Death and Injury

The USFA defines a line-of-duty death as any illness or injury that a firefighter sustains while on-duty or within 24-hours of being on-duty that leads to death (United States Fire Administration, 2012). The USFA considers all fatalities that resulted from a motor vehicle collision regardless of whether it occurred in a fire department-owned firetruck (as defined above) or in a firefighter’s personal vehicle if (s)he was traveling to or from an emergency incident.

Datasets

This paper examines the injury and fatality risk of firefighters traveling in fire department vehicles by evaluating data from two main datasets, the National Highway Traffic Safety Administration’s (NHTSA) Fatality Analysis Recording System (FARS) database and General Estimates System (GES) database. Each dataset was evaluated for errors and adjusted if needed.

FARS Database

NHTSA’s FARS database tracks all fatal crashes in the United States. For this analysis, data from 371,169 crashes between 2000 and 2009 was queried to collect crash characteristics on crashes involving fire vehicles. The database contained 195 cases where firetrucks were involved in fatal cases. In 49 of those cases, at least one fatality occurred in the firetruck.

GES database

The GES database, also established by NHTSA, contains a nationally representative sample of approximately 60,000 police reported crashes collected each year. The data from those crashes is supplemented by statistical weighting factors that are used to estimate the national frequency of motor vehicle crashes that occur in the United States. In rare cases, such as firetruck crashes, the weighting factors have the potential to be less reliable than some of the more common crashes recorded in GES. Since the GES cases are coded from police accident reports, which are known to frequently contain inconsistencies, the data required additional check to ensure the validity of the GES results. For example, one data check involved verifying the body type of the vehicle that was listed as a firetruck. Performing that query resulted in 26 cases of small passenger vehicles that were listed as firetrucks. Since firetrucks should be categorized as a medium or heavy vehicle, the cases involving fire-department passenger vehicles had to be removed from the dataset. This resulted in the removal of a weighted value of 2994 crashes throughout the data collected from 2000–2009. After distilling the data, the weighted cases (n=31,606) presented in this paper are based on a sample size of 497 police reported firetruck crashes.

In FARS and GES, firetruck crashes were identified by the special use variable entitled. By singling out these crashes, it reduces the dataset to a much smaller size and populates it with relevant cases. Other variables of interest included: injury severity, ejection, restraint use, seating position, emergency use, number of vehicles involved in the crash, most harmful event, and rollover type.

In the remainder of this paper, the datasets are combined to present a simplified view of the results. Each table contains GES data that presents the estimated frequency of crashes involving firetrucks and FARS data that presents the cases where a firefighter was fatally injured in the fire apparatus.

RESULTS

FARS and GES were used to assemble a 10-year dataset of information on firetruck crashes. The data was used to examine firefighter deaths in crashes despite advances in vehicle safety across all vehicle types.

As shown in Figure 1, firetruck crashes account for almost 10% of all crashes involving special purpose vehicles, for example, taxis, school buses, city buses, military vehicles, police vehicles, ambulances, and emergency service vehicles.

Figure 1:

Distribution of crashes involving special purpose vehicles.

Data in Table 1 shows that, between the years of 2000 to 2009, there were approximately 31,600 crashes involving fire vehicles, 49 of which lead to the death of at least one firetruck occupant.

Table 1:

Compilation of GES and FARS Crash Data for Crashes Involving Firetrucks

Firetruck Crash Data (2000–2009)	All Crashes (GES)		Fatal Crashes (FARS)
	Unweighted	Weighted	Frequency
Number of Crashes	497	31,606	49
Number of Occupants Involved	997	52,241	476
Number of Fatalities	4	132	50

Note: FARS data is only presented when the fatality occurred within the fire vehicle

Due to the sheer size and mass of fire vehicles, firetruck occupants typically face a very low risk of injury during typical multi-vehicle traffic crashes. Injury severity is coded in FARS and GES with the police-reported KABCO scale. This scale categorizes injury severity ratings into five groups: killed (K), incapacitating injury (A), non-incapacitating injury (B) minor injury (C), and no injury (O). Controversy exists as to whether or not the KABCO scale provides accurate injury severity data since it is limited by the reporting officer’s ability to assess the extent of an injury prior to official medical evaluations. The KABCO scale was used in this analysis since it is commonly coded on police crash reports, but previous studies have shown that KABCO does not correlate well with other more reliable medically-based scales such as the Abbreviated Injury Scale (Farmer, 2010).

As seen in GES data (Table 2), ninety percent of the firetruck occupants involved in all firetruck crashesescape with no injuries. Seventy-five percent (n=146) of fatal crashes involving firetrucks happen in such a manner where the fatality occurs in the other vehicle. However, in firetruck crashes involving the death of a firetruck occupant, the fatality rate per occupant increases from 0.25% to 10.5%.

Table 2:

Distribution Injury Severity Scores for Firetruck Occupants in Motor Vehicle Crashes

Injury Severity (2000–2009)	All Crashes (GES)			Fatal Crashes (FARS)
	Unweighted Cases	Weighted Frequency	Percent	Frequency	Percent
No Injury (O)	786	46,766	89.5	223	46.9
Possible Injury (C)	103	2,548	4.9	77	16.2
Non-incapacitating Evident Injury (B)	53	1,048	2.0	89	18.7
Incapacitating Injury (A)	41	943	1.8	36	7.6
Fatal Injury (K)	4	132	0.3	50	10.5
Injury, severity unknown	3	88	0.2	0	0
Unknown	7	716	1.4	1	0.2
Total	997	52,241	100	476	100

The most harmful event variable was used to investigate what type of impact was causing the most fatalities and injuries in firetruck crashes. The most harmful event is the presumed crash-event which inflicted the most harmful injury or which caused the most vehicle damage. The data in Table 3 indicates that the most harmful event for approximately 75% of all firetruck crashes was a collision with another motor vehicle in transport; however, rollover crashes are the highest contributor to fatal crashes, accounting for almost half of the most harmful fatal events. This indicates that firefighters are exposed to a greater risk of injury or fatality if they are in a rollover crash compared to a non-rollover crash.

Table 3:

Distribution of Firetruck Crashes with Respect to Tripped Rollover or Untripped Rollover Crashes

Type of Rollover (2000–2009)	All Crashes (GES)			Fatal Crashes (FARS)
	Unweighted Cases	Frequency	Percent	Frequency	Percent
No Rollover	479	31,158	98.6	19	38.8
Tripped Rollover	8	285	0.9	16	32.7
Untripped Rollover	0	0	0	13	26.5
Rollover, Unknown Type	10	163	0.5	1	2.0
Total	497	31,606	100	49	100

Rollovers, although found in the most harmful event categories, have their own variable that can be used to glean more information about a case’s crash characteristics. Further analysis of rollover crashes showed that nearly two-thirds of fatal firetruck crashes involved either tripped or untripped rollovers (Table 4). Tripped rollovers include being caused by curbs, guardrails, ditches, soft soil, or other objects or vehicles whereas untripped rollovers are usually caused by sudden steering maneuvers. There was an almost even distribution of tripped and untripped rollovers. Furthermore, when compared with the most harmful crash events, 80% of the crashes that involved rollovers listed the rollover as the most harmful event of the crash.

Table 4:

Distribution of the Most Harmful Event in Firetruck Crashes

Most Harmful Event (2000–2009)	All Crashes (GES)			Fatal Crashes (FARS)
	Unweighted Cases	Frequency	Percent	Frequency	Percent
Rollover/Overturn	14	372	1.2	24	49.0
Collision with a Motor Vehicle in Transport	425	23,254	73.6	10	20.4
Collision with a Parked Motor Vehicle	16	3,512	11.1	0	0
Collision with a Sign Post, Utility Pole or Other Post	13	2,074	6.6	0	0
Collision with a Tree	4	875	2.8	7	14.3
Collision with a Pedestrian	5	119	0.4	0	0
Collision with an Animal	4	521	1.7	0	0
Collision with a Non-fixed Object	3	308	1.0	0	0
Non-Collision Injury (Injured In or Fell From Vehicle)	1	11	0.04	3	6.1
Pavement Surface Irregularity	1	52	0.2	1	2.0
Other	11	507	1.6	4	8.2
Total	497	31,606	100	49	100

The restraint use of each firetruck occupant is recorded in the FARS and GES databases. In each database, the main restraint categories include: lap and shoulder belt, lap belt only, shoulder belt only, and not applicable. In the FARS and GES datasets, the term not applicable applies to cases where seatbelts are not available, but also in cases where seatbelts are available but not worn.

Table 5 presents data on the number of firetruck occupants that were reported as wearing a safety restraint when they were involved in a crash. In FARS and GES, the element named “not applicable” applies to cases where a restraint was either not available or not used. For example, a firefighter in the cab of a firetruck would be coded as not applicable as would a firefighter who was seated but unbuckled. As shown in the data, fire truck occupants have very low restraint usage rates.

Table 5:

Distribution of Occupants’ Ejection Status

Occupant Ejection (2000–2009)	All Crashes (GES)			Fatal Crashes (FARS)
	Unweighted Cases	Frequency	Percent	Frequency	Percent
Not Ejected	981	51,164	97.9	27	54
Totally Ejected	2	15	0.03	17	34
Partially Ejected	0	0	0	5	10
Not on PAR	2	7	0.01	0	0
Not Applicable	12	1,055	2.0	0	0
Unknown	0	0	0	1	2
Total	997	52,241	100	50	100

Another interesting observation from this query is that there were four unweighted cases in GES that indicated children were restrained in child safety seats within the firetruck. Even after verifying the vehicle-type as a heavy vehicle, these cases remained in the dataset. Although this is counterintuitive, one example of this could happen is illustrated by a 2010 crash involving children in a firetruck. A news article on this crash stated that six children who were involved in a firetruck crash (“Fire truck crashes with kids onboard,” 2010). The children were taking part in a fire station demonstration when the emergency brake of the firetruck they were exploring unexpectedly released allowing the truck to roll down a hill and crash into a tree. The children were unharmed, but this case confirms that this observation was not a result of a coding error.

In a severe crash, some circumstances may cause an occupant to be ejected from the vehicle during a crash. The occupant may be only partially ejected for part of the crash, or may be entirely removed from the vehicle.

Data from GES indicates that being ejected from a firetruck is a rare event; however, FARS data suggests that when a firefighter is fully or partially ejected from the vehicle, it puts the firefighter at a much greater injury and fatality risk (Table 6). An occupant is coded as totally ejected if his body is entirely out of the vehicle’s occupant compartment, although he may be coded as totally ejected if he is still physically touching the vehicle after the crash.

Table 6:

Distribution of Firefighter Restraint Use in Police Reported Crashes

Restraint Use (2000–2009)	All Crashes (GES)			Fatal Crashes (FARS)
	Unweighted Cases	Frequency	Percent	Frequency	Percent
Not Used or Not Applicable	121	3,969	7.6	37	74
Shoulder Belt Only	580	31,936	61.1	2	4
Lap Belt Only	82	4,716	9.0	6	12
Shoulder and Lap Belt	7	332	0.6	0	0
Child Safety Seat	4	109	0.2	0	0
None Available	1	73	0.1	0	0
Restraint Used - Type Unknown	76	4,636	8.9	0	0
Other/unknown	126	6,469	12.4	5	10
Total	997	52,241	100	50	100

Total ejection also applies to occupants who are not initially in the occupant compartment, for example hanging onto the side of a truck. A partially ejected occupant is someone who had part of their body outside of the vehicle at any point in time during or immediately after the crash. Ejections are not exclusively associated with rollover crashes. An occupant can be ejected from the vehicle even if a rollover does not occur (Table 7). Although the data lists “partially ejected” as an available coding option, the GES data does not contain any cases where an occupant was partially ejected.

Table 7:

Frequency of Ejected Occupants by Rollover Type in Fatal Firetruck Crashes

Ejection	ROLLOVER TYPE
	No Rollover	Tripped Rollover	Untripped Rollover	Rollover, Unknown Type	Total
Not Ejected	13	6	8	0	27
Totally Ejected	5	7	4	1	17
Partially Ejected	0	4	1	0	5
Unknown if Ejected	1	0	0	0	1
Total	19	17	13	1	50

Seatbelts are an effective countermeasure against ejection. Table 8 shows that over half of the firetruck occupants that were unrestrained in the fire vehicle were ejected or partially ejected from the vehicle. Only one restrained occupant was partially ejected from the fire apparatus and none were fully ejected. As expected, occupants are at a higher risk of being ejected from a vehicle if they are not restrained.

Table 8:

Frequency of Firetruck Fatalities by Seating Position and Restraint System Use

Seating Position	RESTRAINT SYSTEM USE
	Not Used or Not Applicable	Lap Belt Only	Shoulder and Lap Belt	Unknown	Total
Front Seat, Left Side	23	1	3	2	29
Front Seat, Right Side	7	0	2	2	11
Second Seat, Left Side	3	1	1	0	5
Second Seat, Right Side	1	0	0	0	1
Other Passenger in enclosed	1	0	0	0	1
Riding on Exterior of Vehicle	2	0	0	0	2
Unknown	0	0	0	1	1
Total	37	2	6	5	50

Table 9 compares seating position with restraint use in fatal firetruck crashes. Most firetrucks have between two to three rows of seating in the cab, but other seating positions that may be recorded include simply being enclosed in a passenger area but not seated in a proper seat, and riding on the outside of a vehicle. The data indicates that sixty percent of all firetruck fatalities occur in the driver’s seat of the apparatus. Out of 29 drivers-seat fatalities, only three of the drivers were restrained by a lap and shoulder belt.

Table 9:

Frequency of Ejected Occupants versus Occupant Restraint Use in Fatal Firetruck Crashes

Ejection (2000–2009)	RESTRAINT SYSTEM USE
	Not Used or Not Applicable	Lap Belt Only	Shoulder and Lap Belt	Unknown	Total
Not Ejected	16	2	5	4	27
Totally Ejected	16	0	0	1	17
Partially Ejected	4	0	1	0	5
Unknown if Ejected	1	0	0	0	1
Total	37	2	6	5	50

Sometimes it is helpful to determine if the vehicle was being used during emergency operations. Figure 2 indicates that most firetruck crashes occur when the truck is being used during an emergency (66%).

Figure 2:

Crash Occurrence During Emergency Response

DISCUSSION

Previous studies have found that firetruck crashes are most common near intersections, in the PM peak periods, and involved high-risk behaviors like speeding and passing maneuvers (Savolainen, Dey, Ghosh, Karra, & Lamb, 2009), but little research has focused on the injury outcomes of these crashes.

The findings presented in this paper are limited to the data available in FARS and GES. Due to the small sample of crashes in the GES database, actual injury data may be over- or underestimated when using weighting factors. The fatality data from FARS only includes cases where a firetruck occupant sustained fatal injuries as a result of the crash.

The research presented in this paper has shown that restraint use is an important factor to consider when aiming to reduce the number of firefighter fatalities in motor vehicle crashes. Other research projects confirm that increasing belt use will effectively reduce the number of ejected firefighters given that a firetruck crash occurs (National Transportation Safety Board, 1991). The question remains: why are firefighters neglecting their seatbelts?

Firefighters provide a number of explanations as to why they are not wearing seatbelts. In one departmental survey, 42% of the firefighters thought that the size and weight of the vehicle would be enough to prevent against serious injury in the event of a crash (DeMar, 2007), but the most common reasons for firefighters not wearing their seatbelts was that the belts were uncomfortable and did not fit well. Some claim that they are unable to reach the seatbelt when they are in their full turnout gear (Coleman, 2007; DeMar, 2007; Routley, 2009). Others cite incompatibilities with the self-contained breathing apparatus (SCBA) or other personal protective gear (Routley, 2009).

There are currently no nationwide regulations regarding whether firetruck drivers may drive with their gear on. Given that approximately two-thirds of firetruck crashes happen while in route to an emergency situation, as shown in Figure 2, some firefighters may be driving their vehicles while wearing their heavy gear. Some fire departments have local regulations that ban driving in heavy turnout gear, but others leave the decision up to the driver’s personal preference. There is currently no way to identify in FARS or GES what type of uniform or equipment the fatally injured drivers were wearing (Coleman, 2005).

Current Federal Motor Carrier Standards exempt emergency workers from the requirement to wear seatbelts when responding to an emergency. Several states override that regulation and requires seatbelts to be worn at all times while a fire apparatus is in motion, but seventeen states have not augmented the regulation (Coleman, 2007; National Highway Traffic Safety Administration, 2007). Enforcement of these policies fall to a myriad of different people, from the vehicle operator, to the company officers, to relying on the diligence of each individual firefighter to buckle the belt (Coleman, 2007; DeMar, 2007).

Local fire departments and national fire associations are following several different approaches to increase belt use. Some approaches primarily focus on changing firefighters’ mindsets through emotional stories of personal seatbelt-related tragedies (DeMar, 2007; Everyone Goes Home, n.d.) while others focus on the need to revise national policies (National Fire Protection Association, 2009).

The National Fire Protection Association, a nonprofit trade organization, established the NFPA Standard 1901 to improve safety on all types fire vehicles. The newest edition, revised in 2009, endorses new lengths for seatbelts, in addition to requiring seatbelt warning devices to be installed in fire vehicles (National Fire Protection Association, 2009). The seatbelt warning devices will warn the driver when a firetruck occupant is unbelted. Further changes recommend vehicle data recorders to capture data including, but not limited to, vehicle speed, acceleration/deceleration, engine speed, engine throttle position, ABS event, seat occupied and seat belt status, master warning device switch position, and date/time.

Research Needs

In addition to encouraging higher belt use and better seatbelt designs, additional research needs to be performed on crash avoidance technologies for emergency vehicle applications. Example technologies include: vehicle-to-infrastructure communication, vehicle-to-vehicle communication, active crash avoidance technologies, and other technologies that may aid in reducing the number of firetruck crashes.

CONCLUSION

Approximately 500 firefighters are involved in fatal firetruck crashes each year and 1 out of 100 of these occupants dies as a result of the crash. Despite changes in regulations that govern fire vehicle safety, the average fatality rate per year has remained relatively stagnant. Rollovers are the most common crashes that result in firefighter deaths (66% of all fatal firetruck crashes), and a majority of those fatalities were unrestrained occupants. Redesigning and improving firetruck restraint systems could reduce the number of injuries and fatalities that occur in firetruck crashes, but the restraint systems will only be effective if firefighters buckle them in while riding in the apparatus.