Contributing Factors and Issues Associated With Rural Ambulance Crashes: Literature Review and Annotated Bibliography
Posted on: Tuesday, 19 August 2008, 03:00 CDT

By Sanddal, Nels D Albert, Steve; Hansen, Joseph D; Kupas, Douglas F

ABSTRACT Ambulance crashes occur with greater frequency and severity than crashes involving vehicles of similar size and weight characteristics. Crashes in rural areas tend to be more severe in terms of injury or death to vehicle occupants. The purpose of this article was to examine the extant literature, as well as summarize and discuss the overlapping findings of that body of literature. A stepwise literature search was conducted using the following MeSH search terms ambulance; accident, traffic; emergency medical technician; occupational health; and rural in descending combination. MEDLINE was used as the primary database but was augmented by searches of Academic Search Premier, Comprehensive Index of Nursing, Allied Health Literature, and ProQuest Dissertation International. The search resulted in 32 article citations, and of these, 28 were included. An annotated bibliography is followed by a discussion and conclusion that identify opportunities for prevention activities in the areas of education, enforcement, and engineering. Key words: ambulances; accident, traffic; emergency medical services; occupational health; emergencies; emergency medical technician.

PREHOSPITAL EMERGENCY CARE 2008;12:257-267

LITERATURE SEARCH AND SELECTION

The literature search was conducted in a stepwise process. The purpose of the search was to identify published literature that describes the frequency, epidemiology, etiology, typology, and cost (i.e., human and fiscal) of ambulance crashes generally and rural ambulance crashes specifically.

The primary database selected for the literature search was MEDLINE (1996-2007). A secondary search was conducted using Academic Search Premier, Comprehensive Index of Nursing, and Allied Health Literature. A final search was conducted using ProQuest Dissertation International. MeSH search terms used in MEDLINE included ambulance; accident, traffic; emergency medical technician; occupational health; and rural in descending combination. The primary, secondary, and tertiary searches yielded 28 articles on the subject. Of those, four (4) published in trade magazines and one (1) appearing in a foreign medical journal were not retrievable and, therefore, excluded. The authors' personal familiarity of the literature resulted in the addition of four (4) articles not identified by formal search criteria. Staff reviewed the remaining articles for relevance and, ultimately, included 28 in this review. One article encompasses all deaths to EMS personnel, including ambulance crashes.

INTRODUCTION

Unfortunately, ambulance crashes are relatively common. The absence of vehicle performance standards, improper maintenance, variable operator training, and a lack of proper safety restraint use contribute to the human toll caused by at least 6500 ambulance crashes a year.1 As a result of these and other factors, the occupational motor vehicle fatality rate for emergency medical personnel is four times the U.S. average for other occupations.2

Rural ambulance crashes are of great concern, in part because they are usually much more severe than urban crashes.3 Ambulance crashes occurring on rural roadways are more likely to result in death to emergency medical personnel, the patient, and occupants of other vehicles. When a rural crash does not involve injury fatalities, there are often significant delays associated with the continued transportation of the initial patient. Often, multiple ambulances must be dispatched from surrounding communities, resulting in transportation delays for the original patient as well as the "new" patients resulting from the ambulance crash itself.

Unlike helicopter and fixed-wing emergency medical services (EMS) incidents, little is known about ambulance crashes in general and rural ambulance crashes specifically. This paper will review the current literature, discuss the implications for rural EMS agencies and personnel, and provide a sample policy or protocol that could be adapted for use in most communities.

ANNOTATED BIBILIOGRAPHY

To assist the reader, the following papers have been categorized into four sub-headings: 1) Description of the Problem; 2) Safety Issues; 3) Lights and Siren Use; and 4) Legal and Ethical Risks. Articles are organized alphabetically within each category.

Description of the Problem

[Auerback PS, Morris JA, Phillips JB, Redlinger SR, Vaughn WK. An analysis of ambulance accidents in Tennessee. J Am Med Assoc. 1987;258(11):1487-1490.]

The data source for this study was derived from interviews pertaining to 102 consecutive ambulance crashes in Tennessee for the period of January 1983 through June 1986. Only crashes resulting in injury or property damage in excess of $200 were included. Of the 102 inclusions, 29 crashes involved injury to the emergency medical technician (EMT) or patient. The authors note that the failure to use restraints was associated with injury. The self-reported estimate of temporal delay to hospital care following a crash was 9.4 minutes. The authors conclude that safety restraint use should be required and that all traffic signals be obeyed at intersections.

[Centers for Disease Control and Prevention. Ambulance crash- related injuries among emergency medical services workers-United States, 1991-2002. Morbid Mortal Weekly Rev. 2003;52(8):154-156.]

The authors analyze the Fatality Analysis Reporting System for an 11-year period and describe the attributes of 300 fatal crashes involving ambulances that occurred during that time period. The 300 crashes resulted in 82 deaths among the 816 ambulance occupants (patients and emergency personnel). There were an additional 275 deaths to occupants of other vehicles and/or pedestrians.

Although acknowledged to be an imprecise estimate, the authors conjecture that 27 of the deaths were emergency personnel. The authors also cite Maguire et al.'s estimated fatality rate of 12.7 per 100,000 EMS personnel; more than double the national average of on-the-job motor vehicle related mortality.7 The significant findings of the article are the risk to unrestrained occupants. One- third of the fatalities occurred in the front seats of the ambulance where seatbelts were available and could have been used, but weren't, and 22% of the workers killed were working unrestrained in the patient compartment. The report also illustrates three case examples gleaned from the National Institute for Occupational Safety and Health database. In each of the three case reports, the emergency care worker who died was unrestrained at the time of the fatal event.

[Custalow C, Gravitz C. Emergency medical vehicle collisions and potential for preventive intervention. Prehosp Emerg Care 2004;8(2):175-184.]

The authors draw on a database from the Paramedic Division of the Denver Health and Hospital Authority for a nine-year period. During that time period, there were 192 moving collisions involving ambulances. Thirty-nine of these resulted in injuries or death to 81 individuals. These injuries were sustained by 18 emergency vehicle operators, 19 emergency medical providers who were not drivers, 27 civilian drivers (including 2 deaths), 11 civilian passengers, and 2 patients being transported by the ambulance.

While this article does not specifically discuss rural ambulance crashes, it does provide insight into the vectors involved in a crash, identifying the emergency vehicle driver, civilian driver, and environment as each contributing, to some varying degree, to ambulance crashes. The authors note that a disproportionate share (91%) of the crashes occurred while the vehicle was operating with lights and sirens. The study also noted that in 71% of the collisions, the emergency vehicle operator had a record of multiple collisions.

The authors also note that crash-related vehicular claims constitute the greatest liability risk for an EMS agency. They report that in many states, Good Samaritan laws do not protect drivers of emergency vehicles involved in crashes.

[Erich J. Wheels of fortune. Every time you hit the streets, you take your life in your hands-how can you improve your chances? Emerg Med Serv. 2000;29(ll):43-46, 49-50, 52 passim.]

Several case reviews are included in this essay that outline various factors (e.g., emergency vehicle operator error, faulty maintenance, the urgency of a potentially life-saving response, and poor driving habits in the civilian population) involving emergency vehicle crashes. After discussing each of these in some depth, the author concludes that "human error" is the most common and most difficult factor to modify.

[Hunjadi D. From provider to patient. Emerg Med Serv. 2005;34(8):157-160.]

This personal account of the results of an ambulance crash that occurred in rural Wisconsin provides the reader with details of the physical, psychological, social, and economic toll that an ambulance crash can have on those involved. The EMT involved was providing care in the rear compartment of an ambulance that skidded into the median on a rain-soaked highway and rolled. The EMT was in critical condition immediately following the crash and currently is paralyzed below the waist. The economic burden of ongoing medical care on his family has been only partially covered by workers' compensation due to the volunteer nature of the EMS agency and to the fact that a 34- year-old family man does not wish to go to a nursing home to live out his remaining years. [Kahn C, Pirrallo R, Kuhn E. Characteristics of fatal ambulance crashes in the United States: An 11-year retrospective analysis. Prehosp Emerg Care 2001;5(3):261- 269.]

The authors provide a descriptive analysis of fatal ambulance crashes over an 11-year period from 19871997, using data derived from the U.S. DOT's Fatality Analysis Reporting System (PARS). The study's hypothesis was "... that there is no association between emergency use vs. non-emergency use and other fatal ambulance crash characteristics..." (p. 262). There was an inability to reject the null hypothesis for most crash characteristics with only the relationship to an intersection and the manner of the collision showing differences between emergency and nonemergency use.

However, the true value of this analysis lies in the descriptive tables that describe seasonal, temporal, atmospheric, and roadway characteristics in fatal crashes involving ambulances. The authors also note that most ambulance crash fatalities occur among those traveling in the rear compartment, where the patient may not be securely fixed to the chassis and where other occupants are less likely to be wearing seatbelts. The EARS data also revealed that many emergency vehicle operators had poor driving histories.

[Maguire BJ, Hunting KL, Smith GS, Levick NR. Occupational fatalities in emergency medical services: a hidden crisis. Ann of Emerg Med. 2002;40(6):625-632.]

This article is based on data from multiple sources and extrapolates the occupational fatality frequency, rate, and typology of on-duty fatal events involving EMS providers. The primary findings include a fatality rate of 12.7/100,000 more than double the average occupational fatality rate of 5.0/100,000 and approaching the fatality rates for law enforcement and firefighters. When stratified by cause, transportation injury fatality rates for EMS workers was 9.6/100,000 for EMS personnel, exceeding transportation fatality rates for law enforcement (6.1) and firefighters (5.7). The EMS rate is more than four times the average transportation fatality rate for all U.S. workers at 2.0/100,000. This article clearly supports the need for safer driving practices across both rural and urban environments.

[Pratt SG. NIOSH Hazard Review. Work-related roadway crashes: Challenges and opportunities for prevention. Washington, DC: Department of Health and Human Services, Centers for Disease Control and Prevention, National Instituteur Occupational Safety and Health, 2003; 1-92.]

This comprehensive analysis of multiple data sources describes work-related motor vehicle crashes from a variety of perspectives. While it does not specifically address ambulance crashes, it is important to note that trucks (all types) account for 64.9% of all fatal workrelated crashes in rural areas. Ambulances, by the nature of their design, would be captured in this category, again confirming the high-risk nature of rural emergency vehicle operations. Of note, this report discusses in great detail the increased risk associated with both driver fatigue and driver distraction. The report offers employers a set of recommendations to reduce workrelated crashes, including fatigue management, vehicle operations training, and graduated implementation of driving responsibilities for young drivers.

[Ray AM, Kupas DF. Comparison of crashes involving ambulances with those of similar-sized vehicles. Prehosp Emerg Care 2005;9(4):412-415.]

In this analysis of data from the Pennsylvania Department of Transportation from 1997-2001, it was noted that road-surface conditions and weather factors were similar between the ambulance and other truck-type configurations. However, differences were noted in crashes at intersections and traffic signals, with ambulances being more likely to be involved in such events. More people were involved in each ambulance crash, with three or more persons in 84% of the events. There was also a greater preponderance of ambulance crashes occurring during evening and weekend hours. The authors conclude that additional driver training and policies concerning the use of lights and sirens, including enforcement of the "complete stop" rule at intersections and traffic signals, could result in a reduction of ambulance crashes.

[Ray AM, Kupas DF. Comparison of rural and urban ambulance crashes in Pennsylvania. Abstract. Prehosp Emerg Care. 2007;11(4):416-420.]

This study summarizes findings from the Pennsylvania Crash Outcome Data Evaluation System database, which is a probabilistically linked dataset involving a number of separate data systems, including standardized law enforcement accident reports, for the time period of 1997-2001. The analysis identified 1745 ambulance crashes, of which 311 occurred in rural areas. The authors noted that rural crashes were more likely to involve snowy roadway conditions and unlit nighttime roadways. Operator error was the most prevalent contributing factor in both urban and rural crashes, although it was less often the cause in rural environments (75% rural vs. 93% urban). Rural crashes were more likely to involve striking a fixed object. Criteria for distinction between rural and urban was based on the PA Department of Transportation Roadway Management System and in which rural is defined based on traffic volume and municipal population criteria.

[Weiss SJ, ElHs R, Ernst AA, Land RF, Garza A. A comparison of rural and urban ambulance crashes. The American Journal of Emergency Medicine. 2001;19(l):52-56.]

A database comprised of information from mandatory reporting forms completed for all ambulance crashes occurring in the State of Tennessee was analyzed for a five-year period from 1993 to 1997. This dataset includes both fatal and nonfatal crashes. The primary hypothesis of this study was "Rural vehicle accidents will be more severe and have a higher rate of citations than urban accidents... " Rural was defined as a population equal to or less than the fifth- largest county in Tennessee (Montgomery, population 102,000). The authors analyzed characteristics including injury severity, traffic citations, ambulance damage, other vehicle damage, ambulance impact site, temporal, meteorological, and roadway conditions. They also examined the number of people involved, the number of people injured and use of safety belts at the time of the crash. The primary finding was that rural ambulance crashes were more likely to result in injury, and that the injuries sustained during the crash were more likely to be severe. This finding was predominately attributed to the point of collision, which was more likely to involve a frontal impact in rural areas and a rear impact in urban.

Safety Issues

[Barishansty RM. Next generation ambulance puts safety first. EmergMed Serv. 2005;30,34.]

This descriptive article discusses the features of a second- generation ambulance design that "puts safety first." Among updated design functions are external cameras for better driver visibility, improved seat placement and restraints systems for rear crew members, safety cargo netting to reduce the possibility of striking the bulkheads during a crash, more secure equipment storage, turn and brake signal indicators in the rear compartment to provide a visual warning of impending turns or stops, and changes in exterior paint and lighting for higher visibility. Additionally, the new vehicles come with "black box" monitoring and recording systems as standard equipment. The author concludes that these modifications may have an impact in both the avoidance of crashes and in the reduction of injury in the event of a crash.

[Cook RT, Meador SA, Buckingham BD, GroffLV. Opportunity for seatbelt usage by ALS providers. Prehosp Disast Med. 1991#(4);469- 4:71.]

The authors describe self-reported seatbelt use in the patient compartment while attending to patients with varying conditions. The use of seatbelts was inversely related to the perceived criticality of the patient and varied by condition. The overall "perceived need to be unrestrained to properly care for the patient" mean was 41%. It was higher for cardiac arrest (82%) and chest pain (63%) and noted to be lower for shortness of breath (38%) and trauma (41%). Providers stated that appropriate provider and equipment positioning and improved restraint systems could increase the percentage of time that they could wear restraints and still appropriately care for the patient. The study is limited by the self-report nature of the methods and by a limited number of provider responses.

[De Graeve K, Deroo K, Calle P, Vanhaute O, Buylaert W. How to modify the risk-taking behaviour of emergency medical services drivers? Eur J EmergMed. 2003;10(2):111-116.]

This article represents the first of several that have reported the impact of "black boxes" on emergency vehicle driving behavior. The black box is an electronic device that monitors, in real time, several vehicle parameters, such as speed, acceleration, braking, and cornering. It is designed to provide auditory feedback to the driver when predefined limits are exceeded. The authors of this seminal work reported only moderate change resulting from the black box with ongoing feedback and performance monitoring. Interestingly, the authors also note that the change from a Volvo sports wagon to a more traditional ambulance vehicle resulted in less aggressive driving behavior.

[Larmon B, LeGassick T, Schriger D. Differential front and back seat safety belt use by prehospital care providers. Am J Emerg Med. 1993;11(6):595-599.]

This article was among the first to look at the behavior of safety belt use among emergency medical personnel in ambulances. The self-reported data indicated a high use (approaching 100%) of seat belt use when emergency medical personnel are in the front of the ambulance. This was at a time when the civilian seat belt use rate nationally was reported to be 49%. However, when the emergency personnel were in the rear compartment of the ambulance providing patient care, the use rate fell substantially and approached zero if the patient was deemed to be in a "critical" condition. The authors conclude that the findings point to a need for additional training, the investigation of which clinical conditions might warrant the provider being unrestrained, and the need for ambulance redesign to accommodate the needs of the emergency provider in the care of the injured/ill patient. [Levick NR. An optimal solution for enhancing ambulance safety: Implementing a driver performance feedback and monitoring device in ground emergency medical services vehicles. Ann Proc Assoc Mv Automat Med. 2005;49:35-50.]

This pre/post (repeated measures) comparison examined the deployment of a "black box" in an urban ambulance fleet. The black box (onboard computermonitoring device) was placed in the fleet without driver identification and without turning on the auditory alert signal for a period of three months. A number of vehicle operation parameters were measured during this "blind" data- gathering period. The second phase of the research began with an orientation of all personnel to the system, the issuance of key fobs for driver identification, and in the activation of the auditory alert signaling component of the black box. The auditory alert signal was activated when the driver approached preselected speed, braking, and vehicle handling parameters. When the auditory alert signal threshold was exceeded, "penalty points" were also recorded in each individual driver's record. The linear distance interval for auditory alert and penalty point awards went from a baseline low of 0.018 miles to a postdeployment high of 15.8 miles. Significant improvements in front seatbelt use were noted, going from 13,500 seatbelt violations predeployment to four postdeployment. There was also a substantial savings in maintenance costs, netting enough to pay for the acquisition and installment of the black boxes within a short time frame.

In a very brief discussion of the limitations of the study, the authors note that the technology should be further tested across a broad spectrum of systems, including rural/volunteer agencies. However, they also question whether additional research is warranted, or even ethical, considering the dramatic results reported in this paper.

[Lindsey JT. The effects of computer simulation and learning styles on emergency vehicle drivers' competency in training course. 2004. Doctoral dissertation, University of South Florida.]

The focus of this research is the impact of a driver simulator on emergency vehicle operator's performance as measured during a subsequent hands-on driving course. In a comprehensive review of the impact of simulation technology across both the medical and broad vehicle operations field, the author notes that ambulance drivers operate in an environment with multiple distractions, including the patient care activity occurring in the rear of the vehicle itself. He notes that simulators provide a "safe" environment for training emergency medical personnel without endangering themselves, other crew members, the patient, or the public. While the long-term impact of simulator training, coupled with hands-on experience behind the wheel, has not been measured, the author notes that the shortterm impact is substantial and represents both safety achievements and cost savings.

[Proudfoot S. Ambulance crashes: Fatality factors for EMS workers. EmergMed Serv. 2005;34(6):71,73-74.]

This article is a focused summary of Ambulance Crash-Related Injuries among Emergency Medical Services Workers-United States, 1991-2002 (CEXI, 2003) reported above. The conclusions are more prescriptive, stressing the need for driver screening for previous moving violations coupled with initial and ongoing training.

Lights and Siren Use

[Brown LH, Whitney CL, Hunt RC, Addario M, Hague T. Do warning lights and sirens reduce ambulance response times? Prehosp Emerg Care 2000;4(l):70-74.]

In the study, the authors examine the use of lights and sirens pertaining to the response interval from the time of call until arrival at the scene. The treatmentas-usual group (lights and sirens) was compared to a control group that followed the same response route without lights and sirens. In an urban environment, the mean time savings was < 2 minutes. The authors conclude that while the time savings was statistically significant, it would only have been clinically significant in a few, specific instances. The authors conclude by calling for a large, multicenter trial to suggest national policies on the use of lights and sirens during the response phase.

[Ho J, Casey B. Time saved with use of emergency warning lights and sirens during response to requests for emergency medical aid in an urban environment. Ann Emerg Med. 1998;32(5):585-588.]

This prospective study examines the entire response continuum (i.e., leaving the station or deployment area until arrival at the hospital). The authors conclude that there is a 38.5% total time reduction when lights and sirens were used in this urban environment with response distance of 0.20-8.00 (mean 2.3) miles. The authors draw no conclusion about the importance of this time savings on patient outcome, but note that other studies have been very vague about what conditions warrant such time-saving response. The primary limitation of this study is that it is small, representing 64 emergency responses.

[Hunt RC, Brawn LH, Cabinum ES, Whitley TW, Prasad NH, Owens CF. Is ambulance transport time with lights and siren faster than that without? Ann Emerg Med. 1995;25(4):507-511.]

This is one the earliest of several studies that have looked at the time savings of responding with lights and sirens, in this case, specifically during the transport (rather than response) phase of care. In this midsize community of 46,000 people, the average time savings was 43.5 seconds. The authors conclude that such a minimal tune savings does not warrant a lights and siren transport except under very narrowly prescribed circumstances. The authors do note, however, that results of lights and siren transport need to be examined in other venues, specifically mentioning rural environments where transport distances may be much greater.

[Kupas DF, DuIa D], Pino B]. Patient outcome using medical protocol to limit "lights and sirens" transport. Prehosp Disast Med. 1994; 9(4):226-229.]

The authors examine the outcome of patients following the implementation of a protocol governing the use of lights and sirens during transport of the patient from the scene to the hospital. The setting was a rural/suburban county with a mixed EMT-P, EMT-B crew configuration. There were 1625 patients enrolled in the study. Of these, 130 (8%) met the criteria for transport using lights and sirens. Of the 92% of the transports that did not involve the use of lights and sirens, nearly one-half received some advance life support intervention either prior to or during transport. There were no adverse events associated with the nonlights and siren transports. Based on these finding, the authors recommend the establishment of protocols concerning lights and siren transport and the ongoing medical oversight of those protocols (a sample protocol is included as Appendix 1).

[Lacher ME, Bausher JC. Lights and siren in pediatric 911 ambulance transports: are they being misused? Ann Emerg Med. 1997;29(2):223-227.]

The authors reviewed audiotapes of all ambulance responses involving a pediatric medical control center for a seven-month period. The resultant sample was 504 responses, of which 312 involved transport using lights and sirens. The final disposition charts were reviewed for each case and gross determinations of "appropriateness" of lights and sirens was rendered. Of the 312 cases, 189 were determined to be appropriate for the condition of the child and 123 were gauged as inappropriate. Basic life support (BLS) personnel were more likely to engage in an inappropriate lights and siren transport. The authors conclude that EMS personnel require additional education on the benefits of lights and siren use.

[National Association of EMS Physicians and the National Association of State EMS Directors. Use of warning lights and siren in emergency medical vehicle response and patient transport. Prehosp Disast Med. 1994;9(2):133-136.]

This formal position paper by two prominent national organizations concerned with emergency medical care discusses the risks associated with emergency responses involving the use of warning lights and sirens. The paper includes a series of 11 recommendations. Among these recommendations are the need for judicious use of warning lights and sirens based on the patient's medical condition, the need for close oversight by the EMS agency's medical director, and the need for a national reporting system for emergency vehicle collisions. The paper was formally adopted in 1994 and reaffirmed in 2002.

[O'Brien DJ, Price TG, Adams P. The effectiveness of lights and siren use during ambulance transport by paramedics. Prehosp Emerg Care 1999;3(2):127-130.]

This study involves a convenience sample of 75 ambulance calls that, in the opinion of the emergency medical personnel at the scene, required use of lights and sirens during transport to the hospital. The same route was followed by a second vehicle traveling within the normal transportation flow. The authors conclude that there was a significant time savings with using lights and sirens. The mean time savings was 230 seconds. Of these 75 runs, receiving physicians conjecture that four may have clinically benefited from the time savings. There is a correlation between the number of stop lights encountered and traffic intensity with the total time savings. Likewise, there is a relationship between distance traveled and time saved. The authors conclude that lights and siren transports are warranted under certain circumstance, and that availability of advanced life support (ALS) personnel can reduce the frequency of such responses. The patients enrolled in this convenience sample had emergencies of a predominately medical etiology, and the findings may not generalize to a more trauma- oriented setting, common in some rural communities. Legal and Ethical Risks

[Eckstein M. Primum non nocere-first do no harm: an imperative for emergency medical services. Prehosp Emerg Care 2004;8(4):444- 446.]

In this editorial, the author reminds EMS providers that their first responsibility is to "do no harm" and challenges aggressive driving response tactics as violating that tenet. He notes that ambulances are 13 times more likely to be involved in a crash, and these crashes are rive times more likely to result in an injury than other vehicles. He notes that the cost of these crashes exceeds $500 million annually. He suggests that alternative deployment and response characteristics of EMS units may result not only in fewer crashes, but better outcomes.

[Vukrnir RB. Medical malpractice: managing the risk. Med Law 2004;23(3):495-513.]

This article provides a review and analysis of previously published articles pertaining to the likelihood of medicolegal errors. The author describes high-risk encounters for emergency physicians, and more germane to this discussion, notes that, in EMS, the most frequent area for the increased incidence and recovery amounts in verdicts pertained, not to clinical care issues, but rather to ambulance crashes. The author further notes that attention to this high-risk area could help reduce subsequent medicolegal risk, reduce costs, and improve patient care.

[Whiting J, Dunn K,MarchJ, Brown L. EMT knowledge of ambulance traffic laws. Prehosp Emerg Care 1998;2(2):136140.]

This research involves a survey of emergency medical personnel at a statewide conference. The survey measured knowledge of specific traffic laws pertaining to emergency vehicle operations in the state of North Carolina. The sample size was 295. Out of a possible score ranging from O to 5 on the five-question survey, the median response was one (1) correct question. Volunteer emergency medical providers were twice as likely to miss the question pertaining to speed limits than their paid counterparts. Emergency medical personnel who had taken one or more emergency vehicle operations courses were more likely to score above the median and more likely to answer questions concerning yielding to traffic (i.e., same direction and oncoming) correctly. The authors conclude that additional emergency vehicle operation training is warranted.

DISCUSSION

There is an ever-increasing body of knowledge pertaining to crashes involving ambulances. One of the first conclusions that can be drawn from the extant literature is that driving an ambulance is a dangerous process.1,4-11 This finding is, in and of itself, not surprising given the "emergency" nature of the work. Of note is that in comparison to other "emergency" responders, specifically law enforcement and firefighters, emergency medical personnel are at greater risk of a fatal vehicle incident than their public safety colleagues.7 The reason for this increased risk is unknown, but could include the fact that a large portion of the EMS workforce is volunteer in nature and, therefore, has less experience.12-14 Additional factors may include inadequate screening of vehicle operators for previous violations,15,16 nonexistent or inadequate vehicle operations training,6-9,11,15,17-19 fatigue and distraction,1,6,9,15 poor vehicle design,9,15,19,20 poor knowledge concerning driving laws,21 and inadequate or nonexisting policies and procedures.

A key factor in ambulance crashes is operations in an emergency mode with warning lights and sirens engaged. Implicit in the use of these warning devices is the fact that the ambulance is using or requesting certain privileges that may include traveling above the speed limit, expecting traffic to yield, and assuming the right of way at intersections. Arguably, the most heavily researched aspect of ambulance response is the time savings are associated with the use of lights and sirens and the degree to which those savings "might" contribute to positive clinical outcomes. Several authors contend that the time savings is not significant and is unwarranted in all but the most extreme clinical circumstances.5,8,11,15,16,20,22-24,38,39 O'Brien et al. concluded that the use of warning lights and sirens did result in significant time savings (230 seconds), and that there were at least four of 75 cases in which those time savings resulted in improved clinical outcomes.25 Only Kupas et al. specifically define the clinical conditions under which response or transport might warrant the use of warning lights and sirens.24 In this article, only 8% of ambulance transports met the clinical criteria for the use of warning lights and sirens. The increasing practice of terminating unsuccessful resuscitations on scene, and some services' policies of transporting cardiac arrests with no lights and sirens to reduce the possibility of injury to personnel that are unrestrained while performing chest compressions, may further reduce the number of transports that should be done with lights and sirens.

The reluctance of emergency care providers to wear safety restraints, particularly while delivering care in the patient compartment, is also noted by several authors to contribute to the risk of injury or death.3,4,7,15-17,19,26-29,42 Ray and Kupas note that more individuals are likely to be injured or killed in each ambulance crash than in crashes of similarly sized commercial vehicles.11 They conclude that this is due to multiple people (i.e., providers, patients, and family) traveling unrestrained in the rear compartment. Limited discussion is available in the literature about making ambulances more "user friendly" to encourage restraint use for all occupants.19,30

Less is known about rural ambulance crashes, although several of the studies have been conducted in communities of 100,000 or less.23,27,29 Two studies specifically compared rural and urban crashes.3,31 Weiss et al. in their seminal work described and compared the characteristics of rural ambulance crashes from a variety of factors.3 They concluded that rural crashes were more likely to result in injuries, and that the injuries were more serious. This finding compares well to Pratf s work concerning the incidence of fatal crashes in rural areas and, in particular, with those involving "truck-like" vehicles.9 Ray and Kupas concluded that rural crashes were more likely to involve snowy roadway conditions and poorly lit nighttime roads.31 Both Weiss et al. and Ray et al. noted that, during rural crashes, the ambulance is more likely to strike a fixed object, such as a tree, guard rail, or signpost.3,31 While ambulance-specific data are limited, additional information pertaining to rural driving in general supports many of the findings pertaining to poorer road design, longer travel distances, higher rates of speed, inclement weather, and road-surface conditions.9,26,29,32

Little is known about the economic impact of rural ambulance crashes, although one author uses a figure of $500 million annually. Line-of-duty deaths for firefighters are estimated to cost between $900,000 and $1.2 million per occurrence.33"34 These economic costs are increased by long-term economic costs of survivors and further exacerbated by psychosocial impact.27 The impact of a fatal crash in a rural environment can be devastating to a volunteer EMS agency.27 Additional costs are evident in legal fees associated with injuries, fatalities, and property loss to civilians.15,35

One of the persistent challenges in answering any questions about rural ambulance crashes is the application of a consistent definition of rural. Some studies have used a non-Metropolitan Statistical Areas,29 others have used highway department definitions31 and still others have more arbitrarily determined the cut-off in population density hi their state.3 None of the research identified in this literature search used definitions of rural that are more consistent with current thinking in rural health care, such as the Economic Research Service Rural-Urban Continuum Codes.12

The general findings for reducing crashes and improving outcomes of crashes can be fit nicely into the headings of the three E's of prevention (education, engineering, and enforcement) as promulgated by the National Highway Traffic Safety Administration (Table 1).

Education

Many authors suggest the need for additional emergency vehicle operations training.5,6,8,11,16-18,20 Unfortunately, there is a high degree of variability in emergency vehicle driver training programs and little is known about the effectiveness of such courses, although the general injury prevention literature questions the effectiveness of driver's education courses. Additional study of such courses is essential. Because many crashes involve errors in judgment and anticipation, simulator training holds great promise in augmenting traditional "hands-on" courses.18 One challenge associated with this technology is ensuring that rural and frontier EMS providers have access to such simulators.

Enforcement

Policy development, implementation, and enforcement have been shown to have an effect on the "culture" of safety within an organization. Standard policies concerning the use of safety restraint systems, and warning lights and sirens, should be adopted and enforced by all departments.8"9 This intervention is immediately available to all rural EMS agencies and does not require a large outlay of funds to implement. For that reason, we have included a sample agency protocol that can be modified and adopted by any service wishing to do so (Fig. 1) Validation of these and other similar protocols is essential. Engineering

Technology has begun to have an impact on ambulance operations. In particular, the deployment of "black box" and "drive cam" technologies hold great promise in creating a safer driving environment.19,20,28,36 Vehicle modification, including crash avoidance technologies, also holds promise.19 Intelligent transportation systems research also has the potential to contribute to emergency vehicle safety, such as the animal detection and avoidance system created and tested by the Western Transportation Institute.37 As rural EMS agencies go through their normal ambulance purchase cycle, new vehicles should have "black box" or similar technology installed or services should consider installing these in their existing vehicles. However, the effectiveness of these devices is dependent on their consistent use and feedback to all drivers.

Ambulance vehicle redesign is critical to improvements in safety both in terms of crash avoidance and in crash survival. Visibility of the vehicle and EMS personnel should be improved through the use of retroflective clothing and large footprint chevron or similar markings on the vehicle. Changes in lighting color, placements, and patterns can also help in collision avoidance. Of critical importance are the structure, placement, and safety of rear compartment seating. Improvements in this area must be driven by accepted engineering science rather than by "intuition." For instance, some evidence exists that suggests that four-point restraints for side-facing occupants may increase injury severity for restrained EMS personnel. These restraints systems are the centerpiece of many manufacturers' occupant safety improvements and are also available as aftermarket upgrades. Additional testing of these systems is essential to ensure their efficacy.

CONCLUSION

Ambulances are a dangerous place to work. If you happen to work in a rural environment, they are even more dangerous. Findings from a review of the extant literature suggest that there are educational, enforcement, and engineering interventions that can decrease the risk of death and disability to rural and frontier emergency care providers as well as the patients and public that they serve.

This research was funded, in part, by the U.S. Department of Health and Human Services (DHHS), Health Resources and Services Administrations (HRSA), and Office of Rural Health Policy (ORHP) through the Rural Emergency Medical Services and Trauma Technical Assistance Center (REMSTTAC; Jacob Rueda, Project Manager).

The authors gratefully acknowledge Teri L. Sanddal, Heather A. Soucy, Stephen Torgerson, and the REMSTTAC Rural EMS Driving Workgroup for their research, writing, and editorial contributions to this paper.

Performance Parameters:

A. Review for correlation between dispatch classification/ category and documented mode of response to scene.

B. Monitor percentage of "911" calls using L&S during response to EMS calls. Routine or scheduled transports should be excluded. [Potential benchmark <50% of responses with L&S].

C. Review for documentation of reason for L&S transport when patient does not meet criteria listed in section A.13.b-A.13.h.

D. Monitor percentage of urgent/emergent ("911") calls using L&S during transport. [Potential benchmark >900 -95% of patients transported without L&S].

NOTES

1. These guidelines are secondary to and do not supercede the state Motor Vehicle Code.

2. Dispatch centers/PSAPs and EMS regions are encouraged to have medically approved EMD protocols that differentiate emergent responses (e.g., "emergency,""code 3,""red,""Charlie,""Delta," etc... ) from a lesser level of response (e.g., "urgent,""code 2,""yellow,""Alpha,""Bravo," etc... ), based upon medical questions asked by the dispatcher. The dispatch category classification, or determinant that justifies L&S use, should be documented on the patient care record.

3. Firefighters cross-trained as EMS personnel who respond in an EMS vehicle to a fire station or fire incident in order to complete a fire apparatus crew are considered an exception to this policy.

4. In most cases (up to 95% of EMS incidents), EMS personnel can perform the initial care required to stabilize the patienf s condition to a point where the small amount of time gained by L&S transport will not affect the patient's medical condition or outcome. In previous studies and in most situations, L&S transport generally only decreases transport time by a couple of minutes or less.

5. Each of these criteria refers to an acute change in the patient's condition. For example, a patient who is chronically comatose would not automatically require L&S transport because the individual does not follow commands (criterion 2.g.l). Additionally, if the patient improves with treatment and no longer meets the criteria, L&S transport is not necessary.

6. The American Heart Association gives a class III recommendation to L&S transport of patients in cardiac arrest. A Class III indication is not helpful and is potentially harmful. Providing CPR during L&S transport may increase the risk for injury to EMS personnel. L&S may be indicated in some situations where ALS is indicated, but not available or cancelled, because the ALS crew cannot rendezvous with the BLS crew prior to transport to the closest appropriate medical facility.

References

1. Zagaroli L, Taylor A. Ambulance driver fatigue a danger: distractions pose risks to patients, EMTs, traffic. Washington, DC: Detroit News Washington Bureau. 2003;1-8.

2. Levick NR, Swanson J. Managing risk and reducing crashes: Implementing a driver performance-measuring device in ground ambulances. Abstract. Prehosp Emerg Care 2005;9(1): 108.

3. Weiss SJ, Ellis R, Ernst AA, Land RF, Garza A. A comparison of rural and urban ambulance crashes. Am J Emerg Med. 2001;19(1):52- 56.

4. Centers for Disease Control and Prevention. Ambulance crashrelated injuries among emergency medical services workersUnited States, 1991-2002. Morbidity and Mortality Weekly Review 2003;52(8):154-156.

5. Eckstein M. Primum non nocere-first do no harm: An imperative for emergency medical services. Prehosp Emerg Care 2004;8(4):444- 446.

6. Erich J. Wheels of fortune. Every time you hit the streets, you take your life in your hands-how can you improve your chances? Emerg Med Serv. 2000;29(11):43-46,49-50,52.

7. Maguire BJ, Hunting KL, Smith GS, Levick NR. Occupational fatalities in emergency medical services: a hidden crisis. Ann Emerg Med. 2002;40(6):625-632.

8. National Association of EMS Physicians and the National Association of State EMS Directors. Use of warning lights and siren in emergency medical vehicle response and patient transport. Prehosp Disast Med. 1994;9(2):133-136.

9. Pratt SG. NIOSH Hazard Review. Work-related roadway crashes: challenges and opportunities for prevention. Washington, DC: Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health 2003; 1-92.

10. Proudfoot S. Ambulance crashes: fatality factors for EMS workers. Emerg Med Serv. 2005;34(6):71,73-74.

11. Ray AM, Kupas DE. Comparison of crashes involving ambulances with those of similar-sized vehicles. Prehosp Emerg Care 2005;9(4):412-415.

12. Institute of Medicine. Quality through collaboration: the future of rural health. Washington, DC: National Academies Press, 2005.

13. Thompson AM. Rural emergency medical volunteers and their communities: A demographic comparison. J Comm Health 1993;18(6):379- 392.

14. Chng CL, Collins J, Eaddy S. A comparison of rural and urban emergency medical systems (EMS) personnel: a Texas study. Prehosp Disast Med. 2001;16(3):117-123.

15. Custalow C, Gravitz C. Emergency medical vehicle collisions and potential for preventive intervention. Prehosp Emerg Care 2004;8(2):175-184.

16. Kahn C, Pirrallo R, Kuhn E. Characteristics of fatal ambulance crashes in the United States: an 11-year retrospective analysis. Prehosp Emerg Care 2001;5(3):261-269.

17. Larmon B, LeGassick T, Schriger D. Differential front and back seat safety belt use by prehospital care providers. Am J Emerg Med. 1993;11(6):595-599.

18. Lindsey JT. The effects of computer simulation and learning styles on emergency vehicle drivers' competency in training course. Doctoral dissertation, University of South Florida, 2004.

19. Barishansky RM. Next generation ambulance puts safety first. Emerg Med Serv. 2005;30,34.

20. De Graeve K, Deroo K, Calle P, Vanhaute O, Buylaert W. How to modify the risk-taking behaviour of emergency medical services drivers? Eur J Emer Med. 2003;10(2):111-116.

21. Whiting J, Dunn K, March J, Brown L. EMT knowledge of ambulance traffic laws. Prehosp Emerg Care 1998;2(2):136-140.

22. Ho J, casey B. Time saved with use of emergency warning lights and sirens during response to requests for emergency medical aid in an urban environment. Ann Emerg Med. 1998;32(5):585-588.

23. Hunt RC, Brown LH, Cabinum ES, Whitley TW, Prasad NH, Owens CF. Is ambulance transport time with lights and siren faster than that without? Ann Emerg Med. 1995;25(4):507-511.

24. Kupas DF, DuIa DJ, Pino BJ. Patient outcome using medical protocol to limit "lights and sirens" transport. Prehospital and Disaster Medicine. 1994; 9(4):226-229.

25. CCBrien DJ, Price TG, Adams P. The effectiveness of lights and siren use during ambulance transport by paramedics. Prehosp Emerg Care 1999;3(2):127-130.

26. Baker SP, Whitfield RA, O'Neill BB. Geographic variations in mortality from motor vehicle crashes. N Engl J Med. 1987;316(22):1384-1387.

27. Hunjadi D. From provider to patient. Emerg Med Serv. 2005;34(8):157-160.

28. Levick NR. An optimal solution for enhancing ambulance safety: implementing a driver performance feedback and monitoring device in ground emergency medical services vehicles. Ann PTOC/ Assoc Advanc Automot Med. 2005;49:35-50. 29. Maio R, Green P, Becker M, Burney R, Compton C. Rural motor vehicle crash mortality: The role of crash severity and medical resources. Accid Anal Prev. 1992;24(6):631-642.

30. Ferreira J, Hignett S. Reviewing ambulance design for clinical efficiency and paramedic safety. Appl Ergonom. 2004;36:97- 105.

31. Ray AM, Kupas DF. Comparison of rural and urban ambulance crashes in Pennsylvania. Abstract. Prehosp Emerg Care 2007;11(4):416- 420.

32. Moretti F. Fatalities on rural roads-'Main Street' for millions of Americans-occur at a rate two-and-a-half times greater than on all other routes, [press release] Washington, DC: TRIP (a national transportation research group), 2005.

33. National Safety Council. Estimating the costs of unintentional injuries, 2005. Itasca, IL: Statistics Department, National Safety Council, and Children's Safety Network, Economics and Data Analysis Resource Center, 2005.

34. Rice DP, MacKenzie EJ, Associates. Cost of injury in the United States: a report to Congress, 1989. San Francisco, CA: Institute for Health & Aging, University of California, 1989.

35. Vukmir RB. Medical malpractice: managing the risk. Med Law 2004;23(3):495-513.

36. Swanson J, Levick N. Device improves ambulance drivers' performance: cuts crashes and reduces costs for tires and maintenance. EMS Insider. 2005;32(3):3.

37. Jones C. Systems warn drivers of deer in headlights. USA TODAY. October 5, 2006.

38. Lacher ME, Bausher JC. Lights and siren in pediatric 911 ambulance transports: are they being misused? Ann Emerg Med. 1997;29(2):223-227.

39. Brown LH, Whitney CL, Hunt RC, Addario M, Hogue T. Do warning lights and sirens reduce ambulance response times? Prehosp Emerg Care 2000;4(1):70-74.

40. Auerback PS, Morris JA, Phillips JB, Redlinger SR, Vaughn WK. An analysis of ambulance accidents in Tennessee. J Am Med Assoc. 1987;258(11):1487-1490.

41. Christoffel T, Gallagher SS. Injury prevention and public health: practical knowledge, skills, and strategies. Gaithersburg, MD: Aspen Publishers, 1999:139-160.

42. Cook RT, Meador SA, Buckingham BD, Groff LV. Opportunity for seatbelt usage by ALS providers. Prehosp Disast Med. 1991;6(4);469- 471.

Nels D. Sanddal, PhDc, MS, REMT-B, Steve Albert, MA, Joseph D. Hansen, REMT-B, Douglas F. Kupas, MD

Received April 26,2007, from the Critical Illness and Trauma Foundation, Inc., Bozeman, Montana (NDS, JDH), Western Transportation Institute, College of Engineering-Montana State University, Bozeman, MT (SA), and the Department of Emergency Medicine, Geisinger Medical Center, Danville, PA (DFK). Revision received October 18, 2007; accepted for publication October 27,2007.

Address correspondence and reprint requests to: NeIs D. Sanddal, PhDc, MS, REMT-B, Critical Illness and Trauma Foundation, Inc., 300 North Willson Avenue, Suite 502E, Bozeman, MT 59715. e-mail: nsanddal@citmt.org.

APPENDIX 1

MODEL LIGHTS AND SIREN RESPONSE PROTOCOL

Adopted from Pennsylvania Statewide, Basic Life Support Protocols, Pennsylvania Department of Health, Bureau of Emergency Medical Services, November, 2006.

Used with permission.

Criteria:

A. All emergency medical services (EMS) incident responses and patient transports.1

System Requirements:

A. These guidelines provide general information and "best practice" guidelines related to the use of lights and sirens by EMS personnel during incident response and patient transport. Ambulance services may use these guidelines to fulfill the service's requirement for a policy regarding the use of lights and other warning devices as required by state regulation, or regions may use these guidelines in establishing regional treatment and transport protocols.

Policy:

A. Use of lights and other warning devices:

1. Ambulance may not use emergency lights or audible warning devices, unless they do so in accordance with standards imposed by state regulation (relating to Vehicle Code) and are transporting or responding to a call involving a patient who presents or is in good faith perceived to present a combination of circumstances resulting in a need for immediate medical intervention. When transporting the patient, the need for immediate medical intervention must be beyond the capabilities of the ambulance crew using available supplies and equipment.

B. Response to incident:

1. The EMS vehicle driver is responsible for the mode of response to the scene based upon information available at dispatch. If the PSAP or dispatch center provides a response category based upon EMD criteria, EMS services shall respond in a mode (lights and siren [L&S] or non-L&S) consistent with the category of the call at dispatch as directed by the dispatch center.2 Response mode may be altered based upon additional information that is received by the dispatch center while the EMS vehicle is enroute to scene.

2. L&S use is generally NOT appropriate in the following circumstances:

a. "Stand-bys" at the scene of any fire department-related incident that does not involve active interior structural attack, hazardous materials (see below), known injuries to firefighters or other public safety personnel or the need for immediate deployment of a rehabilitation sector.

b. Carbon monoxide detector alarm activations without the report of any ill persons at the scene.

c. Assist to another public safety agency when there is no immediate danger to life or health.

3. Special circumstances may justify L&S use to an emergency incident scene when the emergency vehicle is not transporting a crew for the purposes of caring for a patient:

a. Transportation of personnel or materials resources considered critical or essential to the management of an emergency incident scene.

4. Transportation of human or materials resources considered critical or essential to the prevention or treatment of acute illness/injury at a medical facility or other location at which such a circumstance may occur (i.e., transportation of an amputated limb, organ retrieval, etc).

C. Patient transport:

1. The crewmember primarily responsible for patient care during transportation will advise the driver of the appropriate mode of transportation based upon the medical condition of the patient.

2. L&S should not be used during patient transport unless the patient meets one of the following medical criteria:4'5

a. Emergent transport should be used in any situation in which the most highly trained EMS practitioner believes that the patient's condition will be worsened by a delay equivalent to the time that can be gained by emergent transport. Medical command may be used to assist with this decision. The justification for using this criterion should be documented on the patient care report.

b. Vital signs

i. Systolic BP < 90 mmHg (or < 70 + [2 x age] for patients under 8 years old)

ii. Adults with respiratory rate >32/minute or <10/minute

c. Airway

i. Inability to establish or maintain a patent airway

ii. Upper airway stridor

d. Respiratory

i. Severe respiratory distress. (Objective criteria may include pulse oximetry less than 90%, retractions, stridor, or respiratory rate > 32/minute or < 10 minutes.)

e. Circulatory

i. Cardiac arrest with persistent ventricular fibrillation, hypothermia, overdose/ or poisoning.Note: Most other cardiac arrest patients should not be transported with L&S.6

f. Trauma

i. Patient with anatomic or physiologic criteria for triage to a trauma center (Category 1 Trauma). Refer to Trauma Triage Protocol

g. Neurologic

i. Patient does not follow commands (motor portion of GCS < 5).

ii. Recurrent or persistent generalized seizure activity

iii. Acute stroke symptoms (patient has Cincinnati Prehospital Stroke Scale findings) that began within the last three hours. see Stroke Protocol.

h. Pediatrics

i. Upper airway stridor

ii. When in doubt, contact with the on-line medical director and provide additional direction related to whether there is an urgent need to transport with L&S.

3. No emergency warning lights or siren will be used when advanced life support (ALS) care is not indicated (for example, ALS cancelled by basic life support [BLS] or ALS released by medical command). 7

4. Mode of transport for interfatility transfers will be based upon the medical protocol and the directions of the referring physician or medical command physician who provides the orders for patient care during the transport. Generally, interfacility transport patients have been stabilized to a point where the minimal time saved by L&S transport is not of importance to patient outcome.

5. Exceptions to these policies can be made under extraordinary circumstances (e.g., disaster conditions or a backlog of highpriority calls where the demand for EMS ambulances exceeds available resources). These exceptions should be documented.

D. Other operational safety considerations:

1. The following procedures should be followed for safe EMS vehicle operations:

a. Daytime running lights or low-beam headlights will be on (functioning as daytime running lights) at all times while operating EMS vehicles during L&S and non-L&S driving.

b. L&S should both be used when exercising any moving privilege (examples include proceeding through a red light or stop sign after coming to a complete stop or opposing traffic in an opposing land or one-way street) granted to EMS vehicles that are responding or transporting in an emergency mode.

c. When traveling in an opposing traffic lane, the maximum speed generally should not exceed 20 mph.

d. EMS systems are encouraged to cooperate with the dispatch centers in developing procedures to "downgrade" the response of incoming units to Non-L&S when initial on-scene units determine that there is no immediate threat to life.

e. The dispatch category (e.g., "code 3,""ALS emergency," etc.) that justifies L&S response should be documented on the patient care report. The justification for using L&S during transport should also be documented on the patient care report (e.g., "gunshot would to the abdomen,""systolic BP< 90," etc.). f. Seat belts or restraints will be securely fastened to the following individuals when the vehicle is in motion:

i. All EMS vehicle operators

ii. All patients

iii. All non-EMS passengers (cab and patient compartment)

iv. All EMS practitioners (when patient care allows)

v. All infants and toddlers (these children should be transported in an age appropriate child seat if their condition allows). Children should not be placed in cab passenger seat with airbag.

Copyright Taylor & Francis Ltd. Apr-Jun 2008

(c) 2008 Prehospital Emergency Care. Provided by ProQuest LLC. All rights Reserved.

Source: Prehospital Emergency Care