Skip to main content
PMC home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2015 Sep 15.
Published in final edited form as: Policing. 2015;38(3):563–577. doi: 10.1108/PIJPSM-03-2015-0028

Law enforcement officers’ risk perceptions toward on-duty motor-vehicle events

Hope M Tiesman 1, Rebecca J Heick 2, Srinivas Konda 3, Scott Hendricks 3
PMCID: PMC4570477  NIHMSID: NIHMS718324  PMID: 26380563

Abstract

Purpose

Motor-vehicle-related events (MVEs) are the leading cause of on-duty death for law enforcement officers, yet little is known about how officers view this significant job hazard. The purpose of this paper is to explore officers’ motor-vehicle risk perception and examine how prior on-duty MVEs and the death or injury of a fellow officer influences this perception.

Design/methodology/approach

A state-wide random sample of 136 law enforcement agencies was drawn using publically accessible databases, stratified on type and size of agency. In total, 60 agencies agreed to participate and a cross-sectional questionnaire was distributed to 1,466 officers. Using six-point Likert scales, composite scores for motor-vehicle and intentional violence risk perception were derived. A linear regression multivariable model was used to examine factors affecting motor-vehicle risk perception.

Findings

Motor-vehicle risk perception scores were significantly higher than intentional violence scores. A prior on-duty motor-vehicle crash, prior roadside incident, or knowledge of fellow officer’s injury or death from a MVE significantly increased motor-vehicle risk perception scores. After controlling for potential confounders though, only prior on-duty crashes and roadside incidents impacted motor-vehicle risk perception.

Research limitations/implications

The study comprised primarily small, rural agencies and generalizability may be limited. Also, although the data were collected anonymously, reporting and response biases may affect these findings.

Originality/value

This study involved a large and diverse cohort of officers and explored motor-vehicle risk perception. A better understanding of officers’ risk perceptions will assist in the development and implementation of occupational injury prevention programs, training, and policy.

Keywords: Risk perception, Law enforcement, Motor-vehicle crash, Occupational injury

Introduction

In the USA in 2012, 4,383 workers died while on the job (Bureau of Labor Statistics (BLS), 2012a). This equates to approximately 12 workers dying every day in the USA while at work (2012). In that same year, nearly three million US workers experienced non-fatal work-related illnesses and injuries for an overall incidence rate of 3.4 cases per 100 equivalent full-time workers (BLS, 2013). The costs of occupational injury and illness in the USA are an estimated $250 billion per year (Leigh, 2011). This dollar amount exceeds that of chronic diseases including cancer, diabetes, and chronic obstructive pulmonary disease (2011). However, substantial these numbers may appear, the true toll of work-related injury and illness is estimated to be two to three times greater due to inconsistent official data sources and likely underreporting of injuries by workers and employers (Leigh et al., 2004; United States House of Representatives, 2008).

In 2012, according to the Bureau of Labor Statistics (BLS), 119 law enforcement officers (LEOs) died while on the job (BLS, 2012b). The Federal Bureau of Investigation’s Law Enforcement Officers Killed and Assaulted report found 95 deaths during the same time period (Federal Bureau of Investigation, 2013). Different inclusion criteria may be one explanation for this difference (Tiesman et al., 2013). Law enforcement work is a dangerous occupation due to the numerous hazards faced by LEOs. These hazards significantly increase the risk of dying or being seriously injured on the job and can occur in a dynamic environment. LEOs are exposed to physical hazards including violence and injury, physical exertion, armed and unarmed combat; psychological stressors such as exposure to crime scenes and exposure to death; organizational stressors such as shift work; mental health issues such as anxiety, depression, and post-traumatic stress disorder; cardiovascular disease; exposure to blood and bodily fluids; and increased risk for suicide (Hessl, 2003; Deschamps et al., 2003; Violanti et al., 2012; O’Hara and Violanti, 2009; Zimmerman, 2012; Gershon et al., 2009). LEOs are also routinely exposed to a much less commonly considered job hazard, high-speed and high-risk driving situations as well as work performed along busy roadways (Tiesman et al., 2010).

Across all occupations and industries, motor-vehicle-related events (MVEs) are consistently the leading cause of work-related death in the USA In 2012, MVEs accounted for 35 percent of all occupational injury fatalities in the USA (n = 1,567) (BLS, 2012a). Work-related motor-vehicle crashes are also expensive for employers. One study estimated that a single work-related motor-vehicle fatality cost employers approximately $500,000 in direct and liability costs and a non-fatal crash cost $74,000 (National Highway Traffic Safety Administration, 2003). MVEs were the leading cause of LEO fatalities in 2013 and have been for some time (National Law Enforcement Memorial Fund (NLEOMF), 2013). In the last decade, fatal MVEs, including motor-vehicle crashes and roadside incidents, outnumbered intentional acts of violence such as being shot, strangled, or stabbed (National Law Enforcement Memorial Fund, 2014). Besides spending numerous hours behind the wheel conducting vehicle patrols, LEOs have unique risk factors for MVEs including driving in inclement weather conditions, high-speed driving situations, working alongside interstates and roadways near speeding motor vehicles, and having a multitude of distractions inside the patrol car (Tiesman et al., 2010; Clarke and Zak, 1999). However, given these unique risk factors, LEOs can be complacent toward their increased risk for motor-vehicle crashes and roadside incidents (Wehr et al., 2012).

Risk perception has been defined as acknowledgement of the probability of incurring harm (Cox and Tait, 1991). An individual’s perception of risk plays a crucial role in occupational safety and health (Weyman and Kelly, 1999; Rosenstock et al., 1988; Melia et al., 2008; Arezes and Miguel, 2008). Not only does a worker’s perception of risk accurately predict specific safety behaviors, it has also been identified as an independent predictive factor for work-related injury (Mullen, 2004). An inaccurate perception of risk may lead to a worker’s inability to assess a situation and engage in appropriate safety precautions or behaviors (Huang et al., 2006; Harrel, 1998). This is pivotal for LEOs as they need to continually assess their risk for injury and death in an ever-changing environment. A better understanding of LEO’s risk perception and attitude toward workplace hazards and job tasks is important for the effective development and communication of workplace prevention programs and policies – especially related to motor-vehicle safety.

This special edition journal focusses on traffic and motor-vehicle safety among LEOs and fills a significant gap in the literature dealing with on-duty MVEs. The purpose of this paper is to explore LEO’s perception of risk of on-duty MVEs and identify factors that influence this risk including prior MVEs as well as knowledge of fellow officers’ motor-vehicle-related injury or death using data gathered from a large state-based study. We also compare an officers’ motor-vehicle risk perception to their risk perception of acts of intentional violence such as being assaulted or being shot on the job.

Data and methods

A cross-sectional design utilizing a stratified random sample was employed for this study. Data were collected via paper-and-pencil questionnaire. The questionnaire and research study were approved by the NIOSH Human Subjects Research Board and the US Government Office of Management and Budget. This project was funded by the National Institute for Occupational Safety and Health.

Study population

The study sample was developed in 2010 using publically accessible online resources to identify existing Iowa law enforcement agencies. A list of approximately 400 Iowa law enforcement agencies was compiled and stratified by type of agency (municipal, sheriff, state patrol) and size (small = 20 or fewer officers, medium = 21-50 officers, large = 51 or more officers). A total of 161 agencies were randomly selected for participation including 103 municipal agencies and 58 sheriff’s departments. Although the Iowa State Patrol was not randomly selected, they were included in the study. After removal of 26 sites that were no longer active agencies (two sheriff’s departments and 24 municipal agencies), 136 agencies remained. Recruitment activities, including phone calls, mailings of study flyers, and e-mails were directed at agency leadership. Of the 136 agencies invited, 60 agreed to participate for an overall agency response rate of 44 percent. This included 32 municipal agencies, 27 sheriff’s departments, and the Iowa State Patrol and comprised 1,466 sworn officers.

Data collection

The questionnaire comprised five sections: demographics, occupational characteristics, motor-vehicle safety and operations training, occupational safety practices and perceptions, and prior motor-vehicle crashes and roadside incidents. Sections were derived using existing validated tools when possible. Risk perception questions were independently developed for the study by the authors. The questionnaire was pilot-tested and peer-reviewed prior to use in the field. The questionnaire was distributed between September and December 2011. Questionnaire packets were delivered to agency leadership for distribution to sworn LEOs and included an introduction letter, paper-and-pencil questionnaire, and self-addressed stamped return envelope. Officers used the self-addressed stamped envelope to return completed questionnaires directly to researchers. Questionnaires were coded with a unique alpha-numeric string mapped to each agency, but not to individual officers. These codes were used to monitor survey returns by agency. Four weeks after the distribution of the questionnaires, agency leadership was informed of the number of non-respondents and asked to remind officers to return them. Returning a survey indicated consent to participate in the research study. In total, 79 percent of LEOs (1,157/1,466) returned a questionnaire.

Variable definitions

For this study, MVEs encompassed two types of on-duty events: motor-vehicle crashes and roadside incidents. Motor-vehicle crashes included all crashes, whether they were reported to administration or not. A roadside incident was defined as being struck by or nearly struck by a moving vehicle while working outside of a patrol car. Officers were asked to recall all MVEs occurring in the prior three years. Risk perception was assessed using a six-point Likert-type scale (1 = very unlikely to 6 = very likely) where respondents were asked to estimate the likelihood of being seriously injured or killed on the job related to four hypothetical situations: gunshot wound, assault, motor-vehicle crash, and roadside incident. Single-item stress-level questions have been shown to be reliable, valid, and accurate (Clark et al., 2011; Locke et al., 2007).

A principal component analysis using SAS PROC PRINCOMP was performed to identify the underlying structure of the risk perception variables. The first principal component accounted for 69 percent of the variance between the four variables (eigenvalue = 2.79). The second principal component accounted for 17 percent of the variation (eigenvalue = 0.638). Based on this second principal component, we decided to group the four risk perception variables into two composite scores: motor-vehicle risk perception score (motor-vehicle crash risk + roadside incident risk) and intentional violence risk perception score (gunshot risk + assault risk). Together, these two factors explain 86 percent of the variation within the four variables which is nearly the maximum desired explained variance of 90 percent (Kresta et al., 1994; Hatcher, 1994).

Statistical analyses

Data analyses were conducted using SAS V9.2. Descriptive statistics including counts, proportions, averages, and standard deviations were used to describe the study sample. Mean risk perception scores for each unique situation (assault, gunshot, motor-vehicle crash, and roadside incident) were calculated. Average motor-vehicle and intentional violence risk perception composite scores were then calculated and compared across the two primary risk factors: knowledge of a fellow officer’s death or injury and prior on-duty MVEs. First, mean composite risk perception scores were compared across officers who knew a fellow officer to be injured or killed in the last 12 months due to intentional violence (assault or shot), MVE (motor-vehicle crash or roadside incident), multiple causes, and no serious injury or fatality, with one-way analysis of variance. t-tests were used to compare average risk perception scores across respondents who were in an on-duty motor-vehicle crash in the last three years with those who were not, as well as those who were involved in a roadside incident in the last three years with those who were not.

Multivariate linear regression models were run to identify the effect of prior on-duty MVEs and knowledge of a fellow officer’s motor-vehicle-related death or injury on motor-vehicle risk perception scores, while controlling for potential confounders. Socio-demographic variables including race, ethnicity, age, gender, education, and relationship status were examined as potential confounders. Occupational variables including size of agency, type of agency, years of law enforcement experience, rank, division, shift, hours worked, hours of weekly driving, presence of agency-wide motor-vehicle policy, and yearly number of hours spent on motor-vehicle training were also examined. These variables were further included into models when p-values equaled at least 0.20.

Findings

Sample characteristics

Table I displays the characteristics of the study sample. Officers were primarily male (n = 1,062, 93 percent), white (n = 1,093, 94 percent), and in a committed relationship (n = 1,013, 88 percent). Their average age was 40.1 years (SD = 9.4) (data not shown). In total, 40 percent had a bachelors’ degree or higher (n = 461) and 29 percent had a technical certificate, vocational certificate, or associates degree (n = 333). On average, officers had been in law enforcement for 15.1 years (sd = 9.1) and over 30 percent had been in law enforcement for 20 years or more (n = 359, 31 percent). In total, 41 percent of officers were employed at a sheriff’s agency (n = 479), 38 percent at a municipal agency (n = 449), and 17 percent with the Iowa State Patrol (n = 196). Well over half of the officers were considered “officer,” “officer first class,” “specialist,” or “trooper” (n = 749, 65 percent).

Table I.

Socio-demographic and occupational characteristics of study participants

Characteristic n (%)
Age
21-24 25 (2%)
25-29 147 (13%)
30-34 190 (16%)
35-39 186 (16%)
40-44 214 (18%)
45-49 160 (14%)
50-59 189 (17%)
60+ 22 (2%)
Gender
Female 78 (7%)
Male 1,062 (93%)
Race
White 1,093 (94%)
Non-white 29 (2%)
Relationship status
In a committed relationship 1,013 (88%)
Not in a committed relationship 117 (10%)
Education
High school diploma/GED/some college 346 (30%)
Technical/Vocational Cert./Associate’s degree 333 (29%)
Bachelors or greater 461 (40%)
Rank
Officer/Officer First Class/Specialist/Trooper/Reserve 749 (64%)
Sergeant/Lieutenant/Corporal 226 (19%)
Major/Captain/Chief/Sheriff/Assistant Chief 93 (8%)
Other 86 (7%)
Years in law enforcement
Less than 5 158 (13%)
5-9 203 (17%)
10-14 222 (19%)
15-19 194 (16%)
20+ 359 (31%)
Type of agency
Sheriff 479 (41%)
State Patrol 196 (17%)
Municipal 449 (38%)
Total 1,157
Open in a new tab

Perception of occupational injury and fatality risk

Officers were asked about the likelihood of serious injury or death due to four occupational situations: assault, gunshot wound, motor-vehicle crash, and roadside incident (Table II). Each situation was ranked using a six-point Likert-type scale (1 = very unlikely to 6 = very likely). The occupational situation which resulted in the highest average risk perception score was roadside incident (mean = 3.99, SE = 0.04, SD = 1.31), followed closely by motor-vehicle crash (mean = 3.95, SE = 0.04, SD = 1.27). The percentage of officers who believed it was “very likely” that they could be seriously injured or killed due to a motor-vehicle crash or roadside incident was double that of officers who held this belief due to assault or gunshot wounds (12 and 13 percent vs 6 and 2 percent). The mean composite score for perceived motor-vehicle risk was significantly higher than the mean composite score for perceived risk of acts of intentional violence (7.94 vs 6.70, p < 0.0001).

Table II.

Perception of occupational injury and fatality risk by type of event

Very likely
n (%)		 Likely
n (%)		 Somewhat
likely n (%)		 Somewhat
unlikely n (%)		 Unlikely
n (%)		 Very unlikely
n (%)		 Mean score
(SE, SD)		 Mean composite
score (SE, SD)
Assaulted 68 (6) 333 (30) 198 (18) 204 (18) 254 (23) 70 (6) 3.52
(0.03, 1.30)		 6.70
(0.07, 2.25)
Being shot 20 (2) 238 (21) 147 (13) 208 (19) 417 (37) 93 (8) 3.17
(0.04, 1.17)
Motor-vehicle
crash		 136 (12) 362 (32) 246 (22) 114 (10) 226 (20) 39 (3) 3.95
(0.04, 1.27)		 7.94
(0.07, 2.33)
Roadside
incident		 147 (13) 303 (27) 289 (26) 111 (10) 233 (21) 45 (4) 3.99
(0.04, 1.31)
Open in a new tab

Effect of fellow officer’s occupational injury or death on risk perception

Table III displays the average motor-vehicle and intentional violence risk perception scores stratified by a fellow officer’s occupational injury or death. Four mutually exclusive categories were considered: injury/death due to intentional violence (assaults or gunshot wounds), injury/death due to a motor-vehicle event (motor-vehicle crash or roadside incident), multiple injuries/deaths from any of the four causes, and no known serious injury or death. For both motor-vehicle and intentional violence risk perception, scores were highest for those officers who knew of a fellow officer’s death or injury due to a MVE (mean = 8.31 and mean = 6.67, respectively). Both motor-vehicle and intentional violence risk perception scores were lowest among those officers who knew a fellow officer who had been injured or killed due to an assault or gunshot wound (mean = 7.59 and mean = 6.65, respectively). Knowledge of a fellow officers death or injury significantly impacted motor-vehicle risk perception scores (p = 0.0197), but not intentional violence risk perception scores (p = 0.91).

Table III.

Effect of a fellow officers’ occupational injury or death on risk perception scores

Fellow officers’ occupational injury or death from
Intentional violence
(assault or gunshot)
(n=61)
Mean (SE, SD)		 Motor-vehicle event
(n=254)
Mean (SE, SD)		 Multiple causes
(both intentional and
motor-vehicle)
(n=305)
Mean (SE, SD)		 No officer death or
serious injury
(n=790)
Mean (SE, SD)		 p-value
Perception of
motor-vehicle risk		 7.59 (0.37, 2.27) 8.31 (0.16, 2.61) 8.11 (0.16, 2.73) 7.90 (0.08, 2.32) 0.0197
Perception of
intentional
violence risk		 6.65 (0.32, 2.50) 6.67 (0.14, 2.27) 6.67 (0.13, 2.31) 6.74 (0.08, 2.20) 0.91
Open in a new tab

Effect of prior on-duty motor-vehicle crash or roadside incident on risk perception

Table IV displays the average motor-vehicle and intentional violence risk perception scores stratified by personal experience with on-duty motor-vehicle crashes and roadside incidents. Officers who had been involved in an on-duty motor-vehicle crash in the prior three years had significantly higher motor-vehicle risk perception scores than officers who had not been involved in a crash (8.49 vs 7.79, p = 0.0001). This same trend extended to those officers who had been involved in a roadside incident in the prior three years compared to those who had not (8.65 vs 7.81, p < 0.0001). On-duty motor-vehicle crashes and roadside incidents did not significantly impact risk perception scores for intentional violence (p = 0.09 and p = 0.39, respectively).

Table IV.

Effect of prior on-duty motor-vehicle crash or roadside incident on risk perception

MVC prior 3
years (n=220)
Mean (SE, SD)		 No MVC in prior
3 years (n=894)
Mean (SE, SD)		 p-value Roadside
incident in prior
3 years (n=179)
Mean (SE, SD)		 No roadside
incident in prior 3
years (n=921)
Mean (SE, SD)		 p-value
Perception of
motor-vehicle risk		 8.49 (0.16, 2.43) 7.79 (0.08, 2.42) 0.0001 8.65 (0.18, 2.43) 7.81 (0.08, 2.42) <0.0001
Perception of intentional
violence risk		 6.92 (0.14, 2.13) 6.63 (0.07, 2.26) 0.09 6.83 (0.17, 2.20) 6.68 (0.07, 2.24) 0.39
Open in a new tab

Officers’ perception of specific job hazards

Officers were asked to categorize their perceived level of danger for several types of job tasks using a four-point Likert scale ranging from “not at all dangerous” to “very dangerous” (Table V). Over a third of officers considered responding to scenes of other violence (non-domestic) to be “very dangerous” (n = 411, 36 percent), followed closely by responding to domestic violence calls (n = 397, 35 percent). Driving under non-emergency conditions was generally considered to be the least dangerous. Only 1 percent of officers considered this job task to be “very dangerous” (n = 8, 1 percent).

Table V.

Officers’ perception of specific job hazards

Not at all
dangerous n (%)		 Somewhat
dangerous n (%)		 Dangerous
n (%)		 Very
dangerous
n (%)
Driving
Driving under emergency
conditions		 12 (1) 222 (20) 569 (50) 328 (29)
Driving under non-emergency
conditions		 294 (26) 700 (62) 130 (11) 8 (1)
Response
Responding to domestic
violence calls		 6 (1) 191 (17) 535 (47) 397 (35)
Responding to other scenes of
violence		 7 (1) 148 (13) 563 (50) 411 (36)
Roadway
Working crash scenes on
roadways		 13 (1) 240 (21) 566 (50) 310 (27)
Making traffic stops on
roadways		 14 (1) 329 (29) 568 (50) 219 (19)
Open in a new tab

Multivariable modeling

Multivariable linear regression models were used to predict association with a higher motor-vehicle risk perception score, while controlling for potential confounders (Table VI). In the bivariate analysis, prior motor-vehicle crashes (p = 0.0001), prior roadside incidents (p < 0.0001), and knowledge of a fellow officer’s motor-vehicle-related injury/death (p = 0.006) were significantly associated with higher motor-vehicle risk perception. In the bivariate analysis, several other variables were found to be independently associated with motor-vehicle risk perception scores: age of officer (p = 0.08), size of agency (p = 0.12), type of agency (p = 0.001), rank of officer (p = 0.14), division (p < 0.0001), weekly hours behind the wheel (p < 0.001), years as a LEO (p = 0.20), and education (p = 0.05). All of these variables were considered for further analysis. Since age of officer and years as a LEO were significantly correlated, only years of law enforcement experience was further considered. Interaction terms were examined and none were found to be significant. After controlling for education, size of agency, type of agency, years of law enforcement experience, and driving hours per week, prior motor-vehicle crashes and prior roadside incidents were both found to be significant predictors of high motor-vehicle risk perception scores (p = 0.011 and p = 0.028, respectively).

Table VI.

Multivariable analyses of motor-vehicle crash risk perceptiona

Parameter estimates p-value
On-duty motor-vehicle crash in prior 3 years
Yes −0.484 0.011
No
Roadside incident in prior 3 years
Yes −0.457 0.028
No
Fellow officer motor-vehicle injury/death
Yes 0.233 0.36
No
Open in a new tab

Note:

a

Adjusted for education, size of agency, type of agency, years of law enforcement experience, and driving hours per week

Discussion and conclusion

Our findings indicate that LEOs were aware of the likelihood of injury and death associated with on-duty MVEs. Officers’ motor-vehicle risk perception scores were significantly higher than intentional violence risk perception scores. However, this perception of risk did not extend to officers’ view of the danger of specific job activities. Officers deemed the most hazardous job tasks to be those that involved a risk for intentional injury, such as responding to calls of potential violence. Another main finding from our study was that personal experience with on-duty MVEs significantly impacted officers’ perception of motor-vehicle risk.

With the exception of 2011, MVEs have been the leading cause of on-duty death among LEOs for the prior 16 years (NLEOMF, 2013). Since the 1960s, overall law enforcement fatalities have declined; however, motor-vehicle-related deaths have risen (NLEOMF, 2013). Recently, the law enforcement community has re-focussed their efforts on improving their understanding of MVEs, and more importantly, how best to prevent them (Ashton, 2012; Batiste et al. (2011); Ashton, 2011; Gustafson, 2013). Language on officer safety and health now includes topics like seat belt use, speed caps, cell phone restriction, and intersection crossing policies. Also, programs that raise knowledge of on-duty MVEs such as Below 100 and the MoveOver campaign may also add to this increased awareness among LEOs (Stockton et al., 2010; Moveoveramerica, 2013). It is possible that this messaging is impacting how LEOs view motor-vehicle hazards. Our study demonstrated that officers were aware of the likelihood of occupational injury and death associated with on-duty MVEs as indicated by the high motor-vehicle risk perception scores. This finding is important as it demonstrates that there is a baseline awareness of the seriousness of the hazards associated with motor vehicles in law enforcement.

While officers had significantly higher motor-vehicle risk perception scores compared to intentional violence risk perception scores, officers still deemed the most hazardous job tasks to be those involving the potential for assault. In fact, literature supports that these calls can be very dangerous for LEOs (Kercher et al., 2013). In our study, nearly 90 percent of officers reported that responding to “other” scenes of violence was “dangerous” or “very dangerous.” This is in stark comparison with the 12 percent of officers who believed that driving under non-emergency conditions to be “dangerous” or “very dangerous.” Studies have found that when workers regularly perform tasks and become familiarized with them, they are more likely to underestimate the risks associated with them (Fleming and Buchan, 2002). On average, officers in our study spent over 50 percent of their workweek behind the wheel of a patrol car (mean = 22.4 hrs, data not shown). It is feasible that officers come to view driving-related job tasks as non-threatening because they are routine.

Another possible explanation of this finding is the association between perceived risk and perceived control. Studies have demonstrated that when workers perceive they have control over tasks, they generally perceive less risk for those tasks (Weyman and Kelly, 1999). It has been well-documented that many drivers consider themselves to be more skilled behind the wheel and safer than the “average” driver (Svenson et al., 1985; Dejoy 1989). This “optimism bias” also extends to driver’s views on crash risk. Many drivers believe they are less likely to be involved in a motor-vehicle crash than others (Svenson et al., 1985; Dejoy, 1989). While there are no data specific to LEOs to support this hypothesis, there is also no reason to believe LEOs would be exempt from the optimism bias.

We found that prior on-duty MVEs significantly increased officers’ perception of motor-vehicle risk. Research has demonstrated that an individual’s experience with workplace injuries helps to shape their perception of risk and in turn, response to specific workplace hazards (Nelkin and Brown, 1984; Wiegman et al., 1991; Cree and Kelloway, 1997; Rundmo, 1995). Generally, people with a history of specific injuries score higher on perceived risk measures of those injury events recurring (Watzke and Smith, 1994; Glik et al., 1991; Rundmo, 1995; Tversky and Kahneman, 1973). In our study, the impact of prior MVEs was specific to motor-vehicle risk perception; occurrence of prior MVEs did not significantly impact intentional violence risk perception scores. What is currently unknown is how long after an on-duty MVE does it continue to impact an officers’ perception of risk. Also, is the severity of the MVE correlated with changes in perception of risk? Regardless, we believe that our data support a recommendation to provide LEOs with increased motor-vehicle training soon after an on-duty MVE. Intervening soon after a MVE, regardless of severity, capitalizes on this time period.

Knowledge of a motor-vehicle-related death or injury of a fellow officer was also associated with significantly higher motor-vehicle risk perception scores. Such knowledge also increased intentional violence risk perception scores, although not significantly. It may be that knowledge of a motor-vehicle death or injury affects overall perception of risk among officers. While this finding did not reach statistical significance, it may suggest a possible intervention for agencies. Prior research has found that the use of personal stories has a strong influence on workers (Lavack et al., 2008; Ricketts et al., 2010). Advances in technology such as body-worn and dash-mounted cameras allow agencies to capture events as they unfold (National Law Enforcement and Corrections Technology Center, 2012). Programs like Below 100 and the Street Survival Seminar are widely available in the USA and utilize images captured by these technologies to tell the stories of officers involved in critical incidents (Stockton et al., 2010; Calibre Press, 2013). Agencies could also consider adding personal testimonies of their own officers who have been involved in serious crashes in their driving safety training as a way to increase perception of risk and encourage others to engage in safety behaviors.

In the multivariate analysis, while motor-vehicle crashes in the previous three years had the greatest impact on motor-vehicle risk perception, prior roadside incidents were also an important predictor of motor-vehicle risk perception. While officers are trained on a variety of roadway procedures including parking a vehicle offset and using passenger-side approaches, these cannot fully prevent officers from being struck by moving vehicles. Civilian drivers must also be educated on safe roadway procedures. The Move Over America public health campaign is intended to increase public awareness of move over laws which require motorists to move into another lane of traffic when an emergency vehicle or responder is on the side of the road (Moveoveramerica, 2014). While every US state with the exception of Hawaii has a move-over law in place, a recent phone survey found that 71 percent of Americans had not heard of these laws.

While officers may feel that the prevention of roadside incidents is largely outside of their control, there are several strategies that could significantly reduce officer’s chances of being struck and killed while on the roadway. High-visibility personal protective equipment, such as reflective vests, and the use of flares and cones, alert motorists to crash scenes. The code of federal regulations states that all workers within the right-of-way of a federal-aid highway who are exposed to traffic shall wear high-visibility safety apparel (Code of Federal Regulation, 2007). The safety of officers can be further protected at crash scenes with the use of an incident management system and proper traffic management, especially for large crash scenes or scenes involving multiple vehicles. However, the decision of when to use an incident management system may be made by other non-law enforcement agencies like the Department of Transportation or local highway crews whose primary concern may be traffic backups and delays. Actively including these groups in crash response decisions may help to create a safer work environment for officers.

This study was not without limitations. First, the study was conducted in the state of Iowa and consisted of mostly small and rural law enforcement agencies. It is unknown if these findings are applicable to agencies in larger or more urban locations. Second, since the study was cross-sectional in nature, the results represent associations only and determining causation is not possible. Third, standardized and validated tools were not used to measure the construct of risk perception. Fourth, the recall period used in this study was three years. When deciding on an appropriate recall period to be used in survey research, the trade-off between minimizing recall bias and maximizing rate estimates must be balanced. Here, we used a longer recall period to calculate LEOs’ prior history of on-duty MVEs to improve capture of low-frequency events. However, by doing so we could have increased the likelihood of recall bias (Zwerling et al., 1995). Finally, the participation rate at the agency level was low at 44 percent. It is possible that agencies that chose not to participle differed from those agencies that did. However, among those agencies who did participate, the officer response rate was high at 79 percent.

We found that officers were aware of the likelihood of occupational injury and death associated with on-duty MVEs and this awareness may be due to the recent cultural change in law enforcement toward an emphasis on the prevention of on-duty MVEs. We also found that officers who had been involved in a prior on-duty MVE had higher motor-vehicle risk perception scores. There is little research on the occupational risk perception from a LEO standpoint, or even the risk perception of occupational motor-vehicle crashes with which to compare our results to. Knowledge of how an officer perceives specific job hazards and situations would be useful when developing and implementing evidence-based programs in the workplace. The adoption of protective behaviors and equipment is very much dependent on the workers’ perception of specific hazards (Huang et al., 2006). For example, restrictive driving policies, such as speed caps, may be viewed in a more favorable light if the officer is aware of the true fatality risk associated with job tasks that appear to be mundane. Perhaps working to change the culture of safe motor-vehicle operations among LEOs is a needed first step in the prevention of MVEs. Based on our findings, we believe that capitalizing on officers’ knowledge of this risk is a possible avenue for highly impactful prevention programs, which in turn may foster a safer workplace and potentially reduce occupational injuries and fatalities.

Future research needs

Much more information is needed on the causes, trends, and risk factors for officers’ on-duty motor-vehicle crashes and roadside incidents, especially on a national scale. At the agency level, such data can be obtained from workers’ compensation reports, motor vehicle crash reports, and on-site investigative reports. These quantitative data can be pivotal in furthering our understanding of on-duty MVEs and in the development of comprehensive evidence-based prevention programs. The use of qualitative research methods may also be beneficial in gaining greater insight into both officers’ and leaderships’ perceptions and experiences related to MVEs, allowing for more clear understanding of how and why these events occur, how officers are impacted by them, and how best to reduce the number and severity of these events. Finally, as agencies implement new training programs, standard operating procedures, and policy in an effort to reduce on-duty MVEs, these changes should be evaluated to determine their effectiveness.

Acknowledgments

This project was funded by the National Institute for Occupational Safety and Health.

Footnotes

The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health.

References

  1. Arezes PM, Miguel AS. Risk perception and safety behavior: a study in an occupational environment. Safety Science. 2008;46(6):900–907. [Google Scholar]
  2. Ashton RJ. Could changing our ways mitigate officers’ traffic deaths: highway safety initiatives. The Police Chief. 2011 Aug;78:118–122. [Google Scholar]
  3. Ashton RJ. Speeding cops. The Police Chief. 2012;79:72–74. [Google Scholar]
  4. Batiste JR, Wagers ML, Ashton RJ. Preventing traffic-related line-of-duty deaths. The Police Chief. 2011 Jul;78:52–55. [Google Scholar]
  5. Bureau of Labor Statistics (BLS) Census of Fatal Occupational Injuries Summary, 2012. 2012a available at: www.bls.gov/news.release/cfoi.nr0.htm (accessed June 6, 2014)
  6. BLS Fatal occupational injuries by occupation and event or exposure, all US, (2011) 2012b available at: www.bls.gov/iif/oshwc/cfoi/cftb0263.pdf (accessed June 6, 2014)
  7. BLS Occupational injuries and illnesses. Annual news release. 2013 available at: www.bls.gov/news.release/archives/osh_11072013.htm (accessed June 6, 2014)
  8. Calibre Press Street survival seminar. 2013 available at: http://calibrepress.com/calibre-press/street-survival-seminar/ (accessed June 28, 2014)
  9. Clark MM, Warren BA, Hagen PT, Johnson BD, Werneburg B, Olsen KD. Stress level, health behaviors, and quality of life in employees joining a wellness center. Am J Health Promot. 2011;26(1):21–25. doi: 10.4278/ajhp.090821-QUAN-272. [DOI] [PubMed] [Google Scholar]
  10. Clarke C, Zak MJ. 1999:1–10. Fatalities to Law Enforcement Officers and Firefighters, 1992-1997, Publication CF AR CWC, Department of Labor, Bureau of Labor Statistics, Washington, DC. [Google Scholar]
  11. Code of Federal Regulations Code of Federal Regulations (CFR) 23, Part 634.4. 2007 available at: http://edocket.access.gpo.gov/cfr_2007/aprqtr/pdf/23cfr634.4.pdf (accessed June 11, 2014)
  12. Cox S, Tait R. Reliability, Safety and the Human Factor. Butterworth-Heineman; Stoneham, MA: 1991. [Google Scholar]
  13. Cree T, Kelloway K. Responses to occupational hazards: exit and participation. J Occup Health Psychol. 1997;2(4):304–311. doi: 10.1037//1076-8998.2.4.304. [DOI] [PubMed] [Google Scholar]
  14. Dejoy DM. The optimism bias and traffic accident risk perception. Accid Anal Prev. 1989;21(4):333–340. doi: 10.1016/0001-4575(89)90024-9. [DOI] [PubMed] [Google Scholar]
  15. Deschamps F, Paganon-Badinier I, Marchand AC, Merle C. Sources and assessment of occupational stress in the police. Journal of Occupational Health. 2003;45(6):358–364. doi: 10.1539/joh.45.358. [DOI] [PubMed] [Google Scholar]
  16. Federal Bureau of Investigation 2012 statistics on law enforcement officers killed and assaulted. 2013 available at: www.fbi.gov/about-us/cjis/ucr/leoka/2012/press-release-summary-of-2012/press-release_leoka_-2012 (accessed June 6, 2014)
  17. Fleming M, Buchan D. Risk is in the eye of the beholder. Safety Health Practitioner. 2002;20:30–32. [Google Scholar]
  18. Gershon RM, Barocas B, Canton AN, Li X, Vlahov D. Mental, physical, and behavioral outcomes associated with perceived work stress in police officers. Criminal Justice and Behavior. 2009;36(3):275–289. [Google Scholar]
  19. Glik D, Kronenfeld J, Jackson K. Predictors of risk perceptions of childhood injury among parents of preschoolers. Health Educ Q. 1991;18(3):285–301. doi: 10.1177/109019819101800303. [DOI] [PubMed] [Google Scholar]
  20. Gustafson BG. Realizing safe driving: effects of multi-disciplinary research and intervention in California. The Police Chief. 2013 May;80:14–15. [Google Scholar]
  21. Harrel WA. Perceived risk of work-related injury and intentions to change employment. Psychological Report. 1998;84(1):273–276. doi: 10.2466/pr0.1999.84.1.273. [DOI] [PubMed] [Google Scholar]
  22. Hatcher L. A Step-by-Step Approach to Using the SAS System for Factor Analysis and Structural Equation Modeling. SAS Publishing; Cary, NC: 1994. [Google Scholar]
  23. Hessl SM. Introduction to the history, demographics, and health effects of law enforcement work. Clin Occup Environ Med. 2003;3(3):369–384. [Google Scholar]
  24. Huang YH, Ho M, Smith GS, Chen PY. Safety climate and self-reported injury: assessing the mediating role of employee safety and control. Accid Analy Preevntion. 2006;38(3):425–433. doi: 10.1016/j.aap.2005.07.002. [DOI] [PubMed] [Google Scholar]
  25. Kercher C, Swedler DI, Pollack KM, Webster DW. Homicides of law enforcement officers responding to domestic disturbance calls. Injury Prevention. 2013;19(5):331–335. doi: 10.1136/injuryprev-2012-040723. [DOI] [PubMed] [Google Scholar]
  26. Kresta JV, Marlin TE, MacGregor JF. Development of inferential process models using PLS. Computer and Chemical Engineering. 1994;18(7):597–611. [Google Scholar]
  27. Lavack AM, Magnuson SL, Deshpande S, Basil DZ, Mintz JH. Enhancing occupational health and safety in young workers: the role of social marketing. International Journal of Nonprofit and Voluntary Sector Marketing. 2008;13(3):193–204. [Google Scholar]
  28. Leigh JP. Economic burden of occupational injury and illness in the United States. Milbank Quarterly. 2011;89(4):728–772. doi: 10.1111/j.1468-0009.2011.00648.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Leigh JP, James P, Marcin J, Miller TR. An estimate of the US government’s undercount of nonfatal occupational injuries. Journal of Occupational and Environmental Medicine. 2004;46(1):10–18. doi: 10.1097/01.jom.0000105909.66435.53. [DOI] [PubMed] [Google Scholar]
  30. Locke DE, Decker PA, Sloan JA, Brown PD, Malec JF. Validation of single-item linear analog scale assessment of quality of life in neuro-oncology patients. J Pain Symptom Manag. 2007;34(6):628–638. doi: 10.1016/j.jpainsymman.2007.01.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Melia JL, Mearns K, Silva SA, Lima ML. Safety climate responses and the perceived risk of accidents in the construction industry. Safety Science. 2008;46(6):949–958. [Google Scholar]
  32. Moveoveramerica Protect law enforcement and emergency responders. 2014 available at: www.moveoveramerica.com (accessed June 16, 2014)
  33. Mullen J. Investigating factors that influence individual safety behavior at work. Journal of Safety Research. 2004;35(3):275–285. doi: 10.1016/j.jsr.2004.03.011. [DOI] [PubMed] [Google Scholar]
  34. National Highway Traffic Safety Administration . The Economic Burden of Traffic Crashes on Employers: Costs by State and Industry and by Alcohol and Restraint Use. Department of Transportation; Washington, DC: 2003. [Google Scholar]
  35. National Law Enforcement and Corrections Technology Center . A Primer on Body-Worn Cameras for Law Enforcement. ManTech Advanced Systems International Inc.; Fairmount, WV: 2012. [Google Scholar]
  36. National Law Enforcement Memorial Fund (NLEOMF) Law enforcement fatalities dip to lowest level in six decades. 2013 available at: www.nleomf.org/assets/pdfs/reports/2013-End-of-Year-Preliminary-Report.pdf (accessed June 2, 2014)
  37. National Law Enforcement Memorial Fund Causes of law enforcement deaths. 2014 available at: www.nleomf.org/facts/officer-fatalities-data/causes.html (accessed June 2, 2014)
  38. Nelkin D, Brown MS. Workers at Risk: Voices from the Workplace. University of Chicago Press; Chicago, IL: 1984. [Google Scholar]
  39. O’Hara AF, Violanti JM. Police suicide – a web surveillance of national data. Int J of Emerg Mental Health. 2009;11(1):17–24. [PubMed] [Google Scholar]
  40. Ricketts M, Shanteau J, McSpadden B, Fernandez-Medina KM. Using stories to battle unintentional injuries: narratives in safety and health communication. Social Science and Medicine. 2010;70(9):1441–1449. doi: 10.1016/j.socscimed.2009.12.036. [DOI] [PubMed] [Google Scholar]
  41. Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the health belief model. Health Education Quarterly. 1988;15(2):175–183. doi: 10.1177/109019818801500203. [DOI] [PubMed] [Google Scholar]
  42. Rundmo T. Perceived risk, safety status, and job stress among injured and noninjured employees on offshore petroleum installations. J Safety Res. 1995;26(2):87–97. [Google Scholar]
  43. Stockton D, Graham G, Yates T, Chudwin J, Willis B, Young M. Below 100: an initiative to change our profession – & you. Law Officer Magazine. 2010 Oct 10; available at: www.lawofficer.com/article/below-100/below-100 (accessed June 15, 2014)
  44. Svenson O, Fischhoff B, MacGregor D. Perceived driving safety and seatbelt usage. Accid Anal Prev. 1985;17(2):119–133. doi: 10.1016/0001-4575(85)90015-6. [DOI] [PubMed] [Google Scholar]
  45. Tiesman HM, Hendricks SA, Bell JL, Amandus HA. Eleven years of occupational mortality in law enforcement: the census of fatal occupational injuries, 1992-2002. Am J Ind Med. 2010;53(9):940–949. doi: 10.1002/ajim.20863. [DOI] [PubMed] [Google Scholar]
  46. Tiesman HM, Swedler DI, Konda S, Pollack KM. Fatal occupational injuries among US law enforcement officers: a comparison of national surveillance systems. American Journal of Industrial Medicine. 2013;56(6):693–700. doi: 10.1002/ajim.22182. [DOI] [PubMed] [Google Scholar]
  47. Tversky A, Kahneman D. Availability: a heuristic for judging frequency and probability. Cogn Psychol. 1973;5:207–232. [Google Scholar]
  48. United States House of Representatives Hidden tragedy: underreporting of workplace injuries and illnesses. 2008 available at: www.gpo.gov/fdsys/pkg/CHRG-110hhrg42881/pdf/CHRG-110hhrg42881.pdf (accessed June 6, 2014)
  49. Violanti JM, Fekedulegn D, Andrew ME, Charles LE, Hartley TA, Vila B, Burchfiel CM. Shift work and the incidence of injury among police officers. Am J Ind Med. 2012;55(3):217–227. doi: 10.1002/ajim.22007. [DOI] [PubMed] [Google Scholar]
  50. Watzke JR, Smith DBB. Concern and knowledge of safety hazards among older people: implications for research and prevention. Exp Aging Res. 1994;20(3):177–188. doi: 10.1080/03610739408253964. [DOI] [PubMed] [Google Scholar]
  51. Wehr K, Alpert G, Rojek J. The fear of the ninja assassin: understanding the role of agency culture in injurious and fatal on-duty vehicle collisions. Journal of California Law Enforcement. 2012;46:18–26. [Google Scholar]
  52. Weyman AK, Kelly CJ. Risk perception and risk communication: a review of the literature. Health and Safety Executive. 1999 Research Report No. 248/99. available at: www.hse.gov.uk/index.htm (accessed June 8, 2014)
  53. Wiegman O, Gutteling JM, Boehr H. Verification of information through direct experience with an industrial hazard. Basic and Applied Social Psychology. 1991;12(3):325–339. [Google Scholar]
  54. Zimmerman FH. Cardiovascular disease and risk factors in law enforcement personnel: a comprehensive review. Cardiol Rev. 2012;20(4):159–166. doi: 10.1097/CRD.0b013e318248d631. [DOI] [PubMed] [Google Scholar]
  55. Zwerling C, Sprince NL, Wallace RB, Davis CS, Whitten PS, Heeringa SG. Effect of recall period on the reporting of occupational injuries among older workers in the health and retirement study. Am J Ind Med. 1995;28(5):583–590. doi: 10.1002/ajim.4700280503. [DOI] [PubMed] [Google Scholar]

RESOURCES