Article Text
Abstract
Objective: To critically appraise the published evidence for risk factors for injuries and deaths relating to work-related road traffic crashes.
Design: Systematic review.
Data sources: Electronic databases searched included Medline, EMBASE, PsycINFO, Transport database, and the Australian Transport and Road Index (ATRI) database. Additional searches included websites of relevant organizations, reference lists of included studies, and issues of major injury journals published within the past 5 years.
Inclusion criteria: Studies were included if they investigated work-related traffic crashes or related injuries or deaths as the outcome, measured any potential risk factor for work-related road traffic crash as an exposure, included a relevant comparison group, and were written in English.
Methods: Included studies were critically appraised using the GATE-lite critical appraisal form (www.epiq.co.nz). Meta-analysis was not attempted because of the heterogeneity of the included studies.
Findings: Of 25 studies identified, three of four robust (case–control and case-crossover) studies found an increased injury risk (i) among workers after extended shifts, (ii) for tractor-trailers with brake and steering defects, and (iii) for “double configuration” trucks. The fourth study showed that alcohol and drug use were not risk factors in an industry with a random testing policy. The best cross-sectional studies showed associations between injury and sleepiness, time spent driving, occupational stress, non-insulin-dependent uncomplicated diabetes, and use of narcotics and antihistamines.
Conclusions: Modifiable behavioral and vehicle-related risk factors are likely to contribute to work-related traffic injury. Fatigue and sleepiness—the most commonly researched topics—were consistently associated with increased risk.
Statistics from Altmetric.com
Internationally there is growing recognition that work-related road traffic crashes resulting in injury or death are a significant public health problem. Many occupations require people to drive as part of their work. The work-related Road Safety Task Group in the UK estimated that ∼25% of all road traffic crashes involve someone driving in a work-related capacity.1 Furthermore, traffic crashes comprise the largest category of work-related deaths, accounting for 25–60% of these fatalities.2 Most of these are commuter deaths,3 which are also likely to be underestimated as they are not typically ascribed to a work-related journey.4
Recent occupational health and safety legislation has highlighted the obligations of employers to provide safe work places, explicitly including the work vehicle as a workplace.5 This has implications for organizational policies and procedures with respect to purchasing vehicles, vehicle maintenance, driver selection and training, work schedules, and indeed for the potential influence of the organization’s safety climate as a whole.
Although risk factors for road traffic crashes in general have been well described, with alcohol, speed and sleepiness considered major contributors,6 common modifiable causal factors for road crashes in the work context have received much less attention. The aim of this review was to systematically identify and critically appraise epidemiological studies that have investigated risk factors contributing to work-related road traffic injuries.
METHODS
Inclusion criteria
Studies were included in this review if they specifically investigated work-related road traffic crashes or related injuries or deaths as the outcome, measured any potential risk factor for work-related road traffic crash as an exposure, included a relevant comparison group to enable the magnitude of risk to be quantified, and were written in English. Although many studies investigated factors associated with known risk factors (eg, driver fatigue and sleepiness), these were excluded if they did not investigate the impact of such factors on the occurrence of crashes and associated injuries or death.
Data sources
Electronic databases searched included Medline (1966–2006), EMBASE (1980–2006), PsycINFO (1806–2006), Transport database (1988–2006), and the Australian Transport and Road Index (ATRI) database. The websites of organizations and institutions involved in work-related road traffic safety and work-related injury prevention were searched for relevant publications and reports. Reference lists of included studies were searched for additional relevant studies, as were issues of major injury journals published over the past 5 years (Injury Prevention, Accident Analysis and Prevention, Traffic Injury Prevention, Injury, Journal of Safety Research, and Injury Control and Safety Promotion).
Search terms
Keywords used included both MeSH and text words: MeSH headings included: accidents, traffic/; Accidents, occupational/risk factors/; human engineering/; automobile driving/; human factors/; causality/; risk assessment/; apnea/or sleep apnea syndromes/or sleep apnea, central/or sleep apnea, obstructive/; cannabis/; methamphetamine/; street drugs/; alcohol drinking/; work schedule tolerance/; sleep disorders, circadian rhythm/; cellular phone/; smoking/; quality of life/; morbidity/; wounds and injuries/.
Text words included: car or motor vehicle crash; car or motor vehicle accident; traffic accident or crash; road crash; road trauma; work-related; commercial; professional driver; occupational; industrial; workforce; workplace; occupational road risk; occupational hazard; driving behavior; causation; occupational stress; recreational drugs; marijuana; sleep disorder; driver drowsiness; shift work; work schedule; distraction; inattention; cell or mobile phone; work environment; disability; morbidity; injury; productivity.
Quality assessment
Included studies were classified by design and critically appraised. Two reviewers assessed study quality using the GATE-lite critical appraisal form (www.epiq.co.nz) (GR, SS). Where there was disagreement, a third reviewer was consulted (RJ or SA).
Evidence tables for included studies were developed, reporting participants, exposure, confounders considered, outcomes, results and relevant comments with regard to study quality and biases. Meta-analysis was not attempted because of the heterogeneity of included studies.
RESULTS
The database searches undertaken located over 400 papers. Restricting this search using the limits “English language” and specified epidemiological study designs (case–control, cross-sectional, cohort and case-crossover) further reduced the number of studies to ∼180. Abstracts from each of these studies were examined for eligibility, and the full texts of ∼70 papers were formally appraised. An additional three studies were identified and included from the reference lists of included studies and other reports.
A total of 25 studies fulfilled the inclusion criteria and were included in this review (table 1). Three of these were case–control studies7–9 and there was one case-crossover study.10 The remainder comprised two prospective cohort studies1112 and 19 cross-sectional/retrospective cohort studies that measured exposures cross-sectionally and outcomes retrospectively (ie, outcomes that occurred before exposure measurement). Two studies were based on a combination of routinely collected data and survey information13 and the others were based on surveys.
Studies took place across a wide variety of settings, including the UK,1112 USA,7–1014–18 Canada,1319 Brazil,20 Jordan,21 former Yugoslavia,2223 Turkey,24 Thailand,25 Argentina,26 Finland,27 Israel,28 Australia,2930 and Crete.31 Types of participants included were diverse. Most studies focused on truck drivers, but several also included transit industry employees,7 bus drivers,21242531 taxi drivers,2429 professional drivers,222327 company car drivers,1112 and commercial drivers.16 A few studies focused on medical personnel, specifically interns,10 house staff,15 and nurses.1417
In addition to the standard demographic variables collected by all studies, specific exposures investigated included use of recreational drugs and alcohol,7 work hours and shift work,101429 equipment defects and truck operating characteristics,9 truck configuration and driver characteristics,8 type of freight carried,31 occupational stress,11 medical conditions,131923 carbon monoxide exposure,22 fatigue, drowsiness, sleep disordered breathing, sleep habits, poor sleep quality,15161824–26283031 choice of time and place of rest for long-haul truck drivers,27 difficulty finding parking when tired,28 body mass index, fitness, physical, distracting activities and personal (musculoskeletal) discomfort,17 driver training, incentive schemes, driving more than one car or van as part of their job, hours of work, supervisory status, and type of firm.12
Outcomes were mainly work-related car or truck crash injuries. These were derived from or corroborated by police reports or other routinely collected data (eg, insurance claims)7–111318192329 or self-reported.14–1624–2830 In one study the derivation of the outcome was not specified,22 and in another study, the outcome was “accident liability” defined as the “expected number of accidents that he/she will experience during a given period of time (usually one year)” estimated using self-reported accidents in company car drivers compared with “ordinary” drivers.12
The case–control and case crossover studies—generally robust designs for investigating risk factors—were considered to be of reasonable quality. The key findings of these studies were: (i) extended shifts doubled the risk of a crash for medical interns (OR 2.3 (95% CI 1.6 to 3.3))10; (ii) serious (out-of-service) brake and steering defects increased the risk of crash in tractor–trailer trucks (RR 1.6 (95% CI 1.02 to 2.39)) and (RR 2.6 (95% CI 1.17 to 5.94)), respectively9; (iii) trucks with “double” configuration (“Western” and “Rocky Mountain”) had over three times the risk of crash compared with tractor–trailer trucks (RR 3.17 (95% CI 2.33 to 4.31))8; (iv) alcohol and drugs did not appear to be major risk factors for work-related crashes in companies where random drug and alcohol testing was carried out routinely.7
Only limited conclusions can be drawn from the remaining cross-sectional/retrospective cohort studies. One study noted that the frequency of hypersomnolence (a score >10 using the Epworth Sleepiness Scale) was higher among truck drivers with a history of crashes compared with those without such a history (p = 0.005). In this study, age >28 years was protective against hypersomnolence (OR 0.45 (95% CI 0.24 to 0.84)) but snoring (OR 1.89 (95% CI 1.02 to 3.5)) and working for >10 h without a rest (OR 2.07 (95% CI 1.21 to 3.51)) were independent risk factors for hypersomnolence.20 Sleepiness was also associated with increased risk of a crash in another cross-sectional/retrospective cohort study investigating sleepiness and sleep disordered breathing in commercial vehicle drivers (OR 1.24 (95% CI 1.10 to 1.34)).30 In the same study, the use of narcotics was associated with a significantly higher risk of a crash (OR 2.10 (95% CI 1.30 to 3.38)), and the use of antihistamine medication was associated with a threefold increase in the risk of crashing (OR 3.15 (95% CI 0.97 to 10.28)).30 Furthermore, antihistamines more than doubled the risk of multiple crashes (OR 2.64 (95% CI 1.19 to 5.84)).
Two reasonably well-conducted cross-sectional/retrospective cohort studies investigating medical conditions found an increased risk of truck-crash-related injury in people with uncomplicated diabetes who did not use insulin.1319 The second study suggested a link between occupation stress as measured by a valid occupational stress indicator and road traffic crashes.11
DISCUSSION
This review identified few methodologically sound studies examining risk factors relating to work-related road traffic injuries. Although no definitive conclusions can be drawn about factors that contribute to work-related traffic injury, the most commonly researched topic and consistent evidence related to fatigue- and sleepiness-related factors.10203032 Medical conditions such as non-insulin-dependent diabetes and cardiovascular disease were also found to have an association with work-related road traffic crashes.131923 Use of narcotics and antihistamines was associated with a doubling of crash risk, and the use of antihistamines increased the risk of multiple crashes.30 Of note, alcohol and other drugs may not be a major factor in work-related crashes, particularly where random testing is in place.7
Key points
Work-related road traffic crashes resulting in injury or death are a significant public health problem.
There is a lack of quality epidemiological evidence on risk factors for work-related road traffic crashes.
The few robust studies suggested that modifiable risk factors, including behavioral and vehicle-related factors, are likely to contribute to work-related traffic injury.
Fatigue and sleepiness were the most commonly researched topics and were consistently associated with an increased risk of injury.
This review involved a comprehensive and systematic search of the published literature to locate epidemiological studies, although it was limited to studies published in English.
There are a vast number of published reports about work-related road traffic crashes,133–38 but few epidemiological studies that met our inclusion criteria. Most studies in this review were of a combined cross-sectional/retrospective cohort design involving analyses of self-reported cross-sectional risk factor data (exposures) linked to historical self-reported data on crashes (outcomes). These studies often had low response rates and were prone to recall bias. In addition, injury crashes were often not distinguished from crashes resulting in property damage alone.
This review suggests that shift work, hours of work, and adequate rest breaks are issues that employers may want to consider. Screening for medical conditions such as diabetes may also be important. A recently published case–control study which investigated driver health and crash involvement (not work-related specifically) also found an increased risk for diabetic patients not receiving medication.39 Drivers should be made aware that the use of narcotics and antihistamine tablets may increase their risk of a crash. For the trucking industry, routine maintenance to prevent brake and steering defects will reduce crashes.
Issues relating to work-related road traffic injuries are complex. People who drive for work are a heterogeneous group.35 Whereas company drivers have been shown to have an excess accident liability,12 even within a single company, people at different levels of the organization who drive as part of their work may have varying exposure to risk factors and consequent accident liability.40
For the purposes of this review, we assumed a broad definition of “work-related road traffic crashes” which could include both bystander and commuter crashes. However, it was likely that most studies adopted a more conservative definition, as there were no explicit references to crashes involving commuters or bystanders. Consequently, risk factors for these crashes are not explicitly addressed in this review.
Although there are arguments in favor of blurring the distinction between work and non-work contexts when considering risk factors for injury in general,41 the extent to which established risk factors for non-work driving apply to work-related driving is not known. For example, alcohol, which is one of the major risk factors for road crashes in general, may not be as significant in work-related driving injuries in industries that have introduced random alcohol and drug testing.7 However, other risk factors—for example, distracting activities (cell phone use, eating, etc)—may be more important for work-related driving, where people spend more time in their vehicle and are likely to be operating under time pressures to meet deadlines for appointments or delivery of goods.37 One good-quality case–control study found that holding two or more jobs, working more than 60 h per week, being on a night work schedule, and having fewer hours of sleep were significant risk factors for a road traffic crashes in the general population.42 These findings have relevance when considering risk factors for work-related traffic crashes.
In summary, there is a lack of quality epidemiological evidence on risk factors for work-related road traffic crashes, and the few robust studies cover diverse risk factors in diverse populations. Sleepiness and fatigue-related risk factors featured most often in this review, suggesting that these factors are major causes of work-related traffic injury.
Acknowledgments
We thank the Accident Compensation Corporation NZ who funded this study.
REFERENCES
Footnotes
Competing interests: None.
Contributors: GR carried out the search, selected included papers, critically appraised the papers, developed the evidence tables, and wrote the initial draft of the paper. SS reviewed selected papers to ensure that they met inclusion criteria, independently critically appraised selected papers, and contributed to the writing of the paper. RJ and SA were reviewers for included critically appraised papers and contributed substantially to the interpretation of study findings and writing the paper.