Impact of limited rest areas on truck driver crashes in Saskatchewan: a mixed-methods approach

Background Long-haul truck drivers (LHTDs) suffer from long work hours often resulting in fatigue. Although several studies have reported that fatigue can contribute to crashes, no study has identified the location and patterns of fatigue-related crashes and solicited truck driver feedback on potential mitigation strategies. The purpose of this study is 1) to map the location of fatigue-related crashes and 2) examine the perceptions of truck drivers concerning fatigue-related crashes. Methods Using databases from the Saskatchewan Government Insurance, information on LHTD demographics, crashes and their causes, as well as crash location was analyzed. All fatigue-related crashes were then documented and mapped. Additionally, we interviewed 67 LHTDs (mean age = 53.0 ± 12.9; range 23–89; 95% were men) asking questions about fatigue, access to truck stops/rest areas, and provided recommendations for improvement. All interviews were subsequently analyzed using thematic analyses. Results On average, there were 20 fatigue-related crashes per year over the 10-year period. Fatigue-related crashes were common across Saskatchewan, however, there was a concentration of crashes along major roadways between major cities. There was a significant association between crashes with age and experience. Despite many LHTDs being fatigued, there was a lack of truck stops/rest areas along highway routes. LHTDs suggested having access to truck stops/rest areas 250–400 km apart with running water and washrooms available. Conclusions Additional truck stops and rest areas are needed in Saskatchewan to ensure LHTDs have more opportunities for rest to reduce fatigue in general, as well as to reduce the risk of fatigue-related crashes.


Background
Long-haul truck drivers (LHTDs) are exposed to long working hours and inconsistent sleep schedules, which can cause disruptions in both length and quality of sleep, resulting in over half of drivers reporting sleepiness while driving [1]. Additionally, fatigue may be exacerbated while driving in monotonous environments [2] such as the landscape seen in Central Canada. Multiple studies show that LHTDs have sleep-related problems ranging from 17 and 58% [3][4][5], with 32 to 32.8% of LHTDs reported making serious errors due to sleepiness [1,5].
Fatigue in LHTDs is an occupational and public health concern given its association with crash risk. Fatigue can lead to inefficient processing of the driving environment [6], poorer executive function [7], more hard-braking events [8], and greater variance in steering and lane positioning [9]. Studies found that 6.9-7.5% of truck driver crashes are related to fatigue [1,5]. Similarly, a U.S. study found that fatigue contributed to 4.2% of all LHTD crashes, however, those crashes accounted for 7.2% of non-lifethreatening injuries, 13.1% of collision-related hospital admissions, and 16.5% of all fatal traffic crashes [10].
Although instances of crashes are generally low, it is estimated that the prevalence of fatigue-related accidents is higher than reported in research studies [11,12]. LHTDs may be reluctant to report an accident caused by fatigue for employment and insurance purposes [13]. LHTD are subject to both acute and chronic (extreme) fatigue. Acute fatigue refers to general tiredness resulting from working long hours, poor sleep quality or driving monotonous routes, and without intervention, can become chronic/extreme fatigue after prolonged exposure [14]. Chronic fatigue results in poorer cognitive processing abilities further increasing the risk for crashes [15]. While there are programs available such as the North Fatigue Management Program that addresses fatigue in LHTDs [16], environmental changes are also needed to alleviate fatigue, such as increasing the number of safe parking locations [5].
Prior studies have examined truck parking at rest areas in Canada [17,18]. One study found that 81% of 1788 truck drivers reported having difficulties finding parking in urban areas, which caused them to exceed their hours of service, and more than 88% mentioned that there were not enough rest areas, roadside pull-outs and turnouts, or safe lanes to safely perform inspections. Additionally, amenities at current rest areas such as washrooms, drinking water, cellular phone service (in remote areas), food, showers, convenience stores and security (e.g., camera surveillance, fencing, lighting) were viewed as inadequate. The parking and rest stop shortage contributes to drivers exceeding their hours of service [17], as well as to driver fatigue and illegal parking [18], all of which are related to accidents. In short-haul truck drivers, difficulty finding parking was associated with a 3.7x increased risk of being in a crash [19]. However, none of these studies examined where crashes occurred or where truck stops and rest areas were needed.
Prior studies in industrial safety found that the risk of accidents increased after 2-h of continuous work. A 10min break every two hours, however, reduced the risk of accidents to that of baseline [20]. Soccolich and colleagues [21] found that a 30-min break (non-driving periods) reduced safety-critical events (e.g., crash, nearcrash, unintentional lane deviation) in LHTDs by 28-51%. Additionally, reducing fatigue may also reduce stress, which is also associated with risky driving behaviours [22]. While there are truck stops in the major cities in Saskatchewan (e.g., Saskatoon, Regina, Moose Jaw), little is known about whether truck drivers have adequate access to truck stops and rest areas along major routes that lead into the cities. The objectives of this study are twofold: 1) to examine where fatigue-related accidents occur in Saskatchewan; and 2) to assess the perceptions of LHTDs concerning fatigue and access to truck stops and rest areas. The findings from this study may help policy makers and LHTD stakeholders identify optimal areas of intervention (where truck stops/rest areas could be developed) to reduce the risk of crashes.

Methods
Protocol online directory (https://www.allstays.com/c/truckstops-saskatchewan.htm) listing the number of parking spaces available, restaurants and amenities (e.g., showers, payphone, internet). A truck stop is defined as being on the highway or close to the highway and is used to rest, eat and refuel [23]. Meanwhile, rest stops are defined as areas nearby a highway (or next to the highway) where drivers can stop to rest and use washrooms [24]. Based on this information, there were 45 truck stops; 38 of them being defined as small [25]. Of the small truck stops, 13 did not have parking spaces, 5 provided no information on the number of parking spaces, 15 had between 2 and 10 parking spaces, and 5 had 12-20 parking spaces. As truck drivers do not plan on stopping at small truck stops [26], we only plotted truck stops with more than 20 parking spaces on the map. Only 7 truck stops had more than 20 parking spaces, with two locations in Saskatoon, two in Regina, and one in Swift Current, Whitewood, and Moosomin, respectively. There were 21 rest stops across Saskatchewan with no rest stops specific for truck drivers. Only three rest stops advertised they had parking available for truck drivers, usually 4-5 parking spaces, along with a restroom and running water.
To gather additional information on fatigue, and access to truck stops and rest areas in Saskatchewan, LHTDs were recruited to participate in a semistructured interview from seven truck stops in Alberta and Saskatchewan. Truck drivers were approached to participate in an interview by the researchers at each of the truck stops. To be eligible, truck drivers were required to be: 1) Canadian; 2) have a Class 1 driver's license (or equivalent); and 3) had spent at least one night away from home while delivering a load at the time of recruitment.
Participants were asked open-ended questions on driving routes, access to truck stops and rest areas, availability of truck stop/rest area amenities, and required improvements to reduce fatigue among LHTDs. The interview guide is published elsewhere [27]. The interviews took, on average, 30 min to complete; all interviews were audio-recorded for transcription purposes. Sixty-seven participants completed interviews. The mean age of the sample was 53.0 ± 12.9 (range 23-89; 95% were men). Each participant was compensated $10 for their time. This study was approved by the Research Ethics Board at the University of Saskatchewan.

Analyses
The locations of the crashes were used to create a heat map to show where collisions occurred due to falling asleep, with cooler and hotter colours indicating fewer or more collisions, respectively. Demographic information including age, gender, medical conditions and crashes were described using descriptive statistics (e.g., Mean ± SD or frequency and percent). Group comparisons were performed using either a chi-square or an Analysis of Variance. Correlations, chi-squares or independent t-tests were performed to examine the association between the number of fatigue-related crashes and highway characteristics (e.g., routes, two vs. four-lane highways, length, and whether they have access to a truck stop).
All interviews were transcribed by the Social Sciences Research Laboratories at the University of Saskatchewan. NVivo 11, a qualitative data management tool, was used to analyze all 67 transcripts. Thematic analysis was used to analyze the data which consisted of identifying and recording patterns within the data. Initial codes were developed deductively, based on the research questions, with information separated into comments and recommendations. Subsequent sub-codes were developed by identifying commonalities within the data and organizing information into increasingly specific groups. By grouping common codes together, the final themes were identified. Data was analyzed using two independent coders; one identified the initial codes and sub-codes and the other confirmed their accuracy and developed the final overarching themes.

Results
Over the 10-year period, there were a total of 201 crashes that were caused by extreme fatigue and/or falling asleep while driving (0.6% of all crashes): 79 crashes involved extreme fatigue and 154 crashes were from falling asleep. There were also 32 cases who crashed that were listed as both extremely fatigued and had fallen asleep at the wheel. Only seven drivers who crashed had a diagnosed medical condition: 5 had diabetes, 1 had a stroke, and 1 had heart problems. Table 1 shows the demographic profile of the sample. There was a significant age difference between drivers who crashed that were extremely fatigued and/or whom fell asleep compared to drivers who remained awake (F(2,32,471) = 24.3, p < .001). The Scheffe's Post-Hoc test revealed that the fatigued (Mean = 38.1 ± 13.1 years) and fell asleep (Mean = 38.9 ± 13.6 years) groups were both significantly younger that those who were awake when they crashed (Mean = 45.8 ± 14.4 years, p < .001 for both).
There was a significant relationship between years of experience and crashes (F(2,32,471) = 17.6, p < .001). Drivers who crashed while awake had more experience than both drivers who crashed due to extreme fatigue (Mean = 10.5 ± 7.7 years vs. Mean = 6.5 ± 5.9 years, p = .002) and drivers who crashed due to falling asleep (Mean = 10.5 ± 7.7 years vs. Mean = 7.6 ± 6.5 years, p < .001). There were no differences in working experience between drivers who were fatigued and crashed and those who fell asleep and crashed. There were no gender differences across groups nor were there any significant associations between medical conditions and fatigue. Figure 1 shows where crashes occurred across Saskatchewan over a ten-year period due to drivers being either extremely fatigued or falling asleep. The largest concentrations of crashes were located in major centres: Saskatoon (n = 16) and Regina (n = 14). These two locations account for 13.6% of all crashes from falling asleep. The majority of the crashes outside those two cities are located on highways throughout the province. Table 2 shows the highway number, location and their characteristics (e.g., number of lanes, speed, length, access to truck stops) where fatigue-related crashes occurred. The association between number of crashes and whether there were truck stops/rest areas on 2-lane highways were examined separately. Additionally, the association between the number of crashes and whether there were one or two truck stops on 4-lane highways were assessed. The findings showed no significant association between the number of crashes and availability of truck stops/rest areas on 2-lane (t(11) = 1.46, p = .17) or 4-lane highways (t(4) = 1.40, p = .23). However, when the number of crashes were controlled for by length of highway, the number of crashes were significantly lower on 4-lane highways with two truck stops/rest areas compared to one or no truck stops/rest areas (p < .05). No differences on 2-lane highways emerged.
Overall, truck drivers who participated in the interview overwhelming mentioned that there is a shortage of trucks stops and rest areas in the Western Canadian Provinces, as well as Canada generally. Additionally, there is insufficient parking available at the truck stops available. Due to these parking issues, LHTDs said they will either choose to drive while fatigued or park in unsafe areas. When asked how far apart and what should be provided at truck stops and rest areas, LHTDs stated that they should be 50-200 km apart and should at least have a bathroom and running water.

Access to truck stops and rest areas
Participants were asked whether they were satisfied with the current amount of truck stops and rest areas available to them. Almost all of the participants (91%) stated that there are not enough truck stops and/or rest areas in Canada. Many participants also highlighted Saskatchewan as being particularly problematic for having insufficient truck stops and/or rest areas compared to other Canadian provinces. Participants reported the inability to find parking along highways/driving routes to be quite stressful.
Even those rest stops on, or turn offs, really there isn't enough of them compared to the amount of trucks on the road.

Parking availability
Parking was the primary concern of our participants in relation to insufficient truck stops and rest areas. Almost all participants (96%) stated that there was not enough parking, even when there were truck stops or rest areas. The biggest thing I have found with truck stops is there is no place for us to park, that's the big thing. It drives me crazy. -Participant 36

Driving while fatigued
Many drivers reported often being forced to drive to the next location in hopes that there are spots to park. In some cases, participants reported to continue driving despite being fatigued. This is especially a concern for those that start driving in the morning and finish their shift at night.
If you're tired and you're looking to pull-in and oh, it's all full, now you've got to make the decision: do I keep pushing it or do I do 'til the next one? And if you don't know when the next one is, then you're going to start pushing the envelope top the edge, and you're already tired; that's why you are looking for that rest area. -Participant 15 Just a place to pull over for the night, you get to the truck stops, take a look outside it's loaded. It's filled up by seven o'clock, you can't get in and then try and get out in the morning when you're blocked in.
-Participant 19 Oh sure, yeah I come here quite often but if you don't get here early you're not getting a parking spot. Lack of places that a guy can't stop for a bathroom break or for coffee break. Anything like that eh. And sleeping at night, that's… Depending on where you are, I just run, now I just run Alberta, Saskatchewan, Manitoba. -Participant 91

Impact of hours of service regulations
The lack of truck stops and rest areas were also a concern due to the hours of service regulations. Many drivers reported parking on the side of the road if their hours of service requirements have been met for the day and there were no truck stops/rest areas. This was noted as being stressful as many drivers then park in areas where safety concerns may exist (e.g., side of the highway; unlit and quiet road). For example, some participants reported needing to leave their trucks and walk a distance to access food and other amenities, and during that time, their trucks were either burglarized or had been attempted to be burglarized. Participants noted poor lighting, traffic, inability to sleep due to noise, and being in an unsecure area. The only shortfall I got is not enough safe parking anywhere in any of the provinces of Canada. Now with the electronic logs, everybody is fighting to get to the next truck stop so they can find a safe place to park and most of us are being stuck out on the side of the road. -Participant 30 There is very few rest stops… I've actually parked on the side of the highway, put my four ways on, lights out, put the four ways on, left the truck running. So you still gotta get your sleep. -Participant 89

Recommendations for improvement: amenities
The participants also provided input on the current truck stops/rest areas, as well as what they would want to see in newly developed rest areas. More than half of the participants reported that all truck stops and rest areas should have at least a bathroom and, ideally, running water. Even if truck stops have access to bathrooms and food, many participants noted that the lack of hygiene is a concern. Many participants also wanted showers at the truck stops.

Discussion
Over the 10-year period, there were 201 crashes related to fatigue (on average 20 crashes per year) in Saskatchewan alone. Younger LHTDs (in their 30's) are more likely to suffer from fatigue-related crashes than older LHTDs. Younger CMV drivers may not receive training in dealing with fatigue, may not perceive driving while fatigued to be risky [28] or they lack experience in dealing with fatigue. Another reason could be lack of access to rest areas and truck stops along major highways. It is well known that LHTDs work long hours and often report being fatigued [3][4][5]. Prior studies have reported LHTDs receiving an average of 6 hours of sleep per day [29], which can lead to poorer driving performance and an increased crash risk [1,5]. Our participants mentioned the lack of available rest areas/truck stops along highways and recommended that there should be rest areas and/or truck stops every 50-200 km or roughly 2 hours apart. Distances between cities often go beyond 300-400 km and many drivers reported being fatigued and stressed over the inability to stop and rest. This is consistent with our findings showing that highway routes ranges from 140 to 690 km in length often without access to any rest areas/truck stops.
The number of fatigue-related crashes was highest in the major cities (e.g., Saskatoon and Regina). This trend could be due to various reasons, including LHTDs being unable to rest prior to arriving in major cities, greater cognitive demands, traffic density and volume. However, the rate of fatigue-related crashes was lower on the divided four-lane highway between Regina and Saskatoon compared to similar highways going east or west from those two cities. Our findings show that fatigue-related crashes on divided four-lane highways are reduced on highways where there is more than one truck stop consistent with McArthur and colleagues [30] who found a direct relationship between greater access to truck stops and a lower number of fatigue-related crashes in the U.S.
Our findings show that there are fewer crashes when truck stops/rest areas are within 250 km of each other along highway routes. Having truck stops/rest areas 250-300 apart (or 4 h approximately) was considered adequate by our participants and may lead to a reduction in fatigue and fatigue-related crashes. Our findings, based on the location of crashes and participant recommendations, suggest possibilities for the development and expansion of truck stops and/or rest areas in the Saskatchewan cities of Yorkton, and Prince Albert (see Fig. 2). While there are truck stops in both cities, they have limited parking options and amenities for LHTDs.
Either the development or expansion of these rest areas would reduce travel-related fatigue and provide access to truck stops/rest areas that are at least 4 hours apart between the major cities of Saskatchewan (Regina and Saskatoon) and Alberta and Manitoba. Other recommended locations are Moosomin, Swift Current and Whitewood which have only recently been expanded to include more parking spaces and amenities (e.g., showers, restaurant). This is consistent with the needs of the participants stating the desire for adequate lighting, bathrooms, running water and showers.
While the recommendation of developing truck stops/ rest areas is easy in principle, it requires the coordinated efforts between truck drivers, the public and private sector to balance the costs associated with building and maintaining the truck stops/rest areas. Additionally, Montufar and colleagues [18] suggests developing real time software applications that can provide real time data on parking options, as well as developing a nationwide inventory of parking facilities (and their amenities) so drivers can plan their routes.
While this study provides new data related to the location of truck stops and crashes, there are a few limitations. Convenience sampling was utilized to identify participants for the semi-structured interviews; consequently, the participants may not have been representative of all LHTDs. However, convenience sampling is typically used in qualitative studies and given the large number of interviews performed, it is likely that a wide range of perspectives were gathered. Although we could determine the number of crashes and the location of the crashes, we could not determine the driving conditions (e.g., highway volume or traffic, bad weather) at the time of the crash. We also lacked information on the time of day the accident occurred and could not determine whether crashes occurred in peak traffic, at night or during the daytime. We also did not have data on how long the driver was driving prior to the crash nor any information on their decision to drive particular routes or highways. For this reason, it is difficult to discern whether particular highways can lead to an increased risk of crashes or whether it is simply a product of opportunity (e.g., more drivers on highways leads to more opportunity to crash). Additionally, we could not determine traffic volume on the respective roadways. However, these limitations can be potentially addressed using data from the Canadian Naturalistic Driving Study (CNDS) that prospectively tracked truck drivers over 2 years [31]. The CNDS collected information on truck driver's driving exposure and patterns using GPS devices, on-board diagnostic devices, various cameras and sensors, as well as eye monitors. Using such data could confirm whether LHTDs are indeed involved in fatiguerelated crashes based on eye movements, length of time driving, traffic density and volume, as well as road and weather conditions.

Conclusions
The study results support the existence of an association between fatigue-related crashes and a lack of truck stops and rest areas. The lack of truck stops and rest areas was reinforced by the perceptions of the LHTDs we interviewed. Given the large number of truck drivers on the road and the limited areas for rest, vested stakeholders are urged to consider developing additional truck stops and rest areas with essential amenities to reduce the risk of fatigue-related crashes. Mobile or webbased applications are needed to provide real-time information on parking availability along highway/freeway routes for LHTD and future studies should assess the impact of newly implemented truck and rest stops on driver fatigue and fatigue-related crashes in LHTD.