When driving hurts: characterizing the experience and impact of driving with back pain

Introduction

Driving is considered an instrumental activity of daily living [1]. Driving is likewise an active process requiring cognitive functions such as perception, decision-making, and attention [2], [3], [4]. Studies across multiple domains have reliably demonstrated that pain “demands attention” [5], [6], [7], [8] and can negatively impact performance on tasks that require attentional control [9]. Accordingly, a growing body of research has examined the interface between driving and pain experience [3], [4], [10], [11], [12], [13]. In general, both lab [4], [, 14] and field studies [13] have found deficits in driving performance across such conditions as fibromyalgia [11], [, 15], whiplash [3], [4], [16], and other chronic non-malignant pain conditions [13], [, 14]. Collectively, these findings suggest that the experience of driving in pain is a relatively common phenomenon that deserves further empirical scrutiny.

Low back pain is a leading cause of pain and disability in the United States [17], [, 18], as well as one of the most common reasons for work days lost [19], [, 20]. Individuals with chronic low back pain (CLBP) report significant impairment across domains of daily living, including family, social, and workplace function [21], [22], [23], as well as significant psychological distress [24], [25], [26], [27], [28]. Studies find that over 60% of long haul delivery drivers [29], taxi drivers [30], and bus drivers [12] experience CLBP. While these studies focus on the physical/musculoskeletal repercussions of long driving hours [12], [,], to date, very few studies have examined the potential impact of CLBP on driving experience, including that of non-professional drivers. For example, Hu et al. [33], used epidemiological data from Department of Motor Vehicles (DMV) records to identify back pain as a potential crash risk factor in older women (but not men). Another study reported that back pain was significantly associated with a collision history in truck drivers, which the authors attributed to potential distraction stemming from pain as well as a limited range of movement [34].

In summary, despite the ubiquity of both driving and CLBP, almost no studies to date have characterized the experience of driving with CLBP. Further, no studies have examined the association between driving with pain and key cognitive-psychosocial predictors of pain experience. In particular, pain catastrophizing, defined as an exaggerated negative orientation toward pain [35], has been associated with worse physical and psychological outcomes in CLBP [36], as well as with greater difficulties in performing cognitive tasks while in pain [37]. Of particular interest to the current study, catastrophizing has been associated with attentional capture by pain [7], [, 38] and most recently with cognitive intrusion by pain [5], [7], [38]; the latter was assessed by the Cognitive Intrusion of Pain Scale (ECIP) [5], developed to measure cognitive/attentional interruption by an endogenous or exogenous pain stimulus.

The current study represents a preliminary cross-sectional investigation of self-reported driving experience in a sample of individuals with CLBP, focusing on the experience of pain, affective response (specifically fear, anxiety, and irritation), and self-reported driving behavior and outcomes. The current descriptive findings are intended to serve as a foundation for more future, causally oriented research. We predicted that pain experience, affective responses, pain catastrophizing and the cognitive intrusion by pain would have positive associations not only amongst each other, but also with risky driving behaviors and poor driving outcomes.

Materials and methods

Results

Discussion

To our knowledge, the current study is the first to characterize driving experience specifically among individuals with CLBP, with attention to the relationship among key sensory, affective, and cognitive psychological metrics as well as self-reported driving history and behavior. Broadly, findings suggest that drivers with CLBP experience a wide range of somatic and affective responses. Indeed, over half of our participants reported experiencing pain when they drive, and the majority reported driving to be a source of increased pain intensity. Further, participants reported that pain while driving resulted in substantial irritation at both passengers and other drivers on the road. In line with studies of other pain conditions (e.g., whiplash, fibromyalgia [3], [11], [15], [16]), these data suggest that driving may be a significant source of pain and distress among many individuals with CLBP.

Surprisingly, the current study is the first to assess participants’ affective response to driving in pain, and the first to address anger/irritation in the context of pain and driving. The relationship between anger and pain has been well-documented in the literature [55], [56], [57], [58], [59], [60], as has the relationship between anger and driving [61], [62], [63], [64], [65]. In the context of pain and disability, studies link anger with negative physical, social, and functional outcomes [58], [60], [66]. Studies also reliably implicate anger and irritation in negative driving outcomes, including more fines, traffic violations, accidents, and aggressive driving behaviors [63], [, 65]. In partial support of the above associations, CLBP participants who reported a recent history of collisions while driving likewise reported higher irritation with other passengers and other drivers when driving in pain (note: before controlling for CLBP duration). It is important here to note that pain severity did not relate to the number of collisions; rather, the data points to the importance of psychological factors while driving with pain. While participants did not generally report substantial levels of fear/anxiety when driving in pain, suggesting overall confidence on the road, a substantial portion of the sample reported avoiding driving or letting others drive when in pain.

Of particular interest in the current study were associations observed between participants’ ECIP-Driving and PCS scores with pain, affective, and driving variables. In the current study, the ECIP was used to assess the extent to which pain interrupts or dominates participants’ attention when they drive [5]. Echoing previous findings [6], [, 38] higher scores of attentional capture by pain and pain catastrophizing were consistently associated with elevated pain and negative affective responses in the driving context. Further, elevated ECIP-Driving and PCS scores were associated with more self-reported unsafe driving behaviors (driving errors and violations) and greater likelihood of a having had a collision; this effect remained even after adjusting for CLBP duration. These findings should be considered in the context of an established body of evidence that pain (including back pain) contributes to impaired performance on a variety of attentional tasks [67], [68], [69], [70], [71], [72], and that self-reported cognitive intrusion by pain predicts worse performance on tasks designed to mimic real world challenges [6], [, 7]. Further, pain catastrophizing is generally associated with greater pain hypervigilance and difficulty disengaging from pain-related stimuli [37].

To date, very few studies have examined attentional function in the context of driving with pain. Studies of drivers with whiplash disorders found self-report of reduced concentration/ attention when driving [3], [, 16]. Three studies failed to find laboratory-assessed differences in attention-related functions between healthy controls and participants with pain, including mixed pain [13], fibromyalgia [15], and whiplash [4] samples. At the same time, all studies identified performative driving deficits (e.g., greater weaving, worse coordination) among participants with pain [4], [13], [15]. While the current study did not compare participants to healthy controls, our results suggest that, in addition to standard tests of attention, it would be useful to assess the relative attentional toll of pain among drivers with painful conditions, and that this may provide a more nuanced understanding of driving-related findings. Further, although chronic pain patients have been shown to perform poorly on a range of neurocognitive tasks, these have generally been tasks performed for brief periods of time and which are novel to participants. Driving by comparison is a highly automated process which may therefore show differing relationships between pain and cognitive processing.

It is important to note that participants with CLBP in our sample did not report higher frequency of crashes compared to prior work with comparable samples. Although prior studies have identified driving performance deficits in specific pain populations [4], [, 13], and have even drawn association between back pain and vehicle collisions [34], the current findings do not suggest that CLBP is a risk factor for poor driving outcomes. Rather, our study is the first to link common psychological phenotypes among individuals with CLBP [34] with a number of problematic driving-related outcomes. Given the prevalence of CLBP in the general population [17], [, 18] and in professional drivers [12], [29], [30], the current study provides a foundation for efforts to better understand the mechanisms involved in driving behaviors of those with CLBP as well as the relationship between key cognitive-affective, attentional, and behavioral factors identified in the current study.

While research examining the direct effects of pain on driving performance is currently quite limited, considerable research has examined the effects of analgesic medication, in particular opioids, on driving outcomes [2], [73], [74], [75], [76], [77], [78], [79]. In addition to cognitive and attentional impairments [79], opioids are also related to various driving outcomes, such as collisions [80], [81], [82], longer reaction times [83], and trouble maintaining attention behind the wheel [79]. Although medication use was not the focus of the current study, sample responses to items related to medication use while driving are worthy of note. Approximately half the sample reported some opioid use and about half reported taking pain medication while driving. More frequent use of medication while driving showed some association with pain intensity, negative affective responses, and attentional capture by pain, as well as more driving errors and violations. Additionally, when controlling for opioid use and pain duration, taking medication while driving was no longer significantly associated with irritability and anxiety while driving in pain as well as letting others drive when in pain. Given extensive inquiry regarding risky/illegal driving behavior, the current study chose not to ask specifically regarding opiate use while driving, choosing to remain more general with respect to medication use. However, given that associations between medication use while driving and some affective responses were no longer statistically significant, it is possible that this change was observed because opiates were the medication of choice taken while driving. This question remains pertinent for further study. Further, no causal claims or direction can be asserted regarding medication use and the above outcomes. Rather, combined with self-imposed driving restrictions when in pain (above), our findings point to potentially varied attitudes toward driving and medication use among individuals with CLBP and should be explored in future research.

A clear strength of the current study is that we were able to recruit a representative group of individuals with CLBP which allowed us to examine broad and basic questions with respect to pain during driving. However, the online nature of the data collection also represents a potential limitation, particularly as responses were self-report, with no objective assessment. As a result, responses are subject to potential negative biases [84] and socially desirable responding patterns [85], especially when discussing the possibility of poor driving performance. For example, while DBQ items provided instances of inappropriate driving actions, it is unclear whether participants recognized these as inappropriate or risky (e.g., texting and driving), or whether they even engaged in them. Another possible limitation is the overlapping time windows for study measures, which variably assessed pain-related responses during driving, in general, and in the span of the past week, thus potentially affecting observed correlations. Further, we cannot verify whether all participants reported back pain prior to collision, which was assessed over the past three years, and thus report supplementary results which limit the sample to individuals with at minimum three years of back pain. While overall relationships between variables do not change, observed loss of statistical significance (i.e., PCS and ECIP scores between individuals with and without collision history) suggest either a lack of power or key role of early injury in the relationship between collision and back pain. Further analysis revealed that PCS and ECIP scores in the collision group were substantially higher for those who had CLBP for less than three years than those who had experienced CLBP for more than three years, possibly suggesting particular relevance of these constructs for those with more recent onset of pain.

Although this study laid groundwork for research considering the role of psychological pain variables in driving performance, many of the results and questions posed in this study, including those suggested above, can more rigorously be addressed within an experimental context. For example, studies that actively compare the driving behavior between pain and non-pain groups, controlling strictly for duration of pain, are imperative to gaining a better grasp of the relationships suggested in this study. Possible avenues for objectively measuring participant driving behaviors have been discussed in the literature, with the main two methods being the use of an advanced driving simulator to monitor attentional control and responses to stressful events [2], [4], [14] or use of real world driving tests to measure the amount of vehicle swerving while on the road [2], [13], [78]. Such studies can objectively capture participant attentional and behavioral performance without the potential bias of subject interpretation or reporting bias to self-report questionnaires. Potentially, such experimental work can serve to validate or challenge paper-based responses.

Although causal interpretations cannot be drawn, there are clear clinical implications of the current descriptive findings. As noted above, driving is a critical activity for daily living, be it for attending work, maintaining social relationships, or attending medical appointments. As individuals in pain may have mobility issues which can impair walking or other modes of transport, driving may be even more crucial. The current findings reinforce multiple associations between pain and cognitive-affective variables that have been observed in literature outside the driving context, including pain intensity, anger, inattention, and behavioral disruption. Given that driving is a pervasive, potentially risky behavior that requires some form of cognitive focus and control, the current findings point to a continued need to examine these associations within this specific life context. As part of pain management approaches that facilitate high quality of life and participation, it may be important to consider strategies that address/mitigate pain and pain-related psychological responses behind the wheel, with potentially powerful implications for driving safety.

Conclusion

In conclusion, individuals with CLBP reported significant pain and distress during driving activity, suggesting that the experience of driving with back pain deserves more study and perhaps clinical consideration. Further, our findings suggest that key pain psychological variables, including pain catastrophizing and cognitive/attentional intrusion by pain while driving may be associated with negative emotional responses and problematic driving behavior. The current study is intended to provide a foundation for further scrutiny of this important subject.