Chapter 13. Mental Health and Driving
Joanne E. Taylor
Massey University, Palmerston North, New Zealand
Since the 1940s, researchers in traffic psychology have been interested in the way that mental health affects driving behavior and performance. Early studies investigated various aspects of psychopathology that might contribute to drivers who were considered to be “accident prone” in an attempt to explain their increased motor vehicle crash tendencies. Since then, research in this area has grown exponentially, and literature on alcohol- and substance-related problems, personality traits and disorders, anger, depression, anxiety, attention-deficit/hyperactivity disorder, and stress is reviewed in this chapter. Furthermore, the large amount of research on the mental health effects of driving incidents and motor vehicle collisions is outlined, including post-traumatic stress reactions, driving-related fear, phobia, and travel anxiety. Although these two fields have remained quite separate historically, considering them both together helps to provide a more comprehensive overview of the role of mental health in the driving environment.

1. Mental Health Impacts

2. The Effects of Mental Health on Driving

Driving is a highly complex process. As information processors in the driving system, drivers must constantly receive, process, and respond to information derived from a constantly changing environment as well as modulate their internal states. Therefore, they require efficient cognitive and intrapersonal function. Several factors can influence the efficiency of cognitive function and in different ways, which may place drivers and other road users at increased risk of involvement in an MVC. In an attempt to understand the human error causes of MVCs, researchers have studied an exhaustive array of human factors, including mood, aggression, risk-taking behavior, fatigue, stress, age, gender, brain injury, drug-taking behavior, and psychiatric symptoms (Little, 1970, McDonald and Davey, 1996, Shinar, 1978 and Taylor and Dorn, 2006). Some of these factors are considered in detail elsewhere in this volume (e.g., see Chapter 12 on social, personality, and affective constructs in driving; Chapter 17 on impaired driving; and Chapter 21 on fatigued driving), so the present material focuses on the relationships between mental health and driving. Much of the research in this area has retrospectively examined the prevalence of psychopathology in MVC victims or conducted laboratory-based investigations of driving with groups identified on the basis of particular mental health characteristics, examining factors that may impair and affect driving ability, such as attention, concentration, memory, vigilance, impulse control, judgment, problem solving, reaction time, and psychomotor control.
It has long been suggested that mental health might contribute to road safety, and specifically that those who experience mental health problems, such as psychotic, mood, anxiety, or substance use difficulties, are more likely to be involved in MVCs. Several reviews of this area have been published, although they are somewhat dated (McDonald and Davey, 1996, Metzner et al., 1993, Noyes, 1985, Silverstone, 1988 and Tsuang et al., 1985). Early studies examined what were then called “accident-prone” drivers and reported on their socially deviant characteristics (Tillman & Hobbs, 1949), as well as noting high levels of alcoholism in this group (Selling, 1940). Crancer and Quiring (1969) found that people with personality disorders had an accident rate 144% higher than that of a matched control group, whereas it was 49% higher in a group with psychoneurotic disorders, and there was no increased rate compared with controls for a group with schizophrenia.
Since these early studies, other researchers have reported higher traffic accident rates among people with alcohol use and personality disorders, particularly antisocial personality disorder (Armstrong and Whitlock, 1980, Dumais et al., 2005, Elkema et al., 1970, Selzer et al., 1967 and Waller and Turkel, 1966; see also the review by Tsuang et al., 1985), as well as higher mortality rates in MVCs for these groups (Rorsman et al., 1982 and Schuckit and Gunderson, 1977). Others have found no such increased accident likelihood for those with psychiatric histories, although these studies are limited by methodological issues such as excluding people with substance-related problems (Cushman et al., 1990 and Kastrup et al., 1977). Laboratory- or field-based studies have reported mixed findings, such as slowed driving speed and more errors and collisions on a simulator for a sample with schizophrenia compared with a matched control group (St. Germain, Kurtz, Pearlson, & Astur, 2004) and marked problems with psychomotor performance in a sample of psychiatric outpatients (de la Cuevas Castresana & Alvarez, 2009).
However, the relationship between mental health and driving is complex, and simply examining the differential rates of MVC in those with or without various forms of psychopathology does not provide evidence that psychopathology plays a causal role in accidents. Accidents are often preceded by stressful life events, such as problems in interpersonal relationships, as well as driving under the influence of alcohol and other substances (Noyes, 1985). Various mechanisms have been proposed to explain the relationship between various types of mental health difficulty and MVCs. McDonald and Davey (1996) provide a detailed review of these factors, which are briefly outlined here.

2.1. Alcohol and Substance Use Problems

The physiological, cognitive, and behavioral effects of alcohol, such as slowed reaction time, problems with coordination and attention, and lowered behavioral inhibition, clearly increase MVC risk, and this risk is greater for those who have a pathological problem with alcohol, such as alcohol abuse or dependence (which may also be comorbid with other psychopathology, such as antisocial personality and conduct disorder traits, depression, and post-traumatic stress disorder; del Rio and Alvarez, 2001, del Rio et al., 2001, McDonald and Davey, 1996, McMillan et al., 2008 and Stoduto et al., 2008). Several epidemiological studies have examined psychopathology in those convicted of driving under the influence of alcohol and have reported lifetime rates of alcohol use disorder of 41–91% and other drug use disorder of 26–40% (de Baca et al., 2004, Lapham et al., 2006, Lapham et al., 2001, McCutcheon et al., 2009, Palmer et al., 2007 and Shaffer et al., 2007). Psychotropic medication use is also relevant for drivers with mental health problems, although most of these medications are generally considered not to interfere with driving performance unless alcohol is also taken or the drugs are being abused (Hole, 2007 and Hole, 2008).

2.2. Personality Traits and Disorders

In some cases, such as alcohol abuse, the symptoms of the problem can directly influence vehicle control and therefore contribute to risky or unsafe driving. In others, however, such as personality disorders, the link between the mental health problem and driving behavior is less clear, and is probably influenced by a combination of factors, including personality traits such as aggression, hostility, impulsivity, and sensation-seeking. These latter traits might have indirect effects on MVCs through driving behaviors such as errors and violations (Donovan and Marlatt, 1982, Shaw and Sichel, 1971, Wilson and Jonah, 1988 and Zuckerman and Neeb, 1980).
Several studies have attempted to predict accident involvement from a variety of personality factors and driving behaviors (Kim et al., 1995, Norris et al., 2000, Rimmö and Åberg, 1999, Ulleberg and Rundmo, 2003 and West et al., 1993). Sümer (2003) developed a fairly sophisticated contextual mediated model of personality and behavioral predictors of MVCs that distinguished distal and proximal factors in 295 professional drivers in Turkey. Psychological symptoms, including anxiety, depression, hostility, and psychoticism, had direct effects on aberrant driving behaviors, and dysfunctional drinking had an indirect effect (via aberrant driving behaviors) on the number of accidents (Sümer, 2003). Other researchers have also supported the use of multiple personality predictors of unsafe driving that are related in different ways to different aspects of driving behavior (Dahlen & White, 2006).

2.3. Anger

In addition to being investigated in research on personality traits and driving, anger in the driving situation has increasingly been studied during approximately the past decade, fuelled by interest in the phenomenon of road rage (Galovski et al., 2006 and Hole, 2007). Research has focused on developing instruments to measure driving anger that can distinguish aggressive driving behavior and thoughts as well as risky, nonaggressive behavior (Deffenbacher et al., 2002, Deffenbacher et al., 1994, Deffenbacher et al., 2004, DePasquale et al., 2001 and Lajunen et al., 1998). Others have explored associations of driving anger and aggression with psychopathology and have found that aggressive drivers are more likely to meet criteria for various mental health problems, particularly intermittent explosive disorder, current or past alcohol or substance abuse or dependence, antisocial and borderline personality disorders, conduct disorder, and attention-deficit/hyperactivity disorder (Fong et al., 2001, Galovski et al., 2002 and Malta et al., 2005). Victims of road rage are also at risk of developing mental health problems (Smart, Ashbridge, Mann, & Adlaf, 2003).
Despite the tendency for angry and aggressive drivers to report more risky driving behaviors, most research has found no significant correlations between measures of driving anger and accident involvement (Sullman, 2006 and Van Rooy et al., 2006), although a few have found overall anger, in addition to other variables, to significantly predict crash involvement (Sullman, Gras, Cunill, Planes, & Font-Mayolas, 2007). Driving anger and aggression is generally considered to be a complex problem that depends on the characteristics of the driver and the situation (Lajunen and Parker, 2001 and Shinar, 1998).

2.4. Depression

Rates of depression in accident samples have not been clearly determined given the overlap in some cases of low mood and depression with self-harm and suicide (McDonald & Davey, 1996) and the use of self-report questionnaires instead of diagnostic methods to determine depression (Hilton, Staddon, Sheridan, & Whiteford, 2009). Other research on depression and driving has taken place in the broader context of evaluating the influence of emotions on driving, which may be related to various factors, including prejourney emotions and circumstances, the traffic situation while driving, and thoughts that arise during travel (Bañuls Egeda et al., 1997, Groeger, 1997 and Levelt, 2003). For example, Mesken, Hagenzieker, Rothengatter, and de Waard (2007) found that emotions while driving were related to emotional traits as well as traffic events such that anger was associated with other-blame and events affecting progress, and anxiety was associated with situation-blame and events affecting safety.

2.5. Anxiety

Discussions in the general driving literature that have related anxiety to driving have come from broader studies of personality typologies and disorders (Evans, 1991, Foot and Chapman, 1982, Heimstra et al., 1967, Little, 1970, Loo, 1979, Shinar, 1978, Shoham et al., 1984, Silverstone, 1988 and Wilson and Greensmith, 1983), and stress (Gulian et al., 1988, Gulian et al., 1990, Heimstra, 1970 and Hentschel et al., 1993). Some research suggests that anxiety necessarily impairs driving. Shoham et al. (1984) used a combination of personality variables to predict the likelihood of recidivist traffic accident involvement, and they reported that drivers characterized as anxious manifested high internalization of traffic norms and high levels of anxiety and “were found to have lowered bio-psychogenic [sic] control over the basic mechanisms required for driving” (p. 184). Other authors have considered that anxiety affects driving in a more complex manner and may also have some facilitative or positive effects that are specific to driver behavior and driving skills (Kottenhoff, 1961, O’Hanlon et al., 1995, Payne and Corley, 1994 and Silverstone, 1988). For example, a moderate amount of anxiety may enable the driver to carry out all of the basic skills required for driving, as well as to pay sufficient attention to potential hazards so that the appropriate action can be taken if required (Walklin, 1993), whereas high levels of anxiety could interfere with driving performance and increase the risk of an MVC through errors, indecision, and hesitation (Carbonell et al., 1997, Silverstone, 1988 and Walklin, 1993). Yinon and Levian (1988) found that anxiety about being in the presence of other drivers leads to the division of attention between self- and task-relevant stimuli, although the focus on threat appraisals in anxiety disorders may actually explain why the MVC rate is no higher than population norms for this group (McDonald & Davey, 1996). Indeed, Taylor, Deane, and Podd (2007) found that anxious drivers made more errors than controls in an on-road driving test, but there were no differences in MVC history.

2.6. Attention-Deficit/Hyperactivity Disorder

During approximately the past decade, researchers have started to investigate the influence of adult attention-deficit/hyperactivity disorder (ADHD) on driving behavior, particularly because of the features of difficulties with attention and impulsivity that characterize this condition. ADHD has been found to present risks to safe driving in terms of traffic violations, license suspensions, less safe driving practices, more driving errors, and crashes in simulator performance as well as on the road (Fischer et al., 2007 and Nada-Raja et al., 1997). These risks may be partly accounted for by the fact that ADHD can also be associated with other risk factors, such as tendencies to frustration and aggression (Richards, Deffenbacher, Rosén, Barkley, & Rodricks, 2006), as well as substance use and conduct problems (Jerome, Segal, & Habinski, 2006). Treatment with long-acting methylphenidate improves the driving performance of adolescents and adults with ADHD (Barkley, Murphy, O’Connell, & Connor, 2005), although it is unknown whether this also reduces their risk of MVCs or traffic violations (Barkley, 2010).

2.7. Stress

Although much research has examined the relationship between various types of mental health difficulty and driving performance as well as MVCs, another strand of research has investigated the role of stress as a more general problem linked to mental health that might make people with (or without) psychopathology more vulnerable to MVCs. One difficulty in this research is that stress can act both as a cause and as a consequence of mental health problems, so studies that do not control for psychopathology are limited (McDonald & Davey, 1996).
Matthews (2001) attempted to identify the information-processing functions that mediate the effects of stress on driving performance impairment in developing a transactional model of driver stress (Matthews et al., 1998). Stress variables used have been based on factor analyses of the Driving Behaviour Inventory (Glendon et al., 1993 and Gulian et al., 1989) and its revision, the Driver Stress Inventory (Matthews, Desmond, Joyner, Carcary, & Gilliland, 1997), which are considered to represent vulnerabilities to different types of stress outcome, including aggression and anxiety. Effects of driver stress on driving performance can depend on the nature of the driver’s stress reactions (e.g., appraisal of the demands of the traffic environment, including other drivers; appraisal of personal competence; and coping strategy), the traffic environment, and the demands of the driving task (Matthews, 2001, Matthews et al., 2005, Matthews et al., 1997, Matthews et al., 1998 and Matthews et al., 1999). The approach used to cope with stress has been identified as an important factor influencing the perception of stress and whether it affects driving behavior, and maladaptive coping strategies such as alcohol abuse may increase accident risk (McDonald & Davey, 1996).

3. Effects of Driving on Mental Health

Various aspects of driving can also impact mental health. Most notably, involvement in MVCs can have varied mental health effects, ranging from little or no impact to significant and marked difficulties. Several books have been dedicated to this topic (Blanchard and Hickling, 2004, Duckworth et al., 2008, Hickling and Blanchard, 1999 and Mitchell, 1997). It is also important to note that, in addition to those directly involved and injured in the crash as primary victims, MVCs can also affect (1) those who are uninjured but are still involved in the crash as either participants or witnesses; (2) friends and family who hear descriptions of the event from someone involved; and (3) those involved in dealing with the after effects of MVCs, including police, fire, ambulance, and hospital personnel, as well as those who are charged with preparing medical or legal reports (Mayou, 1997, Mitchell, 1999 and Taylor and Koch, 1995).

3.1. Mental Health Consequences of Motor Vehicle Crashes

The nature and sheer frequency of MVCs suggests that at least some people are likely to suffer psychological repercussions. Research has demonstrated that people who have been involved in MVCs and other common accidents may manifest chronic psychological dysfunction, even in the context of minimal physical injury and good recovery (Horne, 1993; Pilowsky, 1985). Longitudinal research has demonstrated a variety of psychosocial sequelae in injured MVC survivors, including worsened family or spouse relationships as well as reduced social contact, pleasure from leisure activities, and work capacity (Malt, Hϕivik, & Blikra, 1993). Research on the psychiatric consequences of MVCs has documented widespread implications for psychological functioning, including depressive, anxious, and phobic symptoms, as well as multiple disturbances. However, some of these studies have combined MVC victims with victims from a range of industrial and work-related accidents, making it impossible to clearly determine psychiatric morbidity after MVCs (Culpan and Taylor, 1973, Jones and Riley, 1987 and Shalev et al., 1998).
Post-MVC research has ranged from examining specific effects such as depersonalization responses (Noyes, Hoenk, Kuperman, & Slymen, 1977) to comprehensive studies of MVC-induced psychopathology. The most common psychological sequelae of MVCs include driving-related fears and avoidance, post-traumatic stress disorder (PTSD), depression, and pain-related syndromes (Blanchard et al., 1994, Goldberg and Gara, 1990, Koch and Taylor, 1995, Kuch et al., 1994, Malt, 1988, Mayou, 1992 and Mayou et al., 1993). For victims with multiple injuries, depression and anxiety are particularly common (Mayou et al., 1993). Approximately 40% of MVC victims suffer comorbid conditions, such as major depression, panic disorder, specific phobia, eating disorder, substance abuse, and personality disorder (Blanchard et al., 1994). Furthermore, blaming others for the crash has been found to be associated with higher levels of psychological distress and lower psychological well-being for both passengers and drivers (Ho, Davidson, Van Dyke, & Agar-Wilson, 2000). However, consistent prevalence data for the various psychological outcomes have been elusive, likely due to the impact of several methodological issues, such as varying definitions of terms, the use of samples that are not generalizable to MVC victims (e.g., medicolegal samples, victims referred for treatment, and hospital or primary care attendees), and lack of consideration of injury severity (Blaszczynski et al., 1998).
Significant psychological problems have been identified even among victims of minor traffic crashes. For example, one longitudinal study reported that 4 months after a minor MVC (i.e., outpatient treatment only, with no hospitalization or head injury), 13% of 39 participants scored above the cutoff for a positive PTSD diagnosis on a self-report measure, 36% reported symptoms of anxiety, and 16% described avoiding using their car, motorcycle, or bicycle (Smith, MacKenzie-Ross, & Scragg, 2007).

3.1.1. Post-Traumatic Stress Reactions

The most commonly studied psychological outcomes of MVCs are those involving some kind of trauma response, either in the immediate weeks following the crash or in the longer term, and that may range from subthreshold-level symptoms to full-blown clinical syndromes, such as acute and post-traumatic stress disorder. Both are characterized by problematic experiences of anxiety that can occur following a traumatic event involving “actual or threatened death or serious injury, or a threat to the physical integrity of self or others” and in which the person responds to the event with horror, fear, or helplessness (American Psychiatric Association, 2000, pp. 467 and 471). The symptoms of these trauma responses can include psychologically re-experiencing the trauma (e.g., intrusive thoughts and nightmares), increased physical arousal (e.g., exaggerated startle response and irritability), and persistent avoidance related to the crash (e.g., avoidance of or reluctance to drive and avoiding thoughts or conversations about the crash).
A body of research exists that indicates the frequent occurrence of PTSD in MVC victims and its impact on quality of life (Gudmundsdottir, Beck, Coffey, Miller, & Palyo, 2004), and it has been suggested that PTSD thoroughly captures the psychological consequences of MVCs (Burstein, 1989b, Davis and Breslau, 1994, Hickling, blanchard,silverman, et al., 1992, Kuch et al., 1985 and Platt and Husband, 1987). Studies have investigated the treatment of post-traumatic responses after MVCs (Blanchard, Hickling,Taylor, et al., 1996, Brom et al., 1993, Fairbank et al., 1981, Green et al., 1993, McCaffrey and Fairbank, 1985 and Walker, 1981) as well as the complicating nature of PTSD in post-traumatic headache (Davis and Breslau, 1994, Hickling et al., 1992 and Hickling, blanchard,silverman, et al., 1992) and the nature of psychophysiological responding in MVC-related PTSD (Blanchard, Hickling, & Taylor, 1991). However, it has been noted that avoidant symptoms may obscure the identification of PTSD reactions in MVC victims, and PTSD may go unrecognized for some time after the accident (Burstein, 1989a, Burstein, 1989b and Epstein, 1993).
Reported incidence rates of MVC-related PTSD vary considerably across studies, largely due to methodological differences, especially in terms of the sample included and approach used to ascertain PTSD. Table 13.1 provides an overview of studies that have examined adult MVC samples using well-validated structured diagnostic interviews or self-report measures that represent diagnostic criteria. Investigations have differentiated between PTSD among respondents involved in “serious” MVCs, in which there was some degree of physical injury requiring hospitalization, and “non-serious” MVCs, or those not resulting in bodily injury or involving the subjective experience of psychological injury. However, because victims of both serious and non-serious MVCs have been found to experience PTSD, these distinctions may have little utility. Table 13.1 demonstrates this phenomenon, whereby research using seriously injured victims has found a range of rates of PTSD (from 1 to 100%), as have studies that have utilized victims sustaining relatively minor injuries (15–50%). Although there appears to be a larger range for serious MVCs, the overlap in incidence rates is substantial and may be due to definitional (e.g., different criteria for severity of injury, diagnosis, and time since the MVC) and methodological differences (e.g., whether the sample was seeking treatment or not).
TABLE 13.1 Prevalence of PTSD in Adults Following MVCs Using DSM-Based Assessment
NR, not reported.
aDSM-III.
bDSM-III-R.
cDSM-IV.
ReferencesNInjury Severity Criteria% PTSDTime since MVC
Serious Injury
Kuch et al. (1985)30Medical attention sought100aNR
Malt (1988)107Hospitalized16 months
Feinstein and Dolan (1991)48Accidentally injured25
15		6 weeks
6 months
Hickling, Blanchard, Silverman, et al. (1992)20Post-traumatic headache75bNR
Epstein (1993)15Serious injury40bNR
Green et al. (1993)24Severe injury25b18 months
Mayou et al. (1993)188Multiple injury or whiplash neck injury7–9
5–11		3 months
12 months
Blanchard et al. (1994)50Medical attention sought461–4 months
Blanchard, Hickling, Barton, et al., 1996 and Blanchard, Hickling,Taylor, et al., 1996158Medical attention sought39b
12b		1–4 months
12 months
Ehlers et al. (1998)967Emergency department attendees23.1c3 months
Mayou et al. (2002)16.5c
11c		12 months
3 years
Harvey and Bryant (1998)71Hospitalized25.4c6 months
Chubb and Bisson (1999)24Many physically injured56.3c
37.5c		5 weeks
9 months
Koren et al. (1999)74Hospitalized32b12 months
Ursano et al. (1999)122Most hospitalized34.4b
25.3b
18.2b
17.4b
14b		1 month
3 months
6 months
9 months
12 months
Bryant et al. (2000)113Hospitalized21c6 months
Hamanaka et al. (2006)100Severe injuries8.5c6 months
Matsuoka et al. (2008)100In intensive care8c1 month
Yaşan et al. (2009)95Emergency department attendees29.8c
23.1c
17.9c		3 months
6 months
12 months
Non-Serious Injury
Goldberg and Gara (1990)55Not resulting in death or major bodily injury15M=15 months
Kuch et al. (1994)21Minimal injury and chronic pain38bNR
Kupchik et al. (2007)60General health outpatient50cM=44 months
Injury Criteria Not Given
Hickling and Blanchard (1992)20NR50bNR
Horne (1993)7NR43aM=2 years
Dalal and Harrison (1993)86NR32M=2.7 years
Kuch, Cox, and Direnfeld (1995)54NR22M=3.6 years
Chan, Air, and McFarlane (2003)391NR29c9 months
The various studies shown in Table 13.1 indicate that the rate of PTSD for seriously injured victims 1 month following an MVC is approximately 25–56% (Blanchard et al., 1994; Blanchard, Hickling, & Barton, 1996; Blanchard, Hickling,Taylor, et al., 1996, Chubb and Bisson, 1999, Feinstein and Dolan, 1991 and Ursano et al., 1999). Rates at 3–6 months decrease to 7–30% (Bryant et al., 2000, Ehlers et al., 1998, Hamanaka et al., 2006, Harvey and Bryant, 1998, Mayou et al., 1993 and Ursano et al., 1999; Yaşan, Güzel, Tamam, & Ozkan, 2009), and they decrease to 5–32% at 12 months (Blanchard, Hickling, Barton, et al., 1996, Blanchard, Hickling,Taylor, et al., 1996, Ehlers et al., 1998, Green et al., 1993, Koren et al., 1999 and Mayou et al., 1993). Increasingly, studies have documented MVC-related PTSD in children, some of which have reported cases with onset as young as 2 years (Jaworowski, 1992, Jones and Peterson, 1993, McDermott and Cvitanovich, 2000 and Thompson and McArdle, 1993). Symptomatology in such cases has included reliving the MVC through nightmares, conduct difficulties, separation anxiety, enuresis, fear of the dark, trauma-specific fears, sleep disturbance, violent play, reluctance to cross roads or travel by car, and a preoccupation with road safety (Canterbury and Yule, 1997, Jones and Peterson, 1993, Taylor and Koch, 1995 and Thompson and McArdle, 1993).
Studies using approaches that permit diagnostic assessment have typically found higher rates of PTSD in children than have studies using other criteria to determine the presence of PTSD, such as cutoff scores on a measure of PTSD-type symptoms (e.g., 12–18% from 3 to 12 months post-MVC; Landolt et al., 2009, Landolt et al., 2005 and Sturms et al., 2005).
Various predictors of PTSD have been identified in the literature, related to both the early development of the trauma response and whether that response becomes chronic. Predictors of early stage development of PTSD include the presence of acute stress disorder, persistent physical disability, severity of physical injury, a sense of threat to life, dissociation during the crash, low perceptions of coping self-efficacy, and lower perceived social support (Benight et al., 2008, Hamanaka et al., 2006, Koren et al., 1999, Matsuoka et al., 2008 and Yaşan, et al., 2009). Factors that have been shown to predict chronic PTSD (symptoms experienced for 1 year or more) include some of the previously mentioned factors, such as early trauma symptoms (including sleep problems), perceived threat, dissociation during the crash, and persistent health problems, as well as other variables, including hospitalization for injuries, persistent financial problems, litigation, female gender, unemployment, emotional problems prior to the crash (including distress from and amount of prior trauma), alcohol abuse, and poor social support before and after the crash (Ameratunga et al., 2009, Beck et al., 2006, Blanchard, Hickling, Barton, et al., 1996, Buckley et al., 1996, Dörfel et al., 2008, Ehlers et al., 1998, Fujita and Nishida, 2008, Irish et al., 2008, Koren et al., 2002, Mayou and Bryant, 2002, Mayou et al., 2002, Mayou et al., 1997 and Murray et al., 2002).
Several cognitive factors, especially cognitive processes, have also been found to maintain and predict PTSD (in some cases to a greater degree than the established predictors noted previously). These cognitive factors include negative interpretations of intrusive memories, rumination, memory disorganization, thought suppression, anger cognitions, and general negative post-traumatic thoughts (Ehring et al., 2006, Ehring et al., 2008, Ehring et al., 2008, Holeva et al., 2001, Karl et al., 2009 and Murray et al., 2002).
A few studies have examined predictors of PTSD development in children who have experienced MVCs, and early trauma symptom severity has been identified as the strongest predictor (Landolt et al., 2005 and Schäfer et al., 2006), along with severity of MVC-related PTSD in the father. However, variables such as age, gender, injury severity, threat appraisal, prior trauma exposure, prior mental health problems, and maternal MVC-related PTSD have not been found to be significant predictors (Landolt et al., 2005 and Meiser-Stedman et al., 2009). The presence of nightmares with content that exactly matched the trauma has been found to strongly predict PTSD scores at 2 and 6 months post-MVC, although this finding needs to be replicated with a larger sample (Wittman, Zehnder, Schredl, Jenni, & Landolt, 2010). In line with adult research, studies are beginning to document evidence for the role of various cognitive factors in predicting the development and maintenance of PTSD following MVCs in children. For example, research has examined maladaptive cognitive appraisals such as the meaning of the trauma and trauma symptoms, future vulnerability, rumination, anxiety sensitivity, and the quality of trauma memories (Meiser-Stedman et al., 2009). Information about approaches to treatment of PTSD following an MVC is widely available, and readers are referred to the existing literature for more information (Blanchard and Hickling, 2004, Duckworth et al., 2008, Hickling and Blanchard, 1999 and Hickling et al., 2008).

3.1.2. Driving-Related Fear, Phobia, and Travel Anxiety

In addition to post-traumatic reactions, research has also demonstrated that other types of fear reactions to MVCs are common and can be extremely debilitating (Herda et al., 1993, Kuch, 1997, Kuch et al., 1996 and Kuch et al., 1993). Fear of driving is a diverse experience, ranging from mild driving reluctance to driving phobia as a variant of specific phobia. Research on these phobic fear reactions has focused on avoidance of or reduction in driving, endurance of driving with marked discomfort, and the effect of fear on a person’s lifestyle and everyday functioning. However, variations in terminology and definitions used as well as sampling issues have led to vast inconsistencies in prevalence estimates for problems such as driving phobia, accident phobia, travel phobia, driving reluctance, and phobic travel anxiety. In particular, studies that have used broader criteria in which complete avoidance is unnecessary have reported higher rates (e.g., 60–77%; Hickling and Blanchard, 1992 and Kuch et al., 1985) than those that have specified total avoidance for diagnosis (e.g., 2–6%; Blanchard et al., 1994 and Hickling and Blanchard, 1999). Driving fear and phobia can also occur in the absence of MVCs but still be severe and have a marked effect on functioning (Ehlers et al., 1994, Taylor and Deane, 1999, Taylor and Deane, 2000, Taylor et al., 1999 and Taylor et al., 2000).
Driving phobia is most appropriately considered to be a situational type of specific phobia that is characterized by marked and persistent fear that is excessive or unreasonable, is cued by anticipation of or exposure to driving stimuli, is associated with avoidance of driving stimuli or endurance of such stimuli with considerable anxiety or distress, and has a marked impact on the person’s functioning (American Psychiatric Association, 2000). The content of fear can be much broader than the fear of driving and can relate to various aspects of travel and accident-related stimuli, such as fear of riding in a vehicle as a passenger while having no fear of driving (Koch & Taylor, 1995). Blanchard and Hickling (2004) refer to less phobic forms of driving-related fear as driving reluctance, where the person is able to make essential journeys but avoids nonessential travel or tolerates it with some degree of anxiety. Several reviews of driving-related fear and phobia provide more comprehensive information on this topic, along with information on appropriate interventions (Taylor, 2008 and Taylor et al., 2002).

3.1.3. Other Problems

Several studies have documented increased rates of depression and other mood disorders following MVCs that may or may not be comorbid with PTSD, with rates of major depression ranging from 6 to 53% 1 year post-MVC (Blanchard et al., 2004, Blanchard et al., 1995, Dickov et al., 2009 and Ehlers et al., 1998). Blanchard, Hickling, Taylor, et al. (1996) also reported that major depression prior to the MVC was a significant predictor of post-MVC PTSD. Depression has been identified as a common consequence of MVCs and one that can overlap with physical effects such as chronic pain and head injury and contribute markedly to functional limitations following a crash (Duckworth, 2008). Substance use disorders have been examined in relation to MVCs, although the mixed findings reported in the literature suggest that longitudinal research is needed to more clearly identify the relationship between substance use and MVC (O’Donnell, Creamer, & Ludwig, 2008). Psychological factors can also contribute in many ways to pain-related syndromes that occur following MVC-related injury (Duckworth et al., 2008).

4. Summary

The relationship between mental health and driving is complex. Mental health can have an impact on driving behavior and performance, although the relationships are multifaceted and depend on factors related to the specific nature of the problem, other characteristics of the individual, and specific aspects of the traffic environment and driving situation. However, higher accident rates for those with alcohol-related disorders as well as antisocial personality disorder are relatively consistent findings, although these conditions are also frequently comorbid. The absence of clear information from methodologically sound studies about how mental health affects safe driving can present difficulties for health professionals who are required to make decisions regarding fitness to drive in cases in which mental health is an issue (Knapp and VandeCreek, 2009 and Ménard et al., 2006). The complexity of the relationships involved necessitates that current guidelines focus on the importance of individualized assessment and consideration of factors such as acute illness symptoms as well as side effects of, interactions among, and compliance with medication (Carr et al., 2010 and Land Transport Safety Authority, 2002). Several quite different types of mental health problems have tended to emerge from research on the psychological consequences of MVCs and highlight the diverse and complex types of experiences that might also be influenced by pre-accident mental health as well as the specific nature of the incident and response characteristics. Although the two fields of traffic psychology and mental health and driving have historically been considered separately, considering them both provides a more comprehensive overview of the role of mental health in driving, particularly in highlighting the ways that mental health might influence driving and be affected by the driving environment.