Aggressive/hostile personality traits and injury accidents: an eight-year prospective study of a large cohort of French employees -- the GAZEL cohort (2024)

Abstract

Keywords: Accidents, Traffic; statistics & numerical data; Adult; Aggression; psychology; Automobile Driving; statistics & numerical data; Cohort Studies; Employment; statistics & numerical data; Female; Follow-Up Studies; France; epidemiology; Hostility; Humans; Incidence; Male; Middle Aged; Personality Disorders; epidemiology; psychology; Prospective Studies; Questionnaires; Time Factors; Wounds and Injuries; epidemiology

Keywords: aggression / hostility, BDHI, cohort study, negative binomial regression, injury accident.

The phenomenon of “road rage” has in recent years emerged as a new traffic safety issue in some English-speaking countries, including the United States (US), Canada, the United Kingdom (UK) (Harding et al.1998; ; Joint, 1995; Wells-Parker et al. 2002). Despite a lack of consensus, “road rage” was described to be associated with a range of anti-social driving behaviors and/or acts of aggression which occur on the road (Elliot, 1999). According to the US National Highway Traffic Safety Administration (NHTSA), substantial numbers of American drivers report that they have experienced hostile, unsafe or aggressive acts on the part of other drivers (NHTSA, 1999). A survey conducted in the UK (Joint, 1995) in 1995 revealed that 90% of 526 motorists sampled reported having experienced in the past year aggressive acts such as tailgating (close following), and 48% of them reported being a victim of obscene gestures in the same period. In turn, 60% admitted that they themselves had lost their temper while driving at least on occasion.

Although these data suggest that aggressiveness and/or anger behind the wheel in drivers are widespread, it has been argued that dispositional factors like personality traits make some individuals more prone to these behaviors than others (Deffenbacher et al. 2003; ; ; McMillen et al. 1992). An aggressive personality trait or trait aggressiveness has been defined as “a general propensity to engage in acts of physical and verbal aggression, a proneness to anger, and a proneness to hold hostile beliefs about other people across situations” (; ; Berkowitz, 1993; Bushman, 1996).

Although a large body of the literature demonstrates that aggressive driving is associated with an increased risk of traffic accidents (; ; ; ; Parry, 1968; Conger et al. 1959; Mizell, 1997), the relationships between aggressiveness as a personality trait, aggressive driving and traffic accidents remain largely unknown. Earlier and recent studies produced inconsistent findings. In the 1974–1988 period, two studies using the Buss-Durkey Hostility Inventory (BDHI) found a significant association between scores of BDHI, crash involvement, violations and driving-related aggression (Donanan & Marlatt, 1982; ). Other studies conducted (Beirness, 1993) in the same period reported that crash rates were associated with the expression of hostile feelings, increased aggression and with high scores of the Driver Behavior Inventory (DBI). However, further studies carried out in the 1989–1998 period failed to find any significant association between traffic-related injuries, minor accidents and hostility, self-rated irritability, ease of anger arousal, and other aggression factors (; Karlberg et al. 1998; Romanov et al. 1994; Novaco, 1989).

It is noteworthy that the majority of these studies were carried out on small samples with a large spectrum of study designs and measurement tools. Moreover, only one cohort study has been published, on over 21, 000 Finnish participants (Romanov et al. 1994). Unfortunately, this latter study suffered from incomplete consideration of confounding factors (no mileage, alcohol consumption as unchanging with time) and from use of unstandardized measures of hostility.

The present study therefore aimed to examine prospectively the extent to which psychometric measures of aggression/hostility were related to injury accidents in a large cohort of French employees.

METHODS

RESULTS

In 1993, 15 049 participants out of 20 624 returned their BDHI questionnaire. We received responses to the 2001 Driving Behavior and Road Safety (DBRS) questionnaire from 14 226 participants out of 19 894. Some were excluded (n=26) because of data discrepancies with the general cohort database. Data on occupational category or mileage were missing for another 376 participants leaving 13 824 participants with a DBRS questionnaire. Among them, 11 754 participants had also returned their BDHI questionnaire and were included in the analyses. In the 1994–2001 period, 548 participants had one I-A, 47 two I-A and 6 three I-A.

DISCUSSION

Overall aggression/hostility-related personality traits as measured by Buss-Durkey Hostility Inventory (BDHI) did not predict injury accidents in this large cohort of French employees. However, a relatively weak but significant association was found between high irritability, high negativism and I-A risk after major confounding factors were controlled for.

These data were consistent with the most recently published studies of this relationship, suggesting that general aggression/hostility was not systematically associated to road traffic accidents (; ; Karlberg et al. 1998, Romanov et al. 1994; Nonaco, 1989). Contrary to previous studies, we were able to assess incident I-A as outcomes and to include in the analysis a wide range of potential confounders. The exhaustiveness of I-A reports was enhanced by the use of reminders from the medical department, which listed sick leaves for traffic accidents during the period covered by the questionnaire. The large sample size of our prospective study rules out the possibility that the absence of associations observed was due to lack of power. More precisely, with an error of first kind (α) of 0.05 and a power (1-β) of 0.80, we calculated that the fraction detectable rate ratio was 1.29.

In accordance with the literature, in the present study younger participants and unskilled workers were more likely to score higher on overall aggression/hostility (Deffenbacher et al., 2003; Deffenbacher et al., 2000; Deffenbacher et al. 2002). Also consistent with previous findings, participants who exceeded road speed limitations scored higher on overall aggression/hostility measures. Deffenbacher et al. (2003) in their experimental study on anger and aggression found that, in “the familiarization and open road simulations”, high anger drivers drove faster and more erratically than low anger drivers. According to Ulleberg (2001), this suggests that emotional factors and probably lack of control over these are related to risky-driving style. Ulleberg (2001) considers that becoming frustrated and angry in traffic situations can easily trigger responses such as speeding and rule violations. Surprisingly, women in our study were relatively more likely to endorse aggressive/hostile items than men. This is not unlikely, because according to the aggression literature, the role of gender in aggressive behavior and anger is a very complex issue (; Ulleberg (2001); Lajunen et al. (2001)).

Another surprising result, inconsistent with results reported in the literature (Wilson and Jonah, 1988), was that there was no clear significant association between high frequency and high quantity alcohol consumption and high aggression/hostility scores. Alcohol consumption rate is, however, moderate among the participants of the GAZEL cohort as high alcohol consumers were less willing to become members of the cohort. As a consequence, there are almost no heavy drinkers in the cohort (Goldberg et al. 1994). However, some researchers have suggested that demonstrating a clear relationship between alcohol intake and aggression is difficult, because alcohol consumption increases aggressiveness in some individuals but decreases it in others (; , ; ).

Like a recent study conducted in Greece (Chliaoutakis et al. 2002), we found that high irritability was significantly associated with I-A. Theoretically, individuals with high irritability are more prone to lose their temper quickly, and are ready to explode at slight provocation (). Drivers are confronted with wide ranging sources of irritability on roads. A study conducted by EOS Gallup Europe (2003) among 13 673 driving-license holders in 23 countries around the world in the 2002–2003 period reported that the vast majority of respondents (55% to 87%) confirmed that they are sometimes irritated by other drivers. These irritating situations included: aggressive flashing lights, obscene gestures, tailgating, verbal abuse … (EOS Gallup Europe, 2003). A strong relation was also noted between getting irritated and being irritated by other drivers. Highly irritable individuals may run an increased risk of traffic accidents because they are more prone to react negatively to the numerous irritating road situations. A strong reaction could affect the decision-making process (Chliaoutakis et al. 2002), or contribute to the deficit in driving performance because of an increased level of tension or stress (Perry, 1986).

Similarly, individual with high negativism subscore were also slightly more likely to have I-A. Negativistic people tend to be unconventional and individualistic in their response to the world. They have been described as quick to challenge rules or authority deemed arbitrary and unjust. They may also harbour resentment without expressing it directly and may revert to passive-aggressive behaviour to make their feelings known (Strack, 1997). It has been already described that the relationship between mild social deviance and accident involvement was partly mediated by propensity to commit driving violations (). Although the trait could explain why these individuals were slightly at risk of I-A, it is also possible that the finding was due to chance given the large number of statistical tests performed in the study. This caveat could be applied to irritability.

Among the limitations of our study, it should be noted that the sample, despite its large size, was not representative of French drivers as a whole. Although the participants were of diverse trades and socioeconomic groups throughout France, some demographic characteristics were specific. However, previous studies producing results in line with this study were carried out among students aged from 17 to 25, adults aged 25 to 59 years old, and men and women from different socioprofessional categories (Karlberg et al. 1998, Romanov et al., 1994, Novaco, 1989, Magnavita et al. 1997).

It is possible that some accidents might be more closely related to hostility or aggression than others. Because detailed information’s on each recorded injury accident were available, we were able to assess accident’s characteristics associated with the overall BDHI scores. Results did not provide however further interesting insight. Surprisingly, the only significant result was that high overall BDHI scores were associated with a higher proportion of accidents reportedly due to vehicle’s failures. (28.6%, 17.9% and 53.6% for participants with low, intermediate and high overall BDHI scores respectively, p=0.004). It is noteworthy that the power of the analysis was often insufficient when splitting accidents into further categories.

There are several implications that need to be considered in view of the fact that the present study failed to find a significant relation between overall aggression/hostility measures and I-A. Our results suggest that the phenomenon of aggressiveness on the roads and/or anger behind the wheel extend beyond individual’s general propensity toward aggression. Recent studies on the causes of aggressive driving on the roads tend to support this claim. Deffenbacher et al. (2000, 2003) in their studies suggested that a person’s disposition to anger interacts with some irritating road situations, rather than high anger drivers being “ubiquitously angry behind the wheel”. They added that while there is a degree of correlation between anger behind the wheel and general or trait anger, this relationship was not so important as to suggest that anger behind the wheel should be subsumed within general or trait anger (Deffenbacker et al. 2003). In their study of the relationship between general aggressiveness, driver anger and aggressive driving, Lajunen and Parker (2001) also suggested that driver anger and behavioral reaction to the anger seemed to depend partly on situational characteristics. Taking into account that their study failed to find an association between physical aggression and driver anger, they suggested that anger as an emotion state did not always precede aggressive driving behavior.

While much remains to be learned about the factors contributing to anger behind the wheel and/or aggressive driving, it is interesting to note the difficulty of determining one profile of the so-called “aggressive driver”. Mizell (1997) reports that today’s aggressive driver could be “male or female, young or old, educated or uneducated, rich or poor, white or black, Hispanic, or Asian”.

It reasonable however, to suggest that aggressiveness and/or anger behind the wheel may be a phenomenon shared by all road users and that messages on the dangers of aggressive driving and how to deal with it should not be reserved for drivers with aggressive tendencies. It is important in future studies to consider aggression in driving across cultures in order to determine which societal characteristics trigger these behaviors.

Acknowledgments

The project was funded by Renault, Electricité de France-Gaz de France and Fondation MAIF. We thank all those who supported this and in particular the staff of unit 687 of INSERM. Very special thanks go to Sébastien Bonenfant, Nadine Kaniewski, and Nathalie Lopes for their valuable help throughout the study.

Footnotes

References