Impaired driving and motor vehicle crashes among Swedish youth: An investigation into drivers’ sociodemographic characteristics

https://doi.org/10.1016/j.aap.2005.03.001Get rights and content

Abstract

Impaired driving is an important road safety problem, and the characteristics of drivers impaired by alcohol or drugs are relevant to targeted interventions. This study considers young drivers’ sociodemographic attributes: age, sex, class of origin and educational attainment, based on a national cohort of young Swedish drivers (1988–2000) followed up in police registers for their first motor vehicle crash. Of all crashes reported for these drivers, 946 were where the driver was suspected of being under the influence of alcohol or other substances (corresponding to 6% of all first crashes). Impaired driving significantly increases the odds of severe and fatal injuries, regardless of sociodemographic attributes. Observed differences in the proportions of impaired drivers reveal significant excess risks among males, persons aged 18–19, those from households classified as “workers” and “others” (including, e.g. the long-term unemployed and those on long-term sick leave), and those with low educational attainment. Impaired driving is a risk factor in all sociodemographic strata among young people. Members of some groups are more likely to be found than others among impaired drivers on occasion of first motor vehicle crash.

Introduction

It is well known that impaired driving is a major risk factor for motor vehicle crashes (MVCs) (Drummer et al., 2004) and it increases the risk of the severity of the crash (Smink et al., 2005). In 2000, the Swedish police reported 855 motor vehicle drivers who were suspected of being under the influence of alcohol or other substances when involved in a crash leading to personal injury (Statistics Sweden, 2000). This amounts to 3.5% of all MVCs for that year in Sweden.

A comparative study of the five Nordic countries (Sweden, Norway, Iceland, Finland and Iceland) has shown that the highest frequencies of driving under the influence of drugs were detected in samples from Sweden and Norway, although blood alcohol concentrations (BACs) were below the legal limit in 83 and 80% of cases, respectively (Christophersen et al., 1999).

Selection of an appropriate target population is a significant challenge in the design of prevention and intervention efforts aimed at reducing alcohol and other substance related road traffic injuries (RTIs). In this respect, young drivers represent a group of high concern for various two reasons. First, the incidence of both drink driving and drug-associated driving is higher among them that among older drivers (Phebo and Dellinger, 1998, Shults et al., 2002, Kelly et al., 2004, Kruger and Vollrath, 2004). Second, young drivers show higher levels of impairment than older drivers with the same BAC (Phebo and Dellinger, 1998).

As well as being young, being male is also linked to both a higher prevalence of drunk driving and a higher rate of MVCs related to impaired driving (due to either alcohol or drugs) (Baker et al., 2002, Caetano et al., 2004). Yet, there is some evidence that impaired driving prevalence among females has increased in recent years, in particularly in Norway (Skurtveit et al., 2002).

Apart from age and sex, other sociodemographic attributes have been considered to a limited extent when studying the prevalence of impaired driving and to MVCs related to impaired driving. Yet, as there is evidence suggesting that there are important social gradients in crash risk in youth (Laflamme and Engström, 2002, Hasselberg and Laflamme, 2003, Hasselberg and Laflamme, 2004) and that social gradients even exist in both alcohol consumption (Norström and Romelsjö, 1998, Romelsjö and Lundberg, 1996) and impaired driving (Kelly et al., 2004, Karlsson and Romelsjö, 1997) (all age aggregated) increased knowledge of personal sociodemographic characteristics associated with alcohol or other substance-related MVCs may help identify the subgroups at greater risk.

In that vain, it may be worth paying attention to the type of crash in which young people from different sociodemographic groups are involved as there is also evidence that social gradients may vary depending on whether one studies single or multiple vehicle crashes (Hasselberg et al., 2005, Laflamme et al., 2005). There is also a possibility that under-reporting of alcohol or drug use is more frequent in some types of crashes than others (Forsman and Gustafsson, 2004).

This paper broadens investigation into the association between MVCs and drivers’ sociodemographic characteristics. It considers a national cohort of young Swedish drivers followed in police registers with regard to their first car-crash experience. Two main questions are addressed:

  • 1.

    To what extent does impaired driving (due to alcohol or drugs) increase the risk of severe or fatal injury in MVCs among young drivers?

  • 2.

    Does involvement in impaired driving crashes (single or multiple vehicle crashes) leading to personal injury vary between categories of single sociodemographic personal attributes (sex, age, class of origin and educational attainment)?

Section snippets

Study setting and population

The study is register-based and encompasses all individuals born during the 3-year period 1970–1972, as recorded in the Swedish Population and Housing Census of 1985 (Statistic Sweden, 1989) (n = 329,716). Individual records from the census of 1985 were linked to road traffic crashes (RTCs) data that were obtained from the Swedish National Road Administration, which is based on police reported road traffic accidents. These data include details of crash circumstances, severity of collisions,

Results

From 1988 to 2000, police reported a total of 16,178 MVCs (implying 16,178 drivers), of which 10,631 led to personal injury. The police reported 946 drivers who were suspected of being under the influence of alcohol or other substances. This amounts to 6% of all crashes involving motor vehicle drivers.

Table 1 shows the resulting of the logistic regression for severe and fatal injuries. Main effects are found for age, class of origin and educational attainment, and being under the influence of

Main findings

The results from this cohort study strongly support the unsurprising assertion that impaired driving (due to alcohol or drugs) is related to a greater likelihood of severe and fatal injuries regardless of sociodemographic characteristics of drivers. In this sense, the findings of the study echo those of several previous studies of the correlates of MVCs, where impaired driving due to alcohol or other substances has been shown to be an important factor (Kruger and Vollrath, 2004, Movig et al.,

Conclusion

The results of this cohort study provide additional evidence that alcohol and drugs contribute to an increased risk of traffic accidents with a severe and fatal outcome. It also provides new insight into the social distribution of impairment-related MVCs. Impaired driving significantly increases the odds of severe and fatal injuries regardless of sociodemographic characteristics. Differences found between the proportions of impaired drivers reveal significant excess risks for male drivers,

Acknowledgement

The study was funded by the Swedish National Road Administration.

References (47)

Cited by (47)

  • Different demographic and drinking profiles of motorcyclists and car drivers with the first-time offense of driving/riding under the influence of alcohol

    2020, Accident Analysis and Prevention
    Citation Excerpt :

    A nationwide cohort study reported that people with a low socioeconomic status had a higher risk of road-traffic injury than those with the highest socioeconomic status, suggesting that those with a disadvantaged socioeconomic status might be more involved in DUI (Zambon and Hasselberg, 2006). In line with this observation, another study also reported higher proportions of impaired driving among youth with a low socioeconomic status (Vaez and Laflamme, 2005). Motorcyclists constitute a substantial proportion of DUI offenders and are susceptible to severe injury, leading to heavy health and financial burden (Preusser et al., 1995); therefore, we suggest that distinct and appropriate preventive strategies against DUI motorcyclist offenders should be established.

View all citing articles on Scopus
View full text