Short Communication
Previous convictions or accidents and the risk of subsequent accidents of older drivers

https://doi.org/10.1016/S0001-4575(01)00014-8Get rights and content

Abstract

The over-involvement of elderly drivers in collisions has a potentially adverse effect on highway safety. The question for most experts in traffic research is whether we can predict the individual risk of accidents and which variables are the best predictors, especially for this population. For a better understanding of the elderly drivers’ problems, this study aimed to describe the most common types of accidents in the elderly population of drivers living in Quebec (≥65 years of age). The second objective of the study was to analyse the relationship between previous accidents or convictions and the risk of subsequent accidents. The results show that: (1) elderly drivers are characterised by error accidents involving more than one car, especially at intersections, (2) prior accidents are a better predictor for accident risk than prior convictions and (3) these trends steadily increase with each age group (drivers 65 years old to 80 years or more). The results are discussed in relation to the literature on risk behaviour of the elderly drivers.

Introduction

In industrialised countries, drivers over 65 years of age represent a rapidly growing part of the driving population. In terms of the absolute number of accidents, elderly drivers do not appear at a higher risk than the general population. However, they have 1.6–2.4 times more accidents per driven kilometre than drivers between the ages of 25 and 64 (OECD, 1985). Nevertheless, we have to know that the usual interpretation of accidents per driven kilometre as a measure of risk exaggerates the apparent risk of low-mileage groups, like for the older drivers group (Janke, 1991). Older drivers are more often considered responsible for these crashes (Verhaegen, 1995), and they have more fatal accidents (Hakamies-Blomqvist, 1993), especially because they are more vulnerable (Evans, 1988, Dulisse, 1997). Moreover, different studies (Parker et al., 1995, Blockey and Havley, 1995, Stamatiadis and Deacon, 1995, Ryan et al., 1998) have shown that elderly drivers are more frequently involved in specific types of accidents (e.g. situations involving more than one vehicle, especially at intersections).

Many researchers have attempted to determine which variables are the best predictors of accidents for the driving population in general, and more specifically, for elderly drivers. It has been found that the number of prior traffic convictions is one of the best single predictors of accident risk (Gebers, 1990). Gebers and Peck (1992) showed an interaction effect of age and prior convictions on subsequent car accidents. The relationship between the number of prior convictions and the risk of a subsequent accident appeared stronger with the oldest groups of drivers (aged 60–69 and above 70).

The present study is an attempt to replicate the results of Gebers and Peck (1992) using a larger group of elderly drivers. In order to determine if the trend observed by these researchers is confirmed (i.e. a still stronger relationship between prior convictions and subsequent risk of an accident for the oldest of the elderly drivers), it seems important to analyse further the interaction between prior driving record (convictions and accidents) and age, among a large sample (in fact, an entire cohort of elderly drivers) including more drivers over 75. It is known that the elderly population is very heterogeneous and, moreover, that the probability of experiencing a reduction of physical or cognitive capacities (e.g. visual, auditory or visual attention, etc.) increases with age, mostly due to age related diseases. For example, in the active population (excluding hospitalised persons), 31% of people between the ages of 65 and 74 were shown to have an incapacity; this rate increases to 50% for those between the ages of 75 and 84, and to 77% for those over 85 years (Gauthier and Duchesne, 1991).

Section snippets

Population

With the collaboration of La Société de l'Assurance Automobile du Québec (SAAQ), the driving records of 426 408 elderly drivers (60 years old and older) were analysed for a 6-year period (1992–1997). For the period studied, these drivers represent the entire population of elderly persons living in Quebec holding a valid driver's license. In this cohort of drivers, 63.8% are male. The majority of these drivers maintained a good driving record for the 6-year period, with 79% being conviction free

Descriptive analysis of elderly drivers’ accidents

The descriptive analysis of these accident records for the selected population confirms that elderly drivers’ crashes arise most of the time in situations involving more than one car (90.2%). Accidents at intersections are quite frequent (29.4%) with a significant part of these happening when turning left (13.8%). Collisions occur more frequently in good conditions (sunshine: 81.9%, dry road surface: 59.1%, good visibility: 92.8%). In the present study, we observed that these crashes often

Discussion

The descriptive analysis of violation and accident records for a large cohort of elderly drivers confirms what other studies have shown (Hakamies-Blomqvist, 1990, Hakamies-Blomqvist, 1993, Verhaegen, 1995): elderly drivers are characterised by accidents involving more than one car, especially at intersections. Moreover, our study shows that right-angle and left-turn collisions at intersections steadily increase after age 65 while single car accidents on a left curve decrease. Older drivers’

Acknowledgements

The authors are grateful to la Société de l'Assurance Automobile du Québec (SAAQ) and to the le Fonds pour la Formation de Chercheurs et l'Aide à la Recherche (FCAR, Québec) for sponsorship of this project. The authors further thank Diana Raiwet for help in preparing the manuscript.

References (20)

There are more references available in the full text version of this article.

Cited by (52)

  • Backpropagation – Artificial Neural Network (BP-ANN): Understanding gender characteristics of older driver accidents in West Midlands of United Kingdom

    2020, Safety Science
    Citation Excerpt :

    The AgeUK survey in (Holley-Moore and Creighton, 2015) found that forty-three percents of drivers aged 60 and over would stop driving because of health concerns while only one percent of this age group would stop driving because of their age. The cognitive impairment of older drivers slows the perception time and increases the difficulty to comprehend the complex traffic conditions resulting in higher casualty rate (Daigneault et al., 2002). McGwin and Brown (1999) observed the characteristics of different age groups involved in traffic accidents in Alabama, USA, and stated that older drivers were involved in accidents at junctions failing to yield the right-of-way.

  • Estimating likelihood of future crashes for crash-prone drivers

    2015, Journal of Traffic and Transportation Engineering (English Edition)
  • The Canadian Safe Driving Study - Phase i pilot: Examining potential logistical barriers to the full cohort study

    2013, Accident Analysis and Prevention
    Citation Excerpt :

    Older drivers, on the other hand, are experienced and less likely to engage in unsafe driving behaviours (National Highway Traffic Safety Administration). Nevertheless, older drivers have one of the highest rates of collision per mile driven (Braver and Trempel, 2004; Daigneault et al., 2002b; Di Stefano and Macdonald, 2003). One explanation for increased collision risk in this age group is characterized as “low mileage bias” (Gallo et al., 1999; Kostyniuk et al., 2000).

View all citing articles on Scopus
View full text